<%BANNER%>

Obesity and Postpartum Depression

Permanent Link: http://ufdc.ufl.edu/UFE0022424/00001

Material Information

Title: Obesity and Postpartum Depression Does Prenatal Care Utilization Make a Difference?
Physical Description: 1 online resource (207 p.)
Language: english
Creator: Sundaram, Swathy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: bmi, maternal, morbidities, obesity, postpartum, pregnancy, prenatal, utilization
Health Services Research, Management, and Policy -- Dissertations, Academic -- UF
Genre: Health Services Research thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: OBESITY AND POSTPARTUM DEPRESSION: DOES PRENATAL CARE UTILIZATION MAKE A DIFFERENCE? Using a national continuing population-based survey known as Pregnancy Risk Assessment Monitoring System (PRAMS), this study sought to determine the role that PNC utilization plays in the relationship between pre-pregnancy BMI and PPD symptoms. Two years of data, 2004 and 2005 were analyzed among women from 16 states. Two specific aims were examined: 1) the association between pre-pregnancy BMI and PPD symptoms, and 2) the association between pre-pregnancy BMI and PPD symptoms after considering PNC utilization as a moderating variable. It was predicted for the first specific aim that the odds for PPD symptoms would increase as pre-pregnancy BMI increased. For the second objective, it was predicted that the association from the first specific aim would carry over and remain the same (e.g., obese pre-pregnancy BMI would have the highest odds for PPD symptoms), but that within each pre-pregnancy BMI group, the odds for PPD symptoms would decrease as PNC utilization increased (within obese pre?pregnancy BMI, inadequate PNC would have higher odds than intermediate PNC). The general premise for PNC utilization acting as a moderating variable in this study was that PNC can help address the changes that occur during pregnancy with regards to pre-pregnancy BMI (as a biological and psychosocial stressor). Thus, delivering PNC incorporating nutrition, weight and shape changes, and addressing a woman?s concerns about her weight and shape would in-turn, reduce the odds of PPD symptoms. Since the sample used in this study included women from all pregnancy risk statuses, two risk-adjustment approaches were carried out to identify an association between pre-pregnancy BMI and PPD symptoms, and a moderating effect of PNC. One approach included all women in the dataset and used statistical analyses to risk-adjust for pregnancy risk status, and the other approach modified the design of the study by truncating the population of women to include healthy pregnancies only. Results initially showed an association between obesity and PPD symptoms, and PNC and PPD symptoms among the bivariate and multivariate analyses. However, the inclusion of a variety of control variables into the multivariate models removed these associations. Overall, for both approaches, there was no indication of a moderating effect of PNC utilization. However, results from the analyses showed that many of the women were significantly affected by a variety of medical and obstetric problems, many of which were high-risk. It is recommended that future research investigate the possible association of these problems with PPD symptoms. For practice, it is suggested that PNC providers identify the medical and obstetric problems faced by their patients, focus on both the physical and the potential psychosocial consequences of those problems, and establish suitable interventions accordingly.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Swathy Sundaram.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Harman, Jeffrey S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0022424:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022424/00001

Material Information

Title: Obesity and Postpartum Depression Does Prenatal Care Utilization Make a Difference?
Physical Description: 1 online resource (207 p.)
Language: english
Creator: Sundaram, Swathy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: bmi, maternal, morbidities, obesity, postpartum, pregnancy, prenatal, utilization
Health Services Research, Management, and Policy -- Dissertations, Academic -- UF
Genre: Health Services Research thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: OBESITY AND POSTPARTUM DEPRESSION: DOES PRENATAL CARE UTILIZATION MAKE A DIFFERENCE? Using a national continuing population-based survey known as Pregnancy Risk Assessment Monitoring System (PRAMS), this study sought to determine the role that PNC utilization plays in the relationship between pre-pregnancy BMI and PPD symptoms. Two years of data, 2004 and 2005 were analyzed among women from 16 states. Two specific aims were examined: 1) the association between pre-pregnancy BMI and PPD symptoms, and 2) the association between pre-pregnancy BMI and PPD symptoms after considering PNC utilization as a moderating variable. It was predicted for the first specific aim that the odds for PPD symptoms would increase as pre-pregnancy BMI increased. For the second objective, it was predicted that the association from the first specific aim would carry over and remain the same (e.g., obese pre-pregnancy BMI would have the highest odds for PPD symptoms), but that within each pre-pregnancy BMI group, the odds for PPD symptoms would decrease as PNC utilization increased (within obese pre?pregnancy BMI, inadequate PNC would have higher odds than intermediate PNC). The general premise for PNC utilization acting as a moderating variable in this study was that PNC can help address the changes that occur during pregnancy with regards to pre-pregnancy BMI (as a biological and psychosocial stressor). Thus, delivering PNC incorporating nutrition, weight and shape changes, and addressing a woman?s concerns about her weight and shape would in-turn, reduce the odds of PPD symptoms. Since the sample used in this study included women from all pregnancy risk statuses, two risk-adjustment approaches were carried out to identify an association between pre-pregnancy BMI and PPD symptoms, and a moderating effect of PNC. One approach included all women in the dataset and used statistical analyses to risk-adjust for pregnancy risk status, and the other approach modified the design of the study by truncating the population of women to include healthy pregnancies only. Results initially showed an association between obesity and PPD symptoms, and PNC and PPD symptoms among the bivariate and multivariate analyses. However, the inclusion of a variety of control variables into the multivariate models removed these associations. Overall, for both approaches, there was no indication of a moderating effect of PNC utilization. However, results from the analyses showed that many of the women were significantly affected by a variety of medical and obstetric problems, many of which were high-risk. It is recommended that future research investigate the possible association of these problems with PPD symptoms. For practice, it is suggested that PNC providers identify the medical and obstetric problems faced by their patients, focus on both the physical and the potential psychosocial consequences of those problems, and establish suitable interventions accordingly.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Swathy Sundaram.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Harman, Jeffrey S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0022424:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101107_AAAAAW INGEST_TIME 2010-11-07T10:41:09Z PACKAGE UFE0022424_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 3833 DFID F20101107_AAAPVK ORIGIN DEPOSITOR PATH sundaram_s_Page_089thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
76375479fca07b60e348d725cef44511
SHA-1
23207bf0137ddd82138518f6be4a644329315bc3
6685 F20101107_AAAPUV sundaram_s_Page_075thm.jpg
a7b4a085b71d2b05dc14a74a56bd27e4
3376e9f74b260fd5f692a0eeb98f1f0a9adb26ef
1054428 F20101107_AAAOSI sundaram_s_Page_151.tif
acbebbe96098bc3f18b0aad7d62764a3
c7f17f31313f67d3442c30d59312e584bf57a5a1
25271604 F20101107_AAAORT sundaram_s_Page_136.tif
8a5dfc88ff433b187d9c50507f1b0ded
eae6eb40426d7ba5c3c7360d4dae512d2f713367
12221 F20101107_AAAPVL sundaram_s_Page_091.QC.jpg
63a0256c4d30ad6c2322dd342fb0d865
ba5e552e5a0f4b43a023114ec3d3fa461a4d9396
6705 F20101107_AAAPUW sundaram_s_Page_077thm.jpg
f3c9c9ff40c69f125d804869a8cc0105
205da86a699705b5eaa4f9c16a3bdaaa38a8c2e3
F20101107_AAAOSJ sundaram_s_Page_152.tif
0a615b4829c3f1117cef425ed147ec17
c4a1349b37e91d5bd33b4fbaae02d73a5c4535d4
1053954 F20101107_AAAORU sundaram_s_Page_137.tif
5e6da9585c1f385d7c7cfc2535c9f3e1
9f9b03f8efe03d0ec6be485f4de1b3524a8c93e3
23288 F20101107_AAAPWA sundaram_s_Page_104.QC.jpg
b88f78460472fb5bfd09daecba933d29
986174b2b80612de19575ac13b8a53116ee05e91
3825 F20101107_AAAPVM sundaram_s_Page_091thm.jpg
450145f6b3781880e1f38709e8c07a13
87cfc1b10be2d663ef2b81b1ff684f2afe5558d9
18546 F20101107_AAAPUX sundaram_s_Page_078.QC.jpg
6af526734ca111bff4b2a7eef95027df
80cc0887ed2a51cf62312179f5aef332d1611edf
F20101107_AAAOSK sundaram_s_Page_153.tif
0d701a6f69a9ec158bf9208fe191bde9
04c2838494686896d35bd585bad57a45477cebdf
F20101107_AAAORV sundaram_s_Page_138.tif
4518d6422bdc35c903adeac4e6ffe70e
5cb685ec5b08e05bff3fe9a0c4ad27233c78da85
23738 F20101107_AAAPWB sundaram_s_Page_105.QC.jpg
66c97fe21e7808d4a0dafeacb45d90e8
f2ead9b548387b0eb361377242aa87b012461bdf
11060 F20101107_AAAPVN sundaram_s_Page_092.QC.jpg
06a2a45e457a56b1d719b6a4a5ed0135
9960f34593b33d83df1e4b5c3ce4c2b0c9884860
5247 F20101107_AAAPUY sundaram_s_Page_078thm.jpg
ea72893b12f5e9ab7ba2b103c28be456
73046693cee5a015d2c91ca0b893e908ab7e66c2
F20101107_AAAOSL sundaram_s_Page_154.tif
0dcdb80b77409a04ed2bd56ce90bb6cc
edd86196686cdc11434fb9218ec70f082eb53bea
F20101107_AAAORW sundaram_s_Page_139.tif
2c9c6bec292852d9dd871bd948a02a95
e280e0cb25d80f3bc88ede0478320ea1e2450fa6
23400 F20101107_AAAPWC sundaram_s_Page_107.QC.jpg
8f45cc8df325f6603002602f221f68e4
44244f256f9d310e925dc62af0d35e77d8ede84b
3513 F20101107_AAAPVO sundaram_s_Page_094thm.jpg
b9f6a762af9c83cc1e110fe825ebca9b
5e8fc9a5278e0c9ea5b20e7e5e7f5e20af6981c6
16536 F20101107_AAAPUZ sundaram_s_Page_081.QC.jpg
79638ede12987e55527cc7c78c9d1f92
da1080283fa5275d486d5f14e562673e34b6ccfd
F20101107_AAAOSM sundaram_s_Page_155.tif
e8960184e3da8486dbe32dff854eb369
0fe5e018313c34d012d2653f16be49e6ceaf0c2a
F20101107_AAAORX sundaram_s_Page_140.tif
8a01feb0207f026e499ef8f5bb569314
9d132792ad37e11d4fd6d763c8d17ff590b695c5
F20101107_AAAOTA sundaram_s_Page_169.tif
aad6d08b0747a44773b69e515f3d9dbb
7b26610f3197f8dd33c2c2ad327c17986209a7e9
6056 F20101107_AAAPWD sundaram_s_Page_107thm.jpg
1dfd58852550caa77decea6d093d6796
e9bd8c805c9b73464b84c11292de22e258e62f5d
12941 F20101107_AAAPVP sundaram_s_Page_095.QC.jpg
01020f8169e2e17a2715adb3a7ec553c
9b43250b714307b963f3c717a25dc6f1b64cc936
F20101107_AAAOSN sundaram_s_Page_156.tif
8ba50e82d6a8cdb7c5d12121d93ebf90
1ec3ad1fa5db361671fe46b4332b8e6efc68ed1c
F20101107_AAAORY sundaram_s_Page_141.tif
5fae13b637702455367a3d2dacf70355
08e186fc3e9237bb0d614f2848073bbc3181a662
F20101107_AAAOTB sundaram_s_Page_170.tif
e4f00b7a9a692b1997645a01a9ef8e96
3fe526cfeec0e7bd204e3a62fb3c36fc6a132322
8980 F20101107_AAAPWE sundaram_s_Page_108.QC.jpg
41452195089e3a130490b499fcd5f7ce
82aba9b06c2e987fff4a983a7cfe7d56fa3f32a2
3809 F20101107_AAAPVQ sundaram_s_Page_095thm.jpg
51a9aea509a12151e30e01eca2d79dfc
bcf8612f703bc406f57891d228ede912267da8a4
F20101107_AAAOSO sundaram_s_Page_157.tif
20d7c950e8f1815dc04ce2e16651dea6
8871d98084a04269c135eb7c9938643a1d003159
F20101107_AAAORZ sundaram_s_Page_142.tif
f8f000c33fe86250a40f21e23d2119bc
736659ae9b4e2bd2c672c441dc433e508bda26dc
F20101107_AAAOTC sundaram_s_Page_171.tif
a9f0145f8b95499c436aa998bb834b22
81ec23f5dd60b2bfd98cff700eb1fdb013bb7f3c
26916 F20101107_AAAPWF sundaram_s_Page_110.QC.jpg
7f7bad3621861480628c213e2017bf68
92db9f25549d69e7db7841ad6805423ae850d18a
F20101107_AAAOTD sundaram_s_Page_172.tif
41a0f4ba11069b794e728d5ea498e507
61d97bc3e1c65be350407b7ec2d6f15fd375f084
26580 F20101107_AAAPWG sundaram_s_Page_111.QC.jpg
ad7e8bd362142ce363406adf2dc34491
77e113380cf1b3604dc9727ee72c5f1dc865c421
12306 F20101107_AAAPVR sundaram_s_Page_096.QC.jpg
547a3d7c174c851853c074d223da4dad
4a47e09a432c5f8c939f7cc181fe55165396d4ac
F20101107_AAAOSP sundaram_s_Page_158.tif
e7f228bf9aaa2749f3d7104415e2fce0
7bbd8962f8403a5db5707d48807e552554a65840
F20101107_AAAOTE sundaram_s_Page_173.tif
3c2dc19b1ada266d73a51d0508e64cd2
c91153a799ea10c0829f8d546a0fd7c6b43d8f42
7015 F20101107_AAAPWH sundaram_s_Page_111thm.jpg
0d00a11a04d842fa2807f5a00b8f900c
10a52be5fd79b0cf236bc459f384ea0eeed1d8f1
3636 F20101107_AAAPVS sundaram_s_Page_096thm.jpg
74eba155d8fe71e589da7f63980941ad
7489b1b3e7d320c7f0c300a3ee23eaeed1eea0a3
F20101107_AAAOSQ sundaram_s_Page_159.tif
256340286e958d273dc4e648118ebdad
d64c0d345105e32b55e4038d865fd92d28ac9c53
F20101107_AAAOTF sundaram_s_Page_174.tif
985814f63947e31fb3b826b5e9eb3b6d
cbfae04288d80b69d49be1f5d42efb51889e865a
7127 F20101107_AAAPWI sundaram_s_Page_112thm.jpg
1c4198bcb04d97edcfa530d9fff3155e
76b41724fb2d8a0a0e55f22fa052733ca772d828
13678 F20101107_AAAPVT sundaram_s_Page_098.QC.jpg
34674d81732986f3da41d2150bd4bb45
36ede1e22e2335e3cbe4054fe9fc1c429b06ef38
F20101107_AAAOSR sundaram_s_Page_160.tif
a93b31f1e5ec25a93eb96442c8d437e9
050dc0ca1fdc2989d84ec614851b621ce5e85179
F20101107_AAAOTG sundaram_s_Page_175.tif
2ac1686eece805ae81f2e1995bace243
6a2fab76e06a1946f2c9a00d239cd4fedb7168af
7046 F20101107_AAAPWJ sundaram_s_Page_115thm.jpg
0efa0da7466fb1d55755be4c59a391fa
a70c5e06bc8e23b897499b24dfda9aed38b3626d
23011 F20101107_AAAPVU sundaram_s_Page_099.QC.jpg
d33e9647456140c08dd100bb7bdab953
72b3d305fe0252ee4a8abe3567732326244906de
F20101107_AAAOSS sundaram_s_Page_161.tif
352c9c52dcc9f0c04f097c36c9e267bb
f197dd299b4b4d110bc95def7310a5e1a815c884
F20101107_AAAOTH sundaram_s_Page_176.tif
0f7787a092e27ee6b0c82a00dcf3473d
cbdeef64dcb5de37ab447dd8bcb901688e1bcfd7
6915 F20101107_AAAPWK sundaram_s_Page_116thm.jpg
a997363daf17e13ae8be03fb01cff977
cc454a6a17e650ec55131bef3c45a8b58e6539f0
6140 F20101107_AAAPVV sundaram_s_Page_100thm.jpg
2940ee9df63d3dc97bae283020ec081e
64f81cdb7888ca012f85cc8df567f0ae2bfa774d
F20101107_AAAOST sundaram_s_Page_162.tif
bc6932bd26976a77ea87d9ab2b6244c1
836ed14a3b974b8cedbd1d864a7fdc9aa3058b0a
F20101107_AAAOTI sundaram_s_Page_177.tif
a364a3aaf150ef12700082d586a865cd
7728098064dc8003916738aa1b698367b5a38eb9
24383 F20101107_AAAPWL sundaram_s_Page_117.QC.jpg
3382c46c140940a945706f5d8a8aa5e6
c6df1e3f40e180f6ed2b95fb25d6fade1e1dbb7c
23479 F20101107_AAAPVW sundaram_s_Page_101.QC.jpg
bc77ccca3ff7f9d36338afd0a4da05d9
c599b5722542dce927cc883239920b479ee82dc4
F20101107_AAAOTJ sundaram_s_Page_178.tif
912ae5e3bc1e0bc01af45dae26890d7f
84d9b6daa35c0aef542f68e9cd98f63f0ac1c849
F20101107_AAAOSU sundaram_s_Page_163.tif
cfa5f27fcda8dd35690e88fbef302e81
7753ba8d98fd78de7949657f9e8edbb3a0eaaf17
7090 F20101107_AAAPXA sundaram_s_Page_129thm.jpg
387c57ee411553d538bbcd6f863ad83a
b5ba31f05651d6a7eaaa96ad173848387e6b3d7b
6908 F20101107_AAAPWM sundaram_s_Page_117thm.jpg
d991ef8630bdf8fde3cf3d2caf10ef19
d99f443ea9ef2110dc9c4eaf79ceda590c93d8af
6166 F20101107_AAAPVX sundaram_s_Page_101thm.jpg
812a17444c9381f5f108c3dd1114d448
20d59ba6f7e8e274c262c7a6f59c102d5d7c7a7d
F20101107_AAAOTK sundaram_s_Page_179.tif
3e09c314465b5b2ed88a197de6954c82
4162160768d2c1bf1650d77107291c00725ba129
F20101107_AAAOSV sundaram_s_Page_164.tif
cf3badaea6c2a1dd740ace3f30f33bfb
4928dffb00f65e511bf058ac558ca655c8177ca6
27171 F20101107_AAAPXB sundaram_s_Page_131.QC.jpg
e627bcb6cbb16baf3e55604b0ca08550
a58bcf7122b254e1f618d96456af43282d7ba52c
26131 F20101107_AAAPWN sundaram_s_Page_118.QC.jpg
f9586d8f92b645fb839b884e89f26a7a
8d9b99b0a24d2ac69df7fc9ed204e93b808cb780
22039 F20101107_AAAPVY sundaram_s_Page_102.QC.jpg
f24511c1c70814314eeb8c9ee2f8a364
922df274f369fc644b6ddc9a08b810508b86bcda
F20101107_AAAOTL sundaram_s_Page_180.tif
43a6aa17b9546dbd6b10c6055069077c
f94353521c1eb7664401c2b0992507865797a039
F20101107_AAAOSW sundaram_s_Page_165.tif
48b8fd61b3824cfc1fd14e81aeb094bb
120ecd6b046c0bce3409eeeacd488626622f00d6
7197 F20101107_AAAPXC sundaram_s_Page_131thm.jpg
ef86da9f73baaabe4e1762266f07801b
5d1d86b00c3a953709270cb52e0482ded5692ef1
7264 F20101107_AAAPWO sundaram_s_Page_118thm.jpg
8b419f546e09c8b2e047715cf7e84fb3
7f0773721c632aae888b73647003feb4006cc3b0
6792 F20101107_AAAPVZ sundaram_s_Page_103thm.jpg
e13b4082233294be31c4ea92e98e98fb
d61845df52fccd90cbcc41f77a01c3c370a700bd
F20101107_AAAOUA sundaram_s_Page_195.tif
a149844aa13415cecb4b776f6d09cc7c
faca95bc2ab08bd9b7b1ff0daee039e431fbf83d
F20101107_AAAOTM sundaram_s_Page_181.tif
133fff825de0b4cf8e6e3341136fc148
3e6b1ed042910ae59bdfba9b4443cc8e21fcd631
F20101107_AAAOSX sundaram_s_Page_166.tif
4f3164ff3879c86d5aa8e812e94d711f
44a8a8383cfda351d4216b5543aa35247073d945
6754 F20101107_AAAPXD sundaram_s_Page_132thm.jpg
10cfc036f4f9eab210bf98bc4d65308b
3a4d3ffdcc8c810165b01a1b43a2ed31818ee3cd
26645 F20101107_AAAPWP sundaram_s_Page_120.QC.jpg
86df7fc59fbe907cef7f92a44314ef3c
e02b25c0de332f3975ebc756c67b752ce15430ee
F20101107_AAAOUB sundaram_s_Page_196.tif
06eb5bb3ae13e039b46b137c6905be5f
97c14e579c723b147d98317ad67929962ba80d00
F20101107_AAAOTN sundaram_s_Page_182.tif
8ceaad2acce24f0dd77783e9704504ea
fc82637e3787178d96e96178d634c0ef93f95ebf
F20101107_AAAOSY sundaram_s_Page_167.tif
08f8c4cdb91efee8e3691628d5fe876f
42a778d2bb6115b20ded4a1183438d47791f4aa5
6965 F20101107_AAAPXE sundaram_s_Page_133thm.jpg
1b4d03d81f1e39669e5a4d49c58665c4
5dc9e310f75d8d14f1e4141226461dbf45e92f11
7154 F20101107_AAAPWQ sundaram_s_Page_120thm.jpg
a1fa86eda52bfc558301e7a2028e0491
ed24dbee38157ba0f1fd4a78bc64689987e239ec
F20101107_AAAOUC sundaram_s_Page_197.tif
4083acaca2c0c7e5f556238bd6d74ef4
d7d8491e928e20cd64fabef53858c71597740ad6
F20101107_AAAOTO sundaram_s_Page_183.tif
6036da0a3f3501b5f1a4dab1d5f9101d
a705c6180a2c67c68085dcecab22e4b2fcab0cb1
F20101107_AAAOSZ sundaram_s_Page_168.tif
1f2dcb2749932773f75b0c8d50d75aba
9ad224042eb9fb5a3572c8aff9e94e1ab1acae06
7120 F20101107_AAAPXF sundaram_s_Page_134thm.jpg
0b0c097c668875d52ef0e8fa9301a8df
afa90012e7031cca8791c732f6e5f6051875dcb9
7322 F20101107_AAAPWR sundaram_s_Page_121thm.jpg
452a4d8e1f35a8f5730c1b6ab4a230ee
76339f92ef1c9d89981028f70459a2e077ef46fe
F20101107_AAAOUD sundaram_s_Page_198.tif
2ad5b3d2340343d4f8b8a6435403f33a
57018ab697af06ddd97642cc47969625f758cfd7
F20101107_AAAOTP sundaram_s_Page_184.tif
f5163079dda29d6c8204ee195fa3259a
2d3ad6317f210c65169327042d9e0888729cd98f
15156 F20101107_AAAPXG sundaram_s_Page_135.QC.jpg
69b1b8848ea07321294420fdcec6b7b2
1f15164b3c9940f521f22ce2c9a9025c43bca1c5
F20101107_AAAOUE sundaram_s_Page_199.tif
ed504c1dd4d5a630b8440017da19bd42
3702ea74f2c39ffa63ec8bb1ded7bbd174d9a217
4242 F20101107_AAAPXH sundaram_s_Page_135thm.jpg
662ac1e1d357412ede85132b49f2239d
b3eaa805383a8406256bf7cc23f4399f13f5352f
6834 F20101107_AAAPWS sundaram_s_Page_122thm.jpg
d3efc81aeb6f577c6ee13513e093f95f
7db3d156f28e85696987a661f4740792c18930e0
F20101107_AAAOUF sundaram_s_Page_200.tif
67ac42cc581733e397cf82b330800298
38df3fddedeed4c4efec88a27a67d5284843dc3f
F20101107_AAAOTQ sundaram_s_Page_185.tif
51798a97a13ff64d44632de4b162ab6b
e8bcc92953bdf9ca0ffdd272c61ebc846489b7f4
2783 F20101107_AAAPXI sundaram_s_Page_136.QC.jpg
4481e0b72081dcd19573fc1d2c5b3453
b8d01f6735510d8c91f64a7031c33afe5a9bd679
7107 F20101107_AAAPWT sundaram_s_Page_123thm.jpg
78f3c0748837341be85962a0118bce2d
22a651188673ebe4e8a54fcdbf7eddf91ab80d76
F20101107_AAAOUG sundaram_s_Page_201.tif
bfc330a79fa3af2c89f8adcf929f867d
ef1ee97acf8f062be9038688293551b80a774f4c
F20101107_AAAOTR sundaram_s_Page_186.tif
3f33a07cabe464b602535c2ea2dd4632
fd89101b827f4f48681af10d1b034ecdadc0eeb0
29665 F20101107_AAAPAA sundaram_s_Page_145.pro
62d4d81209813e8a7fcfd8c7653989a7
4412586fe75d71f3d183744257846a05289c0a48
1223 F20101107_AAAPXJ sundaram_s_Page_136thm.jpg
e3f20a96fb061c79dd92e7226aa7e77d
d9690b9fd207fe16e462e0a3d4572b03084046a5
7043 F20101107_AAAPWU sundaram_s_Page_124thm.jpg
9f2be60febc254e9cd05376c8db2edea
b98e410242eb447175c99802d2a50fbd2667cd6d
F20101107_AAAOUH sundaram_s_Page_202.tif
0b476a3c1045cc804b6c21947c9ba5c0
51e77f64f07567ea046ea46e22914447b853882e
F20101107_AAAOTS sundaram_s_Page_187.tif
ec6ecbd44876b043c89d2cca22ffc5b2
40181a1f1c7d37876800389be77bfbebd824b643
53262 F20101107_AAAPAB sundaram_s_Page_146.pro
3a1e41d58588cbae5ee421f5bd97973b
5fd4267eff17cdf9b18b907616073a5b4d46fd15
2657 F20101107_AAAPXK sundaram_s_Page_139thm.jpg
303f972c7279c0af990925fdd632106e
02963bc8a242cac521fed5d6d61a617f4d0313c3
26904 F20101107_AAAPWV sundaram_s_Page_127.QC.jpg
9094d363c055fbbc637bcc252c0d1ac3
686c92aca6545a50d4c032611cfccd9147e712a4
8423998 F20101107_AAAOUI sundaram_s_Page_203.tif
c1187ec888968e265c1d7f1fff992930
dd2156dfa9e9d5a5ebd883f262fc8109d4081a9f
F20101107_AAAOTT sundaram_s_Page_188.tif
97fefa3899eb1e115e4d298f4c0e0d23
aabce9ef48c9d421d5786acc95f188207324d1e6
39936 F20101107_AAAPAC sundaram_s_Page_147.pro
025d74c64824bda6171351caca639efb
fab191444754dab60da277aab31a1a9b3a3947c3
10203 F20101107_AAAPXL sundaram_s_Page_140.QC.jpg
b6ba93c7c6a4c221c436192bfdbc5240
afcaf3a5ef050d1f7c4d008c5c96aef5e6095848
7114 F20101107_AAAPWW sundaram_s_Page_127thm.jpg
59fe484982f8b036a6bffcc21df4adec
dbcdf4cfee81108a2fefedb485422c37bd617512
F20101107_AAAOUJ sundaram_s_Page_204.tif
995c4a3890ac99fb95429575b525966e
c0900c96278a4f729853b5dc35494ad19732cf8b
F20101107_AAAOTU sundaram_s_Page_189.tif
a851ee8b8dd46cd7e83afa1a7f3d9ae4
f628b45c68855d79c4831049ad641fc5662e7b6c
48794 F20101107_AAAPAD sundaram_s_Page_148.pro
7bc7ee0b9c04a818890b91a77fb24605
c4ec7e86cb6958a9f5dce094a29021a81a40cb15
7059 F20101107_AAAPYA sundaram_s_Page_159thm.jpg
f4ea3344a6c7b818ee536cb58bca1768
2901d47dec8b32d769091d4e755fc8ef907d0274
3066 F20101107_AAAPXM sundaram_s_Page_140thm.jpg
e2de12e6ebb1d90f9d5e89e8ee4e94c5
e37bfe5cea7ab87a057b133710d2a1df78630beb
27426 F20101107_AAAPWX sundaram_s_Page_128.QC.jpg
c08467230d146fdf765cdb3b9d4553e3
2e1149f74059e6d6c5745c23b85b841f469427f8
F20101107_AAAOUK sundaram_s_Page_205.tif
082411c6029976b93b999ccba3efeed7
2735acbf28619b9f32b43ec769b8f4a1d1e20f41
F20101107_AAAOTV sundaram_s_Page_190.tif
1d811fb54e5ab254fb005546ae675d4f
e1ed6614ad98309cc5b0039e390bcfadba70bc87
41615 F20101107_AAAPAE sundaram_s_Page_149.pro
177152dc81a32717ef44fbb13587c09d
fe4e3231a3e95e6d5fe61cb391dc33353b260fde
26152 F20101107_AAAPYB sundaram_s_Page_160.QC.jpg
ec6831888247882d42e2a0429f595185
4b477d8506209f8fabedee3b54d4f3cd55b5d34c
7853 F20101107_AAAPXN sundaram_s_Page_145.QC.jpg
08878aad23fd73f8f1295aba9784e4fa
f2fa14d7b3fd744b0ab506c62d7b83d0157f19e9
7199 F20101107_AAAPWY sundaram_s_Page_128thm.jpg
d91c944e7997e3d0a5223396923810a9
36f20949103758c0cc6639162284c299753f99fb
F20101107_AAAOUL sundaram_s_Page_206.tif
f8ab47e8b982aff7eafda98263f986fc
901a4e0066ac10b5b15202f8c83051f0769d4ab6
F20101107_AAAOTW sundaram_s_Page_191.tif
205abeb0dba0f2704d766f40992d4003
9f3aaff38cc825296e8e5ca9e42084985bd34166
23696 F20101107_AAAPAF sundaram_s_Page_150.pro
87526f88d2cd91cf8fe3196e57a0360d
d80739ef2a6e37d8f269bdbc0d91d2031a52b8b5
7017 F20101107_AAAPYC sundaram_s_Page_160thm.jpg
da52c6de3c7e79790504cc4918d6b98b
42ec26de4c6d4c427f8a4c204cbba65a7ea0b037
2933 F20101107_AAAPXO sundaram_s_Page_147thm.jpg
8e7922b8101f5352bad867a17404d206
dfe6c2f00361a167c5257c0495c84fbe8740ca2f
27160 F20101107_AAAPWZ sundaram_s_Page_129.QC.jpg
bdcd88ed0c512b4528d8ee11cb9eed5b
824d833eaadb19f1e624b983442b0a96591e5336
F20101107_AAAOUM sundaram_s_Page_207.tif
2e009f009e8c3957827663aa7b88716f
bf1ec7ea65f849153fcfbe761684f8c8fa1aa560
F20101107_AAAOTX sundaram_s_Page_192.tif
18e9773e35db61c2bbc57d04194cdbb3
4eed56e6980f7b798826cfb37c20de393153949a
41435 F20101107_AAAPAG sundaram_s_Page_151.pro
fc365a5cd718b6a96b8893e7196cc723
cfc23361a1dc3b971dfbd453d258c116ff254088
51861 F20101107_AAAOVA sundaram_s_Page_014.pro
d82d734809c604b5cdfdc78a061e6e96
a959bce2d8f9be7f00a1a329085bcbb0437ec4d8
F20101107_AAAPYD sundaram_s_Page_162thm.jpg
f98a6b72b547360b705ed56427a0ce12
0fadb0c107f3754dacb5763711de943d0f348753
10167 F20101107_AAAPXP sundaram_s_Page_148.QC.jpg
895b2fa8a998f8e1d233672c276e6afa
e036834d9f8745287067038cea3dd26d32af82c6
8374 F20101107_AAAOUN sundaram_s_Page_001.pro
b359f5029f5223e53aabd160328dd0e8
2092bcaaf0daa2aebb406ceeabd542891b388ea5
F20101107_AAAOTY sundaram_s_Page_193.tif
850d155331adc629a0e9785325589767
b7e9af7d471206927049f062817076879119ec7d
27349 F20101107_AAAPAH sundaram_s_Page_152.pro
672f6271cf059079459ed8cd7d2c7a57
4514449bdeac55b0ce96d938b3aed2799f593554
49931 F20101107_AAAOVB sundaram_s_Page_015.pro
8b185215ae3e66e0320f2b5f9bfa7e83
379cc515a164f4061c4ff3cd39f10f46ee1a597d
5304 F20101107_AAAPYE sundaram_s_Page_163thm.jpg
3177b0a15d48827bd2055c928243472a
8f8d38a678c221a27df8f48323ebca01c921d603
3116 F20101107_AAAPXQ sundaram_s_Page_148thm.jpg
74695c409b0686f9a5acc0f0724c499d
9097cddb9f0d9c2e267b88fdb6878419cedf0ed7
862 F20101107_AAAOUO sundaram_s_Page_002.pro
36f9b2974a1bdcf4a203cabbf80ccd9e
87ffbc237eb5d8208cc9d9e6cc6f62345061f868
F20101107_AAAOTZ sundaram_s_Page_194.tif
ed7e204748afa07fa89e9ea6dc95e2f6
eae7b12b8507cfe4cee7528c8fd86d65913502fd
52037 F20101107_AAAPAI sundaram_s_Page_153.pro
513d153bd623d4f09d857de22dc07dc5
4e11a11d05093a510cbdefb689fbf7f12128bcdd
54712 F20101107_AAAOVC sundaram_s_Page_016.pro
2e0fa353fdaba09676665773485eb7e2
b86e91c00837495c96199d73ec797009c2ae36d2
10165 F20101107_AAAPYF sundaram_s_Page_164.QC.jpg
0b1e844c67649fe58201bc223c62fd8b
656f381ce53ff5a61e9a381ff83f6ca3b3ccd1d5
10247 F20101107_AAAPXR sundaram_s_Page_149.QC.jpg
65b9004993c0bb72670d0fe6a69aff6d
65db6dd28d96dc7d90936f0d3edd5c15310d6aa3
626 F20101107_AAAOUP sundaram_s_Page_003.pro
b756ac9d2dd674a217c14850728eafe6
d8c039939ec2b593cb5f37311e9160667c9cdb0d
30918 F20101107_AAAPAJ sundaram_s_Page_154.pro
29f128fd5bb132b978262e1499716361
cb632594018baf6f52cb4f61b1c8f2627201205c
52212 F20101107_AAAOVD sundaram_s_Page_017.pro
e2554a53864906348c29f6836ffbb7f6
e2c27c995e6d652e1933d4584f76d38f6fc7fb5d
9889 F20101107_AAAPYG sundaram_s_Page_165.QC.jpg
8d9f0adf37f92dbcc4ca84705f91e6ad
dd83ecfa3be3ba7b455e0e1c2055ce8b17a1067a
6676 F20101107_AAAPXS sundaram_s_Page_150.QC.jpg
ed1987688de12ccbb42e7184a26a4df6
4ece0c6944a143ba2a89faefc4ffbda1ebaa32da
55900 F20101107_AAAOUQ sundaram_s_Page_004.pro
a2d27d354309db7e51277b475db8b265
e78e400d711bb39f77207f51113bc9f21d6431fa
42524 F20101107_AAAPAK sundaram_s_Page_155.pro
f16684fbf9aafe0fcb7ef809c18b7d51
4eb0f9a0bfcd675d9edd1f42c5511fb9caf48db2
54006 F20101107_AAAOVE sundaram_s_Page_018.pro
c8a820f635f3a89eeb79a39ed685ed48
10e23278113259be7406af40fee54dc95120f2b5
3238 F20101107_AAAPYH sundaram_s_Page_166thm.jpg
84e796dbd1d59cbbbd49b5069ed43cec
3a4ee9d16a937ad0654c4ca7d46f271ed745ba69
51396 F20101107_AAAPAL sundaram_s_Page_156.pro
33de408a24fd7f9691e88df16006c85a
54143d2ab7b7e7656e1189251ffd9e1a850d76d3
57830 F20101107_AAAOVF sundaram_s_Page_019.pro
8520adac442b9f6add4334701b1f50df
045c8df8973e7188abfaf267b432815017053b7b
9952 F20101107_AAAPYI sundaram_s_Page_168.QC.jpg
a0ce3b91e36e0e2cf9239214cc236c46
04a3b3d4eb63c7f5c9942baca188002b6d977f81
9108 F20101107_AAAPXT sundaram_s_Page_151.QC.jpg
f1e6e4c909bb24993495954e7770cd71
778b6e4f8ede9076521da1a9be9072bb5c53cb74
19863 F20101107_AAAPBA sundaram_s_Page_171.pro
df8ce44515bc924ff173d765bc75052b
5f5dc14bbd9e8617040f31ef01f220575d019702
6506 F20101107_AAAOUR sundaram_s_Page_005.pro
b77cc1a6b8e733967fd6dc9188d07696
1933e4c010379199cad170b4cae0575630edcfd6
46216 F20101107_AAAPAM sundaram_s_Page_157.pro
c473fce3858a34bda1b0aa3c3cc1d31d
99ffd29641d747d9c342997e9461f3bcdfca4068
17761 F20101107_AAAOVG sundaram_s_Page_020.pro
a2ca7ee95bded67dfb093ec26ec14853
5e54b9c754862364c6c4cd46d04ed08c1dc08962
9462 F20101107_AAAPYJ sundaram_s_Page_169.QC.jpg
63f4c0cd5e778e3d61546d8cb8e58a8e
c8db6d6efc8199ae1f611dfd3d93c0b3245095cf
2140 F20101107_AAAPXU sundaram_s_Page_152thm.jpg
12b7dba3a7692632de61447f1b9a5a2a
4697fa0914da889f1f8ecc83d0009cbc6d8602fd
71752 F20101107_AAAPBB sundaram_s_Page_172.pro
607a3224ab5c96658ae69269b50e9e76
2ef6d6c34d0d3a06e03052d637972648e36d4bf7
97750 F20101107_AAAOUS sundaram_s_Page_006.pro
7675417a722e48aad7b987bb51b05fcc
55965297845636511df71908daf78448a9f98555
54954 F20101107_AAAPAN sundaram_s_Page_158.pro
c477eb520cd3fd0672ced736983ab738
8db2a37d30ed0a4eacc77a45165247bacc665dd5
51033 F20101107_AAAOVH sundaram_s_Page_021.pro
893341ce66dd1c279eeef3e203b72cb9
79ead10db840709f87ef9614c62473360687ac78
3124 F20101107_AAAPYK sundaram_s_Page_169thm.jpg
120bf943b662d60b786d45b6952ddaab
68d0df5144fead10eda35a148fda5628d013eb27
2056 F20101107_AAAPXV sundaram_s_Page_154thm.jpg
2533631bfb2d1e41193a765fcbfed1bc
c212eb051dc1bd45e53273d3c23d6df7135dad60
58038 F20101107_AAAPBC sundaram_s_Page_173.pro
ec1ce233ea59a91f9f211dfc40bc1f10
bc9e2cbb38c19bf017ca2426e9d4d55664677006
114100 F20101107_AAAOUT sundaram_s_Page_007.pro
5d86de7aa96b7a5d1b07a1e40c6fbf83
060a9029eae8e3ac43578bffbfc560ae216e974d
52878 F20101107_AAAPAO sundaram_s_Page_159.pro
31be1ced4264011da6827e07502d07a2
e60bd7889a25ab746a3ceee888b6fa1579dff303
55753 F20101107_AAAOVI sundaram_s_Page_022.pro
2d40668dfa662848b12d0a67ed7581d8
135b6f9d3c608049b6a02c536449e895bc28b491
10138 F20101107_AAAPYL sundaram_s_Page_170.QC.jpg
3434a169f7c77de4928ebccc0e7081f0
974d9691d2b7983ada124fc800eeed4c9f5ed160
6523 F20101107_AAAPXW sundaram_s_Page_156thm.jpg
88cf1a4214027f16df571cd175ca3321
faa4b5bab6ea655db58f3fc9e96c882af858e3a6
60946 F20101107_AAAPBD sundaram_s_Page_174.pro
2f21b6fa2097ae7ed30192bdcc9811c6
60224aa151adba69de176031dcfe95b32983006c
58142 F20101107_AAAOUU sundaram_s_Page_008.pro
2bea48f4d9ade61a2935d5627e567744
c3c7b9140b6c0c88b50cc5a28b1759ca1452ce56
53457 F20101107_AAAPAP sundaram_s_Page_160.pro
76d11866af4d18babb8c1f59f54c76b5
13654315cb33d9fa0895c274c90ca72ba80d536f
56307 F20101107_AAAOVJ sundaram_s_Page_023.pro
4bb8f8292070d56096bcaf21891fe9df
20f042d1c6ca378a373d16d4df9bee8a2ef01137
20207 F20101107_AAAPZA sundaram_s_Page_180.QC.jpg
5481ad1832e8a7d6da36d7b7fb9fd48b
cc8d505878e755c28c6c16aa94ed8ac66e93247b
3283 F20101107_AAAPYM sundaram_s_Page_170thm.jpg
123be5f1bc1ea51540bb71354cf42969
4a4c2caa0c9986adcf2653c6b8a4412ce6e0b538
26628 F20101107_AAAPXX sundaram_s_Page_158.QC.jpg
7ef2f0b4c1ea603257779c46ce0db061
be12c6485e6fca4abbaab876743662b9bd8aebd1
63288 F20101107_AAAPBE sundaram_s_Page_175.pro
89019d35b556e5299508a61ffbef0801
1099e8ec1607b40d3e479fb93bf761d128e2382e
78175 F20101107_AAAOUV sundaram_s_Page_009.pro
ef5bc2ab51f06150515b231cdd9ee69e
d8db22bbfe1c40244f446836c1efe694fde6c2d1
52443 F20101107_AAAPAQ sundaram_s_Page_161.pro
02151da0d12e6ad543654b76ca153a84
9ceaf0cbee344f39eb3c93f68d9999cb22c4d717
56627 F20101107_AAAOVK sundaram_s_Page_024.pro
9f7d2908b7670ccfbefb1fc099841991
f165b65aef869de54a8e0a8ce507075f55b53ace
5286 F20101107_AAAPZB sundaram_s_Page_180thm.jpg
fd5e97da68eb973e1e3bcf19788b2efc
f9425d9507a6d6bda2fdca222e5d831f91ea72e8
6070 F20101107_AAAPYN sundaram_s_Page_171.QC.jpg
547491e9ab6ada56f68b6cfd6c556de4
9db0c9b3c9180feeb09628dc21b0ce8991916fc8
7075 F20101107_AAAPXY sundaram_s_Page_158thm.jpg
8940e259014482288ecbb5f74c1633ed
614df52bf6d8b94ca9b47bcea307b75cbd7532db
57641 F20101107_AAAPBF sundaram_s_Page_176.pro
5c3534f7e6a619f48c35a4f3923d458d
7f4aad420c8647e3c6f553b92be3321a589c7d41
77403 F20101107_AAAOUW sundaram_s_Page_010.pro
304ea2f9f4ae1b4de9ed9d55bd167a6f
eb385abd114b89ef87cdaa75506b1d0c9ebc8d25
53223 F20101107_AAAPAR sundaram_s_Page_162.pro
348b8710d05673000d17d2e3aedd3625
0f430b82f46ed03fc861e57ec32cf9847642e12d
53344 F20101107_AAAOVL sundaram_s_Page_025.pro
162fc830ea9fc35be0461a835bebe2af
f6e882c9b015723098551838f04c0cc889c5edc8
22298 F20101107_AAAPZC sundaram_s_Page_181.QC.jpg
f9eeb7d82b3bee9f555d09aa5d1111c7
988b624976bc95b7720266d1ccb91dc69f58c369
2089 F20101107_AAAPYO sundaram_s_Page_171thm.jpg
15b4eb166ec8a1d37b8f653abea9dfaa
11528ec05b31dbc89dfcda510c6228019b78fcbb
26705 F20101107_AAAPXZ sundaram_s_Page_159.QC.jpg
c18c15b86acbe7f5f8d96f2909c319ff
4acfed96eed8e9871d89afd099eb48ba68b38aa5
55971 F20101107_AAAPBG sundaram_s_Page_177.pro
f23be0d3203c2d165482dbe281789e93
1bfa9ea2bb86cc0d69f9e8fa32f394d27f64ccbd
35380 F20101107_AAAPAS sundaram_s_Page_163.pro
70a5a65401e9fcf4e6334f9fc61f639b
77b952d915f2becc7302aa30cca3d41b04f99224
10519 F20101107_AAAOUX sundaram_s_Page_011.pro
2cf2dded31d1b9c4183ce7ac5d0161ef
52b9b5ff4b9454129a343f7a26354395d37abfad
55034 F20101107_AAAOWA sundaram_s_Page_040.pro
132dc569b5b7d218b51d14c0c0339bad
37604a02254abb68141353efed514e9eb7ecc67c
59122 F20101107_AAAOVM sundaram_s_Page_026.pro
78eb0a1d737b1b2d0dba2f333bf11a3c
b2c135a1e9b7e47b952152b7c80ed9c24f4851cd
20180 F20101107_AAAPZD sundaram_s_Page_182.QC.jpg
2b3d3ec876c30dad4dcfb7567b9618dc
69ebdd11ecd9a89f1ec319e838f026735cbba65e
24834 F20101107_AAAPYP sundaram_s_Page_172.QC.jpg
c579ff988f5d8a969f9bf2049a678329
d27f9802f2de1eff79fb399de4c78f9bbd5728bf
51151 F20101107_AAAPBH sundaram_s_Page_178.pro
40636dc6d3e010edd5cdeea7f0db2a7f
199dfb5572bdd7311e1e526d1eed1035d482ae2e
30800 F20101107_AAAPAT sundaram_s_Page_164.pro
101dc0278a50abb7951afac7dd51d6f1
a6f35a2ee7d7926f0ced102012fa808fa3461563
45393 F20101107_AAAOUY sundaram_s_Page_012.pro
a3bea512a7d9b6c90935d76c82d845c7
58c4e901c694c59a9a8935fdf0871f418cf339c8
56658 F20101107_AAAOWB sundaram_s_Page_041.pro
d7c0f58fbb047658cccd76b438937b06
8d112c1f84bc717081608c935698979e35faf157
57269 F20101107_AAAOVN sundaram_s_Page_027.pro
956d6ccb3ac3977391c0993b4cb4dad5
3c27bc5fc0b0b8d881188aff01963c34d8ed587c
22906 F20101107_AAAPZE sundaram_s_Page_183.QC.jpg
9646a3e7dcf4235241032a64977c1107
b3611f65e60112949bbcf0ea094ae84fb31bfa3c
6297 F20101107_AAAPYQ sundaram_s_Page_172thm.jpg
5e4f5e7a50a5331b7ad0381b43cdf2ae
0d40edabc92c15a9d06dc154edac0f15985c0256
55193 F20101107_AAAPBI sundaram_s_Page_179.pro
84b6fb0522f374c99d29b78a8b1e1175
6e56f0d8fa1446bdadfdc161805b15b1ef31753b
30394 F20101107_AAAPAU sundaram_s_Page_165.pro
4ed42d91c17ec8224fdcf4c6bdb07d28
426c4faefd244e5723650adcaf78bfeb3adcd5ba
40350 F20101107_AAAOUZ sundaram_s_Page_013.pro
cb28d4cb58055c9f0d59a7d94748e8da
b717c95b59de6d89ec6f2cedb529b67ad8a5d764
56398 F20101107_AAAOWC sundaram_s_Page_042.pro
a53b47cf49c1cf1f706bd675ee193b32
188503a99b813ae7dd32ca60db82ed71e266bf23
54702 F20101107_AAAOVO sundaram_s_Page_028.pro
2b014b9a3d4d34bafb83b3ef0834c531
01144865613012eacd2c36229829e5d98135effb
19842 F20101107_AAAPZF sundaram_s_Page_184.QC.jpg
deddf75972451e354f53248ae4b1bef2
717e862dfe5ace269547a9625459ce30a6061127
23397 F20101107_AAAPYR sundaram_s_Page_173.QC.jpg
19556eb6ae4620ad4cfb8497cf21c7e2
e107cb4ddd95e42ccc52ed7dcbcb62bb52c99cd6
51307 F20101107_AAAPBJ sundaram_s_Page_180.pro
334c642255e30464c963465625b32cc1
b046cd318dac04223037216e23ce66176e233f44
29229 F20101107_AAAPAV sundaram_s_Page_166.pro
0b47ce6a0d4f8a2c5710397fba1ce150
2166e554f0770b39082c9fa03174ef552180d5ba
57165 F20101107_AAAOWD sundaram_s_Page_043.pro
fa9265056a774090cc3f690e2270f67d
985bc43f8aa2052c1e28be2c1b73aab426a8dc35
53780 F20101107_AAAOVP sundaram_s_Page_029.pro
ef69e44b5da11c40d103c35807fe7269
84b11bd3fd315977a6a33ffae7586e14d7684195
23420 F20101107_AAAPZG sundaram_s_Page_185.QC.jpg
32e25c2169260fb84f5de86b033c0e7b
dc33bea1b8a928a80a3a964c903a9a61182f0a9f
6261 F20101107_AAAPYS sundaram_s_Page_173thm.jpg
0846483d86402e4317b7939ed7ac6fc2
c6f7ab9fc055ef3c038c713a2ba0d72c420a742e
63366 F20101107_AAAPBK sundaram_s_Page_181.pro
dcc08925ac4145c4bd62de40f7d4bd53
be7de87c116e72f5e879c014d4bfe0de84b73338
28455 F20101107_AAAPAW sundaram_s_Page_167.pro
a8a7e384a681b771cd7e8d0dc1206df8
79771f7383de5b84c912544bd1e94cc32e069e9d
56013 F20101107_AAAOWE sundaram_s_Page_044.pro
440b8da3f08d9ee442d43d8aa54de81d
17421b1357e596ad1ed342e2ae6211a29a7d624a
56798 F20101107_AAAOVQ sundaram_s_Page_030.pro
757831a30324fc69d8d3a8ee6ef8f485
49b50d72b687faf3dbc1c77081f4459f989a0158
6103 F20101107_AAAPZH sundaram_s_Page_185thm.jpg
3eeee1b9b1c881aa760c51f8061983d8
0167f87a0dc3493ef1b64be5afe701a869921a56
5650 F20101107_AAAPYT sundaram_s_Page_174thm.jpg
7f7451e2f3392ed48385fc2ebb13fe9e
63998e5ff20f5ff60001276659b8d61627d6245e
51117 F20101107_AAAPBL sundaram_s_Page_182.pro
8c9a078f8deae18a5f9a992330054445
56f38c590586d916db0b8bd11e3282a3fc5ea180
30392 F20101107_AAAPAX sundaram_s_Page_168.pro
bb58189eb14f5346d4dea22c06da9c04
225d44e2da04c77b32f79d8bc37a6202c8edb864
53673 F20101107_AAAOWF sundaram_s_Page_045.pro
d5110458d3ef79ff8af522f0665be00e
2dc2ac880ac0d6f469e7994fa0bb9b7a0febf773
53027 F20101107_AAAOVR sundaram_s_Page_031.pro
df1dc964c80548c021d77bc538a59606
cc6c8b76304a514ec5bc97c597db8c5a633a9721
5490 F20101107_AAAPZI sundaram_s_Page_186thm.jpg
57f6617ed08e6e68b1cdd31d19d91e17
41578dae8c2ff11b96979c703fbd9352c9ecab13
67250 F20101107_AAAPCA sundaram_s_Page_197.pro
1817cf9a08f2552db0add58e6a170fff
f0b1159dd7b85f90e81c57f0d210b664f5837c66
56137 F20101107_AAAPBM sundaram_s_Page_183.pro
02b3c8b369194a1add658f46c291b3d3
229f6905ba15fba73105d5abb3c91fc910fff596
27542 F20101107_AAAPAY sundaram_s_Page_169.pro
4b7f92f6f5f5a339d6bf4c8b333bf627
ea95ebd1e2f638320fd34fa810fc136980bb10be
56869 F20101107_AAAOWG sundaram_s_Page_046.pro
3e39d1088cb59f700b423f171ff4cba4
f5e2643cd5da236d1ce4415488c44b5e6b733248
7116 F20101107_AAAPZJ sundaram_s_Page_188thm.jpg
e1ef8e086c4044683b45334cc408796e
7915742e5e1c82b1ea6e4129f5e293ed51b60f7a
22355 F20101107_AAAPYU sundaram_s_Page_175.QC.jpg
70de78a619ae43e1fc4746682c840617
c0cb6e9d75718df43742a5fc79d6f2068ac0e16f
69991 F20101107_AAAPCB sundaram_s_Page_198.pro
fe2a59793a2cc6d8dcc84d08a5bc7395
a4e778e44228223452036cb0382cc1f658c1a3f2
50407 F20101107_AAAPBN sundaram_s_Page_184.pro
47078f21fd6ce686800c3710fbcf1ff8
116ce6b444f4483907e2f31f887f21f326275aeb
39519 F20101107_AAAPAZ sundaram_s_Page_170.pro
ac9a95eabeecf4b1f10c621d870f94bd
77a76359c69c6c4fb2b85b6a34cf9da2a36479a8
52384 F20101107_AAAOWH sundaram_s_Page_047.pro
d61b79e12db52220c0d1f3220d139679
fad2c517049d93e413f6724f78af2c5ba73b5b3b
58078 F20101107_AAAOVS sundaram_s_Page_032.pro
65ddfd5582a5ae5879a9254e5e9b3c67
01eb8c85db0d59566ca8790de1a238a462d73feb
27724 F20101107_AAAPZK sundaram_s_Page_189.QC.jpg
ab8a62b9111e75d361889c9bdd688508
2787201d89b1c017a9906f1237e2a85507cbf54e
5984 F20101107_AAAPYV sundaram_s_Page_175thm.jpg
f7a939f63775b3a8e5ef620e12cef247
6d120b75c37fe9945f31736b94ce1b9190f5014b
68833 F20101107_AAAPCC sundaram_s_Page_199.pro
a3864fd6811934a936605cc146a741d5
7479a16fc7fda5a6030ae3ccbfd25192c6fa8b7b
57853 F20101107_AAAPBO sundaram_s_Page_185.pro
d276301a9eb65a7dbd7fd17dd9c0c5de
77e191b4d798948adc84b26a4e7d90b06922522d
52469 F20101107_AAAOWI sundaram_s_Page_048.pro
341bcf932d60b255e860e9d9d1d8cdd8
9a69bdb55a4c2d2c5a8ba4cbccd4b96fc8403a77
50819 F20101107_AAAOVT sundaram_s_Page_033.pro
f2762e54e2d69c91105e38ffc382c826
72429d3ff6c68320163b60bbdada38149afd415e
24476 F20101107_AAAPZL sundaram_s_Page_190.QC.jpg
31f2b1386577942766819c5c9c9d90af
66b4fbe4448c36b3c009722c10db82e2da860e68
22847 F20101107_AAAPYW sundaram_s_Page_177.QC.jpg
72a414e983d6f0cdb676220094576305
741a9dc0e6e72be6477dd615bb1e3afbc7c13560
64390 F20101107_AAAPCD sundaram_s_Page_200.pro
137133f43431349df6d4b4180e7790b1
e0850e8d3ab7c4e96440d74617e3658289f668b0
57435 F20101107_AAAPBP sundaram_s_Page_186.pro
e4ce550874ff32ae9258a7fe218539ef
76cb2e7b047394f48f2a45db5728cb80244e7025
19620 F20101107_AAAOWJ sundaram_s_Page_049.pro
d6b58b69e31a6c15e97010d687d781ff
470f0b5edc117b42467576521cf9256c762e6495
55130 F20101107_AAAOVU sundaram_s_Page_034.pro
71be3aa218dfb715f6b1367f3382c27b
18824748f68375964bfe8a7075dfdfcca7c9ecb9
29152 F20101107_AAAPZM sundaram_s_Page_191.QC.jpg
0afedfad8a41bf7966ce1a92d50aa168
855c13683457d587620d71a35bc430b9e756972c
6000 F20101107_AAAPYX sundaram_s_Page_177thm.jpg
f94222e3a393b9abca605d3c1e20e304
6760efba2206ed68b5d269c63d886d180fe02d66
65865 F20101107_AAAPCE sundaram_s_Page_201.pro
e4f1f553d5b6a8b2bc9f57f5bb52da98
2e839f76be5861790a16e620f236a2e633276f9d
65568 F20101107_AAAPBQ sundaram_s_Page_187.pro
d5989c8d04b2fb6b99d2eb6665924a5f
b2ce4c310259593a2100f2faca016abf1e28f49d
4919 F20101107_AAAOWK sundaram_s_Page_050.pro
b8d84a21009b330e6b009e2da3341a5d
14b4d4c4ebc8356e82e594da7184d9b7fcaafc58
57069 F20101107_AAAOVV sundaram_s_Page_035.pro
7121337bd7924bace367a59314d9a816
bc18c32279654437b6f463a23854ec7357c57980
7863 F20101107_AAAPZN sundaram_s_Page_191thm.jpg
b573c1ab9b571f6e79b42a7a295cbfd8
bcb7a9b3fe241c0a129be4f4cd5c9f76dead442a
5222 F20101107_AAAPYY sundaram_s_Page_178thm.jpg
40ae6354c3e461c4df02c185735d513d
b345b20c430668bbc2c55f1b1df83ff0d417e6f8
66340 F20101107_AAAPCF sundaram_s_Page_202.pro
b9fd1c7e583243d972f96a06c7ef13cd
8bba3c979e5f3b6a21e89b3c82c298290b8b8267
62536 F20101107_AAAPBR sundaram_s_Page_188.pro
27d4c1773ae5888a3773890286c888bc
a9958a171b410bcc128aae4cce7979faa39a2776
48910 F20101107_AAAOWL sundaram_s_Page_051.pro
9b73f3339bf6ac8671e6c6fa569ab274
eca2da35a374ecf5322a723cf17ebe3051efb080
55253 F20101107_AAAOVW sundaram_s_Page_036.pro
6e593fa665e7217b5348662ce3ccc25c
b85636e4e56319c4b6c911af251949eb5170d5a7
25513 F20101107_AAAPZO sundaram_s_Page_192.QC.jpg
53359825ef5b634da1ecc26921fdc37f
d43c6ed13d48eb6396bc83fc5bd6b177c19ddce5
6081 F20101107_AAAPYZ sundaram_s_Page_179thm.jpg
a3bf9c344bc933e0ec3ed1a5957dab0f
e00b919bb12527cfbb622dc9b2a0fb875e9ff412
59399 F20101107_AAAPCG sundaram_s_Page_203.pro
12fe6aab39003eff711bf2120749a8e4
34679e50b5a77bb9bc731acbf5e53980a66cc594
64657 F20101107_AAAPBS sundaram_s_Page_189.pro
f7c24ea58e9328fef8450b197033f47d
2427b3b74d32c678b6ecc01388eb8461245c71f1
57250 F20101107_AAAOXA sundaram_s_Page_066.pro
d30c90d0a0bc2d43d851eb0074a5e1b9
67c773dded7f2f91291e6b7ed71ec38e9acace1f
59840 F20101107_AAAOWM sundaram_s_Page_052.pro
b3d102441082d4a1387dc635604b02ab
96fec326d595642a9e7a899c7c805cbcf0b641ec
54118 F20101107_AAAOVX sundaram_s_Page_037.pro
da21b11451567b3f09a7607ee98db5e8
5f771af102b7b21d53eeb4c67795dbe35389a55d
29288 F20101107_AAAPZP sundaram_s_Page_193.QC.jpg
e4c7a7c761f6f72722c8b90075f71eeb
59b4f33ca60c221f31957e68e53d991c41c42ab1
64148 F20101107_AAAPCH sundaram_s_Page_204.pro
72e19d33c35c4af806acfefc2ba3508f
fb901b2badffec222393159a38be8a85deeabdb0
57288 F20101107_AAAPBT sundaram_s_Page_190.pro
86d88df476be283f8d9903bfcdfcec49
98ab0ca734880039ae37d608291c282b8ef96dc9
25139 F20101107_AAAOXB sundaram_s_Page_067.pro
dc96250b069bd63776796265b3d76a1b
f720d532fd636b3ad0d3acc3eed2bfd7e34093e8
54406 F20101107_AAAOWN sundaram_s_Page_053.pro
f7dd3d5c1912be1df04084c37813a0da
788bf6a8d0fba84f2986993087a404b1cbbd4e7a
54977 F20101107_AAAOVY sundaram_s_Page_038.pro
9f308473489938fa9bb221f6ea8b52aa
207bda2a56b477d5a76ad8d5e7c16640459493c0
7606 F20101107_AAAPZQ sundaram_s_Page_193thm.jpg
d14f3593e46843a3a62364a1cd5cbf82
6ac5e43c7c82c6485c60a884a6f2f3dcbb2995f1
66450 F20101107_AAAPCI sundaram_s_Page_205.pro
7bc382c7d633aee7c5a16fd69b6c66c2
1c6bc8bdbb81659241ea5946b0c88c2265c69ab0
68054 F20101107_AAAPBU sundaram_s_Page_191.pro
49c6a671bcab1e256fd01eb958fcd786
b452e04629c41308c8543c1539b64ca368406404
49818 F20101107_AAAOXC sundaram_s_Page_068.pro
b6d22c4c4f0ce05927757b19a136404b
33b42522f542c2e8ee8ca851067872c217b46b18
39475 F20101107_AAAOWO sundaram_s_Page_054.pro
1492d0d110602e1bfd3b5feab08a51fa
ffbb7402c85fb5bd4ed191c5f2119823183ef66d
56961 F20101107_AAAOVZ sundaram_s_Page_039.pro
8f03cc17624594740fa005c4b019468d
006baa608df72d80870e7a60e77ee916e820a71e
7295 F20101107_AAAPZR sundaram_s_Page_194thm.jpg
6e6d50af738b7ead438d0319adc46936
6ab3c6443f552ddd9bfec94f074689ceed5f0188
44231 F20101107_AAAPCJ sundaram_s_Page_206.pro
efa7cd45e1cf582cebfc9da97bea559f
8b416af996b79543288d8fb88a6500980760d9b9
61109 F20101107_AAAPBV sundaram_s_Page_192.pro
7a0ac2662615a36aaccbde998e458705
dc7bef2e97e033420f106b35a4f2b645b6af6d59
53833 F20101107_AAAOXD sundaram_s_Page_069.pro
dd6ec32a474fab01f0804905a1070f32
648db5a332e07273a2b8cde4b121970f0b44046a
49196 F20101107_AAAOWP sundaram_s_Page_055.pro
3c02e86fc63c5e381dc635360561d699
5f64a946bf8d388c8a5e943742641d35dafd2574
28894 F20101107_AAAPZS sundaram_s_Page_195.QC.jpg
d518eb1b50d565ea164bfaaf6885dbc9
deeabd9bee9c0c5069330aaf7da183ca7fe8701b
29755 F20101107_AAAPCK sundaram_s_Page_207.pro
1f078958e7579342b2dd499f75d1f7b1
4e99e923632e1cdb48c2ab61e4e2b1986ec30087
67953 F20101107_AAAPBW sundaram_s_Page_193.pro
9cf1d1c20e13e2a485eb5472df149874
14dfa3a3a9f11c2e62c76802411859bbc95f9103
55949 F20101107_AAAOXE sundaram_s_Page_070.pro
63bb5c9a3abfd4b58ba833b3441f3753
f371c40bb3cfdc4911d088c384d6269c34dd0b87
48640 F20101107_AAAOWQ sundaram_s_Page_056.pro
594347c8612aa45614457a492b67a0b9
3dc48fc4affba1800055c2cf657e4111981cfb04
7329 F20101107_AAAPZT sundaram_s_Page_195thm.jpg
57fe3b42143175ad668d1fbd501995d3
d8425c9e119cbf528a433d80b53afdfe8c8666af
491 F20101107_AAAPCL sundaram_s_Page_001.txt
ca824e0759953b4b833ab2915359c418
c2ebebee5a9039c0031f34f730f371bcba4939b4
68649 F20101107_AAAPBX sundaram_s_Page_194.pro
3160269eac5ac473f0a2335aad4e245e
5069ef0068f44a1e60cfa514962849c8429f6ce2
56347 F20101107_AAAOXF sundaram_s_Page_071.pro
4bed11ab042d6593db9ae891b268feee
02676374f6eb520247b146d48f9687ae8a8fb0d6
53890 F20101107_AAAOWR sundaram_s_Page_057.pro
ec7bd8fb978bc1ce5310c6e220a47e44
074aca5290b4358cee625ce5e8d39a90fcccb07e
28383 F20101107_AAAPZU sundaram_s_Page_196.QC.jpg
5f0bf1eb2c0266355729ca35dc6f25ef
9ba4727b4b7765f1640bfc3329a1dd8566e4b35a
2169 F20101107_AAAPDA sundaram_s_Page_016.txt
d37bf69d4a583052e0cd252f4cf5adee
7fc8fafc43a23dfd4edfabcc0bd1f343f21fbb12
93 F20101107_AAAPCM sundaram_s_Page_002.txt
a664f572d58222610ea84b70a97615ee
abc2991becbb62a7965501b668cb4d3866213974
63101 F20101107_AAAPBY sundaram_s_Page_195.pro
920c42bc6ef99b11f61f426a1cc93d7e
aad215c31d4c968669a56ba24cbb4028f23abc5f
53057 F20101107_AAAOXG sundaram_s_Page_072.pro
e62e1dfe74a7a35c65a98fe3f7234045
41b0854f2a7f7f0a94777fe51f51152cb252267c
73172 F20101107_AAAOWS sundaram_s_Page_058.pro
10f59643597e0082fd607553982bc2d6
4526b2cd8cb64a7b492c6d147e048dde7ac967e8
2066 F20101107_AAAPDB sundaram_s_Page_017.txt
4cd04e7a9d26539463ed1aba78ad7b7f
69415f042f7781a649866e0ba36ed184495e56ce
84 F20101107_AAAPCN sundaram_s_Page_003.txt
fba33857be850e80f7aebc8b8e5575f6
5abd402609a604f60203ba06da914b07b33cfbe7
72780 F20101107_AAAPBZ sundaram_s_Page_196.pro
d269309711b8abd35469dd4b198eb982
29d22da5833535c2e8a7bff5957f4059e2a39d92
52699 F20101107_AAAOXH sundaram_s_Page_073.pro
2586ba03071a34fad3fd86365878b921
4506c251eb5dc9a4f34516d9298b1304dcda786c
27852 F20101107_AAAPZV sundaram_s_Page_197.QC.jpg
c9ea05ca58dd31bdf27d2cc7a50c31a8
747edd4429fb7f1aa79425aa80abb898b216d0fa
2185 F20101107_AAAPDC sundaram_s_Page_018.txt
144d0a4ab714afd03e693538d6674001
2e13566e62aa8eb1d4ddaecc61990a0ede648d5d
2219 F20101107_AAAPCO sundaram_s_Page_004.txt
f1212f6c65b386f7ff71e65be25d9d39
610cf2c3b3f16a6bf6bf73bc851cd821ac81ad88
52786 F20101107_AAAOXI sundaram_s_Page_074.pro
d42c0bf939276e23cb1fd10460a768f8
72a0c7f59445dc64b95246f9ff4072cba8411c18
44779 F20101107_AAAOWT sundaram_s_Page_059.pro
8ee1621ebe082b67b9bd469527976f82
7dcbaa6a62c8476b91ef367b9300edd14d93c51d
42230 F20101107_AAAOAA sundaram_s_Page_088.jpg
53c56f68d916fd74f3f93330e143e88b
e56f82c3bcbd302c9bf944cc123bef1df47c6899
7369 F20101107_AAAPZW sundaram_s_Page_198thm.jpg
10a8947781f35e9cdc7f832f8ee995ee
1d8b8c12fbc903de8daad4ac1b09c6b4253b2776
2299 F20101107_AAAPDD sundaram_s_Page_019.txt
19dbf90e2ebf6641b0ab1e189eba271a
22bce58e4dc4e99a10fa772e5100b86e8d528a41
261 F20101107_AAAPCP sundaram_s_Page_005.txt
1c499053892c523bfc063f9825b35b07
5411a2cab0fad1575731ab940d35fbb7107e06d6
52693 F20101107_AAAOXJ sundaram_s_Page_075.pro
352d2c13624b9e91c99e3ae0881dce2a
b216ec94942f7994c27966a084f293e4c0eae360
52293 F20101107_AAAOWU sundaram_s_Page_060.pro
f33d53edf80287aabbf0d23afd3bbc26
c0996b6dc17f85bf9caade757d71a1f0f3758bd5
42364 F20101107_AAAOAB sundaram_s_Page_089.jpg
b8f3176ae3da54d2b8970033b8fc4d3b
21eddc7034962cfee0e8d692a609244b6a21e32f
30117 F20101107_AAAPZX sundaram_s_Page_199.QC.jpg
4cbef545846a8d7bbadfe8a90916f82e
bccd2279241c4750ae30c02cc5579cce0bd89720
711 F20101107_AAAPDE sundaram_s_Page_020.txt
fd86da9999057144ef70d5c39a8b54f4
bcf2107954006107439f99b4ecdfaf016a039bad
3939 F20101107_AAAPCQ sundaram_s_Page_006.txt
faa1ba0c1459c65ba4736afdf2788740
61b4b4a75b53497cfa07b1871fa5ecb75be82d89
50119 F20101107_AAAOXK sundaram_s_Page_076.pro
6ba0cdb007808e284f5bca60382782cf
dbc82457c4199f27c2a0b2a0f7ee6dfd422006b1
50355 F20101107_AAAOWV sundaram_s_Page_061.pro
306e952a216b07f68c98388c3c70bd8d
ff4706b2d3025180506f8cef7bbc61e243f08be2
40340 F20101107_AAAOAC sundaram_s_Page_090.jpg
0841cf8e73fabed7c3d61ece55c0722e
5b35380503ad2c6f9ebb3b13ee046f5f02042aa1
7870 F20101107_AAAPZY sundaram_s_Page_199thm.jpg
cce4e61692346d083d5ad370e21b929c
6b1a7bf3f94035a5db47e6e91aaa12fe2ae8b281
2141 F20101107_AAAPDF sundaram_s_Page_021.txt
94434c9d022d56708b4f96d342282ee0
72c38b947c8541981171615019f497d13e0d7d6f
4583 F20101107_AAAPCR sundaram_s_Page_007.txt
9c5d9d957af448a3d965ac537149b587
e43cb960bcc7bd993da25fa21dc665fc32231b21
49591 F20101107_AAAOXL sundaram_s_Page_077.pro
226c138b7aa958ad00c1aab3e9f4b1bd
1605f64220930570e6a9b08cf3fe7b6a2903c73b
49396 F20101107_AAAOWW sundaram_s_Page_062.pro
8e1e2531ad5aed3a12a6cdd150febd8a
98b230401681b23c15f181c08691be3947ada21c
39079 F20101107_AAAOAD sundaram_s_Page_091.jpg
100ce0d579016b96fe625a23e051c420
94cbfb8689f8f6e516e7aba7fdc31f2886fe1ea2
F20101107_AAAPZZ sundaram_s_Page_200thm.jpg
ebc8f362ce9cbf195dc925d13019003c
cae82a3e1573a51b0d1bf22e9d123a01aa3d81b4
2211 F20101107_AAAPDG sundaram_s_Page_022.txt
c928aac2108aa181dbb51cfb258d565d
68389c4f74cfed06d56104ecdcd3ebe3f6ac508c
2376 F20101107_AAAPCS sundaram_s_Page_008.txt
cfc356a2588383d3ab88421fb53af9b0
5e3dbed344abf004c0eed1363fc7108f843c1eea
37077 F20101107_AAAOXM sundaram_s_Page_078.pro
1cf34dc57590976f11f47ccd78e591db
87e5336202b78308ac4dc7d653ce0b2689baa6ab
45211 F20101107_AAAOWX sundaram_s_Page_063.pro
e01edcdd6b188791d10cdfe0e1e2fc80
323b1d3e23dc2ff13c054891c739ad5be0177427
35214 F20101107_AAAOAE sundaram_s_Page_092.jpg
96c009c0156d1e03579777341a10a3e3
a32e0dd195f9068f5db33b7ebfbe21abdb57d792
42923 F20101107_AAAOYA sundaram_s_Page_092.pro
96751a85dc4c9b2c513fb795f9199717
cea448d0d41d36d85041408b3ccf9869d3c27384
2225 F20101107_AAAPDH sundaram_s_Page_023.txt
7d28205e9f8c5e0c48e9019a220926d9
5420b4cd6c637afe066498150658b2b46c1620fa
3155 F20101107_AAAPCT sundaram_s_Page_009.txt
8795c587154aab65ad6291ddc3066e6e
70be916e2db102f07fcc487c49573980e86b6352
37546 F20101107_AAAOXN sundaram_s_Page_079.pro
c62eddc0b9964ad2803b97178bdf9d98
9976971662083176e8bbd7a8e6df485d86dc7c98
41528 F20101107_AAAOWY sundaram_s_Page_064.pro
3decaef138fb30a543e88c4139f44b3c
2e82716a0ce37bdba099133627b93146d4257990
41314 F20101107_AAAOAF sundaram_s_Page_093.jpg
3f8ff0efb27b420422ed177c8bb4295a
566e2c780ef88a723ca67f2befcc7b60a01cd1c7
50113 F20101107_AAAOYB sundaram_s_Page_093.pro
3d6c3360a5c119bfb53f2172aa68c0ad
2aceb0f26805c8233737abd8dd8ee6bf60b48456
2253 F20101107_AAAPDI sundaram_s_Page_024.txt
d5e1604cd08c674cbba57b6cd77f18c4
0ea506dab2ec5c203d08b2c7d87033b5bad7a649
3098 F20101107_AAAPCU sundaram_s_Page_010.txt
7b9e997bb6f787f61182e1dbb6aa5ffc
9dfe581296176e5750502d1a8969c9a571279002
31330 F20101107_AAAOXO sundaram_s_Page_080.pro
0e0609b01e181b7abe10da3e44ff6f0b
b433e4e8a428f690558b74e379b29043bdd59acf
29289 F20101107_AAAOWZ sundaram_s_Page_065.pro
b55d9e006dbe64264f0ffb962c6ab06c
c5750e308d215c9887629b1238313a25c0d01487
39240 F20101107_AAAOAG sundaram_s_Page_094.jpg
17015411af2d587a3bab539c5d74f093
5e60581eb67efd2cc9100b933fc4bd1571ff7f75
46280 F20101107_AAAOYC sundaram_s_Page_094.pro
386a6865dcb568774518b9467cf12cfb
5fd793b172e76a919be81ad8fda9311d5d785bc7
2106 F20101107_AAAPDJ sundaram_s_Page_025.txt
9cb042cdc54fefb986eb1d4b1ecf4e46
9bc9e9b07398273cf011807f648dee76a377fd06
459 F20101107_AAAPCV sundaram_s_Page_011.txt
2bfbb9e666f22bcf4c19cc82dc7efb3e
45a82c866c8c7f649e30104ac49c3c26caf11d1c
29124 F20101107_AAAOXP sundaram_s_Page_081.pro
0f7890ddb2ab9ecdbe0f376b7a7541c8
e7300f92335127df25518a1659cbdd8fb3ae3268
42085 F20101107_AAAOAH sundaram_s_Page_095.jpg
308ca5661871812cd571d3de284c0086
527ce85e044c66b93377353f50dcc07b077095e1
66350 F20101107_AAAOYD sundaram_s_Page_095.pro
ad00ba4603681c296af45c652865936e
688c9c4d6a1377c5725291b5765b3dbd829cdfbe
2318 F20101107_AAAPDK sundaram_s_Page_026.txt
c18ed0e1689eca6db80fa55ab7126454
b3acf69d6ef9de71497ec1280ca6f24ceda76bcb
1980 F20101107_AAAPCW sundaram_s_Page_012.txt
b9773934d481a91e99b64ffc3f11bcd6
b6e919ff7d888cbc6fc9bb913ae8c4c7e691482b
59948 F20101107_AAAOXQ sundaram_s_Page_082.pro
0da95e3d3d723e2f679a49a757c38dff
f4ad602fc7e4c8cb1cdfe1b455a198eb1a38c447
39634 F20101107_AAAOAI sundaram_s_Page_096.jpg
9b38fb2096b656ef0112946bcb2e6b16
f86270fee05a02d1b94ba1c16c90a3267be5bd69
60860 F20101107_AAAOYE sundaram_s_Page_096.pro
1dd1d2f98acfc28b1f2eb3ac0f51a23f
c50fe991676d3699ef388c70f65235640182e189
2275 F20101107_AAAPDL sundaram_s_Page_027.txt
f2de9b68ab78f486868debe1b96ff604
28d8e54b32c06a93858641961f4e6899d3b73b63
1605 F20101107_AAAPCX sundaram_s_Page_013.txt
bf354e1ec49a47e906bc2ef05242dbb9
3808868d4c5d5c273ea7a653d8d8841ab0edcc43
35734 F20101107_AAAOXR sundaram_s_Page_083.pro
9259eb8b8f6b1b5c87f7731344ca4d84
c22b9240ec4ae797a32e2a077ebd1c557d20cd6d
39734 F20101107_AAAOAJ sundaram_s_Page_097.jpg
7172acf2fffe08c623b4abbf284ed30e
8c4ff9d88cca873496ed2d6c3c5c38e6a3399e91
59935 F20101107_AAAOYF sundaram_s_Page_097.pro
8cc3ad57b34ebbb049ba8c64c6e2d1ce
64bccc9b6fb5c0b93fcb390b90e81ff14f277d90
2207 F20101107_AAAPEA sundaram_s_Page_042.txt
c96e9a82a268d79460fc807bfe1933cf
0054394d025ac41ecae6fa914a1d7b24a9b2ca05
2147 F20101107_AAAPDM sundaram_s_Page_028.txt
42718916c7c641a3f428a483b0366a2b
15aba387b847b23448c77798cb060ef8a68e8661
2127 F20101107_AAAPCY sundaram_s_Page_014.txt
88b8455959c83382d4a8ad4416a1270e
24717b8c626e2c9750d4e9988fd71039249e5e30
44805 F20101107_AAAOAK sundaram_s_Page_098.jpg
b9b07bccd3ede19e8a3e84142a3d9d62
327496a950d43f9a093bab08028d09ab37362603
66430 F20101107_AAAOYG sundaram_s_Page_098.pro
6bd4e3322f368c9e45a676980be96294
1936780438b40067ecce6bb6b2c2e17a87bd9b3a
40749 F20101107_AAAOXS sundaram_s_Page_084.pro
f757ad89062eb27f4365a7c68165a56f
80fb4a63fa38b22de9489a5e9f51ffbf04295c1a
2230 F20101107_AAAPEB sundaram_s_Page_043.txt
5053e74cfcbcd9ca59cf51a3ebc9309e
04e915971e6775e809efdb610a7996701fc2eea8
2143 F20101107_AAAPDN sundaram_s_Page_029.txt
c6194ae83babf860e585666b0ea850d1
bd93d1ef939e7f5f6165c534fc598d0bb9a5870c
2000 F20101107_AAAPCZ sundaram_s_Page_015.txt
6656f9b92b8211a307206073341ee2c7
847020cc4b8602a31e7a118088d8feb7021c27e2
78758 F20101107_AAAOAL sundaram_s_Page_099.jpg
e73ba7fd8d4670ada27ce043200f58b8
393a3d1f90a905aa8b92bc232bbb73bab64741b5
58275 F20101107_AAAOYH sundaram_s_Page_099.pro
d2e5cf252476d9d2b232edb560225d84
2fe075bf43d005fc2d0b582de05d696cfa156e08
36752 F20101107_AAAOXT sundaram_s_Page_085.pro
1ca28081af04a4fd2d9840afb288d281
bc0bc38b53e2c37ddbf1d94033b0f471ceab66b3
2188 F20101107_AAAPEC sundaram_s_Page_044.txt
9e90d1c027ef68219cfd6c6a647fdae8
94449df1372d765668b32b9f24a2c07ff835ea6b
2251 F20101107_AAAPDO sundaram_s_Page_030.txt
044a41ee368d8fd96d1a7202e481b5b3
03b7f6da31b5a25042a8ee9b98105ae69941f0a1
76303 F20101107_AAAOBA sundaram_s_Page_114.jpg
91a97d5a0eebab98770cf7aa6159eead
2a41e73a964fd130e0a5d78ac3eb2e5dde8fd0ed
82252 F20101107_AAAOAM sundaram_s_Page_100.jpg
4e5eedac2d5485613e2b289618533970
7e0310bf421256a83d1c21d52008359bc88e16e5
59539 F20101107_AAAOYI sundaram_s_Page_100.pro
6e884021fd3f5488a1e236f8a27058b3
9855c7b480afb815443d0313a85dfb70f467d0ac
2163 F20101107_AAAPED sundaram_s_Page_045.txt
8149ca11dbff785aa70dcf342d90a7cb
fa3a5d570b241222c7b7060a67d86c939daa09ac
2081 F20101107_AAAPDP sundaram_s_Page_031.txt
9803a03b6bcd6a184a204d2efc120688
17bdfc3d8a42fd5c1204c161d18a0364d4234b65
77628 F20101107_AAAOBB sundaram_s_Page_115.jpg
0f17c42c17c768336505b5102e500684
253769219011ecd4c4e651147fb17edfa1a3520b
81362 F20101107_AAAOAN sundaram_s_Page_101.jpg
f499d1ed2a6d9d1ce9d0b488675f2382
238ec90b55b6f668dff0c5ef5992159a56b422b7
57050 F20101107_AAAOYJ sundaram_s_Page_101.pro
4a44129743062377b7b1fe41c01a4c48
49eb2183c5fb80b9d7b3aa9596163e50c4d0dce8
37283 F20101107_AAAOXU sundaram_s_Page_086.pro
e5331f8221d27ff38405e87dc3c72b7b
3d082ec0d51c3c6bc6b92488097bcdcadb6334a9
2242 F20101107_AAAPEE sundaram_s_Page_046.txt
4a50ca3522b70f7cd2f95b00b95a441e
5e6568c8cf04870bcccd99cb0ed546a5bd660312
2279 F20101107_AAAPDQ sundaram_s_Page_032.txt
ff7076cf031dcab82d6394c0276de665
6a63c5d1399d7d17c0c006db32b8e4c14d76de60
77595 F20101107_AAAOBC sundaram_s_Page_116.jpg
a56197c76a850878288031f896d98d84
e15e617c1126357a7757df7d22bd9c5978e5c63e
75738 F20101107_AAAOAO sundaram_s_Page_102.jpg
b8a7936c74a67225836eb8d044d496a0
077978f1d62a609f1f0e0c3f01f85f6c712d4f47
54338 F20101107_AAAOYK sundaram_s_Page_102.pro
7f194475ff480b4e9d5a5d379b070ef7
d7c5fba8d773f9b9da99fe24f29d70988d64b9fb
33836 F20101107_AAAOXV sundaram_s_Page_087.pro
5cdb45b53d943451ad5107e8d993f0dc
946f19b8b209ef2e74a7375f2d11a30e83d1ee79
2093 F20101107_AAAPEF sundaram_s_Page_047.txt
11a0886443c6652004804cfbdf7bdc5a
1d9babc5fdb592074187ac143c60ea1c3556cc35
F20101107_AAAPDR sundaram_s_Page_033.txt
270822b73d432bc91f875444140e3035
d76c17a3ed21c5a50efa5b790ad3fe9b4d3bd473
75385 F20101107_AAAOBD sundaram_s_Page_117.jpg
7ebd2de7864a614e6acc165bd6090f4d
112af36e554b182fc2b3302e7059e7f8608eb898
92015 F20101107_AAAOAP sundaram_s_Page_103.jpg
c39d69d34b8fff7bbf5be5e0c173a627
85751a270e6354985487a29207b7cce1c792782a
68310 F20101107_AAAOYL sundaram_s_Page_103.pro
8e84cc3b106592c4bf685eac18cef6d6
9f042c4f2402aaa2881bad558eb48bf60065f710
49551 F20101107_AAAOXW sundaram_s_Page_088.pro
501ae9e32b982c9d60ac609350cf77e9
155e878436e811a41243c33e221c70181f9af964
2071 F20101107_AAAPEG sundaram_s_Page_048.txt
bccb04fbb2d1f760a923e79434af1226
c32e7d1d1549278a2c5973c34e9061aa5db566d1
2194 F20101107_AAAPDS sundaram_s_Page_034.txt
9d68a4285aa15cda3eac00db908d9105
986a658e2b9747b28313e20191dcc0502aafdf38
78554 F20101107_AAAOAQ sundaram_s_Page_104.jpg
3d1a33f42889e12b038b5f7bc87fcfb0
0506a4194d5fcac312b24fa578e8838f6f6fe0df
54016 F20101107_AAAOZA sundaram_s_Page_118.pro
1bdf22ee1b52876b2b80d628adc9c16e
c3cdf2556ad9b5bd228156fdeb78288d0c3dd171
55535 F20101107_AAAOYM sundaram_s_Page_104.pro
4d828ff06a27d83276f98e9a4f43d151
06b0985615c220c193e188e6095f82c2480d458f
52835 F20101107_AAAOXX sundaram_s_Page_089.pro
19b7bb0c41d62d903af55457fca1257c
35f7d53894a627b29c4bbec571cef0705bc8004e
79054 F20101107_AAAOBE sundaram_s_Page_118.jpg
6a975983bbb78332f50a8d46a18bd243
75e4ec4e94535fe61d5bc8f040485ea471b0bb7b
803 F20101107_AAAPEH sundaram_s_Page_049.txt
e2170a1044ae8405964d64b0f28cfaf5
6dba060e138e11d33c2df9389fc2ad56932650e3
2281 F20101107_AAAPDT sundaram_s_Page_035.txt
056a05d3ad9a5aaa16d0a7ec6a49e799
6bc544c75f7ee196c18cb669ca023438d2d8e273
81693 F20101107_AAAOAR sundaram_s_Page_105.jpg
e46bb36522bb50a06782f04d099a2b2b
666c9f57c6e384f26bccee0913c03a4f1930b7c9
55821 F20101107_AAAOZB sundaram_s_Page_119.pro
6e345df725c28d78d9ad7109f9d4212b
d79d8e85fa56bd9e004b40f23f271aacfb1db63d
59713 F20101107_AAAOYN sundaram_s_Page_105.pro
dbb753978f019b1d1e5cb70fff9fc07a
35056909c2e8b91cc11dc27aa3569badb1a392a1
49786 F20101107_AAAOXY sundaram_s_Page_090.pro
34b85094c0373c5ba3810b53d1409f01
b8b94fc6fc36d4058369382195c0a655c001af3e
80054 F20101107_AAAOBF sundaram_s_Page_119.jpg
d13c348fa74910d32629fbbd1d4d2c01
8d219a70dd1811e49123c65a414f148bce7b4944
282 F20101107_AAAPEI sundaram_s_Page_050.txt
0b762c1d8021b96bb4fcaf78aa143ccf
90ec95111adb55a36d365fe5b86f816fc32f244a
2162 F20101107_AAAPDU sundaram_s_Page_036.txt
be57561e715eba39acb814b0b96aff86
a47f4851e41b875e64a5aa8811cfe83e7ea134af
85714 F20101107_AAAOAS sundaram_s_Page_106.jpg
63aa3f540db7450abd35fd508c54ddc7
dccc7ac26a17c79080f06f131e6c5865691a15d1
54865 F20101107_AAAOZC sundaram_s_Page_120.pro
3907403fd631a886f6c12bcd8104ddd6
9fdba44e459cfc72d9760d6c7456f8e9194b47af
60309 F20101107_AAAOYO sundaram_s_Page_106.pro
24a44c3d1a98f5100ff37a86c906db26
044ee2bf6f4247d503548d7db69bc8b6dbe0012e
47354 F20101107_AAAOXZ sundaram_s_Page_091.pro
818f086a5c31501ff596f216c32e9cdf
a2c3798acb20255f450ea300e5f92fdfdf4315c4
80928 F20101107_AAAOBG sundaram_s_Page_120.jpg
4fda8c27b954c942ab125e6ac08ee096
a7c06a7adebf4527655c21886b8e8594317a3cdb
2046 F20101107_AAAPEJ sundaram_s_Page_051.txt
80e4a83c85c3746b8eab363ce1e09cc7
006550005967c288359d71ceeba4c2989a838b80
F20101107_AAAPDV sundaram_s_Page_037.txt
66bf08d5492237fc0ed97db6aac87f44
6065926881cc83b5e6c20312819294ad264eb848
84403 F20101107_AAAOAT sundaram_s_Page_107.jpg
46592a63fb7ca5d9a566aa014ea74e4b
9fd6e6d1c68f614d44612b6b6075844329e788fa
55972 F20101107_AAAOZD sundaram_s_Page_121.pro
804997223ea8c13a053601728f56a157
a191d6fcef064c133711fb55565cccc3a34a8ec0
59533 F20101107_AAAOYP sundaram_s_Page_107.pro
869c8e99e5904ad1878e12d296ca2ad1
6ece18c0c1001d6d141ab36add206a6ba0df8480
80555 F20101107_AAAOBH sundaram_s_Page_121.jpg
0bb5d2c5aaf2d2980284fe513935f2dc
ae54537c132856ea4ff04f399742ae6de1ac57b8
2419 F20101107_AAAPEK sundaram_s_Page_052.txt
eeabf801deabe7e609508b69c6840086
7bc9281925c03de46b2e34c57a59d25c9f196a44
2174 F20101107_AAAPDW sundaram_s_Page_038.txt
15c3f126f2039fd4241422adde955f6c
e69a0d051bb52c7c7ef4c6ca4684ff0e99bbd702
28844 F20101107_AAAOAU sundaram_s_Page_108.jpg
7910b496b150c9693d2b07acb378ddd3
9d90cb540e4dfddd2b7017cf5d5ed48bd2817968
51693 F20101107_AAAOZE sundaram_s_Page_122.pro
5e4e0f9b4872e39129142426ad52f8f4
9f9fe7d16c084960f081e2b388eaf02207c8319a
11320 F20101107_AAAOYQ sundaram_s_Page_108.pro
d9c3747f1d059d77df29188e9f51f750
62d8c2cd1eb56ea79f918d5cd00d7da96f9ddf46
76353 F20101107_AAAOBI sundaram_s_Page_122.jpg
2b8a9a465f0bf96db5b53e5bf70cd77c
b1cf5d41e6c63788b4052de3b7631871dc059391
2078 F20101107_AAAPFA sundaram_s_Page_068.txt
80e56ffeec30a3aae43cd81ece9108eb
bb82be9a8b920bfc9307e7aafa4d2745ff661980
2196 F20101107_AAAPEL sundaram_s_Page_053.txt
a0cb7242cf21c15d0c0f6d5019161ea1
5aadddb4b2673d6d75f29cd2476055af992109d1
2261 F20101107_AAAPDX sundaram_s_Page_039.txt
9390a9338de7c805131f4ee6b46d9cf5
a4671140a2ee76fb8f31af639cfae28b0ecd7d2b
76243 F20101107_AAAOAV sundaram_s_Page_109.jpg
21c705cb32f03f16acaf270e7739afc9
2841825003cd829d9a62b732f6107e6f6cf4ae78
56075 F20101107_AAAOZF sundaram_s_Page_123.pro
517ae7a455e5009db5ae80358c032c33
45509055f83edcfecf4a1b028390c126d1c894b1
F20101107_AAAOYR sundaram_s_Page_109.pro
419cc95e69bb62835d64e7f91513754a
69f51e988f82583089a6ad9687084afb38c05e7e
79689 F20101107_AAAOBJ sundaram_s_Page_123.jpg
eaf306095b1b5b610e838f02666b8325
013a6f3327853687d189bc99576876942764dcfe
1702 F20101107_AAAPEM sundaram_s_Page_054.txt
9ed5b8433414a7b84fedb02ae71534a0
f8151f80b55d816b07524996eeb130273922dd1b
2155 F20101107_AAAPDY sundaram_s_Page_040.txt
487dded1711baa346dfa4a16e8c7d60e
c854e837d25b89e692e6055ea8008b3c253b2f27
81138 F20101107_AAAOAW sundaram_s_Page_110.jpg
a3fd7f4562923cd45519c16f98904f57
69f0bdc1efc5f76d48022d15ccc2e8a842613730
55552 F20101107_AAAOZG sundaram_s_Page_124.pro
e80d40f6c23f07a26510e1e360034e19
b0af0ed60948329feaf485ca767bc0668247bf73
56588 F20101107_AAAOYS sundaram_s_Page_110.pro
59c5f3f8465e4a5a4f2d0c9d97e81f8b
6a12577d264aa678f823d500bb40df74d02d0a6d
80816 F20101107_AAAOBK sundaram_s_Page_124.jpg
f6e252253b20f14d19d78fce9e0127e5
f2c7109362aa6cf3ce8c35c83ec504918c2c78bb
2113 F20101107_AAAPFB sundaram_s_Page_069.txt
994a883c9fc12b16d1b9ed56fe49bf43
06e0f3e0adc1cee01f01ffe621cccd4a7ad0620c
1958 F20101107_AAAPEN sundaram_s_Page_055.txt
2ea72ed9a5ea1adb1e2dda422c7ab1c4
fe079621cc1f2494423e89045b00ed63c509cae8
2227 F20101107_AAAPDZ sundaram_s_Page_041.txt
dd5ce0b5f29ec42e176b6c5bb18c2856
24952d077402b9f6d5018e6770778cca32139414
79245 F20101107_AAAOAX sundaram_s_Page_111.jpg
64ae9824afd16433e08411d053022c7f
9b077b93872d7f251974e76f51a0852f285c8a9a
54178 F20101107_AAAOZH sundaram_s_Page_125.pro
342ed5c064c0b48d29f0a8ecfe8eeeaa
3efc8f83f58df28f0f575013cba6b36ab6436e27
53155 F20101107_AAAOYT sundaram_s_Page_111.pro
b0adcab81bafe65483daf99fb8d175a6
7d16fb825833776eeb24d758cb54441e1b7f0810
78274 F20101107_AAAOBL sundaram_s_Page_125.jpg
f989436670845d94c8c076ac747f3693
ab8bee576ec97cc271a160a04cc03512b8e89fe1
2190 F20101107_AAAPFC sundaram_s_Page_070.txt
7be76b282c8d25238ff836ac16652ef5
65e249210c3bc9ef9de922f257928b01540db7aa
1927 F20101107_AAAPEO sundaram_s_Page_056.txt
665daadc9368bd08fc608db83cfb4c8e
6060ef5981829c3abc42a093c98e118a5dda4ff2
80181 F20101107_AAAOAY sundaram_s_Page_112.jpg
f57964d4413dcc4daa4235ca6baa8a94
a4a768bab5a245fa7de0e831a618a454200c0a21
54510 F20101107_AAAOZI sundaram_s_Page_126.pro
57ac50e8c710a5214dc5abf1b5c0111c
b58d1e69d8f0859e2b2c21adbb8ae024ce996501
56160 F20101107_AAAOYU sundaram_s_Page_112.pro
018c21a30663a3a28740cd402b34c17e
e68e8a723c74bc19766a21b3c5961174bc046b1f
34350 F20101107_AAAOCA sundaram_s_Page_140.jpg
dcd5ff9b03ab79d66fadd650d8979856
4d3a069c8404d07dde63cea6147ee6356cf9dcca
78778 F20101107_AAAOBM sundaram_s_Page_126.jpg
f5ebc7687b3d0a79d1a46c34e42a293f
84c482cd82c439f84e3f64c8707991b13170ceee
2235 F20101107_AAAPFD sundaram_s_Page_071.txt
579c03da30a504dcbf181d3921b9016f
830eef5243ba298e5a4e80e2be89a4a2bc77d301
2215 F20101107_AAAPEP sundaram_s_Page_057.txt
03e20a180b3e8a7c2a52172bf1857ca2
ed92480b96bd9b649a435b06f3cb7e29e8aed03a
80841 F20101107_AAAOAZ sundaram_s_Page_113.jpg
154fbd077d83dfa6fb9af5313fa3e4c5
0f62c93d7698d9af8bc06123fba400672f877cdd
55772 F20101107_AAAOZJ sundaram_s_Page_127.pro
3da33b30ff135368a3327e7367430469
f7338cb4354353e86f4c6ad5ed4f3c92435bc0e0
26175 F20101107_AAAOCB sundaram_s_Page_141.jpg
6c5be3729d5fb17534c73ee4a704e4a9
05442d666731b0d45e709f3b27ca1e55a4f02fc7
81033 F20101107_AAAOBN sundaram_s_Page_127.jpg
6997d5a65af753552b348eef4926efcd
aacf83cd9b8629daad847b8796d96bb271697723
2080 F20101107_AAAPFE sundaram_s_Page_072.txt
aa0019bfcd52151abe3b70ffada496af
b4bb986ad36195a4c6c1d35a89e2333d6801357c
2956 F20101107_AAAPEQ sundaram_s_Page_058.txt
df37db841515da0ae47601b130c5f23c
efdef24a117733ef07c2dcddab2c530012dc9714
57945 F20101107_AAAOZK sundaram_s_Page_128.pro
6475711751042e1f68c923df827f2d23
cabaea47abc492c6d3dbbed4db769cb84ec5e87d
54659 F20101107_AAAOYV sundaram_s_Page_113.pro
de6f91d998cb274df43ea7d1c207ded1
bb0fc537d42c1971cf341fd6bd0b4ba8547f92ad
37536 F20101107_AAAOCC sundaram_s_Page_142.jpg
9e820f5f54576e851ca6ad771bccd7c3
15508ae74333353fc436e82e4dc45304d7555f8c
81956 F20101107_AAAOBO sundaram_s_Page_128.jpg
6f6673a264b5261d14cbb35d05db165c
1a588bb46b9dd991fad5d706d55ca7f83e8d90e8
2070 F20101107_AAAPFF sundaram_s_Page_073.txt
edf0f5de472acfc0fe3b63527e482cef
c9d44df51a2e235acd4aace6a44658703bbf1436
1817 F20101107_AAAPER sundaram_s_Page_059.txt
2c462ae19af0a62c9b9198371edbf1ca
dbe17262cfc5d89a60fe6a8ec4ec641437752c6f
57173 F20101107_AAAOZL sundaram_s_Page_129.pro
ad95907150fba3be5a07b8e034c074cc
33a18f3122e57f9742a76102f9c70e546318ecdf
52269 F20101107_AAAOYW sundaram_s_Page_114.pro
dc2fd59d43f15285e5f5cf6b8769623c
85d56a8cb9a3ac7c8b1ae56aafd016e61ecc431b
32188 F20101107_AAAOCD sundaram_s_Page_143.jpg
a8107fc1a8a153248f97056218ceecdf
e5edffa4f1cc60d1567be5dc1d486b7d70bcf0cd
80746 F20101107_AAAOBP sundaram_s_Page_129.jpg
9f71d03a51930f94b66d053cd4126412
a6a20f9873087615b27937814ff1a9ebae5d8de9
2076 F20101107_AAAPFG sundaram_s_Page_074.txt
5c8a46014b2e8be5d3f6704e2ebbd5e0
b08c08f540c5b3cc7385dcc289e3af05584f7cd8
2053 F20101107_AAAPES sundaram_s_Page_060.txt
21e37fc11f6780f7a5333174289dd218
152ec2b14d3ad85cfec4133a98fbb932c2a37e54
56934 F20101107_AAAOZM sundaram_s_Page_130.pro
052c0ebd9508968c8bb3b62481253a80
d27e5fabf8ff5c8affab0035950746f4ff0fe393
54246 F20101107_AAAOYX sundaram_s_Page_115.pro
bec445fda330fc694adb0a4456ae67c5
4d7f5ec234c53a577e6c624cab02a7cb770c7f1d
34903 F20101107_AAAOCE sundaram_s_Page_144.jpg
dea45e596e3345a19945ac0e3a7c1f12
fcf2468f7e0b6db8a9c1008039e8591d351cc909
81334 F20101107_AAAOBQ sundaram_s_Page_130.jpg
48962fdc9951749292bdd64416d6d040
0f7d783761807cfd4dc6c8d770ad687890ed51c2
2112 F20101107_AAAPFH sundaram_s_Page_075.txt
107d8f434faecd234b85ca0ef9f847c4
e90043e342476190166252c97b8e0cdd4cf912fb
2109 F20101107_AAAPET sundaram_s_Page_061.txt
d80c4c19df0bdc11ff12140325196b5e
998add6f4d6599fd3d418e36d5e3cc12f182bfc8
56302 F20101107_AAAOZN sundaram_s_Page_131.pro
b80a1c5ed88d4d4a338ab8655a0bc89c
98402d01785a5ffec507bd8b60e485118c2def23
53960 F20101107_AAAOYY sundaram_s_Page_116.pro
9fae37df7645ef9def3198a1f61c39b8
f21f841d1a895da898899add1d6f7f4ff7cfe6e0
27629 F20101107_AAAOCF sundaram_s_Page_145.jpg
486d19ce924fc96036ecf1eb5e2546a4
531dcafeb7c2393251694a135d3130d32039e917
81405 F20101107_AAAOBR sundaram_s_Page_131.jpg
09173737344ae5457dc92d8d005416a3
86ad8bc64a4904418729c44b87504e6a4f445763
1971 F20101107_AAAPFI sundaram_s_Page_076.txt
23dd3c0729ddc1da924ddd9b7e7331b0
e34329d338783d2980857e909f85888778769aed
1951 F20101107_AAAPEU sundaram_s_Page_062.txt
cb0bad6cbd6d9f8c4bf57eb1c3d4cf7f
bd4baf5e90357f8ad568c97a46c8953adbe5ac5f
12151 F20101107_AAANWM sundaram_s_Page_067.QC.jpg
ecc13acfbd1d85492f5d08f1e90de7d3
94f3a627ff507d1e59cd8624b54c2fa4db5bd96f
52163 F20101107_AAAOZO sundaram_s_Page_132.pro
f379db9d968aa50117d5ed7a897360ee
c7a8b10128571f4c258379cb11e6e4cdb46f4b1d
51953 F20101107_AAAOYZ sundaram_s_Page_117.pro
4c1356fb071addc1b559950e229e5682
d66d4c0ccd9693dd5100fac366239d092d112fa0
38147 F20101107_AAAOCG sundaram_s_Page_146.jpg
661f29cb1005fb44afe4c285639b3af2
6b4cd6830612f4342cd17a44b21caf773ab76fa6
100383 F20101107_AAANXA sundaram_s_Page_010.jpg
b54049736d23afe1fba09a9da3aea93d
ea2fb01e2c01acc3db99d01e0c9a1f2e27b233f1
75317 F20101107_AAAOBS sundaram_s_Page_132.jpg
d43a3d0a5163ac8b30fe5d51636b39bc
8bff587739855b1369ce3b92fc010240de2bb948
1959 F20101107_AAAPFJ sundaram_s_Page_077.txt
69e314418fc04f561e314108f5b9f557
c065bedab41fe48ee64e3dfa3196f3454033d708
1800 F20101107_AAAPEV sundaram_s_Page_063.txt
e3cb7513c12a9ae8fc93a41ccdb25859
4f50ed49df16fdae2fd6066f4483d87a2389ced5
F20101107_AAANWN sundaram_s_Page_061.tif
26e9068a444956761fc29a918239da08
d6dd4d400b2ebd07d601688a211a74c4d717e200
55404 F20101107_AAAOZP sundaram_s_Page_133.pro
d50d51c9cc95bf93ee579fa005ddf6a9
d7b8b69ab15783b74e8cfeb1dbac506848a71a2c
33367 F20101107_AAAOCH sundaram_s_Page_147.jpg
30a3c457bf3ad486eaf49e69feac2141
1a11ff24318d49cd30ac7367d7e82c6a9340066f
18168 F20101107_AAANXB sundaram_s_Page_011.jpg
f4ba91a3a5a8cac0545423542a4d07ef
497baec9961a9d78671bdd2a179d9d6bd09bea4e
79146 F20101107_AAAOBT sundaram_s_Page_133.jpg
d2958a92d5a136f456944aa856698f29
c201a154c01258db6c745c39d254b272e7a3d25f
1499 F20101107_AAAPFK sundaram_s_Page_078.txt
f0d1195f049a2cc423b44f45d889ddfa
50d7cb30fb3a391ac42f45b70cfcd37af41c4af8
1671 F20101107_AAAPEW sundaram_s_Page_064.txt
8cec638a8659b21baa5878d0d7716268
611e817064fba270ab64ced55d050aa1d20aebca
239312 F20101107_AAANWO UFE0022424_00001.mets FULL
b15ae8380eabb30b7605b51eb9ab5ab8
2993fe2ac448262f4a4f8f2afb41645268811e58
F20101107_AAAOZQ sundaram_s_Page_134.pro
f858622c5a6820d41ba053397745be94
4280e52a539162a0804c25806c9a23788f46e17a
36638 F20101107_AAAOCI sundaram_s_Page_148.jpg
4d2413ff8ca4ccff375ab5bf91bdcf12
9cdbd011449ab8cb8781825c2edf6860fb0c378c
70237 F20101107_AAANXC sundaram_s_Page_012.jpg
a91544feb3a14b67177a839d2ae12481
dcc822bea3783270aa7d9dbaac5759a657bc0aaf
80397 F20101107_AAAOBU sundaram_s_Page_134.jpg
df9b5ec5ae63f5d4bf20a501dba4b59d
c0657f29872cf87eb3290dfe4d1f3aa75ab228d9
2142 F20101107_AAAPGA sundaram_s_Page_094.txt
84b4a1a4c2b76d9689e540e0ac501e2c
6ec47938ad40e94cdc9f811246796aff92589fa5
1589 F20101107_AAAPFL sundaram_s_Page_079.txt
8b082fe73f7c4ab75f2a0d9ef428e648
a1f06e76d9d6281dfc0ca23f9d7abb4171404f00
1280 F20101107_AAAPEX sundaram_s_Page_065.txt
1be121139f2def4c801dd018a8b2a615
535a7b3c0be476ea411cf0f4e1e5fae57dd7f885
28872 F20101107_AAAOZR sundaram_s_Page_135.pro
bfc05600029fe3f2168f8a4f9572c442
960addd8e96bfb85942b1713676c600671ce595e
35487 F20101107_AAAOCJ sundaram_s_Page_149.jpg
b952882fb507d3d9f8b4e2e82fc49acb
533ec03c1619291c4a140c396e0bad0da74a6cd5
60656 F20101107_AAANXD sundaram_s_Page_013.jpg
16eb8935c4a6afcb0336ff1da3e64616
fef0dc916c14ae58c9ae2868d4172e691f098e47
45042 F20101107_AAAOBV sundaram_s_Page_135.jpg
3a471e6edceed863c61498bfba836fef
aef20012673f4f38434e0032ad06863386b7d225
3173 F20101107_AAAPGB sundaram_s_Page_095.txt
af013a56b6a2e0d2db8ae6f89b78abcb
4a1312daf5a05873a014283cb843c7352feab026
1329 F20101107_AAAPFM sundaram_s_Page_080.txt
60a2131125d658b67665e52d69b85cd7
646a0f701cbdbf99fd45e092d4228abef6d14145
2932 F20101107_AAAPEY sundaram_s_Page_066.txt
82531d6cfd20f8fd31f7cf3e88277dbc
937ae353da1060900e18ef967042a5eaf067d7b3
1525 F20101107_AAAOZS sundaram_s_Page_137.pro
b5aeed0d21cfe8125caef05b1f5f0aa0
ccd839b3ccd237ecead2a10bfa7e929ce53c4e4e
F20101107_AAAOCK sundaram_s_Page_150.jpg
985c68bc8b16449d8f02aa22e48b93a8
9345da178a68b02498a0d016b084b512dac78af0
76605 F20101107_AAANXE sundaram_s_Page_014.jpg
f733e9de61fdce48fc02327bfcdf085f
23e1f09783691cc7902db77b99b4b7ecfbd44278
8863 F20101107_AAAOBW sundaram_s_Page_136.jpg
3c530243eedfdc4ae41e777cb3406498
5fbc2e7170daad4304d306d1e95aac80fc529d3e
1290 F20101107_AAAPFN sundaram_s_Page_081.txt
56a85bdd91f24aa6573722bd99376e31
50b5995304ac9311bae8219ddd1f523d3f744a64
1104 F20101107_AAAPEZ sundaram_s_Page_067.txt
16430ea8a0d795581b14741c08fbc4bb
7bac80c8f7586734068f5e7bdbe02baaa7e5cad8
24579 F20101107_AAANWR sundaram_s_Page_001.jpg
b66a3445eb1d6eaad97e0aa8f84fd4f3
2bb1736b22759d94e2394cd5ee99cd99b421362e
33454 F20101107_AAAOZT sundaram_s_Page_138.pro
3101730358164ed29fb58cd520cab620
1f964c85cc0cc52b73e698f27309da3fa4d878e0
33211 F20101107_AAAOCL sundaram_s_Page_151.jpg
11b0e4a43b02a3d00b59e21a4b4f1057
440a0ac8242e63822865d21445de31a086b8fd04
72577 F20101107_AAANXF sundaram_s_Page_015.jpg
e6f02a4662c037cfa4d876bc024a28f4
e5e0f724bead291d5ab73e6514e656b3b73c49e7
11713 F20101107_AAAOBX sundaram_s_Page_137.jpg
bdaa24f3a9c35da6a5012a211b899105
de50c7a1ab111a7041fc14981ed4537e65d8efd8
2931 F20101107_AAAPGC sundaram_s_Page_096.txt
63c5587a2875fe6a7736265c3087d1c3
be77ac52557450ef331426ff856c6641028a266a
F20101107_AAAPFO sundaram_s_Page_082.txt
b5c721bfdb02503140e5bdb1e029905a
fbbd7036ce8b9f1ae127b9aeb05f4bad4f43f8c6
9960 F20101107_AAANWS sundaram_s_Page_002.jpg
e2bf8ae1f70c7920ffe021b170478b98
a2527922ea576af46a367249810ac2b418225052
44597 F20101107_AAAOZU sundaram_s_Page_139.pro
5cc162a2db5900f072e7737fadce42fe
e4c57cc1c91977cd70d18dd1b1a01a1146c0ac2a
29775 F20101107_AAAODA sundaram_s_Page_166.jpg
3ca0606b2e5611849d65e5973222bb9c
4c0409e1bbae60e017cf5d31b4e6688b2f409288
23407 F20101107_AAAOCM sundaram_s_Page_152.jpg
3a4fc9e36d5ffabed13047ecba1e6670
edc8c1b0c01ce57da284859edecfa4ea2e26dc6b
80344 F20101107_AAANXG sundaram_s_Page_016.jpg
e76ac61d387dc4eacc8094ccc6e35642
21d3cdf30b0aaec20176d477edeee48265485bde
29904 F20101107_AAAOBY sundaram_s_Page_138.jpg
ee5d60076abda10661b8f5ebbff65a48
d9945d88a9d8e4cfd036a8c1288a37379c5f433c
2903 F20101107_AAAPGD sundaram_s_Page_097.txt
a23cc569edd2ab2813b4f4fc1c654a9d
0cbc8581781898b25c624780292fafa487f62a25
1616 F20101107_AAAPFP sundaram_s_Page_083.txt
3e2b3ef87561c85c7407e0faf9d6a8ed
b6e35a76f737305f225b296fac4d4657df99c6ba
9469 F20101107_AAANWT sundaram_s_Page_003.jpg
21f5ad4dfc7e899d924c92c1ea40499d
408970744485bde1657b3f925597b9fc89cf2144
40559 F20101107_AAAOZV sundaram_s_Page_140.pro
bd8005e789de1be730f63b8fe66eebdf
a2eac0e695862b4f97f98c28a929119f518dee83
29917 F20101107_AAAODB sundaram_s_Page_167.jpg
3337faae45b33c625ee84ba9eb51d6c0
ec8850be84888dfe90c43481ec7de68db2e45336
38190 F20101107_AAAOCN sundaram_s_Page_153.jpg
93424f893fe2d83aac86dfd7501399f4
df93057699b067baa7866fb364ccd01ded7627ea
77244 F20101107_AAANXH sundaram_s_Page_017.jpg
bcdeaab408d8ba73107cb7903ec9eeb3
8b6c2a9a71b850ccf022587fa125a44bcb0f7c14
27390 F20101107_AAAOBZ sundaram_s_Page_139.jpg
999840d7808c5914f12b154608eb06fb
7087fac8d0aefb6749a5add26bbc9f84f40ff360
3174 F20101107_AAAPGE sundaram_s_Page_098.txt
1a09220745b903eb3265f0f92d89b07f
416a1b969e95feb3e1aa967b0c35673b6e11237d
1867 F20101107_AAAPFQ sundaram_s_Page_084.txt
a4b872f849c65e73ab0ced7e5a6f01bb
dce87a5a5b40a9f6d23eb6d48884cd8eaaef8c1c
31846 F20101107_AAAODC sundaram_s_Page_168.jpg
eff085199eeee0cc566528b0c04e9ce6
fd082a33a2f3b8daabf14535375faaefd61472bc
24187 F20101107_AAAOCO sundaram_s_Page_154.jpg
094d8a674b117c096bab98451d9a4a08
0e2f0ce426252f2f865bd63d606798c209e31bd1
79440 F20101107_AAANXI sundaram_s_Page_018.jpg
373abaf02d784acf6b3902bd20e75452
6e0d06d20837d8b3370e78633e3a01f239eacb1d
2535 F20101107_AAAPGF sundaram_s_Page_099.txt
30a4e0720c1b53511ad526b746e4d34f
4444f1893bea73b23962e7b27fcee5bd34964e30
1715 F20101107_AAAPFR sundaram_s_Page_085.txt
2b381d4e5396ed48d6b13110bb94d4e4
066dca7cc52268bc7b47c52312e2470b9f07f019
80535 F20101107_AAANWU sundaram_s_Page_004.jpg
1deef61ea356becec38f9245f9fc5c42
4cdfe7dc902ba42a58822f5d6a691f6715df1bc0
28704 F20101107_AAAOZW sundaram_s_Page_141.pro
f8b7182d2b5d3407500477c9d4eb6513
45ae43f80c45370d743bcd66a007c092161e2105
29710 F20101107_AAAODD sundaram_s_Page_169.jpg
31573d918a94f692bdea5c5836fa07fd
6fa1ac2f7884ce37b573f3b43934c16cb8231728
64175 F20101107_AAAOCP sundaram_s_Page_155.jpg
52f57491cdd3d8732fa7adafbb4ffb25
e0474ada82bc4c0a729b61bdf50c99f9c210ef18
81120 F20101107_AAANXJ sundaram_s_Page_019.jpg
223282ee7f5049628e638ee24350bee2
ebc1d16bf8aaede778b9f2c936e0a4103122597c
2374 F20101107_AAAPGG sundaram_s_Page_100.txt
69c577ee7185caf871dfaaa66fbb98c7
b5d6c42dc482759cfdd284679ffc3e9cfc546945
1729 F20101107_AAAPFS sundaram_s_Page_086.txt
10115bc03bffaddc84bf7855f1246813
e40dd102057ddb56f9bc5ff9373b841a1d62ce7c
17000 F20101107_AAANWV sundaram_s_Page_005.jpg
9ec7350384f9bd711fc527c6cc31815a
95d4276be20005927a43fb299668bd5455daaf4d
47553 F20101107_AAAOZX sundaram_s_Page_142.pro
5949fe2e126d186ba948d062179167da
f74164dfccc417558481c4545ef0caf1f6a36a95
34124 F20101107_AAAODE sundaram_s_Page_170.jpg
41305e2d213b584abe4e1f860d8ba22c
287b927c94a7d7acdfda67b8acc7058c9ba68b28
73188 F20101107_AAAOCQ sundaram_s_Page_156.jpg
03f71920309e4e0460523f8fcede00e1
f8e652901fb70807853c4b838e45f2543214baa6
31605 F20101107_AAANXK sundaram_s_Page_020.jpg
9543906a8c128e54ad360735d6def3bf
1b5162b0308309963ef614b72dae1816d0c8758c
F20101107_AAAPGH sundaram_s_Page_101.txt
2e04a336781d93c890acc0cc396d1c16
0d08e581e2517a02fb3cc64de063c782ab4256b9
F20101107_AAAPFT sundaram_s_Page_087.txt
30257f91a4cc61db8fe21502ea468987
536620e03e4f5e128fcb59558da63bc7264d09f7
90765 F20101107_AAANWW sundaram_s_Page_006.jpg
302aa8c4232185b64560cd073de3a30c
2e3179447d54100330887d53391ba4d5f668857c
38727 F20101107_AAAOZY sundaram_s_Page_143.pro
5b52206acd37541a556babde429f932e
efeca7eda977f00fd755b89cf1728598266ef11b
20210 F20101107_AAAODF sundaram_s_Page_171.jpg
85f0b8e31350c4f40a9bb674477943ed
f2076bbf5b0d4092d9ede15a5e1c5b2830e2b963
F20101107_AAAOCR sundaram_s_Page_157.jpg
6b1525ff8f6bb3fd9315a34a11bfa625
0b71018c238dab34440caeedaccfc74c3d213d96
74017 F20101107_AAANXL sundaram_s_Page_021.jpg
e2f869af2a59c1f08164fe79a544efa3
09fc6d63ce9e70a300a14e6cf327661e73f81b31
2180 F20101107_AAAPGI sundaram_s_Page_102.txt
a33c297204f979e4bf2b33cf961248b8
05af2f0ce99dbe31705df25e65bf364808857125
F20101107_AAAPFU sundaram_s_Page_088.txt
ee53b6c9fff60c33ca9cd8df31fa4ec1
cf5e9dbfc2a75227246e59d9c8be3d7fa2c5fca3
95468 F20101107_AAANWX sundaram_s_Page_007.jpg
7d604fb510261b9e886ea8060e3d8177
2f50a9cc69b7a0efa62bc7b62a990dd4a8f1c2b7
42753 F20101107_AAAOZZ sundaram_s_Page_144.pro
28a06e640fa70a86b278911d2daff153
68468fc2899ba17ee5fa67948af02935dda343b9
85597 F20101107_AAAODG sundaram_s_Page_172.jpg
47d84e9461ab29442fe3d8dc7db1beeb
55565c25a22d0158a35e23188654599c6b44979b
F20101107_AAANYA sundaram_s_Page_036.jpg
2e5c92782ca77eccae592bd1924ace5e
c3207d42baa495049db99e8d1b2c45a964557b3a
81065 F20101107_AAAOCS sundaram_s_Page_158.jpg
3a58bc41a2634c26d02537a1dd7fd64e
0bbbd94de7f27dce819d4e0eea0d34d8094f7059
79684 F20101107_AAANXM sundaram_s_Page_022.jpg
656464f83ad080c94cef5448ad96ca17
e2bf32d7a268093d60e4b154d81e389afaeb73f2
2908 F20101107_AAAPGJ sundaram_s_Page_103.txt
c83724a0f52056ce284873c9abfe22d3
0357def55a8b5b7e5042db0f4d6f7a31a9a7ea3b
2365 F20101107_AAAPFV sundaram_s_Page_089.txt
ef7de6216eb2ed22da44bd885806c67e
c5c1f4795ecc72dd474d811b03e65d7ace54f35d
F20101107_AAANWY sundaram_s_Page_008.jpg
283aafb1ca6ab116590e0eee32659629
300c923cd97e798ecc2b7849f395e2a7a5eff2af
81480 F20101107_AAAODH sundaram_s_Page_173.jpg
5a1d8f4540ab90deaa7ed67a6e8074fd
6577ee35ed000d98e10174e3bac2bf65dc4de48f
77011 F20101107_AAANYB sundaram_s_Page_037.jpg
89408df67bd8b2bcc04a79bde259235e
371797f3babfd60ebe1a8cad638282ee828e2811
80109 F20101107_AAAOCT sundaram_s_Page_159.jpg
e498ee41db0f616f0767ae1d05274069
f7736decaa409b4a7ef337f34bf1692132f87430
80197 F20101107_AAANXN sundaram_s_Page_023.jpg
021e68fc9c0f1c376a23e72553d095e2
573ba0a23dfc77f951e76a3ac0972c758c355982
F20101107_AAAPGK sundaram_s_Page_104.txt
283bab61d053acf68ea2641804b61080
cf3d975465415df5a98afc19ca7ffee56d9ed277
2305 F20101107_AAAPFW sundaram_s_Page_090.txt
c6228a3e5a259c13e34cc0753fe91035
8dfb9dde8b2ad8c793529a6d6b3b8984d8af634c
93601 F20101107_AAANWZ sundaram_s_Page_009.jpg
dbe59c48573a6035681376750d4ad162
13ac6f7112abfa4f2bba7ae7d9740ea147c33c05
76820 F20101107_AAAODI sundaram_s_Page_174.jpg
0d7ae6dad29103c477e89b0a40cc4c93
3ce7c9f2751a3ca5caccd78229f28b5f814ba6ea
80832 F20101107_AAANYC sundaram_s_Page_038.jpg
764e5328eb1217aa55ae5a392e65d212
3997a11ffe8274fd8b8b43d02a4e44b6dcaeccb4
78508 F20101107_AAAOCU sundaram_s_Page_160.jpg
b29eb0a98abc2071c03d3686a1d64065
868a2a94e5cff2c6a5b9842772e3d0777d135545
81599 F20101107_AAANXO sundaram_s_Page_024.jpg
501f2e43f4425682104a7f0684dd28f7
6deaa685af146c1649bd66b7ff18dcc3a9cfce8b
F20101107_AAAPHA sundaram_s_Page_120.txt
d795fecb071f862716895eb52f7c4166
90f09443ebba8bf288d329883e76b14266e3ec1f
2534 F20101107_AAAPGL sundaram_s_Page_105.txt
a05053f7a139dcfb973a74a0d2b45b62
86a38b20b8c560a6f5462f48b7df294f630f36d1
2262 F20101107_AAAPFX sundaram_s_Page_091.txt
69a1bd5aaabdb70453a727924ac873e7
8cbf7b0534686caa33d9c86e0b57f639d867d7af
77576 F20101107_AAAODJ sundaram_s_Page_175.jpg
fa418dd5122f42f3bd1e03019fd6bb0d
16b66a433005ff9435580f78daef1e3caaf83f69
83266 F20101107_AAANYD sundaram_s_Page_039.jpg
40449f55c8562dc14a0828502937fe42
2944eafeb529b2c57b0c7ec9060e77897bc35f4e
78433 F20101107_AAAOCV sundaram_s_Page_161.jpg
4eabe349763319621a9dc4935b182d2e
f20c0d43701d20c7a42a5594872192b3031b9285
77131 F20101107_AAANXP sundaram_s_Page_025.jpg
64dd97d7dd3986b59b8c53ccc3f14737
e14c2c36099548320e8ff4b2707d49a870e18d19
2199 F20101107_AAAPHB sundaram_s_Page_121.txt
21a402bc13ecc017d7c5098ec1eaa8d1
50bb69eaa25c286481d155a07599cce9dc2223f2
2425 F20101107_AAAPGM sundaram_s_Page_106.txt
a37f0e78bebc0ad950db900d2ab604c5
fea0b6ae0102177e226c82fcb925d97bf55a8ee2
2051 F20101107_AAAPFY sundaram_s_Page_092.txt
1a615e412fffcc07ebfb02c28a5dd109
c9877e510b51dce69425ddb6b343c6e3f864ad02
79562 F20101107_AAAODK sundaram_s_Page_176.jpg
cb2f8dd006328964dbf33e9ec6c7d276
d89838477434b3e00e3515862547dbe79ae127a6
79263 F20101107_AAANYE sundaram_s_Page_040.jpg
5f244c815fa8b4271ed1772d6e81a5e0
7e373417a36ecae0e2fa24ed491beef415084f5c
78183 F20101107_AAAOCW sundaram_s_Page_162.jpg
e486494d4c9f72637768bfe37488d1ab
f4f42b926ddb4fe9980acd08483e35ae2b053b08
83852 F20101107_AAANXQ sundaram_s_Page_026.jpg
e0e2cd87c0d12e6e406c1aec15c94a3e
110cdd9697064f497a75f87df1a486c390408e18
2059 F20101107_AAAPHC sundaram_s_Page_122.txt
29c575033dc2e77541c3375c81a7da7a
9442c0b8784d4214b579e828df3c0b2c6105b8b4
2540 F20101107_AAAPGN sundaram_s_Page_107.txt
f6cd9da8d8e3090aad62305884f291e1
b5566b350080a361cd5b85fc4ccd7bb04655e054
2342 F20101107_AAAPFZ sundaram_s_Page_093.txt
f36c0bea4a420074d747f51fed6d1855
acdad4c7ab585f81ff0e28a41bdecdf38e9dee0f
78467 F20101107_AAAODL sundaram_s_Page_177.jpg
92da594800f2e245bb38656b38470a1a
ec9044ea43c193167f34920ccb8865b4e9bfe2f5
F20101107_AAANYF sundaram_s_Page_041.jpg
71c4a74522f854eebee05a5f28583f5b
57be6eb5d8efdce393a7a7f4b2f54a903a6b0787
56056 F20101107_AAAOCX sundaram_s_Page_163.jpg
de3bfe5158dc04d033711dc02a5257c7
52464d11b4c47cd6dcca1a52e50cc6627ede4158
82183 F20101107_AAANXR sundaram_s_Page_027.jpg
c16e68d3b1070ee00c49af50a2782290
47b3e3906fcb253221b2dd267352438161d30770
83721 F20101107_AAAOEA sundaram_s_Page_192.jpg
9db8cf0f567f1697ca7fb1b459d56aa8
46ff58c3da359e966323c9004e2c2a14f0144bba
668 F20101107_AAAPGO sundaram_s_Page_108.txt
8aa8941d5645cb6769bcae5d34999d9b
624553fc33103e81ff6abbabca0e88e02e0b22c2
71511 F20101107_AAAODM sundaram_s_Page_178.jpg
3c4629f1b7dce05d2797231191ef5c03
acd84c4488e7b900f8eab66d11014a3d7904c647
80235 F20101107_AAANYG sundaram_s_Page_042.jpg
e53da274646bd5ed3f7510e640c9b14a
017e3b76a0a5f09f5b77ae02c79145411df82589
32554 F20101107_AAAOCY sundaram_s_Page_164.jpg
47b0ce1b2ee33e37438bd1f7fa5e061e
564f06992176d1549e971824b8a4ffdaf64114dd
78851 F20101107_AAANXS sundaram_s_Page_028.jpg
aa72720e3bdd3680db421d76adf34ab4
046aed4d07896b94c2b95f16ab5bcfa6e969a259
2193 F20101107_AAAPHD sundaram_s_Page_123.txt
ef1402e2c506f15319c8cb2e74293648
25353d48b727a726a11d51648e87723bbf6b1274
2125 F20101107_AAAPGP sundaram_s_Page_109.txt
7b0f98fa1c8a744319c3be5d255f4805
744055cdce780d52d709a64a99baf567527860dd
78781 F20101107_AAAODN sundaram_s_Page_179.jpg
001cf0d726334f89924a68e32f98273f
49d17f106f0f4527ed851eb1c2b576f95cbbe7b7
80965 F20101107_AAANYH sundaram_s_Page_043.jpg
f348b54e327da51d2f42a27a3a545625
2f8171fed8f6070c9dfb9309d1c26e01895261d7
31966 F20101107_AAAOCZ sundaram_s_Page_165.jpg
5ae436192809dc16744f1fd9e49052aa
b405e83e5499a67eacc33f170ff0468ddf6b967a
76818 F20101107_AAANXT sundaram_s_Page_029.jpg
f79afbdb092a82355778fede4ed0953f
ef31002ab6133d9da1d13cc961f867db9178cd79
104084 F20101107_AAAOEB sundaram_s_Page_193.jpg
cd9512959c8df33da01f3fc6f3afea93
f90c1e4213a2a8309d52dc07479bbce76b6e40e2
F20101107_AAAPHE sundaram_s_Page_124.txt
6005473631bfbf94e99736138bb2d861
53682b29ead69402354acd53ef8d640b670f1063
2214 F20101107_AAAPGQ sundaram_s_Page_110.txt
8f6ffbbe2e668ffbe6a2a1389f4695b4
8c74c8fd4a845b48b6549c38cbc6ac219bd0d44c
71776 F20101107_AAAODO sundaram_s_Page_180.jpg
62958c2c9ab14f85de2152a8e09c9adc
880537026a51f4645f0fa157fdacd7e260c32496
80505 F20101107_AAANYI sundaram_s_Page_044.jpg
3e0e2dec39e2dccf55d9f04cee25fd5d
75bd2b7fa03681037fa4edfb5988e9810e3204e7
81279 F20101107_AAANXU sundaram_s_Page_030.jpg
34fd9f04e88caf845a1ff244cfb47d8f
c7af6eecfe0ecc0f20298b3208affba54d72150e
95949 F20101107_AAAOEC sundaram_s_Page_194.jpg
76a766bb05556cdafc980eb3979b02c3
fa8a0005d5f42f1eebd893ff6984dbbd9dc19b7b
2122 F20101107_AAAPHF sundaram_s_Page_125.txt
1eaa45eb46374c07ed81735be05ab33a
87efe0afe6a23170c7dfb25fc44d6a0b9c795594
2087 F20101107_AAAPGR sundaram_s_Page_111.txt
66c09e7178c04d3901533864926e49aa
5f2871f4cb418b132470c2391f81cf45a50ce274
76703 F20101107_AAAODP sundaram_s_Page_181.jpg
58cbe1d470f81281007c0060d8d681dc
cd1e49c3c327aa008a9a630be2c6dc818a5a2cd3
77533 F20101107_AAANYJ sundaram_s_Page_045.jpg
eccfd9c7df29eab6c8275bf9ca64f447
54dc9bccea0c1136a3c403ffe0e77ccb91289c47
102025 F20101107_AAAOED sundaram_s_Page_195.jpg
0bcf8f35f6f28fda4312884e454b0e67
2be33c9712f5ed782910a4f643b2a53f42fc8694
F20101107_AAAPHG sundaram_s_Page_126.txt
c82265e8cdca6b0c156e78fc03ff6f14
e4867dcc3582b40024264d6a59b1f4c8ee197168
2197 F20101107_AAAPGS sundaram_s_Page_112.txt
4d50f9cefe6a2f5740c8ba0e9dbf15b3
ec3da278a4390457246f2e378c96a96fb0f3d037
71868 F20101107_AAAODQ sundaram_s_Page_182.jpg
691ab115af330988fc9b76ae740a16ba
d00617247142ec61b04b97775ff633603f93ad4f
80846 F20101107_AAANYK sundaram_s_Page_046.jpg
f60c623e955eab6808a841551f16a1fb
abe9b9eb1c2e4126f63adb79a47bd993ffcd7635
76194 F20101107_AAANXV sundaram_s_Page_031.jpg
2bc6df441334f8b579524a80ea5c9550
5ee0288000da3f595d71901c32561bbe06771569
99975 F20101107_AAAOEE sundaram_s_Page_196.jpg
423727904eaae3afde4a331450ffa582
77d8c1921824ef1ac65d72a06e77e9163a0104a7
2182 F20101107_AAAPHH sundaram_s_Page_127.txt
76343928f4f83fec73f319dfc9118bf7
fbf9f6c6cbb3b7ecac6421e42cbb4e0387ad57dc
F20101107_AAAPGT sundaram_s_Page_113.txt
ffd68aee5b17e25441fe62fb2679a918
aeedf0d6d98b87f1fc6e0ced5bf8d96c4a37e018
78347 F20101107_AAAODR sundaram_s_Page_183.jpg
348d6d53279f51a42b0695b2286e6e39
56a42f6a391d615e5aed903e5b16316f6f5fd1df
75177 F20101107_AAANYL sundaram_s_Page_047.jpg
623349558fe837ca0a90d576e5b1fe57
311f23c3881891ffd488ed9668d588588c88de08
83260 F20101107_AAANXW sundaram_s_Page_032.jpg
ae395cd44b57ec2d3435f83f85057a0c
b27758b72d2da776c4860528c0cdc5ce3f7bc2e7
95094 F20101107_AAAOEF sundaram_s_Page_197.jpg
5d90addf1e71e359e365beb7d00784e5
b52d3bb34bf4752dbbdbcbe84b1373d8a7665240
2259 F20101107_AAAPHI sundaram_s_Page_128.txt
b9123d4ec10b7280fa5e7e160b24edbc
8e0a190655dea55e4cb0e7db037d87586d929c58
F20101107_AAAPGU sundaram_s_Page_114.txt
30bd8fa8e9cda81f206580fb97c28795
a47413f61fb1fb3f0de79453efd48e63d3b4c089
72864 F20101107_AAANZA sundaram_s_Page_062.jpg
8aba0fe4e3deeeaadf58b336669a5443
e4aab4b83167bf992ac11c1436953fc6c56b1f00
70733 F20101107_AAAODS sundaram_s_Page_184.jpg
135ee8b8b0dca1997139dffc5fb709ac
d093e8ef2018b15bb5bc625177bbefb093c9661e
77191 F20101107_AAANYM sundaram_s_Page_048.jpg
87190721b9bda988a1fc31329ee22b99
1a0c58f25e41e1192103dbbf1ba0f848e6c7e391
72988 F20101107_AAANXX sundaram_s_Page_033.jpg
187f80b37f2fa01d22141ec06ab5d542
82e56f2e835e22eed4bdbe27b77157813f527adf
99060 F20101107_AAAOEG sundaram_s_Page_198.jpg
4c7b5e93d19223001b7153e08eec9511
1ef0b72697f27935ff22546fa24f8712952330fd
2277 F20101107_AAAPHJ sundaram_s_Page_129.txt
308432b9df67d353923fbf5178eb4025
3296d7e757d2335aec1f6d9ee86bacb9a7c0b91a
F20101107_AAAPGV sundaram_s_Page_115.txt
6a9b04b7cf9beb7c149970d2af722c86
528f7fe86d8158d5511f1acc3d6e7b2a97b305a7
65731 F20101107_AAANZB sundaram_s_Page_063.jpg
5d0b994896ffd316e9819af3ff9c5780
4704e6bc4749e45fc196bce837d7c4a55cf4b05d
80931 F20101107_AAAODT sundaram_s_Page_185.jpg
48567725eb5c80074a4cf86a8081fcb2
65a611428e4f8d1ba0eafa8f606277384e28a955
33516 F20101107_AAANYN sundaram_s_Page_049.jpg
d4947be066a163067cfce4315e6ccaa5
9bea5c581d1b6699c687335a92657d0d96fc7bb6
80642 F20101107_AAANXY sundaram_s_Page_034.jpg
a278c4f16b67fe879a89f42f3a55c029
e7bea147cda132ac8f7b6c44cc05b5e1f5ee40b9
113184 F20101107_AAAOEH sundaram_s_Page_199.jpg
f36afcb822c521554633da5be7327294
d30014119d04ffb907d0a31ee5b557dcbb4cb72e
2222 F20101107_AAAPHK sundaram_s_Page_130.txt
8c9448d5c6777708b0bf71c9beec3c85
a47dcf52b8d57ccadcbd58e2e794aed755729a45
2115 F20101107_AAAPGW sundaram_s_Page_116.txt
bfb8c9697e9641bf1a9f50cd3f4ed2dd
f4aa5f957778348e6f6186714ae3ff8f77def029
63745 F20101107_AAANZC sundaram_s_Page_064.jpg
9ac80a8af9e252211ed86f373fc365e9
6a5821ee53bfe32f20297483d7b1901be71ae86b
76258 F20101107_AAAODU sundaram_s_Page_186.jpg
8fdfe00679357067c82d158b6b89623c
34018ba76c72c3fe0c74ad6981d7e76c22e79486
20258 F20101107_AAANYO sundaram_s_Page_050.jpg
50325a97398212e38fe7e61f980a6b98
ecfed0d250113fac488a29e1646760e266f662bb
84057 F20101107_AAANXZ sundaram_s_Page_035.jpg
354e7a0bd29d5022ea75f461e3aecaba
b7da4a4f14fe4d8deeb151cffb0cf7a4e82fc41e
84050 F20101107_AAAOEI sundaram_s_Page_200.jpg
802c7435bc89b45809072b4bde289993
f0586e420b88655937c66f570b1e319712144e00
1572 F20101107_AAAPIA sundaram_s_Page_147.txt
39c8afdcf23304a6b86ad78e148a2ea0
82d99063160b48b1de670c625ca0fbb442664009
F20101107_AAAPHL sundaram_s_Page_131.txt
951bd47282485ef8372739abee1f2e2b
9aefa0f2dbd9e54df1ab6c095018afb1839d64f2
2050 F20101107_AAAPGX sundaram_s_Page_117.txt
51641e17b35f3adec60ab7650748e7f4
7f30896da16f952a99a7b661076c5faa569116e7
50854 F20101107_AAANZD sundaram_s_Page_065.jpg
85a006d74c5f1525d88275f4aef5d685
343501133383e45ded9f98d73a4d265d10f5a171
91713 F20101107_AAAODV sundaram_s_Page_187.jpg
cac18068538a472b335e66403f4200bd
0a0709e7dbd54d367a481789e61bc2230e60f2bb
73068 F20101107_AAANYP sundaram_s_Page_051.jpg
827799acaced9525c14641abece430a2
2ba980460f0042f0bbd5a2036a170f81fc981bb3
97206 F20101107_AAAOEJ sundaram_s_Page_201.jpg
cef55799316064354a2d73b70e28974b
de81b007148d9a03112df453216b7947f357a11d
1909 F20101107_AAAPIB sundaram_s_Page_148.txt
0cc9c5d7040daa8288be5775294e52aa
74c05df0506927e19eaa436f4d64961fdba2f90c
2049 F20101107_AAAPHM sundaram_s_Page_132.txt
d7663551c487dfdbd36b51f2175d1f0c
c271380171f9901d24e8977f32cac51be580d542
2124 F20101107_AAAPGY sundaram_s_Page_118.txt
5654ced57ef0a7d6fb29a3b8673af01b
a8ed82be67abfec4ca09cc5f3a4cb5b196deaab7
85945 F20101107_AAAODW sundaram_s_Page_188.jpg
180b9644d9d1d5b89989cf27efa09ec4
bea9d1a8014029779a0bbab15c7d075c4a78c597
81204 F20101107_AAANYQ sundaram_s_Page_052.jpg
757cb91a00b54545a68db600c8e1e3e4
5e3f5060b8bdffed930a5524889864ce2e079c49
92277 F20101107_AAAOEK sundaram_s_Page_202.jpg
ad1076f1da95d48650c233361c0adc7a
fed34ead12607467c18fa9b57f4b8cf1e0bcad75
75973 F20101107_AAANZE sundaram_s_Page_066.jpg
29161dc3ac55cf55b7e55124eb329c07
306e19b54f4da8287bd79ba7c1ddba1a19b457ef
1642 F20101107_AAAPIC sundaram_s_Page_149.txt
ff61ca9c94bd84bdbdf0e18bf4f35d33
9ad1c119faaef9851e42e6536fe70d03791bac68
F20101107_AAAPHN sundaram_s_Page_133.txt
a31c5f17bc6ef3f26bcc939f0687e1d1
ffb70b718018c774f8bdbabf0ec9773c11e04b80
2187 F20101107_AAAPGZ sundaram_s_Page_119.txt
488c0b6fea2dc6ef249f99ebe964b544
48a50466ae768fd10b5eec4aa7c3723b4caf21f3
95773 F20101107_AAAODX sundaram_s_Page_189.jpg
70bcb310b8191e162480b1767e296e90
54e35dcb73a5dd60859dea69fd8a982e215c3960
77792 F20101107_AAANYR sundaram_s_Page_053.jpg
3fd8564a740f1e35f5cc185825334c8a
f6afae74539a5bfdee54d914aebbc4f1605c965d
306910 F20101107_AAAOFA sundaram_s_Page_011.jp2
7ec2b7b032f0c6614743667548a616cc
0e41ff8607716391f77fa736826f730230f2b970
92429 F20101107_AAAOEL sundaram_s_Page_203.jpg
61aea01bc3da7332dca1554d328108a3
c11f0142f00706fde6ee37e5d4bcc2847066e3fd
41167 F20101107_AAANZF sundaram_s_Page_067.jpg
4ef4cd771b1ee82d24f7068092f42ba2
d6a89988a0b0fa9a6a26ae9c7d519d82978049f2
970 F20101107_AAAPID sundaram_s_Page_150.txt
115deba2626a2b36130890897e445ade
09045c67b08ae640c03abaaaf778ea833200f0d8
2204 F20101107_AAAPHO sundaram_s_Page_134.txt
7e27a93e8c874e9277baa867706543d8
20aa935f5a5b7cf5b1a8bfd22fdb41d65d4d2910
76770 F20101107_AAAODY sundaram_s_Page_190.jpg
b2196596ab69d4b119572ad3a34317e4
b7f3e74f47f5468fd8165300b6eeceadbde4b670
59968 F20101107_AAANYS sundaram_s_Page_054.jpg
d05ffd69477a4bdf6294ba2441c2a9fa
be7758bcda1165f39321c3f309e7536fcba04ced
102577 F20101107_AAAOFB sundaram_s_Page_012.jp2
6fa996085320cbb8247f9d5665d0fff5
a6345bbe8f1ae16054ccf09c69f8cdfcd430b351
84986 F20101107_AAAOEM sundaram_s_Page_204.jpg
afad894ea8be17afa2c1c5e111d71397
b074c84731c50182e81a54ad8525d4cb9bf3642a
73277 F20101107_AAANZG sundaram_s_Page_068.jpg
07a4ea93bda44dbcf8a04217c42dc526
99deb2c839848ee0d1c09bc6db6bb757b503e15c
1143 F20101107_AAAPHP sundaram_s_Page_135.txt
57f9ac32e2e13b5e9007291671a4f93f
a4b219e9bb516aa66c9fb413143228f41db3ecc8
106348 F20101107_AAAODZ sundaram_s_Page_191.jpg
5b3fb8b7164040332a0c7d97ba4e5448
16054397862921f96b86b21ef749cca2fb88f64d
70365 F20101107_AAANYT sundaram_s_Page_055.jpg
f0dceb88b33c2a4b9884174f239d42f5
21201ef998e09e1faec5fd9de71353bc92dd74d3
97915 F20101107_AAAOEN sundaram_s_Page_205.jpg
710970eb77694f707ae48aa6db684bbc
2779f1d1bd16fd9168c560e548b30592a92bda55
80715 F20101107_AAANZH sundaram_s_Page_069.jpg
f75616de79ff3fb35a38ee8aac2f081c
045183e1b2b37df8a68059f4bc70523ed4b1340a
1614 F20101107_AAAPIE sundaram_s_Page_151.txt
8bb1c386ad0b4cd54949eb5e27314f17
b25e97a8119973f8abc27207f2a0fe3c37b57a5c
77 F20101107_AAAPHQ sundaram_s_Page_137.txt
63e52f03c524433d3357fe71a65e5a08
7225cf61a66f8b1275069efc4910633dae152f3b
68850 F20101107_AAANYU sundaram_s_Page_056.jpg
98f14a47460db3f58119d5bc1a8e9b4b
ec804bee89f9074bce688ed6ba240463eafebad8
90637 F20101107_AAAOFC sundaram_s_Page_013.jp2
fd146d9af80372679a2f746253dac4a3
72a9a1f94be99f0b53c17a79e9169694c9dc15a9
60303 F20101107_AAAOEO sundaram_s_Page_206.jpg
e8f0c4f54cd6daa2f8bf1b480d06145a
d6ffbd5a4131069c2c8ace3934761052a39282a6
F20101107_AAANZI sundaram_s_Page_070.jpg
60b42cbe12edb32a25c4e0a229838888
5c5acfb493080e83914c203ef4ceda87922cff18
1027 F20101107_AAAPIF sundaram_s_Page_152.txt
25e81d1c696e11c7f376d6b50ceccc0d
372639a853ef5851ecc77fa6f05f2727fe51154c
1328 F20101107_AAAPHR sundaram_s_Page_138.txt
6f69ee93835e29f0c4bc564bafebe021
3b09767381a0896b192a8ea9e54116e58214e53e
77193 F20101107_AAANYV sundaram_s_Page_057.jpg
030d7e3dce40acf67d4371d6e8eb5478
d1e3e073cb751324c0808a043c9ffca09f5e5619
114025 F20101107_AAAOFD sundaram_s_Page_014.jp2
7e73f21efeb33d7f4bff82814c28f5cc
588e8246123892b796fa0bb78d74c719a8a2da8c
46776 F20101107_AAAOEP sundaram_s_Page_207.jpg
6024b6703a743a437f91654909adaf68
8b8acdb5703fe0c7a4a942c04ee92c4929ac2e28
80814 F20101107_AAANZJ sundaram_s_Page_071.jpg
123cd240a4e4cfe16570a1c0e10a7482
a6df1e071e4b0a4d4b114371cb577c212404f2a5
2031 F20101107_AAAPIG sundaram_s_Page_153.txt
3268276c63c27b2996af2d0f5ffcafd1
978db9b69a00de59107f163225e65a051135bcff
2678 F20101107_AAAPHS sundaram_s_Page_139.txt
fd43755d44cbf11e48e60d5cb3ebd8d4
ff681ee41b2197344690c8da800982c07777bf18
112066 F20101107_AAAOFE sundaram_s_Page_015.jp2
d670efaa1f39468bf705e2d08295e890
714642dea9d7949165cfc16ffe1c91674cfd1026
25416 F20101107_AAAOEQ sundaram_s_Page_001.jp2
e2063bda13a36ce1940a870ee74d5108
865cb1448a2d85be42a65d083bda7b99d0c16c66
77788 F20101107_AAANZK sundaram_s_Page_072.jpg
3f9427df2443b3777c0d16120786418e
6f77c4bffa31dabb91fc2170c8adeb8a8972846a
1161 F20101107_AAAPIH sundaram_s_Page_154.txt
d9e9a2c775a5b5ff6685837ab3e426fa
44fd9033bfdf0620cccca3b4e9abefd6140cbc37
F20101107_AAAPHT sundaram_s_Page_140.txt
01c352fb0ad905893c6ed418070beba7
acbf29baa0c7902e4b866b4fd12c1af37769afd1
100821 F20101107_AAANYW sundaram_s_Page_058.jpg
bbe7134690a8c6a57f85902b3534ca6b
d750579bf2e89bc5a4ec3ac1cf18bfd662f00261
121287 F20101107_AAAOFF sundaram_s_Page_016.jp2
9ef3c6773e1c814e0ac0738dc71f12ca
63c954811e90a807961368f5d3bed564d4ae5b4f
5462 F20101107_AAAOER sundaram_s_Page_002.jp2
58ff3ac797e99296d03110af897d5b6e
1d70678b357b841117eb0d7ed290f5e8267b82d9
76028 F20101107_AAANZL sundaram_s_Page_073.jpg
74bac66ed29aed9073d658180b3dddb0
9ff3aac68b3481246e4ed6e0cecb870429ebc380
F20101107_AAAPII sundaram_s_Page_155.txt
3fa1ceb85517bb0c503156169904cca5
28737833e7103636ed58588dd3c1a2b942929850
1149 F20101107_AAAPHU sundaram_s_Page_141.txt
cae2ed7340fc2eda351a97084a234902
42c4bdf444542ce166adb7000ccbbd1f42acb694
66377 F20101107_AAANYX sundaram_s_Page_059.jpg
b2386e52967021872e5ce7d034d397cc
a29aa8b38c4a7362832795b26ea7147f82c366df
115812 F20101107_AAAOFG sundaram_s_Page_017.jp2
03749a33a9db8e1fbbdd7c18744dacd0
69bac4183648562bc36a1c6c71ecd7202bd2fde7
4401 F20101107_AAAOES sundaram_s_Page_003.jp2
401f98663d7f87886e541ed2625b2afd
34a56c72b7b7a7fcd5509bfbf93c573ce27b9648
76790 F20101107_AAANZM sundaram_s_Page_074.jpg
d46c2a2d41a46eef77ea30be820ba78e
67d83b66d54e54869e65129e78af01f8bb0572e7
2189 F20101107_AAAPIJ sundaram_s_Page_156.txt
64f69f272294ce636e8aead0331a5d39
d570eb91f12998ff17a9b00ed894b311f0007bf1
2011 F20101107_AAAPHV sundaram_s_Page_142.txt
92a0154b6c677631fc03f53c62858b7a
66a9b00139b6e0f4a111e07a50c163256c9a2137
78080 F20101107_AAANYY sundaram_s_Page_060.jpg
56dc00c881eb8a6190bdf3b1cfd4caef
da5b263ff9125b581f03d3141e7bab4f9a0fcfee
119221 F20101107_AAAOFH sundaram_s_Page_018.jp2
d43475c1fcc0994e30b2b9060cd1ed84
9f11bdecfcf3610b198f9c1045aa319e4081a547
120721 F20101107_AAAOET sundaram_s_Page_004.jp2
937a7141cb08d794aeba2ee87098afbf
19308b09a9cea71865900c18ffc05eaba232c786
75814 F20101107_AAANZN sundaram_s_Page_075.jpg
abb4f108002d64b800b2867771f61b5f
999631a9ceffe8eddcea2d0206c26b4f1bc63a56
1900 F20101107_AAAPIK sundaram_s_Page_157.txt
2f9fcc8f15f5b730e9508c3ecda421cf
77e7c613fe1bdb89a94ee748ccb0d94ceb17965e
1537 F20101107_AAAPHW sundaram_s_Page_143.txt
bdcb54ed857ce64269540b249aa0f302
941f9e2ead52d40a473e432e312c82227a3fdfd7
72352 F20101107_AAANYZ sundaram_s_Page_061.jpg
abb8e7482f6d2e3dea6206b460c58831
91998c0cfad509b59895d8be84db148095927ab2
126106 F20101107_AAAOFI sundaram_s_Page_019.jp2
9449e123df46b98aa9add165e8ead080
60c6ba4958179680cb2c08c370a70dc5add9435b
17485 F20101107_AAAOEU sundaram_s_Page_005.jp2
ba6eb85244bca7d0977f653fdb24b8e0
f97b0abc3607b6ca031cbc4140f736ec85e24024
74883 F20101107_AAANZO sundaram_s_Page_076.jpg
d34551190b26b09a4013276e9b69f002
3bec2d4d7da8886d9ccf19a6b3ffb9e6900dd35b
2363 F20101107_AAAPJA sundaram_s_Page_173.txt
6a19ebd5342b73435e928c802cbdc4ff
958d6d57a75e6463f7b0ed3b0af9ad3ed9332b20
F20101107_AAAPIL sundaram_s_Page_158.txt
2b6332e1679f5a2b9585170e7dce22bf
49a742ad550c797cec7af2eb2dcefaecf4954824
1681 F20101107_AAAPHX sundaram_s_Page_144.txt
4fde8d2807654e972780998725be5526
981528d17ad6aef63cf8d770af16a4d8037efa01
43106 F20101107_AAAOFJ sundaram_s_Page_020.jp2
d4b4a47b946db9e4b66bbec92e7c14dc
c4ede2f7f7e5ae92b7f256f37e9ad7332c7d0fd7
1051968 F20101107_AAAOEV sundaram_s_Page_006.jp2
42029d8c9cbbfe0776dd195ddf3c9f80
d3a03da7191535a41b1d9e065bc978e454e07fae
72986 F20101107_AAANZP sundaram_s_Page_077.jpg
13b61dcebbbadf57c2cf6055057ce914
3379ce9e62c34b8a255d3b8a9577461dd3b50c6e
2738 F20101107_AAAPJB sundaram_s_Page_174.txt
945d4e2cd94c1a7fda3bbc7a03123637
5b0fd6f159fa4efd15e898252e148c6938f4b7e2
2075 F20101107_AAAPIM sundaram_s_Page_159.txt
ed22f0d90a7566cfeaaf8419e8515ab8
1223161d45c0d53ea29c7e081cdc44e09ffb0631
1187 F20101107_AAAPHY sundaram_s_Page_145.txt
c520f87f98679abcfd56360def537a28
5451f352fa44a9fde3592a68438d7b57ffbd10fa
109611 F20101107_AAAOFK sundaram_s_Page_021.jp2
1d08ea1d83904c8bcc62aaa334b7098c
4f7b5c94079bd23055fc35275c16b8fade67793a
1051986 F20101107_AAAOEW sundaram_s_Page_007.jp2
c274b7c310327cde4c0493862550c9aa
a110cb89c85a13ee2579ae14dd295772b364b167
59323 F20101107_AAANZQ sundaram_s_Page_078.jpg
35f825c1dafeffa34f73a1b4379cd0c9
0713e2d94b75505dfddd4ca0bdf8e1335c8bce35
2927 F20101107_AAAPJC sundaram_s_Page_175.txt
f4b0aeac54b6a543817ebb381038445f
7c931e079adf2ec044bd581d57d0044face15a69
2095 F20101107_AAAPIN sundaram_s_Page_160.txt
553b1b39fd0551128f46dc78be70eab2
a221dc05dad768ef0757850f6172dda1a25ad653
2499 F20101107_AAAPHZ sundaram_s_Page_146.txt
1527640f7fd1974dacb84499ce111b44
80b14f7a6cd2cb03a2f772006bb424b9f6394095
119398 F20101107_AAAOGA sundaram_s_Page_037.jp2
9d627e7969a50cabe8f6a18cc49c158f
6d4caab7eae2f7465f3a76fed9d1110da8c7de99
120902 F20101107_AAAOFL sundaram_s_Page_022.jp2
419627148b93eb4f58f1df68cf769c92
1c7310290ba341fde42c0330e3a6791f8de65b70
1051963 F20101107_AAAOEX sundaram_s_Page_008.jp2
844f7792dec908bc9d1ed47df827c417
9962b103c8c0966f3ced2034336b0e0337c2d55c
58705 F20101107_AAANZR sundaram_s_Page_079.jpg
acdb40d8d54645e4f1f9e8c3e9eea0b3
e528216bc24375d155eeb2e2e39799fb9a4a56b1
2415 F20101107_AAAPJD sundaram_s_Page_176.txt
1d00cb31032f806fe4b243b449ebef50
6a7a759c12e68906d56d239e09403fe3e28a177f
2061 F20101107_AAAPIO sundaram_s_Page_161.txt
61851421619b79b94e4ce36bd81741a0
d3661fc38bbce5d2c6b5706ca6ff9de105ac510b
125461 F20101107_AAAOGB sundaram_s_Page_038.jp2
874063f6da55919af9505db7b7a2c9ad
411a3ae62f556b4c0fd006164c81d09e1b72bf78
122765 F20101107_AAAOFM sundaram_s_Page_023.jp2
6d44b7ec7e1da9917b086c4a973349d9
7728429993e90ba94b291aee81e7ff51a48e36d7
F20101107_AAAOEY sundaram_s_Page_009.jp2
32cb85d9f49246ebd22ec7e08fd08c79
3479f1f1b4c678ee999ce7c488d7e9265c6e439d
52564 F20101107_AAANZS sundaram_s_Page_080.jpg
ec50a7afe5651d80ffb8abae020df83b
9c9ea59f3ba49ec3150e2bd1a2cc628c91081c1d
F20101107_AAAPJE sundaram_s_Page_177.txt
a8639fb2e9e7cdc0b0ef1e58d61ae6ce
52c31eca3575dee7a54d3bd0a00d49be98203511
2085 F20101107_AAAPIP sundaram_s_Page_162.txt
179ce87ef0c9a28f4f205fefb8966985
02f66b69fbba19b37eb112ba96fd9dd0834d1d2d
124369 F20101107_AAAOFN sundaram_s_Page_024.jp2
29bd006bd0b2756fb6ffe6727b89d6ad
8dcff022b02d944a50f6fc198e071d9787b0c900
1051982 F20101107_AAAOEZ sundaram_s_Page_010.jp2
5e46056e3d828be95063a4a065db9892
00089342c76ba1c69ddf914886c1060d568b8975
50642 F20101107_AAANZT sundaram_s_Page_081.jpg
0a2a049525dae556d1c15d05f414399c
afb4d74255dff1c9203104bb7f49d3a193bd47ef
127301 F20101107_AAAOGC sundaram_s_Page_039.jp2
334f86709576d0e19fe3ec18f4b1626b
1be1f3e883536b81bd7daf379b4d33d1c90dd878
1403 F20101107_AAAPIQ sundaram_s_Page_163.txt
d3ffffaf2fb4451521dd3832b37ec703
726b61c734b618bbbde2cd678952d6d22e718816
118290 F20101107_AAAOFO sundaram_s_Page_025.jp2
40cef2b4223eaa53d52c762a2b205f40
961c9bb3c68999d1a55171617f83e0355edd50da
62818 F20101107_AAANZU sundaram_s_Page_082.jpg
e3e9623a3472a15758a7e237b06cdc24
cb2955982f9ed2df834064b629fd91854a43ce66
F20101107_AAAPJF sundaram_s_Page_178.txt
5e29d405319957b9fc1cb8a51b3a1e3c
86866407eeb48ae270333ed23509b3c071f0a66f
1339 F20101107_AAAPIR sundaram_s_Page_164.txt
8a12ee3499600abaa5d324961f196f52
4347634ad75ad9e7e04d06bf4bb39753128e93a9
128865 F20101107_AAAOFP sundaram_s_Page_026.jp2
07d14c58315e36b7faf217babea7aa67
588a3ea0001075f449220942602bfddf66fd17dc
34517 F20101107_AAANZV sundaram_s_Page_083.jpg
038991f9e117cb2c48b8485b6a39164a
a60d6e4fc77864f4557f158563906a87c583fcfa
120610 F20101107_AAAOGD sundaram_s_Page_040.jp2
34feedfb8a9c4044e8653643896d1b66
070f7c29b2e5c71c3a631dcb430dfe9e54795ca9
F20101107_AAAPJG sundaram_s_Page_179.txt
28e6dffb23e9dddc7676a106099b6b5e
e9840be6ee9e18ea363dc82656dd9ae7f21d5dab
1353 F20101107_AAAPIS sundaram_s_Page_165.txt
9b916f086f828be478637a003b48af8a
f8f8f536bcfa953b963d81bc28e3609245222c89
124922 F20101107_AAAOFQ sundaram_s_Page_027.jp2
ac2742c4cf49c7591e929755b32183f6
d353ea04b14934101c28874b739208f43ec2818c
36679 F20101107_AAANZW sundaram_s_Page_084.jpg
e777c89bdc049dc83aad91f4ea8b2708
01f604a2efd5c173bad536bb1f52167770739767
123346 F20101107_AAAOGE sundaram_s_Page_041.jp2
22046637d2490906b1d5665f1e27a687
f1b4f58859ea29aeafb60f2bb7f19f872bd725a7
2153 F20101107_AAAPJH sundaram_s_Page_180.txt
c69198a690410920d69964bdf8995f58
64fe32977f6fa908bf87b865c439358bb1d114fc
1309 F20101107_AAAPIT sundaram_s_Page_166.txt
072b72a6fe7e1eb1a0a57b7cc961f747
85778b4206d34de8af1efca2e84f6c603b752afe
121968 F20101107_AAAOFR sundaram_s_Page_028.jp2
0dd43248380361825d261af4b7a5ba1f
71ff91908048d6942b24823e21bd24d1ae72d170
123922 F20101107_AAAOGF sundaram_s_Page_042.jp2
e4fe49c0a7d39e893507ba0118e1931b
ab1f1256f8a3a7413e153c6081e4bd5ae16d2702
2860 F20101107_AAAPJI sundaram_s_Page_181.txt
292cb8835158273456f53649c89be759
b65807dd070b4f9a611377934a81a8f92d84f5de
1300 F20101107_AAAPIU sundaram_s_Page_167.txt
52e5490105a6e8fdce89ddef21f54ea4
363583b5c25dfdb54200320f0c89fa7b36ea9039
118754 F20101107_AAAOFS sundaram_s_Page_029.jp2
a77d41705a82496c8179d049070e40f4
27d5353e68d0509901f3032d69985d7843f14cc2
34185 F20101107_AAANZX sundaram_s_Page_085.jpg
c5272249e4f64913202467ad617fa6a4
a35330b64dbe62f2d31563c41e2dd764c6ae42fb
124464 F20101107_AAAOGG sundaram_s_Page_043.jp2
71b1d3155d157ec204f5262dd5f333fa
de17ff761e8a3f6b9dbba05c14e337cba8912092
2154 F20101107_AAAPJJ sundaram_s_Page_182.txt
2a37b2fbd130d9ec57230dd55adf2c95
2d2d2060df797160b6768fb7c6ce6920d2935f1f
1370 F20101107_AAAPIV sundaram_s_Page_168.txt
63149299791ae846b6e468ecc67af6ac
46e679e02f8b80733b56b5b992dbed983a8733db
122894 F20101107_AAAOFT sundaram_s_Page_030.jp2
72c022e0737569890396ca35f40a116a
310d92d785cd4e315d43caefe7ca81fddc451006
33722 F20101107_AAANZY sundaram_s_Page_086.jpg
584cca25e2cb4beee6d119475f370b07
631be8c2864ed32a653c6e6f5657ce6a62a65016
122744 F20101107_AAAOGH sundaram_s_Page_044.jp2
ecefa77aac240c2e343d972929c8a7db
d4492a68ca451aa27e09880fc9ade0e2e078ad70
2296 F20101107_AAAPJK sundaram_s_Page_183.txt
65129683f02578dfcc53b18e588e5848
ba31ce90e31fbb2c785c05eeecf6ee3b1e1d9b9b
1243 F20101107_AAAPIW sundaram_s_Page_169.txt
7e202e85f31e7d5fcc8d2473db3d8bba
09fdada2059267470a300ba07432eb185ac7ae4f
114978 F20101107_AAAOFU sundaram_s_Page_031.jp2
1f3b46f9876d9a2cbb20c9ac7af58c51
b72e4c47f4988827b8669400b9605e330a49c529
32158 F20101107_AAANZZ sundaram_s_Page_087.jpg
3a16951cfb1a66d2487ba6a65af098ae
13c0ff5f215f5eecf45c9bccc6f6818b9737f0d4
115701 F20101107_AAAOGI sundaram_s_Page_045.jp2
d0ca9f14c9f51c0601a53695740fa618
3523427feb996838795b495c60655dd9110532f6
2768 F20101107_AAAPKA sundaram_s_Page_199.txt
94409d1f1210df4560857325f21c7fe3
294343c3850252b7de28d2450fb1835096929084
2114 F20101107_AAAPJL sundaram_s_Page_184.txt
29c63462e2fb9b26de9de1e16f26b83d
bd140675bbd4bb44d650bae9cf3393fc6f5cd245
1904 F20101107_AAAPIX sundaram_s_Page_170.txt
463813d2b86e281bbcf7d5590d49fcf4
8a00241262a47d74a9be68bd6177b3696e90ca84
126910 F20101107_AAAOFV sundaram_s_Page_032.jp2
446c0227548450aa14c28a7489f26d58
c6488a1089985c4b3eedf9635f927774712f9e83
123127 F20101107_AAAOGJ sundaram_s_Page_046.jp2
2e5c1352533c59994bfa358dbcd333e0
18b1c9027670f191ff30c37236b3ecdf0b7a1306
2601 F20101107_AAAPKB sundaram_s_Page_200.txt
7fcb7bde1591a0b9dda9868036a6bb1f
5189b07e0093cbfc0def27bfdff6b5abc35d1537
2354 F20101107_AAAPJM sundaram_s_Page_185.txt
68e131c6610c855b27673a4d104b41e7
c8688c901b2ba4d4af8098881c2250b9f6a6311d
927 F20101107_AAAPIY sundaram_s_Page_171.txt
66476cc6a6e256d6185d23644815500f
6516135a3842bfdc768f9f111826591520698518
109467 F20101107_AAAOFW sundaram_s_Page_033.jp2
08609c1735865892815d6bb6a02d7b13
f6a94e7702f504e737702c037aebef10716f3888
115165 F20101107_AAAOGK sundaram_s_Page_047.jp2
44781af2ad1ac703a606769ede4d5263
b456fcd1a54b0c10fd1612a9375664c6ebe6a4e1
2647 F20101107_AAAPKC sundaram_s_Page_201.txt
e3030d220dbb6105b566b5fea499c84f
e356d79a6311f787fa03eb7218c1d8ff8aee899e
2512 F20101107_AAAPJN sundaram_s_Page_186.txt
94ddb8adb112a39b483f4a2345ab9ef4
efffa8734558ddc615c8693e3adbb41707882c86
2852 F20101107_AAAPIZ sundaram_s_Page_172.txt
4ae9638a731370710c1ddd7cc0c235ef
e085e61dfb31a7d8c628b5f111fc7ca64032f99a
121152 F20101107_AAAOFX sundaram_s_Page_034.jp2
1f42b8ba9903fdfcadab2eb6db5d3f4e
fa70dbd0d8f994279a65af833bfa8d0b2f645dd8
99490 F20101107_AAAOHA sundaram_s_Page_063.jp2
55a9fd6790d0cf02b440d89a26614f52
f6c1057216f3e829e39d494b72430d573873c8e2
117340 F20101107_AAAOGL sundaram_s_Page_048.jp2
a524142ee4b1bc5038390251ee45f64b
a23b48dce89c4cf1384fce05055625696eca8539
2685 F20101107_AAAPKD sundaram_s_Page_202.txt
3bf58159e30ac2299f9ef1a40a1ec150
29dd3ae0bae859c434f8c635b2ee21115fc78c77
2635 F20101107_AAAPJO sundaram_s_Page_187.txt
a23beb39c317c2f2f27219f3c8ef6d23
13b9fad94a6cdcc84ba9e3e34fe9308f8ad79e2c
123722 F20101107_AAAOFY sundaram_s_Page_035.jp2
f446398211153a4e273c8c0631de6e2a
2867314cc974ca01b786d1687883bc26776f0926
93800 F20101107_AAAOHB sundaram_s_Page_064.jp2
c0e3ebd6d2e87f77f4fc6b901ffe649f
5afa9a109e65e7fcd8a608a51835ad21c18a9deb
45384 F20101107_AAAOGM sundaram_s_Page_049.jp2
fda88541b79b05ac581f48fbb68cd694
92d9d41cea9cfc383d0462a2cf5001d6e5128c2a
2402 F20101107_AAAPKE sundaram_s_Page_203.txt
a5b6016d43bf9ff8c5b2af8f44506e0e
179b5f3c410b48aa37742104d41968a95a54cf48
2519 F20101107_AAAPJP sundaram_s_Page_188.txt
51af5215a5f3f9f43e596670a395bb32
6bd4ecd1c13a0efb59734b94af6da56bceaa8efa
123131 F20101107_AAAOFZ sundaram_s_Page_036.jp2
dc360bd1376982ec838e2de61dae7da3
d0ae601f04b45c2fed4a8396b70564f0e368762d
63091 F20101107_AAAOHC sundaram_s_Page_065.jp2
72c1c67501ffb3e58714c96714599afc
19c7526f6c1bbdedf5b4b15e9fbe24fd91c0b8c2
19248 F20101107_AAAOGN sundaram_s_Page_050.jp2
49f97bab32fa08b786dd29f548150563
4813cf508c799e7f9570b1e56d637ac794d95b8a
2587 F20101107_AAAPKF sundaram_s_Page_204.txt
774b77a457302dbdb9971ae083e8592c
45cd55ae6ccddd138ada045a4b8a66d170fdec01
2606 F20101107_AAAPJQ sundaram_s_Page_189.txt
96548e20bd5af37eed019db1ae02b465
6ff1b88c560966999f417a1e7109ff0d92d4697b
106188 F20101107_AAAOHD sundaram_s_Page_066.jp2
ed9ef3ca10cfd49c03dade0a64e4829a
c92ee0b0aa3c7dfc55c8204196feb81d78b29b7b
106979 F20101107_AAAOGO sundaram_s_Page_051.jp2
0172a9486817ad68b947db2e34d48926
644ce51dc5d4e9348e9be95d9ea93345351e3df9
2314 F20101107_AAAPJR sundaram_s_Page_190.txt
6c3dafae8ad2ddfd514391a218114b3c
c16b4c2852acfd8ae689dcd85db4ca06fe6a47fc
123833 F20101107_AAAOGP sundaram_s_Page_052.jp2
5d434c3119990b03347af022653aa87a
6917238a639fe1781473cd7bf61035d2ac014f96
2676 F20101107_AAAPKG sundaram_s_Page_205.txt
317017b94062ac7e34566f5a5ffeec61
6ae967610620e34de5cbde2db7a59981050b362b
2736 F20101107_AAAPJS sundaram_s_Page_191.txt
4ac5a6c0ed2c9b3bee7bac3967281af7
e979929fed66927b9a32b0906900a894b620c0ad
56540 F20101107_AAAOHE sundaram_s_Page_067.jp2
8ceecc9ceae53c276d7851f2b38f4aee
de74ef91cfa5ce5d6fb0b47657ef8d9203190a8b
118730 F20101107_AAAOGQ sundaram_s_Page_053.jp2
bea5cba5088f2158cd1e93b747f243d6
2f9029f239808b27cb6dd5013ced7007d7930c38
1801 F20101107_AAAPKH sundaram_s_Page_206.txt
2973004eabf3cab6fe81e918bf36474e
9d3250f8253a733225eab7daffe53f8cbe65acbf
2468 F20101107_AAAPJT sundaram_s_Page_192.txt
ad44d233e2c5aae3cbfdf2d2e40a2905
ea0b3a5a4bb63603d6025c84cd02846fdc782b4a
109525 F20101107_AAAOHF sundaram_s_Page_068.jp2
fd7f1a527b1349ed77802ca96b5069b5
c520798abc3053901a82c79783deba1b913d5af0
88023 F20101107_AAAOGR sundaram_s_Page_054.jp2
8b3e23e80d87a876dcd73c4b1dfc9cf1
b18b9ebf6c2fe22c6fb3855bcf488806fbd73dbc
1219 F20101107_AAAPKI sundaram_s_Page_207.txt
d38c788d3a5da4dcf151be1b83d2b359
cda5d48efa57301e10b97f6c4a046f5f6a768d1f
2733 F20101107_AAAPJU sundaram_s_Page_193.txt
46a70bb60406ebed2c0c9884ad794a65
7bf26b4e5a3b98088ba29dd47c71d1e1b6b54bc8
122730 F20101107_AAAOHG sundaram_s_Page_069.jp2
62065cc275070aea6e15729cdc2c4be6
92650f6f92877ed396537692ab107e260be0ee0a
107281 F20101107_AAAOGS sundaram_s_Page_055.jp2
8d7ed297d4580ff548807fb888c700f9
d73565b7f29c0b39123189071679dbe05868e39c
2361 F20101107_AAAPKJ sundaram_s_Page_001thm.jpg
bf00786b6cc41534f1a92a2292409eda
14b5f6a601df35837e01bc047e33235fc9c10513
2761 F20101107_AAAPJV sundaram_s_Page_194.txt
3796e34c2e71a3dc2e91b4d130f0d508
0422c8b67c96c7d2409f33bd53e3a91439c87b6c
123013 F20101107_AAAOHH sundaram_s_Page_070.jp2
5944774463c9fddbde90789d6c505c2d
fcb035426ed880c947ff366535264ed733980457
104621 F20101107_AAAOGT sundaram_s_Page_056.jp2
e77417c3da70c7c515b4c628ed37ee13
a7afc1089c2d24a2fd0a2ecfd065b1684533b74c
839300 F20101107_AAAPKK sundaram_s.pdf
57bd1ca83dc1d13d6eaa8a057d647ac1
a5ef30e2fdf56b50c1b4ce89776c48076d5ccd7c
F20101107_AAAPJW sundaram_s_Page_195.txt
86023610cc97a74ad744dbf8522c9188
b5ee4eafafdedc8f588d58dee96282cd3ab8b0cd
124768 F20101107_AAAOHI sundaram_s_Page_071.jp2
fd8d77b5a54fe788c5ccca7a4e26d667
1943d9ee9f101b35aad475b038c4832923eaa929
115931 F20101107_AAAOGU sundaram_s_Page_057.jp2
57ee3f5fba3b82d73da16866b503ab24
3ecd54dce4d1aa967ac1047af1d9a314b7f74a94
7091 F20101107_AAAPLA sundaram_s_Page_126thm.jpg
25ab7010f5d82f954c6220be38ff58cb
8c301a92ae33e3c0eee4c94f0dfdfbe65e6e5d74
16432 F20101107_AAAPKL sundaram_s_Page_080.QC.jpg
0a1bd3b9f2a962b648d8e25abae6a056
cf10e261a4c7022d821688915b0ed27f8b08595b
2926 F20101107_AAAPJX sundaram_s_Page_196.txt
efef35d84303f35eb2a34961c1bc7b36
548e1d91241589e55637325ff893959bad296dbb
119572 F20101107_AAAOHJ sundaram_s_Page_072.jp2
803f679b991895f42907d064a73370f5
0bf71ce7597404ded32c43aeb4411921310dfbc0
150280 F20101107_AAAOGV sundaram_s_Page_058.jp2
16fc5291076ab5baaac67555a9aca3f6
ed76a0401a2fd02ac392212f543486c0fc7c31d6
5260 F20101107_AAAPLB sundaram_s_Page_081thm.jpg
90c2a8f6dd3f4b0184c7f8314b933c66
40522ca808c73e4312d5c86e34e48344fe962ab2
24549 F20101107_AAAPKM sundaram_s_Page_132.QC.jpg
8bc09d0ae85cf71c010d68cf66f68b01
2f3c82d5680bd685640d84eb125f2aaa30d68a26
2722 F20101107_AAAPJY sundaram_s_Page_197.txt
657ae52aff5593456a02033760bf6c58
3bf7ea16f7b49dc5d42b613aecf762ca83d78be4
116871 F20101107_AAAOHK sundaram_s_Page_073.jp2
fe0852f9ad0fa9f080a29335fa69a22f
6f7baf4b5521e046ff6464f539bf9b95c44597b2
100069 F20101107_AAAOGW sundaram_s_Page_059.jp2
034f9de4cccb7d39cb539e1464661c59
230c2f5946ada1446d32fddd06e036b9f22bafc9
8982 F20101107_AAAPLC sundaram_s_Page_138.QC.jpg
f7873c852e224011cc8db488620cfa1c
b611a456cdf2a258891758bc9e44505744fd3ecf
7890 F20101107_AAAPKN sundaram_s_Page_139.QC.jpg
3c8d5855d81760bf668a342dd6711905
923317d97309159d8be859a2615580888dd5316a
2817 F20101107_AAAPJZ sundaram_s_Page_198.txt
7c37d25ed5b0b8419560fc7eb31a97c3
924da4a468c7996d2060f07ea4df4f6404841f8b
108077 F20101107_AAAOIA sundaram_s_Page_089.jp2
cf67594c92e6dee6ef250a716f6150b8
b2f276baa6b4f8ce693c0cdea299a08031af23ae
118506 F20101107_AAAOHL sundaram_s_Page_074.jp2
37416c136a940eb2f794dbf6455f6163
1fb5df3ba107d67a0bf9936bc4eba7e4f34f5ac6
116761 F20101107_AAAOGX sundaram_s_Page_060.jp2
f11f6dcc632f32192a60f473b51b13ab
bfa27f68d54b841a375e2d9b58ef49956f96d5ce
2380 F20101107_AAAPLD sundaram_s_Page_050thm.jpg
bf20347d2378c52b08f44d329ac70cc0
b1ffac36afff4be75e4974fad8314359a6005f4f
1437 F20101107_AAAPKO sundaram_s_Page_137thm.jpg
750ad528eec6b0b91f437118c9dc097e
b3d1c62af51074a7b2f96ad47c795f7bf16ec3e3
105346 F20101107_AAAOIB sundaram_s_Page_090.jp2
015262d46fbbdf7c0a12011ca1f236a3
1083dbd541c4507855a15110f3f10ec94c51d531
115402 F20101107_AAAOHM sundaram_s_Page_075.jp2
fb04512caa5b75849807f797afc80f91
9d3660b8f3972e9753be46cde3e19d7c51659c29
110864 F20101107_AAAOGY sundaram_s_Page_061.jp2
9ce1ad1ccead5d95d5b3143386f8ce77
00396363939c50dd086bd8f13342e03156e0a236
9563 F20101107_AAAPLE sundaram_s_Page_147.QC.jpg
c1829c9fbde23a3be7f661035a04917e
eda78d4f73016fec2c8bdc587d760f78906bb3f0
18932 F20101107_AAAPKP sundaram_s_Page_082.QC.jpg
1e303c696739c2dd6714fd62550d3e15
9f66205772fae2f13dbab64917ecbb84593432de
98667 F20101107_AAAOIC sundaram_s_Page_091.jp2
d18560ef24ca341bc85037ee609f5780
2380011fd29c89cb193b81ec523a67b9eb89c5ea
112462 F20101107_AAAOHN sundaram_s_Page_076.jp2
fe1a53f6c59cb26f7d337bb9912dc835
e359c7d1216c2ceda872004bff868ee96aa1e85b
109500 F20101107_AAAOGZ sundaram_s_Page_062.jp2
5aaff89988704ee10554f84c00678cd9
99e472390d7fabf94f14fb158bebaa6ced5b088f
6953 F20101107_AAAPLF sundaram_s_Page_014thm.jpg
e43f35394518c5b770fa4d0fd0ec033e
ff5d8848b33b4b134d13e140ad8329be59a20584
5551 F20101107_AAAPKQ sundaram_s_Page_155thm.jpg
4606eafcf8c79612575d35f2bd63cecb
a329013d94d01904182a652979ef0a729a384967
89794 F20101107_AAAOID sundaram_s_Page_092.jp2
3deb46c2866d41c2106c68b6b865bf71
3b9852f7135fc2919d9cdbccf6d70514c4d642f5
110601 F20101107_AAAOHO sundaram_s_Page_077.jp2
ae0738aa176adf5c53cbda0ff9b5d1aa
8fe3ea526ddcb7497c092acdd92905e0217a67f1
25623 F20101107_AAAPLG sundaram_s_Page_019.QC.jpg
7c0c3236d20bcd5e9559db035026b36d
49facf86f4b14e03b631a9f8063f311de5870d1e
26214 F20101107_AAAPKR sundaram_s_Page_115.QC.jpg
0fa3072aead5921e5c35f5bdb54ef319
14218588fca1d4270af413a31ba425a1240b1a73
105096 F20101107_AAAOIE sundaram_s_Page_093.jp2
d26090c4215baf36ac5ba179c7af3645
9a26adbff9a02e9f4d1b0007f340e68b7e334138
84303 F20101107_AAAOHP sundaram_s_Page_078.jp2
ec59c9d2fda18f1eb3429e04789b712d
382d6583ecca5367c6bf8cbf91b95fca221e1f83
26415 F20101107_AAAPKS sundaram_s_Page_123.QC.jpg
a4b286550245aa7bbf8efb55758c9215
d66f2729744d4adb8687fc8022287ee381772ce9
88812 F20101107_AAAOHQ sundaram_s_Page_079.jp2
1de0ed099033335b49de2780ca5970da
2e6bfb74c54e470b2dbfb9050b822ff5f8fbea43
7273 F20101107_AAAPLH sundaram_s_Page_023thm.jpg
e0ee502876a7736bce88d6075ae054fb
d4a8abb24065f61193e20068108a8e707d5b54e5
6697 F20101107_AAAPKT sundaram_s_Page_190thm.jpg
d700b850ac91888ef51001953c660043
3c7bbf7fe9f3e7e6474f1870597abcb18ae4ca2c
97636 F20101107_AAAOIF sundaram_s_Page_094.jp2
62db982f8bf9a580a58b3e22778c4b52
1187d4df69eb598bc8999a5be4d2f7fb24d1d327
78813 F20101107_AAAOHR sundaram_s_Page_080.jp2
7b68b6ad8724ebab8053b3f40d19bf1f
815b71248988765ee5e6b1a086a86942bd5ae238
7316 F20101107_AAAPLI sundaram_s_Page_189thm.jpg
51b62469b5a7d877044635f2e26a8a5c
b860ddd580f0e94bd8db7910680cad766c507919
7239 F20101107_AAAPKU sundaram_s_Page_027thm.jpg
4b9482a61e2b9f26355631338780cca8
d22e6af34dae0aaf18607b397b4b3e50366a3327
121748 F20101107_AAAOIG sundaram_s_Page_095.jp2
24cce05a572b0930d2b3369541dee2b4
ef559ab63716e9629b34389d5f921660d1930032
75482 F20101107_AAAOHS sundaram_s_Page_081.jp2
dffce4f18c70eacd423f6cdbf06f34d2
b46c69bb269a601ea6efb0335127dc51709fbc2b
6013 F20101107_AAAPLJ sundaram_s_Page_183thm.jpg
18f32dbe89a7b2966d381bb90e5f558f
a017cf627aec86d87deb5a6b8b689a0053678262
5739 F20101107_AAAPKV sundaram_s_Page_176thm.jpg
f0bbb1bf0b2ce5d730e81d988d63e41c
e5563de485beff21b6b362729c6479439c35f99c
112175 F20101107_AAAOIH sundaram_s_Page_096.jp2
f12ccfd0810175246407b20c3205db7d
efa6b24ebde1b6f1887492ecf6751844b3b92713
93790 F20101107_AAAOHT sundaram_s_Page_082.jp2
b4c62334330b087888df9791c808613e
2ca2f87250c9e8df2c081be223d43abcbb4551b6
2821 F20101107_AAAPLK sundaram_s_Page_138thm.jpg
3cf190c5a95c8d0f901327db6130ce94
e41955a8c5e37209597d474471b5e4b88433578f
22801 F20101107_AAAPKW sundaram_s_Page_012.QC.jpg
51d4b95c441b95e27aa1a3be6af9a844
2cb12dce3463fbb2ac26cb460b9f31adbdfa1f27
110252 F20101107_AAAOII sundaram_s_Page_097.jp2
730852d05dea6a76e00b6e8ceb5821f3
5d54ddeeef87d794052fa1e16ff5ba23fa63f743
79614 F20101107_AAAOHU sundaram_s_Page_083.jp2
72e0652f863fdfe5a5279177df91dd1b
5d7550745cd81e6b2c7dd535b1f427a05d744c56
9764 F20101107_AAAPMA sundaram_s_Page_144.QC.jpg
b2f5723e41fa94c3563380b246e6b1a0
2bb79b836383f2d1254fb4764f46d1bc47c91bff
2975 F20101107_AAAPLL sundaram_s_Page_003.QC.jpg
10c1c640788c393b90f0b85ddc962e34
7d99413a0efe6a898b853f18a86de9eab8fbb09b
6870 F20101107_AAAPKX sundaram_s_Page_047thm.jpg
7711a9d012f4af85e86d87a2336905bd
5533f8048821e6cbb29625e3499649e1c2f8c765
123259 F20101107_AAAOIJ sundaram_s_Page_098.jp2
b4beacf42dfd0d8f61553e12810d262f
02de9c2d359a71eee59f89361540484233e7b236
89981 F20101107_AAAOHV sundaram_s_Page_084.jp2
89a0c26191bf576b297b7abee8f35f64
bfe33c3e8ec340e76a2a3c04dd2b33a8549ea1b4
6107 F20101107_AAAPMB sundaram_s_Page_099thm.jpg
356236fd70dd9b1f1a2729d6b5160f33
380e7da779eaccec24b9b4d308a0f21a0d9c825d
5857 F20101107_AAAPLM sundaram_s_Page_054thm.jpg
ed600b0810c22eabe9a47461d0f49d37
1d7e71bd8cb2cb1e9dbe8d953056180954cdd864
3970 F20101107_AAAPKY sundaram_s_Page_098thm.jpg
f76077c7da2d5bc76b3182f42a09aa6e
d7e9c6551b284e6a81847287e16b1133f2211e84
115785 F20101107_AAAOIK sundaram_s_Page_099.jp2
f91d15babd44952abb3b4881b3c53355
98a4ee9fc77d4b8f739ca97321c5d0d8fc3723ec
80884 F20101107_AAAOHW sundaram_s_Page_085.jp2
01a0a601b21626a44d6ffaa9b3c62297
eda03028c39c55d454ab682068a34747a3d32369
2836 F20101107_AAAPMC sundaram_s_Page_151thm.jpg
a1bdfa6dda28d5677be7c9224e1a8c9d
311ce843a570f02bf66c17d3b60034b5e74918eb
26217 F20101107_AAAPLN sundaram_s_Page_133.QC.jpg
135881c0abb0ad41bdc37dedea45a71f
4340fad44d478083deb6701b341214df5f873940
7352 F20101107_AAAPKZ sundaram_s_Page_205thm.jpg
efb483c35d059e1805c902f204865043
d5f4a17363185db6ca14417b35f5dd4b47753bb5
125431 F20101107_AAAOIL sundaram_s_Page_100.jp2
4c8b9fbbe63a8df9502ce64ddf9b7004
11279f8bf0ff3a7e74e0d3d2d6171ff419257025
82592 F20101107_AAAOHX sundaram_s_Page_086.jp2
456b967dcb6b91d4c0dde4c18bb225c5
c26c4bd7d06561ca999a2155a0fd18c956f20a25
120678 F20101107_AAAOJA sundaram_s_Page_115.jp2
bcd01042be7f5f011a84afcfa2725a52
825205b5380d8bf45fc69a00fba2648bafeb235e
26995 F20101107_AAAPMD sundaram_s_Page_023.QC.jpg
443fcc2bcad410b5dc755658ec4da6a7
1ac980516a99840d1f3aa1f554695c4ac6051f74
2957 F20101107_AAAPLO sundaram_s_Page_143thm.jpg
9c08993b4a6d252936bcb5df2fb0760e
2f8180f773972ad442c6367f26bf1abfb3f23266
120510 F20101107_AAAOIM sundaram_s_Page_101.jp2
29168b5420e1edc9a27864214b721009
1a8337f52483f0d8d60444cd1b927ad38cbcc79d
76107 F20101107_AAAOHY sundaram_s_Page_087.jp2
26f98391378f5ce9a4cb36f7435d9f8c
132ce502214ebf33f3d53abb93aabac1660dc654
119190 F20101107_AAAOJB sundaram_s_Page_116.jp2
713715d61804f630daae0718d56468d0
e972efe7f669898e0033251eceec076923ec9506
7584 F20101107_AAAPME sundaram_s_Page_197thm.jpg
0b9b90f7b6d2a16eb14dc70a00e4940b
f4b6c7c79103aadab7a34400bf4b39903013e149
26284 F20101107_AAAPLP sundaram_s_Page_040.QC.jpg
f185787e588f469fbec2001f9fd458e5
3d8289a20d806ec9f0b0b5325a3380edd41ab382
115726 F20101107_AAAOIN sundaram_s_Page_102.jp2
100b96febb6e90cee1334b308e963b68
1ea9c582c5205298e9db7f0f7509e551619b8982
98694 F20101107_AAAOHZ sundaram_s_Page_088.jp2
f9c4b30bb050d00884e267319d2433ae
4feccf41463ff92ed7b5e1764dd0efd4455e32a1
114531 F20101107_AAAOJC sundaram_s_Page_117.jp2
b63e46cbccf51b842adbf30aa621ad64
493fdb3fd89af349cd86be2635e10e5bd6de8815
21502 F20101107_AAAPMF sundaram_s_Page_186.QC.jpg
6f90d5fb9ec3b77ad746139784791323
da1038bf9d740e43b372e12991b71559be11b556
23843 F20101107_AAAPLQ sundaram_s_Page_156.QC.jpg
575315a8356dd51c1a18bddb921cbadc
0d8d545ac9313cf4a058cf219d022663a00c30fb
135321 F20101107_AAAOIO sundaram_s_Page_103.jp2
ce011b0ace6170caf9e328cadce3dcfa
35651824aef80ab5064c34a6e1a79315bd36ea42
118663 F20101107_AAAOJD sundaram_s_Page_118.jp2
4d34c1795e9657efea22ab8c7243a8a2
007f8a5dd171e86d5c937a5edc105b78bc05b2ff
23085 F20101107_AAAPMG sundaram_s_Page_176.QC.jpg
be9538fa3113c0e3d726bf5ddaaacdf3
3cf6cdc0ee502fab403a8651812dea8cecc9a9d5
18058 F20101107_AAAPLR sundaram_s_Page_079.QC.jpg
0d88d58a19c40181799d49c8b6fb03cd
159edb678875171ed7b55da5e98a0df3be992ba6
117411 F20101107_AAAOIP sundaram_s_Page_104.jp2
5492ef1597e7a9364e90389be78b5ec3
4f3052392f5dfca92ec233c75b755c932bf07b08
121609 F20101107_AAAOJE sundaram_s_Page_119.jp2
a4f2018a859f31455f849d97e0c14de2
22e49819453bdb83b903a8aa957fa37cc97d5620
6813 F20101107_AAAPMH sundaram_s_Page_114thm.jpg
8f969019b66dd5fe33a128cce94debe5
5d0e8a9cafbbd3fd7433294fc07ee64bdfe853cb
26091 F20101107_AAAPLS sundaram_s_Page_103.QC.jpg
93562a60101ff762737cb3a6563eefc2
2cb1838383857181ca710e3df0d6b794e4085af6
120965 F20101107_AAAOIQ sundaram_s_Page_105.jp2
2c8e86402b982bb5e7493e4adfa7c919
d281c6ff0246bc030c37ac5065025e02ba4b969e
121757 F20101107_AAAOJF sundaram_s_Page_120.jp2
30e10a20e5aa051eda3790cf7d32d1c9
7341d24661b25b3adfa14c4e21ef5b3198499037
27928 F20101107_AAAPLT sundaram_s_Page_205.QC.jpg
710d123d0a4922e7ee5dc805274c7933
5a8b335aa0381146e340e244c440bf0fb28691a1
127250 F20101107_AAAOIR sundaram_s_Page_106.jp2
61197c67ca7787ce2d0d01b702520a17
fb40bb69d3f020e82e2432e21a37acb8bb34cd5a
3425 F20101107_AAAPMI sundaram_s_Page_084thm.jpg
e26678314ccc5bcb5fa7a018898d2236
935ee53a03b7adb869f7b1a76a894d73504b0955
6873 F20101107_AAAPLU sundaram_s_Page_045thm.jpg
f0f3515ad3b9cc9491583a862d85ae35
2d7069a4a9bbbcfeaa8dd13f53417a5f3edf5823
122465 F20101107_AAAOIS sundaram_s_Page_107.jp2
a452f995b0b509988776542785cdc97e
02d411d9311acf85b58e716bc7340913ae75efe4
122217 F20101107_AAAOJG sundaram_s_Page_121.jp2
b0b7b6379f25b5f7930281f2ea7e20cf
0f6b1c33bc4f437b1d673d09c17ddb2a3f3328c6
F20101107_AAAPMJ sundaram_s_Page_022thm.jpg
58b621b8a3bdbcb03bedcd15c7b3361f
f85edb2602bf2b3c81380f9b1397ef597d1a9b12
12062 F20101107_AAAPLV sundaram_s_Page_090.QC.jpg
017eeaaefbd244ceaea65151feb3112d
659d2b0dec63c82c3cb4c66c5f6fa9a0c9fc683a
484999 F20101107_AAAOIT sundaram_s_Page_108.jp2
b29408a438d8a93580aac5ef1a2a0483
e8e3a9cd4048adf714734d8283e49f621ba5e2c6
114689 F20101107_AAAOJH sundaram_s_Page_122.jp2
d42f569fdde5ad4bf109597235b204a2
47e6b6483728d5148240f51f1e07a1f0042b7d32
25789 F20101107_AAAPMK sundaram_s_Page_017.QC.jpg
ae09f678c4584d96921fa789dc55c74b
b13d07b46a5a266369ac79bef52c2a9259c815ea
9297 F20101107_AAAPLW sundaram_s_Page_166.QC.jpg
3c5e0eaf94a0558bb8a1617f1e9fe5ce
d5837a04eea910ec09e827ba7d978a3d4fba50c5
113316 F20101107_AAAOIU sundaram_s_Page_109.jp2
9ce1bae4f57535e4d559447560b89f91
5e74eb9046c53b74b44ca37fb2a9c43719411590
121951 F20101107_AAAOJI sundaram_s_Page_123.jp2
2bce4230041f299d2f2053fc58489602
c61964da6e8e971bde0e9fab71a94c1ddf684eef
26744 F20101107_AAAPNA sundaram_s_Page_121.QC.jpg
33335918af53f66b8ee9558b9fff5475
9b533b59c00d282dc371379a06d052e894631b37
26726 F20101107_AAAPML sundaram_s_Page_124.QC.jpg
dd1681ea40678734657bfedd717b31ac
0183f6f8c9c2510e5df76e1c36e3bdc4951c376d
7112 F20101107_AAAPLX sundaram_s_Page_161thm.jpg
cc4618a58419f3ca8d54c3757d6ef815
93c0d6050d51e70d7f7015abf1c1186ad8ada1ed
123783 F20101107_AAAOIV sundaram_s_Page_110.jp2
469b8cf24db8dcbd5a4fe031a941035d
62087abbbfc4a78e2a8c9cf1ecb358ec123c07b6
121885 F20101107_AAAOJJ sundaram_s_Page_124.jp2
6a9458eda282619fe1bc195963151ca6
1b7d569776a85ecefadcd05d85cba508586d9af0
6795 F20101107_AAAPNB sundaram_s_Page_076thm.jpg
bb4db2ef37189b5917c69e12e3ef1bb4
f216f19f561500b867269fc825db8454e5dfa460
7104 F20101107_AAAPMM sundaram_s_Page_110thm.jpg
63a875fe9919f84a260919f88735bb84
6c8dc3ca5c5b1851144a480f1a624b26529eda2b
25048 F20101107_AAAPLY sundaram_s_Page_122.QC.jpg
3a7b111769eb608825cb979954eace0e
ba4e9930527c5e4955d6f068df8ed258200d301f
121733 F20101107_AAAOIW sundaram_s_Page_111.jp2
d9e22a8bb9516a8c0b9d2784d0c469d3
a758317602152984606f5bc35911539a3acb5312
118487 F20101107_AAAOJK sundaram_s_Page_125.jp2
e90de88d735eb827519368415ab1a5c1
d3449ba9c29de95e029895044c6746218c72c491
5962 F20101107_AAAPNC sundaram_s_Page_181thm.jpg
990dccbb686548dff2d5c3a57d6f17c6
fd35adeef9614676024e42ea19a58d19a1122c74
22224 F20101107_AAAPMN sundaram_s_Page_174.QC.jpg
85d973e6c353b76ae49a428003e17cb5
fa8ac3a14d5b1ee262d449751c27fa98b4e5bf29
12533 F20101107_AAAPLZ sundaram_s_Page_097.QC.jpg
6e0555326d0acd432beb32a5bb8a331c
dc534c016ec38c58a7da4d911a94aec32de5ff54
121524 F20101107_AAAOIX sundaram_s_Page_112.jp2
3f14bfdb1c3d1c85cbf7b06288f8888e
9a883e8a3386a8ba0f7208968afd06e3f5eb62c8
58030 F20101107_AAAOKA sundaram_s_Page_141.jp2
72b29060a4846a8a080b2a5de5aef9a6
10adaa94b5f75b0b2cdf25b626daa6e104c7e48e
118805 F20101107_AAAOJL sundaram_s_Page_126.jp2
31aa85bcf61f8b2604547e99b5c611d0
1e2d885b6f2a2b21edd4c27c53bba363a890c3fd
26693 F20101107_AAAPND sundaram_s_Page_044.QC.jpg
d2bb54e8986e042ea27aea5dc011c603
97e0e094ef6bdc09e817ba2d065da18095989d7a
2353 F20101107_AAAPMO sundaram_s_Page_150thm.jpg
dde8f031ecbcbcf7b00da7e38bf64ac2
36e44f295cfb5844b5b6879f0c456aea0bcb8838
123354 F20101107_AAAOIY sundaram_s_Page_113.jp2
882bebe5791ae9ef71874d0eb7dfc61c
d967aa80f48b5983656fcfe0375510e1e910ae92
88335 F20101107_AAAOKB sundaram_s_Page_142.jp2
85840fc7217b4b96afd8de883a48adce
3327ce14ed371292d4f0ff22b7475624369b1e0d
122349 F20101107_AAAOJM sundaram_s_Page_127.jp2
e5ccd7e02fa677d1c01f373504ec4983
be3a2ab9b1d5a729aab293ecdd45f2288732b35d
7333 F20101107_AAAPNE sundaram_s_Page_058thm.jpg
b52b2d145bb452aa72c8ce57b1f4f6e4
8440bb5b307d3407ec37ba87d8635da9153c9b72
3156 F20101107_AAAPMP sundaram_s_Page_167thm.jpg
4d2a1bd830d06885ffac5247cc4387d2
eb9d23db3d2f40d4c90425aa25bc49b184652962
115138 F20101107_AAAOIZ sundaram_s_Page_114.jp2
9c72cd6359fcc74bdd9b42c039f25dfd
aa378db5867c587926278232fb56266f78050a18
74552 F20101107_AAAOKC sundaram_s_Page_143.jp2
65349a4ea43806591ecc2f36c2c751cf
0d67937a86803089bfd9fbefcaa09271ee5a3eb0
124668 F20101107_AAAOJN sundaram_s_Page_128.jp2
1008f40d595f2ead3e9ee7dbc6592e4c
1bb0212f62368b4dfeeb6a8cd0249c64d7ff686f
25443 F20101107_AAAPNF sundaram_s_Page_048.QC.jpg
e9dafee36c3115cce798c179c0e730df
97f9c958edd17b70d63bba11a656478a6301f018
24043 F20101107_AAAPMQ sundaram_s_Page_076.QC.jpg
095a6eb0a39820a347bcd681e6577147
7e92b526173f698b6d80a0e2ecf7d433b54d6339
81759 F20101107_AAAOKD sundaram_s_Page_144.jp2
43c0354cf2a92e30bb8e5a17c5681cab
04a13b2fb6e8693c13916a165800ef81ae33082f
121989 F20101107_AAAOJO sundaram_s_Page_129.jp2
6304ac6cc2281daa7090d7401dce3446
dba69ecefa35074eb144f19c02eda45edf13a9e1
9681 F20101107_AAAPNG sundaram_s_Page_167.QC.jpg
b882772922595780cb4de335904cb7d3
05d851db5a6ad7d7d3a1918211a8ac209a3d2d15
22 F20101107_AAAPMR processing.instr
d51fb269b80c1e08b97cef64c901e6ff
1d39002f6cba3643d8fb47818dd4544c2e2af280
63141 F20101107_AAAOKE sundaram_s_Page_145.jp2
312d009a356b88250f528852e8182c5f
482cffad9aaee81f22006494693aa60ca7918d4a
122860 F20101107_AAAOJP sundaram_s_Page_130.jp2
d17cb71bd7c29b944d9910465d4f7bce
01df99bfe3ea9a79f4dacf0f29dd9459738dee33
7446 F20101107_AAAPNH sundaram_s_Page_113thm.jpg
1af476e8d13fd0e274baef5864784e86
1646777916da1ae5e17db735f1ba6b2e1965073b
7241 F20101107_AAAPMS sundaram_s_Page_018thm.jpg
7b7a49745529f3beee74ebf60cc303d0
be1fdd6ff217486ce13a2bd37c29877b5498a9d2
97244 F20101107_AAAOKF sundaram_s_Page_146.jp2
6e004bb4f0488b22fbe30caa2295c1b0
874634152541bafa483ad4a6b2cf961cf5c0ccfe
122505 F20101107_AAAOJQ sundaram_s_Page_131.jp2
bbdae056fd9121d700231c6ed29533c7
cb9b9ff066d127beff004f5f628d47f5b41d10de
10180 F20101107_AAAPNI sundaram_s_Page_146.QC.jpg
06b605958e3398c8177cc5c70f31b31a
4a8e4518727f45e1033b6b457270373333b9c369
5213 F20101107_AAAPMT sundaram_s_Page_079thm.jpg
ec44b9d190005154fa33414c957693e6
94780a276c533d344357badb7d9e62921a2a2f43
79081 F20101107_AAAOKG sundaram_s_Page_147.jp2
4b2f23c6d9ee5c2cf959d0de9dd69f49
469e4981263994d980e1a916f30d7bec1706a169
113037 F20101107_AAAOJR sundaram_s_Page_132.jp2
378d7dd9b8bfa91297413a4c6e5ffb95
43b5bddd48791d2f9315d1a7da2779d8b1fbff64
22897 F20101107_AAAPMU sundaram_s_Page_179.QC.jpg
271a03cbff484d13a461b9a966984f14
18a1b17f12e4b4864a18734af41125f7739f4fc6
120819 F20101107_AAAOJS sundaram_s_Page_133.jp2
a9deba4571baf3d41b6aad1903eb628f
3e4c1a4ab919341d3697a31af351ddded8b22c1b
7262 F20101107_AAAPNJ sundaram_s_Page_071thm.jpg
c07d5ab2bfd00c550ba64a7e48de3c77
eed770188a0e41e5392d8010052cc080e090d6a8
23976 F20101107_AAAPMV sundaram_s_Page_077.QC.jpg
b5f51843984ddb798bdec5fa78ae4085
d4cf6fb0a03e92eae12e95fea8950009b2432e19
89707 F20101107_AAAOKH sundaram_s_Page_148.jp2
318fa63c3e579ec3d883caf1e641f773
8d11011cd63bf8af3383aedd68fca39b9ce74c29
120676 F20101107_AAAOJT sundaram_s_Page_134.jp2
ac0fe368b07a4cb275b41f8752cf7ff0
c427c689c7951474eaa0004ed54bcd79fac4132a
15071 F20101107_AAAPNK sundaram_s_Page_008.QC.jpg
f15d3520c9d070858535a240258f2076
f8a537f9f0a68a0d1085a5270b0a19db94db73f7
7153 F20101107_AAAPMW sundaram_s_Page_119thm.jpg
4104fa8118d210c2fcfcc4a1d7f287c8
f9073335803e3a581d06d2db16a2cb78c8a17bf3
83069 F20101107_AAAOKI sundaram_s_Page_149.jp2
181a74bfd665d54c1ad2ad4c192b015b
8f639c43edcaeaa8f5820539a90e1bf037db93c4
64614 F20101107_AAAOJU sundaram_s_Page_135.jp2
497b39bf66a0c08253c786262b279db1
f5102ec3ec10ffb570d0d5ef90c0d27199ccaa5e
11476 F20101107_AAAPOA sundaram_s_Page_094.QC.jpg
ad7c6e2eaa9738414a1bb96ab6e5703f
fa8ab945d49d072ca5dd473461f15d14448ec4b6
25969 F20101107_AAAPNL sundaram_s_Page_188.QC.jpg
2ed7260b340069ef9401df0ae874cc9a
02225807d985147acc8846511b05836c098c3413
22265 F20101107_AAAPMX sundaram_s_Page_157.QC.jpg
fbd89a614eadf73d0a5274ce77c4a134
6ddad4f3bc216e5cbfef766a999304a1a00aa143
49956 F20101107_AAAOKJ sundaram_s_Page_150.jp2
e6433db90d5bb69b0da3f340c9ba37d2
6a5b72351662bde09be9a6223ea130bd867b2cdc
7758 F20101107_AAAOJV sundaram_s_Page_136.jp2
c7f7eec75eda757ba3ab8f79aaa5d501
236313481c40e4a0049195d92d92714ebea46d1e
24804 F20101107_AAAPOB sundaram_s_Page_109.QC.jpg
3149207191180d27177e70b18bca6683
c2ab45b1f543afd1ace4a8dfe2693b9ae9ec48e4
6681 F20101107_AAAPNM sundaram_s_Page_154.QC.jpg
ec61cea52550bcf5d714d876abb2e351
577a0b3a55403a72d7627b3991d2fed30f52fad6
27825 F20101107_AAAPMY sundaram_s_Page_194.QC.jpg
05d1f2a3078a0baeebfddf2e30ebbcdb
c98b789bfe9540f23977ccf18344a7b0438268a1
78264 F20101107_AAAOKK sundaram_s_Page_151.jp2
f47dbac29280422c429edb2be62d9a9c
14842c9e0b7af429dfb714355fbbdd85fc9c03cb
7723 F20101107_AAAOJW sundaram_s_Page_137.jp2
1b1d39c66e85f65bcfe5e604e4ef556e
a30cfcc3426d567619f4a6d885b45166ad23a33a
26061 F20101107_AAAPOC sundaram_s_Page_162.QC.jpg
253b43338c25318cd6ac1473abfc5b95
38aec15f3eeb988f495722efe6a85e4feb1cbe7b
6540 F20101107_AAAPNN sundaram_s_Page_152.QC.jpg
1b1fe7a2cc70b535d476fcab91563c50
86d939e67bed31d9a049b8c0a93bfabdd81d1c6c
23139 F20101107_AAAPMZ sundaram_s_Page_006.QC.jpg
2027fe5a57a9534a1ea0e05fe659a1ab
92cf26986bc564eef5535e0072cbb554ea4d5482
52139 F20101107_AAAOKL sundaram_s_Page_152.jp2
fbded55a244bf013acff3bea6beb9d8c
0c4227b0b72a2fe45c45cd1604c4d1768e3b00b5
67182 F20101107_AAAOJX sundaram_s_Page_138.jp2
4473fd6361e6a75dfeaf0aa75dfdf3d3
9ab4fdc3bcc0e5afe9298ccf60fe5f8745b93548
67522 F20101107_AAAOLA sundaram_s_Page_167.jp2
81a16c68779340232629abf44d8b8e3f
90ddd7179a5129837faf3b73d304e1a0f879cbb4
19813 F20101107_AAAPOD sundaram_s_Page_054.QC.jpg
c8789acfec3f2e76e8d47926b9018db5
7ab48ec1a39859fe3ca22296f2f34af81b9c6508
25523 F20101107_AAAPNO sundaram_s_Page_114.QC.jpg
5eaa3bf1364db876e1e43944f0e1156b
c35f638a46bd19f555c1cacaf1e48d2857fa9220
93466 F20101107_AAAOKM sundaram_s_Page_153.jp2
331f78b5f117e2c95f7a1c5f115468c1
58d17aa28356bb2c3915a228f2e44a40925fddc7
61806 F20101107_AAAOJY sundaram_s_Page_139.jp2
56b85fd7ec9669871fd40117a01668c8
a856c056e8482215b884fed820f523a93bbd2c77
73552 F20101107_AAAOLB sundaram_s_Page_168.jp2
fc0ce90ece18e14af5baf18760acd2a7
156d00b257091173f83f6773b2484aa83857c3d3
3855 F20101107_AAAPOE sundaram_s_Page_137.QC.jpg
4ea0fb2840614ef7ff3749bde61f39cc
fac99b8f6589dd969acda488450d0a92cc3e1862
3898 F20101107_AAAPNP sundaram_s_Page_093thm.jpg
29195b3eb26d298e64ba18986f0aa2a8
a10d016aeecda725e7afc188ce403866fd6d3a4b
54693 F20101107_AAAOKN sundaram_s_Page_154.jp2
ce7bffc083a4435b4b975b6e771f9a90
be740cb0c19f66a114a431aade0ad81719ea2dc4
81301 F20101107_AAAOJZ sundaram_s_Page_140.jp2
d76163e76446d31b83f2dc3f42780153
6673f8fd92bea19d2083cec9c208799c8d15871b
66605 F20101107_AAAOLC sundaram_s_Page_169.jp2
8df87adf352553c7a789f6e9cb820117
2d55c1e850a95bc19e9c3f4a5accf0472b210b6d
5170 F20101107_AAAPOF sundaram_s_Page_184thm.jpg
0c60cc5e0d4b90763733986460bcfdd3
01c8f6ca51c2149ccc7b852b74ce87cf9545aaa8
24093 F20101107_AAAPNQ sundaram_s_Page_033.QC.jpg
c68819c68aff6e39d8654df1f9571cc3
9ca6d63c5ae39bba7844e991329d19f0f279c2a3
94268 F20101107_AAAOKO sundaram_s_Page_155.jp2
fcc0ca62ebad54f4f01b323d1d9d5fd9
477da59009831717562e1d7dbc9ed9bc7d906cb8
78056 F20101107_AAAOLD sundaram_s_Page_170.jp2
9760185d1a050082b0918c3d627b97f2
e7a6d811fa0bcb4eac87f567a34c67a398d9224c
24742 F20101107_AAAPOG sundaram_s_Page_031.QC.jpg
c3269d1d1610ba74c849a2aa8c3f4e13
abab78a0624f70c8a360fe3d56db608a14c0a54c
3113 F20101107_AAAPNR sundaram_s_Page_165thm.jpg
79401f84932926c27cae057e99fe936e
bc8b4ffa78244f0d0f9cb2ee49ad6a14ad217779
111292 F20101107_AAAOKP sundaram_s_Page_156.jp2
97c72320c6aa46f27ade46394018bace
bc248e64240a5400189a4621485dd7921de29dc5
39984 F20101107_AAAOLE sundaram_s_Page_171.jp2
332b6a87220408e1c5c04f9ada45ec57
35972ef300612747b9d447dc8e40b4302918dd0a
5480 F20101107_AAAPOH sundaram_s_Page_011.QC.jpg
b9031ecea3ef2d4d09b1729d458f6cc7
faad1646a793b398ea0b7cd782d1a8ba050cf47b
5694 F20101107_AAAPNS sundaram_s_Page_102thm.jpg
40716e72a58013819bbcfa274589a10b
f403500ac0f38820fd530c1ceb99ec69630b9ef7
102049 F20101107_AAAOKQ sundaram_s_Page_157.jp2
0cf053e8d5e10473a69c017a4f965be3
9b245603064630e1ab86281b12d77dd20d6e1928
132715 F20101107_AAAOLF sundaram_s_Page_172.jp2
d2e089f5506faed509186f83f4001e46
eec7501155f045ed883a238aa72794e6ccd86be8
6384 F20101107_AAAPOI sundaram_s_Page_105thm.jpg
28bf37e39548d0f5fbd3d2707595390b
e9c090fdf61d3f43afc917763abc49dee939c2e2
7366 F20101107_AAAPNT sundaram_s_Page_032thm.jpg
598afc4745705485ab5a2d53cc1e55e4
b1e38e5629a05471a3e5f290100c294fe6346124
121454 F20101107_AAAOKR sundaram_s_Page_158.jp2
38edf6e518e0d4fb4924a5f06d2d1454
516190ab80ca431d2436ffbea1615034f188b67c
121281 F20101107_AAAOLG sundaram_s_Page_173.jp2
76ad44b91a2bb674e258aa39ea6c4652
168ad8b913074e9bcdb72880f3864ace00dd38e7
7152 F20101107_AAAPOJ sundaram_s_Page_130thm.jpg
2ab726580f4b4933a1e31792f240e114
d6b54a79db7b7db0b0655c116e73806d97ffcfd3
10895 F20101107_AAAPNU sundaram_s_Page_153.QC.jpg
92bfbb9f4ec25669441f0b21b1728321
10b142c99c3595e298203a941f44c703a596719c
121106 F20101107_AAAOKS sundaram_s_Page_159.jp2
8cf27fef35ae2c90b138d25cd3bc00f3
f24e6b46b0f9ab42d3566fa13532bf0a4c484d09
116620 F20101107_AAAOLH sundaram_s_Page_174.jp2
1735dfe9fda6676dae5a794aab949f15
113ee577ba8734d262fa9e255a0abb5443daaaff
1362 F20101107_AAAPNV sundaram_s_Page_002thm.jpg
a4358c30f57b6c3ef7eb53466ab633bd
6011a5ef59868bf6e1cf4f685183e9bc9ce368f0
118866 F20101107_AAAOKT sundaram_s_Page_160.jp2
1d0f47628a59bd45fa93c27d49ce0aba
38584f4ccdf89360e7678b7acfae6c64eafbcb9a
6826 F20101107_AAAPOK sundaram_s_Page_017thm.jpg
77b17802379851d9b5c757879171998b
3658bcec0cffbdda8f11251fe3a75bd8f9c0238d
5923 F20101107_AAAPNW sundaram_s_Page_066thm.jpg
14ff032de3f140c95c95115ec58e9a7c
1e45d543e4aee2810e77eb9ac265ef1fa49e479c
119876 F20101107_AAAOKU sundaram_s_Page_161.jp2
375637035bdf6e6c85d40acd01e477b7
606a3c0987644c0a2b6050608ebd11e3da33b3d1
116746 F20101107_AAAOLI sundaram_s_Page_175.jp2
4354c9fc132d0b8593c43c4569e9e807
3efda148c8dc92e6960a15601caf0caba5798182
25730 F20101107_AAAPPA sundaram_s_Page_057.QC.jpg
67b3374beacc775b848a9dfa158e4b4f
8abf0d58481768ec80a0952f768e81c58e9e63c7
6290 F20101107_AAAPOL sundaram_s_Page_106thm.jpg
c5c3c8d7c7ef56e4e924043f5edb3529
6c42ca4b5fc46f19b7322a65942dc48716612522
6405 F20101107_AAAPNX sundaram_s_Page_062thm.jpg
2e897206464af9ea1540f491b1ebdef8
1f34bebe1705d8c198769cbe7a574e1ead92198a
118228 F20101107_AAAOKV sundaram_s_Page_162.jp2
1f22f85b894d5554d77f16329fc61117
74e22b99b91e03813d32d2c65a29cb6ff310a8be
119030 F20101107_AAAOLJ sundaram_s_Page_176.jp2
270d452025c4a8aaced4c9ee22b51af8
d93751f30802fb4d8904415cae89b087898ce469
5163 F20101107_AAAPPB sundaram_s_Page_080thm.jpg
bccc1545ba2e34adf2b2a0da179c2681
3ef47c8a89a90e3fbacf489fa6556dbc933ace50
F20101107_AAAPOM sundaram_s_Page_149thm.jpg
0f8eef194a0ee78546434cebbd574133
b5236c0822a6f46d3c2193631640b177421c1f5f
19641 F20101107_AAAPNY sundaram_s_Page_155.QC.jpg
59d07e8aa949d46a4e5f6150d1116075
32d2d89d870672786c17b84d947ffef86ff4cb86
81667 F20101107_AAAOKW sundaram_s_Page_163.jp2
8bb9d43ed972aae6ddd7d2cf39695b96
59490bbad3131eaa55ca08d1e85cf43295b635e7
118100 F20101107_AAAOLK sundaram_s_Page_177.jp2
ee3d8cd1c54e219c94716c10a69ca35b
f689e3ddaf4c5229d2868ca84fbea02bf8e54793
6544 F20101107_AAAPPC sundaram_s_Page_021thm.jpg
6d98959065203d674e14fcac58d2cacf
184ce3eea8f36ca9b7214c3ef0340ced6aba8eb3
23820 F20101107_AAAPON sundaram_s_Page_062.QC.jpg
d580148d4c8e14df70f5ad24c8a06e8d
6c155b518ce148714576d9a124157d68caec14fb
26270 F20101107_AAAPNZ sundaram_s_Page_126.QC.jpg
2ec8460be7c790e3aad374d9fd652412
3adfd560d6c6a05f06dd3b58f1f07e828e6a8508
73596 F20101107_AAAOKX sundaram_s_Page_164.jp2
c50b04195d3f27c917d8ae6febe9306d
d14b6d64969047dc769f1ef2d515a109445d4b64
1051904 F20101107_AAAOMA sundaram_s_Page_193.jp2
d1700216ac1ce5433978c2e786d5a290
3c14ffae7f821225dcb79cc72a9a3b8d518e7700
107462 F20101107_AAAOLL sundaram_s_Page_178.jp2
f32d31a6423bd63ee8af7581146d2a46
80a45bc0fddc7a987e984b8d8fee7f6648450d23
6968 F20101107_AAAPPD sundaram_s_Page_053thm.jpg
101cb93ba69232201ae7c49dde7eeeb3
0fbc4dc3eefbb8454a60c9c981eb209b43003fd2
3519 F20101107_AAAPOO sundaram_s_Page_086thm.jpg
5f4222dd4c913b77b599a67866e50141
6347066f7a3efaec39c1a2511102627181f1fb67
73402 F20101107_AAAOKY sundaram_s_Page_165.jp2
f05361a513249929a86b83b90778b555
9e920069606be9280358a7fa4dc15e5358e27a0c
148286 F20101107_AAAOMB sundaram_s_Page_194.jp2
d7fe8ff83a02a2a9dca92d71a92b3982
b593aeb13aa8d75f77c3c4e6c4bec257f004397b
117410 F20101107_AAAOLM sundaram_s_Page_179.jp2
3e6f70d59f0fe4940d08e578f560c107
e714ffbb44fdaf577ffd09d8424a58f7f36a1cb4
5939 F20101107_AAAPPE sundaram_s_Page_059thm.jpg
bd0439e298d2bc80fe4495093f433fc1
4efdde676845ffc672238db37254569b6253fcc8
3853 F20101107_AAAPOP sundaram_s_Page_090thm.jpg
f5f2ed33ca91c4708ceab9022c1d1e2d
686be31b0f6e5ae3b3e66f33a6a49de3cf4e39ff
71056 F20101107_AAAOKZ sundaram_s_Page_166.jp2
85081e90b680d06a7a671e567e5d8522
8645f6208ed96c0b2249d23edaebbbfcc78396e5
1051894 F20101107_AAAOMC sundaram_s_Page_195.jp2
4f55e2e44bad6d8cb6a6430818f3cd25
ecaf6d1966141e9fbc28d7c044e1cdf623993533
107694 F20101107_AAAOLN sundaram_s_Page_180.jp2
6ea87fa905b4a2961bd7c498ed9e2c03
e5c85e7f425f7d77973db590b10f7d57f0a2ce29
7664 F20101107_AAAPPF sundaram_s_Page_196thm.jpg
0fe1c9d13f9c91de7bee67489921213e
58437e57f69622e94a39922888bde0157e07755d
25861 F20101107_AAAPOQ sundaram_s_Page_125.QC.jpg
eb294d9dcbe5d95275976c39c254d43f
0c898d60c6c7f5a22fa6da75d667aad6447dbef1
152223 F20101107_AAAOMD sundaram_s_Page_196.jp2
e447fdb0007facdcef71e97ee7ce6913
cd9d22fcbde7405a44c56ed815effd7952585e64
114503 F20101107_AAAOLO sundaram_s_Page_181.jp2
e9bb2f33d97638e93bd9e9cc664a8b09
3fafd1d6c2dbadb0f35ba61ac72766997a0cc4d2
7266 F20101107_AAAPPG sundaram_s_Page_187thm.jpg
ac93a17791bbe1d8ce8c61f3c92818fc
15d0d2f12612e4291427b2ec007cb7e2dd87c991
5317 F20101107_AAAPOR sundaram_s_Page_206thm.jpg
122e804b6bd10c34120012958bba8b41
a8bc144fa4d7988b65f3ef5192edc191e1e8100f
144903 F20101107_AAAOME sundaram_s_Page_197.jp2
e1715d210c58fb29f2c7437be4483f35
1b34f973ea70307ed177d7bdb22828da7dc940b9
107350 F20101107_AAAOLP sundaram_s_Page_182.jp2
bcc5ecefc6bd19894ce6e00393f830d3
0ec218c3623f96ec50dd52d2112ff73fb9238817
7030 F20101107_AAAPPH sundaram_s_Page_004thm.jpg
0db6ab9e23ef947c04f8f6e0999703fc
79e65dfcadd906d32e433c3a36784114145448fc
5884 F20101107_AAAPOS sundaram_s_Page_104thm.jpg
9183b354d412df57d29928102e8b24d5
6ca3f127946b913c25d99a61aeb1d26db051ef8d
147666 F20101107_AAAOMF sundaram_s_Page_198.jp2
5e84e52fbd26f6f46c53e30862bf0afe
a1881f2336ec8cdb51db53163aa84eba67574f1d
118106 F20101107_AAAOLQ sundaram_s_Page_183.jp2
1415be0c2afc7378e9d9bac7e83ec0b8
c6388a08a1c89c9f39a5f0d69109f171372c4be6
6598 F20101107_AAAPPI sundaram_s_Page_050.QC.jpg
120d2c2351942b551c477136138a4e9f
a38bb92de04005e0151348bfa548e5c0b447f616
7256 F20101107_AAAPOT sundaram_s_Page_201thm.jpg
6fec6dc6b4795347d3d3e4a7eb8ddd98
e78b8f7962fbf167dc571207020e3d1ebc2f2410
1051979 F20101107_AAAOMG sundaram_s_Page_199.jp2
f8ecd8df285c59e9484519012989fc46
4ee215ead0ba2d6ddc2235ac00aab4b933942be1
106738 F20101107_AAAOLR sundaram_s_Page_184.jp2
900ee8d81331fae1af5f6e03b210c4cf
d55a8c05d2e4c5a19fbdec5bcd86c0a8f1538905
27365 F20101107_AAAPPJ sundaram_s_Page_113.QC.jpg
4fcb3fbeb7f64e044254ae220fc017dd
903721ae3ba66ecf7ea0ce80a1844f25efc7f0cd
5573 F20101107_AAAPOU sundaram_s_Page_082thm.jpg
9447dd6a71ec5eedeaeb8b99c1786844
b49e693a7c8500bb91a0ed415221ea8f86914f2f
139082 F20101107_AAAOMH sundaram_s_Page_200.jp2
63900812dc6dc8fda03cd0f759bba9d9
9b25551eef960b7a42a922e9f78b93b4b7b70d38
121442 F20101107_AAAOLS sundaram_s_Page_185.jp2
95860c67ac740c4df1cdb6b248a3e8c3
51a88c5443367c1eb2e00918a842ef98d7e076fd
3143 F20101107_AAAPPK sundaram_s_Page_146thm.jpg
6fa27519dc68ab3f88f095ed0b965b21
97361d0ca8a03e1cf8043363adb0bb903e5c6a04
25997 F20101107_AAAPOV sundaram_s_Page_161.QC.jpg
c821355d98849d21c5e98719c91e6ffe
67b6c9bf6104f2299298669fc163e24d66d88bec
141672 F20101107_AAAOMI sundaram_s_Page_201.jp2
2dfa7ed26304cb50928bbf7dabac636f
fa08ad9030f12f3fc97c0a5077bcfd5d86645888
115558 F20101107_AAAOLT sundaram_s_Page_186.jp2
9059d7457cc22433bd91d4aaeb54f799
383f55e97aef551ff09d4da87d5575c88a420087
19032 F20101107_AAAPOW sundaram_s_Page_163.QC.jpg
f4a0fdd6d934cebabe7ab94a51e328a8
678705c925cc3fc6958cc6b227254b128c9f4db1
143772 F20101107_AAAOLU sundaram_s_Page_187.jp2
7fdaedd7a7d43062fed54b06caa43da7
2dd95c4b9d9b15bdb15c178302b739dae057c3a7
7205 F20101107_AAAPQA sundaram_s_Page_202thm.jpg
a0a81f9819c01c56dbd2c28900d18232
2816912ee75e46f6f21f2347ec4d34a4e36698f1
24545 F20101107_AAAPPL sundaram_s_Page_007.QC.jpg
a1688cc91bc3d283080e18175c495a95
5ce49e2ee84db9e79b7cc5af3358eb6d823cc66f
7056 F20101107_AAAPOX sundaram_s_Page_125thm.jpg
47965c18ed23e90e3421a7164a940bd2
27a53280657f7c7a4dbe182ba379511fd5d04d48
144256 F20101107_AAAOMJ sundaram_s_Page_202.jp2
812498b76f8d099ebedbd40e5e1465fd
30b9b0f4e5970246bc05e8dc7b863c7f5fc10966
134178 F20101107_AAAOLV sundaram_s_Page_188.jp2
b939e0ee66c861718dd9ff3c75bb46ed
e1ed4aa4457b3905bb546898e4ad7d5ad18284e8
25663 F20101107_AAAPQB sundaram_s_Page_072.QC.jpg
7537ec003b22cc56cca76a75f70f87cb
31f52ff5e8a28226c0d2aad2e39b473fedad81d5
26685 F20101107_AAAPPM sundaram_s_Page_112.QC.jpg
98db3e6b15ca65336fe609885a8aa4db
dc892db44841fb717a7dd8a44dad9e6a7f171883
6766 F20101107_AAAPOY sundaram_s_Page_192thm.jpg
bb93c3b45327582a550ff7d82cfbe9ce
60f1a5118d2abcca82d5e5ff7bf7165e9283b4b6
1051981 F20101107_AAAOMK sundaram_s_Page_203.jp2
f1fa99608bb12c6637aea2cae698229c
f13660fba75463d7ddcda2f9db05a3f6d063c735
144236 F20101107_AAAOLW sundaram_s_Page_189.jp2
b611242d07e80750b833659d28c79ba2
33e9810946486f828259661032a70ec11b191e1f
28516 F20101107_AAAPQC sundaram_s_Page_198.QC.jpg
da25b0119887ce684960297feaa92d02
76b3193fdcfdcc9e004754b23e376ba1cbc393da
5296 F20101107_AAAPPN sundaram_s_Page_182thm.jpg
c40f0bc8e446821d776d5ecc6816059a
77dc781f00d0c59cbf2fa541e202e3600396c280
21954 F20101107_AAAPOZ sundaram_s_Page_059.QC.jpg
4a1be62cad4df895e8f478929efbb747
c04135102dbc51c03ce15670a35c75041f521cf7
F20101107_AAAONA sundaram_s_Page_012.tif
bcf62a36c78f1ba7947378c1a5e380a3
e6814eeacfe71844e7c54e4bb0700e2e080e0fa5
138667 F20101107_AAAOML sundaram_s_Page_204.jp2
cd630f75bf7ec20ac5fe326f5f83f4be
3728f09adfee6b79a1cac440b004514acc45d964
126948 F20101107_AAAOLX sundaram_s_Page_190.jp2
d8fca5bd07d9718e914bdbd670f9f786
d1534ea67b81954b24dc7e40fffc91507094c290
23105 F20101107_AAAPQD sundaram_s_Page_100.QC.jpg
3df26f3d8d49f6db72be9e028b79e73f
01e36a3a92a3b86258f738fac077378fab666748
4471 F20101107_AAAPPO sundaram_s_Page_065thm.jpg
368b64e8502a6848e3744d8ccc923308
3b86794c29530c0fce61ac34b6375d017aea5944
F20101107_AAAONB sundaram_s_Page_013.tif
6e8e5f2d0c8a546142e0a4291f89afe4
92d2518b05b470ec28e64f4a7e320a27996aa0d6
143063 F20101107_AAAOMM sundaram_s_Page_205.jp2
cc3b73a5f4fbafce0a63a37e0887cdd1
86461a62060b681a2b4d16365b36f0f89ece1d73
1051970 F20101107_AAAOLY sundaram_s_Page_191.jp2
2df6bb9f9a9494c1fd356075826c8e79
b2b84a213be6549607b34d5ef89a35bd41aee673
3694 F20101107_AAAPQE sundaram_s_Page_097thm.jpg
20d2a18d8ad083e475480a9731cfdbd1
5cccffeb47c09014d8d198d81817c5e174e546cd
25794 F20101107_AAAPPP sundaram_s_Page_029.QC.jpg
a5cf339f5fe660d0a4ce23f543e3c652
c1b4b913511aedea59d04121f825f48837a05c89
F20101107_AAAONC sundaram_s_Page_014.tif
03a9c71bf0ebfc5f9d7bd2cd3d06803b
6de35ef4648b1889e01aa7b54256b2715fbe9037
98400 F20101107_AAAOMN sundaram_s_Page_206.jp2
502176cfb8476a0718bb69fd8ad4583d
9f7bae66d1cac8f54a8264ec3fad1ace60d20c18
134812 F20101107_AAAOLZ sundaram_s_Page_192.jp2
347c33e2fbc143601dcc1718575b688e
2c24b6557ecfd3a1b4d8e0830e49d7466888310a
10088 F20101107_AAAPQF sundaram_s_Page_142.QC.jpg
e62c5a0d0b068925117ba862db5d1a91
03785e9e6f018601e3baa4ac4332bfe2bfc521a4
23881 F20101107_AAAPPQ sundaram_s_Page_015.QC.jpg
0f78ef54ea0f699af0b062e2f51ae4ab
512ffa59ad90aa04b6896a20a403f75cdbef9dd3
F20101107_AAAOND sundaram_s_Page_015.tif
a994c7ea2cc496d8aa7d01b73af0fa86
2795f8bdd74ec7826b220668dfdbcf532a8f7f1c
66164 F20101107_AAAOMO sundaram_s_Page_207.jp2
3622e6ee2bc504c96939d6e98d456005
f38aabff6c72654fafde70d379882255171b7ccb
F20101107_AAAPQG sundaram_s_Page_074thm.jpg
ef4b3311611c9ff3d63b739a2f33a7e8
72eeaebcdad8c4cabb02cada99398240318f7fef
24611 F20101107_AAAPPR sundaram_s_Page_106.QC.jpg
cb686dfa04870d1c55094a1718d50899
2e604378793dfba7aa9a7bcf8d2807b23ef79d59
F20101107_AAAONE sundaram_s_Page_016.tif
a3a05099edcc7c3a0e8f027886d6f639
cb098fc9a058484a82f0f5254cdc5c1a566b9d73
F20101107_AAAOMP sundaram_s_Page_001.tif
a89eb323032cf2d3b08e769c0eef03aa
acbe3e4ce7eddecf8fa9f254911c1d465eda7e62
6927 F20101107_AAAPQH sundaram_s_Page_109thm.jpg
c86dde146e46cd19722d41429a9048f2
9a441b08d077f27b043a7c61861951a24d728a93
27212 F20101107_AAAPPS sundaram_s_Page_130.QC.jpg
aa2e2263398773bf424d734ab88eca3c
91a47f4862fe822cfb8093dbd4bf80ed871351b7
F20101107_AAAONF sundaram_s_Page_017.tif
c428d42552214d0c4022890080c22cbc
d9edf78980414173f3a205ef9083fc607ca89c75
F20101107_AAAOMQ sundaram_s_Page_002.tif
3ac18920fcfc698d0d95e1c4501c6a81
edb860568dc35c2e9e87ac4078161f913821fb7e
25440 F20101107_AAAPQI sundaram_s_Page_073.QC.jpg
22439a08836f2c25a625ca0dd15989d0
ce257b6a8327d69aab961c7cdb458b0e1573061c
3306 F20101107_AAAPPT sundaram_s_Page_168thm.jpg
52778ce352a9023ed29c6ed87793ffa9
f187a10428a4b9cc6139c3e821b8edf2e231c1f8
F20101107_AAAONG sundaram_s_Page_018.tif
afc81ea25b4f5b09359c8f418eff9126
a198935c05a1a8e0c69ae141617cce67aab70adb
F20101107_AAAOMR sundaram_s_Page_003.tif
a536d2735db4c207e9ea7e7327db36ee
d4ebd56d47da8b364634f410bf3a583cb5190903
25597 F20101107_AAAPQJ sundaram_s_Page_052.QC.jpg
4d922cb6e903e2b9f44dcd2fd006d101
dbf288b02865b2a0df0d9a11d567a78fd1713056
20140 F20101107_AAAPPU sundaram_s_Page_178.QC.jpg
8b4ff03b0f13d14600122b1ed47c0473
2993699f2928917c59c4e7893c0a8e402d3a6897
F20101107_AAAONH sundaram_s_Page_019.tif
d6f29149b3229cfb2d1e07f6096be192
0664526d4a36e83ab597f51c09e614f6debe17e5
F20101107_AAAOMS sundaram_s_Page_004.tif
b91a8418e0344cff0650bcdb5a989fbf
2b444146de41066cc0b5cc507e1a30e50a6c10fe
24162 F20101107_AAAPQK sundaram_s_Page_061.QC.jpg
541fa2d91933c3faf201ff32bd69248b
4b54b9e423d1202bf15f5b4b4b58ca3d272147b3
6977 F20101107_AAAPPV sundaram_s_Page_016thm.jpg
4a2f242723a99b3e02a997de15905766
2b62e2328e3e3cc32433bba06baccfab2a116dde
F20101107_AAAONI sundaram_s_Page_020.tif
165547fc0731029f1a2f33d746a46de8
e5e133d8f4052c6636f7b60921be25b87066fdf4
F20101107_AAAOMT sundaram_s_Page_005.tif
9ade196d133f488a95c8c430d26db6e4
a0f6e259a342068d0dd0ad706f73e3d6fb0b803d
3063 F20101107_AAAPQL sundaram_s_Page_108thm.jpg
cd48c57d46b8a8cc8ec9b9db1daca2b6
830fc4a04dbedfcd5a94059a69e5e0968bfabf22
6525 F20101107_AAAPPW sundaram_s_Page_033thm.jpg
0a3833bbbc28a9fe8bd9846c3af89a76
f4d8b829042a59b3ff8236fdec2104f2dee85394
F20101107_AAAONJ sundaram_s_Page_021.tif
4a05cf021a7cd1e3e65542fc9d46dfc2
9577f8dbe17a9997e76f1a78d82380eebfe385d6
F20101107_AAAOMU sundaram_s_Page_006.tif
27bcdd83df5207c2099ec8ee43aba546
4047a621642f690078614223c86ea40ed4d205fd
12607 F20101107_AAAPRA sundaram_s_Page_093.QC.jpg
8b2d54f4b407959a3e31077a173d447a
c20e9b73db2faf68ab5710b4537d0947a07fae77
26652 F20101107_AAAPPX sundaram_s_Page_187.QC.jpg
8173754f6e2fbf49eaf1dc3bc976702a
32b4a9e5df7b5bb91baac97ae5eaa2d55df7f30a
F20101107_AAAOMV sundaram_s_Page_007.tif
6e6ef9473a4447f28c9b007450638fac
094b45330c17945c0aabd7984ce4c65b2ddf7e1b
3645 F20101107_AAAPRB sundaram_s_Page_092thm.jpg
58b9459845f35d83d9a606a28f8d7bb0
74d8611f0d071e8acc551417217ee8f1fc05c326
3245 F20101107_AAAPQM sundaram_s_Page_087thm.jpg
882026b958c5bccb9be0433c58a777e6
cb618d90c652d16dbac368d3214f131346585998
22524 F20101107_AAAPPY sundaram_s_Page_056.QC.jpg
a1dd22166dd23d12705686fdc064712f
52a30c3bc86b6116fb9525e58b3f5861d20f69f6
F20101107_AAAONK sundaram_s_Page_022.tif
fd8149cc436c55155260210b0a781f05
a970d0401b5f40b381bdec55ba4c94c3700e0dc6
F20101107_AAAOMW sundaram_s_Page_008.tif
c40edc89fb8a06f8869f9e6a59db9660
8cb869f6fcc865d972e3202105d901f42968680f
2014 F20101107_AAAPRC sundaram_s_Page_005thm.jpg
cbc3c3a7f04bdf1df9bdff897110cc6e
b1be9549943fa9b075a9e1bc60a5c6ccf4c976f1
6879 F20101107_AAAPQN sundaram_s_Page_048thm.jpg
5ec413135148ce14b5f34fdbc8d97e84
ca7f9ead9335a1287bebedbb7e778d3bc78e2b49
3247 F20101107_AAAPPZ sundaram_s_Page_153thm.jpg
eece90224a417d6ffd5daaa18ae23c80
c3e7610b1cffca1165a32303a18fd7acbb9eadf3
F20101107_AAAONL sundaram_s_Page_023.tif
19f08d18542f48bebcc1d4ca3c12a270
5d38e4db06ff8b02a4a84ea5fa8a83c4f5dd5286
F20101107_AAAOMX sundaram_s_Page_009.tif
4c5a239a2d68295699c7fe6553734550
7c5d07f46b91b926a61750449436e56f61f9b340
F20101107_AAAOOA sundaram_s_Page_038.tif
7aa27316749c87de7cbd5cacfc9f8828
fc94d9de606464805f9331b2cdec589ab05a6713
2507 F20101107_AAAPRD sundaram_s_Page_145thm.jpg
747ac8db66a5cf6577e6a97ecae36b1f
b58f3af0597f188872275bceece3e53a001a3eb9
2980 F20101107_AAAPQO sundaram_s_Page_144thm.jpg
e311aa584a6560f1f4ae79d0810a1b4c
eaa362d1bb7838e8fc044c54b88820e83def8a33
F20101107_AAAONM sundaram_s_Page_024.tif
424342b2f19b266aebc2c5422242689b
485ed239286ba64d411aee7b0785cf139d38563d
F20101107_AAAOMY sundaram_s_Page_010.tif
9ca1e0912173ad8a9f7249e0063a8c1f
e666d04091e369d5e6a31f4272a576b61ac0b804
F20101107_AAAOOB sundaram_s_Page_039.tif
4d4d3cfc61a97d6f317c04998f693dcb
b2bfe7c6e901d157c41a79d25d57edb446fe30a1
21930 F20101107_AAAPRE sundaram_s_Page_063.QC.jpg
26f0399208e45599dce383f4ed48251c
0a36e4d6bab513d8db5c0c64741d45f49e9e1b5c
3132 F20101107_AAAPQP sundaram_s_Page_142thm.jpg
3d903c56afdbe35f660a98450782d543
274307475ce266512bdeea1bfd0b1cffae0e1db3
F20101107_AAAONN sundaram_s_Page_025.tif
5e26c3f9b376f3f0f945329f1b25acf9
dc8a07852a703d20933b0141466337074700acba
F20101107_AAAOMZ sundaram_s_Page_011.tif
8f6cc8fbcebff95a99be9d0996ac2d1a
c94648995330d0815b00e6e071326dd741b7a223
F20101107_AAAOOC sundaram_s_Page_040.tif
a46e305251ef5cbc97ebf3f76f7e76f0
6f1084b856c22010364dc256fe6eaea1d2820c83
2583 F20101107_AAAPRF sundaram_s_Page_141thm.jpg
f0446e8f1fd2c8e4b9e4e01182a7e1b0
4bb8557ebc5135b0cb3ccbee658d4c72fb98bab6
7108 F20101107_AAAPQQ sundaram_s_Page_034thm.jpg
b1146c8cb7bb532eba72e1b582949286
176b2977906442bfa950c087b9c08a0802d407d1
F20101107_AAAONO sundaram_s_Page_026.tif
8207d59f232a8abc0baee4157a7cf245
935f1838b2060237709e7a177c41bd852961d8e3
F20101107_AAAOOD sundaram_s_Page_041.tif
aef4cf39df2e3a18a40f2573d4df2dbd
a689281c444b7db3129318dafa8ffa5ba6e0a4b7
F20101107_AAAPRG sundaram_s_Page_141.QC.jpg
09b969ed82d1bf625889e99360f466f8
92a69a2965f3fbee8f68d47be6941ea5950e25e6
26653 F20101107_AAAPQR sundaram_s_Page_200.QC.jpg
ddc197bd1f3a3802438a4f37304a58c5
d26da83801a5b7144c11b2f4d318c8a3043cfc28
F20101107_AAAONP sundaram_s_Page_027.tif
969b66337b2a441e4efdd2222ab6c676
87f7284fd1a3f1aa849f48d9d8d9ca2d2705061d
F20101107_AAAOOE sundaram_s_Page_042.tif
b23510ae06be6db768cfcc44a4263da3
fbd90892d185688363c6cef2bb3ae7b9c7092645
25396 F20101107_AAAPRH sundaram_s_Page_116.QC.jpg
560a82340cbc8279daf72cff602c41f5
364602fa4df51782ed6e7c9e4cbcaa8bb063bf9f
3947 F20101107_AAAPQS sundaram_s_Page_008thm.jpg
f95f85c363a2cf39094c88f9f6ad2aa6
1a53b5dad661d118d89a2be53e725be36ca0bc21
F20101107_AAAONQ sundaram_s_Page_028.tif
248d94ec8576bc5e14220e0a7c724f58
317268827d9ca6f3b049a00af94b6786d6e22533
F20101107_AAAOOF sundaram_s_Page_043.tif
a80a5a48496089b3c48b61d1310a8f61
c9872f6763ae9c7ec518fe34a4435913cb34d8de
26413 F20101107_AAAPRI sundaram_s_Page_119.QC.jpg
2fccd8ea47b050fef806ad74ac1d7638
dbdc3282e9245a3f0f4dbb2c307236377f174389
5933 F20101107_AAAPQT sundaram_s_Page_157thm.jpg
22b0ddddc6439fb4444ff6f4ac8c3d7a
d4cec67b0f4590777002cdf12282b082a59801a2
F20101107_AAAONR sundaram_s_Page_029.tif
c011cafa03dc86d4b98f62999bf09f3b
eb10fb9150cfc59a8a3f3ede15f2877163dcf366
F20101107_AAAOOG sundaram_s_Page_044.tif
f10094915e4763bb01acbfb4125e77cd
00932b098c9d0e4198ce708ab5e75001836793a4
26800 F20101107_AAAPRJ sundaram_s_Page_030.QC.jpg
d0022a8ff6288d516f6db53360c55555
8e8a896c8bda84b5c67ca1ff946d924fdf027311
9357 F20101107_AAAPQU sundaram_s_Page_143.QC.jpg
22454e103e8e0db3c574700eb0277bc7
741dbf68a349f921b071e97ebba7add931c97be7
F20101107_AAAONS sundaram_s_Page_030.tif
ffba18ecb5570ae2c42aa918a528f436
cbf120722b198a200e7254a90a5ee1aa1586c8af
F20101107_AAAOOH sundaram_s_Page_045.tif
1de0d916ce15880d6bff2f709a78e9c1
eada9e98a9303a0a205b9901f8fd71513da64910
310539 F20101107_AAAPRK UFE0022424_00001.xml
860eb21515fb02b00ef49c50aa62f3c6
194ba225bde6bfdf5a6a2da1c05958cf27660ff5
7125 F20101107_AAAPQV sundaram_s_Page_069thm.jpg
ee9402e6555d75fa1d463e8f4cab091b
9b43ce592399b4d89ff01f338a506489a6a20dd1
F20101107_AAAONT sundaram_s_Page_031.tif
07a8a698e153bbb21d9d6fd5dafc9550
c7d9788645cd9cdaadb781f0db744193fb7c0029
F20101107_AAAOOI sundaram_s_Page_046.tif
5eb0ce8306cb0dd5578e6f673f27b842
876def8b0624780f3792c3503a5a96ec9a9ec38c
7412 F20101107_AAAPRL sundaram_s_Page_001.QC.jpg
2c47a628edac5ef3d629f7c23e07d6d6
800e56b6d1166ac6ec90a5c912c84690d2e7cb3a
26748 F20101107_AAAPQW sundaram_s_Page_134.QC.jpg
661943aed8bfe535afc8c6598443ec2b
12f5d6848c4cb31a808a88a5b3ad73d104d58917
F20101107_AAAONU sundaram_s_Page_032.tif
56bbc2d64daa8779865e957a080b6a02
0ee3d0cc11fabc857f2e14c01f74ca38c35a8ad6
F20101107_AAAOOJ sundaram_s_Page_047.tif
2cb140fe8f0cb58fcc8d62ee4bf20bf7
5deead0f57b8a6edbde9e4325ccdb0acd7592129
6460 F20101107_AAAPSA sundaram_s_Page_015thm.jpg
6d4083ce964abfc645ccb0367a82aaf9
1faffde11f16d154b58a267e6b18465ddb463433
3073 F20101107_AAAPRM sundaram_s_Page_002.QC.jpg
89ee3a05cb7855092ff19c29373a35aa
f7f8d4cf32b057bbc4d9b7a7142463ca977ffeca
3373 F20101107_AAAPQX sundaram_s_Page_164thm.jpg
6689c95f25052744a3d333e374d3347b
42c427fdc4ea51ef9d34f8370c021b2e56d8b99c
F20101107_AAAONV sundaram_s_Page_033.tif
b1963ae0a9002ac1fd51bd7471a89449
8324c6e519b3a43fb806725642bb1c2422854bce
F20101107_AAAOOK sundaram_s_Page_048.tif
f13b82da5910eae8e9b6309b8aa3058f
b500173a3d6af45fba97729482e1f7b5167ad0c6
F20101107_AAAPSB sundaram_s_Page_016.QC.jpg
7b7f94e20f9eaedb41901dd0d57aba17
8d4f84af12dc58b02d1ec23d79c41a2c95509217
10595 F20101107_AAAPQY sundaram_s_Page_020.QC.jpg
a4cdbf20ca6cf75a2cdab0cfbd35ec82
f471ac1422af061a038a199feb7206e9a44ecff6
F20101107_AAAONW sundaram_s_Page_034.tif
372b98e32c3a5a75cddf15a7bb12f7b6
7ea11d6e30172b603788b3bfde579ca89beddbbf
26384 F20101107_AAAPSC sundaram_s_Page_018.QC.jpg
e2f932e30a0a464119c0b0ef6f609472
5d2a6e6aca5d377f079268b96d6c7a688eaa9e6c
1307 F20101107_AAAPRN sundaram_s_Page_003thm.jpg
3a5398f245b9e4fafe02c11624aad095
c3f4559e4e167f4d3ed6bd1f16a18cb4ee40727a
5979 F20101107_AAAPQZ sundaram_s_Page_005.QC.jpg
53d3cea97af7fda50ce5cdd8b5dfe8f5
4a7e4d0b05e5aae26681c1fb01a480537a5249f2
F20101107_AAAOPA sundaram_s_Page_065.tif
2c678f2f535077a000351617bce5e129
712b96338bb017666fb0384fbda2bf1114cd68c7
F20101107_AAAOOL sundaram_s_Page_049.tif
0829086bdde4e6855206699c533d5891
e5a7fb390876e3e067464d29b688bcb73a15d178
F20101107_AAAONX sundaram_s_Page_035.tif
380265f025ac559431ad8c2461812094
6340157502b013d62f31de5588a1644354834fe7
7058 F20101107_AAAPSD sundaram_s_Page_019thm.jpg
36f0ad188b05494eda3cc78269330dde
4832973bf20d461514e76e8bddbb69fc3e9dcd55
26143 F20101107_AAAPRO sundaram_s_Page_004.QC.jpg
2960399d2e8f2621a7dd49decce5b20a
93e16864226ff193e1beb4cf69a3361ee3a43308
F20101107_AAAOPB sundaram_s_Page_066.tif
1cf1d96b8a0f4efdfc99fe72eb85ea3e
6a928cbc5ac0d12156f2f10c79ad507c1d4a50ee
F20101107_AAAOOM sundaram_s_Page_050.tif
70561ca5d2bf27929b15b6ed466fac24
b62db5e6e37cdf69494d6120c9cc31524686f75f
F20101107_AAAONY sundaram_s_Page_036.tif
98fecce19522f928a4f921820bb1e610
b039a86ba52a9ec013768b50717911f6786e2cb3
3214 F20101107_AAAPSE sundaram_s_Page_020thm.jpg
ab3101481a662887b7cc558b2186e9d5
75571e56370ac675800d471d9f34ba1adcbddbde
5872 F20101107_AAAPRP sundaram_s_Page_006thm.jpg
a6043561efd60fcc37e3166eaf3b7ca5
d89a4c4875c29ed86cb918715d2b873a6b6b71ae
F20101107_AAAOPC sundaram_s_Page_067.tif
3cfeba0fee7fbb25ea58846eaf0026c5
48de0d808b00a22869bf0916f26be4dbcaa272fa
F20101107_AAAOON sundaram_s_Page_051.tif
48aa139f4947f48ef9b5f815e0fa1b4b
236c079ef031be50ffd4b6fbceb6dc0cd8339967
F20101107_AAAONZ sundaram_s_Page_037.tif
c484031fdab69662cf7ebc2590c4c792
df347d99fa2a878958379e62a3cc85fc709329d2
24460 F20101107_AAAPSF sundaram_s_Page_021.QC.jpg
4809d61dd796480196798c23f05f3269
0a4c34ea5c8bf819c51ed7fbc45a737f852ed5fd
6184 F20101107_AAAPRQ sundaram_s_Page_007thm.jpg
b08f8dfdc1076377c9b3a6cb8c5db914
cdb81f267713d179db63e281f73b7d5c8d12479d
F20101107_AAAOPD sundaram_s_Page_068.tif
9e8216ac3177621074f8f0b709266d7c
10b27ef8f86e1dae28caa1ea306403b003d9a104
F20101107_AAAOOO sundaram_s_Page_052.tif
62355c70ab850517ab8b1555c2d15931
d9f7f90beab47d89900cfea3ea102e8dece64b9b
26751 F20101107_AAAPSG sundaram_s_Page_022.QC.jpg
42e23d85f4579ff4eb1a85905494bc75
7273480cf0df460c53da8bd2d2d90a5ecdfba8a3
26258 F20101107_AAAPRR sundaram_s_Page_009.QC.jpg
a225558bd5f017b51182ee2648b4fa23
6eb0330da3ce7fb6b9a7d7c83958d0bce53ddf95
F20101107_AAAOPE sundaram_s_Page_069.tif
ae7c912d43a77343e21034122a273225
417fe8008815941484986bf7d6301c449e33543c
F20101107_AAAOOP sundaram_s_Page_053.tif
9b9614f335f719e3aa65d90421275949
6ad159dce294a789753f367f9bd670d6a09dba4e
F20101107_AAAPSH sundaram_s_Page_024.QC.jpg
9ff0eccb474f960b6ffaa305ce6442e5
5e805c80a4766539b7d99905f2cd95050c332f64
7066 F20101107_AAAPRS sundaram_s_Page_009thm.jpg
48870774ba08c7c1a618bcbd96e7882a
d272dca4ea4799f0e09d6642878d779030f49ccf
F20101107_AAAOPF sundaram_s_Page_070.tif
16734b74d5ee20b028eb65b4c73a7cd3
3de50c736b0750b4ea859b7d1fc6384d08eaa749
F20101107_AAAOOQ sundaram_s_Page_054.tif
dfbeb9dc111aa36434b16ba1f5e4bf48
20c6a22c9a4c8c9433c613cdcf274ba202adfa73
7289 F20101107_AAAPSI sundaram_s_Page_024thm.jpg
c5183866793e8495882c72b92b4ee80c
d1e201e23b597d6021ebc3c90ad3b8aa19dfff1c
27869 F20101107_AAAPRT sundaram_s_Page_010.QC.jpg
7d500d5e52e59ab62199288faa5d0360
51042b346000fa2a816370cc40977ab2c6cd20c1
F20101107_AAAOPG sundaram_s_Page_071.tif
54e426bfcd800b7bd4ff3717dcb0f95f
8de53e3d2404b62031daa9ad287af9f87c65cde3
F20101107_AAAOOR sundaram_s_Page_055.tif
c3f72f5d90ee2455e8ccb8eaf21f4af2
019f12f579640e3d3bd1fe605ac475d06cc582e1
25445 F20101107_AAAPSJ sundaram_s_Page_025.QC.jpg
645454e869576ebc80871bb664fd6ac7
b49041ee1478625c00f067c3790d5266a777c680
7103 F20101107_AAAPRU sundaram_s_Page_010thm.jpg
e7ce6aab529bb9a56de09f1f108030fd
449d97c30876abe58c8a679b77bec6a63186cd16
F20101107_AAAOPH sundaram_s_Page_072.tif
19149dd1000c90e3ddf2f5b7fd2adfce
065c0c5254307a51004f49e7ade17f4da98ad47a
F20101107_AAAOOS sundaram_s_Page_056.tif
7eee45f90691494861cfc058c7c060ab
9eeecf90850ba098e73aad3690ce558667dd0775
6841 F20101107_AAAPSK sundaram_s_Page_025thm.jpg
4bdc0a9b67c836dfb2ff9d42ff4abb17
52c89cb8477bb2e4e56cdd264f6ef29471b8505f
1944 F20101107_AAAPRV sundaram_s_Page_011thm.jpg
4b34bd5cee309394075d73dec6eabf22
35ca9cfa6950922743182712cdc70a83f63e9296
F20101107_AAAOPI sundaram_s_Page_073.tif
e22edb634e0a1abed6780093d367a3c7
dd938b28fdee216d2399f74a9bb708da64f920bd
F20101107_AAAOOT sundaram_s_Page_057.tif
e7d463ae67242ef534ed084a5a8b0b16
f289a711fb77bde7199182216af1364bba00d976
27549 F20101107_AAAPSL sundaram_s_Page_026.QC.jpg
769f18808d0b771882783f18d3104537
2be554c55727682d2d51ac4c5834cfde96823ab2
6212 F20101107_AAAPRW sundaram_s_Page_012thm.jpg
39a318a52598c3cc4464ff3da65b434e
17b1f64c0352f0e58413748e706c90bbb9aa913b
F20101107_AAAOPJ sundaram_s_Page_074.tif
9cf6f0545ba0151fcf3fdd99bd0f5e3e
65ebe9979a0248a12556a50c4ff06bac7df03ab6
F20101107_AAAOOU sundaram_s_Page_058.tif
bd1f1a0ff11394cb4ce72f63dbd83af5
ae6b376ec2bbaec3867f450092f5b663bbdebba7
F20101107_AAAPTA sundaram_s_Page_037thm.jpg
ab3f4e4a47242273e19d13d6a260760b
7bcc6236e1df5ebcf96d88f4b6abd2207eabbd97
7311 F20101107_AAAPSM sundaram_s_Page_026thm.jpg
046e6a67c022292c65e3bc9b15547719
294d12d84da27bc08f3822caf6973a36464f1e69
19795 F20101107_AAAPRX sundaram_s_Page_013.QC.jpg
cbad19b5cf613d6c3490bddebc9d8057
6c41ca890fba2df84ae5e92d4d50e9dea86bb965
F20101107_AAAOPK sundaram_s_Page_075.tif
b87ce83fcf96c9f6dbe5b84f39967e7c
6315f4d8d6c2e622992670f93d5a469b73c7af97
F20101107_AAAOOV sundaram_s_Page_059.tif
886d462599f8a4b4fbf1863a2e6bad0a
65c47a349aff3ee4abcebcbdc3d072f2e0ca99eb
26715 F20101107_AAAPTB sundaram_s_Page_038.QC.jpg
ff907bbd507ce870a098bac19403f4a7
fccd91352ffddd9e70a9bde6a54f80f292bb845f
26999 F20101107_AAAPSN sundaram_s_Page_027.QC.jpg
c2ad5a09b10a6a7ee32c5ad703f09864
00301bc18e8e156c8121292f7f7d990dd2a4663c
5472 F20101107_AAAPRY sundaram_s_Page_013thm.jpg
cf06fce7d5ee7ba6cc22e3405760329b
0e39f58239e0f308f2f52bf61a219cc97e877042
F20101107_AAAOPL sundaram_s_Page_076.tif
1027e5baf40f2ba8d05aa3efcf25f570
07e4d0e63dc5dc4f0f6d697822cc0ca85de11399
F20101107_AAAOOW sundaram_s_Page_060.tif
717830d5ba57f0afca3284bc1dafab19
532f04866cb2ab805d3186e133c5672b5aa99f74
7110 F20101107_AAAPTC sundaram_s_Page_038thm.jpg
bdc30cd2eb32cfb0005f9025c9340d4d
74d469e771b9a42a81843b85285da6a8fb1ed2e2
25196 F20101107_AAAPRZ sundaram_s_Page_014.QC.jpg
5aee5f7cc75536eb39c83ebdd191ffeb
a43019eedc24a2083b20bc17d3c4ad075909abd8
F20101107_AAAOOX sundaram_s_Page_062.tif
bae0da4098b4ada3b4631246c01ea9db
5c7e46fb65d9478e9c72c57dfd63e570853b6687
F20101107_AAAOQA sundaram_s_Page_091.tif
aa523b70a0b1881b9a4753ea355d6b3a
34562cafb1589eb82543524aa4684659a98d6a77
27159 F20101107_AAAPTD sundaram_s_Page_039.QC.jpg
77b2ed78d2cc7d6bf536703d8c2a3692
959fbb7085dd0499aa5c0516547ef46c6ca1ff9a
25955 F20101107_AAAPSO sundaram_s_Page_028.QC.jpg
d1b5fcf69c1c91b3584797c5cac696d2
e92b5e8d38264b10a3b17d48a11fa331b7528898
F20101107_AAAOPM sundaram_s_Page_077.tif
2d6b851ac21da5a851a625d0c2e4567c
a5e20be11d090f4b2bf91a12433b927fb312e15a
F20101107_AAAOOY sundaram_s_Page_063.tif
7d3f4de489062a32a6a02a03f2efd65a
1c883afd8dbf5a4b9b93866974004c750538e455
F20101107_AAAOQB sundaram_s_Page_092.tif
d4b54f42ca08aef5649643f09a21ec7d
49793f5766c365114e739e918ad0b350359a68ff
7093 F20101107_AAAPTE sundaram_s_Page_039thm.jpg
e901020c6494546b70e232a2caec7bd0
e2d1ff94abf9ff7095b48daac2fc86f9aac91c58
6918 F20101107_AAAPSP sundaram_s_Page_028thm.jpg
86248c5c159a4c05bad10b341cd266f9
3c943827722f44eecf7dc8a3287dfd8d54f49d0a
F20101107_AAAOPN sundaram_s_Page_078.tif
f38e8a2a8a5dd055f9082b517e0c68a5
19327f1e86e7d5db1e086bb2900d6613cb47a203
F20101107_AAAOOZ sundaram_s_Page_064.tif
44260d8619be519e5bc21a05e9d2d792
5821f6d3db82bc0c07d1eb1010c4fc7083db858a
F20101107_AAAOQC sundaram_s_Page_093.tif
10a20d5591e35672825d0c309fc2d300
c643f2b5a814a9772df72239e03b824a7547b1e0
F20101107_AAAPTF sundaram_s_Page_040thm.jpg
029f63b5a0930f60928f01713357e4a1
37176045f6afb164a68214872483fbcd8114c3f7
6707 F20101107_AAAPSQ sundaram_s_Page_029thm.jpg
57c83209558a8cda7196c141e5e49552
7e3dc9765206335d2939cf8bc5b8a9b1a9313d87
F20101107_AAAOPO sundaram_s_Page_079.tif
b107379ccbdafe9eb93590dbb4521e22
5eb3875defcb98ad4d55b117a1db21b8163cacfd
F20101107_AAAOQD sundaram_s_Page_094.tif
851912ad7e8c109592b1fc93c3377f04
cf3ee1ffe9809cc471ec62ca9f9c5ba0835e1a4c
27028 F20101107_AAAPTG sundaram_s_Page_041.QC.jpg
abe24ab2fe3b74f4fe48452d18555ade
f353a9b00a5646c78b84491675dfb9f1dd472deb
7214 F20101107_AAAPSR sundaram_s_Page_030thm.jpg
9970cabe5e99ef3212e6a7c94e04a57a
18539cb34d19faf7a69d0e27abc94b6b6d8b0558
F20101107_AAAOPP sundaram_s_Page_080.tif
4dbc57c74db1f4979892a986026cf528
af26493162fa138f91d51853f0413b1066105d6f
F20101107_AAAOQE sundaram_s_Page_095.tif
0ae779fec4a6c1557e7c1ce477a263c9
516fa5bbc203d315347dcfda388af78045966126
7216 F20101107_AAAPTH sundaram_s_Page_041thm.jpg
964a868986b6bb6818650411d351be55
5c0954a0babbf49c6ce4eaff2874b542aee27818
6888 F20101107_AAAPSS sundaram_s_Page_031thm.jpg
90df4cf9469ab7265caff64f180151e6
eb9b198e17c94a139733b28eba68e62d8facc69e
F20101107_AAAOPQ sundaram_s_Page_081.tif
b12ea59e2656f5b7fca719b2db1694d1
9abc78a44f5c5c1968c7796df432bc94b5198087
F20101107_AAAOQF sundaram_s_Page_096.tif
c2d557007801b70efd9662464137f02d
bf037d047faa9dc222eb7d77890858d1752240bb
26569 F20101107_AAAPTI sundaram_s_Page_042.QC.jpg
e4cdd268c92e02dae24144db3f74a035
91fe52496e5402879370d878c6e1faae7fe3cc3c
27533 F20101107_AAAPST sundaram_s_Page_032.QC.jpg
5ff8acf6eefa2f4fd9c2da4fc660b1e0
48a2fb552469a26fb032f8a830efe829092f5738
F20101107_AAAOPR sundaram_s_Page_082.tif
f7be371b6c6619a4c8b68ff81c27ed04
7487d2c2dd63ec3bb1b7cdc0744d50a7cd18491a
F20101107_AAAOQG sundaram_s_Page_097.tif
912f429e5e9269cf31c9104546d0ebc5
6b2ea4fb97430cf6c5ac93611f5d481ea8e1816f
F20101107_AAAPTJ sundaram_s_Page_042thm.jpg
58f04d1562bc5e227812559b2bd6b9f2
26f499a2e49074d9bf27b3f6f48063c19db7c7a4
26089 F20101107_AAAPSU sundaram_s_Page_034.QC.jpg
d28a48aa12cf54d5063036a15d90776c
639485d7a1548df387a4de617ebc4144933bcccb
F20101107_AAAOPS sundaram_s_Page_083.tif
944b59c6bb7612717742688da043a074
107d18b58eeccfc73dbcf4cab4bd084457d1d3b5
F20101107_AAAOQH sundaram_s_Page_098.tif
61a5640c48107a9fa788e21b80e239d8
027ebb15fac33ad937f523e14b01663c3278317f
26742 F20101107_AAAPTK sundaram_s_Page_043.QC.jpg
efef7de86e0640118b562214676d30ea
aabcba2b1c0c8c23ed221fa98c891e0d718b880a
25534 F20101107_AAAPSV sundaram_s_Page_035.QC.jpg
1600a1a933d8d0c69ecc96eb8bdcd77c
d9caeb067ff1bfa0876ba3559a3142350dc90599
F20101107_AAAOPT sundaram_s_Page_084.tif
11bc352335cb717fb0ffaf06439af168
ab75441233a2a1668cffa8373b3603033139d004
F20101107_AAAOQI sundaram_s_Page_099.tif
8396eee849c54ff43a2108f4aea06c64
df9e3c7f94866ec6f01ce022098297c96b852c8b
7164 F20101107_AAAPTL sundaram_s_Page_043thm.jpg
0c82f675b1fa190cbe267addf8e45058
c2c4ed08216ef6fd45f19739dec8dd2091b43790
6868 F20101107_AAAPSW sundaram_s_Page_035thm.jpg
96051aa5bcf78ac92f861dfdd2b9747b
7416e9f2914e080ae035f8f444fe2e3286f8ae26
F20101107_AAAOPU sundaram_s_Page_085.tif
f9bfd4d0ea530d474fe8e66adae4a6fc
619459d166fef0a3cd5a341a0c6a18e76340d8c2
F20101107_AAAOQJ sundaram_s_Page_100.tif
815f4fbc7d354f65cd7685097022dad8
8ab71233e8d8797212ef2ff47119453faaa9a280
F20101107_AAAPUA sundaram_s_Page_057thm.jpg
82aa1812af869b948df51b435bb57cbd
887d4223228b26406050616a23bf160b7ff9109f
F20101107_AAAPTM sundaram_s_Page_044thm.jpg
081a5fa0faf646bf65019aa9be07a6af
ecc0a0e63b00789d40498d1e7717dacd3559c08a
26605 F20101107_AAAPSX sundaram_s_Page_036.QC.jpg
a097b7c960fc63b58c5ffce6417c8866
1ad0cfb32d188683f654aceaea51e65486da95ba
F20101107_AAAOPV sundaram_s_Page_086.tif
277d75fa49094c8ef77c590a34169581
8c45963558a1b5ba0e801f73bbcc45b18c1b5b93
F20101107_AAAOQK sundaram_s_Page_101.tif
45f75a1a6668e0732d428b6b43f17e06
11531510e04e539868acf7ebbf7a00e442c6a66e
28070 F20101107_AAAPUB sundaram_s_Page_058.QC.jpg
c96c352d391d93613c312b85c4a7a06f
6bfda23cb74d6aa005dcbf2a3abd8b7ef20face1
25608 F20101107_AAAPTN sundaram_s_Page_045.QC.jpg
28096add87af941e58b7a37f68f97e37
3d2e215df7feb60f3738890552181748ad69e41f
7161 F20101107_AAAPSY sundaram_s_Page_036thm.jpg
df73d4de5cfba6c665e4360095b8fef7
79b5b4404d42758ebb390f588d5bfa682812b051
F20101107_AAAOPW sundaram_s_Page_087.tif
782d66d2286121267766d7e42c606739
ebbdfd229ab0542afd6459dfb76086a8c2b62ceb
F20101107_AAAOQL sundaram_s_Page_102.tif
e6008d5f81e69636bbfcad31b2408bc7
8faba4d78f20827726587b20b67e5c33eadf93aa
25984 F20101107_AAAPUC sundaram_s_Page_060.QC.jpg
6842e0a2e51da9abb8dd141213d4a01d
3126bf888664c20a4d384cbcf36a30745b2f8ca0
27397 F20101107_AAAPTO sundaram_s_Page_046.QC.jpg
43f68c0a3d28fa1aa477f9d90527f76a
819ee21caf263fb80d3161a149818127bdf6ab53
25944 F20101107_AAAPSZ sundaram_s_Page_037.QC.jpg
21cf2bd77084893929d9cbbd5975b0ba
afe700f9e2a9bb04971dc3070745d62c844abd77
F20101107_AAAOPX sundaram_s_Page_088.tif
874ab05e1143e3d3387cfe4a79d74eb9
b71f674c5e5f38ad8db7ec4ace4fe35bbefdc81f
F20101107_AAAORA sundaram_s_Page_117.tif
a0ac77d8518d52b0c19966a2bf277582
4dedd6aacf280e89643f7aa8680dcc2a2375d303
F20101107_AAAOQM sundaram_s_Page_103.tif
780cc94b6f44d1166a808c5a76589513
a7a8d1f965d6136f45620a7bc0c335ea5a857865
6896 F20101107_AAAPUD sundaram_s_Page_060thm.jpg
378e88de8dd674ba0663b7bb192ca144
71f940adbed62fd2d32fae4e214b056c1250be75
F20101107_AAAOPY sundaram_s_Page_089.tif
08b78c51b624d270297a77dd79f85f4c
def675fa0947d63d6ce730cb85f4dab6c2bad15d
F20101107_AAAORB sundaram_s_Page_118.tif
84d81303573f80c27da2be11f3491d8b
43980ec8e8fd6db1061138155954847c2d2fd1a2
6590 F20101107_AAAPUE sundaram_s_Page_061thm.jpg
47291ae18a0f5e5ebb113ca4f7f3069b
c205f0c2ed3328816510282e6058e5fc54d7fbbe
7265 F20101107_AAAPTP sundaram_s_Page_046thm.jpg
da04e5753612604f29776acdfdf1ac17
fbbd53184246a7421c2d59c425a609186c2454ef
F20101107_AAAOPZ sundaram_s_Page_090.tif
d6b15bc1d0f4f444d35d39246bb710ed
eb7379f245dae791b6cb60899a55e5ea9f5bc68b
F20101107_AAAORC sundaram_s_Page_119.tif
043daa0ec9e1dfc8ae0230af4e1f41dc
ee6a8d8aabd732f8d7717b0b6c52e3b4705a91b7
F20101107_AAAOQN sundaram_s_Page_104.tif
e01fcda9e4cecaa87ba8a259103b976f
a2edece00b3b13fd9fc7edcda0c05c817e44fcaf
6192 F20101107_AAAPUF sundaram_s_Page_063thm.jpg
a8e72936f3e4de111475dd3fb9a1d4b8
2373df90959e78c5b5f018797cfbb3354ecb1f88
24720 F20101107_AAAPTQ sundaram_s_Page_047.QC.jpg
d203e6dce85d29ac88c6c53512289ea6
dcc1cac0b214d6cb1db4dea4475f6b693fb0bb79
F20101107_AAAORD sundaram_s_Page_120.tif
63dc90b2ada54f25fe505bd81fafa107
09de1aa044657eb5db86bcdad56bdf6c973d0fd7
F20101107_AAAOQO sundaram_s_Page_105.tif
2dacc3d359133eac6980c61f07d5c2f3
4e22dabc05ffe80bfb598ec0117c6ed1fe19df93
28074 F20101107_AAAQAA sundaram_s_Page_201.QC.jpg
d8bdb38b108d346904b6eb59d0c3cdfd
9ee6ab25bace8ab53fd30a666136378bb1cb455a
21116 F20101107_AAAPUG sundaram_s_Page_064.QC.jpg
9d859b475cd598417ef4f61000fcfe2e
e6dcf9eee249f952a1715c31ac05e69d42b89a8d
9838 F20101107_AAAPTR sundaram_s_Page_049.QC.jpg
4f18368a8f28b8e4f6627e757f512e2f
198892b513bba33d8c24f6e7f1d259c7ae4779bb
F20101107_AAAORE sundaram_s_Page_121.tif
f511c30cc4ea87c5032b245642873cf7
e2426687f5b6f123cbd8db2b053f006b02ddd02f
F20101107_AAAOQP sundaram_s_Page_106.tif
0868d4fedaee069a74e90dc293ea5f97
9a851b37061e7ee27cb1314e2d16961cbc6bf651
26676 F20101107_AAAQAB sundaram_s_Page_202.QC.jpg
e44dd4fe54fce27837335815471ed636
0009b1fdf617298a761ff689cd14b078e26713f8
6044 F20101107_AAAPUH sundaram_s_Page_064thm.jpg
a4ac438892cff0f4ae73923ac7929ceb
f6dfadb3e8aa0abcacddc3ad22f9cb06c97b8cd4
3023 F20101107_AAAPTS sundaram_s_Page_049thm.jpg
592d22cf90be62cac89aee63e1fded68
0838d53400af0fe863f1b3beb7f7671bdd22b746
F20101107_AAAORF sundaram_s_Page_122.tif
296b6a41d39ff8c389fc332c77cad1e9
72e858cb0b2ba5aeaed4e59de41fcaba1aabf65d
F20101107_AAAOQQ sundaram_s_Page_107.tif
bbf6f94ff8875e8f2d0fc3ded85f5d83
b2a666027c8fd093c03f1e623e692c1049a3392b
25772 F20101107_AAAQAC sundaram_s_Page_203.QC.jpg
828ee97e630d5bf7da5a536331446f65
aeef4ca220598b711dcfe6975800189c895b37e4
15142 F20101107_AAAPUI sundaram_s_Page_065.QC.jpg
41a992cbbf1a5bb42fdccad3e85c510c
b8a7bd3da21e84c4327e136902a4eaa609b941d1
23713 F20101107_AAAPTT sundaram_s_Page_051.QC.jpg
36f15c7a8b7a244dc5027651b41e9227
80d7ac144af1cb93e4038d4dff455bedda8a55ce
F20101107_AAAORG sundaram_s_Page_123.tif
e2eb8388ccaba302fa0408705429dfaa
73a2de3b3e217a0302fb11b5f191d19972de0e9a
25265604 F20101107_AAAOQR sundaram_s_Page_108.tif
24c319040f7705fe14414de9fb642edd
6317d5c33498fa4a1a3861068f5b46d88ab54917
6806 F20101107_AAAQAD sundaram_s_Page_203thm.jpg
e97c769ee6175afffbe46c15f2a22fd4
7abdf88081cd0d6a855b5f3a88289218e3370c85
22406 F20101107_AAAPUJ sundaram_s_Page_066.QC.jpg
fe0cb1704792898f9f779cd627e65cbe
b07fadeb8593e0a604977f7036be3c70517eb432
6507 F20101107_AAAPTU sundaram_s_Page_051thm.jpg
6c0603f273bc7badae063786f9184623
7be4904ec2579cdf61462141b11796cd848ac46a
F20101107_AAAORH sundaram_s_Page_124.tif
fc67364ad79c97d2127eecc14ba1243e
c8d9ecfc2fb41a0f5adcc92e55ec4e22b3c8a40a
F20101107_AAAOQS sundaram_s_Page_109.tif
a33c6344254f303409bccd92186533ec
7fc99ffe38134434db6a54d4b20274c1de2e2978
26001 F20101107_AAAQAE sundaram_s_Page_204.QC.jpg
c536aa179fddbde18d30a0b339ed8c5d
cbbe562b60ad96c500b5b2756e81feb8d3898865
3450 F20101107_AAAPUK sundaram_s_Page_067thm.jpg
4641265a502591954b1d204db77f4965
7fd968aaebb6c435a41bca0576725fd6f46f5bc3
7119 F20101107_AAAPTV sundaram_s_Page_052thm.jpg
d22e5f2bc7f339f2b03074f54a71ffc9
40944ec51e3b794e18e72cdba0f405cec172474e
F20101107_AAAORI sundaram_s_Page_125.tif
207631d334f32863d5646cfe0a297e7c
d6d44520db54d99f547ef841dbece80e5034e2cd
F20101107_AAAOQT sundaram_s_Page_110.tif
bb2a0827db2db8610f5c45095b5b340c
0c31f532e5adfa863f05fe541a3d585121920755
7054 F20101107_AAAQAF sundaram_s_Page_204thm.jpg
6c512a24f413649eb8dde47b73191f81
a0b77d555c8f0ab9a0632234b113e87c19276117
23788 F20101107_AAAPUL sundaram_s_Page_068.QC.jpg
eeea9442b2831c297e250b67e108b590
6c8cf2081a9bdebc57a34014791ae19d52a3d9b5
25848 F20101107_AAAPTW sundaram_s_Page_053.QC.jpg
4453a5788f25e97bb67701bae179a531
cab91bdae09ea9b6dddb969edfbb895918897bdf
F20101107_AAAORJ sundaram_s_Page_126.tif
5f71f677af87e5abadc42de7a3197d43
acfd1344a89a7856dae454cfba7def6c8f1957ba
F20101107_AAAOQU sundaram_s_Page_111.tif
3f507ad46894219f6842ed3093b9fc91
863a668e98b26f124afef4301bd767024e679876
19319 F20101107_AAAQAG sundaram_s_Page_206.QC.jpg
9de0325d751eed254cbae2e3676ca884
0e321e8ed3f34781d5f7c431cdfd23153b4d5680
11020 F20101107_AAAPVA sundaram_s_Page_083.QC.jpg
6d99842678ba3958e00b2bf7edba0c06
5404af50db8d23368600e8e773f09abee9d0c913
6600 F20101107_AAAPUM sundaram_s_Page_068thm.jpg
8abdf6fdc367cef401da8de8196b1116
6b009d442422ad98bcbdfe51c450049645c5f070
22993 F20101107_AAAPTX sundaram_s_Page_055.QC.jpg
3672ed236ae1bd7d2de78e10b5ea9b47
604c16befc5bbf8d7ab6f986b073b33b14b38d64
F20101107_AAAORK sundaram_s_Page_127.tif
33e2f4c49aaeef79fc4167ae1a7de618
1ab4365e74589834bedbb182e92ca7e51319c53f
F20101107_AAAOQV sundaram_s_Page_112.tif
5c321ebcbf1024567fa21e77f94715be
76912383b1cbc98f0069c57728979b846587c884
15281 F20101107_AAAQAH sundaram_s_Page_207.QC.jpg
84bb0ed60dbdfe31e29bbfa195d91b59
44bf534e2d46e7c3dd1280e51e41b68f22373959
3408 F20101107_AAAPVB sundaram_s_Page_083thm.jpg
a3d58b2498dfecf0514d35444453d6b3
22d57481f6ec5c86c1a0a5d0e208a7464726c21e
26587 F20101107_AAAPUN sundaram_s_Page_069.QC.jpg
87dade53deeae0cec641b20dd2dab773
32857410bc7965bcf1cab65cc4f6172a29795d93
6328 F20101107_AAAPTY sundaram_s_Page_055thm.jpg
9419f0d600b47fb8ef4c0026b2e49997
de17e7658bd20f88965f9d0ac3ffa72c38800bc4
F20101107_AAAORL sundaram_s_Page_128.tif
f716924470a8152e6a21d6e19469021e
52d1e830143fe8bee379b67d4d7de135338067b2
F20101107_AAAOQW sundaram_s_Page_113.tif
3dd55cd0a44cef17f7bca4f2768367ed
bfee8b8813e27d8800f505207c0d2b06faa5ac8a
4348 F20101107_AAAQAI sundaram_s_Page_207thm.jpg
6e3ef064ac5cbaa90563c275c5eb04b5
0b32cbdf915798a67d6bf1c712d019658512c338
11414 F20101107_AAAPVC sundaram_s_Page_084.QC.jpg
24cb2dbbd758bc01e0355a6be38aeb93
bed321ddd1870173c2f91eda2bfcce31e486b326
26860 F20101107_AAAPUO sundaram_s_Page_070.QC.jpg
48832ac3319d64b464c70df5aadc03de
ec43a905a56df403576803ccbbda92b78f7efc6d
6234 F20101107_AAAPTZ sundaram_s_Page_056thm.jpg
895f12ee73eb5dc00d840d840abf0933
29fb73ccca900ab86da8ad3c1789fa862d46cb01
F20101107_AAAOSA sundaram_s_Page_143.tif
697efad26ed449c45dd2b1c6b7dc4f27
eecd664de0a8ae47582c4d4085aac8a5387052dd
F20101107_AAAORM sundaram_s_Page_129.tif
4737b964b6ba6fb90b3f7f09ea8c8d50
bda362aa58817b6d72c74b984744cc0cc334d291
F20101107_AAAOQX sundaram_s_Page_114.tif
14382def71ff17c51054e22ffdfaeea3
76e564392fd97d8c600e2b790699741b20c039f7
10548 F20101107_AAAPVD sundaram_s_Page_085.QC.jpg
9ee0dee1efa2565b2d71ca9e39c48cc8
4f0428975619c81592c4a59c4def302e9bebe604
7160 F20101107_AAAPUP sundaram_s_Page_070thm.jpg
81e8f9e7ed58299bbc787c35c67a4d5d
5895ae9d5b9217cfceafc099d45739f8152a57ae
F20101107_AAAOSB sundaram_s_Page_144.tif
377615ac21a1195506266261dc917029
5af221413e7db8338eba01e98324c26db1f1e5ad
F20101107_AAAORN sundaram_s_Page_130.tif
09f00209f2a1666f62afc7283348d02d
edb1fa7934586cfa5364d3e438ce3e00cfc1fd86
F20101107_AAAOQY sundaram_s_Page_115.tif
5f1cac09e80595a8961d62534f381ad0
3127f95cc70761b9d5dc496263c010fa97604d66
3433 F20101107_AAAPVE sundaram_s_Page_085thm.jpg
9da715493523b03d17463b393b4902dc
a53f3a155c273b167f0365239128d719d61589cd
F20101107_AAAOSC sundaram_s_Page_145.tif
c1e7e3f471bf842ae21ef8c401c198d0
b33f385462263e53cb488cc34a1d3e206b8c79a3
F20101107_AAAOQZ sundaram_s_Page_116.tif
43b3bf01dcdafad60670f1cd448ef7c2
7df6856b813b2b70610ded2e05b0f2054cd796d7
10386 F20101107_AAAPVF sundaram_s_Page_086.QC.jpg
71d8ea24f5b236c7d247062804b1ba84
d43eb8328f611385ce27440cfdd9c66f8f1c8fc7
27613 F20101107_AAAPUQ sundaram_s_Page_071.QC.jpg
46e745a89494eb680f52d9fff638d294
79c98e39866ba81741d9f87b2d995b7495bf57cb
F20101107_AAAOSD sundaram_s_Page_146.tif
a82327b1604d4254cf12175f34a6e987
29591f26c28d632cbc3f07fa4e7129c47ac3b2e1
F20101107_AAAORO sundaram_s_Page_131.tif
564b34b844bc34d184bc54c724e1df29
d7701d1226a9697dfaafc331c0b4a64d462eee6c
10278 F20101107_AAAPVG sundaram_s_Page_087.QC.jpg
c050ff4ad232e341b43bc3894b965d61
53d1c14904825c390e7a00f848c2232d5f63a495
6971 F20101107_AAAPUR sundaram_s_Page_072thm.jpg
6eb8deeec68a9d32789d06bf157f3b57
dd0889e7341748fbfa12b9826b93793ee9ac63dc
F20101107_AAAOSE sundaram_s_Page_147.tif
6f4018ca7f8eeff79feae23f17c57d35
08ac42c38bc5ed51941bd92f7267c89fe671e817
F20101107_AAAORP sundaram_s_Page_132.tif
c122eeed428f4dbeb87fa1330c9ce9c7
accc9ee3fc9ccf6f1f3b89ae50638de34d30afc9
12124 F20101107_AAAPVH sundaram_s_Page_088.QC.jpg
65feccf64070fb2a2f54da4a335afeb7
9a4ed7fe1ae909824420d2af082c2664d6dfd279
6892 F20101107_AAAPUS sundaram_s_Page_073thm.jpg
f65e494a80ddfaf8bcd0a403e7fc03fb
5e26294cba14b81c1d5438374f06ee42c5ff9454
F20101107_AAAOSF sundaram_s_Page_148.tif
f9c5be172a6aee31a864fc2d4b049a42
625210264603794817db33fd425a273ecdc8b4c4
F20101107_AAAORQ sundaram_s_Page_133.tif
083b268086b6835686b1db3b24a04781
137c261aff7139c4e1a5e5a4f7c73263b7034019
3785 F20101107_AAAPVI sundaram_s_Page_088thm.jpg
09c98bb0618ca4a7f5031865e6645a14
b0932fa07b841ae1f4e30d1c35bbd7d0db252d46
25286 F20101107_AAAPUT sundaram_s_Page_074.QC.jpg
7dcf5dd1df50280bb51c0e5f34aed440
85ea7b682f81d71931ebee93fc6d3bf3e9565558
F20101107_AAAOSG sundaram_s_Page_149.tif
d992de14dbcdd9980b09688a482a3b88
75af3a0e64387a2b5e68dbc73ba4192fa550d987
F20101107_AAAORR sundaram_s_Page_134.tif
e565fddb3a22077a86391ad1aef95b38
251d92e05b92d819093645a282d56fe39deff9f4
12736 F20101107_AAAPVJ sundaram_s_Page_089.QC.jpg
be42f6e010d4f4ec32e4cd51615789b9
41c7e79d6bbb766c3055905c26533dcd5d4b0acc
24923 F20101107_AAAPUU sundaram_s_Page_075.QC.jpg
4853a266c9b35c60257156b140514556
2ff21fe57434a9be9f1d0c6673ef7a2375d8aa4b
F20101107_AAAOSH sundaram_s_Page_150.tif
e71cd697db6d11760f424a34ba20d813
7abf3553ce0e4ab3dcccf4250c79285ae40e70f0
F20101107_AAAORS sundaram_s_Page_135.tif
d003983d618caff9b5a3bb9b131b0716
e985c818af0eacf0aeabf8d33618e91a6b6409c6







OBESITY AND POSTPARTUM DEPRESSION: DOES PRENATAL CARE UTILIZATION
MAKE A DIFFERENCE?




















By

SWATHY SUNDARAM


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2009



































2009 Swathy Sundaram




































To my parents









ACKNOWLEDGMENTS

I would like to express a very gracious "thank you" to my entire committee for being my

"gurus," teaching me and passing on knowledge of everything they know, and providing their

unlimited guidance and support all the way. I would especially like to thank my chair, Dr. Jeffrey

Harman, for 1) always encouraging me to approach my dissertation work in the most promising

and challenging ways, 2) teaching me about health services research methods and providing

much knowledge on the software that I used for my analyses, 3) aiding me and always taking the

time to answer all of my questions and concerns during the progression of my dissertation work,

and 4) providing extreme guidance and support both as my chair and my mentor. I would like to

thank Dr. Mary Peoples-Sheps for 1) helping me find a dissertation topic, 2) directing me to a

suitable dataset to carry out my dissertation and connecting me with the Centers for Disease

Control, and 3) teaching me about prenatal care, and providing endless knowledge and sources

on prenatal care, all of which were valuable to and strengthened this study. I would like to thank

Dr. Allyson Hall, for her assistance with my theoretical framework, and for always cheering me

on in my endeavor to graduate. Finally, I would like to thank Dr. Sharleen Simpson for helping

to strengthen my dissertation with her valuable clinical experience. I would also like to thank Dr.

Paul Duncan, for offering his words of wisdom when I needed guidance throughout the course of

my doctoral education.

To my parents, Dr. Kalpathy Sundaram and Mrs. Girija Sundaram, who have always

believed in me and provided many means of support throughout my life. Their endless support

and encouragement has allowed me to achieve many successes in life. To my husband, Mahesh

Sundaresan, for being supportive all the way. To all of my friends, especially Karen Mounger,

Kezia Awadzi, Keva Thompson, for always providing a "shoulder to lean on," and for always

reassuring me by saying, "you can do it," and "you'll be fine." Finally, I would like to thank the









Centers for Disease Control, particularly, Ms. Mary Rogers, Ms. Denise D'Angelo, Mr. Brian

Morrow, and the entire PRAMS committee, for without their generosity and assistance, I would

not have had a dataset to carry out this dissertation.









TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ............................................................................................... ............... 4

L IST O F T A B L E S ... .. .................. ..... ..................................................................... .............. ....... 9

LIST OF FIGURES ..................................... .............. ............... 11

A B S T R A C T ............................................................................................... ...... 12

CHAPTER

1 STATEMENT OF PROBLEM ........... ............................. 14

Why is Postpartum Depression of Concern? ........................................ 14
W hy is Obesity of Concern? ................................................................. ....... 15
What is the Importance of Prenatal Care? .................................... ........................ ......... 16
Gaps in the Literature: Weight Issues Related to the Pregnancy and Postpartum Periods..... 16
Gaps in the Literature: Prenatal Care and Postpartum Depression ........................................ 17
Purpose of T his Stu dy ...................................................... 18
Specific Aims and Hypotheses ................................. .......................... .......... 18

2 LITERA TU RE REV IEW ... ........................................................ ............ ............... 21

Postpartum D expression (PPD ) ................................................................ .. ......................... 21
C on sequ en ces of P P D ............................................................................................... 2 1
Screening for PPD ........................................ ...................... 22
Postpartum Depression (PPD): Women and Men..................................................23
Obesity ................................................... ............................ 24
Psychosocial Consequences of Obesity: Stereotypes and Stigmas ..................... ...............25
Psychosocial Consequences of Obesity: Weight as a Chronic Stressor .............................26
O b esity an d D ep re ssion .................................................................................. ......................... 2 7
Obesity and PPD .............................................. ...... 29
Prenatal C are (PN C )....................................................... .................... ............... 30
Prenatal Care (PNC), Nutrition, and Weight: Behavior Modifications ............... ...............32
What Should the Content of PNC Entail? ................... ...... ..................... 34
Weight as a Stressor and the Importance of PNC.............................................................37
Pregnancy and Excessive W eight Gain .................................................... ..................... 38
Postpartum W eight Retention........................................ .. ....................... 39
Relation Between Pregnancy, Weight, and Postpartum Distress..............................41
Expected PNC Content Versus Actual PNC Content ......................................................45
E ffectiv en ess of P N C ..................................................................................................... ...... 4 5
Prenatal Care (PN C) and Postpartum Outcom es................................................................ 46
T h eo retical F ram ew ork ................................................................. ........................................ 4 7










3 M E TH O D S .................................................................... ......... ....... 51

Data Overview: Pregnancy Risk Assessment Monitoring System (PRAMS) ........................51
D ata C collection P rocedu res ................................................ .................................................. 5 1
W fighting of D ata ............................................... ....................... .......... ....... 53
Rationale for Using PRAMS Data .......................................................... 53
Postpartum D expression (Dependent Variable)........................................ ..................... 54
Obesity (M ain Independent Variable)........... ..................... ................. ....... ...... ......... 57
Prenatal Care Utilization (M operating Variable).................................... ..................... 57
Control V ariables............................................ ............... 59
A naly sis ...................................................... ................ ............... 59
Primary Risk-Adjusted Logistic Regression................................................................ 60
Specific aim 1 ................................................. 60
Specific aim 2 ........................................ 61
Secondary Risk-Adjusted Logistic Regression..................... .................................... 61
W a ld T e st ............................................................................................................................. 6 3
M odel F it ..................................................................... ........ ..... ........................ 6 3
Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Merging
Strata............ ...... ...... ..... ........................... ... ...... ................. 63
Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Dropped
C ases............ ....... ...... ........................... ... ........................... 64
Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Imputed
D ata ............................................... 64

4 RESULTS ........................... ...... ............... 68

U n iv ariate A n aly se s ................................................................................. ............................. 6 8
B iv ariate A n a ly se s ...................................................................................................................... 7 1
M ultivariate Analyses ............................................................ ..................... 75
Primary Risk-Adjusted Logistic Regression Analysis .................................................. 76
B aselin e m odel .................................................... 76
Specific aim 1 ................................................. 76
Specific aim 2 ............................... ...... ................................... 77
Secondary Risk-Adjusted Logistic Regression: Subpopulation With Healthy
Pregnancies.................................. .. ...... .............. 77
B aselin e m odel .................................................... 77
Specific aim 1 ................................................. 78
Specific aim 2 ................................................. 78
W a ld T e st ............................................................................................................................. 7 8
M o d el F it ................... ...................7...................8..........

5 D ISC U S SIO N ................... ...................1............................9

U n iv ariate A n aly se s .................................................................................................................. 10 9
Bivariate A analyses .................................. ..... ... .. ... ....... .............................. 113
Multivariate Analyses: Primary Risk-Adjusted Logistic Regression Analysis ..................... 118



7









B a selin e M o d el .................................................................................................................. 1 1 8
S p e cific A im 1 ................................................................................................................... 12 0
Specific A im 2 ....................................... ... ........... .. ....... ....... ................. 12 1
Multivariate Analyses: Secondary Risk-Adjusted Logistic Regression (Subpopulation
W ith H healthy P pregnancies) ............................................ ................................................. 122
B a selin e M o d el .................................................................................................................. 1 2 2
S p e cific A im 1 ................................................................................................................... 12 4
Sp ecific A im 2 .................................................................. 12 5
Summary of Multivariate Results .......... ............................. 126
Lim stations ................................................. ........ ............... ......... 129
Importance of This Study/Implications ..................................... 131

APPENDIX

A SUMMARY OF LITERATURE ON PNC CONTENT ......................................................... 137

B MULTIVARIATE SUB-ANALYSES ........................................ 155

A dequ ate P lu s .................... ... ...................................................................... ...... 155
Sensitivity Analysis: Ordinal Logistic Regression ...................................... 156
Women, Infants, and Children (WIC)................................... ............................. 156
In co m e .............. ...................57............................
W eight Gain Discussion ......................................... 157
R results of Sub-A nalyses .................................................... 158

LIST OF REFERENCES .......................................................................... 187

BIOGRAPHICAL SKETCH .............. .............................. ........................207









LIST OF TABLES


Table page

3-1 Specific aims, dependent, and independent variables............................... ..................... 65

3-2 Classification of body m ass index (BM I) ........................................ ......................... 65

3-3 Step 1: Gestational age calculation into expected number of visits for the APNCU
In d e x ................ ......... ...................................................................................... ............ 6 5

3-4 Step 2: Month of PNC initiation calculation into number of missed visits for the
APNCU Index ......................................... ................ 65

3-5 Step 3: Categorization of index into categories for the APNCU Index.............................66

3-6 Characteristics of the the APNCU Index groups ....................................................... 66

3-7 Adequacy of Prenatal Care Utilization (APNCU) Index frequencies before recoding .....66

3-8 Adequacy of Prenatal Care Utilization (APNCU) Index frequencies of incorrect
codings (observations incorrectly coded into other PNC utilization categories that
were recorded into "inadequate PNC utilization" based on the month of initiation) ..........66

3-9 Step 4: Adequacy of Prenatal Care Utilization (APNCU) Index frequencies after
recoding all cases from Table 3-8 into "inadequate PNC utilization"...............................66

3-10 Coding for raw income variable categories before collapsing categories........................ 67

3-11 Coding for collapsed income variable categories .................................................... 67

4-1 Univariate statistics for all categorical variables included in the bivariate and
m u ltiv ariate an aly ses....................................................................... .......... ............... ...... 7 9

4-2 U nivariate statistics (continuous variable)................................................. ... ... ............... 82

4-3 Chi-square analyses comparing 41 characteristics among women with postpartum
depressive (PPD) symptoms versus women without postpartum depressive (PPD)
sy m p to m s ................... ................... ....................................................... .. 8 3

4-4 Maternal age (continuous variable) and postpartum depression (PPD) symptoms t-
te st resu lts .... ............... ....................................... ........ .......... ...... 8 8

4-5 Chi-square analyses comparing 40 characteristics among women from four body
m ass in dex (B M I) group p s .............................................. .................................................. 89

4-6 Chi-square analyses comparing 39 characteristics among women from four prenatal
care (P N C ) utilization group s ..................................................................... ..................... 95









4-7 Primary baseline logistic regression with the main effect independent variables .............99

4-8 Specific aim 1: Primary risk-adjusted logistic regression with the main effect
independent variables and control variables.............................................. ............... 99

4-9 Specific aim 2: Primary risk-adjusted logistic regression with the main effect
independent variables, interaction effect variables, and control variables....................... 101

4-10 Wald tests for pre-pregnancy BMI/PNC interaction terms: Primary risk-adjusted
logistic regression....................................................... 103

4-11 Secondary baseline risk-adjusted logistic regression (healthy pregnancies only) with
pre-pregnancy BM I and PNC utilization................................................. ........ ....... 103

4-12 Specific Aim 1: Secondary risk-adjusted logistic regression (healthy pregnancies
only) with the main effect independent variables, interaction effect variables, and
control variables .... ... ............ ........................... ................ 104

4-13 Specific aim 2: Secondary risk-adjusted logistic regression (healthy pregnancies
only) with the main effect independent variables, interaction effect variables, and
control variables .... ... ............ ........................... ................ 105

4-14 Wald tests for pre-pregnancy BMI/PNC interaction terms: Secondary risk-adjusted
logistic regression (healthy pregnancies only) ............... ...................................... 107

B-l Chi-square analyses comparing 40 characteristics among women who utilized
adequate plus PNC versus women who utilized "other quantities of PNC" .................... 164

B-2 Maternal age (continuous variable) and adequate plus PNC t-test results .......................171

B-3 Logistic regression for adequate plus PNC to determine significant predictors of
ad e qu ate p lu s P N C ............................................................................ ............................. 17 2

B-4 Risk-adjusted ordinal logistic regression sensitivity analysis with the main effect
independent variables, interaction effect variables, and control variables..................... 173

B-5 Logistic regression for women who received WIC services during pregnancy............. 175

B-6 Logistic regression for women with very low income.................. ................................... 177

B-7 Logistic regression for women with low income ...................................................179

B-8 Logistic regression for women with moderate income................................................ 181

B-9 Logistic regression for women with higher income.......................................................... 183

B-10 Logistic regression for women who received weight gain discussion............................ 185









LIST OF FIGURES


Figure page

2-1 T theoretical fram ew ork ................................................................... .................... 50

4-1 Primary risk-adjusted logistic regression with postpartum depression (PPD)
symptom odds ratios for each interaction effect variable............................................... 108









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

OBESITY AND POSTPARTUM DEPRESSION: DOES PRENATAL CARE UTILIZATION
MAKE A DIFFERENCE?

By

Swathy Sundaram

August 2009

Chair: Jeffrey S. Harman
Major: Health Services Research

Using a national continuing population-based survey known as Pregnancy Risk

Assessment Monitoring System (PRAMS), this study sought to determine the role that PNC

utilization plays in the relationship between pre-pregnancy BMI and PPD symptoms. Two years

of data, 2004 and 2005 were analyzed among women from 16 states. Two specific aims were

examined: 1) the association between pre-pregnancy BMI and PPD symptoms, and 2) the

association between pre-pregnancy BMI and PPD symptoms after considering PNC utilization as

a moderating variable. It was predicted for the first specific aim that the odds for PPD symptoms

would increase as pre-pregnancy BMI increased. For the second objective, it was predicted that

the association from the first specific aim would carry over and remain the same (e.g., obese pre-

pregnancy BMI would have the highest odds for PPD symptoms), but that within each pre-

pregnancy BMI group, the odds for PPD symptoms would decrease as PNC utilization increased

(within obese pre-pregnancy BMI, inadequate PNC would have higher odds than intermediate

PNC). The general premise for PNC utilization acting as a moderating variable in this study was

that PNC can help address the changes that occur during pregnancy with regards to pre-

pregnancy BMI (as a biological and psychosocial stressor). Thus, delivering PNC incorporating









nutrition, weight and shape changes, and addressing a woman's concerns about her weight and

shape would in-turn, reduce the odds of PPD symptoms.

Since the sample used in this study included women from all pregnancy risk statuses, two

risk-adjustment approaches were carried out to identify an association between pre-pregnancy

BMI and PPD symptoms, and a moderating effect of PNC. One approach included all women in

the dataset and used statistical analyses to risk-adjust for pregnancy risk status, and the other

approach modified the design of the study by truncating the population of women to include

healthy pregnancies only. Results initially showed an association between obesity and PPD

symptoms, and PNC and PPD symptoms among the bivariate and multivariate analyses.

However, the inclusion of a variety of control variables into the multivariate models removed

these associations. Overall, for both approaches, there was no indication of a moderating effect

of PNC utilization. However, results from the analyses showed that many of the women were

significantly affected by a variety of medical and obstetric problems, many of which were high-

risk. It is recommended that future research investigate the possible association of these

problems with PPD symptoms. For practice, it is suggested that PNC providers identify the

medical and obstetric problems faced by their patients, focus on both the physical and the

potential psychosocial consequences of those problems, and establish suitable interventions

accordingly.









CHAPTER 1
STATEMENT OF PROBLEM

Why is Postpartum Depression of Concern?

Postpartum depression (PPD) is a mood disorder that involves a variety of symptoms

including fatigue, fears, anxiety, despair, thoughts of compulsion, loss of libido, and feelings of

inadequacy (Horowitz, Damato, Solon, Von Metzsch, & Gill, 1995). The relationship between a

mother and baby is crucial for healthy maternal and child health outcomes. A woman can

experience PPD anytime during the first year after the birth of her child (Epperson, 1999).

Symptoms include mood swings, sadness, anxiety, loneliness, and inconsistent sleeping patterns.

However, when these symptoms reach a level of intensity that begins to affect the well-being of

a woman and her daily functioning, a woman should seek treatment as these symptoms may

indicate PPD. A new mother may be unaware of 1) the normal physical changes that occur after

giving birth, and 2) her ability to care for the infant (American Academy of Pediatrics (AAP) &

American College of Obstetricians and Gynecologists (ACOG), 1992). Approximately 4-6

weeks after the delivery, the AAP & ACOG (1992) recommend that a woman should see her

physician for a postpartum examination that includes evaluation of the mother's current health

status and her adaptation to her infant. Since many women experience emotional distress to some

extent in the postpartum period, 1) the emotional status of a woman should be evaluated, and 2)

any counseling with regards to a woman's postpartum emotional distress should address future

health and future pregnancies.

Consequences of PPD include maternal aggression, neglect of the infant, and infanticide

(Reck et al., 2004). Other psychosocial factors associated with PPD include child care stress,

poor marital satisfaction, and low self-esteem (Appolonio, & Fingerhut, 2008). Mothers with

PPD are portrayed as being unresponsive to their infants, passive and intrusive, displaying









avoidance and withdrawal, and displaying low levels of influence, or affect (Reck et al., 2004).

The processes associated with childbearing (e.g., pregnancy, childbirth, childrearing, etc.)

warrant attention because they remain responsible for many maternal morbidities and mortality

(e.g., complications such as preeclampsia, hemorrhage, self-acceptance in the postpartum period,

etc.) (Misra & Grason, 2006). Thus, not only can these processes have a biological influence on

a woman's health, but they can also have a psychosocial influence on a woman's health and

well-being that occur during this time (Wisner et al., 2006).

Why is Obesity of Concern?

Obesity continues to burden our society in terms of increased prevalence of other diseases

(e.g., heart disease), increased health care costs (e.g. treatment), and poses increased risk for

disability and death. In addition to this, obesity presents many social, emotional, and aesthetic

problems, especially in developed countries (Rubinstein, 2006). For women who are obese and

pregnant, pregnancy-related consequences of obesity that result from high pre-pregnancy body

mass indices [(weight in pounds/square of height in inches) x 703] include increased risk for

other diseases (e.g., gestational diabetes) and a lower survival rate for premature babies (Colditz,

2002). Obesity and BMI have also been associated independently with delivery complications

including excessive blood loss, greater operating time, and increased likelihood for cesearean

section (American College of Obstetricians and Gynecologists, 2005). The association between

pre-pregnancy BMI and postpartum depressive symptoms has been demonstrated with this

association increasing as pre-pregnancy BMI increases (Carter, Wood Baker, Brownell, 2000;

LaCoursiere, Baksh, Bloebaum, & Varner, 2006; Andersson, Sundstrom-Poromaa, Wulff,

Astrom, & Bixo, 2006).









What is the Importance of Prenatal Care?

Since the health of the infant is determined significantly by the health of the mother,

addressing issues during pregnancy itself can minimize adverse maternal and infant as well as

child outcomes later on. According to Healthy People 2010 (2000), prenatal care (PNC) should

start early on in the pregnancy and continue all through the pregnancy period; the effectiveness

of PNC is more likely if PNC is received early in the pregnancy. An ideal setting to discuss

issues (e.g., weight) during pregnancy is during prenatal care. It is suggested that pregnancy is a

good time to target changes in health behavior due to a woman's motivation to maximize the

health of her child (Birdsall, Wvya, Khazaezadeh, & Otegn-Ntim, 2009). Prenatal care (PNC) is

defined as the care a woman receives in the period during pregnancy, leading up to the time she

gives birth; adequate PNC is vital for both the mother and her developing baby (National

Institute of Child Health and Human Development, 2007). Prenatal care (along with obesity) has

been noted as two of the four special concerns for women's health (Torpy, Burke, & Glass,

2006).

Gaps in the Literature: Weight Issues Related to the Pregnancy and Postpartum Periods

Regarding previous research, Walker, Timmerman, Minseong, & Sterling (2002) found

weight as a leading factor for postpartum dissatisfaction among women of all ethnicities.

However, their sample size was not nationally representative and did not include all income

groups. Lebanon. Fox, & Yamaguchi (1997) found that women who were overweight before

pregnancy were more likely to have positive changes in body image during pregnancy compared

to normal weight; However, the women who were overweight before pregnancy also had more

negative concerns about body shape than normal weight women. This study did not address

associations between body weight/body shape and PPD, and the sample size was limited to

women in London. Moran, Holt, & Martin (1997) found that among postpartum health concerns,









the highest percentage of women in the sample wanted more information on nutrition, exercise,

and dieting. However, the study did not look at PPD. LaCoursiere et al. (2006) found a

significant association between pre-pregnancy obesity and moderate or greater postpartum

depressive symptoms. However, though all BMI categories were included, their sample size was

not nationally representative. Andersson et al. (2006) did not find any significant associations

between first-trimester BMI and a new-onset episode of postpartum depression. Also, BMI data

was missing for 8% of their sample and their sample size was limited to women in Sweden.

Carter, Baker, & Brownell (2000) found an association between BMI and anxiety/postpartum

depressive symptoms. However, their study had a small sample size and their BMI categories

included those with BMI <27, and those with BMI >27.

Gaps in the Literature: Prenatal Care and Postpartum Depression

Chaaya et al. (2002) looked at the determinants of PPD. Though they commented on the

importance of PNC in addressing the needs of pregnant women, they did not include PNC or

BMI in their analysis, and their sample size was limited to women in Lebanon. El-Kak, Chaaya,

Campbell, & Kaddour (2004) found that more PNC visits were associated with fewer cases of

PPD. However, their sample size was also limited to women in Lebanon. Nalepka & Coblentz

(1995) hypothesized that educating women on PPD prior to childbirth would reduce the

likelihood of PPD among the women compared to the women who did not received education;

however, they found no significance and the education was delivered in childbirth classes during

pregnancy as opposed to PNC, per se.

Therefore, since 1) there are limitations of previous studies on weight and postpartum

distress, 2) there is a paucity of literature on prenatal care and postpartum depression, and 3)

many significant changes occur during pregnancy, changes which pregnant women can be









educated on during the delivery of PNC, the relationship between PNC and PPD in the U.S can

further be ascertained in the literature (Merritt, Kuppin, & Wolper, 2001).

Purpose of This Study

Since studies have confirmed both an association between obesity and PPD, and PNC and

PPD, I would like to combine these two relationships. I propose that PNC can be seen as a means

for addressing weight concerns in order to ensure healthy pregnancy and postpartum outcomes

for her and her baby. To my knowledge, no study has determined if 1) an association exists

between pre-pregnancy BMI and PPD in the U.S., and 2) if an association exists between pre-

pregnancy BMI and PPD after considering PNC utilization as a moderator. The purpose of this

study is to determine if any existing association between pre-pregnancy BMI and PPD symptoms

is weakened after considering a woman's pre-pregnancy body mass index. Since women from all

pre-pregnancy body mass index groups seek PNC, I would like to see if the association of pre-

pregnancy BMI and PPD symptoms differ by PNC utilization level. The primary rationale for

PNC utilization acting as a moderating variable is based on the premise that PNC can be seen as

a means for providers to help women address any negative attitudes towards weight gain and

body image that may develop during pregnancy. I hypothesize that PNC plays an important role

in reducing the likelihood of PPD among women from all pre-pregnancy body mass index

groups. Identifying 1) an association between pre-pregnancy BMI groups and PPD among

pregnant women, and 2) seeing if the association differs for women from different PNC

utilization groups can substantiate the role (e.g., prevention) that PNC plays in reducing the

likelihood of an adverse maternal and child health outcomes (PPD symptoms).

Specific Aims and Hypotheses

Little is known about the relationship of BMI and PPD after considering a woman's PNC

utilization. Since the literature has suggested 1) an association between pre-pregnancy BMI and









weight gain during pregnancy, 2) an association between weight gain during pregnancy and

weight issues postpartum (e.g. weight retention), 3) an association between obesity and

depression, 4) an association between pre-pregnancy BMI and PPD, and 5) an association

between PNC and PPD, in accordance with the recommendations suggested in the literature, I

propose that PNC can help assuage effects that pre-pregnancy BMI (as a biological and

psychosocial stressor) and the changes that occur during pregnancy may have on women, such as

its influence on any negative attitudes about weight gain and body image that may arise either

during pregnancy or in the postpartum period. Consequently, I predict that if health care

providers deliver PNC information incorporating nutrition, weight and shape changes, and

address a woman's concerns about her weight and shape, this will in-turn, reduce the likelihood

of PPD.

The objective of this study is to determine the importance of PNC in acting as a

moderating variable in the relationship between pre-pregnancy BMI and PPD (e.g., weight issues

and lifestyle behaviors addressed through PNC may reduce the likelihood of weight issues

experienced after delivery, and reduce the likelihood of PPD). The specific aims are as follows:

* Specific aim 1: What is the association of pre-pregnancy body mass index (BMI) with
subsequent development of postpartum depression (PPD) symptoms?

* Hypothesis: I predict that women who had a pre-pregnancy BMI of obese will have the
highest odds for PPD symptoms, followed by overweight BMI, and finally underweight
BMI (lowest odds for PPD symptoms). Women who had a normal pre-pregnancy BMI will
be the reference group.

* Specific aim 2: Does PNC moderate the relationship between pre-pregnancy BMI and PPD
symptoms?

* Hypothesis: Within each pre-pregnancy BMI category, the likelihood a woman will
experience PPD symptoms will decrease as prenatal care increases: inadequate PNC will
have the highest odds for PPD symptoms, followed by adequate plus PNC, and finally
intermediate PNC (lowest odds for PPD symptoms). Women who utilized adequate PNC
will be the reference group.









Thus, if PNC acts as a moderating variable, as predicted, after looking at the relationship

between pre-pregnancy BMI and PPD symptoms only, that relationship will change as the

likelihood for PPD symptoms will change when considering PNC. For example, there will be a

difference in a woman with a pre-pregnancy BMI of obese who received adequate plus PNC

versus a woman with a pre-pregnancy BMI of obese who received intermediate PNC. The next

chapter elaborates on the literature to propose an idea of what occurs during PNC in addressing

concerns related to weight that arise during pregnancy, explaining how this may in-turn, reduce

the likelihood of PPD symptoms.









CHAPTER 2
LITERATURE REVIEW

Postpartum Depression (PPD)

Postpartum depression (PPD) is a mood disorder that involves a variety of symptoms

including fatigue, fears, anxiety, despair, thoughts of compulsion, loss of libido, and feelings of

inadequacy (Horowitz et al., 1995). PPD is known to be a very common illness, and affects

approximately one in every eight mothers to a point that affects her ability to carry out her

maternal responsibilities (Wisner, Parry, & Piontek, 2002). PPD is divided into three categories:

1) blues, which affect roughly 50-80% of new mothers, and is considered to be normal, 2) non-

psychotic postpartum depression, which affects roughly 10-15% of new mothers, the incidence

being on average 13%, and 3) postpartum psychosis, which is rarer than the other two types and

occurs in roughly 1-2 out of every 1000 pregnancies or 0.1-0.2% of mothers (Miller, 2002; Evins

& Theofrastous, 1997; Negus Jolley & Betrus, 2007). It is the conditions of labor, delivery, and

the postpartum period that are predicted to bring about a traumatic level of stress that can trigger

postpartum depressive symptoms (Dietz et al., 2007).

Consequences of PPD

Da Costa, Dritsa, Lowensteyn, & Khalife (2006) showed that women experiencing PPD

suffered significant reductions in health related quality of life, with the association continuing

even after controlling for depression severity. PPD has been shown to have negative

consequences on the child's behavior and development, mother-child interaction, and parenting

practices (Minkovitz et al., 2005). PPD may also affect a mother's health care utilization for her

child. For example, if fewer preventive measures are taken for the child such as lack of

vaccinations, this can in-turn, affect the physical health of her child, which can bring about an

increase in acute care utilization for the child (Minkovitz et al., 2005). This relationship is









suggested because health promotion activities taken on by the mother for her child are associated

in part with the functional capacity of the mother. The functional capacity may be affected if the

mother's psychological well-being is compromised. (Rahman, Iqbal, Bunn, Lovel, & Harrington,

2004). Also, children of a depressed parent have a higher likelihood of experiencing negative

cognitive and social outcomes (e.g., lack of social competence), and have a higher rate of mental

illnesses that can continue into adulthood, with this likelihood increasing if both parents

experience depression (NICHD Early Child Care Research Network, 1999; Lieberman, 1977;

Weissman et al., 2006; Goodman & Gotlib, 1999). Since the primary figure in a child's life tends

to be the parent, usually the mother in many families, parental depression may affect the quality

of the relationship between the parent and child, and even cause behavioral problems for the

child later on in life, such as anxiety (Radke-Yarrow, Cummings, Kuczynski, & Chapman, 1985;

Lieberman, 1977). Children of mothers with PPD also have a higher likelihood of receiving

lower scores on measures of mental and motor development, have more difficult temperaments,

react more negatively to stress, and lower self-esteem (Goodman & Gotlib, 1999). Finally,

physical consequences associated with depression for the mother around the time of childbirth

include low birth weight and impaired growth for the child (Rahman, Iqbal, Bunn, Lovel, &

Harrington, 2004).

Screening for PPD

Recognizing that mental health, as well as physical health, is important for the

mother is essential for her overall well-being (Wisner et al., 2006). Since a new mother may be

unaware of 1) the normal physical changes that occur after giving birth, and 2) any limits of her

ability to care for her infant (AAP & ACOG, 1992), screening is the first step in detecting PPD

(Negus Jolley & Betrus, 2007). Approximately 4-6 weeks after she gives birth, the American

Academy of Pediatrics and the American College of Obstetricians and Gynecologists (1992)









recommend that a woman should see her OB-GYN for a postpartum examination to determine

her current health status and her adaptation to her infant. Since many women experience

emotional distress to some extent in the postpartum period, the emotional status of a woman

should be evaluated during this time. Also, any counseling regarding a woman's postpartum

emotional distress should address future health and future pregnancies.

A wealth of literature exists that supports screening as an effective way to combat the

consequences of PPD. Identifying women at risk for PPD has been previously identified as a

preventive method for PPD (Boyce & Hickey, 2005). A universal screening system for PPD is

suggested which includes screening for PPD as soon as two weeks after birth and no later than a

year after birth (Wisner et al., 2006). Questions however exist as to whether PPD screening

actually leads to improved maternal and child health outcomes (Gaynes et al., 2005). It is advised

that detecting for women who are at-risk for PPD symptoms can be done in the late stages of

pregnancy (Josefsson, Berg, Nordi, & Sydsjo, 2001). Screening for risk factors and/or depressive

symptoms would be conducive to early detection and initiation of treatment (Miller, 2002). This

screening can indeed be incorporated into both prenatal clinics (during PNC delivery) and in

pediatric clinics (during postpartum check-up visits) (Miller, 2002).

Postpartum Depression (PPD): Women and Men

Research has shown that women are two times more likely than men to suffer from

depression (with the dominance of depression affecting women consistent across developed

nations), and that the first onset tends to be during the reproductive years (Weissman & Olfson,

1995). In addition, women tend to experience a longer duration of depression and a higher

frequency than that of men (Sargeant, Bruce, Florio, & Weissman, 1990). Men experience PPD

as well as women; however, the literature suggests that women, unlike men, have a higher

likelihood of suffering PPD due to hormonal withdrawal (e.g., gonadal steroids such as estrogen









and progesterone) experienced by a woman in the postpartum period, with this likelihood

increasing with parity (Bloch et al., 2000). Maternal PPD is correlated with paternal PPD, and

both can affect family health by affecting other relationships within the family and eventually the

well-being of the family (Deater-Deckard, Pickering, Dunn, & Golding, 1998; Goodman, 2004).

Since men are becoming more involved in the experience of having a newborn in the house,

compared to previous decades, there are now greater possibilities for men to experience PPD

(Goodman, 2004). Combined PPD of both the mother and father puts the child at an increased

risk for developmental problems than would occur with maternal PPD alone (Goodman, 2004).

Obesity

Obesity is defined as having a body mass index (BMI) of 30 or greater (National Institutes

of Health, 1998). Body mass index, a measure that represents the comparative weight to height,

is recommended by the Centers for Disease Control (CDC) as a reliable body fat indicator (it is

significantly correlated with the total fat content in the body) and as an excellent method for

assessing both overweight and obesity (National Institutes of Health, 1998). Obesity has been

shown to be associated with an increased risk for other diseases, including hypertension,

cardiovascular disease, cholecystectomy, non-insulin dependent diabetes mellitus, and colon

cancer (National Institutes of Health, 1998). It is estimated that about 65% of Americans 21

years and older have a BMI more than 25 (overweight classification), 30.5% have a BMI of30

or more (obese classification), and 4.9% have a BMI of 40 or more (extremely obese

classification) (Sarwer, Allison, Gibbons, Markowitz, & Nelson, 2006). Rates of obesity

continue to increase, especially among childbearing women. According to Lu et al. (2001) the

average maternal weight of women in the initial prenatal care visit increased by 20%, and the

percentage of women classified as obese increased from 7.3% to 24.4% over a period of 20

years. Along with the growing rates of obesity and the physical consequences of obesity on









health, there exist many psychosocial consequences that can have effects on the mental health of

individuals who are obese.

Psychosocial Consequences of Obesity: Stereotypes and Stigmas

In addition to its effects on physical health as identified above, obesity also poses

psychosocial consequences. There exists a plethora of negative attitudes and stigmas associated

with being obese. Examples include discrimination and prejudice with respect to arenas such as

health care and employment (Crerand, Wadden, Foster, & Gary, 2007). This discrimination may

have health care access implications for obese postpartum women who are in need of mental

health care to address PPD. According to Wooley & Wooley (1979), and apart from skin, having

excess body fat is known to be the most stigmatized physical feature. However, unlike skin,

excess body fat can be voluntarily controlled. The results of stigmatization of overweight/obese

individuals include self-victimization and having to suffer differential treatment due to physical

appearance, with women suffering more than men (Wooley & Wooley, 1979). Stereotypes

affiliated with being overweight and obese include being self-indulgent, less self-disciplined, less

attractive, less happy, and lazier (compared to thin counterparts) (Tiggeman & Rothblum, 1988).

Also, it is noted that obesity is more prevalent in women than it is in men (Hedley et al., 2004).

Women have double the likelihood over men to experience a major weight gain with an increase

over a period of 10-years and overweight women in the 25-44 year age group have the highest

incidence for significant weight gain a compared to all other groups (Williamson, Kahn,

Remington, & Anda, 1990); these years are crucial regarding child-bearing years. It is suggested

that although both men and women experience stigmatization with obesity; the effect is more

profound towards women than for men as women possess a greater propensity for obesity (Hebl

& Turchin, 2005; De Garine & Pollock, 2005).









Psychosocial Consequences of Obesity: Weight as a Chronic Stressor

Looking at attitudes and beliefs held by many women, there are many chronic stressors

that American women experience on a daily basis. The association of a slender figure with

attractiveness and beauty and its effects on women's body perceptions and body images remains

one of the more significant chronic stressors (Attie & Brooks-Gunn, 1987). This is due to

cultural and societal influences that can be influential to an extent that a woman does not

question the validity of her perceptions that she is overweight or has an undesirable figure; even

if she has alternatives to negative body perceptions (e.g., accepting her weight and body shape),

she may face confrontation to these alternative perceptions from outside influences (e.g., peers,

spouse, etc.) (Attie & Brooks-Gunn, 1987). For women, especially in the Western world, the

body image tends to support a slim body, and there continues to exist a dilemma between desire

versus control (De Garine & Pollock, 2005). In fact, many American women tend to experience

negative ramifications when their weight exceeds society's expectations (Cameron et al., 1996).

What is socially accepted is a thin, lean body, which remains symbolic of characteristics such as

self-control, hard work, attractiveness, success, acceptance, being physically fit and healthy, and

in general, having desirable personal qualities (Brownell, 1991). The phrase "thin is beautiful"

tends to lead to fear of and prejudice against overweight/obese individuals (Attie & Brooks-

Gunn, 1987). Women tend to experience greater pressures to conform to being thin as there are

more positive attitudes towards thinness (De Garine & Pollock, 2005). Karlsson, Taft, Sjostrom,

Torgerson, & Sullivan (2003) found that obese women reported having more psychosocial

problems related to weight than men. Society's focus on body image can take a toll on a

woman's emotional and physical health to an extent where thinness may take primacy over

health (Paquette & Raine, 2004). For example, this focus can cause many women to 1) undertake

dangerous weight loss behaviors, 2) experience poor body image and low self-esteem issues, and









3) general unhappiness that could lead to depression (Battle & Brownell, 1996). In fact, many

studies have confirmed as association between obesity and depression.

Obesity and Depression

Not only is obesity known as the most common chronic illness in today's society, but

depression is considered to be the second most prevalent psychological illness in today's society,

following anxiety disorders (Dixon, Dixon, & O'Brien, 2003). Depression is known as the

leading cause of disability globally (Kruijshaar, Hoeymans, Spijker, Stouthard, & Essink-Bot,

2005). Depression is an independent risk factor for premature morbidity and mortality, especially

when combined with congestive heart failure, hypertension, and/or stroke, and is associated with

higher health care utilization and higher total healthcare expenditures, and loss of productivity

(Schulz et al., 2000; Olfson & Klerman, 1992; Greenberg et al., 2003).

Both obesity and depression are known to be among the most prevalent and most costly

public health problems in the United States, and are associated with increased health care

utilization, which can in-turn result in increased health care costs (Kress, Peterson, & Hartzell,

2006). The combined effect of both can increase an individual's risk for loss of function as both

may affect one another (Markowitz, Friedman, & Arent, 2008). There is increasing evidence that

obesity and depression are related; those who are overweight or obese are more likely to feel

depressed at least one week during the month (Aberdour, 2006). Plutchik (1976) found that the

greater the degree of being overweight, the greater the tendency to experience problems with

depression. Roberts, Deleber, Strawbridge, & Kaplan (2003) found that obesity at baseline was

associated with depression and that obesity predicts depression subsequently. Since many obese

individuals are forced to endure discrimination and the stigmas associated with obesity, this can

certainly contribute to psychosocial distress, if not depression; however, Dixon et al. (2003)

confirmed that obesity is associated with depression. Increased BMI is associated with a higher









risk for depression, thoughts of suicide, and suicide attempts (Carpenter, Hasin, Allison, & Faith,

2000). Onyike, Crum, Lee, Lykestsos, & Eaton (2003) found that obese persons in their sample

had a higher prevalence of depression in the past-month than their normal weight counterparts.

For the women in the sample, there was an 82% increase in the odds compared to the 73% higher

odds in men (which was non-significant). The prevalence of depression was the highest for those

obese who were obese with the strongest association remaining for those who were severely

obese (BMI >/= 40). Since treatment for both depression and obesity is costly and only available

to few people prevention is important in curbing obesity (Battle & Brownell, 1996).

Women also tend to be at a higher risk for obesity-related costs, and there is an increased

risk of depression among women who are obese (Kress, Peterson, & Hartzell, 2006). Research

has shown that obesity is associated with depression in females (Dong, Sanchez, & Price, 2004).

Heo, Pietrobelli, Fontaine, Sirey, & Faith (2006) found that among young women, those

overweight and obese were significantly more likely to have experienced depressive moods

compared to young women who were not overweight or obese, with those who are Hispanic

being more susceptible. Linde et al. (2007) found that being overweight or being obese is

associated with depression, especially among women. Jorm et al. (2003) found that in women,

obesity was associated with more depressive symptoms and lower well-being whereas in men,

the associations were weak and inconsistent.

Studies have also shown that being underweight is associated with depression. Lox,

Osborn, & Pellet (1998) found that women who perceive themselves as underweight experience

similar psychosocial issues (e.g., self-esteem, depression, body dissatisfaction, anxiety) as

women who perceive themselves as overweight. Carpenter et al. (2000) found a U-shaped

relationship among BMI values and their association with increased probability of depression;









low and high BMI values were associated with an increased probability of having experienced

major depression. However, since the relationship between obesity and depression has been

demonstrated to a greater extent in the literature, suggesting that obese women have a higher

likelihood for experiencing mental distress and depression associated with weight, taking this

relationship further and looking at a special group of women, postpartum women, could further

add to the literature on the effects of weight, specifically obesity, on mental health; in this case,

during the postpartum period.

Obesity and PPD

A woman undergoes a significant amount of stress during the course of pregnancy. Since

giving birth is an important time in a woman's life and safely delivering a healthy baby is a

critical concern, this can be stressful for a woman in ensuring a healthy pregnancy outcome. In

addition, the effect of pregnancy on body perceptions and/or self-esteem can also act as a chronic

stressor (Hobfoll & Leiberman, 1987). In fact, even though dietary restraints are less prevalent in

pregnant women (Davies, & Wardle, 1994), body perceptions of women tend to be increasingly

negative, especially during the early to mid-second trimester of pregnancy (Skouteris, Carr,

Wertheim, Paxton, & Duncombe, 2005). Though some research has shown that these perceptions

may become less negative in the postpartum period, body perceptions during the postpartum

period tend to be less positive than before pregnancy (Strang & Sullivan, 1985). In fact, negative

body images are likely to be associated with weight distress during the postpartum period.

Obesity has been shown to have a significant association with a new-onset episode of postpartum

psychiatric disorder (LaCoursiere, Baksh, Bloebaum, & Varner, 2006; Andersson et al., 2006).

In addition, after controlling for marital status and income, pre-pregnancy obesity (defined as

having a BMI greater than 29) was found to be associated with having moderate or greater









postpartum depressive symptoms, with the strength of this association increasing as body mass

increased above the normal BMI range (LaCoursiere et al., 2006.)

Prenatal Care (PNC)

It is believed that what is vital for every person is his/her health and the aptitude to work

efficiently. An important resource for society is an infant who is born with the ability to function

well in society. If an infant is born at a disadvantage, with a condition that may prevent

maximum functioning, this may be detrimental for the individual and the community. The period

of pregnancy is a time that provides an opportunity to address lifestyle behaviors that remain

important to a woman both during the course of her pregnancy and after she gives birth (e.g.,

smoking, nutrition, exercise, violence, etc.), many of which may have implications for infant and

child outcomes (e.g., maternal obesity is a strong predictor for metabolic syndrome among

children) (McCormick & Siegel, 2001; Boney, Verma, Tucker, & Vohr, 2005). One venue that

may be used to alter the health behaviors of pregnant women is the advice and encouragement

rendered by health care providers through prenatal care (PNC) (Kogan, Kotelchuck, Alexander,

& Johnson, 1994). Prenatal care may be seen as a means to allow women to participate in their

own health; an example being the change of their health behaviors during pregnancy to

incorporate healthy eating (McCormick & Siegel, 2001). American women deem routine PNC to

be essential as they strongly believe in the importance and efficacy of PNC and will invest

efforts to "make a good baby" (Press & Browner, 1997; Rubin 1984, p.65); that is the message

delivered by their health care providers, even though there is limited evidence to support the

direct benefits of routine PNC on birth outcomes. For many women, the most significant aspect

of PNC observed is that they are provided with information about the pregnancy and the growing

fetus. This is information that women find to be encouraging and empowering (Press & Browner,

1997). The purpose of PNC is to 1) find the pregnant women with problems, 2) assure









management of the problems identified, 3) prepare both women and their partners for delivery

and child care, 4) provide information, 5) provide health education, and 6) provide support to all

pregnant women (Hemminki, 1988). Thus, PNC includes services that are intended to improve

outcomes for the mother and infant, as well as promote educated decision-making among the

mother, family members and friends, with regard to health care during the pregnancy (Daniels,

Fuji Noe, & Mayberry, 2006), and even in the postpartum period.

Prenatal care is known as a key preventive service for pregnant women (Kogan,

Alexander, Kotelchuck, Nagey, & Jack, 1994) and has been accepted as an important conduit to

prevent harm for the mother and child. Prenatal care is rendered in a variety of settings

including: 1) private clinics of physicians, osteopaths, and midwives, 2) university hospital

clinics, 3) health maintenance organizations, 4) community health centers, 5) public health

departments, 6) migrant health centers, 7) community hospital clinics, 8) university hospital

clinics, 9) schools, and 10) military facilities. Those involved with the delivery of PNC include

family practice physicians, obstetricians and gynecologists, midwives including nurse midwives,

osteopaths, nurses, and nurse practitioners (Peoples-Sheps, Kalsbeek, & Siegel, 1988). It is

important to note that for many women who are pregnant for the first time, PNC may be the first

point of adult contact with the health care system; thus, their experiences with PNC may

influence subsequent use of the health care system for themselves, their partners, and their

children (Alexander & Kotelchuck, 2001). For all the women who seek PNC, most of them see a

physician at some point during the course of their pregnancy; however, many women see

multiple providers. (Peoples-Sheps et al., 1988). Given that women are provided with

information from their health care providers during the course of their pregnancy, it is believed









that pregnancy is a time when a woman has the highest likelihood to make changes in her

lifestyle behaviors, more than any other time during her life (Higgins, Frank, & Brown, 1994).

Prenatal Care (PNC), Nutrition, and Weight: Behavior Modifications

It is confirmed that the pregnancy period is a time when women make significant changes

in their health behaviors (Baric & MacArthur, 1977) as most women are motivated to do what is

necessary to enhance the likelihood of having a healthy baby (Higgins & Woods, 1999); for

example, many women are motivated to change and/or modify their nutrition and fitness health

behaviors during pregnancy (Wood Baker, Carter, Cohen, & Brownell, 1999). Pregnancy is also

an ideal time to encourage women to initiate healthy lifestyle behaviors such as exercise (e.g.,

walking) and dietary habits (e.g., proper foods) (Morin & Reilly, 2007). The most common

modifications made during pregnancy include exercise, nutrition, and reduced substance abuse

behaviors (Higgins, Clough, Frank, & Wallerstedt, 1995). Kline, Martin, & Deyo (1998) found

that women reported that the pregnancy and postpartum periods motivated them to reduce any

risky behaviors due to the fear that their children would be affected. The health behaviors that

occur, the quality of the diet consumed, and the amount of weight gained during the pregnancy

period are significant because of 1) its impact directly on the health and well-being of the mother

(both short-term and long-term), and 2) its impact on the development of the growing fetus

(Robb-Todter, 1996). Thus, commonly addressed concerns during pregnancy include weight

gain and nutrition (e.g., nutrient intake) because both can affect the health of the mother and

infant. Chomitz, Cheung, & Lieberman (1995) suggest that adopting healthy lifestyle behaviors

during pregnancy can result in positive long-term health for the women and their

infants/children. For example, having a balanced diet is important for women in their

reproductive years, especially for women who are pregnant, in order to enhance the health,

survival, and development of their children (Mora & Nestel, 2000). Among the things a woman









can do to heighten the likelihood of giving birth to a normal, healthy child are modifying

lifestyle behaviors such as unhealthy dietary habits that may pose a risk of affecting the

likelihood of delivering a healthy infant (Chomitz et al., 1995). Habits that are detrimental can be

difficult to modify, but can be done with the support and assistance from family members and

other close individuals, the health care system, and society; for example, modifications can be

accomplished with the assistance of education that is relayed through PNC (Chomitz et al.,

1995). In fact, many women seek pregnancy-related information through their PNC provider. For

example, Risica & Phipps (2006) noted that the information topics that were most frequently

requested by women to discuss with their PNC provider were eating well and staying fit,

followed by caring for a newborn, breastfeeding, healthy weight gain, gestational diabetes,

genetic testing for their baby, and smoking cessation. Other requests for information included

depression during and after pregnancy, working after giving birth, and preterm labor. Though the

success of using written materials to deliver nutrition education has been demonstrated in the

literature (Beresford et al., 1997) found that a self-help book endorsed by physicians and given to

patients who were looking to modify their dietary lifestyle habits was successful in helping those

patients decrease their fat intake over the course of a year and increase their fiber intake), the

authors found in their study that the women preferred to receive this information from a PNC

nurse or provider rather than from other materials (e.g., printed materials, videos, classes,

internet, CD-ROM). Thus, it is the PNC providers that can educate, assist guide, and work with

their pregnant patients in helping them make healthy behavioral changes and/or modifications

related to nutrition and wellness.









What Should the Content of PNC Entail?

In order to encourage pregnant women to modify and/or adopt healthy behavioral changes

related to nutrition and wellness, it is ideal for PNC to be comprehensive and inclusive of care

that priorities nutrition and wellness. PNC content is generally comprised of prevention (e.g.,

education), detection (e.g., birth defects), and treatment services as well as interventions

designed to focus on psychosocial issues (e.g., stress) and change health behaviors that may

prevent healthy pregnancy outcomes (e.g., poor eating habits). Lederman, Alfasi, & Deckelbaum

(2002) recommend that PNC should include a significant focus on helping women optimize their

weight during pregnancy (e.g., informing women about the Institute of Medicine guidelines on

weight gain based on BMI). Many prenatal interventions such as PNC have focused on

overweight/obese women in efforts to prevent excessive postpartum weight retention/weight gain

(Walker, 2007). Prenatal care discussions should also include what it means to gain weight

during pregnancy and how the weight is distributed in the woman's body (e.g., between the

uterus, placenta, fetus, etc.). Also, discussions should include well-balanced diets that are high in

protein and would consequently have positive effects on a woman's body shape and her weight

(Moore, 1978).

Since the content of PNC had grown to include services related to nutrition and are

considered a vital part component of PNC (Wheatley, Kelley, Peacock, & Delgado, 2008),

nutrition education and guidance should be a critical component of all PNC services (Bronner &

Baldwin, 1999; Klohe-Lehman et al., 2006). Both the Institute of Medicine (1990), and the

American College of Obstetricians and Gynecologists (2005) have issued recommendations for

obstetricians-gynecologists including: (Institute of Medicine, 1990):

1) Health care providers should use reliable procedures for measuring height and weight of
pregnant women at each visit, set goals for weight gain, and monitor weight gain
throughout the term.









2) Calculate the woman's pre-pregnancy BMI

3) Estimate the woman' s gestational age

4) Determine a weight gain goal together with the woman at the beginning of her initial
prenatal care visit

5) Explain to the woman why weight gain is important

6) Discuss the recommended range of weight gain and the pattern of weight gain depending
on her pre-pregnancy BMI: 25-35 lbs for normal weight women, 15-25 lbs for overweight
women, and 15 lbs for obese women, record height and weight for women at all PNC visits

7) Monitor the woman's pattern of weight gain throughout her term and identify any
abnormal patterns that may necessitate the health care provider intervening.

8) Upon identifying any abnormal patterns (if applicable), determine the cause of the
abnormal weight gain and then determine ways to rectify the problem with the woman.

9) Evaluate a woman's dietary habits via a food history or a food frequency questionnaire;
include questions about problems or conditions which may affect her dietary habits and
behaviors.

10) Offer nutrition consultation to obese women and encourage them to adhere to an exercise
program during the pregnancy and postpartum, discuss pregnancy-related complications
due to weight.

Overall, it is recommended that PNC care include 1) a routine dietary assessment to

determine dietary needs (e.g. nutrient supplementation), and 2) guidance and support for women

on achieving a healthy, balanced diet and maintaining healthy behaviors that will support

adequate weight gain optimal health for the women and their fetuses (Institute of Medicine,

1990).

An association between weight gain advice from PNC providers during pregnancy and

actual weight gain during pregnancy has been demonstrated, suggesting that women can be

successfully encouraged to gain the appropriate amount of weight during their pregnancy (Taffel,

Keppel, & Jones, 1993). Keppel & Taffel (1993) showed that White women who had pregnancy

weight gain within the Institute of Medicine's guidelines retained fewer than four pounds in the

postpartum period, and had a median of 1.6 pounds. However, White women who gained more









weight during pregnancy than the recommended ranges had a higher likelihood of retaining nine

pounds or more (with a median of 4.9 pounds) compared to the women who did not exceed the

recommended amount of weight gain. As for Black women, they were more likely to retain

weight and were thus heavier in the postpartum period than White women (median of seven

pounds) even though the gestational weight gain was similar for both groups of women, and they

also consume higher total calories, a diet with a higher portion of calories from fat, and less

physical activity during the prenatal and postpartum periods than White women (Keppel &

Taffel, 1993; Boardley, Sargent, Coker, Hussey, & Sharpe, 1995). Also, as prenatal weight

increased, the postpartum weight retained increased. Among 4,218 women in the sample,

Carmichael, Abrams, & Selvin, (1997) found that 40% of the women gained weight within the

recommended ranges; out of this 40% of women, 53% of them were underweight BMI, 35% of

them were normal BMI, 24% of them were overweight BMI, and 27% of them were obese BMI.

While delivering PNC related to nutrition and wellness, PNC providers should be aware

and knowledgeable about the attitudes and feelings that many pregnant women hold regarding

body image and perceptions, especially closer to childbirth and in the postpartum period, as they

tend to be negative during these times (Moore, 1978; Stein & Fairbum, 1996). It is the PNC

providers that can assist women in viewing pregnancy and its associated physical changes as

ones that are normal and beautiful (Moore, 1978). It is recommended that health care

professionals intervene early in the pregnancy to assist women in modifying their dietary

patterns to achieve appropriate weight during these periods (Lederman, Paxton, Heymsfield,

Wang, Thornton, & Pierson, 1997); Since weight can be a stressor for a number of women

during pregnancy and in the postpartum period, health care professionals can 1) help women

engage in healthy behaviors to minimize weight retention in the postpartum period, and 2) help









women psychosocially (e.g., help women feel better about themselves and their weight) during

pregnancy and in the postpartum period.

Weight as a Stressor and the Importance of PNC

Affonso & Mayberry (1990) found that the commonly reported stressors of pregnant

women included weight gain and body changes for women in their first and third trimesters

(second most commonly reported stressor), and for women in the postpartum period (fourth most

commonly reported stressor). Among the total sample, body image changes was the second

highest commonly reported stressor. The authors suggest that assessments made in both the

prenatal and postpartum periods must address these stressors and determine what they mean to

women with regards to body image and body perception judgments, and how they can produce

discomfort and uneasiness if a woman feels that she is losing control over managing her body.

Addressing these issues during the prenatal period (e.g., during PNC) may assist women in

handling the intense emotions that follow childbirth and the imbalances that occur between

stressors commonly experienced after childbirth (e.g., weight and body image issues) (Hiser,

1987).

It is known that pregnant women experience a transition as their babies are developing, and

this transition includes changes in body images and even during pregnancy (as well as the

postpartum period), there is a significant amount of attention and importance that is given to

appearance as is in the pre-pregnancy period (McCarthy, 1998). The physical changes that occur

during pregnancy are extensive due to 1) growth of the woman due to growth of the child, and 2)

changes in boundaries of the body that occur in the third trimester of the childbearing period

(e.g., thinning of the abdominal and uterine walls that make the abdomen and uterus tight due to

stretching) (Rubin, 1984). The resulting stress that may occur due to changes in body shape and

size may cause distress both during the pregnancy and in the postpartum period (O'Hara,









Schlechte, Lewis, & Varner, 1991). For some women, the stress may be more profound if the

weight gained during pregnancy is excessive.

Pregnancy and Excessive Weight Gain

Women who gain excessive weight during pregnancy may be at a higher likelihood to

retain the weight after giving birth. For example, Lederman, Alfasi, & Deckelbaum (2002) found

that women who were obese prior to pregnancy were more likely to experience excessive weight

gain during pregnancy and postpartum. Wells, Schwalberg, Noonan, & Gabor (2006) showed the

following: 1) being underweight was associated with inadequate weight gain, but protective for

excessive weight gain during pregnancy, 2) being obese was associated with both excessive and

inadequate weight gain during pregnancy, and 3) being overweight was associated with

excessive weight gain, but protective against inadequate weight gain during pregnancy.

Olafsdottir, Skuladottir, Thorsdottir, Hauksson, & Steingrimsdottir (2006) found after comparing

women with a BMI less than 25, and 25 or greater that those in the latter category, specifically

those with a pre-pregnancy BMI between 25-29 were the most likely to gain excessive weight

during pregnancy. Thus, special attention should be given to women who are overweight before

their pregnancy because they are the most likely to experience excessive weight gain during

pregnancy. Consequently, they are also the most likely to experience pregnancy and delivery

complications such as preeclampsia (LaCoursiere, Bloebaum, Duncan, & Varner, 2005;

Saravanakumar, Rao, & Cooper, 2006; Cedergren, 2004; Rosenberg, Garbers, Lipkind, &

Chiasson, 2005; Rosenberg, Garbers, Chavkin, & Chiasson, 2003; Mahmood, 2009; Baeten,

Bukusi and Lambe, 2001; Cnattingius, Bergstrom, Lipworth, & Kramer, 1998) as well as

struggle with overweight/obesity issues after birth. These struggles may consequently result in

weight distress in the postpartum period due to issues such as weight retention (e.g., rigorous

dieting) (Olafsdottir et al., 2006; Shepard, Hellenbrand, Bracken, 1986).









Postpartum Weight Retention

In general, many women are concerned about excessive weight gain during pregnancy due

to their apprehension about postpartum weight retention (Keppel & Taffel, 1993). Excessive

weight gain during pregnancy can cause obesity issues postpartum, which may cause women to

restrict their food consumption in efforts to lose weight quickly. These restrictions in-turn may

weaken breastfeeding capabilities, and thus, obese women may cease breastfeeding earlier than

non-obese women due to weight issues (Lederman, Paxton, Heymsfield, Wang, Thornton, &

Pierson, 1997). According to Gunderson & Abrams (2000), factors that influence postpartum

weight retention include pre-pregnancy weight, race/ethnicity, parity, lactation capabilities, and

weight gained during pregnancy. Jenkin & Tiggeman (1989) found that women were more

dissatisfied with their postpartum weight than their pre-pregnancy weight, and that postpartum

weight was associated with psychological well-being with this association between weight and

dissatisfaction increasing as weight increased. The authors concluded that postpartum weight is a

predictor of psychological well-being in the postpartum period. Polley, Wing, & Sims, (2002)

found a significant and strong association between weight gain during pregnancy and postpartum

weight retention. They suggest that normal-weight women tended to retain less than the control

group, and for overweight women, they tended to retain more compared to the control group.

Lederman, Paxton, Heymsfield, Wang, Thornton, & Pierson (1997) suggest that high pregnancy

weight gain is likely to be associated with postpartum weight because many women have

difficulty in adjusting to energy intake and expenditure during those periods.

It is suggested that the effects of postpartum weight retention should be studied from a

psychosocial context with an emphasis placed on weight management (Walker, 1997). This is

because there is a significant amount of distress (associated with higher BMI) that is experienced

due to a high dissatisfaction with weight; and this may be followed by a lowered self-esteem









(Walker, 1997). Walker, Timmerman, Kim, & Sterling (2002) found that weight was the area

that women in the postpartum period experienced the most dissatisfaction, followed by distress

about the waist, hips, legs, and muscle tone (all of which tend to be areas where fat is retained in

the postpartum period). This dissatisfaction with body image was significantly associated with

postpartum depressive symptoms at six-weeks postpartum among women from all ethnic groups.

However, Suttie (1998) found that postpartum women were more concerned about their fitness,

less concerned with their appearance, and they also felt healthier compared to non-postpartum

women. Results also showed that women in the early postpartum period had similar body images

to women in the latter part of the postpartum period; however, the women in the earlier

postpartum period reported feeling healthier than the women in the latter part of the postpartum

period. Other changes expressed by the women in the postpartum period, though not as

distressing as changes with weight and figure included stretch marks, wrinkly skin, and

discoloration marks. Hiser (1987) found that concerns of women in the second postpartum week

included the following: weight was reported as a worry by 35% of the women, and 40% of the

women reported concern regarding having a flabby figure, while stretch marks were reported by

70% of the women as not being a general concern in the postpartum period. In general, the

women in the sample cited weight, flabby figure and returning their figures to normal as frequent

concerns in the postpartum period.

Given the extent that postpartum weight retention is suggested to pose long-term effects on

a woman's health and well-being (e.g., weight increases later on in life) (Walker, 2007), PNC

providers should address this issue during the delivery of PNC, Many women are unaware of the

postpartum consequences of not managing weight during pregnancy. Since the body rarely

immediately returns to its preconception shape following childbirth, many women tend to be









surprised by the extent to which they retain weight after childbirth (Stein & Fairburn, 1996;

Wood Baker, Carter, Cohen, & Brownell, 1999). For example, Fairburn & Welch (1990) found

in their sample that 38% of the women had no intention of trying to lose weight as they felt their

weight would return to normal. It is important for weight gain to be monitored during pregnancy

in order to determine if it is within recommended ranges. PNC providers should educate their

patients on how to lose weight following delivery in addition to educating women on how much

weight to gain during pregnancy (Keppel & Taffel, 1993). Thus, it is important that pregnant

women be given information related to food, nutrition, and weight during pregnancy. Olafsdottir

et al. (2006) recommend that women should be given guidelines about weight gain and lifestyle

modifications during pregnancy. Along with weight gain guidelines, pregnant women should

also receive advice on changes with respect to eating patterns and habits, as well as changes in

body shape and image that that are common among pregnant women. It has been suggested that

women who are overweight and/or obese may overestimate their prenatal physical activity

(Lichtman et al., 1992); Hence, if PNC providers work closely with their pregnant patients,

monitor physical activity, and engage in discussion about weight, nutrition, and physical activity

at PNC visits, this may, in-turn prepare women (e.g., less distress) for managing changes in body

shape and weight in the postpartum period (e.g., women who are encouraged to engage in

healthy behaviors and do so during pregnancy may carry on these behaviors into the postpartum

period).

Relation Between Pregnancy, Weight, and Postpartum Distress

Russell (1974) found that worrying about loss of figure was among the top five concerns

experienced by women in the postpartum period and thus, this can significantly affect a woman

during this time. Harris, Ellison, & Clement (1999) found that women tend to experience more

dissatisfaction with their body during the postpartum period than they were in the preconception









period. Four reasons may explain these feelings: 1) It is possible that mothers may experience a

stronger drive towards thinness in the postpartum period compared to the preconception period,

2) Mother may perceive that they are heavier postpartum than they were before pregnancy; this

may be due to an increase in caloric consumption that often accompanies pregnancy and may

carry over into the postpartum period, 3) Mothers may actual be heavier in the postpartum period

than they were before pregnancy, and 4) Mothers may put their preconception figures on a

pedestal, hence, having a stronger desire to return to those figures after childbirth. Thus, because

many women worry about returning to the normal weight and body shape during the postpartum

period, they may possess negative attitudes and feelings towards their bodies (Strang & Sullivan,

1985). Strang & Sullivan (1985) found that the women in the sample experienced more negative

feelings towards their body image during pregnancy than they did in the preconception period.

However, in the postpartum period, the women experienced more positive feelings towards their

body image than they did in the last trimester of their pregnancy.

Since body image is comprised of many components (e.g., physical appearance such as

weight and skin, posture, sense of fashion, etc.), it can affect a woman's personality, self-image,

identity, and behaviors and determine they way she responds to input from others and society

(e.g., media, positive feedback, respectively). Changes in body images can be a reflection of

what society and others define as beautiful and/or acceptable. Women are highly influenced by

the feedback received from others and what society tells us is the paradigm of being fit and

attractive; many American women connect beauty and attractiveness with success (Moore,

1978). It is important to note that pregnant women tend to react stronger to feedback received

about their body image more so than non-pregnant women, as many feel the need for the

endorsement from society regarding their weight gain and body shape changes in order to feel









that she is successfully experiencing the pregnancy and is capable of becoming a mother (Moore,

1978). The standards in our society today for weight and body size and shape do not allow

women to feel proud of their pregnant and postpartum bodies (Jenkin & Tiggemann, 1997).

Therefore, it is important that weight and body image issues be addressed in the pregnancy and

the postpartum periods. Women tend to feel that the physical effects of pregnancy on the body

(e.g., weight gain, breast changes) can bring about changes in self-esteem and body image; for

some women, these changes are negative (e.g., negative body image, lowered self-esteem), and

for others, these changes are positive (e.g., interpreting the changes as successful nurturing of the

fetus, resulting in increases in self-esteem) (Kline, Martin, & Deyo, 1998). However, in the

postpartum period, many women desire to "get their body back" (p.845) as they feel that

pregnancy and childbirth is a time when control cannot be taken over the physical changes in the

body. Thus, it is the PNC providers who can assist women engaging in proper nutrition and

exercise habits, which may potentially result in a more positive self-esteem and body image.

Since 1) weight is a prominent concern with respect to a women's well-being, and 2) weight

change is a significant feature of pregnancy, it is suggested that changes in weight and body

shape may play an important role in the development of PPD symptoms (Cameron et al., 1996).

The attitudinal and behavioral changes with regards to weight that often take place during the

course of pregnancy tend to be positive, but nonetheless, concerns about eating habits and weight

continue to exist and may even extend into the postpartum period. In fact, a woman entering the

postpartum period is vulnerable to experience concerns about her weight, because many women

tend to retain more weight, and thus, weigh more than they did in the preconception period.

Thus, much of the concerns may be more relentless that they were during the preconception

period, and, many women do not necessarily attribute weight gain in a positive manner as they









did during the pregnancy. However, women's responses toward weight gain during pregnancy

may differ depending on acceptance of the role of motherhood that will be undertaken following

childbirth, and perceptions of physical changes of the body as indications of fetal growth (Lacey

& Smith, 1987). These weight concerns may contribute to anxiety or depressive symptoms in the

postpartum period (Carter, Baker, & Brownell, 2000), which may result in more weight loss

attempts. Wood Baker et al. (1999) found that few women reported attempts to lose weight

during pregnancy in comparing the preconception, pregnancy, and postpartum periods; the most

weight loss attempts were made in the postpartum periods. For these women that engage in

healthful behaviors during the postpartum period, it is important to be cognizant of the factors

that motivate them to engage in exercise and weight management following childbirth (Keller,

Allan, & Tinkle, 2005). However, Harris, Ellison, & Clement (1999) found that time for exercise

and fitness was compromised among many of the women due to the demands of motherhood.

Thus, it is suggested that women adopt a healthy diet/fitness regimen during pregnancy (with the

guidance of their PNC provider) that can be easily transitioned into the postpartum period. It is

vital that health promotion activities that begin during pregnancy and continue into the

postpartum period target sources of chronic stress (e.g., BMI) and work towards supporting

women's self-esteem, as they may cultivate positive mental health during both periods (Hall,

Kotch, Browne, & Rayens, 1996). In addition, health promotion activities and interventions

should seek to determine eating attitudes and behaviors of all women as both can be influential

on postpartum distress. This is given that pregnancy is a time when 1) weight and body shape

changes are expected, and 2) many women feel that pregnancy is a "license" not to have weight

concerns (Stein & Fairburn, 1996; Fairburn & Welch, 1990, p. 158), this can affect the eating

attitudes and/or eating behaviors of women during pregnancy.









Expected PNC Content Versus Actual PNC Content

It is important to note that the content of PNC is yet to be standardized and the extent to

which prenatal health behavior advice is given to women during their PNC is not consistent

among all PNC rendered (Kogan, Kotelchuck, Alexander, & Johnson, 1994). Even though the

IOM, ACOG, and the U.S. Public Health Service Expert Content on Prenatal Care have issued

guidelines and recommendations on PNC content with respect to nutrition education and

guidance, there remains differences in the PNC that is actually delivered. A number of studies

have compared and contrasted the content PNC that is actually delivered versus the PNC content

that is recommended (without considering a birth outcome in the study). To illustrate the

differences that exist among the PNC rendered across the U.S., Appendix A (pp. 137-155)

summarizes the literature that has been conducted on the content of PNC using nationally

representative samples of women. According to Appendix A, there is a considerable amount of

variability that exists with respect to the care that is actually delivered. However, the literature in

Appendix A suggests that a significant portion of PNC providers in the U.S. are discussing

nutritionally-related content in the delivery of PNC (e.g., weight gain, exercise, proper nutrition,

etc.). In addition to the literature that exemplifies that PNC providers are carrying out nutrition

and wellness within the delivery of PNC, there is a collection of literature addressing the

effectiveness of PNC for pregnancy outcomes as well as the potential impact on postpartum

outcomes.

Effectiveness of PNC

The efficacy of the content of PNC has been addressed in the literature, though not

sufficiently (Alexander & Kotelchuck, 2001). In addressing the effectiveness of PNC, PNC is

often publicized as a health care service that is necessary for improving pregnancy outcomes

among women in the United States (Alexander & Kotelchuck, 2001). The benefits of receiving









early and continuing PNC in the U.S. has been advertised as critical to promoting healthy

pregnancy outcomes (Alexander & Kotelchuck, 2001). Some studies have shown that early

initiation of PNC (in the first trimester) results in improved pregnancy outcomes as opposed to

later or no PNC at all (Daniels et al., 2006). Accurately measuring PNC utilization is vital for

determining the need for health services, monitoring trends in health care utilization, and

determining associations between PNC and pregnancy outcomes (Kogan et al., 1998). PNC also

seems to result in healthier pregnancies even for women who have no disease because PNC

appears to associate well with the prevention of adverse pregnancy outcomes (Rosen, 1989). For

example, if conditions such as obesity are not addressed in the pre-conception period, as

evidence indicates that preconception care can improve pregnancy outcomes (Atrash et al., 2008;

Frieder, Dunlop, Culpepper, & Bernstein, 2008), it is recommended that PNC should address risk

factors such as obesity as its association with pregnancy complications has been shown (e.g.,

preeclampsia, respiratory problems, cesarean section, fetal death, etc.) (LaCoursiere et al., 2005;

Saravanakumar et al., 2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005;

Mahmood, 2009; Baeten et al., 2001; Cnattingius et al., 1998). It may also be beneficial for

postpartum health to be addressed during PNC delivery.

Prenatal Care (PNC) and Postpartum Outcomes

Alexander & Kotelchuck (2001) suggest that experiences in PNC that occur through

education and support services may positively impact the postpartum health of the mother and

the infant, including health status, health behaviors, and health care utilization. However, it is

important to note that some women may not feel the need to see their PNC provider for a

postpartum check-up. This may be due to women believing that unless they experience adverse

symptoms in the postpartum period that warrant a postpartum check-up, there is no need for a

postpartum evaluation of their health. This is unlike the prenatal period when women are









concerned about the health of their fetus and doing what is necessary to reduce the likelihood of

pregnancy-related complications.

More research is needed on the relationship between PNC and postpartum behaviors,

particularly postpartum depression (PPD) (Alexander & Kotelchuck, 2001). Though there is an

absence of literature addressing the impact of PNC on PPD in the U.S., the association between

PNC and PPD has been previously demonstrated outside the U.S., with more PNC visits

inversely associated with the onset of PPD among high-risk women (El-Kak, Chaaya, Campbell,

& Kaddour, 2004). Though the sample of women in this study were Lebanese, this study

demonstrates the importance of PNC in preventing adverse postpartum outcomes (e.g., PPD),

and warrants attention for the impact of PNC on PPD in the U.S. Since this study suggests an

impact of PNC on a postpartum outcome (PPD), and thus, the experiences that a woman

undergoes through her PNC may affect the outcomes she experiences in the postpartum period

(Alexander & Kotelchuck, 2001). If a woman seeks PNC and consequently complies with the

advice/recommendations given to her by her PNC provider), this may help prevent adverse

postpartum outcomes.

In addition, postpartum care should be aligned with the PNC in addressing similar issues

that can impact weight and health during the postpartum period. This would help women obtain

access to professional assistance in addressing obesity issues postpartum.

Theoretical Framework

For this study, since PNC as an intervention through which PNC providers can educate

help address any concerns with a woman about her weight, and provide guidance to patients, this

delivery of health care will be seen as "informational support." Homan & Korenbrot (1998)

looked at support delivered at each of five PNC ambulatory practice settings (community clinic,

health department, public hospital clinic, private hospital clinic, private physician office).









Support service delivery was defined as "psychosocial, health education, and nutrition" (using

the USPHS recommendations) and the authors concluded that a woman who has her needs

addressed during PNC (psychological support, nutrition, or health information) has a higher

likelihood of experiencing a positive birth outcome (poor obstetric outcomes was defined as

experiencing preterm birth, having a low birth weight infant, fetal death, ectopic pregnancy, or

spontaneous abortion) than a woman who does not have any of her needs addressed during PNC.

Obesity: Research has shown that an association exists between pre-pregnancy weight and

pregnancy weight gain, as well as postpartum weight retention. Pre-pregnancy BMI was chosen

as the measure for BMI because it is known to be the strongest predictor of future obesity,

including excess perinatal weight gain and future weight gain (Krummel, 2007; Gore, Brown, &

Smith West, 2003), and is also suggested to be connected to postpartum BMI distress (Walker,

1998).

Prenatal Care (PNC): The intervention of interest in this study will be PNC (health

education and guidance on healthy behaviors during pregnancy). In general, PNC includes 1)

risk assessment (e.g., medical and psychosocial history, physical examination, laboratory tests),

2) health promotion activities (e.g., counseling to promote healthy behaviors and providing

general knowledge about pregnancy and parenting such as physiological and emotional changes,

symptoms of preterm labor, fetal growth and development) and 3) a proposed pregnancy plan

that is tailored to each woman's needs (USPHSEPCPNC, 1989). The following nutritional

intervention is recommended for the PNC provider during the delivery of care (Institute of

Medicine, 1990):

1) Encourage the women to achieve a healthy, balanced diet to support adequate weight gain
(e.g., IOM recommended ranges)

2) Evaluate a woman's dietary habits (e.g., food history, food frequency questionnaire)









3) Calculate the woman's pre-pregnancy BMI

4) Estimate the woman' s gestational age

5) Conduct a routine dietary assessment to determine dietary needs

6) Be aware and knowledgeable about the attitudes and feelings that many pregnant women
hold regarding body image and perceptions, especially closer to childbirth and in the
postpartum period, as they tend to be negative during these times (Moore, 1978; Stein &
Fairburn, 1996). It is the PNC providers that can assist women in viewing pregnancy and
its associated physical changes as ones that are normal and beautiful (Moore, 1978).

Postpartum depression symptoms: If a woman engages in healthy nutrition behaviors

during pregnancy, her likelihood for PPD symptoms will be reduced.














T. PPD


SObesity
(pre-
pregnancy
BMI)













Figure 2-1. Theoretical framework


Prenatal care (informational
support): Education and
guidance on weight gain,
health behaviors









CHAPTER 3
METHODS

Data Overview: Pregnancy Risk Assessment Monitoring System (PRAMS)

This study used data from the 2004 and 2005 Pregnancy Risk Assessment Monitoring

System (PRAMS) (CDC, 2007). This is a continuing population-based survey maintained by the

Centers for Disease Control (CDC) that collects data on maternal behaviors, experiences, and

characteristics in the pre-pregnancy, pregnancy, and postpartum period among randomly selected

woman who delivered a live infant. Started in 1987 because the incidence of low birth weight

infants had changed very little during the previous 20 years, and because the infant mortality rate

was not decreasing as fast as it had in previous years, this database provides state-specific data.

Currently, 30 states participate in PRAMS. States can use these data to measure the performance

of health programs and to obtain information on maternal experiences within their respective

populations, which can contribute to improving maternal and child health. These data can also be

used to 1) identify the women and infants who are at the highest-risk for maternal and child

health problems, 2) monitor their health status, and 3) assess their progress in efforts to improve

the health of these mothers and their infants.

Data Collection Procedures

Each state that participates samples 1,300 to 3,400 women annually. Women are initially

contacted through mail, and those who do not respond are contacted and interviewed via

telephone. The questionnaire includes two components: the core questionnaire and the standard

questionnaire. The core questionnaire contains 56 questions which all states include in their

survey, and the standard questionnaire contains 185-questions from which states can choose

which to include in their survey (questions are options for survey inclusion). The standard









questionnaire allows the collection of data to best meet the needs of each state. Data collection

methods and instruments are standardized across states and occur as follows (CDC, 2007):

1) A preletter is sent to the mother to inform her that she will be receiving a PRAMS
questionnaire

2) The initial mail questionnaire packet is sent to all the mothers randomly chosen for the
sample 3-7 days after the preletter is sent

3) The tickler is the name given to the thank you note and a reminder to complete the
questionnaire. This document is sent 7-10 days after the initial packet

4) The second mail questionnaire packet is sent to all the mothers who do not respond 7-14
days after the tickler is sent out

5) The third mail questionnaire packet is sent to all the mothers who do not respond 7-14 days
after the second questionnaire is sent out

6) A telephone follow-up occurs for all the mothers who do not respond 7-14 days after the
third questionnaire is sent out (with up to 15 phone attempts in efforts to reach the mother).
During this follow-up, interviewers may coordinate times with the mother to administer the
questionnaire over the phone. Thus, PRAMS involves mixed methodology: the self-
administered questionnaire that is mailed out, or a phone-administered questionnaire in the
event the mother does not complete and return the questionnaire prior to the phone follow-
up.

The questionnaire packet includes the following: a cover letter that 1) explains PRAMS

and why the mother was chosen to participate, 2) provides directions for completion of the

survey, 3) describes potential incentives/rewards, and 4) provides a contact number to address

further questions. In 2004, the cover letter was divided into two components: an introductory

portion, and a document of informed consent. These mailings are sent to the mothers

approximately 2-4 months after they have given birth, and the data collection process lasts for

approximately 60-95 days. The states included in the final data set that is released include those

who achieved a 70% response rate or higher.

Data collection is attempted every month, by randomly selecting a sample of

approximately 100-250 women from the current birth certificates, and then sending out the









mailings to the mothers chosen. Since the birth certificates are linked to the mothers' responses,

variables on the birth certificate can also be accessed for analysis. Some groups of women, who

comprise high-risk populations, are sampled at a higher rate so that a sufficient quantity of data

are available (CDC, 2007). The oversampling of subpopulations of interest is accomplished by

the stratification of the PRAMS sample (Shulman, Gilbert, & Lansky, 2006). For the 2004 and

2005 years of PRAMS data used in this study, the stratification variables included: birthweight,

Medicaid, maternal age, geographic area, maternal race/ethnicity, county density, and smoking

status (CDC, 2009).

Weighting of Data

Data are weighted to account for characteristics of the mothers that may influence the

response rates (e.g., single marital status). These nonresponsee" weights assume that those

mothers who did not respond would have given similar answers to questions as mothers who did

respond to the questionnaire. For each cell, there are at least 25 respondents. For this study, the

following weights were set prior to the analysis: a finite population correction factor (fpc), a

strata weight, and a final analysis weight comprised of a non-response weight, a non-coverage

weight, and a sampling weight. Thus, the final analysis weight in-part, corrects for groups (e.g.,

obese pre-pregnancy BMI) that have higher than normal response rates. Weights are applied so

that each state's data are representative of the women who gave birth in that state.

Rationale for Using PRAMS Data

The Pregnancy Risk Assessment Monitoring System (PRAMS) is a population-based data

set that incorporates random sampling techniques from 30 states, 16 of which are included in the

main analysis. For this study, a large sample size was used that was representative of the

different regions around the U.S (e.g. Southeast, Northeast, Midwest, etc.) and hence, the

different populations that reside in these regions. In addition, PRAMS includes a comprehensive









list of variables regarding maternal health and behaviors during the pre-conception, antenatal,

and postpartum periods. There exist no other national databases that include the significant

amount of variables that PRAMS includes.

Measures/Procedures

Postpartum Depression (Dependent Variable)

PRAMS data include 12 measures of PPD. However, participating states have the option

of choosing whether to use measures of PPD, if any (PPD questions are part of the standard

questionnaire). For the main analysis, two measures of PPD are used:

1. Since your new baby was born, how often have you felt down, depressed, or hopeless?

Always
Often
Sometimes
Rarely
Never

2. Since your new baby was born, how often have you had little interest or little pleasure in
doing things?

Always
Often
Sometimes
Rarely
Never

The states that include these PPD questions include: Alaska, Colorado, Georgia, Hawaii,

Illinois, Maine, Minnesota, North Carolina, Nebraska, New Mexico, Oregon, Rhode Island,

South Carolina, Utah, Vermont, and Washington. Other questions that pertain to PPD in PRAMS

include whether a woman received a professional diagnosis and/or whether she sought treatment

for her PPD. However, these questions were excluded from the primary analysis because the

extent to which a woman is either 1) diagnosed with PPD and/or 2) received treatment for her

PPD symptoms may depend on the proactive nature of the provider from whom she 1) sought









PNC from and or 2) sought her postpartum check-up visit from (as they may be two different

providers and each may or may not be cognizant of or look out for PPD symptoms).

The main analysis consisted of a logistic regression model, which included the 16 states

from which data were requested. Control variables for the main analysis came from the PRAMS

Core Questionnaire only, since all the states are mandated to include the core questions in their

state surveys (see page 59 for a list of the control variables).

In coding PPD for the logistic regression model as either "1" or "0," a scoring system

similar to the Patient Health Questionnaire was used. The Patient Health Questionnaire (PHQ) is

a three-page, patient self-administered, criteria-based instrument that was created to diagnose

mental disorders (e.g., major depressive disorder, panic disorder, other anxiety disorders, etc.)

(Kroenke, Spitzer, & Williams, 2001). Not only have the sensitivity and specificity been

established (e.g., 68-95%, 84-95% respectively) (Kroenke et al., 2001), as well as criterion

validity, construct validity, test-retest reliability, and internal reliability, but the condensed 9-

item module that is used to diagnose major depression and determine its severity is based on the

criteria that are used in the DSM-IV to diagnose depressive disorders. Stemming from the PHQ-

9, the PHQ-2 was extracted to offer clinicians a more concise measure of depression diagnosis

and severity to accommodate the busy clinical settings that exist in today's health care system

(Kroenke, Spitzer, & Williams, 2003). The following two items comprise the PHQ-2 (Kroenke et

al., 2003, p.1285):

"Over the last 2 weeks, how often have you been bothered by any of the following

problems?"

a) "Little interest or pleasure in doing things"
b) "Feeling down, depressed, or hopeless"









The response and scoring system are as follows: "Not at all" (score = 0), "several days"

(score = 1), "more than half the days" (score = 2), "nearly everyday" (score = 3) (p. 1285). Since

a score is given for each of the two question, and the two scores are added to obtain one total

score, the highest possible score than can be given is a 6, and the lowest possible score that can

be given is a 0. A score of 3 or greater indicates major depressive disorder. The sensitivity and

specificity (83% and 92%, respectively) for scores greater than or equal to three, as well as the

criterion and construct validities of the entire PHQ-2 have been demonstrated (Kroenke et al.,

2003). Since the questions on the PHQ-2 mirror the PRAMS PPD measures that were used for

this study, a similar scoring system was used to determine PPD symptoms. However, since the

PPD measure consists of a 5-item response scale, and the PHQ-2 consists of a 4-item response

scale, the "sometimes" and "rarely" responses were scored as 1. Thus, the following point

system was assigned to the PPD responses:

3 = Always
2 = Often
1 = Sometimes
1 = Rarely
0 = Never

For the purpose of this study, scores from 0-2 were coded as "0" (no PPD symptoms), and

scores from 3-6 will be coded as "1" (PPD symptoms), similar to the coding of the PHQ-2. In

addition to the logistic regression model, an ordinal logistic regression was also estimated in

order to test the association of PNC on PPD when PPD is analyzed as a non-dichotomous

outcome variable. For the ordinal logistic regression model, since there are two questions to

represent the dependent variable (PPD), the PHQ-2 scoring system was also used in the analyses.

*Note: the scores for PPD in the logistic regression represent the total score after adding the

scores for both PPD questions.









Obesity (Main Independent Variable)

BMI is calculated using the following formula:

Weight in pounds
BMI = ---------------------- X 703 (conversion factor)
Square of height in inches (3-1)

However, since maternal pre-pregnancy BMI is a variable included in the PRAMS data,

there was no need to separately calculate and categorize a woman's pre-pregnancy BMI. Table

3-1 shows the BMI categories, one of which a woman's pre-pregnancy BMI is classified into.

Prenatal Care Utilization (Moderating Variable)

The Adequacy of Prenatal Care Utilization (APNCU) is an index measure of PNC that

considers both the adequacy of PNC initiation, as well as the adequacy of the number of PNC

services received (Kotelchuck, 1994). The benefits for using this index are as follows: 1) the

APNCU offers a suitable index for assessing the degree of prenatal care utilization once care is

initiated, 2) this index also separates the initiation of care from compliance with the number of

visits as recommended by the American College of Gynecologists once care is initiated, 3) this

index can include women who did not receive any PNC, 4) this index is valuable for research

that is aimed towards enhancing PNC, 5) this index has a separate category for high-risk

pregnancies (adequate plus care), and 6) this index, among others, provides the most serious

depiction of prenatal care utilization (Alexander & Kotelchuck, 1996). The APNCU is calculated

based on the time of PNC initiation and the number of PNC visits, adjusting for gestational age.

Gestational age refers to the period from the first day of the woman's last menstrual cycle to the

date of the baby's birth (National Institutes of Health, 2007). Adjusting for gestational age is

important, as it may be associated with the duration of PNC as well as problems experienced

during pregnancy/postpartum. The index is defined as the ratio of observed (actual) prenatal care

visits (determined from the birth certificate)/expected number of PNC visits (based on









recommendations from the American College of Obstetricians and Gynecologists). The

percentage obtained is then categorized into one four PNC categories (see Table 3-5)

(Kotelchuck, 1994).

The APNCU Index was calculated in this study by the following methods:

1) Gestational age (included in PRAMS as number of weeks) was recorded based on the
number of expected visits that should occur based on a woman's gestational age. The rule
of thumb is that there should be one PNC visit each month until 28 weeks (or seven
months), which then changes to a PNC visit once every 2 weeks until 36 weeks (or nine
months) with a PNC visit weekly until a woman gives birth (Kotelchuck, 1994). The
month that a woman gave birth was not counted as a PNC visit because it was unclear
whether the woman received her PNC visit for that month or not before she gave birth.
Table 3-3 lists the number of expected visits dependent on the gestation age.

2) Next, month of PNC initiation was recorded into number of missed visits. Table 3-4 lists
the number of missed visits depending on when a woman initiated her first PNC visit.
Missed visits were calculated based on the maximum number of PNC visits a woman
could miss given the month she initiated her PNC.

3) Since steps 1 and 2 comprise the denominator, a variable was next calculated subtracting
the number of missed visits from the number of expected visits. This variable was labeled
"indexdenominator."

4) The index was then calculated by dividing the indexdenominator variable into the "actual
number of PNC visits" variable (measured in PRAMS).

5) Table 3-5 shows how each index was categorized into one of the APNCU categories. Table
3-6 presents the characteristics of the APNCU Index categories.

6) Since some women many have received the quantity of expected visits even though they
initiated PNC late, all women who initiated care after the fourth month who were not
initially coded into inadequate care were re-coded into inadequate care. The month of PNC
initiation takes precedence over the quantity of visits when determining the APNCU
(Kotelchuck, 1994). Thus, women who initiated PNC after the fourth month were filtered
out of the sample, and recorded to inadequate care. Finally, the filter was removed to allow
those women to remain as part of the sample; however, now with the correct APNCU.
Table 3-7 contains the frequencies of women before recoding occurred, while Table 3-8
contains the frequencies of women who were incorrectly coded, and Table 3-9 contains the
frequencies after recoding was accomplished.

After creating the APNCU, dummy variables were first created for each of the main effects

(each pre-pregnancy BMI and each PNC utilization category). Then, dummy variables, labeled









as interaction terms, were created for each combination of pre-pregnancy BMI and PNC (9

groups) to detect any differences among any of the combinations. Normal pre-pregnancy BMI

and adequate PNC were the reference groups.

1) Obese/Inadequate care
2) Overweight/Inadequate care
3) Underweight/Inadequate care
4) Obese/Intermediate care
5) Overweight/Intermediate care
6) Underweight/Intermediate care
7) Obese/Adequate-plus care
8) Overweight/Adequate-plus care
9) Underweight/Adequate-plus care

Control Variables

Control variables included maternal age, maternal education, maternal income (12

months before), maternal race, maternal ethnicity, birthweight, gender of infant, vaginal delivery,

alcohol and smoking behaviors during pregnancy, participation in Women, Infants, and Children,

breastfeeding practice, if the infant ended up in the ICU, pregnancy intention, how PNC was paid

for (representing insurance), and maternal morbidities. All control variables were categorical

except for maternal age.

Analysis

All analyses were conducted using Stata v.10. Both univariate and bivariate analyses were

conducted, with chi-square tests used to test significance between groups. Significant association

of PPD and maternal age was tested using a t-test, while ANOVA was used to test the

association of pre-pregnancy BMI and maternal age, and PNC utilization and maternal age. The

primary multivariate analysis held PPD as a dichotomous variable via a logistic regression (logit)

model, while a secondary multivariate analyses held PPD as a dichotomous variable via a

logistic regression (logit) model, and manually risk-adjusted for high-risk pregnancies after









removing observations that met any of the criteria for a high-risk pregnancy. Thus, the sample

analyzed in the secondary logistic regression model included healthy pregnancies only.

Primary Risk-Adjusted Logistic Regression

Specific aim 1:

What is the association of pre-pregnancy BMI with subsequent development of postpartum

depression (PPD) symptoms?

Hypothesis: Women, who were obese before pregnancy will have the highest likelihood of

PPD, followed by overweight women, then underweight women. Women who had a pre-

pregnancy BMI of normal were the reference group. A dummy variable was created for each

BMI category and the reference group for this analysis was normal pre-pregnancy BMI. Women

who had a pre-pregnancy BMI of obese are predicted to have the highest likelihood for PPD. The

p coefficients of interest in this model, and in accordance with the hypothesis, were obese pre-

pregnancy BMI and underweight pre-pregnancy BMI which were expected to have the most

positive (meaning the likelihood will be the highest for women in this group) and the least

positive (the PPD likelihood will be the lowest for women in this group) association with PPD,

respectively, relative to normal pre-pregnancy BMI (the reference group). The odds ratios were

provided by STATA (exponentiating the 0 coefficients). Thus, it was hypothesized that the

women who had a pre-pregnancy BMI of obese, relative to women who had a pre-pregnancy

BMI of normal, would have the highest likelihood for PPD symptoms and that the pre-pregnancy

BMI of overweight and underweight groups would have the second to highest and lowest

likelihood for PPD symptoms, respectively. Since this is a logistic regression, PPD was

analyzed as a dichotomous variable, using the scoring system similar to the Patient Health

Questionnaire-2 [for a complete description of the scoring system, see pp.60-61: Postpartum

depression (dependent variable)]. The model specification was as follows:










P(Y = 1)
logit = log P(Y 0) = a+ 3(underweight) + 32(overweight) + 33(obese)+ 3kXk

(control variables) + s (3-2)

Specific aim 2:

Does PNC moderate the relationship between pre-pregnancy BMI and PPD?

Hypothesis: This aim tested whether the association of pre-pregnancy BMI with PPD

symptoms varied with the level of PNC. The association between pre-pregnancy BMI and PPD

symptoms was expected to remain after estimating PNC as a moderating variable. This means

that women who had a pre-pregnancy BMI of obese would continue to have the highest

likelihood for PPD symptoms, and women who had a pre-pregnancy BMI of underweight would

have the lowest likelihood, relative to women who had a pre-pregnancy BMI of normal.

However, for each pre-pregnancy BMI group, it was predicted that relative to women who

received adequate PNC, the highest likelihood for PPD symptoms would decrease as follows:

inadequate PNC, adequate plus PNC, intermediate PNC. Each pre-pregnancy BMI category had

its own reference group: women in the BMI group who received adequate care. For example,

obese inadequate, obese intermediate, and obese adequate-plus were compared to obese adequate

women. The model specification for the main logistic regression was as follows:

PlP(Y = 1)
logit = log+ = 1 (obese/adequate plus care) + 32(overweight/adequate plus
LP(Y = 0)]
care) + 33(underweight adequate plus care) + 34(obese/intermediate care) +
P3(overweight/intermediate care) + 36(underweight/intermediate care)
+P7(obese/inadequate care) +Ps(overweight/inadequate care) +
39(underweight/inadequate care) +3kXk(control variables) + s (3-3)

Secondary Risk-Adjusted Logistic Regression

Adequate plus, sometimes referred to as "intensive" care, is defined as PNC that is

initiated by the fourth month (inclusive), and/or a woman received 110% or more of the expected









number of PNC visits (Kotelchuck, 1994). A number of women in the adequate plus PNC

category are considered to be a high-risk pregnancy group of women who receive more PNC

than the standards as established by the American College of Obstetricians and Gynecologists

(Kotelchuck, 1994). Compared to the other categories of PNC in the APNCU: inadequate,

intermediate, and adequate PNC, a woman may receive adequate plus PNC if she possesses

health risks such as medical conditions that can put her at a greater risk for complications and/or

adverse pregnancy outcomes; hence, terming the pregnancy as "high-risk" (Chism, 1997;

Kotelchuck, 1994). Thus, the increased number of PNC visits in this category is meant to allow

for additional monitoring of a woman's pregnancy by her PNC providerss. However, some

adequate plus PNC is delivered to women who seek extra PNC due to other unobserved

characteristics unrelated to risk (e.g., motivation to seek more PNC than necessary because of a

woman's desires to do so). Since the APNCU index does not risk-adjust, but rather, remains a

conventional index that looks at the quantity and initiation of PNC (Kotelchuck, 1994), this study

sought to risk-adjust for women who received adequate plus PNC in two ways: 1) risk-adjusting

by controlling for high-risk characteristics, and 2) risk-adjusting by removing observations with

high-risk characteristics (e.g., women with preterm labor) from the sample. Risk-adjustment for

the second model was accomplished by removing observations that met one or more of the

following criteria during pregnancy:

1) Birthweight less than 2,500 grams
2) Women less than 18 years of age or greater than 40 years of age
3) Diabetes before pregnancy
4) Incompetent cervix
5) Preterm labor
6) Placenta previa or placenta abruptio
7) Bedrest
8) Car crash injury
9) Blood transfusion









10) Medical risk factors
11) Hospitalized during pregnancy

Thus, after removing these observations from the sample, it was assumed that the

observations in the sample constituted pregnancies with standard, common characteristics (e.g.,

nausea, infant birthweight of 2,500 g or greater, etc.). A sub-analysis was estimated for the first

risk-adjustment model, holding adequate plus PNC as the dependent variable and subsequently

controlling for the significant variables in the main model. A further description of the sub-

analysis for the first model is provided in Appendix B (pp. 156-187). A sensitivity analysis was

estimated for the second risk-adjustment model. The specific aims, hypotheses, and

specification for this model remained the same as the primary risk-adjusted logistic regression

model.

Wald Test

Finally, to test the equality of the interaction term coefficients within each BMI category,

which would test the moderating effect of PNC on the BMI and PPD association, Wald tests

were performed for each model that addressed the second specific aim. Two types of Wald tests

were run: 1) to determine if the coefficients were equal to one another, and 2) to determine if the

coefficients were equal to 0.

Model Fit

To test the model fit of the primary and secondary logistic regression models, the Del

measure was calculated. This goodness of fit measure takes the scope and precision of the model

into account.

Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Merging Strata

The PSU in this study was the woman herself. Since STATA would not analyze strata with

only one PSU, frequencies were run on the strata weight variable and four strata were identified









with less than ten PSUs (1 strata with 1 PSU, 2 stratas with 2 PSUs, and 1 strata with 9 PSUs).

These strata were then merged into the strata with the largest number of PSUs, after which the

analysis was accomplished.

Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Dropped Cases

Since the loss of an infant can be traumatic for a mother, women who reported having an

infant that did not live were removed from the analysis (N= 1,032).

Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Imputed Data

Since the income variable included in the analysis had 3,346 missing values (more missing

values than most of the other variables included in the analysis), the data were imputed via the

conditional means approach using the following steps:

1) An OLS regression was estimated on demographic variables included in the multivariate
analyses.

2) Predicated values were calculated after running the OLS regression.

3) The missing values were replaced with the predicted values.

However, prior to imputation, because some states include different types of questions

with respect to the same income category, data were collapsed to reflect four income categories.

Table 3-10 shows the raw income variable's categories and Table 3-11 shows the collapsed

income variable before it was imputed.

In addition, the variables representing how PNC was paid for, and the maternal morbidities

were imputed using a conditional means approach. This was accomplished due to the large

number of missing values for some of the variables in these categories.









Table 3-1. Specific aims, dependent, and independent variables
Specific aim Dependent variable Independent variable(s)
1 Postpartum Pre-pregnancy BMI


depression
Postpartum
depression


Pre-pregnancy body mass index (BMI), prenatal care
utilization (PNC), pre-pregnancy BMI/PNC interaction
terms (moderator)


Table 3-2. Classification of body mass index (BMI)
BMI Weight status
> 18.5 Underweight
18.5-24.9 Normal
25.0-29.9 Overweight
>/= 30.0 Obese
(National Institutes of Health, 1998)

Table 3-3. Step 1: Gestational age calculation into expected number of visits for the APNCU


Index
Gestational age (in weeks)
18-20
21-24
25-28
29-30
31-32
33-34
35-36
37
38
39
40
41
42
43


Expected number of visits


Table 3-4. Step 2: Month of PNC initiation calculation into number of missed visits for the
APNCU Index
Month of PNC initiation Number of missed visits
1st month 0
2nd month 1
3rd month 2
4th month 3
5th month 4
6th month 5
7th month 6
8th month 8
9th month 10









Table 3-5. Step 3: Categorization of index into categories for the APNCU Index
APNCU group Included indices
Inadequate Lowest thru 49.99
Intermediate 50.00 thru 79.99
Adequate 80.00 thru 109.99
Adequate plus 110.00 thru highest

Table 3-6. Characteristics of the APNCU Index groups
APNCU group Characteristics
Inadequate care Initiated after the 4th month; under 50% of
expected visits were received; can be divided to
include those who did not receive PNC
Intermediate care Initiated by the fourth month; between 50-79% of
expected visits were received.
Adequate care Initiated by the fourth month; 80-109% of
expected visits were received
Adequate plus care Initiated by the fourth month; 110% or more of
expected visits were received
(Kotelchuck, 1994)

Table 3-7. Adequacy of Prenatal Care Utilization (APNCU) Index frequencies before recoding
APNCU index Frequency Percentage
Inadequate 1,743 3.6%
Intermediate 7,631 15.6%
Adequate 20,678 42.2%
Adequate plus 18,952 38.7%
Total 49,004 100%

Table 3-8. Adequacy of Prenatal Care Utilization (APNCU) Index frequencies of incorrect
codings (observations incorrectly coded into other PNC utilization categories that
were recorded into "inadequate PNC utilization" based on the month of initiation)
APNCU index Frequency Percentage
Intermediate 1,680 42.4%
Adequate 1,237 31.2%
Adequate plus 1,047 26.4%
Total # of observations to be recorded into "inadequate PNC" 3,964 100%

Table 3-9. Step 4: Adequacy of Prenatal Care Utilization (APNCU) Index frequencies after
recoding all cases from Table 3-8 into "inadequate PNC utilization"
APNCU Index Frequency Percentage
Inadequate 5,707 11.6%
Intermediate 6,584 13.4%
Adequate 19,441 39.7%
Adequate plus 17,272 35.2%
Total 49,004 100%









Table 3-10. Coding for raw income variable categories before collapsing categories
Code Category
1 Less than 10,000
2 $10,000 to $14,999
3 $15,000 to $19,999
4 $20,000 to $24,999
5 $25,000 to $34,999
6 $35,000 to $49,999
7 $50,000 or more
8 Less than $8,000
9 $8,000 to $9,999
10 $50,000 to $74,999
11 $75,000 or more


Table 3-11. Coding for collapsed income variable categories


Code


Category
Less than 10,000
$10,000 to $24,999
$25,000 to $49,999
$50,000 or greater


4
Changes made
**2
**3 and 4
** 5

**6
** 7
**8
**9
**10
**11


Expanded to include income up to $24,999
Collapsed into category 2
Category number changed to 3 and expanded to include
income up to $49,999
Collapsed into category 3
Category number changed to 4
Collapsed into category 1
Collapsed into category 1
Collapsed into category 4
Collapsed into category 4









CHAPTER 4
RESULTS

In this chapter, 1) the PRAMS sample is described through univariate analyses, and 2)

bivariate analyses indicating percentages and chi-square significance for a) the dependent

variable (PPD), b) each of the primary independent variables of interest, pre-pregnancy BMI and

PNC utilization, and c) adequate plus PNC. Then, the results from the multivariate analyses are

discussed via the primary risk-adjusted logistic regression model and the secondary risk-adjusted

logistic regression, which removed high-risk pregnancies from the sample.

Univariate Analyses

The description of the sample is presented in Table 4-1 for the categorical variables, and

Table 4-2 for the one continuous variable in this study, maternal age. Table 4-1 describes 1) the

distribution of the dependent variable (PPD symptoms), 2) the distribution of the main

independent variables (pre-pregnancy BMI and PNC utilization), and 3) the distribution of the 39

categorical control variables included in the analyses. Control variables were organized in the

table by "main variables," "demographic control variables," "insurance control variables,"

"pregnancy and delivery control variables," "high-risk maternal morbidity control variables,"

and "non high-risk maternal morbidity control variables." The latter two labels were added since

the secondary analysis sought to distinguish "high-risk" from "healthy" pregnancies. Table 4-2

presents the mean, standard deviation, maximum and minimum values for maternal age. Basic

demographic frequencies for the entire sample show that the largest percentage of women in the

sample were of White race (63.64%), about half of the women received a secondary education or

less (49.45%), and 63.29% of the women were married. Finally, the income distribution

consisted of about 20% each for women who received less than $10,000 and women who









received between $25,000 to $49,999 (22.80%), and about 30% each for women who received

between $10,000 to $24,999 (28.18%) and women who received $50,000 or greater (28.69%).

PPD symptom frequencies for the sample analyzed in the primary risk-adjusted logistic

regression (including all pregnancies), and the sample analyzed in the secondary risk-adjusted

logistic regression (including healthy pregnancies only) were provided in Table 4-1 to compare

the distribution of PPD symptoms in a sample including women comprising healthy and high-

risk pregnancies, versus a sample that removed high-risk pregnancies and included healthy

pregnancies only. For PPD symptoms, the frequencies showed that among 45,285 women

included in the primary risk-adjusted logistic regression model, the prevalence of postpartum

"blues" was 84.6%, while the prevalence of PPD symptoms in the sample was approximately

15.4%. Removing the high-risk pregnancies from the analysis, to include only healthy

pregnancies with "common" pregnancy characteristics (e.g., nausea) showed that among the

15,443 women included in this sample, the prevalence of postpartum "blues" was higher, at

88.5% and the prevalence of PPD symptoms was lower, at 11.5%.

Frequencies for the primary main effect independent variable, pre-pregnancy BMI, showed

a high number of women who had an obese pre-pregnancy BMI (n= 10,270; 22.15%). This group

of women comprised the second highest group in the sample, following the reference group,

women with a normal pre-pregnancy BMI, who comprised about half of the sample (n=23,834;

51.40%). The number and percent of women who had an obese pre-pregnancy BMI was nearly

twice as great as those of the group of women who were the lowest in number: women who had

an overweight pre-pregnancy BMI (n=5,888; 12.70%). Finally, the number of women who had

an underweight pre-pregnancy BMI was not considerably higher than the number of women who

had an overweight pre-pregnancy BMI (n=6,379; 13.76%). For the secondary main effect









(moderating) independent variable, PNC utilization, the highest percentage was also comprised

of the reference group, women who utilized an "adequate" quantity of PNC (n=19,258; 40.03%).

However, the second highest number was comprised of women who utilized an "adequate plus"

quantity of PNC (n=16,748; 34.82%). This group of women nearly tripled the number of women

who utilized an "inadequate" quantity of PNC (n=5,568; 11.58%). Finally, the number of women

who utilized an "intermediate" quantity of PNC was not considerably different from the number

of women who utilized an "inadequate" quantity of PNC" (n=6,529; 13.57%).

Due to the high number of women who utilized "adequate plus" PNC, it was speculated

that reasons for this high number could be attributed to either 1) high-risk status, or 2) a woman's

own desire to seek more PNC than necessary (as established in medical guidelines). In attempts

to distinguish pregnancies that were high-risk versus pregnancies that experienced normal,

standard pregnancy-related troubles (e.g., nausea), 17 maternal morbidity control variables were

included in this study. The morbidity variables representative of high-risk and removed, in-part,

from the secondary sample (to include healthy pregnancies only) included nine variables: 1)

Diabetes before pregnancy, 2) incompetent cervix, 3) preterm labor, 4) placenta previa or

abruptio, 5) bedrest, 6) car crash injury, 7) blood transfusion, 8) having general medical risk

factors, and 9) hospitalization during pregnancy. The distribution of the high-risk morbidity

variables was as follows: 1) four of the variables were comprised of less than 5% of the sample

(less than 5% of the women answered "yes" to having had the morbidity during pregnancy), with

the lowest percentage representing women who had a blood transfusion during pregnancy

(n=677; 1.31%), 2) one variable was comprised of a percentage between 5-10%, and 3) the

remaining four variables were comprised of percentages between 10-40%, with the highest

percentage representing women who had general medical risk factors during pregnancy









(n=18,209; 35.29%). Since the largest percentage of the high-risk morbidity variables

represented a general measure (women who answered "yes" to having had medical risk factors

during pregnancy), the highest percentage representing a single, defined morbidity was

comprised of about 25% of the sample: women who had preterm labor (n=13,595; 26.35%).

Among the non high-risk morbidity variables, the distribution occurred as follows: 1) one

variable was comprised of less than 10% of the sample, with the lowest percentage representing

women who had gestational diabetes (n=4,786; 9.28%) 2) four variables were comprised of

percentages between 10-20%, and 3) the remaining two variables were comprised of percentages

between 20-40%, with labor/delivery complications comprising the highest percentage

(n=18,131; 35.14%). Similar to the high-risk morbidity variables, since the largest percentage for

the non high-risk morbidity variables represented a general measure (women who answered

"yes" to having had labor/delivery complications), the highest percentage representing a single,

defined morbidity involved a very common morbidity in pregnant women: nausea, comprising

about 30% of the sample (n=15,104; 29.27%). To further examine characteristics of women who

utilized "adequate plus" PNC versus women who utilized other quantities of PNC, beyond

frequencies, chi-square analyses, presented in the next section, were carried out.

Bivariate Analyses

Chi-square analyses, presented in Tables 4-3, 4-5, and 4-6 respectively, were performed to

examine relationships between 1) the dependent variable (PPD) and each of the following: a)

pre-pregnancy BMI, b) PNC utilization, and c) each of the control variables, 2) the primary main

effect independent variable (pre-pregnancy BMI) and each of the control variables, and 3) the

secondary main effect independent variable (the moderating variable, PNC utilization) and each

of the control variables. Also, since there were many women who utilized adequate plus PNC,

the highest quantity of PNC, chi-square analyses, presented in Appendix B (pp. 156-187), were









performed with adequate plus PNC to further describe characteristics of women who utilized this

quantity of PNC. All chi-square analyses were carried out in order to test the specific aims, and

to determine the significance of the relationships between the variables of interest (PPD

symptoms, pre-pregnancy BMI, and PNC utilization) and the control variables. Table 4-4

presents the results from the t-test for the continuous variable, maternal age, with PPD

symptoms.

Since PPD symptoms was the dependent variable in this study, chi-square analyses were

estimated to examine the relationship of characteristics among women who were categorized as

experiencing PPD symptoms versus those who were categorized as not experiencing PPD

symptoms. Table 4-3 shows that there was a significant difference in percentages of PPD

symptoms across the four pre-pregnancy BMI categories (p<0.05), with normal pre-pregnancy

BMI displaying the highest percentage of PPD symptoms, followed by women who had an obese

pre-pregnancy BMI. Overweight pre-pregnancy BMI displayed the lowest percentage of PPD

symptoms, while the percentage of women who had an underweight pre-pregnancy BMI was not

considerably higher than the percentage of women who had an overweight pre-pregnancy BMI.

In fact, the percentage of women who had an obese pre-pregnancy BMI with PPD symptoms was

twice as much as the percentage of women who had an overweight pre-pregnancy BMI. As for

PNC utilization, there was a significant difference in percentages of PPD symptoms among the

four PNC utilization categories. Women who utilized adequate plus PNC displayed the highest

percentage for PPD symptoms, followed by women who utilized adequate PNC. Women who

utilized intermediate PNC displayed the lowest percentage for PPD symptoms, while the

percentage of women who utilized inadequate PNC was not considerably higher. Overall, chi-

square results in comparing chi-square statistics for 41 variables showed a significant difference









in percentages for PPD symptoms across all the variables (p<0.05), except for three variables

that did not show any significant difference (PNC paid by: Native American Health Services,

gender of the infant, and labor abnormalities), and one variable that was significantly different at

a p<0.10 (labor/delivery complications).

Since pre-pregnancy BMI was the primary independent variable in this study, chi-square

analyses were estimated to examine the relationship of characteristics among women from all

four pre-pregnancy BMI categories. Table 4-5 shows that there was a significant difference for

prenatal care (PNC) utilization frequencies among all pre-pregnancy BMI groups (p<0.05). The

percentages of PNC utilization for each pre-pregnancy BMI group were not considerably

different from the percentages presented in the univariate analysis for PNC utilization. It was

only for obese pre-pregnancy BMI that the percentage of women who utilized adequate plus

PNC slightly increased (about 4% higher) than the average percentage for the other pre-

pregnancy BMI groups.

Further addressing bivariate results for pre-pregnancy BMI, since women who had an

obese pre-pregnancy BMI were the focus of this study, it is especially worthy to note the

bivariate characteristics of women who had an obese pre-pregnancy BMI. For example, in

looking at the frequencies of the 17 maternal morbidities presented in Table 4-5 with pre-

pregnancy BMI, given the 14 morbidities in which there was a significant difference, except for

two variables: preterm labor and placental problems (placenta previa or abruptio), women who

had an obese pre-pregnancy BMI always comprised the highest percentage of women who

experienced those morbidities when comparing the percentages for all four pre-pregnancy BMI

groups. The highest percentage for preterm labor and placental problems was for women who

had an underweight pre-pregnancy BMI. To further note relevant bivariate statistics, women who









had an underweight pre-pregnancy BMI also had the highest percentage (26.6%) of low birth

weight babies (between 1,500 to 2,499 grams), while women who had an obese pre-pregnancy

BMI had the highest percentage (7.95%) of "very low birth weight" babies (less than 1,500

grams), compared to women from the three other pre-pregnancy BMI groups. Overall, chi-square

results in comparing chi-square statistics among women from all pre-pregnancy BMI groups for

40 variables showed a significant difference among all the characteristics (p<0.05), except for six

variables that did not show any significant difference (PNC paid by the military, gender of the

infant, weight gain talk during pregnancy, car crash injury, blood transfusion, and labor/delivery

complications).

Since PNC utilization was the secondary independent variable (the moderating variable) in

this study, chi-square analyses were estimated to examine the relationship of characteristics

among women from all four PNC utilization categories (Table 4-6). Overall, chi-square results

among all women in PNC utilization groups for 39 variables showed a significant difference

among all the characteristics (p<0.05), except for one variable that did not show any significant

difference (gender of the infant).

Due to a large number of women who utilized "adequate plus" PNC, chi-square statistics,

presented in Appendix B (pp. 156-187), were estimated to examine the relationship of

characteristics among women who were categorized as utilizing adequate plus PNC versus those

who were categorized as utilizing other quantities of PNC. Comparing pre-pregnancy body mass

indices (BMIs) among women who did not receive adequate plus PNC versus women who

utilized adequate plus PNC, there was a larger percentage of women in each pre-pregnancy BMI

group for women who did not receive adequate plus PNC, except for women who had an obese

pre-pregnancy BMI; hence, 24.6% of women who utilized adequate plus PNC had an obese pre-









pregnancy BMI, while 20.6% of women who "utilized other quantities of PNC" had an obese

pre-pregnancy BMI.

Finally, in comparing the morbidity variables among women who utilized adequate plus

PNC, overall results showed that the women who received adequate plus PNC were medically

and obstetrically high risk. These results provided a basis for controlling for high-risk

characteristics in the multivariate analyses.

A number of multivariate models were estimated in addition to 1) the univariate statistics

that gave a general overview of the sample characteristics, many of which were noteworthy, and

2) the bivariate statistics that gave an indication of frequencies and significance between each

variable of interest: PPD symptoms, pre-pregnancy BMI, PNC utilization, and adequate plus

PNC, and the other variables included in this study (e.g., control variables). Since the chi-square

analyses demonstrated significant differences between women with PPD symptoms versus

women without PPD symptoms, among all four pre-pregnancy BMI groups and all four PNC

utilization groups (the two independent variables of interest), a variety of multivariate analyses

were conducted to determine if these relationships would remain. The multivariate analyses,

which included a variety of logistic regression models, were estimated to 1) determine if

significance with the variables of interest (e.g., PPD symptoms, pre-pregnancy BMI) would be

demonstrated, thus supporting the specific aims and hypotheses, and 2) find the best model fit for

the specific aims and hypotheses proposed for this study.

Multivariate Analyses

A variety of logistic regression (logit) models were estimated in attempts to demonstrate

1) an association between pre-pregnancy BMI and PPD symptoms, and 2) a moderating effect of

PNC in the association between pre-pregnancy BMI and PPD symptoms. Logistic regression

models for both specific aims were estimated since this study sought to detect a moderating









effect of PNC utilization as opposed to a mediating effect. Six of these models are presented in

the main analyses, and the other ten models are presented in B (pp. 156-187). All odds ratios

were calculated using a 95% confidence interval.

Primary Risk-Adjusted Logistic Regression Analysis

Baseline model

Table 4-7 presents the results from the baseline logit model that was estimated for pre-

pregnancy BMI and PNC utilization only, without the control variables and interaction effects.

This model showed significance for obese pre-pregnancy BMI, suggesting an association

between women from this pre-pregnancy BMI category only, and PPD symptoms. Compared to

women who had a pre-pregnancy BMI of normal, women who had a pre-pregnancy BMI of

obese had 15% greater odds for PPD symptoms (OR=1.15, p=0.02). All PNC utilization

categories were significant in the baseline model, suggesting a general association between PNC

utilization and PPD symptoms. Compared to women who utilized adequate PNC, women who

utilized inadequate PNC had 84% greater odds (OR=1.84, p<0.0001) for PPD symptoms, while

women who utilized intermediate PNC had approximately one-fifth greater odds (OR=1.19,

p=0.02), and finally, women who utilized the highest quantity of PNC, adequate plus, had 28%

greater odds for PPD symptoms (OR=1.28, p<0.0001).

Specific aim 1

The model presented in Table 4-8 sought to detect an association between the main

effects of pre-pregnancy BMI and PPD symptoms, while taking into account the control

variables identified in the chi-square analyses. However, unlike the baseline model, only

borderline significance was found for underweight pre-pregnancy BMI. In contrast to the

hypothesis, this group had lower odds for PPD symptoms compared to women who had a normal

pre-pregnancy BMI (OR=0.87, p=0.08). Thus, specific aim 1 was not supported. In addition, the









association between PNC utilization and PPD symptoms was not found in this controlled

analysis.

Specific aim 2

The model presented in Table 4-9 sought to detect a moderating effect of PNC in the

association between pre-pregnancy BMI and PPD by including interaction terms between each

combination of pre-pregnancy BMI and PNC utilization category. This "all-inclusive" model

included the main effects, interaction effects, and control variables. However, despite the

significance seen for underweight pre-pregnancy BMI in the previous model, this effect

disappeared as significance was not seen for any of these main effects. No moderating effect of

PNC was apparent for any of the pre-pregnancy BMI groups either. Figure 4-2 presents a graph

of the odds ratios for the interaction effects within each pre-pregnancy BMI group.

Secondary Risk-Adjusted Logistic Regression: Subpopulation With Healthy Pregnancies

Another approach was taken to test the specific aims. The logistic regression models

carried out under this approach consisted of women who had normal, healthy pregnancy

characteristics. For a list of the characteristics that constituted high-risk characteristics and were

removed from these analyses, please refer to pp.68-70. The risk-adjustment process was

accomplished by removing high-risk pregnancies from the analysis.

Baseline model

This model, presented in Table 4-11, showed significance in the baseline model among

women who utilized inadequate PNC. Compared to women who utilized adequate PNC, they had

97% greater odds (OR=1.97, p<0.0001) for PPD symptoms (Table 4-15). Compared to the

baseline model for the entire population (Table 4-7), the significant relationship of obese pre-

pregnancy BMI with PPD symptoms, and the significant relationships of intermediate and

adequate plus PNC utilization with PPD symptoms disappeared.









Specific aim 1

The presented in Table 4-12, which included the main effects and the control variables, did

not show any significance among the main effects. Thus, after adding the control variables to the

baseline model for the logistic regression that removed high-risk pregnancies, the significance of

inadequate PNC also disappeared.

Specific aim 2

The all-inclusive model, presented in Table 4-13, showed borderline significance among

women who had a pre-pregnancy BMI of obese and utilized inadequate PNC. However, contrary

to what was expected, women who had a pre-pregnancy BMI of obese and utilized inadequate

PNC had roughly half the odds (OR=0.51, p<0.08) for PPD symptoms compared to women who

had a pre-pregnancy BMI of obese and utilized adequate PNC.

Wald Test

Tables 4-10 and 4-14, present the results from the main risk-adjusted logistic regression

and the risk-adjusted logistic regression that removed the high-risk pregnancies (respectively).

The results of the Wald tests indicated that the PNC categories within each pre-pregnancy BMI

group were not different from each other; hence, no moderating effect of PNC was seen after

performing these tests.

Model Fit

The Del measures calculated for the primary and secondary logistic regression models

were 0.03 and 0.02, respectively. The Del values of both models indicated a poor model fit in the

ability to predict PPD symptoms.









Table 4-1. Univariate statistics for all categorical variables included in the bivariate and
multivariate analyses
Categorical variables N Frequency (%)
Main variables
Primary analysis dependent variable: Postpartum depressive 45,285
symptoms (risk-adjusted logistic regression including all


pregnancies)
No
Yes
Secondary analysis dependent variable: Postpartum
depressive symptoms (risk-adjusted logistic regression
including healthy pregnancies only)
No
Yes
Main independent variable: Pre-pregnancy body mass index
(BMI)
Underweight
Normal
Overweight
Obese
Main independent variable: Adequacy of Prenatal Care
Utilization (APNCU) Index
Inadequate
Intermediate
Adequate
Adequate plus
Demographic control variables
Maternal race: White
No
Yes
Maternal race: Black
No
Yes
Maternal race: Other
No
Yes
Hispanic
Not Hispanic
Hispanic
Maternal education
0-8 years
9-11 years
12 years
13-15 years
16+ years


15,443




46,371






48,103







49,119


49,119


49,119


48,972


50,765


38,320 (84.62%)
6,965 (15.38%)



13,662 (88.5%)
1,781 (11.5%)


6,379 (13.76%)
23,834 (51.40%)
5,888 (12.70%)
10,270 (22.15%)


5,568 (11.58%)
6,529 (13.57%)
19,258 (40.03%)
16,748 (34.82%)


17,858 (36.36%)
31,261 (63.64%)

41,400 (84.29%)
7,719 (15.71%)

38,980 (79.36%)
10,139 (20.64%)

39,792 (81.55%)
9,000 (18.45%)

2,266 (4.46%)
7,294 (14.37%)
15,542 (30.62%)
11,871 (23.38%)
13,792 (27.17%)









Table 4-1. Continued
Categorical variables
Income (12 months prior)
Less than $10,000
$10,000 to $24,999
$25,000 to $49,999
$50,000 or more
Marital status
Married
Other
Insurance control variables
PNC paid by income
No
Yes
PNC paid by insurance/HMO
No
Yes
PNC paid by Medicaid
No
Yes
PNC paid by military


N
51,600


51,564


51,600


51,600


51,600


51,600


51,600



51,559


Yes
PNC paid by Native American Health Services
No
Yes
Pregnancy and delivery control variables
Birthweight
Less than 1,500 g
1,500 g -2,499 g
2,500 g or greater
Smoking during pregnancy
No
Yes
Vaginal delivery
No
Yes
Gender of infant
Male
Female
Infant in the intensive care unit (ICU)
No
Yes


51,147


51,544


51,599


50,758


Frequency (%)

10,489 (20.33%)
14,542 (28.18%)
11,765 (22.80%)
14,804 (28.69%)

32,635 (63.29%)
18,929 (36.71%)


41,349 (80.13%)
10,251 (19.87%)

25,598 (49.61%)
26,002 (50.39%)

30,395 (58.91%)
21,205 (39.16%)

50,488 (97.84%)
1,112 (2.16%)

51,097 (99.03%)
503 (0.009%)


2,754 (5.34%)
10,920 (21.18%)
37,885 (73.48%)

45,569 (89.09%)
5,578 (10.91%)

16,286 (31.60%)
35,258 (68.40%)

26,061 (50.51%)
25,538 (49.49%)

40,259 (79.32%)
10,499 (20.68%)









Table 4-1. Continued
Categorical variables
Pregnancy intention
No
Yes
Breastfed (ever)
No


Alcohol consumption in the last three months of
pregnancy
No
Yes
Women, Infants, and Children (WIC) during pregnancy
No
Yes
Subpopulation variable (Appendix B): Weight gain talk
during pregnancy
No
Yes
High-risk maternal morbidity control variables
Diabetes before pregnancy


Incompetent cervix
No
Yes
Preterm labor
No
Yes
Placenta previa or placenta abruptio
No
Yes
Bedrest
No
Yes
Car crash injury
No
Yes
Blood transfusion


Yes
Medical risk factors
No


N
50,893


50,408


50,603



50,878


9,377




51,600


51,600


51,600


51,600


51,600


51,600


51,600


51,600


Frequency (%)

26,143 (51.37%)
24,750 (48.63%)

9,454 (18.75%)
40,954 (81.25%)


47,291 (93.45%)
3,312 (6.55%)

27,037 (53.14%)
23,841 (46.86%)


2,088 (22.27%)
7,289 (77.73%)


50,505 (97.88%)
1,095 (2.12%)

50,640 (98.14%)
960 (1.86%)

38,005 (73.65%)
13,595 (26.35%)

48,142 (93.30%)
3,458 (6.70%)

40,099 (77.71%)
11,501 (22.29%)

50,698 (98.25%)
902(1.75%)

50,923 (98.69%)
677(1.31%)

33,391 (64.71%)
18,209 (35.29%)









Table 4-1. Continued
Categorical variables N Frequency (%)
Hospitalized during pregnancy 51,600
No 41,638 (80.69%)
Yes 9,962 (19.31%)
Non high-risk maternal morbidity control variables
Gestational diabetes 51,600
No 46,814 (90.73%)
Yes 4,786 (9.28%)
Kidney/bladder infection 51,600
No 42,178 (81.74%)
Yes 9,422 (18.26%)
Nausea 51,600
No 36,496 (70.73%)
Yes 15,104 (29.27%)
High blood pressure 51,600
No 43,739 (84.77%)
Yes 7,861 (15.23%)
Vaginal bleeding 51,600
No 42,682 (82.70%)
Yes 8,918 (17.28%)
Premature rupture of membrane (PROM) 51,600
No 46,282 (89.69%)
Yes 5,318 (10.31%)
Labor abnormalities 51,600
No 40,934 (79.33%)
Yes 10,666 (20.67%)
Labor/delivery complications 51,600
No 33,469 (64.86%)
Yes 18,131 (35.14%)
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI) and prenatal care (PNC)
utilization. The population for this table included all pregnancies and the years of PRAMS data
collection were for 2004 & 2005.

Table 4-2. Univariate statistics (continuous variable)
Variable N Mean Standard Minimum Maximum
(continuous) deviation
Maternal age 51,596 27.46 6.195 12 55
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variable was maternal age. The population for this table included all
pregnancies and the years of PRAMS data collection were for 2004 & 2005.










Table 4-3. Chi-square analyses comparing 41 characteristics among women with postpartum depressive (PPD) symptoms versus
women without postpartum depressive (PPD) symptoms
Categorical variables Dependent n for no PPD Dependent variable: n for yes PPD N P-value
variable: No PPD symptoms Yes PPD symptoms symptoms
symptoms (Frequency, %)
(Frequency, %)


Main variables
Main independent variable: Pre-
pregnancy body mass index
(BMI)
Underweight
Normal
Overweight
Obese
Main independent variable:
Adequacy of Prenatal Care
Utilization Index (APNCU)
Inadequate
Intermediate
Adequate
Adequate plus
Demographic control variables
Maternal race: White
No
Yes
Maternal race: Black
No
Yes
Maternal race: Other
No
Yes


34,427


4,663 (13.5%)
17,921 (52.1%)
4,352 (12.6%)
7,491 (21.8%)


6,291


6,413


40,718 <0.0001*


903 (14.4%)
2,979 (47.4%)
806 (12.8%)
1,603 (25.5%)


35,802


3,841 (10.7%)
4,848 (13.5%)
14,641 (40.9%)
12,472 (34.8%)


12,733 (35.2%)
23,430 (64.8%)

31,220 (86.3%)
4,943 (13.7%)

28,373 (78.5%)
7,790 (21.5%)


36,163


36,163


36,163


1,026 (16.0%)
913 (14.2%)
2,218 (34.6%)
2,256 (35.2%)


3,039 (45.6%)
3,672 (55.1%)

5,299 (79.5%)
1,367 (20.5%)

4,994 (74.9%)
1,672 (25.1%)


42,215 <0.0001*


6,666


6,666


6,666


42,829 <0.0001*


42,829 <0.0001*


42,829 <0.0001*










Table 4-3. Continued
Categorical variables Dependent n for no PPD Dependent variable: n for yes PPD N P-value
variable: No PPD symptoms Yes PPD symptoms symptoms
symptoms (Frequency, %)
(Frequency, %)


35,957 6,614 42,571 <0.0001*


Hispanic
Not Hispanic
Hispanic
Maternal education
0-8 years
9-11 years
12 years
13-15 years
16+ years
Income (12 months prior)
Less than $10,000
$10,000 to $24,999
$25,000 to $49,999
$50,000 or more
Marital status
Married
Other
Insurance control variables
PNC paid by income
No
Yes
PNC paid by insurance/HMO
No
Yes
PNC paid by Medicaid
No
Yes
PNC paid by military
No
Yes


37,695


38,320


29,905 (83.2%)
6,052 (16.8%)

1,486 (3.94%)
4,824 (12.8%)
11,252 (29.9%)
9,021 (23.9%)
11,112 (29.5%)

6,913 (18.0%)
10,258 (26.8%)
9,177 (23.9%)
11,972 (31.2%)

25,517 (66.6%)
12,775 (33.4%)


30,259 (79.0%)
8,061 (21.0%)

17,779 (46.4%)
20,541 (53.6%)

23,928 (62.4%)
14,392 (37.6%)

37,335 (97.4%)
985 (2.57%)


5,377 (81.3%)
1,237 (18.7%)

347 (5.08%)
1,492 (21.8%)
2,497 (36.5%)
1,559 (22.8%)
941 (13.8%)

2,316 (33.3%)
2,289 (32.9%)
1,392 (20.0%)
968 (13.9%)

3,425 (49.2%)
3,536 (50.8%)


5,739 (82.4%)
1,226 (17.6%)

4,528 (65.0%)
2,437 (35.0%)

3,108 (44.6%)
3,857 (55.4%)

6,849 (98.3%)
116(1.67%)


38,320


38,320


38,320


38,320


6,836


44,531 <0.0001*


6,965


45,285 <0.0001*


6,961


6,965


6,965


6,965


6,965


45,253 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


38,292










Table 4-3. Continued
Categorical variables Dependent n for no PPD Dependent variable: n for yes PPD N P-value
variable: No PPD symptoms Yes PPD symptoms symptoms
symptoms (Frequency, %)
(Frequency, %)


38,320 6,965 45,285 0.18


37,921 (98.9%)
399 (1.04%)


6,880 (98.8%)
85 (1.22%)


38,294


PNC paid by Native American
Health Services
No
Yes
Pregnancy and delivery control
variables
Birthweight
Less than 1,500 g
1,500 g to 2,499 g
2,500 g or greater
Smoking during pregnancy
No
Yes
Vaginal delivery
No
Yes
Gender of infant
Male
Female
Infant in the intensive care unit
(ICU)
No
Yes
Pregnancy intention
No
Yes
Breastfed (ever)
No
Yes


37,954


38,276


38,319


37,966


6,952


526 (7.57%)
1,589 (22.9%)
4,837 (69.6%)

5,647 (81.8%)
1,259 (18.2%)

2,263 (32.5%)
4,693 (67.5%)

3,552 (51.0%)
3,413 (49.0%)


5,150 (75.3%)
1,693 (24.7)

4,440 (64.7%)
2,419 (35.3%)

1,564 (23.3%)
5,147 (76.7%)


1,914 (5.00%)
7,702 (20.1%)
28,678 (74.9%)

34,126 (98.9%)
3,828 (10.1%)

11,933 (31.2%)
26,343 (68.8%)

19,370 (50.5%)
18,949 (49.5%)


30,591 (80.6%)
7,375 (19.4%)

18,514 (48.9%)
19,335 (51.1%)

6,589 (17.4%)
31,189 (82.6%)


45,246 <0.0001*


6,906


6,956


6,965


6,843


44,860 <0.0001*


45,232 0.025*


45,284 0.49


44,809 <0.0001*


6,859


6,711


44,708 <0.0001*


44,489 <0.0001*


37,849


37,778










Table 4-3. Continued
Categorical variables Dependent n for no PPD Dependent variable: n for yes PPD N P-value
variable: No PPD symptoms Yes PPD symptoms symptoms
symptoms (Frequency, %)
(Frequency, %)


37,749 6,841 44,590 0.034*


35,359 (93.7%)
2,390 (6.33%)


21,370 (56.4%)
16,551 (43.6%)


37,921


6,361 (93.0%)
480 (7.00%)


2,641 (38.4%)
4,233 (61.6%)


6,874


4,990


Alcohol consumption in the last
three months of pregnancy
No
Yes
Women, Infants, and Children
(WIC) during pregnancy
No
Yes
Subpopulation variable (Appendix
B): Weight gain talk during
pregnancy
No
Yes
High-risk maternal morbidity
control variables
Diabetes before pregnancy
No
Yes
Incompetent cervix
No
Yes
Preterm labor
No
Yes
Placenta previa or placenta
abruption
No
Yes
Bedrest
No
Yes


37,576 (98.0%)
744 (1.94%)

37,661 (98.3%)
659 (1.72%)

28,714 (74.9%)
9,606 (25.1%)


35,818 (93.5%)
2,502 (6.53%)

30,153 (78.7%)
8,167 (21.3%)


44,795 <0.0001*


5,794 <0.0001*


224 (27.9%)
580 (72.1%)


38,320


38,320


38,320


38,320


6,746 (96.9%)
219(3.14%)

6,794 (97.5%)
171 (2.46%)

4,510 (64.8%)
2,455 (35.2%)


6,409 (92.0%)
556 (7.98%)

4,918 (70.6%)
2,047 (29.3%)


38,320


6,965


6,965


6,965


6,965


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


6,965


45,285 <0.0001*


1,084 (21.7%)
3,906 (78.3%)










Table 4-3. Continued
Categorical variables Dependent n for no PPD Dependent variable: n for yes PPD N P-value
variable: No PPD symptoms Yes PPD symptoms symptoms
symptoms (Frequency, %)
(Frequency, %)


38,320 6,965 45,285 <0.0001*


Car crash injury
No
Yes
Blood transfusion
No
Yes
Medical risk factors
No
Yes
Hospitalized during pregnancy
No
Yes
High-risk maternal morbidity
control variables
Gestational diabetes
No
Yes
Kidney/bladder infection
No
Yes
Nausea
No
Yes
High blood pressure
No
Yes


38,320


38,320


38,320


6,797 (97.6%)
168 (2.41%)

6,812 (97.8%)
153 (2.20%)

4,304 (61.8%)
2,661 (38.2%)

5,269 (75.6%)
1,696 (24.4%)


37,686 (98.4%)
634 (1.65%)

37,895 (98.9%)
425 (1.11%)

25,029 (65.3%)
13,291 (34.7%)

31,156 (81.3%)
7,164 (18.7%)



34,890 (91.0%)
3,430 (8.95%)

31,839 (83.1%)
6,481 (16.9%)

27,787 (72.5%)
10,533 (27.4%)

32,478 (84.8%)
5,842 (15.2%)


6,210 (89.2%)
755 (10.8%)

5,172 (74.3%)
1,793 (25.7%)

4,137 (59.4%)
2,828 (40.6%)

5,760 (82.7%)
1,205 (17.3%)


6,965


6,965


6,965


6,965


6,965


6,965


6,965


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


45,285 <0.0001*


38,320


38,320


38,320


38,320










Table 4-3. Continued
Categorical variables Dependent n for no PPD Dependent variable: n for yes PPD N P-value
variable: No PPD symptoms Yes PPD symptoms symptoms
symptoms (Frequency, %)
(Frequency, %)
Vaginal bleeding 38,320 6,965 45,285 <0.0001*


No
Yes
Premature rupture of membrane
(PROM)
No
Yes
Labor abnormalities
No
Yes
Labor/delivery complications
No
Yes


31,937 (83.3%)
6,383 (16.7%)


34,536 (90.1%)
3,784 (9.87%)

29,741 (77.6%)
8,579 (22.4%)

25,010 (65.3%)
13,310 (34.7%)


38,320


5,486 (14.3%)
1,479 (21.2%)


6,119 (87.9%)
846 (12.1%)

5,356 (76.9%)
1,609 (23.1%)

4,463 (64.1%)
2,502 (35.9%)


38,320


38,320


6,965


45,285 <0.0001*


6,965


6,965


45,285 0.19


45,285 0.06**


The dependent variable for this table was postpartum depression (PPD), while the main independent variables were pre-pregnancy
body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included all pregnancies and the years of
PRAMS data collection were for 2004 & 2005.

Table 4-4. Maternal age (continuous variable) and postpartum depression (PPD) symptoms t-test results
Group Observations Mean Standard error Standard deviation 95% Confidence interval
(PPD symptoms) (lower, upper)
No 38,317 27.65 .0313 6.12 (27.58, 27.71)
Yes 6,964 25.95 .0749 6.25 (25.81, 26.10)
Combined 45,281 27.38 .0290 6.17 (27.33, 27.44)
Difference -------- 1.692 .0799 -------- (1.535, 1.849)
The dependent variable for this table was postpartum depression (PPD), while the main independent variable was maternal age. The
population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.










Table 4-5. Chi-square analyses comparing 40 characteristics among women from four body mass index (BMI) groups


Categorical variable


Main variable
Adequacy of Prenatal
Care Utilization
Index (APNCU)
Inadequate
Intermediate
Adequate
Adequate plus
Demographic control
variables
Maternal race: White
No
Yes
Maternal race: Black
No
Yes
Maternal race: Other
No
Yes
Hispanic
Not Hispanic
Hispanic
Maternal education
0-8 years
9-11 years
12 years
13-15 years
16+ years


Dependent
variable:
Underweight
pre-pregnancy
BMI


n Dependent
variable:
Normal pre-
pregnancy
BMI


5,964


747(12.5%)
798 (13.4%)
2,374 (39.8%)
2,045 (34.3%)


2,349 (37.0%)
3,993 (63.0%)

5,507 (86.8%)
835 (13.2%)

4,828 (76.1%)
1,514 (23.9%)

5,458 (86.8%)
829 (13.2%)

187(3.00%)
1,080 (17.3%)
1,944 (31.1%)
1,309 (20.9%)
1,735 (27.7%)


2,460 (11.1%)
3,243 (14.6%)
9,179 (41.2%)
7,384 (33.2%)


6,342


6,342


6,342


6,287


6,255


7,976 (33.7%)
15,711 (66.3%)

20,525 (86.7%)
3,162 (13.3%)

18,873 (79.7%)
4,814 (20.3%)

19,895 (84.6%)
3,618 (15.4%)

639 (2.70%)
3,075 (13.1%)
6,711 (28.6%)
5,374 (22.9%)
7,659 (32.6%)


n Dependent
variable:
Overweight
pre-pregnancy
BMI


22,266


642(11.7%)
757(13.8%)
2,216 (40.3%)
1,880 (34.2%)


23,687


23,687


23,687


23,513


23,458


2,223 (38.0%)
3,624 (62.6%)

4,764 (81.5%)
1,083 (18.5%)

4,707 (80.5%)
1,140 (19.5%)

4,717(18.4%)
1,076 (18.6%)

226 (3.90%)
803 (13.7%)
1,835 (31.7%)
1,499 (25.9%)
1,427 (24.6%)


n Dependent
variable:
Obese pre-
pregnancy
BMI


5,495


n N P-value


9,513 43,238 <0.0001*


1,058 (11.1%)
1,190 (12.5%)
3,579 (37.6%)
3,686 (38.7%)


5,847


5,847


5,847


5,793


5,790


4,236 (41.6%)
5,952 (58.4%)

7,909 (77.6%)
2,279 (22.4%)

8,231 (80.8%)
1,957 (19.2%)

8,511 (83.9%)
1,629 (16.1%)

347 (3.43%)
1,357 (13.4%)
3,570 (35.3%)
2,898 (28.7%)
1,939 (19.2%)


10,188 46,064 <0.0001*


10,188 46,064 <0.0001*


10,188 46,064 <0.0001*


10,140 45,733 <0.0001*


10,111 45,614 <0.0001*










Table 4-5. Continued
Categorical variable


Income (12 months
prior)
Less than $10,000
$10,000 to $24,999
$25,000 to $49,999
$50,000 or more
Marital status
Married
Other
Insurance control
variables
PNC paid by income
No
Yes
PNC paid by
insurance/HMO
No
Yes
PNC paid by
Medicaid
No
Yes
PNC paid by military
No
Yes
PNC paid by Native
American Health
Services
No
Yes


Dependent
variable:
Underweight
pre-pregnancy
BMI


1,485 (23.3%)
1,899 (29.8%)
1,248 (19.6%)
1,747 (27.4%)

3,836 (60.2%)
2,540 (39.8%)



5,170 (81.0%)
1,209 (19.0%)


3,364 (52.7%)
3,015 (47.3%)


3,647 (57.2%)
2,732 (42.8%)

6,237 (97.8%)
142 (2.23%)


6,343 (99.4%)
36 (0.56%)


n Dependent
variable:
Normal pre-
pregnancy BMI


6,379


6,376




6,379


6,379



6,379



6,379


6,379


4,402 (18.5%)
6,105 (25.6%)
5,229 (21.9%)
8,098 (34.0%)

15,630 (65.6%)
8,202 (34.4%)



18,964 (79.6%)
4,870 (20.4%)


10,677 (44.8%)
13,157 (55.2%)


15,148 (63.6%)
8,686 (36.4%)

23,237 (97.5%)
597(2.50%)


n Dependent
variable:
Overweight
pre-pregnancy
BMI
23,834


23,832




23,834


23,834



23,834



23,834


23,834


23,615 (99.1%)
219 (0.92%)


1,150 (19.5%)
1,691 (28.7%)
1,418 (24.1%)
1,629 (27.7%)

3,737 (63.5%)
2,147 (36.5%)



4,747(80.6%)
1,141 (19.4%)


2,917 (49.5%)
2,971 (50.5%)


3,418 (58.1%)
2,470 (41.9%)

5,750 (97.7%)
138 (2.34%)


5,801 (98.5%)
87 (1.48%)


n Dependent
variable: Obese
pre-pregnancy
BMI


5,888


5,884




5,888


5,888



5,888



5,888


5,888


n N P-value


10,270 46,371 <0.0001*


2,166 (21.1%)
3,103 (30.2%)
2,684 (26.1%)
2,317 (22.6%)

6,345 (61.8%)
3,917 (38.2%)



8,272 (80.5%)
1,998 (19.5%)


5,365 (52.2%)
4,905 (47.8%)


5,441 (53.0%)
4,829 (47.0%)

10,053 (97.9%)
217(2.11%)


10,262 46,354 <0.0001*




10,270 46,371 0.02*


10,270 46,371 <0.0001*



10,270 46,371 <0.0001*



10,270 46,371 0.15


10,270 46,371 <0.0001*


10,129 (98.6%)
141 (1.37%)










Table 4-5. Continued


Categorical variable




Pregnancy and
delivery control
variables
Birthweight
Less than 1,500 g
1,500 g to 2,499 g
2,500 g or greater
Smoking during
pregnancy
No
Yes
Vaginal delivery
No
Yes
Gender of infant
Male
Female
Infant in the
intensive care unit
(ICU)
No
Yes
Pregnancy intention
No
Yes
Breastfed (ever)
No
Yes


Dependent
variable:
Underweight
pre-pregnancy
BMI




318(4.99%)
1,697 (26.6%)
4,358 (68.4%)


5,522 (87.3%)
805 (12.7%)

1,586 (24.9%)
4,787 (75.1%)

3,161 (49.6%)
3,218 (50.4%)



5,002 (79.9%)
1,258 (20.1%)

3,451 (54.9%)
2,841 (45.1%)

1,219 (19.6%)
4,986 (80.4%)


n Dependent
variable:
Normal pre-
pregnancy
BMI


6,373



6,327



6,373


6,379


6,260




6,292


6,205


1,075 (4.51%)
5,052 (21.2%)
17,692 (74.3%)


21,258 (90.0)
2,365 (10.0%)

6,847 (28.8%)
16,960 (71.2%)

12,019 (50.4%)
11,814 (49.6%)



18,922 (80.6%)
4,564 (19.4%)

11,634 (49.4%)
11,913 (50.6%)

3,885 (16.6%)
19,449 (83.4%)


n Dependent
variable:
Overweight
pre-pregnancy
BMI


23,819



23,623



23,807


23,833


23,486




23,547


23,334


338 (5.74%)
1,127 (19.2%)
4,415 (75.1%)


5,216 (89.4%)
621 (10.6%)

1,946 (33.1%)
3,934 (66.9%)

3,014 (51.2%)
2,874 (48.8%)



4,605 (79.3%)
1,205 (20.7%)

3,051 (52.4%)
2,769 (47.6%)

1,161 (20.1%)
4,606 (79.9%)


n Dependent
variable: Obese
pre-pregnancy
BMI


5,880



5,837



5,880


5,888


5,810




5,820


5,767


816(7.95%)
2,000 (19.5%)
7,449 (72.6%)


9,000 (88.4%)
1,176 (11.6%)

4,381 (42.7%)
5,877 (57.3%)

5,195 (50.6%)
5,075 (49.4%)



7,633 (75.4%)
2,494 (24.6%)

5,566 (54.9%)
4,572 (45.1%)

2,315 (23.1%)
7,707 (76.9%)


n N P-value








10,265 46,337 <0.0001*



10,176 45,963 <0.0001*



10,258 46,318 <0.0001*


10,270 46,370 0.332


10,127 45,683 <0.0001*




10,138 45,797 <0.0001*


10,022 45,328 <0.0001*










Table 4-5. Continued
Categorical variable Dependent
variable:
Underweight
pre-pregnancy
BMI


Alcohol
consumption in the
last three months of
pregnancy
No
Yes
Women, Infants, and
Children (WIC)
during pregnancy
No
Yes
Subpopulation
variable (Appendix
B): Weight gain talk
during pregnancy
No
Yes
High-risk maternal
morbidity control
variables
Diabetes before
pregnancy
No
Yes
Incompetent cervix
No
Yes
Preterm labor
No
Yes


5,837 (93.3%)
422 (6.74%)



3,364 (53.5%)
2,927 (46.5%)


N Dependent
variable:
Normal pre-
pregnancy
BMI
6,259


6,291




1,008


213 (21.1%)
795 (78.9%)


21,631 (92.5%)
1,750 (7.48%)



13,885 (59.1%)
9,627 (40.9%)


n Dependent
variable:
Overweight
pre-pregnancy
BMI
23,381


23,512




3,543


830 (23.4%)
2,713 (76.6%)


6,379


6,327 (99.2%)
52 (0.82%)

6,268 (98.3%)
111(1.74%)

4,462 (69.9%)
1,917 (30.1%)


6,379


6,379


5,437(94.1%)
342 (5.91%)



3,002 (51.6%)
2,811 (49.4%)


n Dependent
variable: Obese
pre-pregnancy
BMI


5,779


5,813




835


199 (23.8%)
636 (76.2%)


23,834


23,531 (98.7%)
303 (1.27%)

23,440 (98.3%)
394(1.65%)

17,673 (74.2%)
6,161 (25.8%)


23,834


23,834


9,657 (95.5%)
454 (4.49%)



4,629 (45.6%)
5,516 (54.4%)


n N P-value


10,111 45,530 0.034*


10,145 45,761 <0.0001*


1,365 6,751 0.359


328 (24.0%)
1,037 (76.0%)


5,888


5,752 (97.7%)
136 (2.30%)

5,762 (97.9%)
126 (2.14%)

4,352 (73.9%)
1,536 (26.1%)


5,888


5,888


10,270 46,371 <0.0001*


9,808 (95.5%)
462 (4.50%)

10,023 (97.6%)
247 (2.40%)

7,476 (72.8%)
2,794 (27.2%)


10,270 46,371 <0.0001*


10,270 46,371 <0.0001*










Table 4-5. Continued


Categorical variable


Placenta previa or
placenta abruption
No
Yes
Bedrest
No
Yes
Car crash injury
No
Yes
Blood transfusion
No
Yes
Medical risk factors
No
Yes
Hospitalized during
pregnancy
No
Yes
Non high-risk
maternal morbidity
control variables
Gestational diabetes
No
Yes
Kidney/bladder
infection
No
Yes


Dependent
variable:
Underweight
pre-pregnancy
BMI


5,872 (92.1%)
507 (7.94%)

4,978 (78.0%)
1,401 (22.0%)

6,261 (98.2%)
118(1.85%)

6,278 (98.4%)
101 (1.58%)

4,366 (68.4%)
2,013 (31.6%)


5,122 (80.3%)
1,257 (19.7%)


6,023 (94.4%)
356 (5.58%)


5,166 (81.0%)
1,213 (19.0%)


n Dependent
variable:
Normal pre-
pregnancy
BMI
6,379


6,379


6,379


6,379


6,379


6,379


6,379


6,379


22,261 (93.4%)
1,573 (6.60%)

18,914 (79.4%)
4,920 (20.6%)

23,435 (98.3%)
399(1.67%)

23,516 (98.7%)
318(1.33%)

16,083 (67.5%)
7,751 (32.5%)


19,436 (81.5%)
4,398 (18.5%)


22,191 (93.1%)
1,643 (6.89%)


19,781 (83.0%)
4,053 (17.0%)


n Dependent
variable:
Overweight
pre-pregnancy
BMI
23,834


23,834


23,834


23,834


23,834


23,834


23,834


23,834


5,521 (93.8%)
367 (6.23%)

4,509 (76.6%)
1,379 (23.4%)

5,785 (98.3%)
103 (1.75%)

5,819 (98.3%)
69(1.17%)

3,750 (63.7%)
2,138 (36.3%)


4,761 (80.9%)
1,127 (19.1%)


5,244 (89.1%)
644 (10.9%)


4,770 (81.0%)
1,118 (19.0%)


n Dependent
variable:
Obese pre-
pregnancy
BMI
5,888


5,888


5,888


5,888


5,888


5,888


5,888


5,888


9,542 (92.9%)
728 (7.10%)

7,372 (71.8%)
2,898 (28.2%)

10,070 (98.1%)
200 (1.95%)

10,145 (98.8%)
125 (1.22%)

5,776 (56.2%)
4,494 (43.8%)


7,881 (76.7%)
2,389 (23.3%)


8,641 (84.1%)
1,629 (15.9%)


8,110 (79.0%)
2,160 (21.0%)


n N P-value


10,270 46,371 <0.0001*


10,270 46,371 <0.0001*


10,270 46,371 0.34


10,270 46,371 0.15


10,270 46,371 <0.0001*


10,270 46,371 <0.0001*


10,270 46,371 <0.0001*


10,270 46,371 <0.0001*










Table 4-5. Continued
Categorical variable




Nausea
No
Yes
High blood pressure
No
Yes
Vaginal bleeding
No
Yes
Premature rupture of
membrane (PROM)
No
Yes
Labor/delivery
complications
No
Yes
Labor abnormalities
No
Yes


Dependent
variable:
Underweight
pre-pregnancy
BMI

4,504 (70.6%)
1,875 (29.4)

5,848 (91.7%)
531(8.32%)

5,277 (82.7%)
1,102 (17.3%)


5,683 (89.1%)
696 (10.9%)


4,181 (65.5%)
2,198 (34.5%)

5,197 (81.5%)
1,182 (18.5%)


n Dependent
variable:
Normal pre-
pregnancy
BMI


6,379


6,379


6,379


6,379



6,379



6,379


17,330 (72.7%)
6,504 (27.3%)

20,768 (87.1%)
3,066 (12.9%)

19,844 (83.3%)
3,990 (16.7%)


21,465 (90.1%)
2,369 (9.94%)


15,554 (65.3%)
8,280 (34.7%)

19,288 (80.9%)
4,546 (19.1%)


n Dependent
variable:
Overweight
pre-pregnancy
BMI


23,834


23,834


23,834


23,834



23,834



23,834


4,101 (69.7%)
1,787 (30.3%)

4,850 (82.4%)
1,038 (17.6%)

4,849 (82.4%)
1,039 (17.6%)


5,289 (89.8%)
599(10.2%)


3,814 (64.8%)
2,074 (35.2%)

4,667 (79.3%)
1,221 (20.7%)


n Dependent
variable:
Obese pre-
pregnancy
BMI


5,888


5,888


5,888


5,888



5,888



5,888


6,880 (67.0%)
3,390 (33.3%)

7,655 (74.5%)
2,615 (25.5%)

8,305 (80.9%)
1,965 (19.1%)


9,135 (88.9%)
1,135 (11.1%)


6,614 (64.4%)
3,656 (35.6%)

7,888 (76.8%)
2,382 (23.2%)


n N P-value




10,270 46,371 <0.0001*


10,270 46,371 <0.0001*


10,270 46,371 <0.0001*


10,270 46,371 0.007*



10,270 46,371 0.36



10,270 46,371 <0.0001*


The dependent variable for this table was pre-pregnancy body mass index (BMI), while the main independent variables were pre-
prenatal care (PNC) utilization and control variables The population for this table included all pregnancies and the years of PRAMS
data collection were for 2004 & 2005.











Table 4-6. Chi-square analyses comparing 39 characteristics among women from four prenatal care (PNC) utilization groups


Categorical variable




Demographic control
variables
Maternal race: White
No
Yes
Maternal race: Black
No
Yes
Maternal race: Other
No
Yes
Hispanic
Not Hispanic
Hispanic
Maternal education
0-8 years
9-11 years
12 years
13-15 years
16+ years
Income (12 months prior)
Less than $10,000
$10,000 to $24,999
$25,000 to $49,999
$50,000 or more
Marital status
Married
Other
Insurance control variables
PNC paid by income
No
Yes
PNC paid by insurance/HMO
No
Yes


Dependent
variable:
Inadequate
PNC
utilization



2,579 (47.5%)
2,846 (52.5%)

4,336 (79.9%)
1,089 (20.1%)

3,935 (72.5%)
1,490 (27.5%)

3,974 (73.6%)
1,429 (26.4%)

462 (8.48%)
1,443 (26.5%)
1,937 (35.6%)
1,027 (18.9%)
579 (10.6%)

2,023 (36.3%)
2,096 (37.6%)
900 (16.2%)
549 (9.86%)

2,362 (42.5%)
3,202 (57.5%)


4,590 (82.4%)
978 (17.6%)

4,168 (74.9%)
1,400 (25.1%)


n Dependent
variable:
Intermediate
PNC utilization


5,425


5,425


5,425


5,403


5,448





5,568




5,564



5,568


5,568


2,652 (41.6%)
3,717 (58.4%)

5,412 (85.0%)
957 (15.0%)

4,674 (73.4%)
1,695 (26.6%)

4,956 (78.5%)
1,361 (21.5%)

357 (5.56%)
985 (15.4%)
1,965 (30.6%)
1,494 (23.3%)
1,618 (25.2%)

1,403 (21.5%)
1,918 (29.4%)
1,473 (22.6%)
1,735 (26.7%)

4,083 (62.5%)
2,445 (37.5%)


5,299 (81.2%)
1,230 (18.8%)

3,478 (53.3%)
3,051 (46.7%)


n Dependent
variable:
Adequate PNC
utilization


6,369


6,369


6,369


6,317


6,419





6,529




6,528



6,529


6,529


6,057 (33.1%)
12,248 (66.9%)

15,922 (87.0%)
2,383 (13.0%)

14,631 (79.9%)
3,674 (20.1%)

14,986 (82.3%)
3,220 (17.7%)

719 (3.67%)
2,282 (11.6%)
5,523 (28.2%)
4,547 (23.2%)
5,935 (30.3%)

3,237 (16.8%)
4,973 (25.8%)
4,754 (24.7%)
6,294 (32.7%)

13,125 (68.2%)
6,128 (31.8%)


15,239 (79.1%)
4,019 (20.9%)

8,500 (44.1%)
10,758 (55.9%)


n Dependent variable:
Adequate plus PNC
utilization


18,305


18,305


18,305


18,206


19,600





19,258




19,253



19,258


19,258


4,912 (31.3%)
10,789 (68.7%)

13,093 (83.4%)
2,608 (16.6%)

13,397 (85.3%)
2,304 (14.7%)

13,322 (85.4%)
2,280 (14.6%)

530 (3.20%)
2,013 (12.2%)
5,048 (30.5%)
4,046 (24.4%)
4,928 (29.8%)

2,944 (17.6%)
4,495 (26.8%)
3,887 (23.2%)
5,422 (32.4%)

11,142 (66.6%)
5,588 (33.4%)


13,294 (79.4%)
3,454 (3454%)

7,418 (44.3%)
9,330 (55.7%)


n N P-value







15,701 45,800 <0.0001*


15,701 45,800 <0.0001*


15,701 45,800 <0.0001*


15,602 45,528 <0.0001*


16,565


47,438 <0.0001*


16,748 48,103 <0.0001*




16,730 48,075 <0.0001*



16,748 48,103 <0.0001*


16,748 48,103 <0.0001*











Table 4-6. Continued
Categorical variable




PNC paid by Medicaid
No
Yes
PNC paid by military
No
Yes
PNC paid by Native American
Health Services
No
Yes
Pregnancy and delivery
control variables
Birthweight
Less than 1,500 g
1,500 g to 2,499 g
2,500 g or greater
Smoking during pregnancy
No
Yes
Vaginal delivery
No
Yes
Gender of infant
Male
Female
Infant in the intensive care
unit (ICU)
No
Yes
Pregnancy intention
No
Yes
Breastfed (ever)
No
Yes


Dependent
variable:
Inadequate
PNC
utilization

2,326 (41.8%)
3,242 (58.2%)

5,489 (98.6%)
79(1.42%)


5,472 (98.3%)
96(1.72%)



304 (5.46%)
1,178 (21.2%)
4,082 (73.4%)

4,558 (82.8%)
950 (17.2%)

1,538 (27.6%)
4,029 (72.4%)

2,785 (50.0%)
2,783 (50.0%)


4,267 (78.9%)
1,141 (21.1%)

3,870 (70.6%)
1,614 (29.4%)

1,337 (25.2%)
3,961 (74.8%)


n Dependent
variable:
Intermediate
PNC utilization


5,568


5,568


5,568


5,564



5,508


5,567


5,408



5,484


5,298


3,800 (58.2%)
2,729 (41.8%)

6,247 (95.7%)
282 (4.32%)


6,410 (98.2%)
119(1.82%)



157 (2.41%)
846 (13.0%)
5,526 (84.6%)

5,840 (90.1%)
644 (9.93%)

1,708 (26.2%)
4,819 (73.8%)

3,360 (51.5%)
3,169 (48.5%)


5,502 (85.8%)
909 (14.2%)

3,382 (53.4%)
3,056 (48.2%)

1,066 (16.7%)
5,299 (83.3%)


n Dependent
variable:
Adequate PNC
utilization


6,529


6,529


6,529


6,529



6,484


6,527


6,529


6,411



6,338


6,365


12,308 (63.9%)
6,950 (36.1%)

18,874 (98.0%)
474 (2.46%)


19,096 (99.2%)
162 (0.84%)



403 (2.09%)
2,531 (13.2%)
16,320 (84.8%)

17,420 (90.9%)
1,750 (9.13%)

5,335 (27.7%)
13,911 (72.3%)

9,758 (50.7%)
9,500 (49.3%)


16,633 (87.4%)
2,395 (12.6%)

9,098 (47.9%)
9,903 (52.1%)

3,148 (16.6%)
15,788 (83.4%)


n Dependent variable:
Adequate plus PNC
utilization


19,258


19,258


19,258


19,254



19,170


19,246


19,258


19,028



19,001


18,936


10,143 (60.6%)
6,605 (39.4%)

16,526 (98.7)
222 (1.33%)


16,649 (99.4%)
99 (0.59%)



1,622 (9.68%)
5,664 (33.8%)
9,462 (56.5%)

14,793 (88.8%)
1,875 (11.2%)

6,542 (39.1%)
10,195 (60.9%)

8,437 (50.4%)
8,310 (49.6%)


11,308 (68.6%)
5,185 (31.4%)

7,822 (47.3%)
8,701 (52.7%)

3,256 (19.8%)
13,174 (80.2%)


n N P-value




16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,095 <0.0001*


16,668


47,830 <0.0001*


16,737 48,077 <0.0001*


16,747 48,102 0.332


16,493 47,340 <0.0001*



16,523 47,446 <0.0001*


16,430 47,029 <0.0001*











Table 4-6. Continued
Categorical variable




Alcohol consumption in the
last three months of
pregnancy
No
Yes
Women, Infants, and Children
(WIC) during pregnancy
No
Yes
Subpopulation variable
(Appendix B): Weight gain
talk during pregnancy
No
Yes
High-risk maternal morbidity
control variables
Diabetes before pregnancy
No
Yes
Incompetent cervix
No
Yes
Preterm labor
No
Yes
Placenta previa or placenta
abruption
No
Yes
Bedrest
No
Yes
Car crash injury
No
Yes
Blood transfusion
No
Yes


Dependent
variable:
Inadequate
PNC
utilization


n Dependent
variable:
Intermediate
PNC utilization


5,419


5,074 (93.6%)
345 (6.37%)


2,137 (39.2%)
3,314 (60.8%)



225 (27.1%)
605 (72.9%)



5,427 (97.5%)
141 (2.53%)

5,468 (98.2%)
100 (1.80%)

4,159 (74.7%)
1,409 (25.3%)


5,281 (94.8%)
287 (5.15%)

4,467 (80.2%)
1,101 (19.8%)

5,464 (98.1%)
104 (1.87%)

5,469 (98.2%)
99(1.78%)


5,451



830


5,568


5,568


5,568



5,568


5,568


5,568


n Dependent
variable:
Adequate PNC
utilization


6,381


5,932 (93.0%)
449 (7.0%)


3,313 (51.5%)
3,126 (48.5%)



176 (25.4%)
518 (74.6%)



6,446 (98.7%)
83 (1.27%)

6,448 (98.8%)
81(1.24%)

5,334 (81.7%)
1,195 (18.3%)


6,242 (95.6%)
287 (4.4%)

5,526 (84.6%)
1,003 (15.4%)

6,435 (98.6%)
94(1.44%)

6,460 (98.9%)
69(1.06%)


6,439



694


6,529


6,529


6,529


6,529



6,529


6,529


6,529


n Dependent variable:
Adequate plus PNC
utilization


18,935


17,616 (93.0%)
1,319 (6.97%)


10,953 (57.6%)
8,064 (42.4%)



887 (22.4%)
3,082 (77.7%)



18,986 (98.6%)
272(1.41%)

19,044 (98.9%)
214(1.11%)

15,482 (80.4%)
3,776 (19.6%)


18,289 (95.0%)
969 (5.03%)

16,144 (83.8%)
3,114 (16.2%)

18,948 (98.4%)
310(1.61%)

19,093 (99.1%)
165 (0.09%)


19,017



3,969


19,258


19,258


19,258


19,258



19,258


19,258


19,258


n N P-value


16,446 47,181 0.034*


15,486 (94.2%)
960 (5.84%)


9,066 (54.8%)
7,467 (45.2%)



691 (20.5%)
2,675 (79.5%)



16,245 (97.0%)
503 (3.00%)

16,271 (97.2%)
477 (2.84%)

10,528 (62.9%)
6,220 (37.1%)


15,111 (90.2%)
1,637(9.77%)

11,295 (67.4%)
5,453 (32.6%)

16,431 (98.1%)
317(1.89%)

16,466 (98.3%)
282 (1.68%)


16, 533 47,440 <0.0001*



3,366 8,859 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*



16,748 48,103 <0.0001*


16,748 48,103 0.04*


16,748 48,103 <0.0001*











Table 4-6. Continued
Categorical variable




Medical risk factors
No
Yes
Hospitalized during
pregnancy
No
Yes
Non high-risk maternal
morbidity control variables
Gestational diabetes
No
Yes
Kidney/bladder infection
No
Yes
Nausea
No
Yes
High blood pressure
No
Yes
Vaginal bleeding
No
Yes
Premature rupture of
membrane (PROM)
No
Yes
Labor abnormalities
No
Yes
Labor/delivery complications
No
Yes


Dependent
variable:
Inadequate
PNC
utilization

3,464 (62.2%)
2,104 (37.8%)


4,582 (82.3%)
986 (17.7%)



5,130 (92.1%)
438 (7.87%)

4,360 (78.3%)
1,208 (21.7%)

3,940 (70.8%)
1,628 (29.2%)

4,778 (85.8%)
790 (14.2%)

4,771 (85.9%)
797 (14.3%)


5,033 (90.4%)
535 (9.60%)

4,385 (78.8%)
1,183 (21.2%)

3,507 (63.0%)
2,061 (37.0%)


n Dependent
variable:
Intermediate
PNC utilization


5,568


5,568


5,568


5,568


5,568


5,568


5,568


5,568



5,568


5,568


4,636 (71.0%)
1,893 (29.0%)


5,733 (87.8%)
796 (12.2%)



6,051 (92.7%)
478 (7.32%)

5,477 (83.9%)
1,052 (16.1%)

4,761 (72.9%)
1,768 (27.1%)

5,850 (89.6%)
679 (10.4%)

5,641 (86.3%)
888 (13.6%)


6,097 (93.4%)
432 (6.61%)

5,193 (79.5%)
1,336 (20.5%)

4,401 (67.4%)
2,128 (32.6%)


n Dependent
variable:
Adequate PNC
utilization


6,529


6,529


6,529


6,529


6,529


6,529


6,529


6,529



6,529


6,529


13,217 (68.6%)
6,041 (31.4%)


16,932 (87.9%)
2,326 (12.1%)



17,774 (92.3%)
1,484 (7.71%)

16,131 (83.8%)
3,127 (16.2%)

14,024 (72.8%)
5,234 (27.2%)

17,108 (88.8%)
2,150 (11.2%)

16,435 (85.3%)
2,823 (14.7%)


18,147 (94.2%)
1,111 (5.77%)

15,800 (82.0%)
3,458 (18.0%)

12,593 (65.4%)
6,665 (34.6%)


n Dependent variable:
Adequate plus PNC
utilization


19,258


19,258


19,258


19,258


19,258


19,258


19,258


19,258



19,258


19,258


9,553 (57.0%)
7,195 (43%)


11,641 (69.5%)
5,107 (30.5%)



14,739 (88.0%)
2,009 (12.0%)

13,405 (80.0%)
3,343 (20.0%)

11,378 (67.9%)
5,370 (32.1%)

13,027 (77.8%)
3,721 (22.2%)

12,944 (77.3%)
3,804 (22.7%)


3,949 (83.3%)
2,799 (16.7%)

13,543 (80.9%)
3,205 (19.1%)

10,298 (61.5%)
6,450 (38.5%)


n N P-value




16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


16,748 48,103 <0.0001*



16,748 48,103 <0.0001*


16,748 48,103 <0.0001*


The dependent variable for this table was prenatal care (PNC) utilization, while the main independent variables were pre-pregnancy
body mass index (BMI) and the control variables. The population for this table included all pregnancies and the years of PRAMS data
collection were for 2004 & 2005.









Table 4-7. Primary baseline logistic regression with the main effect independent variables
Dependent variable: Postpartum Odds ratio P-value 95% Confidence
(PPD) depression symptoms interval
(lower, upper)
Main effect independent variable:
Pre-pregnancy BMI
Underweight 1.047 0.54 (0.905, 1.212)
Overweight 1.05 0.52 (0.903, 1.224)
Obese 1.15 0.02* (1.023, 1.302)
Main effect independent variable:
PNC utilization
Inadequate 1.84 <0.0001* (1.589, 2.142)
Intermediate 1.19 0.02* (1.025, 1.380)
Adequate plus 1.28 <0.0001* (1.141, 1.438)
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI) and prenatal care (PNC)
utilization. The population for this table included all pregnancies and the years of PRAMS data
collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval (CI).

Table 4-8. Specific aim 1: Primary risk-adjusted logistic regression with the main effect


independent variables and control variables
Dependent variable: Postpartum depression Odds
(PPD) symptoms ratio


Main effect independent variable: Pre-pregnancy
BMI
Normal (reference)
Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Very low income: Less than $10,000


1.00
0.87
0.94
0.91


1.00
1.10
1.08
1.08

1.00
1.36
1.51
0.997
0.91
0.99
1.00
2.14


P-value 95% Confidence
interval
(lower, upper)


0.08**
0.46
0.15



0.26
0.33
0.23


<0.0001*
<0.0001*
0.97
0.003*
0.14

<0.0001


----------------
(0.735, 1.018)
(0.798, 1.107)
(0.792, 1.036)


----------------
(0.931, 1.309)
(0.923, 1.273)
(0.952, 1.232)

----------------
(1.159, 1.592)
(1.305, 1.741)
(0.848, 1.172)
(0.851, 0.969)
(0.981, 1.003)

(1.689, 2.712)
(1.689, 2.712)









Table 4-8. Continued
Dependent variable: Postpartum depression
(PPD) symptoms

Low income: $10,000-$24,999
Moderate income: $25,000-$49,999
Marital status
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children during pregnancy
High-risk maternal morbidity control variables
Diabetes before pregnancy
Cervix sewn shut (incompetent)
Preterm labor
Placenta previa or placenta abruptio
Bedrest during pregnancy
Medical risk factors
Hospitalized during pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
High blood pressure
Premature rupture of membrane (PROM)


Odds
ratio

1.68
1.51
1.01

1.00
0.90
1.09
0.89
0.73

1.05
0.79
0.91
1.06
1.27
0.82
0.93
1.27
1.05

1.36
0.95
1.44
1.04
1.16
1.12
1.06
1.18
1.27
1.46
0.95
0.69


P-value 95% Confidence
interval


<0.0001*
<0.0001*
0.86


0.18
0.30
0.56
0.06**

0.45
0.004*
0.11
0.28
0.006*
0.001*
0.31
0.02*
0.47

0.12
0.80
<0.0001*
0.76
0.047*
0.07**
0.50
0.08**
0.001*
<0.0001*
0.54
0.001*


(lower, upper)
(1.366, 2.065)
(1.269, 1.785)
(0.882, 1.163)

----------------
(0.764, 1.051)
(0.926, 1.280)
(0.592, 1.325)
(0.518, 1.016)

(0.927, 1.184)
(0.676, 0.930)
(0.805, 1.021)
(0.954, 1.177)
(1.073, 1.504)
(0.722, 0.919)
(0.818, 1.065)
(1.033, 1.569)
(0.912, 1.220)

(0.928, 1.980)
(0.629, 1.427)
(1.262, 1.641)
(0.826, 1.297)
(1.002, 1.333)
(0.992, 1.247)
(0.900, 1.241)
(0.979, 1.424)
(1.100, 1.455)
(1.282, 1.656)
(0.814, 1.114)
(0.549, 0.862)


The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI) and prenatal care (PNC)
utilization. The population for this table included all pregnancies and the years of PRAMS data
collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a
double asterisk corresponds to a 90% confidence interval (CI).










Table 4-9. Specific aim 2: Primary risk-adjusted logistic regression with the main effect
independent variables, interaction effect variables, and control variables


Dependent variable: Postpartum depression
(PPD) symptoms

Main effect independent variable: Pre-pregnancy
BMI
Normal (reference)
Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate plus PNC
Overweight BMI/Adequate PNC (reference)
Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate plus PNC
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Very low income: Less than $10,000
Low income: $10,000-$24,999
Moderate income: $25,000-$49,999
Marital status
Insurance control variables
PNC paid by income (reference)


Odds P-value 95% Confidence


ratio


1.00
0.84
0.86
0.92


1.00
1.11
1.05
1.06


1.00
0.71
1.20
0.99
1.00
1.40
0.95
1.18
1.00
1.13
0.98
1.03

1.00
1.36
1.50
0.997
0.91
0.99
1.00
2.16
1.69
1.50
1.01

1.00


0.18
0.25
0.43



0.40
0.67
0.55



0.11
0.36
0.97

0.19
0.83
0.42

0.63
0.92
0.86


<0.0001*
<0.0001*
0.97
0.003*
0.17

<0.0001*
<0.0001*
<0.0001*
0.86


interval
(lower, upper)


----------------
(0.656, 1.081)
(0.656, 1.116)
(0.746, 1.133)


(0.874, 1.400)
(0.836, 1.324)
(0.882, 1.266)


(0.459, 1.083)
(0.813, 1.770)
(0.735, 1.343)
----------------
(0.848, 2.318)
(0.588, 1.528)
(0.793, 1.743)
----------------
(0.697, 1.822)
(0.587, 1.620)
(0.703, 1.522)

----------------
(1.159, 1.591)
(1.303, 1.737)
(0.848, 1.162)
(0.850, 0.968)
(0.981, 1.003)
----------------
(1.702, 2.736)
(1.371, 2.073)
(1.267, 1.784)
(0.882, 1.163)









Table 4-9. Continued
Dependent variable: Postpartum depression
(PPD) symptoms

PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children during pregnancy
High-risk maternal morbidity control variables
Diabetes before pregnancy
Cervix sewn shut (incompetent)
Preterm labor
Placenta previa or placenta abruptio
Bedrest during pregnancy
Medical risk factors
Hospitalized during pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
High blood pressure
Premature rupture of membrane (PROM)


Odds
ratio

0.90
1.09
0.89
0.73

1.05
0.79
0.91
1.06
1.28
0.82
0.93
1.27
1.05

1.35
0.95
1.44
1.03
1.16
1.11
1.05
1.18
1.27
1.46
0.96
0.69


P-value 95% Confidence
interval


0.18
0.29
0.58
0.07**

0.44
0.004*
0.11
0.28
0.004*
0.001*
0.31
0.03*
0.49

0.12
0.81
<0.0001*
0.80
0.04*
0.07**
0.55
0.08**
0.001*
<0.0001*
0.61
0.001*


(lower, upper)
(0.764, 1.051)
(0.928, 1.283)
(0.596, 1.335)
(0.521, 1.022)

(0.929, 1.186)
(0.677, 0.930)
(0.806, 1.022)
(0.954, 1.177)
(1.080, 1.514)
(0.724, 0.922)
(0.818, 1.065)
(1.028, 1.564)
(0.910, 1.218)

(0.929, 1.975)
(0.631, 1.432)
(1.261, 1.640)
(0.822, 1.291)
(1.006, 1.337)
(0.992, 1.247)
(0.894, 1.235)
(0.981, 1.433)
(1.103, 1.459)
(1.282, 1.655)
(0.817, 1.118)
(0.553, 0.869)


The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, and the pre-pregnancy BMI/PNC utilization interaction effect variables. The
population for this table included all pregnancies and the years of PRAMS data collection were
for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a double asterisk
corresponds to a 90% confidence interval (CI).









Table 4-10. Wald tests for pre-pregnancy BMI/PNC interaction terms: Primary risk-adjusted
logistic regression
Interaction effect variables P-value
Interaction effect variables tested equal to each other within a pre-pregnancy
BMI group
Obese BMI/Inadequate PNC Obese BMI/Intermediate PNC = 0 0.11
Obese BMI/Inadequate PNC Obese BMI/Adequate plus PNC = 0
Overweight BMI/Inadequate PNC Overweight BMI/Intermediate PNC = 0 0.42
Overweight BMI/Inadequate PNC Overweight BMI/Adequate plus PNC = 0
Underweight BMI/Inadequate PNC Underweight BMI/Intermediate PNC = 0 0.89
Underweight BMI/Inadequate PNC Underweight BMI/Adequate plus PNC = 0
Each interaction effect variable tested equal to 0
Obese BMI/Inadequate PNC = 0 0.21
Obese BMI/Intermediate PNC = 0
Obese BMI/Adequate plus PNC = 0
Overweight BMI/Inadequate PNC = 0 0.48
Overweight BMI/Intermediate PNC = 0
Overweight BMI/Adequate plus PNC = 0
Underweight BMI/Inadequate PNC = 0 0.96
Underweight BMI/Intermediate PNC = 0
Underweight BMI/Adequate plus PNC = 0
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were the pre-pregnancy BMI/PNC utilization interaction effect
variables. The population for this table included all pregnancies and the years of PRAMS data
collection were for 2004 & 2005.

Table 4-11. Secondary baseline logistic regression (healthy pregnancies only) with the main
effect independent variables
Dependent variable: Postpartum Odds ratio P-value 95% Confidence
depression (PPD) symptoms interval
(lower, upper)
Main effect independent variable:
Pre-pregnancy BMI
Underweight 0.97 0.82 (0.744, 1.263)
Overweight 1.13 0.36 (0.870, 1.475)
Obese 1.12 0.30 (0.900, 1.403)
Main effect independent variable:
PNC utilization
Inadequate 1.97 <0.0001* (1.529, 2.541)
Intermediate 1.05 0.68 (0.827, 1.336)
Adequate plus 1.11 0.37 (0.885, 1.388)
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI) and prenatal care (PNC)
utilization. The population for this table included healthy pregnancies only, and the years of
PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence
interval and a double asterisk corresponds to a 90% confidence interval (CI).









Table 4-12. Specific aim 1: Secondary risk-adjusted logistic regression (healthy pregnancies
only) with the main effect independent variables, interaction effect variables, and
control variables


Dependent variable: Postpartum depression
(PPD) symptoms

Main effect independent variable: Pre-
pregnancy BMI
Normal (reference)
Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Very low income: Less than $10,000
Low income: $10,000-$24,999
Moderate income: $25,000-$49,999
Marital status
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS
Pregnancy and delivery control variables
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children during
pregnancy


Odds ratio


1.00
0.89
1.00
0.94


1.00
1.17
0.94
1.03

1.00
1.32
1.81
1.08
0.94
0.99
1.00
2.29
1.52
1.36
1.01

1.00
0.99
1.34
1.15
1.33

0.82
0.86
1.16
1.22
0.86
1.09
1.14
0.94


P-value 95% Confidence
interval
(lower, upper)


0.44
0.98
0.62



0.30
0.67
0.83


0.07**
<0.0001*
0.58
0.30
0.23

<0.0001*
0.02*
0.03*
0.91


0.93
0.046*
0.66
0.51

0.23
0.19
0.12
0.30
0.15
0.50
0.47
0.62


(0.672, 1.187)
(0.746, 1.329)
(0.737, 1.199)


(0.872, 1.562)
(0.729, 1.223)
(0.805, 1.312)

----------------
(0.977, 1.778)
(1.426, 2.300)
(0.820, 1.424)
(0.843, 1.055)
(0.965, 1.009)
----------------
(1.523, 3.445)
(1.080, 2.147)
(1.031, 1.802)
(0.780, 1.287)

----------------
(0.749, 1.302)
(1.005, 1.774)
(0.617, 2.142)
(0.575, 3.066)

(0.591, 1.135)
(0.693, 1.075)
(0.963, 1.386)
(0.839, 1.764)
(0.698, 1.056)
(0.850, 1.397)
(0.804, 1.610)
(0.730, 1.205)










Table 4-12. Continued
Dependent variable: Postpartum depression
(PPD) symptoms


Odds ratio P-value


95% Confidence
interval
(lower, upper)


Non high-risk maternal morbidity control
variables
Gestational diabetes 1.49 0.09** (0.940, 2.376)
Vaginal bleeding 0.80 0.15 (0.541, 1.096)
Kidney/bladder infection 1.63 <0.0001* (1.267, 2.092)
High blood pressure 0.83 0.43 (0.528, 1.314)
Nausea 1.57 <0.0001* (1.259, 1.948)
Premature rupture of membrane (PROM) 1.58 <0.0001* (1.259, 1.948)
Labor abnormalities 1.32 0.01* (1.057, 1.640)
Labor/delivery complications 0.94 0.60 (0.756, 1.177)
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI) and prenatal care (PNC)
utilization. The population for this table included healthy pregnancies only, and the years of
PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence
interval and a double asterisk corresponds to a 90% confidence interval (CI).

Table 4-13. Specific aim 2: Secondary risk-adjusted logistic regression (healthy pregnancies
only) with the main effect independent variables, interaction effect variables, and


control variables
Dependent variable: Postpartum depression
(PPD) symptoms

Main effect independent variable: Pre-pregnancy
BMI
Normal (reference)
Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate Plus PNC
Overweight BMI/Adequate PNC (reference)


Odds
ratio



1.00
0.90
0.81
1.15


1.00
1.25
0.98
1.03


1.00
0.51
0.87
0.61
1.00


P-value 95% Confidence
interval
(lower, upper)


0.61
0.33
0.42



0.26
0.92
0.85



0.08**
0.67
0.11


----------------
(0.606, 1.343)
(0.531, 1.236)
(0.819, 1.612)


----------------
(0.847, 1.836)
(0.692, 1.394)
(0.743, 1.435)


----------------
(0.243, 1.090)
(0.456, 1.654)
(0.331, 1.126)
----------------









Table 4-13. Continued
Dependent variable: Postpartum depression
(PPD) symptoms

Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate Plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate Plus PNC
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Very low income: Less than $10,000
Low income: $10,000-$24,999
Moderate income: $25,000-$49,999
Marital status
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS
Pregnancy and delivery control variables
Smoking during pregnancy
Vaginal delivery
Gender of infant
Intensive care unit
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children during pregnancy
Non high-risk maternal morbidity control
variables
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
High blood pressure
Nausea
Premature rupture of membrane (PROM)


Odds
ratio

1.70
1.03
1.73
1.00
0.88
0.89
1.13

1.00
1.31
1.82
1.08
0.95
0.99
1.00
2.30
1.53
1.35
1.03

1.00
0.99
1.34
1.18
1.27

0.82
0.86
1.16
1.23
0.86
1.09
1.12
0.94


1.48
0.76
1.63
0.83
1.57
1.65


P-value 95% Confidence
interval


0.21
0.93
0.16

0.70
0.78
0.75


0.08**
<0.0001*
0.57
0.33
0.26

<0.0001*
0.02*
0.03*
0.82


0.94
0.04*
0.61
0.59

0.23
0.19
0.12
0.27
0.15
0.48
0.52
0.61


0.096**
0.14
<0.0001*
0.43
<0.0001*
0.49


(lower, upper)
(0.738, 3.902)
(0.465, 2.299)
(0.810, 3.698)
----------------
(0.373, 2.062)
(0.386, 2.046)
(0.538, 2.367)

----------------
(0.969, 1.770)
(1.430, 2.307)
(0.823, 1.430)
(0.846, 1.058)
(0.966, 1.009)
----------------
(1.527, 3.451)
(1.082, 2.153)
(1.023, 1.789)
(0.810, 1.303)

----------------
(0.750, 1.304)
(1.010, 1.785)
(0.633, 2.184)
(0.535, 2.994)

(0.594, 1.134)
(0.693, 1.076)
(0.965, 1.389)
(0.849, 1.780)
(0.700, 1.058)
(0.852, 1.403)
(0.790, 1.591)
(0.730, 1.203)


(0.933, 2.355)
(0.536, 1.088)
(1.266, 2.090)
(0.525, 1.315)
(1.261, 1.952)
(0.393, 6.955)









Table 4-13. Continued
Dependent variable: Postpartum depression Odds P-value 95% Confidence
(PPD) symptoms ratio interval
(lower, upper)
Labor abnormalities 1.31 0.02* (0.052, 0.491)
Labor/delivery complications 0.93 0.53 (0.746, 1.163)
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, and the pre-pregnancy BMI/PNC utilization interaction effect variables. The
population for this table included healthy pregnancies only, and the years of PRAMS data
collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a
double asterisk corresponds to a 90% confidence interval (CI)

Table 4-14. Wald tests for pre-pregnancy BMI/PNC interaction terms: Secondary risk-adjusted
logistic regression (healthy pregnancies only)
Interaction effect variables P-value
Interaction effect variables tested equal to each other within a pre-pregnancy
BMI group
Obese BMI/Inadequate PNC Obese BMI/Intermediate PNC = 0 0.46
Obese BMI/Inadequate PNC Obese BMI/Adequate plus PNC = 0
Overweight BMI/Inadequate PNC Overweight BMI/Intermediate PNC = 0 0.47
Overweight BMI/Inadequate PNC Overweight BMI/Adequate plus PNC = 0
Underweight BMI/Inadequate PNC Underweight BMI/Intermediate PNC = 0 0.77
Underweight BMI/Inadequate PNC Underweight BMI/Adequate plus PNC = 0
Each interaction effect variable tested equal to 0
Obese BMI/Inadequate PNC = 0 0.22
Obese BMI/Intermediate PNC = 0
Obese BMI/Adequate plus PNC = 0
Overweight BMI/Inadequate PNC = 0 0.37
Overweight BMI/Intermediate PNC = 0
Overweight BMI/Adequate plus PNC = 0
Underweight BMI/Inadequate PNC = 0 0.92
Underweight BMI/Intermediate PNC = 0
Underweight BMI/Adequate plus PNC = 0
The dependent variable for this table was postpartum depression (PPD) symptoms, while the
main independent variables were the pre-pregnancy BMI/PNC utilization interaction effect
variables. The population for this table included healthy pregnancies only, and the years of
PRAMS data collection were for 2004 & 2005.











1.6
1.4 -
1.2


0.8
0.6
0.4
0.2

.- j "- cu' .. 0 "






Figure 4-1. Primary risk-adjusted logistic regression with postpartum depression (PPD) symptom odds ratios for each interaction





108
L) 07 *05 o0 C) LM 07 U (D U) U Uu
0) (D (D (D Z ( D 7
D E U 0a- D E 0 3- ( 3 ( E (D3 Z







the horizontal axis).

















108









CHAPTER 5
DISCUSSION

This study sought to determine the role that PNC utilization plays in the relationship

between pre-pregnancy BMI and postpartum depression (PPD) symptoms among a sample of

51,600 women in the United States. These women represented 16 states across the nation:

Alaska, Colorado, Georgia, Hawaii, Illinois, Maine, Minnesota, North Carolina, Nebraska, New

Mexico, Oregon, Rhode Island, South Carolina, Utah, Vermont, and Washington. Even though a

consistent moderating effect of PNC was not seen throughout the association between pre-

pregnancy BMI and PPD symptoms, patterns that warrant attention and provide insight into this

model with pre-pregnancy BMI, PNC utilization, and PPD symptoms were seen among the

univariate analyses, the bivariate analyses, and the multivariate analyses

Univariate Analyses

The frequencies reported in the univariate analyses showed that there are characteristics of

the sample that call for discussion. With regards to the frequencies of the dependent variable,

PPD symptoms, removing the high-risk pregnancies from the analysis did not change the

prevalence drastically. In comparing the frequencies of PPD symptoms with previous literature,

the prevalence of postpartum "blues" was a little higher than what has been reported in previous

literature, approximately 50-80%, while the prevalence of PPD symptoms was within the

average range for PPD symptoms as reported in previous literature for non-psychotic PPD, or

approximately 10-15% (Miller, 2002; Evins & Theofrastous, 1997; Negus Jolley & Betrus,

2007). However, in comparing PPD frequencies between this study and previous studies, it

should be noted that the PPD measure used for this study was self-reported PPD symptoms

whereas many of the frequencies reported in previous literature reflect those obtained from

screening instruments that were used to diagnose PPD (not self-reported). However, given that









the measure was self-reported, the methodology used to categorize observations into PPD

symptoms versus "no" PPD symptoms mirrored an instrument that has demonstrated sensitivity

and specificity, criterion and construct validities for diagnosing major depressive disorder

(Kroenke et al., 2003). Also, with regard to these PPD frequencies, it is important to note that the

method of categorizing PPD symptoms (e.g., "blues" versus actual depressive symptoms) for the

primary risk-adjusted logistic regression model and the secondary risk-adjusted logistic

regression model that removed high-risk pregnancies may have included both women who self-

reported non-psychotic PPD symptoms and women who reported PPD psychosis symptoms.

Scores of 3 or greater in the PHQ-2 are not meant to diagnose the severity of depression, but are

rather used to screen for depression (Kroenke et al., 2003). Therefore, the prevalence for non-

psychotic PPD symptoms in this study may have been less than 15.4% and 11.5% (as reported in

this study) for the risk-adjusted primary and secondary logistic regression models, respectively.

Frequencies for the main effect independent variables showed some results that call for

discussion. Though frequencies for pre-pregnancy BMI showed that over half of the women in

the sample were classified as having had a normal pre-pregnancy BMI (the high number of

women in this group was expected), the next highest percentage in the sample was for women

who had an obese pre-pregnancy BMI (n =10,270). Some groups of women, who comprise high-

risk populations, are sampled at a higher rate, via stratification variables in PRAMS, so that a

sufficient quantity of data are available (CDC, 2007). Since obesity is a risk-factor that has been

associated with outcomes such as complications during pregnancy (e.g., preeclampsia,

respiratory problems, etc.) (LaCoursiere et al., 2005; Saravanakumar et al., 2006; Cedergren,

2004), in this study, obesity was considered to be a general risk factor for PPD symptoms.

However, since the stratification variables in PRAMS for the 2004 and 2005 years of data (see p.









53 for a list of the stratification variables) did not include pre-pregnancy BMI, this group was not

oversampled. If response rates are higher than normal for groups of women in PRAMS, they are

automatically adjusted for by the analysis weight that was incorporated into the PRAMS dataset

and used in the analyses for this study. However, in comparing these percentages with two

previous studies that used PRAMS to look at pre-pregnancy BMI, Kim, Dietz, England, Morrow,

& Callaghan (2007), who looked at 2002-2003 PRAMS data, showed the following ranges for

pre-pregnancy BMI among 9 states: underweight (13%-16%), normal (45%-54%), overweight

(11%-14%), and obese (18%-26%), while D'Angelo et al. (2007) showed the following ranges

for pre-pregnancy BMI among 26 states using 2004 PRAMS data: underweight (10%-17%),

normal (reference group), overweight (11%-15%), and obese (15%-26%). Thus, even though the

PRAMS dataset used for this study was for 2004 and 2005, the percentages of pre-pregnancy

BMI in this study were within average ranges compared to previous studies. However, the

percentages of this study differ (in descending order) from the results demonstrated by

LaCoursiere et al. (2006), who also sampled women through the PRAMS 2000-2001 years of

data (in the state of Utah only). Results from their study showed that the highest percentage of

women had a normal pre-pregnancy BMI (who represented about half of their sample), women

who had an underweight pre-pregnancy BMI were the second largest in percentage (about 20%

of the sample), women who had an obese pre-pregnancy BMI comprised the next group at 16%,

and then women who had an overweight pre-pregnancy BMI was the lowest percentage in the

sample (about 11%).

For PNC utilization, it was expected that the highest percentage would be comprised of

women who utilized an "adequate" quantity of PNC. However, the next highest percentage,

which included "high-risk" pregnancies, was comprised of women who utilized an "adequate









plus" quantity of PNC. Even though adequacy of PNC was previously a stratification variable in

PRAMS (e.g., 1990 and 1991 PRAMS data, 1995 PRAMS data) (Goodwin et al., 2006, Centers

for Disease Control, 1995), it was not one of the stratification variables in the 2004 and 2005

years of data from which oversampling occurred. Thus, as mentioned for women who had an

obese pre-pregnancy BMI, the high response rate for women in this PNC utilization group was

adjusted for by the analysis weight that was incorporated into the PRAMS dataset and used in the

analyses for this study. Comparing these percentages with trends in PNC over time, according to

Kogan et al. (1998), who used national birth records, the percentage of utilization (according to

the APNCU Index) occurred as follows: there was a decrease in inadequate PNC utilization from

12% in 1981 to 8.9% in 1995, there was a decrease in intermediate PNC utilization from 23.2%

in 1981 to 17.2% in 1995, there was a slight decrease in adequate PNC utilization from 45.1% in

1981 to 43.9% in 1995, and there was a significant increase in adequate plus PNC (intensive)

from 18.4% in 1981 to 28.8% in 1995. Thus, given the 10-year gap between Kogan et al. (1998)

and the use of the APNCU Index, calculated from the PNC information provided on the birth

certificates and linked with the PRAMS data, it seems that there were slight decreases in

intermediate and adequate PNC utilization, as slight increases in inadequate and adequate plus

PNC utilization with the data used for this study.

To further explain characteristics that may, in part, have contributed to the large number of

adequate plus PNC observations, a variety of maternal morbidities that affected women in the

sample were investigated, many of which provide insight into the risk-status of a large portion of

the women in the sample. The univariate results for the maternal morbidities showed that there

were a variety of potential health risks (maternal morbidities) that may have prompted the

delivery of adequate plus PNC, as it appears that there were a number of women in the sample









who possessed high-risk characteristics. One thing that is unclear, however, is whether women

who answered "yes" to having medical risk factors during pregnancy in the PRAMS

questionnaire had risk factors that actually warranted utilization of adequate plus PNC.

Bivariate Analyses

Given that there was a high number of women (as shown in the univariate results) who had

a normal pre-pregnancy BMI, this may explain that the highest percentage of women who

experienced PPD symptoms was also women from this BMI category; however, the result

showing that one-fourth of women with PPD symptoms had an obese pre-pregnancy BMI, with

this result being statistically significant (p<0.0001*) provides evidence for a significant

difference in percentages for PPD symptoms across all pre-pregnancy BMI categories. The chi-

square results presented for pre-pregnancy BMI and PPD symptoms mostly countered the results

for pre-pregnancy BMI and PPD symptoms by LaCoursiere et al. (2006). This study showed a

significant difference in percentages for PPD symptoms across pre-pregnancy BMI categories in

the following order: women who had a normal pre-pregnancy BMI, who had the highest

percentage of PPD symptoms, followed by women who had an obese pre-pregnancy BMI, then

followed by women who had an underweight pre-pregnancy BMI, and finally, women who had

an overweight pre-pregnancy BMI, who had the lowest percentage of PPD symptoms.

LaCoursiere et al. (2006) showed the highest percentage of PPD symptoms among women who

had an obese pre-pregnancy BMI, followed by women who had an underweight pre-pregnancy

BMI, then followed by women who had an overweight pre-pregnancy BMI, and finally, women

who had a normal pre-pregnancy BMI, who had the lowest percentage of PPD symptoms. The

results for women without PPD symptoms for this study matched the results for women with

PPD symptoms (in order), but were shown as different by LaCoursiere et al. (2006) as highest

for women who had a normal pre-pregnancy BMI, followed by women who had an underweight









pre-pregnancy BMI, then followed by women who had an obese pre-pregnancy BMI, and finally,

women who had an overweight pre-pregnancy BMI, who had the lowest percentage. So

surprisingly, and contrary to what was hypothesized for this study, these results demonstrated

that women who had an overweight pre-pregnancy BMI reported the lowest percentage of self-

reported PPD symptoms, whereas LaCoursiere et al. (2006) demonstrated that women who had a

normal pre-pregnancy BMI reported the lowest percentage of self-reported "moderate" or

"greater" PPD symptoms.

It was not surprising that there was a significant difference in percentages for PPD

symptoms across the four PNC utilization groups. It is worthy to note that the highest percentage

for PPD symptoms was among women who utilized adequate plus PNC. Though the outcome

was a birth outcome and not a postpartum outcome, Kotelchuck (1994) revealed a "U-shaped"

association between PNC utilization and low birthweight rates, with women who utilized

inadequate PNC and adequate plus PNC having the highest rates of low birthweight babies.

Though the adequate plus groups were on the highest end in both this study and Kotelchuck

(1994), on the low end, inadequate was higher than intermediate in this study, but not as high so

as to form a "U-shape" as was the case for Kotelchuck (1994); rather, the shape in this study was

a "J-shape."

Considering the variables in which there were no significant differences in percentages for

PPD symptoms, it was surprising that "gender of the infant" did not demonstrate a significant

difference. This is considering that women receive pressures, especially in Asian cultures, to give

birth to a boy, and, giving birth to a girl, when a boy is expected, has been found to increase the

risk for PPD with this risk increasing as the number of female children increase (Chan, Levy,

Chung, & Lee, 2002, Dindar & Erdogan, 2007; Patel, Rodrigues, & DeSouza, 2002; Rahman,









Iqbal, & Harrington, 2003). Given that the women in the sample comprised a variety of

races/ethnicities, it was expected that "gender of the infant" would demonstrate a significant

difference in percentages for PPD symptoms between women who gave birth to a female versus

a male infant.

With regards to the chi-square analyses on pre-pregnancy BMI, the lack of a significance

difference in percentages for weight gain discussion during PNC across the four pre-pregnancy

BMI groups was surprising as significant difference was expected for this variable. This is

considering that 1) obesity has been shown in a number of studies as a risk-factor for

complications during pregnancy and/or adverse pregnancy outcomes LaCoursiere et al., 2005;

Saravanakumar et al., 2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005;

Mahmood, 2009; Baeten et al., 2001; Cnattingius et al., 1998), 2) it has been found that women

who were overweight or obese prior to pregnancy were more likely to experience excessive

weight gain during pregnancy (Lederman et al., 2002; Olafsdottir et al., 2006), and 3) women

who have an underweight pre-pregnancy BMI have a higher odds for delivering a preterm infant

(Siega-Riz, Adair, & Hobel, 1996). However, it is recommended that 1) clinicians support and

encourage women during PNC delivery in gaining the appropriate amount of weight during

pregnancy (Lederman, 2001), and 2) young obese women who are planning a pregnancy be

cautioned of the possible complications during pregnancy and/or at birth (Dietl, 2005). Further

considering that obesity has been previously shown as a risk-factor for complications during

pregnancy and/or adverse pregnancy outcomes LaCoursiere et al., 2005; Saravanakumar et al.,

2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005; Mahmood, 2009; Baeten

et al., 2001; Cnattingius et al., 1998), it was expected that labor/delivery complications would be

significantly different in percentages across the four pre-pregnancy BMI groups. However, in









comparing this result with results from the multivariate analyses, significance was not

demonstrated for labor/delivery complications as a high-risk variable (p. 167) with the logit

model holding adequate plus PNC as the dependent variable, and for the secondary logistic

regression models (that removed the high-risk pregnancies form the analyses), this variable was

not significant for either specific aim 1 or specific aim 2. Thus, consistency remained with the

results of this variable.

Though this may or may not have been related to PPD symptoms in this study, the chi-

square analyses addressing pre-pregnancy BMI and the maternal morbidities (Table 4-5) suggest

that the majority of these morbidities affected women who had an obese pre-pregnancy BMI the

most (compared to the three other pre-pregnancy BMI groups). Interestingly enough, the

bivariate results showed that the highest percentage of preterm labor was among women who had

an underweight pre-pregnancy BMI, with these women also having the highest percentage of low

birth weight babies (between 1,500 to 2,499 grams). This may be expected considering a

consequence of preterm birth for the infant may be low birth weight. Surprisingly, women who

had an obese pre-pregnancy BMI had the highest percentage of "very low birth weight" babies

(less than 1,500 grams), compared to women from the three other pre-pregnancy BMI groups.

This result is surprising considering that previous studies have found higher/excessive pre-

pregnancy weight to be associated with giving birth to a macrosomic infant (Baeten et al., 2001;

Rosenberg, Garbers, Chavkin, & Chiasson, 2003; Cedergren, 2004). Though it is unclear in this

sample whether giving birth to a very low birth weight baby caused enough postpartum distress

(e.g., depression) for a woman that was at a non-psychotic severity, previous studies have

suggested an association between low birth weight and psychological distress that may call for

emotional support (Singer et al., 1999; Kersting et al., 2004).









Further addressing maternal morbidities, depending on the number of morbidities a woman

had, as well as the duration and severity of those morbidities, future research should seek to

determine the impact of these factors, after stratifying by pre-pregnancy BMI on postpartum

distress, if not on the severity of PPD. Though there are many studies to support an association

between pregnancy and/or delivery complications and PPD (Josefsson et al., 2002; Leidner,

Singer, Sicherman, Francoise, & Divon, 2008; Adewuya, Fatoye, Ola, Ijaodola, & Ibigbami,

2005; Campbell & Cohen, 1991), and studies that refute the association between pregnancy

and/or delivery complications and PPD (Nielsen, Videbech, Hedegaard, Dalby, & Secher, 2000;

Johnstone, Boyce, Hickey, Morris-Yates, & Harris, 2001), it is recommended that the study of

the pathophysiology of PPD include, in-part, pregnancy complications such as bedrest,

gestational diabetes, and preeclampsia (Stowe & Nemeroff, 1995).

Regarding the chi-square analyses addressing PNC utilization, it was not surprising that

there was no significant difference in percentages for "gender of the infant" across the four PNC

utilization groups, considering that there are no previous studies that have confirmed an

association between these two variables.

With regards to the chi-square analyses that compared women who utilized adequate plus

PNC versus women who utilized "other quantities of PNC," the results showed a significant

difference in percentages across the four pre-pregnancy BMI. Women who had an obese pre-

pregnancy BMI were the only group of women with a higher percentage in the group that

utilized adequate plus PNC, compared to the group that utilized "other quantities of PNC." This

result is worthy to note considering that 1) there were a large number of women who utilized

adequate plus PNC, and 2) the primary population of focus in this study was women who had an

obese pre-pregnancy BMI.









Multivariate Analyses: Primary Risk-Adj usted Logistic Regression Analysis

Baseline Model

Baseline results for the main model were somewhat consistent with conclusions made by

Carter et al. (2000), and LaCoursiere et al. (2006) in that higher pre-pregnancy BMI is associated

with a higher likelihood for PPD symptoms; a linear trend was seen in this study in that higher

pre-pregnancy BMI increased odds for PPD symptoms. However, as for significance, it was only

women in the highest pre-pregnancy BMI category, obese, which had significantly greater odds

of having PPD symptoms relative to women who had a normal pre-pregnancy BMI in these

models; this result matched what was predicted for women who had an obese pre-pregnancy

BMI.

Looking into the association between PNC and PPD symptoms, the baseline model

demonstrated that the highest likelihood for PPD symptoms among levels of PNC utilization was

for inadequate PNC, followed by adequate plus PNC, and intermediate PNC, relative to adequate

PNC. However, after including control variables and risk-adjusting for high-risk pregnancies that

received adequate plus PNC due to a medical necessity, the statistical significance disappeared

for all PNC utilization levels, demonstrating that there is no association between PNC utilization

and PPD symptoms. These results were contrary to the one previous study that has looked at

quantity of PNC and PPD: El-Kak et al. (2004). The authors demonstrated a linear relationship

between PNC and PPD for high-risk women, where a higher number of PNC visits were

associated with fewer cases of PPD. However, even though El-Kak et al. (2004) controlled for a

variety of characteristics (e.g., parity, education, area of residence, employment during

pregnancy) as this study also controlled for a variety of characteristics, this study varied in the

measures used for PNC and PPD (compared to El-Kak et al., 2004). For example, this study used

the APNCU Index (Kotelchuck, 1994) to categorize PNC utilization into inadequate,









intermediate, adequate, and adequate plus, by considering the frequency of PNC visits,

gestational age, and the timing of PNC initiation. El-Kak et al. (2004) analyzed each PNC

measure separately: the initiation of the first PNC visit was categorized by trimester (1st, 2nd, 3rd),

the frequency of PNC visits was categorized into three categories (1-4, 5-9 and 10 + visits), and

gestational age was categorized into pretermm" or "term." With regards to pregnancy risk-status,

they stratified risk status into two categories (low-risk versus high-risk), whereas this study used

two different approaches to risk-adjustment: 1) including high-risk characteristics as control

variables, and 2) removing the high-risk pregnancies from the analyses. This study used a 5-tier

likert scale (from the PRAMS questionnaire) to categorize PPD symptoms into yes/no, whereas

El-Kak et al. (2004) categorized PPD into yes/no based on the occurrence of PPD symptoms for

the women in their sample. Thus, it is postulated that the differences seen in the results for both

studies can be explained by the difference in the measures used for both studies.

Regarding previous studies that have used PNC indices to examine adequacy of PNC with

birth outcomes, many studies have examined the effectiveness of PNC on low birth weight.

Though some studies have shown that adequacy of PNC is not associated with birth weight,

several studies, many of which have used indices as measures to represent PNC (Alexander &

Kotelchuck, 1996), have shown the benefits of PNC utilization on birth weight (Gortmaker,

1979; Showstack, Budetti, & Minkler, 1984; Quick, Greenlick, & Roghmann, 1981; Mustard &

Roos, 1994). Though the moderating effect of PNC utilization on a postpartum outcome, PPD

symptoms, was not demonstrated in this study, it is recommended that research look further into

the relationship between the effects of PNC on postpartum preventive behaviors (Alexander &

Kotelchuck, 2001), while perhaps focusing on women who are experiencing different severities

of PPD (e.g., postpartum blues versus non-psychotic PPD).









Specific Aim 1

"What is the association of pre-pregnancy body mass index (BMI) with subsequent

development of postpartum depression (PPD) symptoms?"

It was predicted in the hypothesis that women who had an obese pre-pregnancy BMI would

have the highest odds of PPD symptoms, followed by women who had an overweight BMI, and

finally women who had an underweight BMI, who would have the lowest odds for PPD

symptoms (compared to women who had a pre-pregnancy BMI of normal). The results for this

logistic regression showed borderline significance only among women who had an underweight

pre-pregnancy BMI, with this significance apparent at a 90% confidence interval only. Contrary

to the hypothesis, this group of women had lower odds for PPD symptoms compared to women

who had a normal pre-pregnancy BMI. Thus, the significant relationships demonstrated for obese

pre-pregnancy BMI, and all the PNC utilization groups, all of which displayed significant higher

odds for PPD symptoms in the baseline primary risk-adjusted logistic regression, disappeared

after adding all the control variables. A new significant relationship for women who had an

underweight pre-pregnancy BMI appeared in that they had lower odds for PPD symptoms

compared to women who had a normal pre-pregnancy BMI. This result was contrary to the initial

hypothesized that women who had an underweight pre-pregnancy BMI would have greater odds

for PPD symptoms. The significance that appeared for women who had an underweight pre-

pregnancy BMI only after adding control variables can be attributed to controlling for variables

in this model that were associated with underweight pre-pregnancy BMI and positively

associated with PPD symptoms. Hence, not controlling for those variables initially in the

baseline model did not allow for the borderline significant association between underweight pre-

pregnancy BMI and PPD symptoms to appear; after controlling for these variables, this

significance was revealed. Future research should further study the significant control variables









included in this model to determine which variables (e.g., race, education, income, etc.) are

positively associated with PPD symptoms and associated with the appearance of this borderline

significance for underweight pre-pregnancy BMI upon its inclusion in the model.

To explain the disappearance of the significance for obese pre-pregnancy BMI and all the

PNC utilization groups, it might be that one or more of the control variables included in this

model that was significantly associated with PPD symptoms was also confounding variable

(associated with obese pre-pregnancy BMI and the PNC utilization groups in the previous model

(but not controlled for). Thus, after adding the control variable(s) in the model, there was no

"true association" of either obese pre-pregnancy BMI or any of the PNC utilization groups with

PPD symptoms. Fifteen control variables added in this model were significantly associated with

PPD symptoms; it is hypothesized that one or more of these variables were responsible for (but

not controlled for) the initial significance seen for the four variables in the baseline model.

Future research should seek to determine the control variables that were responsible for the

significance seen for each of the four variables in the baseline model (as confounders)

Specific Aim 2

"Does PNC moderate the relationship between pre-pregnancy BMI and PPD symptoms?"

It was predicted in the hypothesis that within each pre-pregnancy BMI category, the

likelihood a woman will experience PPD symptoms would decrease as PNC increased: Women

who utilized inadequate PNC would have the highest odds for PPD symptoms, followed by

women who utilized adequate plus PNC, and finally women who utilized intermediate PNC, who

would have the lowest odds for PPD symptoms (compared to women who received adequate

PNC). The results showed that there were no significant relationships present among the main

effects (pre-pregnancy BMI and PNC utilization) or the pre-pregnancy BMI/PNC utilization

interaction variables (no moderating effect of PNC). Thus, the significance demonstrated for









underweight pre-pregnancy BMI disappeared after including the interaction variables in the

model. It appeared that interacting the quantity of PNC utilization with women who had an

underweight pre-pregnancy BMI removed the previous significant association seen with

underweight pre-pregnancy BMI and PPD symptoms. So, there was an overall effect seen for

underweight pre-pregnancy BMI previously that was the "average" effect of all the interaction

terms. However, after including the interaction terms, the specific effect of each interaction term

was subsequently removed from that "average" effect, leaving only the effect of underweight

pre-pregnancy BMI alone in the main effect variable; hence, although the direction of the effect

remained the same and contrary to what was initially hypothesized (women who had an

underweight pre-pregnancy BMI had lower odds instead for PPD symptoms compared to women

who had a normal pre-pregnancy BMI), underweight pre-pregnancy BMI was no longer

significant.

Multivariate Analyses: Secondary Risk-Adjusted Logistic Regression (Subpopulation With
Healthy Pregnancies)

Baseline Model

After adjusting for high-risk pregnancies in the logistic regression model, the significance

seen in the primary logistic regression model for women who had an obese pre-pregnancy BMI

disappeared. Even though a linear trend was seen in that increasing pre-pregnancy BMI

increased odds for PPD symptoms, none of these effects were significant; thus, refuting

conclusions made by Carter et al. (2000) and LaCoursiere et al. (2006) in that higher pre-

pregnancy BMI is associated with a higher likelihood for PPD symptoms. However, in

comparing the sample characteristics of this analysis versus those of the studies conducted by

Carter et al. (2000) and LaCoursiere et al. (2006) this study used two risk-adjustment

approaches: 1) controlling for the characteristics associated with high-risk pregnancies and 2)









removing the high-risk pregnancies from the analysis. It was not evident that Carter et al. (2000)

and LaCoursiere et al. (2006) risk-adjusted for the women in their samples via their exclusion

criteria or in the description of their samples. Thus, in this study, comparing the primary analysis

that included all pregnancies versus the secondary analysis that included only the healthy

pregnancies showed that significance is only demonstrated for women who had an obese pre-

pregnancy BMI when the high-risk pregnancies are also included. Since 1) obesity has been

shown in a number of studies as a risk-factor for complications during pregnancy and/or adverse

pregnancy outcomes (LaCoursiere et al., 2005; Saravanakumar et al., 2006; Cedergren, 2004;

Rosenberg et al., 2003; Rosenberg et al., 2005; Mahmood, 2009; Baeten et al., 2001; Cnattingius

et al., 1998), and 2) previous studies support an association between pregnancy and/or delivery

complications and PPD (Josefsson et al., 2002; Leidner et al., 2008; Adewuya et al., 2005;

Campbell & Cohen, 1991), it is hypothesized that it might be the high-risk pregnancies, among

women who had an obese pre-pregnancy BMI, that were responsible for the significant

association seen in the primary analysis between obese pre-pregnancy BMI and PPD symptoms.

The chi-square results for pre-pregnancy BMI suggest that many of the women who had an

obese pre-pregnancy BMI experienced high-risk morbidities. For example, among the nine high-

risk morbidities analyzed in the chi-square analyses, women who had an obese pre-pregnancy

BMI had the highest percentage for seven of the nine morbidities.

What is especially worth further noting is that the significance in the baseline model

disappeared for women who utilized adequate plus PNC after adding control variables in the

primary analyses (models for specific aims 1 and 2). These control variables included the high-

risk maternal morbidities. What was surprising was that unlike the baseline model for the

primary analysis, which included all pregnancies and did not control for high-risk characteristics,









the baseline model for the secondary analysis, which included healthy pregnancies only, did not

demonstrate significance for women who utilized adequate plus. Thus, after risk-adjusting by 1)

controlling for high-risk characteristics by including the maternal morbidities as appropriate

control variables (primary analyses), and 2) modifying the design of the model by removing the

high-risk pregnancies from the sample (secondary analyses), the significance for women who

utilized adequate plus ceased to appear as it had appeared in a model that did not include any

control variables and included women from all pregnancy risk-statuses. What may explain this

result is that the high-risk pregnancies were responsible for the significance that previously

appeared, and that among all the women in the sample who utilized adequate plus PNC, it is only

the women who are high-risk that are the women who are subsequently at higher odds for PPD

symptoms, compared to women who utilized adequate PNC that were not high-risk.

Specific Aim 1

"What is the association of pre-pregnancy body mass index (BMI) with subsequent

development of postpartum depression (PPD) symptoms?"

Similar to the risk-adjusted primary logistic regression model, it was also predicted in the

hypothesis for this model that women who had an obese pre-pregnancy BMI would have the

highest odds for PPD symptoms, followed by women who had an overweight BMI, and finally

women who had an underweight BMI, who would have the lowest odds for PPD symptoms

(compared to women who had a pre-pregnancy BMI of normal). Results showed there was no

significance among the pre-pregnancy BMI main effects. However, the main effects for PNC

utilization showed that significance among women who utilized inadequate PNC disappeared.

Similar to the reasoning provided for the disappearance of significance for obese pre-pregnancy

BMI, and the PNC utilization main effects between the primary baseline logistic regression and

the primary logistic regression that addressed the first specific aim, it might be that one or more









of the control variables included in this model that was significantly associated with PPD

symptoms was also a confounding variable in the previous model (but not controlled for). Thus,

after adding the control variable(s) in the model, there was no "true association" of inadequate

PNC utilization on PPD symptoms. Eight control variables added in this model were

significantly associated with PPD symptoms; it is hypothesized that one or more of these

variables were responsible for (but not controlled for) the initial significance seen in the baseline

model. Future research should seek to determine the control variables that were responsible for

the significance seen for the four variables in the secondary baseline logistic regression model

(as confounders).

Specific Aim 2

"Does PNC moderate the relationship between pre-pregnancy BMI and PPD symptoms?"

Similar to the risk-adjusted primary logistic regression model, it was also predicted in the

hypothesis for this model that within each pre-pregnancy BMI category, the likelihood a woman

will experience PPD symptoms would decrease as PNC increased. That is, women who utilized

inadequate PNC would have the highest odds for PPD symptoms, followed by women who

utilized adequate plus PNC, and finally women who utilized intermediate PNC, who would have

the lowest odds for PPD symptoms (compared to women who received adequate PNC).

However, after removing high-risk pregnancies from the sample, the results for this model

showed that there was no significance among any of the main effects (pre-pregnancy BMI or

PNC utilization). Surprisingly, significance appeared for an interaction variable, but in the

opposite direction to what was hypothesized: women who had an obese pre-pregnancy BMI and

utilized inadequate PNC had lower odds for PPD symptoms compared to women who had an

obese pre-pregnancy BMI and utilized adequate PNC. Thus, even though the results support that

a significant moderating effect for one interaction group only appeared after adding the









interaction terms, the effect was in an unexpected direction, opposing what was initially

hypothesized. One reason to explain why women who had an obese pre-pregnancy BMI and

utilized inadequate PNC had lower odds for PPD symptoms compared to women who had an

obese pre-pregnancy and utilized adequate PNC is that women in the former group were 1)

healthy, 2) content with their weight and their bodies, or 3) if they feel healthy, they may not see

the need to seek PNC (their self-perceived health status is excellent/healthy). Since this model

included healthy pregnancies only, it is possible that women in this group had a high self-

perception of their health status during pregnancy. Other reasons for utilizing inadequate PNC,

previously found in one study, include denial and/or concealment of pregnancy, or financial

reasons, as concluded by Friedman, Heneghan, & Rosenthal (2009).

The significance seen for inadequate PNC utilization disappeared in this model after

adding the interaction terms. Similar to the reasoning provided for the disappearance of

significance for women who had an underweight pre-pregnancy BMI after adding the interaction

variables to the model, the significance seen initially in the model with only the main effect

variables (pre-pregnancy BMI and PNC utilization) and the control variables was an overall,

"average" effect of all the pre-pregnancy BMI inadequate PNC groups. However, after including

the interaction terms between each pre-pregnancy BMI group and inadequate PNC, the specific

effect of each interaction term was subsequently removed from that "average" effect, leaving

only the effect of inadequate PNC alone in the main effect variable; hence, inadequate PNC was

no longer significant.

Summary of Multivariate Results

Overall, the inconsistency of results between the baseline logistic regression models, the

primary and secondary logistic regression models addressing the first specific aim, and the

primary and secondary logistic regression models addressing the second specific aim showed









that there is no moderating effect of PNC on the relationship between pre-pregnancy BMI and

PPD symptoms. However, the inconsistency of results with respect to some groups of women

(e.g., women who had an underweight pre-pregnancy BMI) shed some light, possibly warranting

further exploration on these groups of women to determine reasons and additional variables

responsible for the inconsistency of results. For example, this study revealed that many medical

and obstetric problems are faced by women during pregnancy, and should be further examined.

Perhaps further exploring an unexpected result in this study, specifically, lower odds for PPD

symptoms among women who had an obese pre-pregnancy BMI who sought little to minimal

(inadequate) PNC (compared to women who had an obese pre-pregnancy BMI who received

adequate PNC), and the extent to which these women are affected by medical and obstetric

problems, would prove to be valuable. Also, another population to further investigate would be

women who utilized adequate plus PNC who are high-risk. This study showed that the

significance for having a higher odds for PPD symptoms among women who utilized adequate

plus PNC, compared to women who utilized adequate PNC, disappeared after incorporating two

different approaches of risk-adjustment, suggesting that it is the high-risk adequate plus PNC

women who are at risk for PPD symptoms. Thus, further research that looks into women who

utilize adequate plus PNC may confirm this, and/or provide insight as to why the significance

between women who utilized adequate plus PNC and PPD symptoms was present when control

variables (including morbidity control variables as predictors of PPD) were not included, and the

sample represented by that result included both high-risk women and non high-risk women.

Overall, this inconsistency of results did not agree with results from previous studies (e.g.,

an association between pre-pregnancy BMI and PPD symptoms). Reasons to explain the

inconsistency of results within this study may be attributed to the paucity of variables in the data









to address the content of PNC with respect to nutrition and wellness (e.g., excessive weight gain,

exercise, etc.) discussions, discussion of PPD during the delivery of PNC, discussion of what to

possibly expect regarding weight retention in the postpartum period (e.g., returning back to pre-

pregnancy weight), the type and number of PNC providers) since different disciplines may be

trained to deliver PNC differently to some extent and having multiple providers may affect the

content and quality of PNC delivered, and a weight gain discussion measure for all the states to

include. The one question on weight gain discussion was only available for two states. However,

if this question were included in the PRAMS Core Questionnaire (mandated for all states to ask),

the results might have been affected; this is hypothesized considering the results for the logistic

regression estimated on a subpopulation of women who received weight gain discussion from

their PNC (pp. 186-187) showed significance for three variables (p<0.05), all within the same

pre-pregnancy BMI group: underweight pre-pregnancy BMI, and two interaction groups for

underweight pre-pregnancy BMI: inadequate and intermediate PNC. The PNC questions

included in PRAMS referred to topics such as the discussion of vitamins, if the women got PNC

as early as she wanted, how satisfied she was with the staff and the waiting time. Many of the

PNC variables that are suggested, and were not a part of the data, may alter the results shown in

this study, possibly leading to more consistency. Though the database was extensive and

included a variety of variables, many of which were included as control variables, the lack of

variables related to the discussion of delivery of PNC related to nutrition and wellness leaves

much room for questions on the standardization of the content of PNC among the 51,600 women

included in the study. Appendix A (pp. 137-155), which includes a collection of literature on the

content of PNC, suggests that though many PNC providers are discussing nutrition and wellness

matters in the delivery of PNC, there is still a considerable amount of variability that exists in the









delivery of PNC with respect to nutrition and wellness. Hence, this may have had an impact on

the results of this study since there were minimal measures representing this content of PNC.

This study included 34 control variables in the primary logistic regression models and 29

control variables in the secondary logistic regression models. The inclusion of a variety of

control variables in the model helped to limit the amount of omitted variable bias in this study.

Had some of the control variables not been included in the multivariate analyses, further

significance may have been demonstrated with the variables of interest (e.g., interaction terms),

but the significance would not have been a "true" significance. Though the results suggested that

the inclusion of many control variables removed the significance demonstrated among some of

the variables of interest in the absence of those control variables (e.g., PNC utilization), it

simultaneously assured that overall, there is no "true" moderating effect of PNC in this sample.

Limitations

Since PRAMS is a self-reported survey, the likelihood of recall bias (e.g. frequency a

woman experienced depressive symptoms, pre-pregnancy weight) remains among the

participants. A woman may not remember her pre-pregnancy weight, especially because it was

her weight more than nine months ago. If pre-pregnancy weight is reported inaccurately, this

could cause error in the frequency of women within a pre-pregnancy BMI category, which could

in-turn bias regression estimates. Second, regarding the reporting of PPD symptom severity,

because of the stigmas that exist regarding mental disorders in general, a woman may have a

tendency to under-report her symptom severity and/or overlook the frequency and severity of her

symptoms (e.g., for fear of being termed a "bad mother"). This may be especially so if societal

pressures to be a good mother and a mother's desire to try and do everything she feels is

necessary for her baby prompt her to not want to succumb to openly admitting to PPD symptoms

(Epperson, 1999). If symptom severity was inaccurately reported, the prevalence of PPD









symptoms would be under-reported for the purpose of this study. Third, the amount of weight a

woman gained during pregnancy was not controlled for, and there were no measures indicating if

a woman experienced pregnancy-related obesity or eating disorders that some women experience

during pregnancy. Evidence exists that both are predictors of postpartum distress (Krummel,

2007; Franko & Spurrell, 2000). Not controlling for either of these factors can undermine the

true effect that pre-pregnancy BMI may have on PPD symptoms. Fourth, it is expected that in a

self-reported survey, women may have the tendency to either under-report or over-report their

weight. For women who had a pre-pregnancy BMI of obese or overweight, they may have a

higher tendency to under-report their weight, and for women who had a pre-pregnancy BMI of

underweight, they may have a tendency to over-report their weight; thus, the weight indicated on

the survey may not be reliable for some of the women. Similar to the reasoning explained

previously for recall bias, if pre-pregnancy weight is reported inaccurately, this could cause error

in the frequency of women within a pre-pregnancy BMI category, which could in-turn bias

regression estimates. Fifth, the APNCU does not include quality or content of PNC in its

measure, which could impact the likelihood of PPD. For example, the satisfaction a woman feels

during the delivery of her PNC may impact whether she seeks "adequate PNC," especially if she

feels uncomfortable in discussing certain pregnancy (psychosocial) issues with her PNC provider

and/or she feels that she is unable to reap the benefits of PNC due to the lack of the duration for

each visit. Not accounting for the quality of PNC could either underestimate the effectiveness

that PNC may have on reducing the likelihood for PPD symptoms. However, the sub-analyses

included a model on a subpopulation of women who had weight gain discussed during their

pregnancy; this is one measure regarding the content of PNC that was included, though the

quality of this content was not included. Sixth, there may be other variables that could impact









PPD symptomatology and were not controlled for in this study; hence, resulting in omitted

variable bias. However, given the breadth of the database used for the analysis, this study

attempted to control for the factors that would be most associated with pre-pregnancy BMI, PNC

utilization, and PPD symptoms based on the existing literature. Seventh, this study did not

control for depression history or depressive episodes during pregnancy, both of which are shown

to increase the likelihood of PPD symptoms (Gotlib, 1989; Gotlib, Whiffen, Wallace, & Mount,

1991; Beck, 2001). Having a systematic form of screening both during pregnancy and

monitoring women who show symptoms of depression during pregnancy may help prevent

depression postpartum, especially because many women suffer depression silently during

pregnancy (Marcus, Flynn, Blow, & Barry, 2003; Smith et al., 2004). Not controlling for

depression history could also undermine the effect of the association between pre-pregnancy

BMI and PPD in that a woman may have had an even higher likelihood for PPD symptoms

because of previous episodes of depression. Eighth, because it was unclear whether the woman

included in this study received an actual diagnosis of PPD, any conclusions were made with

respect to PPD symptomatology, and not PPD itself. There could be a significant difference

between a woman's perception of her PPD symptoms versus a professional diagnosis of PPD

symptoms; a woman may not recognize the severity of her symptoms (Epperson, 1999). The

effect of this could lead to a misconception of the true prevalence of PPD in this study. Finally,

and most importantly, this study was observational in nature and not causal.

Importance of This Study/Implications

Though this study determined through multiple models and two different approaches to

risk-adjustment (e.g., statistically risk-adjusting versus truncating the population) that there is no

consistent moderating effect of PNC in the association between pre-pregnancy BMI and PPD

symptoms, this study provides a great deal of insight regarding PNC delivery, for researchers,









policymakers, and clinicians, much of which involve and can be heightened with collaboration.

Further research should look in-depth into discussions between pregnant women and a variety of

health care providers (e.g., physicians, nurses, midwives, etc.), while also looking into the

relationships between women and each of these health care providers to assess interpersonal

communication that may provide some insight into the BVI/PNC/PPD relationship among

different providers. Even though the purpose of this study sought to find that pre-pregnancy BMI

is a potential marker for imminent PPD symptoms, the risk-adjustment processes carried out in

this study generated strong evidence that many of the women experienced an array of medical

and obstetric problems during pregnancy, many of which were associated with PPD symptoms in

this study. Thus, further research should also look into the content of discussions between

patients and providers regarding 1) the identification of problems and the risk factors that prompt

providers to deliver suitable interventions, and 2) the extent to which one or more of these

medical and/or obstetric problems are associated with PPD symptoms and/or other psychosocial

consequences.

In general, research focusing on the content of PNC in addition to the quantity of PNC

delivered over the recent years could determine if the results are consistent with the existing

literature and the protocols to be followed during PNC delivery. Ongoing research should

address whether different PNC providers are adhering to PNC guidelines and to what extent they

are adhering to them. In today's practice, if it is found that there is an inconsistency with the

delivery of PNC content regarding weight, nutrition, and wellness, perhaps policymakers should

seek to standardize the delivery of PNC through policy initiatives such as periodic

accountability. Policy regarding evaluation of PNC in a variety of settings with a variety of









providers may also provide insight into the quality and content of PNC delivered such as the

strengths and weaknesses of current PNC.

Regarding the discussion of PPD as a disorder that may affect women in the year following

the birth of the child, research should look into the extent to which PNC providers are discussing

this with their patients in the pregnancy period. Also, PNC providers should address postpartum

depression and educate patients during PNC, as many women may be unaware of this disorder

and/or when the severity of symptoms necessitates medical attention. With regard to access to

health care in the postpartum period, it may be of worth to look into the extent to which woman

are undiagnosed with PPD symptoms, if not PPD itself, due to a lack of access to health care that

ceases to exist after a woman's six-week postpartum check-up. It is suggested that preventive

care resources tend to be available during pregnancy, but may not be as readily available during

other times outside of pregnancy (Kopelman et al., 2008). Since 1) PPD is underdiagnosed

and/or overlooked in the United States (O'Hara, & Gorman, 2004; Clayton, 2004), 2) the use of

screening instruments for PPD remains uncommon in the U.S. (Seehusen, Baldwin, Runkle, &

Clark, 2005; Georgiopoulous, Bryan, Wollan, & Yawn, 2001), and 3) the association of PPD

screening with higher rates of symptom recognition, diagnosis, and treatment as well as the

feasibility and appropriateness of screening has been suggested (Georgiopoulous et al., 1999;

Georgiopoulous et al., 2001), policy initiatives should seek to facilitate PPD screening

periodically during the first-year postpartum, working towards standardizing PPD screening.

Policy initiatives should also seek to train health care providers in being cognizant of

signs/symptoms of PPD, screening for PPD, and the importance of addressing and screening for

PPD during pregnancy, hospitalization for delivery, and in the postpartum period (Seehusen et

al., 2005). Further research should also investigate screening for women at-risk for PPD









symptoms during PNC as it has been suggested that detection rates for depressive disorders are

lower in obstetric settings compared to other primary care settings (Smith et al., 2004). Looking

at the extent to which PPD screening takes place during PNC delivery would help 1) identify

those women who are at-risk for PPD due to previous and/or current history of depression, and

2) subsequently provide additional medical attention to those women who are identified as

having a history of depression. The research implications suggested from the results of this study

also encourage policy initiatives to help cultivate research for PPD as a plethora of questions

remain regarding PPD in general among women in the United States.

This study makes many contributions. First, this study is among the first in the United

States that stratified the quantity of PNC among women of different pre-pregnancy BMI groups,

while looking at the effect on the likelihood of PPD symptoms. Secondly, because women from

a variety of PPD symptom severities were included in this study, the importance of both

screening for PPD during PNC, the delivery and postpartum hospitalization period, and the six-

week postpartum check-up are stressed, because all severities of PPD symptoms are prevalent

among women in the U.S., according to the data. Thus, PNC, the hospitalization period, and the

postpartum check-up remain critical points to screen for PPD. Also, if a postpartum woman

seeks her postpartum check-up visit through her PNC provider, this study can affirm the

importance of PNC providers in facilitating a healthy relationship with their patients (e.g.,

tailoring PNC to each woman's needs, supporting an environment that is conducive for a woman

to openly address her concerns as a pregnant individual by encouraging open discussion, etc.).

This may increase the likelihood that patients will seek care from their PNC provider in the

postpartum period through a postpartum check-up visit. Also, because some relationships were

found among BMI and PPD after stratifying by PNC utilization, this study adds to the literature









that stresses the importance of addressing obesity and PPD because both are rising public health

concerns in the United States and globally. Finally, and most importantly, although the results

showed that there is no association between pre-pregnancy BMI and PPD symptoms (after

including suitable control variables), and PNC utilization does not generally moderate this

relationship, the results uncovered that many of the women were significantly affected by a

variety of medical and obstetric problems, many of which were high-risk and associated with

PPD symptoms. For future research, it is strongly recommended that the possible association of

these problems with PPD symptoms be further investigated. For practice, it is suggested that 1)

PNC providers recognize the risk factors for and the prevalence of medical and obstetrical

morbidities during pregnancy, including, but not limited to those featured in this study, 2)

identify and diagnose the morbidities that surface in their patients, 3) establish suitable

interventions, and finally, 4) follow-up on their patients accordingly.









APPENDIX A
SUMMARY OF LITERATURE ON PNC CONTENT










Authors


Covington &
Rice (1997)


Objectives


To explore the
association
between patient
receipt of
recommended
prenatal care
interventions
and infant birth
weight


Independent/
dependent
variables
Prenatal care
initial
interventions
(weighted and
measured), health
promotion advice
received (eat
proper foods,
gain weight),
birth weight


Sample
characteristics

3,905 African-
American
women


Data


1988
National
Maternal and
Infant Health
Survey


Type of
prenatal care
guidelines
U.S. Public
Health
Service
Expert Panel
on the
Content of
Prenatal Care


Results (statistical tests, p-values)


1) Height/weight taken at initial
PNC visit
a) 98% with a VLBW infant
(<1,500 grams) or a "moderately
low birth weight" (1,500-2,499
grams) infant
b) 97% with a normal birth
weight infant (2,500 grams or
greater)

2) Receiving advice on proper
foods
a) 92% with a VLBW infant
b) 93% with a MLBW infant
c) 90% with a NBW infant

3) Receiving advice on weight
gain:
a) 64% with a VLBW infant
b) 65% with a MLBW infant
c) 71% with a NBW infant

3) Association between women
who did not receive all types of
health promotion advice and birth
weight: OR: 1.28 to give birth to
a VLBW infant (adjusted for
LBW risk)











Authors Objectives Independent/ Sample Data Type of Results (statistical tests, p-
dependent characteristics prenatal values)
variables care
guidelines
Freda, To compare the Prenatal care 159 women Questionnaires U.S. Public 1) Nutrition information received
Andersen, type of information (80 who Health a) Private settings: 85% of
Damus, & information received, prenatal received care in Service women
Merkatz given to care delivery site a public setting, Expert b) Public settings: 96% of
(1993) women who 79 who Panel on women
sought prenatal received care in the Content
care in public a private of Prenatal 2) Exercise during pregnancy
and private setting) in Care a) Private settings: 56% of
clinics and the Bronx, New (1989) women
degree to York. b) Public settings: 64% of
which the women at public settings
women were reported receiving this
satisfied with information (p=0.3)
the information 3) Patient satisfaction with
they were information received: patients
given during were more likely to experience
their prenatal satisfaction on any PNC topic if
care providers initiated discussion










Authors


Kogan,
Alexander,
Kotelchuck,
Nagey, &
Jack (1994)


Objectives


What percent of
women reported
receiving PHS
recommended
procedures and
health behavior
advice and how do
they differ based on
health insurance,
site of care, and
sociodemographics?


Sample
characteristics

9,932 women,
nationally
representative


Data


1988
National
Maternal and
Infant Health
Survey


Type of
prenatal care
guidelines
U.S. Public
Health
Service's
Expert Panel
on the
Content of
Prenatal Care
Report
(1989)


Independent/
dependent
variables
USPHS
recommended
national
guidelines for
prenatal care
(health
behavior
advice and
prenatal care
procedures),
reports of
receiving
different types
of prenatal
care
procedures and
health
behavior
advice


Results (statistical tests, p-
values)

Percentages reported for:
1) Weight/height taken at 1st or
2nd visit:
a) Maternal education: 95.3-
98.2%
b) Household income, 95.8-
98.5%
c) Marital status: 96.7-98%
d) Race/ethnicity: 92.4-98.7%
e) Trimester care began: 94.6-
98%
f) Site of care: 96-98.3%

2) Proper foods advice:
a) Maternal education: 87.5-
94.2%
b) Maternal age: 90.8-93.4%
c) Household income: 90.6-
94.4%
d) Marital status: 91.5-93.3%
e) Race/ethnicity: 88.3-94%
f) Parity: 91.4-94.3
g) APNCU: 85.7-94.2%

3) Weight gain advice:
a) Maternal education: 64.9-73
b) Maternal age: 64.9-74.7
c) Household income: 68.9-74.4
d) Marital status: 70.9-76.6
e) Race/ethnicity: 62.2-74.1
f) Parity: 64.7-78
g) APNCU: 59.7-74.5










Authors


Kogan,
Alexander,
Kotelchuck,
& Nagey
(1994)


Independent/
dependent
variables
Health behavior
advice and initial
prenatal care
procedures, low
birth weight
(<2,500 g)


Sample
characteristics

9,394 women
(nationally
representative)


Data


1988
National
Maternal and
Infant Health
Survey


Objectives


To examine the
relationship
between
maternal
reports of
health behavior
advice received
and initial
prenatal care
procedures
performed
during the first
two visits and
low birth
weight


Type of
prenatal care
guidelines
U.S. Public
Health
Service's
Expert Panel
on the
Content of
Prenatal Care
Report
(1989)


Results (statistical tests, p-values)


1) Health behavior advice: the
following results were reported
a) 8,670 women who received
advice on proper diet: 5.6% gave
birth to LBW infants (p=0.06)
b) 6,770 women who received
advice on weight gain: 5.3% gave
birth to LBW infants (p<0.01)

2) For initial prenatal care
procedures, among 9,159 women
who had their weight recorded,
5.6% gave birth to LBW infants
(p=0.03)











Authors Objectives Independent/ Sample Data Type of Results (statistical tests, p-values)
dependent characteristics prenatal
variables care
guidelines


Kotelchuck,
Kogan,
Alexander, &
Jack (1997)


To assess if
site of
prenatal
care
delivery
influences
the content
of prenatal
care given
to low-
income
women


Recommended
initial prenatal
care
procedures,
recommended
prenatal care
advice, site of
prenatal care
delivery


3,405 low
income
women


1988
National
Maternal
and
Infant
Health
Survey


U.S.
Public
Health
Service
Expert
Panel on
the
Content of
Prenatal
Care


1) Between 89.6-92.6% of the women reported
received advice on proper foods to eat during
pregnancy (not-significant) and 64.5-79.9% of
women reported receiving advice on weight
gain during pregnancy (p<0.001)

2) Comparing the content of PNC at different
sites, between 95.1-97.9% of women were
measured and weighed at their initial PNC visit
(p=0.006), and between 75.4-87.4% of women
had their health history taken (p<0.001)

3) Women who received their PNC at a private
office were 1.52 times (CI: 1.18-1.95) more
likely not to receive all the PNC procedures
(e.g., blood pressure taken, height and weight
measured, blood work taken, etc.) at their initial
visit, and 1.76 times (CI: 1.34-2.32) more likely
not to receive all the types of PNC advice (e.g.,
alcohol and smoking cessation, proper foods to
eat, vitamins to take, weight to gain, etc.) as
recommended by the U.S. Public Health
Service.

4) Women who received care at sites other than
a private office, public clinic, an HMO, or a
hospital clinic were 1.73 times (CI: 1.04-2.83)
more likely not to receive all the PNC
procedures at their initial visit










Authors


Libbus &
Sable (1991)


Objectives


To examine the
relationship
between
absence of
prenatal care
educational
content and the
risk of adverse
birth outcomes


Independent/
dependent
variables
Six educational
prenatal care
content areas, 10
risk areas, term
low birthweight,
and preterm low
birthweight


Sample
characteristics

1,484 women
from three
regions in
Missouri


Data


Data from a
previous
study on
barriers to
prenatal care
sponsored by
the Missouri
Department
of Health and
the Missouri
Perinatal
Association


Type of
prenatal care
guidelines
Though no
source is
mentioned in
choosing the
areas of
prenatal
content, the
Institute of
Medicine and
the U.S.
Public Health
Service
Expert Panel
on the
Content of
Prenatal Care
were
included in
the reference
list.


Results (statistical tests, p-values)


1) 20.4% of the women reported
receiving diet counseling

2) 14.2% of the women possessed
a nutritional risk

3) Not receiving diet education
was significantly associated with
the risk of delivering a preterm
low birthweight infant in the
bivariate analyses, but not in the
multivariate analyses.

4) Adequacy of care (care
initiated w/in 1st 4 months of
gestation & atleast 8 visits (term
infants) or 5 visits (infants bom
a) Adequate care: 720 women:
13.2% did not receive diet
counseling
b) Inadequate care: 764 women:
27.1% did not receive diet
counseling











Authors


Sable &
Herman
(1997)


Independent/
dependent
variables
Prenatal care
advice, birth
weight


Sample
characteristics

2,205 women
from the state
of Missouri


Data


National
Institute of
Child Health
and Human
Development/
Missouri
Maternal and
Infant Health
Survey


Type of
prenatal care
guidelines
U.S. Public
Health
Service
Expert Panel
on the
Content of
Prenatal
Care


Objectives


To 1) examine
the relationship
between the U.S.
Public Health
Service Expert
Panel on the
Content of
Prenatal Care
recommendations
and the risk of
low birth weight,
and 2) to describe
the type and
frequency of
health behavior
advice given to a
sample of
pregnant women.


Results (statistical tests, p-values)


1) 54.8% of the women received
advice on improving diet and
nutrition and eating proper foods

2) Regarding weight gain, 62.1%
of women received this advice
during the course of their parental
care

3) For receiving advice on diet
and nutrition, 31.6% of the
women were told to watch their
caloric intake and to avoid
excessive weight gain

4) Regarding exercise factors,
29.3% of women were told to get
more exercise, and 16.7% of
women were told to restrict their
exercise

5) In looking at birth weight,
women who did not report
receiving all the seven types of
advice during their PNC as
recommended by the U.S. Public
Health Service Expert Panel were
1.49 times (CI: 1.10-1.88) more
likely to give birth to a baby that
was of "very low birth weight"
(less than 1,500 grams) than they
were to give birth to a baby of
normal birth weight











Authors


Baldwin,
Raine,
Jenkins, Hart,
& Rosenblatt
(1994)


Objectives


To what extent
do obstetric
providers
follow ACOG
guidelines?


Independent/
dependent
variables
Components of
first prenatal care
visit, number of
prenatal care
visits, ACOG
recommended
laboratory tests,
prenatal content
monitoring
(subsequent
visits)


Sample
characteristics

Providers: 54
urban OB-
GYN's, 29
rural OB-
GYN's, 59
urban FP's, 67
rural FP's, 43
urban MW's;
2,357 female
patients


Data


The Content
of Obstetrical
Care Study


Type of
prenatal care
guidelines
American
College of
Obstetricians
and
Gynecologists


Results (statistical tests, p-
values)

1) Pre-pregnancy weight was
recorded as follows (ANOVA,
p 68%; UFP: 83%; RFP: 79%;
UNW: 98%

2) Maternal weight at 1st visit
(ANOVA, p 87%; ROB: 98%; UFP: 95%;
RFP: 90%; UNW: 96%

3) Maternal height at 1st visit
(ANOVA, p 80%; ROB: 73%; UFP: 58%;
RFP: 59%; UNW: 98%


4) Weight at subsequent visits
(ANOVA, p 97%; ROB: 99%; UFP: 96%;
RFP: 98%; UNW; 98%











Authors Objectives Independent/
dependent
variables
Conway To Adequate care
& determine index
Kutinova the efficacy (APNCU=1),
(2006) of prenatal prenatal are
care and the measures:
policies advice about
designed to weight gain,
improve advice about
access to eating,
prenatal excessive
care maternal
(Medicaid) hospitalization,
BMI status
change (became
overweight after
conception;
become
underweight
after
conception)


Sample
characteristics

7,464
observations


Data


1988
National
Maternal
and
Infant
Health
Survey


Type of
prenatal care
guidelines
American
College of
Obstetricians
and
Gynecologists


Results (statistical tests, p-values)


1) Statistical significance was found between
receiving advice about eating and excessive
maternal hospitalization (if the mother's length of
stay was longer than her infant). No significant
associations were found between receiving weight
gain advice and excessive maternal hospitalization.

2) Receiving advice about eating was significantly
associated with having a BMI change to
"underweight" afterbirth.

3) A significant association was found for
receiving advice about weight gain, and a change
in BMI status after birth to "underweight."

4) Inverse associations were found for receiving
advice about weight gain and having a change in
BMI status after birth to o\ i c iglui "No
significant associations however were found
between receiving either advice about eating or
weight gain during pregnancy, and a change in
BMI status to o\ ilci glil "

5) For women with "adequate PNC" (APNCU
index), approximately 93-94% of the women
received advice about eating. For the women who
did not receive "adequate care," 87-94% of the
women received advice about eating.

6) 70-75% of women who received "adequate
care" received advice about weight gain during
pregnancy.

7) 61-72% of women who did not receive
"adequate care," received advice about weight gain
during pregnancy.










Objectives


To determine
the prevalence
nutrition advice
received by
women who
sought prenatal
care (self-
report)


Independent/
dependent
variables
Nutrition advice
(seven
categories),
maternal
characteristics


Sample
characteristics

9,639 mothers
who gave birth
to a live infant,
4,955 mothers
who did not
give birth to a
live infant


Data


1988
National
Maternal and
Infant Health
Survey


Authors


Yu &
Jackson
(1995)


Type of
prenatal care
guidelines
Though no
guidelines
were noted in
the literature
review, the
American
Academy of
Pediatrics,
American
College of
Obstetrics
and
Gynecology,
and the
Institute of
Medicine
were
included in
the reference
list.


Results (statistical tests, p-values)


1) Among mothers who gave birth
to a live infant, 72.8% of White
women received advice about
weight gain during pregnancy,
70.1% of Black women received
advice about weight gain during
pregnancy, 63% of Asian and
Pacific Islander women received
advice about weight gain during
pregnancy, and 73.8% of Eskimo,
Aleut, and American Indian
women received advice about
weight gain during pregnancy

2) Regarding eating properly, this
advice was received by 93% of
White women, 92.7% of Black
women, 90.2% of Asian and
Pacific Islander women, and
89.3% of Eskimo, Aleut, and
American Indian women

3) Among mothers who did not
give birth to a live infant, 87.3%
received advice on eating
properly, and 63.8% received
advice on weight gain











Authors Objectives


To evaluate
if weight
gain advice
given from
a health
care
provider, a
woman's
target
gestational
weight gain,
and actual
weight gain
are in
congruence
with the
IOM
guidelines


Independent/
dependent
variables
Advised
weight gain,
target weight
gain, and
actual weight
gain


Sample
characteristics

2,237 women


Data


Prenatal
questionnaire,
neonatal
questionnaire


Type of
prenatal care
guidelines
Institute of
Medicine


Cogswell,
Scanlon,
Fein, &
Schieve
(1999)


Results (statistical tests, p-values)


1) 27% did not receive weight gain advice during
PNC

2) Advice about weight gain and IOM
recommendations
a) 14% were advised to gain less weight than IOM
guidelines
b) 22% advised to gain more weight than IOM
guidelines

3) Target weight gain and IOM recommendations
a) 19% had a target weight gain less than IOM
guidelines
b) 22% had a target weight gain higher than IOM
guidelines

3) Actual weight gain and IOM recommendations
a) 23% actually gained less than IOM guidelines
b) 42% of women gained more than IOM guidelines

4) Women in the "very high" pre-pregnancy BMI
category were 15 times more likely to receive advice
to gain more weight than as recommended by the
IOM and 0.9 times as likely to receive advice to gain
less weight than as recommended by the IOM

5) Women in the "high" pre-pregnancy BMI
category were 31.8 times more likely to receive
advice to gain more weight than as recommended by
the IOM and 0.1 times as likely to receive advice to
gain less weight than as recommended by the IOM

6) Women in the "low" pre-pregnancy BMI category
were 0.5 times as likely to receive advice to gain
more weight than as recommended by the IOM and
0.8 times as likely to receive advice to gain less
weight than as recommended by the IOM










Authors


Stotland,
Haas,
Brawarsky,
Jackson,
Fuentes-
Afflick, &
Escobar
(2005)


Independent/
dependent
variables
Pre-pregnancy
BMI, women's
target gestational
weight gain,
provider weight
gain advice


Objectives


To study the
relationship
between pre-
pregnancy
BMI, women's
target
gestational
weight gain,
and provider
weight gain
advice


Data


Project
WISH
(Women and
Infants
Starting
Healthy)


Type of
prenatal care
guidelines
Institute of
Medicine


Sample
characteristics

1,198 women
in the state of
California who
received PNC
at 1) an urban
public hospital,
2) an urban
community
hospital, 3) a
university
hospital, or 4) 1
of 3 medical
centers
affiliated with
an MCO.
Weight gain
advice was
received from
either a
physician,
nurse, or a
nutrition
counselor.


Results (statistical tests, p-values)


1) Relationship between BMI and
pregnancy target weight gain that
was below IOM guidelines:
a) "Low" BMI (OR: 0.63)
b) "Overweight" BMI (OR: 0.05)
c) "Obese" BMI (OR: 0.18)

2) Relationship between BMI and
pregnancy target weight gain that
was above IOM guidelines:
a) "Overweight" BMI (OR: 3.79)
b) "Obese" BMI (OR: 2.39)

3) Association between provider
advice and a woman's target
weight gain during her pregnancy
a) Advice to gain weight gain
below IOM guidelines and target
weight gain below IOM
guidelines (OR: 3.17)
b) Advice to gain weight above
IOM guidelines and target weight
gain above IOM guidelines (OR:
3.39)
c) No advice and target weight
gain below IOM guidelines (OR:
1.72)











Authors


Petersen,
Connelly,
Martin, &
Kupper
(2001)


Independent/
dependent
variables
Reports of
preventive health
counseling during
prenatal care
(e.g., nutrition)


Sample
characteristics

24,620 women
from 14 states


Data


Pregnancy
Risk
Assessment
Monitoring
System


Type of
prenatal care
guidelines
The U.S.
Preventive
Services
Task Force
Guide to
Clinical
Preventive
Services


Results (statistical tests, p-values)


Between 84-92% of the women
(depending on the state) received
counseling on nutrition during
pregnancy


Objectives


To determine
1) the
prevalence of
preventive
health
counseling
given during
prenatal care,
2) the
prevalence of
women who
are in higher
need of
counseling
about specific
health
concerns, and
3) if women
who are in
higher need of
counseling are
more likely
than the
women in
lower need to
have received
counseling.











Authors


Levine,
Wigren,
Chapman,
Kemer,
Bergman, &
Rivlin (1993)


Objectives


To examine the
degree to
which primary
care physicians
in the U.S.
report
practicing the
basic
nutritional
competencies
in the delivery
of care.


Independent/
dependent
variables
Nutrition related
attitude
statements,
nutrition-related
behaviors


Sample
characteristics

3,416 primary
care physicians


Data


A
demographic
survey, an
attitude
survey, and a
behavior
survey


Type of
prenatal care
guidelines
N/A


Results (statistical tests, p-values)


1) 75% or more of the physicians
agreed or strongly agreed with the
following: a) Continuing medical
education courses should devote
time to nutritional-related issues,
b) it is important to have an
understanding of food
composition and preparation to
provide reliable nutritional
counseling, c) in many cases,
medication could be reduced or
eliminated if patients followed a
recommended diet, d) nutrition
will have an increasingly
important role in the prevention
and treatment of disease, e)
doctors should spend more time
exploring dietary habits during
patient evaluation

2) Statements towards which 75%
or greater of physicians in the
sample disagreed or strongly
disagreed, this included: a) Most
doctors are very knowledgeable
about nutrition, b) physicians are
well prepared to provide
nutritional counseling, c) nutrition
is important only in certain
medical specialties, d) dietary
counseling is a waste of time
because people don't change their










habits anyway, and e) nutrition
education is not the responsibility
of the physician.

3) Regarding specific nutritional
advice on what physicians usually
or always practice, the authors
found that a) 61% of the
physicians reported advising or
teaching their patients about the
rationale for dietary
modifications, b) 60% of the
physicians reported advising or
teaching their patients about
achievements and maintenance of
health habits such as exercise, c)
56% of physicians reported
advising or teaching their patients
about the achievements of
desirable weight, d) 55% of
physicians reported prescribing to
their patients dietary
modifications such as sugar or
salt intake reductions, weight
reduction, e) 54% of physicians
reported monitoring their
patients' nutrition status and
progress in response to treatments
recommended, and f) 52% of
physicians reported prescribing to
their patients exercise depending
on their age, physical
conditionss, and their health
status.










Authors


Splett,
Reinhardt, &
Fleming
(1994)


Objectives


To 1) identify
physicians'
need and
expectations
regarding
quality
nutrition
services
rendered in
prenatal care 2)
to rank the
characteristics
of the services
rendered by
importance in
making
nutrition
referral
decisions, and
3) identify
nutrition
services
physicians
would most
likely add to
their delivery
of care.


Independent/
dependent
variables
Nutrition care
services,
availability of
services:
currently
available/likely to
add/unlikely to
add; physicians'
rating of nutrition
services


Sample
characteristics

130 prenatal
care OB-GYN
physicians


Data


Quality
Service
Management
Model


Type of
prenatal care
guidelines
Though no
guidelines
were noted in
the literature
review, the
Institute of
Medicine
was included
on the
reference list


Results (statistical tests, p-values)


1) 63.8% of the physicians had an
ongoing monitoring of patient
weight gain and dietary patterns,
while 16.9% of physicians reported
that they would likely start
practicing the service, and 13.1%
of the physicians reported that they
would unlikely to start practicing
the service

2) Regarding initial PNC screening
of women to detect their nutritional
risk, 61.5% of physicians reported
that they provided that service,
24.6% of physicians reported that
they would likely start practicing
the service, and 10.8% of the
physicians reported that they would
unlikely start practicing the service

3) Regarding follow-up of women
who were identified as having
nutrition problems during their
pregnancy, 42.3% reported that
they currently provided the service,
43.8% reported that they would
likely start practicing the service,
and 5.4% of the physicians reported
that they would unlikely start
practicing the service


4) Regarding nutritional
consultation for each woman
during her PNC, 40% of the
physicians reported providing the










service, 43.1% of the physicians
reported that they would likely start
practicing the service, and 11.5%
of the physicians reported that they
would unlikely start practicing the
service

5) Nutritional assessment and
planning of care for women with a
"high-risk" pregnancy, was
currently conducted by 36.2% of
physicians, while 53.8% reported
that they would likely start
practicing the service, and 6.2% of
physicians reported that they would
unlikely start practicing the service

6) 26.2% of physicians reported
that they currently practice
postpartum weight counseling,
while 61.5% of physicians reported
that they would likely start
practicing the service, and 6.2% of
physicians reported that they would
unlikely start practicing the service.

7) 32% of the women engaged in
discussions with their physician for
seeking advice on nutrition
problems, while 12% of the women
engaged in discussions with a
registered dietitian, 8% sought
information from PNC classes, 8%
sought information from brochures
and pamphlets given in the office
where the PNC was provided, and
3% sought assistance through WIC









APPENDIX B
MULTIVARIATE SUB-ANALYSES

In addition to the main logit models (PPD), sub-analyses were conducted to further test the

moderating effect of PNC in the association between pre-pregnancy BMI and PPD symptoms.

One of the sub-analysis logit models was estimated by specifying the dependent variable (PPD

symptoms) differently (a sensitivity analysis via an ordinal logistic regression model), while

another model was estimated with a different dependent variable (adequate plus PNC), and five

of the sub-analysis logit models comprised of five different subpopulations: one consisting of

women who utilized WIC services, and each of the four remaining models comprising a different

subpopulation income group.

Adequate Plus

Unlike the secondary analysis that risk-adjusted for high-risk pregnancies by removing

observations that met any of the high-risk criteria defined for this study, a logistic regression was

estimated holding adequate plus as the dependent variable, to determine high-risk characteristics

(denoted by significance). Seventeen morbidities, among other characteristics (e.g.,

demographics) were included in this model to determine factors that could label a pregnancy as

"high-risk." These morbidities included:

1) Pre-pregnancy diabetes
2) Gestational diabetes
3) Vaginal bleeding
4) Kidney/bladder infection
5) Nausea
6) Hospitalization
7) Preterm labor
8) Premature rupture of membranes (PROM)
9) Placenta abrutio or placenta previa
10) Incompetent cervix (cervix closed)
11) High blood pressure
12) Blood transfusion
13) Car crash injury
14) Bed rest









15) Labor/delivery complications (in general)
16) Pregnancy abnormalities (in general)
17) Medical risk factors (in general)

Sensitivity Analysis: Ordinal Logistic Regression

To further test the moderating effect of PNC, a sensitivity analysis employed an ordinal

logistic regression and kept PPD symptom response as seven categories (scores of 0-6), prior to

grouping the scores into "yes" or "no" to create a dichotomous variable for the primary logit

model. This model was used to estimate PPD symptoms as an ordinal, categorical variable to

determine whether the results are sensitive to the way in which the dependent variable is

specified. Reference groups remained the same as the primary logistic regression model. With

the dependent variable specified into seven categories, it was also predicted that the highest

likelihood for increasing in PPD severity was for women who had a pre-pregnancy BMI of obese

and received inadequate care.

As shown in the model specification below, each PPD response (assigned a score) had its

own a, value (while the 3 coefficients of the independent variables for each PPD response

remained the same):

P(Y < j)
logit = log -< P J) = + Pi3(obese/adequate plus care) + 32(overweight/adequate
(1 P(Y :!j)
plus care) + 33(underweight adequate plus care) + 34(obese/intermediate care) +
P3(overweight/intermediate care) + 36(underweight/intermediate care) + 37(obese/inadequate
care) +38(overweight/inadequate care) + 39(underweight/inadequate care) + 3kXk(control
variables) + s (B-l)

(Wherej represents the PPD score assigned to each woman)

Women, Infants, and Children (WIC)

A logistic regression was estimated only on women who received WIC services during

their pregnancy. Eligibility for WIC during pregnancy is based on socioeconomic status. This

program is a food and nutrition service that primarily targets low-income women, infants, and









children, who may be nutritionally at-risk, in order to provide health care referrals, information

on nutrition, and nutritious food to these women. Since this program may be a potential avenue

for women to receive advice and guidance on nutrition, weight, and fitness during pregnancy, a

model inclusive of this sub-population of women was estimated to determine if a moderating

effect of PNC could be detected among women who also received WIC services in addition to

the quantity of PNC received.

Income

To determine if there is a PNC moderating effect after stratifying by income category, and

to see if PNC is more effective for certain income groups versus others, a logistic regression was

estimated for each income sub-population. Each logit model mirrored the main PPD analysis, but

included only women from each income group. Income groups were stratified as follows:

1) Less than $10,000
2) $10,000 to $24,999
3) $25,000- $49,999
4) $50,000 or more
Weight Gain Discussion

Since discussion of weight gain during PNC, in addition to nutrition and wellness during

pregnancy, is the premise of the theory posed for this study, a logistic regression model was

estimated with a sub-population of women who answered "yes" to the following question: "Did

your health care professional discuss how much weight to gain?" The weight gain discussion

variable was selected from the PRAMS Standard Questionnaire. This model included the same

variables as the main PPD analysis (PNC, BMI, and the control variables). However, because

this additional control variable is optional for inclusion in state surveys, the sample size only

contained women from Utah and Vermont; thus, reducing the sample size and limiting the

external validity.









Results of Sub-Analyses

Sub-analyses were estimated with six logit models, each using a different sub-population

of women from the sample used originally for the main models (PPD). Table B-l presents the

chi-square results to describe the model of women who received adequate plus PNC versus

women who received other quantities of PNC. Table B-2 presents the t-test results with maternal

age and adequate plus PNC. Table B-3 presents the results from the logit model for adequate

plus PNC, including the twelve maternal morbidities that were significant (p<0.1). Table B-4

presents the results of the ordinal logistic regression model inclusive of the main effects and

control variables. Similar to the primary logistic regression model that addressed the first specific

aim, this model also showed that women who had an underweight pre-pregnancy BMI had lower

odds for PPD symptoms compared to women who had a normal pre-pregnancy BMI. Thus,

women who had a normal pre-pregnancy BMI had 11% greater odds for PPD symptoms

compared to women who had an underweight pre-pregnancy BMI (OR=0.90, p<0.05). However,

in continuing to comparing this model with the primary logistic regression model addressing the

first specific aim, the significance for women who had an underweight pre-pregnancy BMI

increased: p<0.10 versus p<0.05, respectively. In addition, the odds of PPD symptoms for

women with a normal pre-pregnancy BMI slightly decreased. The first logistic regression (logit)

model held adequate plus PNC as the dependent variable in order to determine the maternal

morbidities predictive of high-risk adequate plus PNC.

The next logit model (Table B-5) included a sub-population of women who received

services during pregnancy from Women, Infants, and Children (WIC). Results showed that for

the main effects in this model, only overweight pre-pregnancy BMI had a significant association

with PPD symptoms. However, the odds for PPD symptoms among women who had a pre-

pregnancy BMI of normal were greater by 64% compared to women who had a pre-pregnancy









BMI of overweight (OR=0.61, p=0.005). This model also demonstrated a moderating effect of

PNC for two pre-pregnancy BMI groups of women who received inadequate PNC. Women who

had an obese pre-pregnancy BMI and received adequate PNC had 59% greater odds for PPD

symptoms compared to women who had an obese pre-pregnancy BMI and received inadequate

PNC (OR=0.63, p=0095). However, the reverse moderating effect of PNC was seen for women

who had an overweight pre-pregnancy BMI and received inadequate PNC in that they had 92%

greater odds for PPD symptoms compared to women who had an overweight pre-pregnancy BMI

and received adequate PNC (OR=1.92, p=0.048). For women who received WIC services, it is

suggested that among women who had an obese pre-pregnancy BMI, compared to women who

received adequate PNC, those who received inadequate PNC were perhaps healthy and happy

with their bodies and/or pregnancy and did not see a need to seek PNC. Also, women who

received adequate PNC may have been the women who were more anxious and worried about

their pregnancy. However, for women who had an overweight pre-pregnancy BMI, the PNC may

have been beneficial along with the WIC services; thus, resulting in the higher odds for PPD

symptoms among women who received inadequate PNC and the lower odds for PPD symptoms

among women who received adequate PNC. Looking at a sub-population of women who

received WIC services in this model appeared to show a moderating effect of PNC to some

extent. Further research should address whether receiving PNC and WIC services simultaneously

during pregnancy (in which weight, nutrition, and wellness are addressed in both) has a

beneficial effect on reducing the likelihood for PPD symptoms.

The next four logit models were estimated using a sub-population of women from each

income category to see if PNC was more effective for one income category versus another.

These models were estimated in efforts to demonstrate a moderating effect of PNC after









stratifying the sample by income category. The first logit model, estimated on women with an

income of less than $10,000 (Table B-6), showed a moderating effect of PNC for women from

two pre-pregnancy BMI groups of women. Women who had an obese pre-pregnancy and

received adequate PNC had 96% greater odds for PPD symptoms compared to women who had

an obese pre-pregnancy BMI and received inadequate PNC (OR=0.51, p=0.07). On the other

hand, women who had an overweight pre-pregnancy BMI and received inadequate PNC had 2.32

times greater odds for PPD symptoms compared to women who had an overweight pre-

pregnancy BMI and received adequate PNC (OR=2.32, p=0.06). It is interesting to note that

these results remain consistent with the results shown for the logit model estimated on women

who received WIC services during their pregnancy, because WIC primarily targets low-income

women. These results also support further research to focus on women who receive WIC

services and PNC services simultaneously, to see if there is additional benefit, compared to

women who receive either one or the other.

The next logit model, estimated using a sub-population of women with an income

between $10,000 and $24,999 (Table B-7) did not show a consistent moderating effect of PNC

as shown in the two previous logit models (with sub-populations of WIC and women with an

income of less than $10,000). In fact, there was no significance for any of the main effects and

PPD symptoms, nor the interaction effects and PPD symptoms. Since this model did not show a

similar moderating effect as shown in the logit models estimated with WIC women and women

with an income of less than $10,000, it is suggested that perhaps, when considering women who

receive WIC services during pregnancy but get minimal to no PNC (inadequate PNC), the

moderating effect occurs more readily for women with a very low income as opposed to a low

income, as stratified in this study. It is surprising though that among women who received WIC









services during pregnancy (Table B-5), women with an income less than $10,000 had a higher

odds for PPD (OR=1.96, p=0.002) compared to women who had an income of $50,000 or

greater, while women with an income between $10,000 and $24,000 (OR=1.48, p=0.07)

compared to women who had an income of $50,000 or greater.

Next, in the logit model inclusive of women with an income between $25,000 and

$49,000 (Table B-8), only inadequate PNC was significant with PPD symptoms in that women

from this PNC utilization category had 72% greater odds compared to women who received

adequate PNC (OR=1.72, p=0.046). Though a moderating effect of PNC was not seen after

interacting each pre-pregnancy BMI group with each PNC utilization category, the significance

shown for inadequate PNC among this group of women with an income between $25,000 and

$49,999 suggests further studying the relationship between weight and PPD symptoms amongst

women from this income group who receive no to minimal PNC. Understanding the reasons that

play a role in explaining why some women from this income group seek no to minimal PNC may

help explain these results, especially because it appears that PNC has a beneficial effect on

women from this income group who received adequate PNC.

Finally, a logit model estimated for a sub-population of women with an income of $50,000

or greater (Table B-9) did not show significance between any of the main effects and PPD

symptoms, but a moderating effect was seen for one group of women. Women who had a pre-

pregnancy BMI of underweight and received intermediate PNC had a 2.4 times greater odds for

PPD symptoms compared to women who had a pre-pregnancy BMI of underweight and received

adequate PNC (OR=2.40, p=0.06). It is interesting to note that despite lack of significance,

women from this pre-pregnancy BMI category who received inadequate PNC had 28% greater

odds for PPD symptoms compared to women who had an underweight pre-pregnancy BMI who









received adequate PNC. These odds are lower than the women who received the next level of

PNC quantity (as the reverse was predicted). For women from this income group and from this

pre-pregnancy BMI category, it may be of worth to ascertain psychosocial reasons to explain

why it is that women in between those receive no to minimal PNC, and those who receive

adequate PNC, have more than double the odds for PPD symptoms than those who receive no to

minimal PNC. In comparing the overall results from the logit models estimated on a sub-

population of women from each income group, it seems that a moderating effect of PNC was the

most worthy to note in the model that included women with an income of less than $10,000 (very

low income).

Table B-10 presents the results from the logistic regression that looked a subpopulation of

women who had weight gain discussed during their PNC. The variables "PNC paid by military"

and "PNC paid by Native American health services" were not included in this model due to

collinearity. For example, women who answered "yes" to receiving payment for PNC from

either of these organizations also answered "yes" to having weight gain discussed by their PNC

provider, hence, resulting in collinearity where the variables (not the women) were automatically

removed from the model by Stata. Looking at the main effects, among women who had weight

gain discussed by their PNC provider, women who had a normal pre-pregnancy BMI had double

the odds of PPD symptoms compared to women who had an underweight pre-pregnancy BMI

(OR=0.48, p=0.02). However, when looking at the interaction effects among women who had an

underweight pre-pregnancy BMI, those who received inadequate PNC had about four times

greater odds for PPD symptoms (OR=4.10, p=0.01), and women who received intermediate PNC

had about 12.8 times greater odds for PPD symptoms (OR=12.79, p=0.002). To explain this

inconsistency of results, table 4-5 shows that among women who had weight gain discussed









during their PNC, 795 of women who had an underweight pre-pregnancy BMI had the

discussion versus 2,713 women with a normal pre-pregnancy BMI who had the discussion. Thus,

when comparing pre-pregnancy BMI groups, only, even with the higher number of women with

a normal pre-pregnancy BMI who had weight gain discussed, it was beneficial more-so for

women who had an underweight pre-pregnancy BMI as these women had lower odds for PPD

symptoms. However, when looking only at women who had an underweight pre-pregnancy BMI,

there were some women in which weight gain discussion was not beneficial; thus, these women

at a higher odds for PPD symptoms compared to women in that same pre-pregnancy BMI group

who received adequate PNC. It is suggested that perhaps there that women who receive

inadequate and intermediate levels of PNC may be more sensitive about their weight, which may

prompt them not to seek as much PNC as women who had an underweight pre-pregnancy BMI

and received adequate PNC. The latter may also be more proactive about gaining the right

amount of weight for their health of their baby. Hence, a selection effect may be the reason to

explain these results. Further research should seek to determine the feelings and attitudes of

women who are underweight towards receiving weight gain discussion from their PNC

providerss.









Table B-1. Chi-square analyses comparing 40 characteristics among women who utilized adequate plus PNC versus women who
utilized "other quantities of PNC"
Categorical control No for adequate n (no Yes for adequate n (yes for N P-value
variable plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)


Main variable:
Pre-pregnancy
body mass index
(BMI)
Underweight
Normal
Overweight
Obese
Demographic
control variables
Maternal race:
White
No
Yes
Maternal race:
Black
No
Yes
Maternal race:
Other
No
Yes
Hispanic
Not Hispanic
Hispanic


28,243


3,919 (13.9%)
14,882 (52.7%)
3,615 (12.8%)
5,827 (20.6%)


14,995


43,238 <0.0001*


2,045 (13.6%)
7,384 (49.2%)
1,880 (12.5%)
3,686 (24.6%)


30,099


11,288 (37.5%)
18,811 (62.5%)


25,670 (85.3%)
4,429 (14.7%)


23,240 (77.2%)
6,859 (22.8%)

23,916 (79.9%)
6,010 (20.1%)


30,099


30,099


15,701


4,912 (31.3%)
10,789 (68.7%)


13,093 (83.4%)
2,608 (16.6%)


13,397 (85.3%)
2,304 (14.7%)

13,322 (85.4%)
2,280 (14.6%)


29,926


15,701


15,701



15,602


45,800 <0.0001*



45,800 <0.0001*


45,800 <0.0001*



45,528 <0.0001*









Table B-1. Continued
Categorical control No for adequate n (no Yes for adequate n (yes for N P-value
variable plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)


Maternal education
0-8 years
9-11 years
12 years
13-15 years
16+ years
Income (12 months
prior)
Less than $10,000
$10,000 to $24,999
$25,000 to $49,999
$50,000 or more
Marital status
Married
Other
Insurance control
variables
PNC paid by
income
No
Yes
PNC paid by
insurance/HMO


30,873


1,538 (4.98%)
4,710 (15.3%)
9,425 (30.5%)
7,068 (22.9%)
8,132 (26.3%)


6,663 (21.3%)
8,987 (28.7%)
7,127 (22.7%)
8,578 (27.4%)

19,570 (62.4%)
11,775 (37.6%)


31,355


16,565


530(3.19%)
2,013 (12.2%)
5,048 (30.5%)
4,046 (24.4%)
4,928 (29.7%)


2,944 (17.6%)
4,495 (26.8%)
3,887 (23.2%)
5,422 (32.4%)

11,142 (66.6%)
5,588 (33.4%)


31,345


31,355


25,128 (80.1%)
6,227 (19.9%)


16,146 (51.5%)
15,209 (48.5%)


31,355


16,748






16,730


16,748


13,294 (79.4%)
3,454 (20.6%)


7,418 (44.3%)
9,330 (55.7%)


16,748


47,438 <0.0001*


48,103 <0.0001*






48,075 <0.0001*




48,103 0.047*



48,103 <0.0001*









Table B-1. Continued
Categorical control No for adequate n (no Yes for adequate n (yes for N P-value
variable plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)


PNC paid by
Medicaid
No
Yes
PNC paid by
military
No
Yes
PNC paid by
Native American
Health Services
No
Yes
Pregnancy and
delivery control
variables
Birthweight
<1,500 g
1,500 g to 2,499 g
2,500+ g
Smoking during
pregnancy
No
Yes
Vaginal delivery
No
Yes


31,355


18,434 (58.8%)
12,921 (41.2%)


30,520 (97.3%)
835 (2.67%)


31,355


16,748


10,143 (60.6%)
6,605 (39.4%)


16,526 (98.7%)
222 (1.32%)


31,355


30,978 (98.8%)
377 (1.20%)


16,748



16,748


48,103 <0.0001*


48,103 <0.0001*



48,103 <0.0001*


16,649 (99.4%)
99 (0.59%)


31,347


864 (27.6%)
4,555 (14.5%)
25,928 (82.7%)


27,818 (89.3%)
3,344 (10.7%)

8,581(27.4%)
22,759 (72.6%)


31,162


1,622 (9.68%)
5,664 (33.8%)
9,462 (56.5%)


14,793 (88.8%)
1,875 (11.2%)

6,542 (39.1%)
10,195 (60.9%)


31,340


16,748



16,668



16,737


48,095 <0.0001*



47,830 0.08**



48,077 0.025*









Table B-1. Continued


Categorical control
variable


No for adequate n (no Yes for adequate n (yes for N P-value
plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)
31,355 16,747 48,102 0.48


15,903 (50.7%)
15,452 (49.6%)


8,437 (50.4%)
8,310 (49.6%)


30,847


Gender of infant
Male
Female
Infant in the
intensive care unit
(ICU)
No
Yes
Pregnancy
intention
No
Yes
Breastfed (ever)
No
Yes
Alcohol
consumption in the
last three months
of pregnancy
No
Yes
Women, Infants,
and Children
during pregnancy
No
Yes


30,923


11,308 (68.6%)
5,185 (31.4%)


7,822 (47.3%)
8,701 (52.7%)

3,256 (19.8%)
13,174 (80.2%)


30,599


30,735


15,486 (94.2%)
960(5.84%)


30,907


16,403 (53.1%)
14,504 (46.9%)


16,493




16,523



16,430


16,446






16,533


47,340 <0.0001*




47,446 <0.0001*



47,029 <0.0001*


47,181 0.034*






47,440 <0.0001*


9,066 (54.8%)
7,467 (45.2%)


26,402 (85.6%)
4,445 (14.4%)


16,350 (52.9%)
14,573 (47.1%)

5,551 (18.1%)
25,048 (81.9%)




28,622 (93.1%)
2,113 (6.87%)









Table B-1. Continued
Categorical control No for adequate n (no Yes for adequate n (yes for N P-value
variable plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)


Weight gain talk
during pregnancy
No
Yes
High-risk maternal
morbidity control
variables
Diabetes before
pregnancy
No
Yes
Incompetent cervix
No
Yes
Preterm labor
No
Yes
Placenta previa or
placenta abruptio
No
Yes
Bedrest
No
Yes
Car crash injury
No
Yes


5,493


1,288 (23.4%)
4,205 (76.6%)


3,366


8,859 0.0001*


691 (20.5%)
2,675 (79.5%)


31,355


30,859 (98.4%)
496 (1.58%)

30,960 (98.7%)
395 (1.26%)

24,975 (79.7%)
6,380 (20.3%)


29,812 (95.1%)
1,543 (4.92%)

26,137 (83.4%)
5,218 (16.6%)

30,847 (98.4%)
508 (1.62%)


31,355


31,355


31,355


16,748


16,245 (97.0%)
503 (3.00%)

16,271 (97.2%)
477 (2.85%)

10,528 (62.9%)
6,220 (37.1%)


15,111 (90.2%)
1,637 (9.77%)

11,295 (67.4%)
5,453 (32.6%)

16,431 (98.1%)
317 (1.89%)


31,355


31,355


16,748


16,748


16,748



16,748


16,748


48,103 <0.0001*


48,103 <0.0001*


48,103 <0.0001*


48,103 <0.0001*



48,103 <0.0001*


48,103 0.03*









Table B-1. Continued
Categorical control No for adequate n (no Yes for adequate n (yes for N P-value
variable plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)


Blood transfusion
No
Yes
Medical risk
factors
No
Yes
Hospitalized
during pregnancy
No
Yes
Non high-risk
maternal morbidity
control variables
Gestational
diabetes
No
Yes
Kidney/bladder
infection
No
Yes
Nausea
No
Yes


31,022 (98.9%)
333 (1.06%)


21,317(68.0%)
10,038 (32.0%)


27,247 (86.9%)
4,108 (13.1%)


31,355


31,355


16,466 (98.3%)
282 (1.68%)


9,553 (57.0%)
7,195 (43.0%)


11,641 (69.5%)
5,107 (30.5%)


31,355


31,355


28,955 (92.3%)
2,400 (7.65%)


25,968 (82.8%)
5,387 (17.2%)

22,725 (72.5%)
8,630 (27.5%)


31,355


16,748


16,748



16,748


16,748


14,739 (88.0%)
2,009 (12.0%)


13,405 (80.0%)
3,343 (20.0%)

11,378 (67.9%)
5,370 (32.1%)


31,355


16,748



16,748


48,103 <0.0001*


48,103 <0.0001*



48,103 <0.0001*


48,103 <0.0001*



48,103 <0.0001*



48,103 <0.0001*









Table B-1. Continued
Categorical control No for adequate n (no Yes for adequate n (yes for N P-value
variable plus PNC adequate plus plus PNC adequate
(Frequency, %) PNC) (Frequency, %) plus PNC)
High blood 31,355 16,748 48,103 <0.0001*
pressure
No 27,736 (88.5%) 13,027 (77.8%)
Yes 3,619 (11.5%) 3,721 (22.2%)
Vaginal bleeding 31,355 16,748 48,103 <0.0001*
No 26,847 (85.6%) 12,944 (77.3%)
Yes 4,508 (14.4%) 3,804 (22.7%)
Premature rupture 31,355 16,748 48,103 <0.0001*
of membrane
(PROM)
No 29,277 (93.4%) 13,949 (83.3%)
Yes 2,078 (6.62%) 2,799 (16.7%)
Labor 31,355 16,748 48,103 0.84
abnormalities
No 25,378 (80.9%) 13,543 (80.9%)
Yes 5,977 (19.1%) 3,205 (19.1%)
Labor/delivery 31,355 16,748 48,103 <0.0001*
complications
No 20,501 (65.4%) 10,298 (61.5%)
Yes 10,854 (34.6%) 6,450 (38.5%)
The dependent variable for this table was adequate plus PNC, while the main independent variable was pre-pregnancy body mass
index (BMI). The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.
An asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a 90% confidence interval (CI).









Table B-2. Maternal age (continuous variable) and adequate plus PNC t-test results
Group Observations Mean Standard error Standard deviation 95% Confidence interval
(Adequate plus PNC) (Lower, Upper)
No 31353 27.22 .0347 6.14 (27.15, 27.28)
Yes 16746 27.96 .0481 6.22 (27.86, 28.05)
Combined 48099 27.48 .0282 6.18 (27.42, 27.53)
Difference -------- -.7374 .0591 -.8531 -.6215
The dependent variable for this table was adequate plus PNC, while the main independent variable was maternal age. The population
for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.










Table B-3. Logistic regression for adequate plus PNC to determine significant predictors of


adequate plus PNC
Dependent variable: Adequate plus PNC


Main effect independent variable: Pre-pregnancy
BMI
Normal (reference)
Obese
Overweight
Underweight
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Very low income: Less than $10,000
Low income: $10,000-$24,999
Moderate income: $25,000-$49,999
Marital status
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Alcohol during pregnancy
Women, Infants, and Children during pregnancy
Maternal morbidity control variables
Diabetes before pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
Cervix sewn shut (incompetent)
High blood pressure during pregnancy
Nausea
Preterm labor
Premature rupture of membrane (PROM)
Placenta previa or placenta abruptio
Bedrest
Car crash injury
Blood transfusion
Medical risk factors
Labor abnormalities
Labor/delivery complications
Hospitalized during pregnancy


Odds
ratio



1.00
1.09
0.95
1.12

1.00
0.89
0.71
0.80
1.05
1.01
1.00
0.74
0.80
0.86
0.92

0.52
1.02
0.85
0.76
1.21

1.75
1.35
1.13
1.12
1.35
1.36
1.01
1.46
1.49
1.16
1.47
0.81
0.79
1.17
1.07
0.96
1.26


P-value






0.07**
0.33
0.04*


0.049*
<0.0001*
<0.0001*
0.02*
0.02*

<0.0001"
<0.0001*
0.004*
0.07**

<0.0001*
0.76
<0.0001*
<0.0001*
<0.0001*

<0.0001*
<0.0001*
0.01*
0.02*
0.048*
<0.0001*
0.89
<0.0001*
<0.0001*
0.065**
<0.0001*
0.14
0.18
<0.0001*
0.15
0.35
<0.0001*


95% Confidence
interval
(Lower, Upper)


----------------
(0.994, 1.193)
(0.848, 1.056)
(1.006, 1.252)

----------------
(0.801, 0.999)
(0.638, 0.798)
(0.720, 0.898)
(1.007, 1.096)
(1.001, 1.016)
----------------
(0.635, 0.850)
(0.705, 0.901)
(0.781, 0.953)
(0.830, 1.008)

(0.486, 0.565)
(0.989, 1.160)
(0.782, 0.915)
(0.662, 0.875)
(1.100, 1.327)

(1.345, 2.284)
(1.190, 1.522)
(1.030, 1.248)
(1.017, 1.226)
(1.002, 1.821)
(1.225, 1.509)
(0.928, 1.090)
(1.328, 1.595)
(1.272, 1.736)
(0.991, 1.351)
(1.337, 1.617)
(0.609, 1.072)
(0.556,1.119)
(1.081, 1.259)
(0.977, 1.169)
(0.893, 1.041)
(1.127, 1.411)


The dependent variable for this table was adequate plus PNC, while the main independent variable was pre-
pregnancy body mass index (BMI). The population for this table included all pregnancies and the years of PRAMS
data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval (CI) and a double
asterisk corresponds to a 90% confidence interval (CI).










Table B-4. Risk-adjusted ordinal logistic regression sensitivity analysis with the main effect
independent variables, interaction effect variables, and control variables


Dependent variable: Postpartum depressive
(PPD) symptoms

Main effect independent variable: Pre-pregnancy
BMI
Normal BMI (reference)
Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate plus PNC
Overweight BMI/Adequate PNC (reference)
Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate plus PNC
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Very low income: Less than $10,000
Low income: $10,000-$24,999
Moderate income: $25,000-$49,999
Marital status
Insurance control variables
PNC paid by income (reference)


Odds
ratio


1.00
0.94
0.96
1.06


1.00
0.99
1.08
1.04


1.00
0.87
0.84
1.01
1.00
1.15
1.10
0.998
1.00
1.05
0.75
0.94

1.00
0.87
1.12
0.83
1.05
0.99
1.00
1.80
1.36
1.27
1.07

1.00


P-value 95% Confidence
interval
(Lower, Upper)


0.43
0.60
0.36




0.93
0.26
0.52




0.38
0.20
0.95

0.53
0.50
0.99

0.79
0.07**
0.64


0.03*
0.03*
0.001*
0.03*
0.09**

<0.0001*
<0.0001*
<0.0001*
0.20


----------------
(0.820, 1.087)
(0.830, 1.114)
(0.937, 1.198)


----------------
(0.829, 1.188)
(0.946, 1.229)
(0.930, 1.156)


----------------
(0.635, 1.188)
(0.645, 1.096)
(0.833, 1.216)
----------------
(0.747, 1.764)
(0.836, 1.441)
(0.782, 1.273)
----------------
(0.720, 1.540)
(0.546, 1.027)
(0.744, 1.199)

----------------
(0.775, 0.988)
(1.013, 1.241)
(0.741, 0.931)
(1.005, 1.096)
(0.987, 1.001)
----------------
(1.541, 2.106)
(1.209, 1.536)
(1.163, 1.382)
(0.967, 1.178)










Table B-4. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms


Odds P-value 95% Confidence


ratio


PNC paid by insurance/HMO 1.05 0.37
PNC paid by Medicaid 1.04 0.56
PNC paid by military 0.95 0.66
PNC paid by Native American/Alaskan HS 0.72 0.005*
Pregnancy and delivery control variables
Birthweight 1.10 0.05**
Smoking during pregnancy 0.83 0.004*
Vaginal delivery 0.94 0.11
Gender of infant 0.97 0.29
Infant in the intensive care unit (ICU) 1.13 0.06**
Pregnancy intention 0.79 <0.0001*
Breastfed 1.13 0.01*
Alcohol during pregnancy 1.48 <0.0001*
Women, Infants, and Children during pregnancy 1.02 0.68
High-risk maternal morbidity control variables
Diabetes before pregnancy 1.27 0.18
Gestational diabetes 1.19 0.01*
Vaginal bleeding 1.24 <0.0001*
Kidney/bladder infection 1.38 <0.0001*
Cervix sewn shut (incompetent) 0.75 0.10
High blood pressure during pregnancy 1.01 0.84
Preterm labor 1.38 <0.0001*
Premature rupture of membrane (PROM) 0.82 0.01 *
Placenta previa or placenta abruptio 1.08 0.33
Bedrest during pregnancy 1.11 0.05**
Medical risk factors during pregnancy 1.05 0.22
Hospitalized during pregnancy 1.05 0.39
The dependent variable for this table was postpartum depressive (PPD)


interval
(lower, upper)
(0.944, 1.169)
(0.922, 1.162)
(0.737, 1.215)
(0.566, 0.904)

(0.999, 1.204)
(0.726, 0.941)
(0.873, 1.014)
(0.904, 1.031)
(0.995, 1.283)
(0.731, 0.847)
(1.027, 1.246)
(1.318, 1.660)
(0.924, 1.129)

(0.897, 1.784)
(1.040, 1.351)
(1.130, 1.369)
(1.255, 1.524)
(0.531, 1.061)
(0.909, 1.124)
(1.262, 1.518)
(0.698, 0.953)
(0.926, 1.256)
(0.999, 1.222)
(0.973, 1.128)
(0.934, 1.191)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a
90% confidence interval (CI).









Table B-5. Logistic regression for women who received WIC services during pregnancy
Dependent variable: Postpartum depressive Odds P-value 95% Confidence
(PPD) symptoms ratio interval
Subpopulation: Women who received WIC (lower, upper)
during pregnancy
Main effect independent variable: Pre-pregnancy
BMI
Normal (reference) 1.00--- -------
Underweight 0.86 0.38 (0.608, 1.207)
Overweight 0.61 0.005* (0.424, 0.861)
Obese 0.88 0.37 (0.664, 1.164)
Main effect independent variable: PNC
utilization
Adequate (reference) 1.00--- -------
Inadequate 1.10 0.53 (0.818, 1.475)
Intermediate 1.08 0.65 (0.780, 1.489)
Adequate plus 0.97 0.80 (0.747, 1.250)
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference) 1.00--- ----
Obese BMI/Inadequate PNC 0.63 0.095** (0.370, 1.083)
Obese BMI/Intermediate PNC 1.27 0.37 (0.755, 2.121)
Obese BMI/Adequate plus PNC 0.97 0.88 (0.647, 1.449)
Overweight BMI/Adequate PNC (reference) 1.00--- ------
Overweight BMI/Inadequate PNC 1.92 0.048* (1.005, 3.653)
Overweight BMI/Intermediate PNC 1.23 0.51 (0.655, 2.327)
Overweight BMI/Adequate plus PNC 1.54 0.11 (0.912, 2.602)
Underweight BMI/Adequate PNC (reference) 1.00--- ------
Underweight BMI/Inadequate PNC 1.20 0.55 (0.657, 2.206)
Underweight BMI/Intermediate PNC 0.97 0.94 (0.491, 1.927)
Underweight BMI/Adequate plus PNC 0.93 0.79 (0.543, 1.595)
Demographic control variables
Maternal race: White (reference) 1.00--- -------
Maternal race: Black 1.32 0.004* (1.095, 1.591)
Maternal race: Other 1.47 <0.0001* (1.214, 1.767)
Hispanic ethnicity 0.96 0.68 (0.787, 1.168)
Maternal education 0.94 0.16 (0.865, 1.024)
Maternal age 0.99 0.07** (0.973, 1.001)
Higher income: $50,000 or more (reference) 1.00--- -------
Very low income: Less than $10,000 1.96 0.002* (1.271, 3.012)
Low income: $10,000-$24,999 1.48 0.07** (0.970, 2.247)
Moderate income: $25,000-$49,999 1.36 0.16 (0.889, 2.096)
Marital status 1.02 0.80 (0.867, 1.204)









Table B-5. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms
Subpopulation: Women who received WIC
during pregnancy
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
High-risk maternal morbidity control variables
Diabetes before pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
Cervix sewn shut (incompetent)
High blood pressure during pregnancy
Preterm labor


Odds
ratio


1.00
0.96
0.98
1.06
0.71

1.15
0.81
0.84
0.99
1.43
0.94
0.89
1.42


P-value 95% Confidence
interval
(lower, upper)


0.71
0.82
0.84
0.10

0.09**
0.03*
0.03*
0.88
0.002*
0.44
0.16
0.03*


1.12 0.62
1.11 0.39
1.49 <0.0001*
1.45 <0.0001*
1.04 0.88
0.92 0.43
1.49 <0.0001*


Premature rupture of membrane (PROM) 0.69 0.01 *
Placenta previa or placenta abruptio 1.02 0.91
Bedrest during pregnancy 1.17 0.11
Medical risk factors during pregnancy 1.21 0.01
Hospitalized during pregnancy 1.07 0.519
The dependent variable for this table was postpartum depressive (PPD)


----------------
(0.789, 1.175)
(0.813, 1.177)
(0.610, 1.831)
(0.466, 1.069)

(0.979, 1.356)
(0.666, 0.982)
(0.715, 0.985)
(0.859, 1.139)
(1.144, 1.783)
(0.799, 1.104)
(0.757, 1.047)
(1.037, 1.937)

(0.712, 1.763)
(0.871, 1.423)
(1.232, 1.802)
(1.237, 1.706)
(0.623, 1.743)
(0.750, 1.131)
(1.253, 1.760)
(0.519, 0.917)
(0.747, 1.385)
(0.968, 1.405)
(1.039, 1.404)
(0.870, 1.317)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a
90% confidence interval (CI).










Table B-6. Logistic regression for women with very low income
Dependent variable: Postpartum depressive Odds P-value
(PPD) symptoms ratio
Subpopulation: Women with very low income
(less than $10,000)
Main effect independent variable: Pre-pregnancy
BMI
Normal (reference) 1.00 ---


Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate plus PNC
Overweight BMI/Adequate PNC (reference)
Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate plus PNC
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Marital status
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS


0.98
0.89
0.88


1.00
1.12
0.996
1.05


1.00
0.51
1.49
0.97
1.00
2.32
0.75
1.08
1.00
0.93
0.65
0.81

1.00
1.18
1.16
0.89
1.02
1.01
0.996

1.00
0.76
0.87
0.36
0.84


0.92
0.67
0.57



0.59
0.99
0.81



0.07**
0.29
0.93

0.06**
0.56
0.86

0.85
0.37
0.58


0.22
0.29
0.42
0.74
0.62
0.98


0.14
0.36
0.02*
0.60


95% Confidence
interval
(lower, upper)



----------------
(0.595, 1.600)
(0.517, 1.531)
(0.574, 1.356)


(0.749, 1.664)
(0.634, 1.566)
(0.716, 1.535)


(0.246, 1.049)
(0.712, 3.121)
(0.526, 1.801)
----------------
(0.979, 5.504)
(0.283, 1.971)
(0.468, 2.469)
----------------
(0.415, 2.060)
(0.258, 1.646)
(0.371, 1.748)

----------------
(0.907, 1.537)
(0.886, 1.505)
(0.670, 1.183)
(0.907, 1.148)
(0.985, 1.026)
(0.778, 1.276)

----------------
(0.530, 1.091)
(0.652, 1.169)
(0.148, 0.874)
(0.436, 1.614)









Table B-6. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms
Subpopulation: Women with very low income
(less than $10,000)
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children
High-risk maternal morbidity control variables
Diabetes before pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
Cervix sewn shut (incompetent)
High blood pressure during pregnancy
Preterm labor


Odds
ratio


1.04
0.89
0.89
0.91
1.32
1.08
0.85
0.96
0.998


P-value 95% Confidence
interval
(lower, upper)


0.74
0.38
0.32
0.39
0.09**
0.55
0.16
0.84
0.99


2.14 0.007*
0.87 0.42
1.18 0.25
1.53 <0.0001*
0.98 0.94
0.81 0.175
1.30 0.03*


Premature rupture of membrane (PROM) 0.77 0.17
Placenta previa or placenta abruptio 1.08 0.73
Bedrest during pregnancy 1.33 0.02*
Medical risk factors during pregnancy 1.15 0.20
Hospitalized during pregnancy 0.96 0.75
The dependent variable for this table was postpartum depressive (PPD)


(0.832, 1.294)
(0.689, 1.151)
(0.702, 1.121)
(0.747, 1.120)
(0.955, 1.812)
(0.845,1.370)
(0.670, 1.066)
(0.644, 1.427)
(0.755, 1.321)

(1.237, 3.706)
(0.611, 1.229)
(0.892, 1.554)
(1.217, 1.928)
(0.524, 1.824)
(0.605, 1.095)
(1.030, 1.650)
(0.523, 1.120)
(0.703, 1.652)
(1.040, 1.710)
(0.928, 1.433)
(0.728, 1.258)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a
90% confidence interval (CI).









Table B-7. Logistic regression for women with low income
Dependent variable: Postpartum depressive Odds
(PPD) symptoms ratio
Subpopulation: Women with low income
($10,000 to $24,999)
Main effect independent variable: Pre-pregnancy
BMI
Normal (reference) 1.00


Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate plus PNC
Overweight BMI/Adequate PNC (reference)
Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate plus PNC
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Marital status
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS


0.77
0.69
0.90


1.00
0.91
1.28
0.92


1.00
0.97
0.98
1.22
1.00
1.26
0.74
1.45
1.00
1.44
0.75
1.32

1.00
1.22
1.74
1.08
0.96
0.98
1.02

1.00
1.14
1.19
1.01
0.32


P-value 95% Confidence
interval
(lower, upper)


0.26
0.11
0.58



0.66
0.23
0.62



0.94
0.95
0.47

0.58
0.47
0.27

0.39
0.56
0.42


0.13
<0.0001*
0.57
0.46
0.06**
0.83


0.27
0.16
0.98
0.001*


(0.485, 1.214)
(0.430, 1.095)
(0.613, 1.316)


(0.610, 1.370)
(0.855, 1.928)
(0.662, 1.277)


(0.472, 2.001)
(0.489, 1.962)
(0.715, 2.082)
----------------
(0.558, 2.857)
(0.331, 1.657)
(0.746, 2.806)
----------------
(0.630, 3.284)
(0.287, 1.957)
(0.673, 2.598)

----------------
(0.941, 1.599)
(1.368, 2.222)
(0.832, 1.401)
(0.859, 1.071)
(0.963, 1.001)
(0.832, 1.257)

----------------
(0.901, 1.453)
(0.936, 1.512)
(0.489, 2.091)
(0.167, 0.630)









Table B-7. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms
Subpopulation: Women with low income
($10,000 to $24,999)
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children
High-risk maternal morbidity control variables
Diabetes before pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
Cervix sewn shut (incompetent)
High blood pressure during pregnancy
Preterm labor


Odds
ratio


1.21
0.69
0.83
1.11
1.41
0.81
0.85
1.78
0.99


P-value 95% Confidence
interval
(lower, upper)


0.08**
0.005*
0.07**
0.26
0.02*
0.047*
0.14
0.003*
0.95


0.88 0.74
1.36 0.06**
1.31 0.03*
1.46 0.001*
0.76 0.43
1.07 0.61
1.43 0.003*


Premature rupture of membrane (PROM) 0.76 0.21
Placenta previa or placenta abruptio 0.83 0.37
Bedrest during pregnancy 1.28 0.05**
Medical risk factors during pregnancy 1.10 0.33
Hospitalized during pregnancy 1.14 0.37
The dependent variable for this table was postpartum depressive (PPD)


(0.978, 1.506)
(0.536, 0.896)
(0.671, 1.016)
(0.924, 1.333)
(1.069, 1.872)
((0.654, 0.997)
(0.676, 1.058)
(1.210, 2.627)
(0.795, 1.240)

(0.417, 1.867)
(0.988, 1.877)
(1.024, 1.688)
(1.172, 1.811)
(0.386, 1.497)
(0.817, 1.409)
(1.129, 1.804)
(0.487, 1.175)
(0.557, 1.242)
(0.998, 1.649)
(0.904, 1.345)
(0.860, 1.505)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a
90% confidence interval (CI).










Table B-8. Logistic regression for women with moderate income
Dependent variable: Postpartum depressive Odds P-value
(PPD) symptoms ratio
Subpopulation: Women with moderate income
($25,000 to $49,999)
Main effect independent variable: Pre-pregnancy
BMI
Normal (reference) 1.00 ---
Underweight 0.87 0.61
Overweight 0.999 0.99
Obese 0.92 0.68
Main effect independent variable: PNC
utilization
Adequate (reference) 1.00 ---
Inadequate 1.72 0.046*
Intermediate 0.76 0.28
Adequate plus 1.06 0.73
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference) 1.00 ---
Obese BMI/Inadequate PNC 0.72 0.51
Obese BMI/Intermediate PNC 1.46 0.36
Obese BMI/Adequate plus PNC 0.90 0.73
Overweight BMI/Adequate PNC (reference) 1.00 ---
Overweight BMI/Inadequate PNC 0.66 0.50
Overweight BMI/Intermediate PNC 1.12 0.82
Overweight BMI/Adequate plus PNC 1.17 0.69
Underweight BMI/Adequate PNC (reference) 1.00 ---
Underweight BMI/Inadequate PNC 1.06 0.92
Underweight BMI/Intermediate PNC 1.75 0.33
Underweight BMI/Adequate plus PNC 0.81 0.63
Demographic control variables
Maternal race: White (reference) 1.00 ---
Maternal race: Black 1.77 0.001*
Maternal race: Other 1.53 0.01*
Hispanic ethnicity 1.02 0.93
Maternal education 0.79 0.001*
Maternal age 0.995 0.68
Marital status 0.94 0.66
Insurance control variables
PNC paid by income (reference) 1.00 ---
PNC paid by insurance/HMO 0.85 0.30
PNC paid by Medicaid 1.13 0.48
PNC paid by military 1.27 0.42


95% Confidence
interval
(lower, upper)



----------------
(0.504, 1.498)
(0.599, 1.670)
(0.617, 1.373)


----------------
(1.009, 2.934)
(0.459, 1.251)
(0.744, 1.522)


----------------
(0.277, 1.887)
(0.651, 3.256)
(0.496, 1.632)
----------------
(0.194, 2.227)
(0.420, 2.990)
(0.540, 2.535)
----------------
0.360, 3.124)
(0.571, 5.332)
(0.345, 1.906)

----------------
(1.247, 2.502)
(1.104, 2.109)
(0.712, 1.451)
(0.688, 0.901)
(0.973, 1.018)
(0.694, 1.262)

----------------
(0.628, 1.155)
(0.801, 1.600)
(0.713, 2.267)









Table B-8. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms
Subpopulation: Women with moderate income
($25,000 to $49,999)
PNC paid by Native American/Alaskan HS
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children
High-risk maternal morbidity control variables
Diabetes before pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
Cervix sewn shut (incompetent)
High blood pressure during pregnancy
Preterm labor


Odds
ratio


P-value 95% Confidence
interval
(lower, upper)


1.21 0.53


0.96
0.79
0.94
1.09
1.18
0.77
1.02
1.06
1.07


0.75
0.23
0.64
0.46
0.36
0.03*
0.88
0.82
0.64


1.52 0.33
1.02 0.92
1.63 0.001*
1.34 0.04*
1.60 0.41
1.01 0.95
1.47 0.006*


Premature rupture of membrane (PROM) 0.59 0.03 *
Placenta previa or placenta abruptio 1.13 0.64
Bedrest during pregnancy 0.84 0.25
Medical risk factors during pregnancy 1.14 0.64
Hospitalized during pregnancy 1.10 0.61
The dependent variable for this table was postpartum depressive (PPD)


(0.662, 2.229)

(0.723, 1.263)
(0.538, 1.163)
(0.732, 1.211)
(0.873, 1.353)
(0.827, 1.690)
(0.607, 0.971)
(0.761, 1.375)
(0.634, 1.783)
(0.809, 1.412)

(0.661, 3.496)
(0.675, 1.545)
(1.222, 2.161)
(1.021, 1.762)
(0.526, 4.883)
(0.735, 1.387)
(1.114, 1.937)
(0.371, 0.941)
(0.682, 1.865)
(0.618, 1.134)
(0.682, 1.865)
(0.763, 1.590)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI).










Table B-9. Logistic regression for women with higher income
Dependent variable: Postpartum depressive Odds P-value 95% Confidence
(PPD) symptoms ratio interval
Subpopulation: Women with higher income (lower, upper)
($50,000 or more)
Main effect independent variable: Pre-pregnancy
BMI
Normal (reference) 1.00--- -------


Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate Plus PNC
Overweight BMI/Adequate PNC (reference)
Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate Plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate Plus PNC
Demographic control variables
Maternal race: White (reference)
Maternal race: Black
Maternal race: Other
Hispanic ethnicity
Maternal education
Maternal age
Marital status
Insurance control variables
PNC paid by income (reference)
PNC paid by insurance/HMO
PNC paid by Medicaid
PNC paid by military
PNC paid by Native American/Alaskan HS


0.78
0.96
0.90


1.00
0.72
0.98
1.32


1.00
1.23
1.37
0.84
1.00
1.40
1.89
1.01
1.00
1.28
2.40
1.29

1.00
1.90
1.55
0.92
0.84
0.99
1.34

1.00
0.92
1.22
0.30
0.66


0.37
0.88
0.68


0.45
0.95
0.14



0.76
0.51
0.62

0.70
0.22
0.98

0.74
0.06*
0.55


0.008*
0.008*
0.75
0.02*
0.30
0.28


0.68
0.57
0.02*
0.33


(0.460, 1.334)
(0.527, 1.736)
(0.553, 1.475)


(0.303, 1.689)
(0.604, 1.602)
(0.912, 1.919)


(0.315, 4.830)
(0.538, 3.512)
(0.426, 1.662)
----------------
(0.253, 7.690)
(0.685, 5.199)
(0.412, 2.486)
----------------
(0.297, 5.524)
(0.967, 5.975)
(0.558, 2.970)

----------------
(1.180, 3.053)
(1.123, 1.235)
(0.563, 1.509)
(0.714, 0.976)
(0.958, 1.013)
(0.786, 2.275)

----------------
(0.611, 1.376)
(0.613, 2.434)
(0.110, 0.801)
(0.290, 1.516)









Table B-9. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms
Subpopulation: Women with higher income
($50,000 or more)
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Intensive care unit
Pregnancy intention
Breastfed
Alcohol during pregnancy
Women, Infants, and Children
High-risk maternal morbidity control variables
Diabetes before pregnancy
Gestational diabetes
Vaginal bleeding
Kidney/bladder infection
Cervix sewn shut (incompetent)
High blood pressure during pregnancy
Preterm labor


Odds
ratio


0.92
1.09
1.08
1.13
1.15
0.67
1.28
1.36
0.99


P-value 95% Confidence
interval
(lower, upper)


0.58
0.81
0.58
0.32
0.55
0.003*
0.23
0.12
0.97


0.75 0.64
1.70 0.02*
0.96 0.78
1.46 0.04*
0.82 0.67
0.98 0.92
1.64 0.003*


Premature rupture of membrane (PROM) 0.64 0.08**
Placenta previa or placenta abruptio 1.25 0.35
Bedrest during pregnancy 1.07 0.72
Medical risk factors during pregnancy 0.97 0.81
Hospitalized during pregnancy 0.91 0.65
The dependent variable for this table was postpartum depressive (PPD)


(0.673, 1.247)
(0.549, 2.165)
(0.827, 1.405)
(0.889, 1.440)
(0.726, 1.821)
(0.518, 0.869)
(0.856, 1.921)
(0.923, 2.004)
(0.566, 1.734)

(0.230, 2.451)
(1.108, 2.613)
(0.707, 1.298)
(1.022, 2.098)
(0.315, 2.105)
(0.669, 1.438)
(1.179, 2.280)
(0.391, 1.059)
(0.783, 1.994)
(0.752, 1.508)
(0.741, 1.263)
(0.603, 1.371)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a
90% confidence interval (CI).










Table B-10. Logistic regression for women who received weight gain discussion
Dependent variable: Postpartum depressive Odds P-value 95% Confidence
(PPD) symptoms ratio interval
Subpopulation: Women who received weight (lower, upper)
gain discussion during PNC
Main effect independent variable: Pre-pregnancy
BMI
Normal (reference) 1.00--- -------


Underweight
Overweight
Obese
Main effect independent variable: PNC
utilization
Adequate (reference)
Inadequate
Intermediate
Adequate plus
Interaction effect variables: Pre-pregnancy
BMI/PNC utilization
Obese BMI/Adequate PNC (reference)
Obese BMI/Inadequate PNC
Obese BMI/Intermediate PNC
Obese BMI/Adequate plus PNC
Overweight BMI/Adequate PNC (reference)
Overweight BMI/Inadequate PNC
Overweight BMI/Intermediate PNC
Overweight BMI/Adequate plus PNC
Underweight BMI/Adequate PNC (reference)
Underweight BMI/Inadequate PNC
Underweight BMI/Intermediate PNC
Underweight BMI/Adequate plus PNC
Demographic control variables
Maternal Race: White (reference)
Maternal Race: Black
Maternal Race: Other
Hispanic ethnicity
Maternal education
Maternal age
Higher income: $50,000 or more (reference)
Income: Very low income: Less than $10,000
Income: Low income: $10,000-$24,999
Income: Moderate income: $25,000-$49,999
Marital status
Insurance control variables
PNC paid by income (reference)


0.48
0.61
1.59


1.00
0.90
0.60
1.40


1.00
0.67
1.47
0.48
1.00
0.81
1.86
0.79
1.00
4.10
12.79
1.65

1.00
0.75
1.86
0.68
0.79
1.01
1.00
1.51
1.24
1.11
1.58

1.00


0.02*
0.19
0.12


0.75
0.29
0.18



0.50
0.63
0.14

0.81
0.63
0.74

0.01*
0.002*
0.31


0.71
0.049
0.13
0.01*
0.70

0.25
0.48
0.68
0.03*


(0.258, 0.908)
(0.293, 1.267)
(0.892, 2.830)


(0.475, 1.708)
(0.233, 1.543)
(0.858, 2.280)


(0.212, 2.143)
(0.304, 7.103)
(0.177, 1.287)
----------------
(0.142, 4.642)
(0.148, 23.35)
(0.193, 3.205)
----------------
(1.409, 11.95)
(2.616, 62.50)
(0.624, 4.376)

----------------
(0.162, 3.452)
(1.002, 3.438)
(0.418, 1.117)
(0.653, 0.945)
(0.971, 1.044)
----------------
(0.745, 3.061)
(0.674, 2.296)
(0.673, 1.842)
(1.043, 2.385)









Table B-10. Continued
Dependent variable: Postpartum depressive
(PPD) symptoms
Subpopulation: Women who received weight
gain discussion during their PNC
PNC paid by insurance/HMO
PNC paid by Medicaid
Pregnancy and delivery control variables
Birthweight
Smoking during pregnancy
Vaginal delivery
Gender of infant
Infant in the intensive care unit (ICU)


Odds
ratio


P-value 95% Confidence
interval
(lower, upper)


0.98 0.93
1.63 0.05**


0.91
0.82
0.79
0.79
0.84


0.61
0.47
0.20
0.14
0.52


Pregnancy intention 0.61 0.005*
Breastfed 0.75 0.21
Alcohol during pregnancy 0.81 0.57
Women, Infants, and Children during pregnancy 0.64 0.02*
High-risk maternal morbidity control variables
Diabetes before pregnancy 2.76 0.096**
Gestational diabetes 0.85 0.64
Vaginal bleeding 1.05 0.81
Kidney/bladder infection 1.65 0.006*
Cervix sewn shut (incompetent) 0.85 0.86
High blood pressure during pregnancy 0.84 0.48
Preterm labor 1.10 0.63
Premature rupture of membrane (PROM) 0.63 0.13
Placenta previa or placenta abruptio 1.48 0.19
Bedrest during pregnancy 1.31 0.17
Medical risk factors during pregnancy 0.73 0.05**
Hospitalized during pregnancy 1.04 0.88
The dependent variable for this table was postpartum depressive (PPD)


(0.612, 1.565)
(0.997, 2.651)

(0.646, 1.293)
(0.470, 1.419)
(0.544, 1.135)
(0.579, 1.076)
(0.488, 1.437)
(0.427, 0.859)
(0.480, 1.174)
(0.392, 1.671)
(0.434, 0.935)

(0.834, 9.148)
(0.425, 1.697)
(0.696, 1.590)
(1.152, 2.376)
(0.131, 5.485)
(0.5151, 1.368)
(0.753, 1.599)
(0.3445, 1.152)
(0.825, 2.671)
(0.888, 1.925)
(0.530, 1.001)
(0.604, 1.796)
symptoms, while the


main independent variables were Pre-pregnancy body mass index (BMI), prenatal care (PNC)
utilization, pre-pregnancy BMI/PNC utilization interaction terms. The population for this table
included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An
asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a
90% confidence interval (CI).









LIST OF REFERENCES


Aberdour, S. (2006). Sugar, obesity, and depression: Cause and effect? Alive: Canadian Journal
of Health & Nutrition, 288, 82-83.

Adewuya, A.O., Fatoye, F.O., Ola, B.A., Ijaodola, O.R., & Ibigbami, S-M, O. (2005).
Sociodemographic and obstetric risk factors for postpartum depressive symptoms in
Nigerian women. Journal of Psychiatric Practice, 11, 353-358.

Affonso, D.D., & Mayberry, L.J. (1990). Common stressors reported by a group of childbearing
American women. Health Care for Women International, 11, 331-345.

Albright, A. (1993). Postpartum depression: An overview. Journal of Counseling &
Development, 71, 316-320.

Alexander G.R., & Kotelchuck, M. (1996). Quantifying the adequacy of prenatal care: A
comparison of indices. Public Health Reports, 111, 408-418.

Alexander, G.R., & Kotelchuck, M. (2001). Assessing the role and effectiveness of prenatal care:
History, challenges, and directions for future research. Public Health Reports, 116, 306-
316.

Alkazaleh, F., Thomas, M., Grenbenyuk, J., Glaude, L., Savage, D., Johnnnesen, J., Caetano, M.,
& Windrim, R. (2004). What women want: Women's preferences of caregiver behavior
when prenatal sonography findings are abnormal. Ultrasound in Obstetrics & Gynecology,
23, 56-62.

Allison, K.C., Sarwer, D.B., & Pare, E. (2007). Issues related to weight management during
pregnancy among overweight and obese women. Expert Reviews in Obstetrics and
Gynecology, 2, 249-254.

Alvik, A., Haldorsen, T., & Lindermann, R. (2006). Alcohol consumption, smoking and
breastfeeding in the first six months after delivery. Acta Paediatrica, 95, 686-693.

American Academy of Pediatrics and the American College of Obstetrics and Gynecologists.
(1992). Postpartum and follow-up care. In R.K. Freeman & R.L. Poland (Eds.), Guidelines
for Perinatal care (3rd ed.). (pp. 91-116).

American College of Obstetricians and Gynecologists. (2005). ACOG Issues Guidance to Ob-
Gyns on Impact of Obesity During Pregnancy. Retrieved October 24, 2007, from
http://www.acog.org/fromhome/publications/pressreleases/nr08-31-05-2. cfm

Andersen, R.M. (1995). Revisiting the behavioral model and access to medical care: Does it
matter? Journal of Health and Social Behavior, 36, 1-10.

Andersson, L., Sundstrom-Poromaa, I., Wulff, M., Astrom, M., & Bixo, M. (2006). Depression
and anxiety during pregnancy and six months postpartum: A follow-up study. Acta
Obstetricia et Gynecologica Scandinavica, 85, 937-944.









Appolonio, C.K.K., & Fingerhut, R. (2008). Postpartum depression in a military sample. Military
Medicine, 173, 1085-1091.

Atrash, H., Jack, B.W., Johnson, K., Coonrod, D.V., Moos, M-K., Stubblefield, P.G., et al.
(2008). Where is the "W"omen in MCH? American Journal of Obstetrics and Gynecology,
199, S259-S265.

Baeten, J.M., Bukusi, E.A., & Lambe, M. (2001). Pregnancy complications and outcomes
among overweight and obese nulliparous women. American Journal of Public Health, 91,
436-440.

Baldwin, L-M., Raine, T., Jenkins, L.D., Hart, G., & Rosenblatt, R. (1994). Do providers adhere
to the ACOG standards? The case of prenatal care. Obstetrics & Gynecology, 84, 549-556.

Balon, R. (2006). Mood, anxiety, and physical illness: Body and mind, or mind and body?
Depression andAnxiety, 23, 377-387.

Baric, L., & MacArthur, C. (1977). Health norms in pregnancy. British Journal of Preventive
Social Medicine, 31, 30-38.

Battle, E.K., & Brownell, K.D. (1996). Confronting a rising tide of eating disorders and obesity:
Treatment vs. prevention and policy. Addictive Behaviors, 21, 755-765.

Beck, C.T. (2001). Predictors of postpartum depression. Nursing Research, 50, 275-285.

Beine, K., Fullerton, J., Palinkas, L., & Anders, B. (1995). Conceptions of PNC among Somali
women in San Diego. Journal ofNurse-Midwifery, 40, 376-381.

Beresford, S. A. A., Curry, S.J., Kristal, A., Lazovich, D., Feng, Z., & Wagner, E.H. 1997. A
dietary intervention in primary care practice: The eating patterns study. American Journal
of Public Health, 87, 610- 616.

Birdsall, K.M., Vyas, S., Khazaezadeh, N., & Oteng-Ntim, E. (2009). Maternal obesity: A
review of interventions. The International Journal of Clinical Practice, 63, 494-507.

Bloch, M., Schmidt, P.J., Danaceau, M., Murphy, J., Neiman, L., & Rubinow, D.R. (2000).
Effects of gonadal steroids in women with a history of postpartum depression. American
Journal ofPsychiatry, 157, 924-930.

Boardley, D.J., Sargent, R.G., Coker, A.L., Hussey, J.R., & Sharpe, P.A. (1995). The
relationship between diet, activity, and other factors, and postpartum weight change by
race. Obstetrics & Gynecology, 86, 834-838.

Boney, C.M., Verma, A., Tucker, R., & Vohr, B.R. (2005). Metabolic syndrome in childhood:
Association with birth weight, maternal obesity, and gestational diabetes mellitus.
Pediatrics, 115, e290-e296.









Boyce, P., & Hickey, A. (2005). Psychosocial risk factors to major depression after childbirth.
Social Psychiatry and Psychiatric Epidemiology, 40, 605-612.

Brewerton, T.D. (1995). Toward a unified theory of serotonin dysregulation in eating and related
disorders. Psychoneuroendocrinology, 20, 561-590.

Bronner, Y., & Baldwin, K. (1999). The nutritional status and needs of women of reproductive
age. In H. Grason, J. Hutchins, & G. Silver (Eds.), Charting a course for the future of
women's andperinatal health (pp. 183-207). Timonium, MD: Printing Corporation of
America.

Brown, P.J. (1991). Culture and the evolution of obesity. Human Nature, 2, 31-57.

Cameron, R.P., Grabill, C.M., Hobfoll, S.E., Crowther, J.H., Ritter, C., & Lavin, J. (1996).
Weight, self-esteem, ethnicity, and depressive symptomatology during pregnancy among
inner-city women. Health Psychology, 15, 293-297.

Campbell, S.B. & Cohn, J. F. (1991). Prevalence and correlates ofpostpartum depression in first
time mothers. Journal ofAbnormal Psychology, 100, 594-599.

Carmichael, S., Abrams, B., & Selvin, B. (1997) The pattern of maternal weight gain in women
with good pregnancy outcomes. American Journal of Public Health, 87, 1984-1988.

Carpenter, K.M., Hasin, D.S., Allison, D.B., & Faith, M.S. (2000). Relationship between obesity
and DSM-IV major depressive disorder, suicide ideation, and suicide ideation and suicide
attempts: Results from a general population study. American Journal ofPublic Health, 90,
251-257.

Carter, A.S., Wood Baker, C., & Brownell, K.D. (2000). Body mass index, eating attitudes, and
symptoms of depression and anxiety in pregnancy and the postpartum period.
Psychosomatic Medicine, 62, 264-200.

Cedergren, M.I. (2004). Maternal morbid obesity and the risk of adverse pregnancy outcomes.
Obstetrics & Gynecology, 103, 219-224.

Centers for Disease Control. (2005). Appendixes: PRAMS 1995 Surveillance Report. Retrieved
April 21, 2009, from http://www.cdc.gov/PRAMS/PDFs/95pramsapp.pdf

Centers for Disease Control. BM- Body Mass Index: About BMI for Adults. Retrieved July 25,
2007, from
http://www.cdc.gov/nccdphp/dnpa/bmi/adultBMI/aboutadultBMI.htm#Interpreted

Centers for Disease Control. (2007). Pregnancy Risk Assessment Monitoring System (PRAMS).
Retrieved July 24, 2007, from http://www.cdc.gov/prams/

Centers for Disease Control. (2007). Pregnancy Risk Assessment Monitoring System (PRAMS):
Phase 5 Core Questionnaire.









Centers for Disease Control. (2007). Pregnancy Risk Assessment Monitoring System (PRAMS):
Phase 5 Standard Questionnaire.

Centers for Disease Control. (2009). [Pregnancy Risk Assessment Monitoring System (PRAMS):
2004 & 2005 response rates]. Unpublished raw data.

Chauhan, S.P., & Henrichs, C.E. (2003). Obesity in pregnancy: Risks and interventions by
gestational stage. OBG Management, 15, 27-38.

Chan, S.W-C., Levy, V., Chung, T.K.H., & Lee, D. (2002). A qualitative study of the
experiences of a group of Hong Kong Chinese women diagnosed with postnatal
depression. Journal ofAdvanced Nursing, 39, 571-579.

Chism, D.M. (1997). Part 1: Introduction: What makes a pregnancy high-risk? In The high-risk
pregnancy sourcebook (pp. 3-6). Los Angeles, CA: Lowell House.

Chomitz, V.R., Cheung, L.W.Y., & Lieberman, E. (1995). The role of lifestyle in preventing
low birth weight. The Future of Children, 5, 121-138.

Clayton, A.H. (2004). Considerations in women's mental health: Postpartum mood disorders.
Primary Psychiatry, 11, 20-21.

Clearly, P.D., & McNeil, B.J. (1988). Patient satisfaction as an indicator of quality care. Inquiry,
25, 25-36.

Cnattingius, S., Bergstrom, R., Lipworth, L., & Kramer, M.S. (1998). Prepregnancy weight and
the risk of adverse pregnancy outcomes. The New England Journal ofMedicine, 338, 147-
152.

Cogswell, M.E., Scanlon, K.S., Fein, S.B., & Schieve, L.A. (1999). Medically advised, mother's
personal target, and actual weight gain during pregnancy. Obstetrics & Gynecology, 94,
616-622.

Colditz, G.A. (1992). Economic costs of obesity. American Journal of ClinicalNutrition, 55,
503S-507S.

Conway, K.S., & Kutinova, A. (2006). Maternal health: Does prenatal care make a difference?
Health Economics, 15, 461-488.

Covington, M.T., & Rice, R.J. (1997). Receipt of recommended prenatal interventions and birth
weight among African-American women: Analysis of data from the 1988 National
Maternal and Infant Health Survey. Maternal and Child Health Journal, 1, 157-164.

Crerand, C.E., Wadden, T., Foster, G., & Gary, D. (2007). Changes in obesity-related attitudes in
women seeking weight reduction. Obesity, 15, 740-747.









D'Angelo, D., Williams, L., Morrow, B., Cox, S., Harris, N., Harrison, L., Posner, S.F., Hood,
J.R., & Zapata, L. (2007). Preconception and interconception health status of women who
recently gave birth to a live-born infant-Pregnancy Risk Assessment Monitoring System
(PRAMS), United States, 26 reporting areas, 2004. Morbidity andMortality Weekly
Report, 56, 1-35.

Da Costa, M. Dritsa, N. Rippen, I. Lowensteyn, & S. Khalife (2006). Health-related quality of
life in postpartum depressed women. Archives of Women's MentalHealth, 9, 95-102.

Daniels, P. Fuji Noe, G., & Mayberry, R. (2006). Barriers to prenatal care among Black women
of low socioeconomic status. American Journal of Health Behavior, 30, 188-198.

Davies, K., & Wardle, J. (1994). Body image and dieting in pregnancy. Journal of
Psychosomatic Research, 28, 787-799.

Deater-Deckard, K., Pickering, K., Dunn, J.F., & Golding, J. (1998). Family structure and
depressive symptoms in men preceding and following the birth of a child. American
Journal ofPsychiatry, 155, 818-823.

De Garine, I. & Pollock, N.J. (1995). Introduction: An overview of obesity issues across several
cultures. In SocialAspects of Obesity (pp. xiii-xxxiii). Luxembourg: Gordon & Breach.

Dennis, C-L. (2005). Psychosocial and psychological interventions for prevention of postnatal
depression: Systematic review. British Medical Journal, 331, 15-18.

Dennis, C-L., & Ross, L (2006). Women's perceptions of partner support and conflict in the
development of postpartum depressive symptoms. Journal of Advanced Nursing, 56, 588-
599.

Dietl, J. (2005). Maternal obesity and complications during pregnancy. Journal ofPerinatal
Medicine, 33, 100-105.

Dietz, P.M., Williams, S.V., Callaghan, W.M., Bachman, D.J., Whitlock, E.P., & Hornbrook,
M.C. (2007). Clinically identified maternal depression before, during, and after
pregnancies ending in live births. American Journal ofPsychiatry, 164, 1515-1520.

Dindar, I., & Erdogan, S. (2007). Screening of Turkish women for postpartum depression within
the first postpartum year: The risk profile of a community sample. Public Health Nursing,
24, 176-183.

Dixon, J.B., Dixon, M.E., & O'Brien, P.E. (2003). Depression in association with severe obesity.
Archives of internal Medicine, 163, 2058-2065.

Dong, C., Sanchez, L.E., & Price, R.A. (2004). InternationalJournal of Obesity, 28, 790-795.

El-Kak, F., Chaaya, M., Campbell, O., & Kaddour, A. (2004). Patterns of antenatal care in low-
versus high-risk pregnancies in Lebanon. Eastern Mediterranean Health Journal, 10, 268-
276.









Epperson, C.N. (1999, April 15). Postpartum major depression: Detection and treatment.
Retrieved July 28, 2006, from http://www.aafp.org/afp/990415ap/2247.html

Evans, G.G., & Theofrastous, J.P. (1997). Postpartum depression: A review of postpartum
screening. Primary Care Update for Obstetricians and Gynecologists, 4, 241-246.

Fairburn, C.G., & Welch, S.L. (1990). The impact of pregnancy on eating habits and attitudes to
shape and weight. International Journal ofEating Disorders, 9, 153-160.

Fink, A. (1993). Evaluation fundamentals: Guiding health programs, research, and policy.
Newbury Park, CA: Sage.

Finkelstein EA, Ruhm CJ, Kosa KM. (2005). Economic causes and consequences of obesity.
AnnualReview of Public Health, 26, 239-257.

Fox, P., & Yamaguchi, C. (1997). Body Image Change in Pregnancy: A Comparison of Normal
Weight and Overweight Primigravidas. Birth, 24, 35-40.

Franko, D.L., & Spurrell, E. B. (2000). Detection and management of eating disorders during
pregnancy. Obstetrics & Gynecology, 95, 942-946.

Freda, M.C., Andersen, H.F., Damus, K., & Merkatz, I.R. (1993). Are there differences in
information given to private and public prenatal patients? American Journal of Obstetrics
& Gynecology, 169, 155-160.

Freeman, M.P., Wright, R., Watchman, M., Wahl, R.A., Sisk, D.J., Fraleigh, L., & Weibrecht,
J.M. (2005). Postpartum depression assessments at well-baby visits: Screening feasibility,
prevalence, and risk factors. Journal of Women's Health, 14, 929-935.

Frieder, A., Dunlop, A.L., Culpepper, L., & Bernstein, P.S. (2008). The clinical content of
preconception care: Women with psychiatric conditions. American Journal of Obstetrics &
Gynecology, 199, S328-S332.

Friedman, S. H., Heneghan, A., & Rosenthal, M. (2009). Characteristics of women who do not
seek prenatal care and implications for prevention. Journal of Obstetrics, Gynecology and
Neonatal Nursing, 38, 174-181.

Frisbie, W.P., Biegler, M., de Turk, P., Forbes, D., & Pullum, S.G. (1997). Racial and ethnic
differences in determinants of intrauterine growth retardation and other compromised birth
outcomes. American Journal ofPublic Health, 87, 1977-1983.

Galtier-Dereure, F., Boegner, C., & Bringer, J. (2000). Obesity and pregnancy: Complications
and cost. American Journal of Clinical Nutrition, 71, 1242S-1248S.









Gaynes, B.N., Gavin, N., Meltzer-Brody, S., Lohr, K.N., Swinson, T., Gartlehner, G., Brody, S.,
& Miller, W.C. (2005). Perinatal depression: Prevalence, screening, accuracy, and
screening outcomes. Rockville, MD: Agency for Healthcare Research and Quality; 2005.
Evidence Report/Technology Assessment 119. AHRQ Publication 05-E006-2. Retrieved
April 14, 2008, from
http://www.ahrq.gov/downloads/pub/evidence/pdf/peridepr/peridep.pdf

Georgiopoulous, A.M., Bryan, T.L., Yawn, B.P., Houston, M.S., Rummans, T.A., & Therneau,
T.M. (1999). Population-based screening for postpartum depression. Obstetrics &
Gynecology, 93, 653-657.

Georgiopoulous, A.M., Bryan, T.L., Wollan, P., & Yawn, B.P. (2001). Routine screening for
postpartum depression. Journal ofFamily Practice, 50, 117-122.

Glanz, K., Rimer, B.K. & Lewis, F.M. (2002). Health Behavior andHealth Education. Theory,
Research and Practice. San Francisco, CA: Wiley & Sons.

Goodman, J.H. (2004). Paternal postpartum depression, its relationship to maternal postpartum
depression, and implications for family health. Journal ofAdvanced Nursing, 45, 26-35.

Goodwin, M.M., Zahniser, C.S., Kogan, M.D., Johnson, C., Bruce, C., & Gunter, E.P. (no date).
Behavioral risk factors before and during pregnancy (Centers for Disease Control
publication). Atlanta: GA: Author.

Gore, S.A., Brown, D.M., & Smith West, D. (2003). The role of postpartum weight retention in
obesity among women: A review of the evidence. Annals ofBehavioral Medicine, 26, 149-
159.

Gortmaker, S.L. (1979). The effects of prenatal care upon the health of the newborn. American
Journal of Public Health, 69, 653-660.

Gotlib, I.H. (1989). Prevalence rates and demographic characteristics associated with depression
in pregnancy and the postpartum. Journal of Consulting and Clinical Psychology, 57, 269-
274.

Gotlib, I.H., Whiffen, V.E., Wallace, P.M., & Mount, J.H. (1991). Propsective investigation of
postpartum depression: Factors involved in onset and recovery. Journal ofAbnormal
Psychology, 100, 122-132.

Greenberg, P.E., Kessler, R.C., Bimbaum, H.G., Leong, S.A., Lowe, S.W., Berglund, P.A., &
Corey-Lisle, P.K. (2003). The economic burden of depression in the United States: How
did it change between 1990 and 2000? Journal of Clinical Psychiatry, 63, 1465-1475.

Gross, K.H., Wells, C.S., Radigan-Garcia, A., & Dietz, P.M. (2002). Correlates of self-reports of
being very depressed in the months after delivery: Results from the Pregnancy Risk
Assessment Monitoring System. Maternal and Child Health Journal, 6, 247-253.









Gunderson, E.P., & Abrams, B. (2000). Epidemiology ofgestational weight gain and body
weight changes after pregnancy. Epidemiologic Reviews, 22, 261-274.

Guyll, M., Matthews, K.A., & Bromberger, J.T. (2001). Discrimination and unfair treatment.
Relationship to cardiovascular reactivity among African American and European
American women. Health Psychology, 20, 315-325.

Halbreich, U. (2005). The association between pregnancy processes, preterm delivery, low birth
weight, and postpartum depressions-The need for interdisciplinary integration. American
Journal of Obstetrics and Gynecology, 193, 1312-1322.

Hall, L.A., Kotch, J.B., Browne, D., & Rayens, M.K. (1996). Self-esteem as a mediator of the
effects of stressors and social resources on depressive symptoms in postpartum mothers.
Nursing Research, 45, 231-238.

Harris, H.E., Ellison, G.T.H., & Clement. S. (1999) Do the psychosocial and behavioral changes
that accompany motherhood influence the impact of pregnancy on long-term weight gain?
Journal ofPsychosomatic Obstetrics and Gynecology, 20, 65-79.

Haslam, D.M., Pakenham, K, I., & Smith, A. (2006). Social support and postpartum depressive
symptomatology. Infant Mental Health Journal, 27, 276-291.

Hayes, D., & Ross, C.E. (1987). Concern with appearance, health beliefs, and eating habits.
Journal of Health and Social Behavior, 28, 120-130.

Health Resources and Services Administration and the Centers for Disease Control. (2000).
Healthy People 2010: 16, Maternal, Infant, and Child Health. Retrieved March 20, 2009,
from http://www.healthypeople.gov/Document/HTML/Volume2/16MICH.htm

Heaney, C.A., & Israel, B.A. (2002). Social networks and social support. In Glanz, K., Rimer,
B.K., & Lewis, F.M. (Eds.), Health Behavior and Health Education: Theory, Research, and
Practice. (pp. 185-209). San Francisco, CA: Jossey-Bass.

Hebl, M.R., & Turchin, J. M. (2005). The stigma of obesity: What about men? Basic and applied
social psychology, 27, 267-275.

Hedley, A.A., Ogden, C.L., Johnson, C.L., Carroll, M.D., Curtin, L.R., & Flegal, K.M. (2004).
Prevalence of overweight and obesity among U.S. children, adolescents, and adults, 1999-
2002. Journal of the American Medical Association, 291, 2847-2850.

Heo, M., Pietrobelli, A., Fontaine, K.R., Sirey, J.A., & Fiath, M.S. (2006). Depressive mood and
obesity in U.S. adults: Comparison and moderation by sex, age, and race. International
Journal of Obesity, 30, 513-519.

Higgins, P. G., Frank, B., & Brown, M. (1994). Changes in health behaviors made by pregnant
women. Health Care for Women International, 15, 149-156.









Higgins, P. G., Clough, D. H., Frank, B., & Wallerstedt, C. (1995). Changes in health behaviors
made by pregnant substance users. The International Journal of the Addiction, 30, 1323-
1333.

Higgins, P.G., & Woods, P.J. (1999). Reasons, health behaviors, and outcomes of no prenatal
care: Research that changed practice. Health Care for Women International, 20, 128-136.

Hiser, P. L. (1987). Concerns of multiparas during the second postpartum week. Journal of
Obstetrics, Gynecology, and Neonatal Nursing, 16, 195-202.

Hobfoll, S.E., & Leiberman, J.R. (1987). Personality and social resources in immediate and
continued stress resistance among women. Journal ofPersonality and Social Psychology,
52, 18-26.

Homan, R.K., & Korenbrot, C.C. (1998). Explaining variation in birth outcomes of Medicaid-
eligible women with variation in the adequacy of prenatal support services. Medical Care,
36, 190-201.

Horowitz, J.A.,Damato, E., Solon, L., von Metzsch, G.,& Gill,V. (1995). Postpartum depression:
Issues in clinical assessment. Journal ofPerinatology, 15, 268-278.

Horowitz, J.A., Damato, E.G., & Duffy, M.E. (2005). The relationship of maternal attributes,
resources, and perceptions of postpartum experiences to depression. Research in Nursing
andHealth, 28, 159-171.

Institute of Medicine. (1990). Nutrition during pregnancy. Washington, DC: National Academy
Press.

Janz, N.K., & Becker, M.H. (1984). The health belief model: A decade later. Health Education
Behavior, 11, 1-47.

Jenkin, W., & Tiggemann, M. (1997). Psychological effects of weight retained after pregnancy.
Women & Health, 25, 89-98.

Johnson, R.W., & Broadnax, P.A. (2003). A perspective on obesity. ABNFJournal, 14, 69-70.

Johnstone, S.J., Boyce, P.M., Hickey, A.R., Morris-Yates, A.D., & Harris, M.G. (2001).
Obstetric risk factors for postnatal depression in urban and rural community samples.
Australian and New Zealand Journal ofPsychiatry, 35, 69-74.

Jorm, A.F., Korten, A.E., Christensen, H., Jacomb, P.A., & Rodgers, B., Parslow, R.A. (2003).
Association of obesity with anxiety, depression, and emotional well-being: A community
survey. Australian & New Zealand Journal of Public Health, 27, 434-440.

Josefsson, A., Berg, G., Nordi, C., & Sydsjo, G. (2001). Prevalence of depressive symptoms in
late pregnancy and postpartum. Acta obstetricia et gynecologica Scandinavica, 80, 251-
255.









Josefsson, A., Angelsioo, L., Berg, G., Ekstrom, C-M., Bunnervik, C., Nordin, C., & Sydsjo, G.
(2002). Obstetric, somatic, and demographic risk factors for postpartum depressive
symptoms. Obstetrics & Gynecology, 99, 223-228.

Kabakian-Khasholian, T., Campbell, O., Shediac-Rizkallah, M., & Ghorayeb, F. (2000).
Women's experiences of maternity care: Satisfaction or passivity? Social Science &
Medicine, 51, 103-113.

Karlsson, J., Taft, C., Sjostrom, L., Torgerson, J.S., & Sullivan, M. (2003). Psychosocial
functioning in the obese before and after weight reduction: Construct validity and
responsiveness of the Obesity-Related Problems Scale. International Journal of Obesity,
27, 617-630.

Keller, C., Allan, J., & Tinkle, M.B. (2005). Stages of change, processes of change, and social
support for exercise and weight gain in postpartum women. Journal of Obstetrics,
Gynecology, and Neonatal Nursing, 35, 232-240.

Keppel, K.G., & Taffel, S.M. (1993). Pregnancy-related weight gain and retention: Implications
of the 1990 Institute of Medicine Guidelines. American Journal of Public Health, 83,
1100-1103.

Kersting, A., Dorsch, M., Wesselmann, U., Ludorff, K., Witthaut, J., Ohrmann, P., Hornig-
Franz, I., Klockenbusch, W., Harms, E., & Arolt, V. (2004). Maternal posttraumatic stress
response after the birth of a very low-birth-weight infant. Journal of Psychosomatic
Research, 57, 473-476.

Kim, S.Y., Dietz, P.M., England, L., Morrow, B., & Callaghan, W.M. (2007). Trends in pre-
pregnancy obesity in nine states, 1993-2003. Obesity, 15, 986-993.

Kim, O., & Kim, K. (2001). Body weight, self-esteem, and depression in Korean female
adolescents. Adolescence. 36, 315-322.

Klein, Sm., Burke, L.E., Bray, G.A., Blair, S., Allison, D.B., Pi-Sunyer, X., Hong, Y., & Eckel,
R.H. (2004). Clinical implications of obesity with specific focus on cardiovascular disease.
A statement for professionals from the American Heart Association Council on Nutrition,
Physical Activity, and Metabolism. Circulation, 110, 2952-2967.

Kline, C.K., Martin, D.P., & Deyo, R.A. (1998). Health consequences of pregnancy and
childbirth as perceived by women and clinicians. Obstetrics & Gynecology, 92, 842-848.

Klohe-Lehman, D.M., Freeland-Graves, J., Anderson, E.R., McDowell, T., Clarke, K.K., Hanss-
Nuss, H., Cai, G., Puri, D., & Milani, T.J. (2006). Nutrition knowledge is associated with
greater weight loss in obese and overweight low-income mothers. Journal of the American
Dietetic Association, 106, 65-75.

Kogan, M.D., Alexander, G.R., Kotelchuck, M., Nagey, D.A., & Jack, B.W. (1994). Comparing
mothers' reports on the content of prenatal care received with recommended national
guidelines for care. Public Health Reports, 109, 637-646.









Kogan, M.D., Kotelchuck, M., Alexander, G.R., & Johnson, W.E. (1994). Racial disparities in
reported PNC advice from health care providers. American Journal of Public Health, 84,
82-88.

Kogan, M.D., Alexander, G.R., Kotelchuck, M., & Nagey, D.A. (1994). Relations of the content
of prenatal care to the risk of low birth weight. Journal of the American Medical
Association, 271, 1340-1345.

Kogan, M.D., Martin, J.A., Alexander, G.R., Kotelchuck, M., Ventura, S.J., & Frigoletto, F.D.
(1998). The changing pattern of prenatal care utilization in the United States, 1981-1995,
using different prenatal care indices. Journal of the American Medical Association, 279,
1623-1628.

Koroukian, S.M., & Rimm, A.A. (2002). The "Adequacy of Prenatal Care Utilization" (APNCU)
Index to study low birth weight: Is the index biased? Journal of Clinical Epidemiology, 55,
296-305.

Kotelchuck, M. (1994). An evaluation of the Kessner adequacy of prenatal care index and a
proposed Adequacy of Prenatal Care Utilization Index. American Journal ofPublic
Health, 94, 1414-1420.

Kotelchuck, M. (1994). The Adequacy of Prenatal Care Utilization Index: Its U.S. distribution
and association with low birth weight. American Journal of Public Health, 84, 1486-1489.

Kotelchuck, M., Kogan, M.D., Alexander, G.R., & Jack, B.W. (1997). The influence of site of
care on the content of PNC for low-income women. Maternal and Child Health Journal, 1,
25-34.

Kress, A.M., Peterson, M.R., & Hartzell, M.C (2006). Association between obesity and
depressive symptoms among U.S. military active duty service personnel, 2002. Journal of
Psychosomatic Research, 60, 363-271.

Kroenke, K., Spitzer, R.L., & Willians, J.B.W. (2001). The PHQ-9. Validity of a brief depression
severity measure. Journal of General Internal Medicine, 16, 606-613.

Kroenke, K., Spitzer, R.L., & Willians, J.B.W. (2003). The Patient Health Questionnaire-2.
Validity of a two-item depression screener. Medical Care, 41, 1284-1192.

Kruijshaar, M.E., Hoeymans, N., Spijker, J., Stouthard, M.E.A., & Essink-Bot, M-L, (2005). Has
the burden of depression been overestimated? Bulletin of the WorldHealth Organization,
83, 443-448.

Lacey, J.H., & Smith, G. (1987). Bulimia nervosa: The impact of pregnancy on mother and baby.
British Journal of Psychiatry, 150, 777-781.

LaCoursiere, D.Y., Bloebaum, L., Duncan, J.D., & Varner, M.W. (2005). Population-based
trends and correlates of maternal overweight and obesity, Utah 1991-2001. American
Journal of Obstetrics and Gynecology, 192, 832-839.









LaCoursiere, D.Y., Baksh, L., Bloebaum, L., & Varner, M.W. (2006). Maternal body mass index
and self-reported postpartum depressive symptoms. Maternal and ChildHealth Journal,
10, 385-390.

Laxton-Kane, M., & Slade, P. (2002). The role of maternal prenatal attachment in a woman's
experience of pregnancy and implications for the process of care. Journal of Reproductive
and Infant Psychology, 4, 253-266.

Lederman, S.A., Alfasi, G., & Deckelbaum, R.J. (2002). Pregnancy-associated obesity in Black
women in New York City. Maternal and Child Health Journal, 6, 37-42.

Lederman, S.A., Paxton, A., Heymsfield, S.B., Wang, J., Thornton, J., & Pierson, R.N. (1997).
Body fat and water changes during pregnancy in women with different body weight and
weight gain. Obstetrics & Gynecology, 90, 483-488.

Leidner, N.F., Singer, T., Sicherman, N., Francoise, V., & Divon, M.Y. (2008). Postpartum
depression; correlation with obstetric, medical and demographic risk factors. American
Journal of Obstetrics and Gynecology, 199, S150-S150.

Leung, W.C., Kung, F., Lam, J., Leung, T.W., & Ho, P.C. (2002). Domestic violence and
postnatal depression in a Chinese community. International Journal of Gynecology and
Obstetrics, 79, 159-166.

Levine, B.S., Wigren, M.M., Chapman, D.S., Kerner, J.F., German, R.L., & Rivlin, R.S. (1993).
A national survey of attitudes and practices of primary-care physicians relating to nutrition:
Strategies for enhancing the use of clinical nutrition in medical practice. American Journal
of Clinical Nutrition, 57, 115-119.

Libbus, M.K., & Sable, M.R. (1991). Prenatal education in a high-risk population: The effect on
birth outcomes. Birth, 18, 78-82.

Lichtman, S.W., Pisarska, K., Berman, E.R., Pestone, M., Dowling, H., Offenbacher, E., Weisel,
H., Heshka, S., Matthews, D.E., & Heymsfield, S.B. (1992). Discrepency between self-
reported and actual caloric intake and exercise in obese subjects. New England Journal of
Medicine, 327, 1893-1898.

Lieberman, A.F. (1977). Preschoolers' competence with a peer: Relations with attachment and
peer experience. ChildDevelopment, 48, 1277-1287.

Linde, J. A., Jeffery, R. W. Levy, R. L. Sherwood, N. E. Utter, J, Pronk, N. P. Boyle, R. G.
(2004). Binge eating disorder, weight control, self-efficacy, and depression in overweight
men and women. International Journal of Obesity & Related Metabolic Disorders, 28,
418-425.

Linde, J.A., Jeffery, R.W., Finch, E.A., Simon, G.E., Ludman, E.J., Operskalski, B.H., Ichikawa,
L., & Rohde, P. (2007). Relation of body mass index to depression and weighing
frequency in overweight women. Preventive Medicine, 45, 75-79.









Long, A., Reed, R., & Lehman, G. (2006). The cost of lifestyle health risks: Obesity. Journal of
Occupational & Environmental Medicine, 48, 244-251.

Lox, C.L., Osborn, M.C., & Pellet, T. (1998). Body image and affective experiences of
subjectively underweight females: Implications for exercise behavior. Journal ofApplied
Biobehavioral Research, 3, 110-118.

Lu, G.C., Rouse, D.J., DuBard, M., Cliver, S., Kimberlin, D., & Hauth, J.C. (2001). The effect of
the increasing prevalence of maternal obesity on perinatal morbidity. American Journal of
Obstetrics and Gynecology, 185, 845-849.

Luke, B., Dickinson, C., & Petrie, R.H. (1981). Intrauterine growth: Correlations of maternal
nutritional status and rate of gestational weight gain. European Journal of Obstetrics,
Gynecology, and Reproductive Biology, 12, 113-121.

Mahmood, T. A. (2009). Review: Obesity and pregnancy: An obstetrician's view. The British
Journal ofDiabetes & Vascular Disease, 9, 19-22.

Markowitz, S., Friedman, M.A., & Arent, S.M. (2008). Understanding the relation between
obesity and depression: Causal mechanisms and implications for treatment. Clinical
Psychology: Science and Practice, 15, 1-20.

McCarthy, M.E. (1998). Longitudinal study of body image and adjustment during pregnancy and
thepuerperium. Retrieved from Proquest: Dissertations and Theses on April 6, 2008.

McCormick, M.C., & Siegel, J.E. (2001). Recent evidence on the effectiveness of PNC.
Ambulatory Pediatrics, 1, 321-325.

McCoy, S.J.B., Beal, J.M., Shipman, S.B.M., Payton, M.E., & Watson, G.H. (2006). Risk factors
for postpartum depression: A retrospective investigation at 4-weeks postnatal and a review
of the literature. JAOA: Journal of the American Osteopathic Association, 106, 193-198.

Mercer, R.T., & Ferketich, S.L. (1988). Stress and social support as predictors of anxiety and
depression during pregnancy. Advances in Nursing Science, 10, 26-39.

Merritt, T., Kuppin, S., & Wolper, M. (2001). Postpartum depression causes and correlates.
International Electronic Journal of Health Education, 31, 57-63.

Miller, L.J. (2002). Postpartum depression. Journal of the American Medical Association, 287,
762-765.

Minkovitz, C.S., Strobino, D., Scharfstein, D., Hou, W., Miller, T., Mistry, K.B., & Swartz, K.
(2005). Maternal depressive symptoms and children's receipt of health care in the first 3
years of life. Pediatrics, 115, 306-314.

Misra, D. P., & Grason, H. (2006). Achieving safe motherhood: Applying a life course and
multiple determinants perinatal health framework in public health. Women's Health Issues,
16, 159-175.









Moore, D.S. (1978). The body image in pregnancy. Journal ofNurse-Midwifery, 22, 17-27.

Mora, J.O., & Nestel, P.S. (2000). Improving prenatal nutrition in developing countries:
Strategies, prospects, and challenges. American Journal of Clinical Nutrition, 71, 1353S-
1363S.

Morin, K.H., & Reilly, L. (2007). Caring for obese pregnant women. Journal of Obstetrics,
Gynecology, and Neonatal Nursing, 36, 482-489.

Mustard, C.A., & Roos, N.P. (1994). The relationship of prenatal care and pregnancy
complications to birthweight in Winnipeg, Canada. American Journal ofPublic Health, 84,
1450-1457.

Nalepka, C.D., & Coblentz, M.J. (1995). Incidence of postpartum depression following
antepartal intervention. Journal ofPerinatal Education, 4, 29-37.

National Institute of Child Health and Human Development. (2007). Prenatal Care. Retrieved
July 27, 2007 from http://www.nichd.nih.gov/health/topics/Prenatal_Care.cfm

National Institutes of Health. (1998). Clinical guidelines on the identification, evaluation, and
treatment of overweight and obesity in adults. Retrieved December 8, 2007 from
http://www.nhlbi.nih.gov/guidelines/obesity/ob_gdlns.pdf

National Institutes of Health (Medline Plus). (2008). Gestational age. Retrieved July 11, 2007,
from http://www.nlm.nih.gov/medlineplus/ency/article/002367.htm

Negus Jolley, S.N., & Betrus, P. (2007). Comparing postpartum depression and major depressive
disorder: Issues in assessment. Issues in MentalHealth Nursing, 28, 765-780.

Nielsen, D., Videbech, P., Hedegaard, M., Dalby, J., & Secher, N.J. (2000). Postpartum
depression: Identification of women at risk. BJOG: An International Journal of Obstetrics
& Gynaecology, 107, 1210-1217.

O'Hara, M.W., Schlechte, J.A., Lewis, D.A., & Varner, M.W. (1991). Controlled prospective
study of postpartum mood disorders: Psychological, environmental, and hormonal
variables. Journal ofAbnormal Psychology, 100, 63-73.

O'Hara, M.W., & Gorman, L.L. (2004). Can postpartum depression be predicted? Primary
Psychiatry, 11, 42-47.

Olafsdottir, A.S., Skuladottir, G.V., Thorsdottir, I., Hauksson, A., & Steingrimsdottir, L. (2006).
Maternal diet in early and late pregnancy in relation to weight gain. International Journal
of Obesity, 30, 492-499.

Olfson, M., & Klerman, G.L. (1992). Depressive symptoms and mental health service utilization
in a community sample. Social Psychiatry and Psychiatric Epidemiology, 27, 161-167.


200









Onyike, C.U., Crum, R.M., Lee, H.B., Lykestsos, C.G., & Eaton, W.W. (2003). Is obesity
associated with major depression? Results from the Third National Health and Nutrition
Examination Survey. American Journal ofEpidemiology, 158, 1139-1147.

Orstead, C., Arrington, D., Kamath, S.K., Olson, R., & Kohrs, M.B. (1985). Efficacy of prenatal
nutrition counseling: Weight gain, infant birth weight and cost effectiveness. Journal of the
American Dietetic Association, 85, 40-46.

O'Toole, M.L., Sawicki, M.A., & Artal, R. (2003). Structured diet and physical activity prevent
postpartum weight retention. Journal of Women's Health, 12, 991-998.

Paquette, M-C., & Raine, K. (2004). Sociocultural context of women's body image. Social
Science and Medicine, 58, 1047-1058.

Patel, V., Rodrigues, M., & DeSouza, N. (2002). Gender, poverty, and postnatal depression: A
study of mothers in Goa, India. American Journal ofPsychiatry, 159, 43-47.

Peoples-Sheps, M.D., Kalsbeek, W.D., & Siegel, E. (1988). Why we know so little about PNC
nationwide: An assessment of required methodology. Health Services Research, 23, 359-
379.

Petersen, R., Connelly, A., Martin, S.L., & Kupper, L.L. (2001). Preventive counseling during
prenatal care: Pregnancy Risk Assessment Monitoring System (PRAMS). American
Journal of Preventive Medicine, 20, 245-250.

Petitti, D.B., Hiatt, R.A., Chin, V., & Croughan-Minihane, M. (1994). An outcome evaluation of
the content and quality of PNC. Birth, 18, 21-25.

Plutchik, R. (1976). Emotions and attitudes related to being overweight. Journal of Clinical
Psychology, 32, 21-24.

Polley, B.A., Wing, R.R., & Sims, C.J. (2002). Randomized controlled trial to prevent excessive
weight gain in pregnancy women. InternationalJournal of Obesity, 26, 1494-1502.

Press, N., & Browner, C.H. (1997). Why women say yes to prenatal diagnosis. Social Science &
Medicine, 45, 978-989.

Quick, J.D., Greenlick, M.R., & Roghmann, K.J. (1981). Prenatal care and pregnancy outcome
in an HMO and general population. American Journal ofPublic Health, 71, 381-390.

Radke-Yarrow, M., Cummings, M.E., Kuczynski, L., & Chapman, M. (1985). Patterns of
attachment in two-and three-year-olds in normal families and families with parental
depression. Child Development, 56, 884-493.

Rahman, A., Iqbal, Z., & Harrington, R. (2003). Life events, social support and depression in
childbirth: Perspectives from a rural community in the developing world. Psychological
Medicine, 33, 1161-1167.









Rahman, A., Iqbal, Z., Bunn, J., Lovel, H., & Harrington, R. (2004). Impact of maternal
depression on infant nutritional status and illness. Archives of General Psychiatry, 61, 946-
952.

Rahman, A., & Creed, F. (2006). Outcome of prenatal depression and risk factors associated with
persistence in the first postnatal year: Prospective study from Rawalpindi, Pakistan.
Journal of Affective Disorders, 100, 115-121.

Reck, C., Hunt, A., Fuchs, T., Weiss, R., Noon, A., Moehler, E., Downing, G., Tronick, E.Z., &
Mundt, C. (2004). Interactive regulation of affect in postpartum depressed mothers and
their infants: An overview. Psychopathology, 37, 272-280.

Rich-Edwards, J.W., Kleinman, K., Abrams, A., Harlow, B.L., McLauhglin, T.J., Joffe, H., &
Gillman, M.W. (2006). Sociodemographic predictors of antenatal and postpartum
depressive symptoms among women in a medical group practice. Journal ofEpidemiology
& Community Health, 60, 221-227.

Risica, P.M., & Phipps, M.G. (2006). Educational preferences in a prenatal clinic. International
Journal of Chilbi, i th Education, 21, 4-7.

Ritenbaugh, C. (1982). Obesity as a culture-bound syndrome. Culture, Medicine, andPsychiatry,
6, 347-361.

Robb-Todter, G.E. (1996). Women's experience of weight and shape changes during pregnancy.
Dissertation Abstracts International, 57 (01), A. (UMI No. 9616020)

Roberts, R.E., Deleber, S., Strawbridge, W.J., & Kaplan, G.A. (2003). Prospective association
between obesity and depression: Evidence from the Alameda County study. International
Journal of Obesity, 27, 514-521.

Rosen, M.G. (1989). Public Health Service: Caring for our future: the content ofprenatal care.
A report of the Public Health Service's Expert Panel on the Content of Prenatal Care.
Washington, D. C: U.S. Government Printing Office. Retrieved November 27, 2007 from
http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content storage_01/0000019b/80/23/
Of/3d.pdf

Rosenberg, T.J., Garbers, S., Chavkin, W., & Chiasson, M.A. (2003). Prepregnancy weight and
adverse perinatal outcomes in an ethnically diverse population. Obstetrics & Gynecology,
102, 1022-1027.

Rosenberg, T.J., Garbers, S., Lipkind, H., & Chiasson, M.A. (2005). Maternal obesity and
diabetes as risk factors for adverse pregnancy outcomes: Differences among 4 racial/ethnic
groups. American Journal of Public Health, 95, 1544-1551.

Rubeinstein, G. (2006). The big five and self-esteem among overweight dieting and non-dieting
women. Eating Behaviors, 7, 355-361.


202









Rubin, R. (1984). Maternal identity and the maternal experience. New York, NY: Springer
Publishing Company.

Ruderman, M., & O'Campo, P. (1999). Depression in women. In H. Grason, J. Hutchins, & G.
Silver (Eds.), Charting a course for the future of women's and perinatal health (pp. 147-
166). Timonium, MD: Printing Corporation of America.

Russell, C.S. (1974). Transition to parenthood: Problems and gratifications. Journal ofMarriage
and the Family, 36, 294-3 02.

Sable, M.R., & Herman, A.A. (1997). The relationship between prenatal health behavior advice
and low birth weight. Public Health Reports, 112, 332-339.

Sackett, K., Pope, R.K., & Erdley, W.S. (2004). Demonstrating a positive return on investment
for a prenatal program at a managed care organization. Journal ofPerinatal andNeonatal
Nursing, 18, 117-127.

Saravanakumar, K., Rao, S.G., & Cooper, G.M. (2006). Obesity and obstetric anaesthesia.
Anlethehtiia, 61, 36-48.

Sargeant, J.K., Bruce, M.L., Florio, L.P., & Weissman, M.M. (1990). Factors associated with 1-
year outcome of major depression in the community. Archives of General Psychiatry, 47,
519-526.

Sarwer, D., Allison, K., Gibbons, L., Markowitz, J., & Nelson, D. (2006). Pregnancy and
obesity: A review and agenda for future research. Journal of Women's Health, 15, 720-
733.

Schuckit, M.A. (1996). Alcohol, anxiety, and depressive disorders. Alcohol Health & Research
World, 20, 81-86.

Schulz, R., Beach, S.R., Ives, D.G., Martire, L.M., Ariyo, A.A., & Kop, W.J. (2000). Association
between depression and mortality in older adults. Archives oflnternal Medicine, 160,
1761-1768.

Sharan, H., Kaplan, B., Weizer, N., Sulkes, J., & Merlob, P. (2006) Early screening of
postpartum depression using the Edinburgh Postnatal Depression Scale. International
Journal of Risk & Safety in Medicine, 18, 213-218.

Shepard, M.J., Hellenbrand, K.G., & Bracken, M.B. (1986). Proportional weight gain and
complications of pregnancy, labor and delivery in healthy women of normal pre-pregnant
stature. American Journal of Obstetrics & Gynecology, 155, 947-954.

Sheppard, V.B., Zambrana, R.E., & O'Malley, A.S. (2004). Providing health care to low-income
women: A matter of trust. Family Practice, 21, 484-491.


203









Showstack, J.A., Budetti, P.P., & Minkler, D. (1984). Factors associated with birthweight: An
exploration of the roles of prenatal care and length of gestation. American Journal of
Public Health, 74, 1003-1008.

Shulman, H.B., Gilbert, B.C., & Lanksky, A. (2006). The Pregnancy Risk Assessment
Monitoring System (PRAMS): Current methods and evaluation of 2001 response rates.
Public Health Reports, 121, 74-83.

Siega-Riz, A.M., Adair, L.S., & Hobel, C.J. (1996). Maternal Underweight Status and
Inadequate Rate of Weight Gain during the Third Trimester of Pregnancy Increases the
Risk of Preterm Delivery. The Journal ofNutrition, 126, 146-153

Singer, L.T., Salvatore, A., Guo, S., Collin, M., Lilien, L., & Baley, J. (1999). Maternal
psychological distress and parenting stress after the birth of a very low-birth-weight infant.
Journal of the American Medical Association, 281, 799-805.

Skouteris, H., Carr, R., Wertheim, E.H., Paxton, S.J., & Duncombe, D. (2005). A prospective
study of factors that lead to body dissatisfaction during pregnancy. Body Image, 2, 347-
361.

Splett, P.L., Reinhardt, M.A., & Fleming, P. (1994). Physicians' expectations for quality
nutrition expertise and service in PNC. Journal of the American Dietetic Association, 94,
1375-1380.

StataCorpLP. (2007). STATA: Statistical Analysis and Data Software (Version 10) [Computer
software]. College Station, TX.

Stewart, A.L., Napoles-Springer, A., & Perez-Stable, E.J. (1999). Interpersonal processes of care
in diverse populations. Milbank Quarterly 77, 305-339.

Stotland, N.E., Haas, J.S., Brawarsky, P., Jackson, R.A., Fuentes-Afflick, E., & Escobar, G.J.
(2005). Body mass index, provider advice, and target gestational weight gain. Obstetrics &
Gynecology, 105, 633-638.

Stowe, Z.N., & Nemeroff, C.B. (1995). Women at risk for postpartum-onset major depression.
American Journal of Obstetrics and Gynecology, 173, 639-645.

Strang, V.R., & Sullivan, P.L. (1985). Body image attitudes during pregnancy and the
postpartum period. Journal of Obstetrics, Gynecology, and Neonatal Nursing, 14, 332-337.

Suttie, J.S. (1998). Body image issues in postpartum women and their relationship to postpartum
depression. Retrieved from Proquest: Dissertations and Theses on April 5, 2008.

Taffel, S.M., & Keppel, K.G. (1986). Advice about weight gain during pregnancy and actual
weight gain. American Journal ofPublic Health, 76, 1396-1399.


204









Taffel, S.M., Keppel, K.G., & Jones, G.K. (1993). Medical advice on maternal weight gain and
actual weight gain results from the 1988 National Maternal and Infant Health Survey.
Annals of the New York Academy ofSciences, 678, 293-305.

Tiggeman, M., & Rothblum, E.D. (1988). Gender differences in social consequences of
perceived overweight in the United States and Australia. Sex Roles, 18, 75-86.

Torpy, J.M., Burke, A.E., & Glass, R.M. (2006). Women's Health. Journal of the American
Medical Association, 295, 1474.

U.S. Public Health Service. (1989). Caringfor our future: The content ofprenatal care.
Retrieved January 10, 2008, from
http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content storage_01/0000019b/80/23/
Of/3d.pdf

Walker, L.O. (1997). Weight and weight-related distress after childbirth: Relationships to stress,
social support, and depressive symptoms. Journal of Holistic Nursing, 15, 389-405

Walker, L.O. (1998). Weight-related distress in early months after childbirth. Western Journal of
Nursing Research, 20, 30-44.

Walker, L.O., Timmerman, G.M., Kim, M., & Sterling, B. (2002). Relationships between body
image and depressive symptoms during postpartum in ethnically diverse, low income
women. Women &Health, 36, 101-121.

Walker, L.O. (2007). Managing excessive weight gain during pregnancy and the postpartum
period. Journal of Obstetrics, Gynecology, & Neonatal Nursing, 36, 490-500.

Watkins, M.L., Rasmussen, S.A., Honein, M.A., Botto, L.D., & Moore, C.A. (2003). Maternal
obesity and risk for birth defects. Pediatrics, 111, 1152-1158.

Webb, D. A., Bloch, J.R., Coyne, J.C., Chung, E.K., Bennett, I.M., & Culhane, J.F. (2008).
Postpartum physical symptoms in new mothers: Their relationship to functional limitations
and well-being. Birth, 35, 179-187.

Weissman, M.M., & Olfson, M. (1995). Depression in women: Implications for health care
research. Science, 269, 799-800.

Weissman, M.M., Pilowsky, D.J., Wickramaratne, P.J., Talati, A., Wisniewski, S.R., Fava, M.,
Hughes, C.W., Garber, J., Malloy, E., King, C.A., Cerda, G., Sood, A.B., Alpert, J.E.,
Trivedi, M.H., Rush, A.J., & Star*D-Child Team. (2006). Remissions in maternal
depression and child psychopathology: A STAR*D-child report. Journal of the American
Medical Association, 295, 1389-1398.

Wells, C.S., Schwalberg, R., Noonan, G., & Gabor, V. (2006). Factors influencing inadequate
and excessive weight gain in pregnancy: Colorado, 2000-2002. Maternal and Child Health
Journal, 10, 55-62.


205









Wheatley, R.R., Kelley, M.A., Peacock, N., & Delgado, J. (2008). Women's narraties on quality
in prenatal care: A multicultural perspective. Qualitative Health Research, 18, 1586-1598.

Williamson, C.S. (2006). Nutrition in pregnancy. Nutrition Bulletin, 31, 28-59.

Williamson, D.F., Kahn, H.S., Remington, P.L., & Anda, R.F. (1990). The 10-year incidence of
overweight and major weight gain in US adults. Archives ofInternal Medicine, 150, 555-
672.

Wisner, K.L., Parry, B.L., & Piontek, C.M. (2002). Postpartum depression. New England
Journal of Medicine, 347, 195-199.

Wisner, K.L., Chambers, C., & Sit, D.K.Y. (2006). Postpartum depression: A major public
health problem. JAMA: Journal of the American Medical Association, 296, 2616-2618.

Wong, S.T., Korenbrot, C.C., & Stewart, A.L. (2004). Consumer assessment of the quality of
interpersonal processes of prenatal care among ethnically diverse low-income women:
Development of a new measure. Women's Health Issues, 14, 118-129.

Wooley, S.C., & Wooley, O.W. (1979). Obesity and women-I. A closer look at the facts.
Women's Studies International Quarterly, 2, 69-79.

Wood Baker, C., Carter, A.S., Cohen, L.R., & Brownell, K.D. (1999). Eating attitudes and
behaviors in pregnancy and postpartum: Global stability versus specific transitions. Annals
ofBehavioral Medicine, 21, 143-148.

Wyatt, S.B., Winters, K.P., & Dubbert, P.M. (2006). Overweight and obesity: Prevalence,
consequences, and causes of a growing public health problem. American Journal of the
Medical Sciences, 331, 166-174.

Yu, S.M., & Jackson, R.T. (1995). Need for nutrition advice in PNC. Journal of the American
Dietetic Association, 95, 1027-1029.


206









BIOGRAPHICAL SKETCH

Swathy Sundaram was born in Hamilton, Ontario, Canada. She moved with her parents to

the United States at the age of five and has lived in Florida since then. She always possessed an

interest in health and medicine, in both her personal and academic life. She graduated with a

bachelor's in microbiology/molecular biology from the University of Central Florida in 2002.

Interested in pursuing further education in health and medicine, she received a master's in public

health from Nova Southeastern University in 2004. While in her master's program, she

developed an interest in research. Wanting to pursue a field in which she could incorporate

health, medicine, and research, she enrolled in a doctoral program in health services research at

the University of Florida in 2005. During the course of the program, she developed a strong

interest in maternal and child health. In 2008, she became an American Public Health

Association maternal and child health student fellow from which, she furthered her interests in

maternal and child health. She received her doctoral degree in the summer of 2009.


207





PAGE 1

1 OBESITY AND POSTPARTUM DEPRESSION: DOES PRENATAL CARE UTILIZATION MAKE A DIFFERENCE? By SWATHY SUNDARAM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMEN TS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Swathy Sundaram

PAGE 3

3 To my parents

PAGE 4

4 ACKNOWLEDGMENTS I would like to express a very gracious thank you to my entire committee f or being my gurus, teaching me and passing on knowledge of everything they know, and providing their unlimited guidance and support all the way. I would especially like to thank my chair, Dr. Jeffrey Harman, for 1) always encouraging me to approach my di ssertation work in the most promising and challenging ways, 2) teaching me about health services research methods and providing much knowledge on the software that I used for my analyses, 3) aiding me and always taking the time to answer all of my question s and concerns during the progression of my dissertation work, and 4) providing extreme guidance and support both as my chair and my mentor. I would like to thank Dr. Mary Peoples -Sheps for 1) helping me find a dissertation topic, 2) directing me to a suit able dataset to carry out my dissertation and connecting me with the Centers for Disease Control, and 3) teaching me about prenatal care, and providing endless knowledge and sources on prenatal care, all of which were valuable to and strengthened this study. I would like to thank Dr. Allyson Hall, for her assistance with my theoretical framework, and for always cheering me on in my endeavor to graduate. Finally, I would like to thank Dr. Sharleen Simpson for helping to strengthen my dissertation with her va luable clinical experience. I woul d al s o l i k e t o t h a nk D r. P a u l D u n c a n fo r o ffering his words of wisdom when I needed guidance throughout the course of my doctoral education. To my parents, Dr. Kalpathy Sundaram and Mrs. Girija Sundaram, who have always believed in me and provided many means of support throughout my life. Their endless support and encouragement has allowed me to achieve many success es in life. To my husband, Mahesh Sundares an for being supportive all the way. To all of my friends, especially Karen Mounger, Kezia Awadzi, Keva Thompson, for always providing a shoulder to lean on, and for always reassuring me by saying, you can do it, and youll be fine. Finally, I would like to thank the

PAGE 5

5 Centers for Disease Control, particularly, Ms. Mary Rogers, Ms. Denise DAngelo, Mr. Brian Morrow, and the entire PRAMS committee, for without their generosity and assistance, I would not have had a dataset to carry out this dissertation.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 9 LIST OF FIGURES ............................................................................................................................ 11 ABSTRACT ........................................................................................................................................ 12 CHAPTER 1 STATEMENT OF PROBLEM .................................................................................................. 14 Why is Postpartum Depression of Concern? ............................................................................. 14 Why is Obesity of Concern? ....................................................................................................... 15 What is the Importance of Prenatal Care? ................................................................................. 16 Gaps in the Literature: Weight Issues Related to the Pregnancy and Postpartum Periods ..... 16 Gaps in the Literature: Prenatal Care and Postpartum Depression .......................................... 17 Purpose of This Study ................................................................................................................. 18 Specific Aims and Hypothese s ................................................................................................... 18 2 LITERATURE REVIEW ........................................................................................................... 21 Postpartum Depression (PPD) .................................................................................................... 21 Conse quences of PPD ................................................................................................................. 21 Screening for PPD ....................................................................................................................... 22 Postpartum Depression (PPD): Women and Men ..................................................................... 23 Obesity ......................................................................................................................................... 24 Psychosocial Consequences of Obesity: Stereotypes and Stigmas .......................................... 25 Psychosocial Consequences of Obesity: Weight as a Chronic Stressor .................................. 26 Obesity and Depression .............................................................................................................. 27 Obesity and PPD ......................................................................................................................... 29 Prenatal Care (PNC) .................................................................................................................... 30 Prenatal Care (PNC), Nutrition, and Weight: Behavior Modifications ................................... 32 What Should the Content of PNC Entail? ................................................................................. 34 Weight as a Stressor and the Importance of PNC ..................................................................... 37 Pregnancy and Excessive Weight Gain ..................................................................................... 38 Postpartum Weight Retention ..................................................................................................... 39 Relation Between Pregnancy, Weight, and Postpartum Distress ............................................. 41 Expected PNC Content Versus Actual PNC Content ............................................................... 4 5 Effectiveness of PNC .................................................................................................................. 45 Prenatal Care (PNC) and Postpartum O utcomes ....................................................................... 46 Theoretical Framework ............................................................................................................... 47

PAGE 7

7 3 METHODS .................................................................................................................................. 51 Data Overview: Pregnancy Risk Assessment Monitoring System (PRAMS) ........................ 51 Data Collection Procedures ........................................................................................................ 51 Weighting of Data ....................................................................................................................... 53 Rationale for Using PRAMS Data ............................................................................................. 53 Postpartum Depression (Dependent Variable) ................................................................... 54 Obesi ty (Main Independent Variable) ................................................................................ 57 Prenatal Care Utilization (Moderating Variable) ............................................................... 57 Control Variables ................................................................................................................. 59 Analysis ....................................................................................................................................... 59 Primary Risk -Adjusted Logistic Regression ...................................................................... 60 Specific aim 1 ............................................................................................................... 60 Specific aim 2 ............................................................................................................... 61 Secondary Risk -Adjusted Logistic Regression .................................................................. 61 Wald Test ............................................................................................................................. 63 Model Fit .............................................................................................................................. 63 Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Merging Strata ......................................................................................................................................... 63 Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Dropped Cases ......................................................................................................................................... 64 Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Imputed Data .......................................................................................................................................... 64 4 RESULTS .................................................................................................................................... 68 Univariate Analyses .................................................................................................................... 68 Bivariate Analyses ...................................................................................................................... 71 Multivariate Analyses ................................................................................................................. 75 Primary Risk -Adjusted Logistic Regression Analysis ...................................................... 76 Baseline model ............................................................................................................. 76 Specific aim 1 ............................................................................................................... 76 Specific aim 2 ............................................................................................................... 77 Secondary Risk -Adjusted Logistic Regression: Subpopulation With Healthy Pregnancies ....................................................................................................................... 77 Baseline model ............................................................................................................. 77 Specific aim 1 ............................................................................................................... 78 Specific aim 2 ............................................................................................................... 78 Wald Test ............................................................................................................................. 78 Model Fit .............................................................................................................................. 78 5 DISCUSSION ............................................................................................................................ 109 Univariate Analyses .................................................................................................................. 109 Bivariate Analyses .................................................................................................................... 113 Multivariate Analyses: Primary Risk -Adjusted Logistic Regression Analysis ..................... 118

PAGE 8

8 Baseline Model .................................................................................................................. 118 Specific Aim 1 ................................................................................................................... 120 Specific Aim 2 ................................................................................................................... 121 Multivariate Analyses: Secondary Risk -Adjusted Logistic Regression (Subpopulation With Healthy Pregnancies) ................................................................................................... 122 Baseline Model .................................................................................................................. 122 Specific Aim 1 ................................................................................................................... 124 Specific Aim 2 ................................................................................................................... 125 Summary of Multivariate Results ............................................................................................ 126 Limitations ................................................................................................................................. 129 Importance of This Study/Implications ................................................................................... 131 APPENDIX A SUMMARY OF LITERATURE ON PNC CONTENT ......................................................... 137 B MULTIVARIATE SUB -ANALYSES .................................................................................... 155 Adequate Plus ............................................................................................................................ 155 Sensitivity Analysis: Ordinal Logistic Regression ................................................................. 156 Women, Infants, and Children (WIC) ...................................................................................... 156 Income ........................................................................................................................................ 157 Weight Gain Discussion ........................................................................................................... 157 Results of Sub-Analyses ........................................................................................................... 158 LIST OF REFERENCES ................................................................................................................. 187 BIOGRAPHICAL SKETCH ........................................................................................................... 207

PAGE 9

9 LIST OF TABLES Table page 3 1 Specific aims, dependent, and independent variables .......................................................... 65 3 2 Classifica tion of body mass index (BMI) ............................................................................. 65 3 3 Step 1: Gestational age calculation into expected number of visits for the APNCU Index ........................................................................................................................................ 65 3 4 Step 2: Month of PNC initiation calculation into number of missed visits for the APNCU Index ........................................................................................................................ 65 3 5 Step 3: Categorization of index into categories for the APNCU Index .............................. 66 3 6 Characteristics of the the APNCU Index groups ................................................................. 66 3 7 Adequacy of Prenatal Care Utilization (APNCU) Index frequencies before recoding ..... 66 3 8 Adequacy of Prenatal Care Utilization (APNCU) Index frequencies of incorrect codings (observations incorrectly coded into other PNC utilization categories that were recoded into inadequate PNC utilization based on the month of initiation) .......... 66 3 9 Step 4: Adequacy of Prenatal Care Utilization (APNCU) Index frequencies after recoding all cases from Table 3 8 into inadequate PNC utili zation ................................ 66 3 10 Coding for raw income variable categories before collapsing categories .......................... 67 3 11 Coding for collapsed income varia ble categories ................................................................ 67 4 1 Univariate statistics for all categorical variables included in the bivariate and multivariate analyses .............................................................................................................. 79 4 2 Univariate statistics (continuous variable) ............................................................................ 82 4 3 Chi -square analyses comparing 41 characteristics among women with postpartum depressive (PPD) symptoms versus women without postpartum depressive (PPD) symptoms ................................................................................................................................ 83 4 4 Maternal age (continuous variable) and postpartum depression (PPD) symptoms t test results ............................................................................................................................... 88 4 5 Chi -square analyses comparing 40 characteristics among women from four body mass index (BMI) groups ...................................................................................................... 89 4 6 Chi -square analyses comparing 39 characteristics among women from four prenatal car e (PNC) utilization groups ................................................................................................ 95

PAGE 10

10 4 7 Primary baseline logistic regression with the main effect independent variables .............. 99 4 8 Specifi c aim 1: Primary risk adjusted logistic regression with the main effect independent variables and control variables ......................................................................... 99 4 9 Specific aim 2: Primary risk adjusted logistic regression with the ma in effect independent variables, interaction effect variables, and control variables ....................... 101 4 10 Wald tests for pre -pregnancy BMI/PNC interaction terms: Primary risk adjusted logistic regression ................................................................................................................. 103 4 11 Secondary baseline risk adjusted logistic regression (healthy pregnancies only) with pre pregnancy BMI and PNC utilization ............................................................................ 103 4 12 Specific Aim 1: Secondary risk adjusted logistic regression (healthy pregnancies only) with the main effect independent variables, interaction effect variables, and control variables ................................................................................................................... 104 4 13 Specific aim 2: Secondary risk adjusted logistic regression (healthy pregnancies only) with the main effect independent variables, interaction effect variables, and control variables ................................................................................................................... 105 4 14 Wald tests for pre -pregnancy BMI/PNC interaction terms: Secondary risk adjusted logistic regression (healthy pregnancies only) ................................................................... 107 B1 Chi -square analyses comparing 40 characteris tics among women who utilized adequate plus PNC versus women who utilized other quantities of PNC .................... 164 B2 Maternal age (continuous variable) and adequate plus PNC t -test results ....................... 171 B3 Logistic regression for adequate plus PNC to determine significant predictors of adequate plus PNC ............................................................................................................... 172 B4 Risk adjusted ordinal logistic regression sensitivity analysis with the main effect independent variables, interaction effect variables, and control variables ....................... 173 B5 Logistic regression for women who received WIC ser vices during pregnancy ............... 175 B6 Logistic regression for women with very low income ....................................................... 177 B7 Logistic regression for women with low income ............................................................... 179 B8 Logistic regression for women with moderate income ...................................................... 181 B9 Logistic regression for women with higher income ........................................................... 183 B10 L ogistic regression for women who received weight gain discussion .............................. 185

PAGE 11

11 LIST OF FIGURES Figure page 2 1 Theoretical framework ........................................................................................................... 50 4 1 Primary risk adjusted logistic regression with postpartum depression (PPD) symptom odds ratios for each interaction effect variable .................................................. 108

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy OBESITY AND POSTPARTUM DEPRESSION: DOES PRENATAL CARE UTILIZATION MAKE A DIFFERENCE? By Swathy Sundaram August 2009 Chair: Jeffrey S. Harman Major: Health Services Research Using a national continuing population -based survey known as Pregnancy Risk Assessment Monitoring System (PRAMS), this s tudy sought to determine the role that PNC utilization plays in the relationship between pre -pregnancy BMI and PPD symptoms. Two years of data, 2004 and 2005 were analyzed among women from 16 states. Two specific aims were examined: 1) the association betw een pre pregnancy BMI and PPD symptoms, and 2) the association between pre -pregnancy BMI and PPD symptoms after considering PNC utilization as a moderating variable. It was predicted for the first specific aim that the odds for PPD symptoms would increase as pre -pregnancy BMI increased. For the second objective, it was predicted that the association from the first specific aim would carry over and remain the same (e.g., obese pre pregnancy BMI would have the highest odds for PPD symptoms), but that within e ach pre pregnancy BMI group, the odds for PPD symptoms would decrease as PNC utilization increased (within obese pre pregnancy BMI, inadequate PNC would have higher odds than intermediate PNC). The general premise for PNC utilization acting as a moderating variable in this study was that PNC can help address the changes that occur during pregnancy with regards to pre pregnancy BMI (as a biological and psychosocial stressor). Thus, delivering PNC incorporating

PAGE 13

13 nutrition, weight and shape changes, and address ing a womans concerns about her weight and shape would in turn, reduce the odds of PPD symptoms. Since the sample used in this study included women from all pregnancy risk statuses, two risk adjustment approaches were carried out to identify an associat ion between pre -pregnancy BMI and PPD symptoms, and a moderating effect of PNC. One approach included all women in the dataset and used statistical analyses to risk adjust for pregnancy risk status, and the other approach modified the design of the study b y truncating the population of women to include healthy pregnancies only. Results initially showed an association between obesity and PPD symptoms, and PNC and PPD symptoms among the bivariate and multivariate analyses. However, the inclusion of a variety of control variables into the multivariate models removed these associations. Overall, for both approaches, there was no indication of a moderating effect of PNC utilization. However, results from the analyses showed that many of the women were significant ly affected by a variety of medical and obstetric problems, many of which were high risk. It is recommended that future research investigate the possible association of these problems with PPD symptoms. For practice, it is suggested that PNC providers identify the medical and obstetric problems faced by their patients, focus on both the physical and the potential psychosocial consequences of those problems, and establish suitable interventions accordingly.

PAGE 14

14 14 CHAPTER 1 STATEMENT OF PROBLEM Why is Postpartum Depression of Concern? Postpartum depression (PPD) is a mood disorder that involves a variety of symptoms including fatigue, fears, anxiety, despair, thoughts of compulsion, loss of libido, and feelings of inadequacy (Horowitz, Damato, Solon, Von Metzsch, & Gill, 1995). The relationship between a mother and baby is crucial for healthy maternal and child health outcomes. A woman can experience PPD anytime during the first year after the birth of her child (Epperson, 1999). Symptoms include mood swings, sadness, anxiety, loneliness, and inconsistent sleeping patterns. However, when these symptoms reach a level of intensity that begins to affect the well -being of a woman and her daily functioning, a woman should seek treatment as these symptoms may indicate PP D. A new mother may be unaware of 1) the normal physical changes that occur after giving birth, and 2) her ability to care for the infant (American Academy of Pediatrics (AAP) & American College of Obstetricians and Gynecologists (ACOG), 1992). Approximate ly 4 6 weeks after the delivery, the AAP & ACOG (1992) recommend that a woman should see her physician for a postpartum examination that includes evaluation of the mothers current health status and her adaptation to her infant. Since many women experience emotional distress to some extent in the postpartum period, 1) the emotional status of a woman should be evaluated, and 2) any counseling with regards to a womans postpartum emotional distress should address future health and future pregnancies. Consequences of PPD include maternal aggression, neglect of the infant, and infanticide (Reck et al., 2004). Other psychosocial factors associated with PPD include child care stress, poor marital satisfaction, and low self -esteem (Appolonio, & Fingerhut, 2008). M others with PPD are portrayed as being unresponsive to their infants, passive and intrusive, displaying

PAGE 15

15 15 avoidance and withdrawal, and displaying low levels of influence, or affect (Reck et al., 2004). The processes associated with childbearing (e.g., pregnancy, childbirth, childrearing, etc.) warrant attention because they remain responsible for many maternal morbidities and mortality (e.g., complications such as preeclampsia, hemorrhage, self acceptance in the postpartum period, etc.) (Misra & Grason, 2006). Thus, not only can these processes have a biological influence on a womans health, but they can also have a psychosocial influence on a womans health and well -being that occur during this time (Wisner et al., 2006). Why is Obesity of Concern? Obesity continues to burden our society in terms of increased prevalence of other diseases (e.g., heart disease), increased health care costs (e.g. treatment), and poses increased risk for disability and death. In addition to this, obesity presents many social, e motional, and aesthetic problems, especially in developed countries (Rubinstein, 2006). For women who are obese and pregnant, pregnancy related consequences of obesity that result from high pre -pregnancy body mass indices [(weight in pounds/square of heigh t in inches) x 703] include increased risk for other diseases (e.g., gestational diabetes) and a lower survival rate for premature babies (Colditz, 2002). Obesity and BMI have also been associated independently with delivery complications including excessi ve blood loss, greater operating time, and increased likelihood for cesearean section ( American College of Obstetricians and Gynecologists, 2005). The association between pre pregnancy BMI and postpartum depressive symptoms has been demonstrated with this association increasing as pre -pregnancy BMI increases (Carter, Wood Baker, Brownell, 2000; LaCoursiere, Baksh, Bloebaum, & Varner, 2006; Andersson, Sundstrom -Poromaa, Wulff, Astrom, & Bixo, 2006).

PAGE 16

16 16 What is the Importance of Prenatal Care? Since the health o f the infant is determined significantly by the health of the mother, addressing issues during pregnancy itself can minimize adverse maternal and infant as well as child outcomes later on. According to Healthy People 2010 (2000), prenatal care (PNC) should start early on in the pregnancy and continue all through the pregnancy period; the effectiveness of PNC is more likely if PNC is received early in the pregnancy. An ideal setting to discuss issues (e.g., weight) during pregnancy is during prenatal care. I t is suggested that pregnancy is a good time to target changes in health behavior due to a womans motivation to maximize the health of her child (Birdsall, Wvya, Khazaezadeh, & Otegn Ntim, 2009). Prenatal care (PNC) is defined as the care a woman receives in the period during pregnancy, leading up to the time she gives birth; adequate PNC is vital for both the mother and her developing baby (National Institute of Child Health and Human Development, 2007). Prenatal care (along with obesity) has been noted a s two of the four special concerns for womens health (Torpy, Burke, & Glass, 2006). Gaps in the Literature: Weight Issues Related to the Pregnancy and Postpartum Periods Regarding previous research, Walker Timmerman, Minseong, & Sterling (2002) found w eight as a leading factor for postpartum dissatisfaction among women of all ethnicities. However, their sample size was not nationally representative and did not include all income groups. Lebanon. Fox, & Yamaguchi (1997) found that women who were overweig ht before pregnancy were more likely to have positive changes in body image during pregnancy compared to normal weight; However, the women who were overweight before pregnancy also had more negative concerns about body shape than normal weight women. This study did not address associations between body weight/body shape and PPD, and the sample size was limited to women in London. Moran, Holt, & Martin (1997) found that among postpartum health concerns,

PAGE 17

17 17 the highest percentage of women in the sample wanted mo re information on nutrition, exercise, and dieting. However, the study did not look at PPD. LaCoursiere et al. (2006) found a significant association between pre -pregnancy obesity and moderate or greater postpartum depressive symptoms. However, though all BMI categories were included, their sample size was not nationally representative. Andersson et al. (2006) did not find any significant associations between first trimester BMI and a new -onset episode of postpartum depression. Also, BMI data was missing f or 8% of their sample and their sample size was limited to women in Sweden. Carter Baker & Brownell (2000) found an association between BMI and anxiety/postpartum depressive symptoms. However, their study had a small sample size and their BMI categories included those with BMI <27, and those with BMI >27. Gaps in the Literature: Prenatal Care and Postpartum Depression Chaaya et al. (2002) looked at the determinants of PPD. Though they commented on the importance of PNC in addressing the needs of pregnant women, they did not include PNC or BMI in their analysis, and their sample size was limited to women in Lebanon. El -Kak, Chaaya, Campbell, & Kaddour (2004) found that more PNC visits were associated with fewer cases of PPD. However, their sample size was also limited to women in Lebanon. Nalepka & Coblentz (1995) hypothesized that educating women on PPD prior to childbirth would reduce the likelihood of PPD among the women compared to the women who did not received education; however, they found no signifi cance and the education was delivered in childbirth classes during pregnancy as opposed to PNC, per se. Therefore, since 1) there are limitations of previous studies on weight and postpartum distress, 2) there is a paucity of literature on prenatal care a nd postpartum depression, and 3) many significant changes occur during pregnancy, changes which pregnant women can be

PAGE 18

18 18 educated on during the delivery of PNC, the relationship between PNC and PPD in the U.S can further be ascertained in the literature (Merr itt, Kuppin, & Wolper, 2001). Purpose of This Study Since studies have confirmed both an association between obesity and PPD, and PNC and PPD, I would like to combine these two relationships. I propose that PNC can be seen as a means for addressing weight concerns in order to ensure healthy pregnancy and postpartum outcomes for her and her baby. To my knowledge, no study has determined if 1) an association exists between pre -pregnancy BMI and PPD in the U.S., and 2) if an association exists between pre pre gnancy BMI and PPD after considering PNC utilization as a moderator. The purpose of this study is to determine if any existing association between pre -pregnancy BMI and PPD symptoms is weakened after considering a womans pre pregnancy body mass index. Si nce women from all pre pregnancy body mass index groups seek PNC, I would like to see if the association of pre pregnancy BMI and PPD symptoms differ by PNC utilization level. The primary rationale for PNC utilization acting as a moderating variable is bas ed on the premise that PNC can be seen as a means for providers to help women address any negative attitudes towards weight gain and body image that may develop during pregnancy. I hypothesize that PNC plays an important role in reducing the likelihood of PPD among women from all pre -pregnancy body mass index groups. Identifying 1) an association between pre -pregnancy BMI groups and PPD among pregnant women, and 2) seeing if the association differs for women from different PNC utilization groups can substan tiate the role (e.g., prevention) that PNC plays in reducing the likelihood of an adverse maternal and child health outcomes (PPD symptoms). Specific Aims and Hypotheses Little is known about the relationship of BMI and PPD after considering a womans PNC utilization. Since the literature has suggested 1) an association between pre -pregnancy BMI and

PAGE 19

19 19 weight gain during pregnancy, 2) an association between weight gain during pregnancy and weight issues postpartum (e.g. weight retention), 3) an association bet ween obesity and depression, 4) an association between pre -pregnancy BMI and PPD, and 5) an association between PNC and PPD, in accordance with the recommendations suggested in the literature, I propose that PNC can help assuage effects that pre pregnancy BMI (as a biological and psychosocial stressor) and the changes that occur during pregnancy may have on women, such as its influence on any negative attitudes about weight gain and body image that may arise either during pregnancy or in the postpartum peri od. Consequently, I predict that if health care providers deliver PNC information incorporating nutrition, weight and shape changes, and address a womans concerns about her weight and shape, this will in turn, reduce the likelihood of PPD. The objective of this study is to determine the importance of PNC in acting as a moderating variable in the relationship between pre pregnancy BMI and PPD (e.g., weight issues and lifestyle behaviors addressed through PNC may reduce the likelihood of weight issues expe rienced after delivery, and reduce the likelihood of PPD). The specific aims are as follows: Specific aim 1: What is the association of pre -pregnancy body mass index (BMI) with subsequent development of postpartum depression (PPD) symptoms? Hypothesis: I predict that women who had a pre -pregnancy BMI of obese will have the highest odds for PPD symptoms, followed by overweight BMI, and finally underweight BMI (lowest odds for PPD symptoms). Women who had a normal pre -pregnancy BM I will be the reference group. Specific aim 2: Does PNC moderate the relationship between pre-pregnancy BMI and PPD symptoms? Hypothesis: Within each pre -pregnancy BMI category, the likelihood a woman will experience PPD symptoms will decrea se as prenatal care increases: i nade quate PNC will have the highest odds for PPD symptoms, followed by adequate plus PNC, and finally intermediate PNC (lowest odds for PPD symptoms) Women who utilized adequate PNC will be the reference group

PAGE 20

20 20 Thus, if PNC acts as a moderating variable, as predicted, afte r looking at the relationship between pre -pregnancy BMI and PPD symptoms only, that relationship will change as the likelihood for PPD symptoms will change when considering PNC. For example, there will be a difference in a woman with a pre -pregnancy BMI of obese who received adequate plus PNC versus a woman with a pre -pregnancy BMI of obese who received intermediate PNC. The next chapter elaborates on the literature to propose an idea of what occurs during PNC in addressing concerns related to weight that a rise during pregnancy, explaining how this may in turn, reduce the likelihood of PPD symptoms.

PAGE 21

21 21 CHAPTER 2 LITERATURE REVIEW Postpartum Depression (PPD) Postpartum depression (PPD) is a mood disorder that involves a variety of symptoms including fatigue, f ears, anxiety, despair, thoughts of compulsion, loss of libido, and feelings of inadequacy (Horowitz et al., 1995). P P D is known to be a very common illness, and affects approximately one in every eight mothers to a point that affects her ability to carry out her maternal responsibilities (Wisner, Parry, & Piontek, 2002). PPD is divided into three categories: 1) blues, which affect roughly 50 80% of new mothers, and is considered to be normal, 2) nonpsychotic postpartum depression, which affects roughly 10 15% of new mothers, the incidence being on average 13%, and 3) postpartum psychosis, which is rarer than the other two types and occurs in roughly 12 out of every 1000 pregnancies or 0.1 0.2% of mothers (Miller, 2002; Evins & Theofrastous, 1997; Negus Jolley & Betrus, 2007). It is the conditions of labor, delivery, and the postpartum period that are predicted to bring about a traumatic level of stress that can trigger postpartum depressive symptoms (Dietz et al., 2007). Consequences of PPD Da Costa, Drit sa, Lowensteyn, & Khalife (2006) showed that women experiencing PPD suffered significant reductions in health related quality of life, with the association continuing even after controlling for depression severity. PPD has been shown to have negative cons equences on the childs behavior and development, mother -child interaction, and parenting practices (Minkovitz et al., 2005). PPD may also affect a mothers health care utilization for her child. For example, if fewer preventive measures are taken for the child such as lack of vaccinations, this can in -turn, affect the physical health of her child, which can bring about an increase in acute care utilization for the child (Minkovitz et al., 2005). This relationship is

PAGE 22

22 22 suggested because health promotion activities taken on by the mother for her child are associated in part with the functional capacity of the mother. The functional capacity may be affected if the mothers psychological well being is compromised. (Rahman, Iqbal, Bunn, Lovel, & Harrington, 2004). Also, children of a depressed parent have a higher likelihood of experiencing negative cognitive and social outcomes (e.g., lack of social competence), and have a higher rate of mental illnesses that can continue into adulthood, with this likelihood incre asing if both parents experience depression (NICHD Early Child Care Research Network, 1999; Lieberman, 1977; Weissman et al., 2006; Goodman & Gotlib, 1999). Since the primary figure in a childs life tends to be the parent, usually the mother in many famil ies, parental depression may affect the quality of the relationship between the parent and child, and even cause behavioral problems for the child later on in life, such as anxiety (Radke Yarrow, Cummings, Kuczynski, & Chapman, 1985; Lieberman, 1977). Chil dren of mothers with PPD also have a higher likelihood of receiving lower scores on measures of mental and motor development, have more difficult temperaments, react more negatively to stress, and lower self -esteem (Goodman & Gotlib, 1999). Finally, physic al consequences associated with depression for the mother around the time of childbirth include low birth weight and impaired growth for the child (Rahman, Iqbal, Bunn, Lovel, & Harrington, 2004). Screening for PPD Recognizing that mental health, as well a s physical health, is important for the mother is essential for her overall well being (Wisner et al., 2006). Since a new mother may be unaware of 1) the normal physical changes that occur after giving birth, and 2) any limits of her ability to care for h er infant (AAP & ACOG, 1992), screening is the first step in detecting PPD (Negus Jolley & Betrus, 2007). Approximately 4 6 weeks after she gives birth, the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists (1992)

PAGE 23

23 23 recommend that a woman should see her OB GYN for a postpartum examination to determine her current health status and her adaptation to her infant. Since many women experience emotional distress to some extent in the postpartum period, the emotional status o f a woman should be evaluated during this time. Also, any counseling regarding a womans postpartum emotional distress should address future health and future pregnancies. A wealth of literature exists that supports screening as an effective way to combat the consequences of PPD. I dentifying women at risk for PPD has been previously identified as a preventive method for PPD (Boyce & Hickey, 2005). A universal screening system for PPD is suggested which includes screening for PPD as soon as two weeks after birth and no later than a year after birth (Wisner et al., 2006). Questions however exist as to whether PPD screening actually leads to improved maternal and child health outcomes (Gaynes et al., 2005). It is advised that detecting for women who are at ri sk for PPD symptoms can be done in the late stages of pregnancy (Josefsson, Berg, Nordi, & Sydsjo, 2001). Screening for risk factors and/or depressive symptoms would be conducive to early detection and initiation of treatment (Miller, 2002). This screening can indeed be incorporated into both prenatal clinics (during PNC delivery) and in pediatric clinics (during postpartum check up visits) (Miller, 2002). Postpartum Depression (PPD): Women and Men Research has shown that women are two times more likely th an men to suffer from depression (with the dominance of depression affecting women consistent across developed nations), and that the first onset tends to be during the reproductive years (Weissman & Olfson, 1995). In addition, women tend to experience a l onger duration of depression and a higher frequency than that of men ( Sargeant, Bruce, Florio, & Weissman, 1990). Men experience PPD as well as women; however, the literature suggests that women, unlike men, have a higher likelihood of suffering PPD due to hormonal withdrawal (e.g., gonadal steroids such as estrogen

PAGE 24

24 24 and progesterone) experienced by a woman in the postpartum period, with this likelihood increasing with parity (Bloch et al., 2000). Maternal PPD is correlated with paternal PPD, and both can af fect family health by affecting other relationships within the family and eventually the well -being of the family (Deater Deckard, Pickering, Dunn, & Golding, 1998; Goodman, 2004). Since men are becoming more involved in the experience of having a newborn in the house, compared to previous decades, there are now greater possibilities for men to experience PPD (Goodman, 2004). Combined PPD of both the mother and father puts the child at an increased risk for developmental problems than would occur with mater nal PPD alone (Goodman, 2004). Obesity Obesity is defined as having a body mass index (BMI) of 30 or greater (National Institutes of Health, 1998). Body mass index, a measure that represents the comparative weight to height, is recommended by the Centers f or Disease Control (CDC) as a reliable body fat indicator (it is significantly correlated with the total fat content in the body) and as an excellent method for assessing both overweight and obesity (National Institutes of Health, 1998). Obesity has been s hown to be associated with an increased risk for other diseases, including hypertension, cardiovascular disease, cholecystectomy, non-insulin dependent diabetes mellitus, and colon cancer (National Institutes of Health, 1998). It is estimated that about 65% of Americans 21 years and older have a BMI more than 25 (overweight classification), 30.5% have a BMI of 30 or more (obese classification), and 4.9% have a BMI of 40 or more (extremely obese classification) ( Sarwer, Allison, Gibbons, Markowitz, & Nelson, 2006). Rates of obesity continue to increase, especially among childbearing women. According to Lu et al. (2001) the average maternal weight of women in the initial prenatal care visit increased by 20%, and the percentage of women classified as obese incr eased from 7.3% to 24.4% over a period of 20 years. Along with the growing rates of obesity and the physical consequences of obesity on

PAGE 25

25 25 health, there exist many psychosocial consequences that can have effects on the mental health of individuals who are obese. Psychosocial Consequences of Obesity: Stereotypes and Stigmas In addition to its effects on physical health as identified above, obesity also poses psychosocial consequences. There exists a plethora of negative attitudes and stigmas associated with b eing obese. Examples include discrimination and prejudice with respect to arenas such as health care and employment (Crerand, Wadden, Foster, & Gary, 2007). This discrimination may have health care access implications for obese postpartum women who are in need of mental health care to address PPD. According to Wooley & Wooley (1979), and apart from skin, having excess body fat is known to be the most stigmatized physical feature. However, unlike skin, excess body fat can be voluntarily controlled. The resul ts of stigmatization of overweight/obese individuals include self -victimization and having to suffer differential treatment due to physical appearance, with women suffering more than men (Wooley & Wooley, 1979). Stereotypes affiliated with being overweight and obese include being self -indulgent, less self -disciplined, less attractive, less happy, and lazier (compared to thin counterparts) (Tiggeman & Rothblum, 1988). Also, it is noted that obesity is more prevalent in women than it is in men (Hedley et al., 2004). Women have double the likelihood over men to experience a major weight gain with an increase over a period of 10 years and overweight women in the 2544 year age group have the highest incidence for significant weight gain a compared to all other groups (Williamson, Kahn, Remington, & Anda, 1990); these years are crucial regarding child-bearing years. It is suggested that although both men and women experience stigmatization with obesity; the effect is more profound towards women than for men as wom en possess a greater propensity for obesity (Hebl & Turchin, 2005; De Garine & Pollock, 2005).

PAGE 26

26 26 Psychosocial Consequences of Obesity: Weight as a Chronic Stressor Looking at attitudes and beliefs held by many women, there are many chronic stressors that Am erican women experience on a daily basis. The association of a slender figure with attractiveness and beauty and its effects on womens body perceptions and body images remains one of the more significant chronic stressors (Attie & Brooks -Gunn, 1987). This is due to cultural and societal influences that can be influential to an extent that a woman does not question the validity of her perceptions that she is overweight or has an undesirable figure; even if she has alternatives to negative body perceptions ( e.g., accepting her weight and body shape), she may face confrontation to these alternative perceptions from outside influences (e.g., peers, spouse, etc.) (Attie & Brooks Gunn, 1987). For women, especially in the Western world, the body image tends to support a slim body, and there continues to exist a dilemma between desire versus control (De Garine & Pollock, 2005). In fact, many American women tend to experience negative ramifications when their weight exceeds societys expectations (Cameron et al., 1996). What is socially accepted is a thin, lean body, which remains symbolic of characteristics such as self -control, hard work, attractiveness, success, acceptance, being physically fit and healthy, and in general, having desirable personal qualities (Brow nell, 1991). The phrase thin is beautiful tends to lead to fear of and prejudice against overweight/obese individuals (Attie & Brooks Gunn, 1987). Women tend to experience greater pressures to conform to being thin as there are more positive attitudes to wards thinness (De Garine & Pollock, 2005). Karlsson, Taft, Sjostrom, Torgerson, & Sullivan (2003) found that obese women reported having more psychosocial problems related to weight than men. Societys focus on body image can take a toll on a womans emot ional and physical health to an extent where thinness may take primacy over health (Paquette & Raine, 2004). For example, this focus can cause many women to 1) undertake dangerous weight loss behaviors, 2) experience poor body image and low self -esteem iss ues, and

PAGE 27

27 27 3) general unhappiness that could lead to depression (Battle & Brownell, 1996). In fact, many studies have confirmed as association between obesity and depression. Obesity and Depression Not only is obesity known as the most common chronic illnes s in todays society, but depression is considered to be the second most prevalent psychological illness in todays society, following anxiety disorders (Dixon, Dixon, & OBrien, 2003). Depression is known as the leading cause of disability globally (Kruij shaar, Hoeymans, Spijker, Stouthard, & Essink Bot, 2005). Depression is an independent risk factor for premature morbidity and mortality, especially when combined with congestive heart failure, hypertension, and/or stroke, and is associated with higher hea lth care utilization and higher total healthcare expenditures, and loss of productivity (Schulz et al., 2000; Olfson & Klerman, 1992; Greenberg et al., 2003). Both obesity and depression are known to be among the most prevalent and most costly public heal th problems in the United States, and are associated with increased health care utilization, which can in turn result in increased health care costs (Kress, Peterson, & Hartzell, 2006). The combined effect of both can increase an individuals risk for loss of function as both may affect one another (Markowitz, Friedman, & Arent, 2008). There is increasing evidence that obesity and depression are related; those who are overweight or obese are more likely to feel depressed at least one week during the month ( Aberdour, 2006). Plutchik (1976) found that the greater the degree of being overweight, the greater the tendency to experience problems with depression. Roberts, Deleber, Strawbridge, & Kaplan (2003) found that obesity at baseline was associated with depre ssion and that obesity predicts depression subsequently. Since many obese individuals are forced to endure discrimination and the stigmas associated with obesity, this can certainly contribute to psychosocial distress, if not depression; however, Dixon et al. (2003) confirmed that obesity is associated with depression. Increased BMI is associated with a higher

PAGE 28

28 28 risk for depression, thoughts of suicide, and suicide attempts (Carpenter, Hasin, Allison, & Faith, 2000). Onyike, Crum, Lee, Lykestsos, & Eaton (2003) found that obese persons in their sample had a higher prevalence of depression in the past -month than their normal weight counterparts. For the women in the sample, there was an 82% increase in the odds compared to the 73% higher odds in men (which was non -significant). The prevalence of depression was the highest for those obese who were obese with the strongest association remaining for those who were severely obese (BMI >/= 40). Since treatment for both depression and obesity is costly and only avail able to few people prevention is important in curbing obesity (Battle & Brownell, 1996). Women also tend to be at a higher risk for obesity -related costs, and there is an increased risk of depression among women who are obese (Kress, Peterson, & Hartzell, 2006). Research has shown that obesity is associated with depression in females (Dong, Sanchez, & Price, 2004). Heo, Pietrobelli, Fontaine, Sirey, & Faith (2006) found that among young women, those overweight and obese were significantly more likely to ha ve experienced depressive moods compared to young women who were not overweight or obese, with those who are Hispanic being more susceptible. Linde et al. (2007) found that being overweight or being obese is associated with depression, especially among wom en. Jorm et al. (2003) found that in women, obesity was associated with more depressive symptoms and lower well -being whereas in men, the associations were weak and inconsistent. Studies have also shown that being underweight is associated with depression Lox, Osborn, & Pellet (1998) found that women who perceive themselves as underweight experience similar psychosocial issues (e.g., self -esteem, depression, body dissatisfaction, anxiety) as women who perceive themselves as overweight. Carpenter et al. (2 000) found a U -shaped relationship among BMI values and their association with increased probability of depression;

PAGE 29

29 29 low and high BMI values were associated with an increased probability of having experienced major depression. However, s ince the relationshi p between obesity and depression has been demonstrated to a greater extent in the literature, suggesting that obese women have a higher likelihood for experiencing mental distress and depression associated with weight, taking this relationship further and looking at a special group of women, postpartum women, could further add to the literature on the effects of weight, specifically obesity, on mental health; in this case, during the postpartum period. Obesity and PPD A woman undergoes a significant amoun t of stress during the course of pregnancy. Since giving birth is an important time in a womans life and safely delivering a healthy baby is a critical concern, this can be stressful for a woman in ensuring a healthy pregnancy outcome. In addition, the ef fect of pregnancy on body perceptions and/or self -esteem can also act as a chronic stressor (Hobfoll & Leiberman, 1987). In fact, even though dietary restraints are less prevalent in pregnant women (Davies, & Wardle, 1994), body perceptions of women tend t o be increasingly negative, especially during the early to mid -second trimester of pregnancy (Skouteris, Carr, Wertheim, Paxton, & Duncombe, 2005). Though some research has shown that these perceptions may become less negative in the postpartum period, bod y perceptions during the postpartum period tend to be less positive than before pregnancy (Strang & Sullivan, 1985). In fact, negative body images are likely to be associated with weight distress during the postpartum period. Obesity has been shown to have a significant association with a new onset episode of postpartum psychiatric disorder (LaCoursiere, Baksh, Bloebaum, & Varner, 2006; Andersson et al., 2006). In addition, after controlling for marital status and income, pre -pregnancy obesity (defined as h aving a BMI greater than 29) was found to be associated with having moderate or greater

PAGE 30

30 30 postpartum depressive symptoms, with the strength of this association increasing as body mass increased above the normal BMI range (LaCoursiere et al., 2006.) Prenatal Care (PNC) It is believed that what is vital for every person is his/her health and the aptitude to work efficiently. An important resource for society is an infant who is born with the ability to function well in society. If an infant is born at a disadva ntage, with a condition that may prevent maximum functioning, this may be detrimental for the individual and the community. The period of pregnancy is a time that provides an opportunity to address lifestyle behaviors that remain important to a woman both during the course of her pregnancy and after she gives birth (e.g., smoking, nutrition, exercise, violence, etc.), many of which may have implications for infant and child outcomes (e.g., maternal obesity is a strong predictor for metabolic syndrome among children) (McCormick & Siegel, 2001; Boney, Verma, Tucker, & Vohr, 2005). One venue that may be used to alter the health behaviors of pregnant women is the advice and encouragement rendered by health care providers through prenatal care (PNC) (Kogan, Kotel chuck, Alexander, & Johnson, 1994). Prenatal care may be seen as a means to allow women to participate in their own health; an example being the change of their health behaviors during pregnancy to incorporate healthy eating (McCormick & Siegel, 2001). Ame rican women deem routine PNC to be essential as they strongly believe in the importance and efficacy of PNC and will invest efforts to make a good baby (Press & Browner, 1997; Rubin 1984, p.65); that is the message delivered by their health care provider s, even though there is limited evidence to support the direct benefits of routine PNC on birth outcomes. For many women, the most significant aspect of PNC observed is that they are provided with information about the pregnancy and the growing fetus. This is information that women find to be encouraging and empowering (Press & Browner, 1997). The purpose of PNC is to 1) find the pregnant women with problems, 2) assure

PAGE 31

31 31 management of the problems identified, 3) prepare both women and their partners for deliv ery and child care, 4) provide information, 5) provide health education, and 6) provide support to all pregnant women (Hemminki, 1988). Thus, PNC includes services that are intended to improve outcomes for the mother and infant, as well as promote educated decision -making among the mother, family members and friends, with regard to health care during the pregnancy (Daniels, Fuji Noe & Mayberry, 2006), and even in the postpartum period. Prenatal care is known as a key preventive service for pregnant women ( Kogan, Alexander, Kotelchuck, Nagey, & Jack, 1994) and has been accepted as an important conduit to prevent harm for the mother and child. Prenatal care is rendered in a variety of settings including: 1) private clinics of physicians, osteopaths, and midwi ves, 2) university hospital clinics, 3) health maintenance organizations, 4) community health centers, 5) public health departments, 6) migrant health centers, 7) community hospital clinics, 8) university hospital clinics, 9) schools, and 10) military faci lities. Those involved with the delivery of PNC include family practice physicians, obstetricians and gynecologists, midwives including nurse midwives, osteopaths, nurses, and nurse practitioners (Peoples Sheps, Kalsbeek, & Siegel, 1988). It is important to note that for many women who are pregnant for the first time, PNC may be the first point of adult contact with the health care system; thus, their experiences with PNC may influence subsequent use of the health care system for themselves, their partners and their children (Alexander & Kotelchuck, 2001). For all the women who seek PNC, most of them see a physician at some point during the course of their pregnancy; however, many women see multiple providers. (Peoples Sheps et al., 1988). Given that women are provided with information from their health care providers during the course of their pregnancy, it is believed

PAGE 32

32 32 that pregnancy is a time when a woman has the highest likelihood to make changes in her lifestyle behaviors, more than any other time during her life (Higgins, Frank, & Brown, 1994). Prenatal Care (PNC), Nutrition, and Weight: Behavior Modifications It is confirmed that the pregnancy period is a time when women make significant changes in their health behaviors (Baric & MacArthur, 1977) as m ost women are motivated to do what is necessary to enhance the likelihood of having a healthy baby (Higgins & Woods, 1999); for example, many women are motivated to change and/or modify their nutrition and fitness health behaviors during pregnancy (Wood Ba ker, Carter, Cohen, & Brownell, 1999). Pregnancy is also an ideal time to encourage women to initiate healthy lifestyle behaviors such as exercise (e.g., walking) and dietary habits (e.g., proper foods) (Morin & Reilly, 2007). The most common modifications made during pregnancy include exercise, nutrition, and reduced substance abuse behaviors (Higgins, Clough, Frank, & Wallerstedt, 1995). Kline, Martin, & Deyo (1998) found that women reported that the pregnancy and postpartum periods motivated them to redu ce any risky behaviors due to the fear that their children would be affected. The health behaviors that occur, the quality of the diet consumed, and the amount of weight gained during the pregnancy period are significant because of 1) its impact directly o n the health and well -being of the mother (both short -term and longterm), and 2) its impact on the development of the growing fetus (Robb Todter, 1996). Thus, commonly addressed concerns during pregnancy include weight gain and nutrition (e.g., nutrient i ntake) because both can affect the health of the mother and infant. Chomitz, Cheung, & Lieberman (1995) suggest that adopting healthy lifestyle behaviors during pregnancy can result in positive longterm health for the women and their infants/children. For example, having a balanced diet is important for women in their reproductive years, especially for women who are pregnant, in order to enhance the health, survival, and development of their children (Mora & Nestel, 2000). Among the things a woman

PAGE 33

33 33 can do t o heighten the likelihood of giving birth to a normal, healthy child are modifying lifestyle behaviors such as unhealthy dietary habits that may pose a risk of affecting the likelihood of delivering a healthy infant (Chomitz et al., 1995). Habits that are detrimental can be difficult to modify, but can be done with the support and assistance from family members and other close individuals, the health care system, and society; for example, modifications can be accomplished with the assistance of education th at is relayed through PNC (Chomitz et al., 1995). In fact, many women seek pregnancy related information through their PNC provider. For example, Risica & Phipps (2006) noted that the information topics that were most frequently requested by women to discu ss with their PNC provider were eating well and staying fit, followed by caring for a newborn, breastfeeding, healthy weight gain, gestational diabetes, genetic testing for their baby, and smoking cessation. Other requests for information included depressi on during and after pregnancy, working after giving birth, and preterm labor. Though the success of using written materials to deliver nutrition education has been demonstrated in the literature (Beresford et al., 1997) found that a self help book endorsed by physicians and given to patients who were looking to modify their dietary lifestyle habits was successful in helping those patients decrease their fat intake over the course of a year and increase their fiber intake), the authors found in their study t hat the women preferred to receive this information from a PNC nurse or provider rather than from other materials (e.g., printed materials, videos, classes, internet, CD ROM). Thus, it is the PNC providers that can educate, assist guide, and work with thei r pregnant patients in helping them make healthy behavioral changes and/or modifications related to nutrition and wellness.

PAGE 34

34 34 What Should the Content of PNC Entail? In order to encourage pregnant women to modify and/or adopt healthy behavioral changes relat ed to nutrition and wellness, it is ideal for PNC to be comprehensive and inclusive of care that prioritizes nutrition and wellness. PNC content is generally comprised of prevention (e.g., education), detection (e.g., birth defects), and treatment services as well as interventions designed to focus on psychosocial issues (e.g., stress) and change health behaviors that may prevent healthy pregnancy outcomes (e.g., poor eating habits). Lederman, Alfasi, & Deckelbaum (2002) recommend that PNC should include a significant focus on helping women optimize their weight during pregnancy (e.g., informing women about the Institute of Medicine guidelines on weight gain based on BMI). Many prenatal interventions such as PNC have focused on overweight/obese women in efforts to prevent excessive postpartum weight retention/weight gain (Walker, 2007). Prenatal care discussions should also include what it means to gain weight during pregnancy and how the weight is distributed in the womans body (e.g., between the uterus, pl acenta, fetus, etc.). Also, discussions should include well -balanced diets that are high in protein and would consequently have positive effects on a womans body shape and her weight (Moore, 1978). Since the content of PNC had grown to include services r elated to nutrition and are considered a vital part component of PNC (Wheatley, Kelley, Peacock, & Delgado, 2008), nutrition education and guidance should be a critical component of all PNC services (Bronner & Baldwin, 1999; Klohe -Lehman et al., 2006). Bot h the Institute of Medicine (1990), and the American College of Obstetricians and Gynecologists (2005) have issued recommendations for obstetricians -gynecologists including: (Institute of Medicine, 1990): 1 ) Health care providers should use reliable procedur es for measuring height and weight of pregnant women at each visit, set goals for weight gain, and monitor weight gain throughout the term.

PAGE 35

35 35 2 ) Calculate the womans pre -pregnancy BMI 3 ) Estimate the womans gestational age 4 ) Determine a weight gain goal together with the woman at the beginning of her initial prenatal care visit 5 ) Explain to the woman why weight gain is important 6 ) Discuss the recommended range of weight gain and the pattern of weight gain depending on her pre -pregnancy BMI: 25 35 lbs for normal weigh t women, 1525 lbs for overweight women, and 15 lbs for obese women, record height and weight for women at all PNC visits 7 ) Monitor the womans pattern of weight gain throughout her term and identify any abnormal patterns that may necessitate the health care provider intervening. 8 ) Upon identifying any abnormal patterns (if applicable), determine the cause of the abnormal weight gain and then determine ways to rectify the problem with the woman. 9 ) Evaluate a womans dietary habits via a food history or a food fr equency questionnaire; include questions about problems or conditions which may affect her dietary habits and behaviors. 10) Offer nutrition consultation to obese women and encourage them to adhere to an exercise program during the pregnancy and postpartum, discuss pregnancy related complications due to weight. Overall, it is recommended that PNC care include 1) a routine dietary assessment to determine dietary needs (e.g. nutrient supplementation), and 2) guidance and support for women on achieving a healt hy, balanced diet and maintaining healthy behaviors that will support adequate weight gain optimal health for the women and their fetuses (Institute of Medicine, 1990). An association between weight gain advice from PNC providers during pregnancy and actu al weight gain during pregnancy has been demonstrated, suggesting that women can be successfully encouraged to gain the appropriate amount of weight during their pregnancy (Taffel, Keppel, & Jones, 1993). Keppel & Taffel (1993) showed that White women who had pregnancy weight gain within the Institute of Medicines guidelines retained fewer than four pounds in the postpartum period, and had a median of 1.6 pounds. However, White women who gained more

PAGE 36

36 36 weight during pregnancy than the recommended ranges had a higher likelihood of retaining nine pounds or more (with a median of 4.9 pounds) compared to the women who did not exceed the recommended amount of weight gain. As for Black women, they were more likely to retain weight and were thus heavier in the postpa rtum period than White women (median of seven pounds) even though the gestational weight gain was similar for both groups of women, and they also consume higher total calories, a diet with a higher portion of calories from fat, and less physical activity d uring the prenatal and postpartum periods than White women (Keppel & Taffel, 1993; Boardley, Sargent, Coker, Hussey, & Sharpe, 1995) Also, as prenatal weight increased, the postpartum weight retained increased. Among 4,218 women in the sample, Carmichael Abrams, & Selvin, (1997) found that 40% of the women gained weight within the recommended ranges; out of this 40% of women, 53% of them were underweight BMI, 35% of them were normal BMI, 24% of them were overweight BMI, and 27% of them were obese BMI. W hile delivering PNC related to nutrition and wellness, PNC providers should be aware and knowledgeable about the attitudes and feelings that many pregnant women hold regarding body image and perceptions, especially closer to childbirth and in the postpartu m period, as they tend to be negative during these times (Moore, 1978; Stein & Fairburn, 1996). It is the PNC providers that can assist women in viewing pregnancy and its associated physical changes as ones that are normal and beautiful (Moore, 1978). It is recommended that health care professionals intervene early in the pregnancy to assist women in modifying their dietary patterns to achieve appropriate weight during these periods (Lederman, Paxton, Heymsfield, Wang, Thornton, & Pierson, 1997); Since wei ght can be a stressor for a number of women during pregnancy and in the postpartum period, health care professionals can 1) help women engage in healthy behaviors to minimize weight retention in the postpartum period, and 2) help

PAGE 37

37 37 women psychosocially (e.g. help women feel better about themselves and their weight) during pregnancy and in the postpartum period. Weight as a Stressor and the Importance of PNC Affonso & Mayberry (1990) found that the commonly reported stressors of pregnant women included weigh t gain and body changes for women in their first and third trimesters (second most commonly reported stressor), and for women in the postpartum period (fourth most commonly reported stressor). Among the total sample, body image changes was the second highe st commonly reported stressor. The authors suggest that assessments made in both the prenatal and postpartum periods must address these stressors and determine what they mean to women with regards to body image and body perception judgments, and how they c an produce discomfort and uneasiness if a woman feels that she is losing control over managing her body. Addressing these issues during the prenatal period (e.g., during PNC) may assist women in handling the intense emotions that follow childbirth and the imbalances that occur between stressors commonly experienced after childbirth (e.g., weight and body image issues) (Hiser, 1987). It is known that pregnant women experience a transition as their babies are developing, and this transition includes changes in body images and even during pregnancy (as well as the postpartum period), there is a significant amount of attention and importance that is given to appearance as is in the pre pregnancy period (McCarthy, 1998). The physical changes that occur during pr egnancy are extensive due to 1) growth of the woman due to growth of the child, and 2) changes in boundaries of the body that occur in the third trimester of the childbearing period (e.g., thinning of the abdominal and uterine walls that make the abdomen a nd uterus tight due to stretching) (Rubin, 1984). The resulting stress that may occur due to changes in body shape and size may cause distress both during the pregnancy and in the postpartum period (OHara,

PAGE 38

38 38 Schlechte, Lewis, & Varner, 1991). For some women the stress may be more profound if the weight gained during pregnancy is excessive. Pregnancy and Excessive Weight Gain Women who gain excessive weight during pregnancy may be at a higher likelihood to retain the weight after giving birth. For example, Lederman, Alfasi, & Deckelbaum (2002) found that women who were obese prior to pregnancy were more likely to experience excessive weight gain during pregnancy and postpartum. Wells, Schwalberg, Noonan, & Gabor (2006) showed the following: 1) being underwei ght was associated with inadequate weight gain, but protective for excessive weight gain during pregnancy, 2) being obese was associated with both excessive and inadequate weight gain during pregnancy, and 3) being overweight was associated with excessive weight gain, but protective against inadequate weight gain during pregnancy. Olafsdottir, Skuladottir, Thorsdottir, Hauksson, & Steingrimsdottir (2006) found after comparing women with a BMI less than 25, and 25 or greater that those in the latter category, specifically those with a pre -pregnancy BMI between 25 29 were the most likely to gain excessive weight during pregnancy. Thus, special attention should be given to women who are overweight before their pregnancy because they are the most likely to exper ience excessive weight gain during pregnancy. Consequently, they are also the most likely to experience pregnancy and delivery complications such as preeclampsia ( LaCoursiere, Bloebaum, Duncan, & Varner, 2005; Saravanakumar, Rao, & Cooper, 2006; Cedergren, 2004; Rosenberg, Garbers, Lipkind, & Chiasson, 2005; Rosenberg, Garbers, Chavkin, & Chiasson, 2003; Mahmood, 2009; Baeten, Bukusi and Lambe, 2001; Cnattingius, Bergstrom, Lipworth, & Kramer, 1998) as well as struggle with overweight/obesity issues after b irth. These struggles may consequently result in weight distress in the postpartum period due to issues such as weight retention (e.g., rigorous dieting) (Olafsdottir et al., 2006; Shepard, Hellenbrand, Bracken, 1986).

PAGE 39

39 39 Postpartum Weight Retention In gener al, many women are concerned about excessive weight gain during pregnancy due to their apprehension about postpartum weight retention (Keppel & Taffel, 1993). Excessive weight gain during pregnancy can cause obesity issues postpartum, which may cause wome n to restrict their food consumption in efforts to lose weight quickly. These restrictions inturn may weaken breastfeeding capabilities, and thus, obese women may cease breastfeeding earlier than non -obese women due to weight issues (Lederman, Paxton, Heymsfield, Wang, Thornton, & Pierson, 1997). According to Gunderson & Abrams (2000), factors that influence postpartum weight retention include pre -pregnancy weight, race/ethnicity, parity, lactation capabilities, and weight gained during pregnancy. Jenkin & Tiggeman (1989) found that women were more dissatisfied with their postpartum weight than their pre -pregnancy weight, and that postpartum weight was associated with psychological well -being with this association between weight and dissatisfaction increasi ng as weight increased. The authors concluded that postpartum weight is a predictor of psychological well being in the postpartum period. Polley, Wing, & Sims, (2002) found a significant and strong association between weight gain during pregnancy and postpartum weight retention. They suggest that normal -weight women tended to retain less than the control group, and for overweight women, they tended to retain more compared to the control group. Lederman, Paxton, Heymsfield, Wang, Thornton, & Pierson (1997) suggest that high pregnancy weight gain is likely to be associated with postpartum weight because many women have difficulty in adjusting to energy intake and expenditure during those periods. It is suggested that the effects of postpartum weight retentio n should be studied from a psychosocial context with an emphasis placed on weight management (Walker, 1997). This is because there is a significant amount of distress (associated with higher BMI) that is experienced due to a high dissatisfaction with weigh t; and this may be followed by a lowered self -esteem

PAGE 40

40 40 (Walker, 1997). Walker, Timmerman, Kim, & Sterling (2002) found that weight was the area that women in the postpartum period experienced the most dissatisfaction, followed by distress about the waist, hi ps, legs, and muscle tone (all of which tend to be areas where fat is retained in the postpartum period). This dissatisfaction with body image was significantly associated with postpartum depressive symptoms at six-weeks postpartum among women from all ethnic groups. However, Suttie (1998) found that postpartum women were more concerned about their fitness, less concerned with their appearance, and they also felt healthier compared to non-postpartum women. Results also showed that women in the early postpar tum period had similar body images to women in the latter part of the postpartum period; however, the women in the earlier postpartum period reported feeling healthier than the women in the latter part of the postpartum period. Other changes expressed by t he women in the postpartum period, though not as distressing as changes with weight and figure included stretch marks, wrinkly skin, and discoloration marks. Hiser (1987) found that concerns of women in the second postpartum week included the following: we ight was reported as a worry by 35% of the women, and 40% of the women reported concern regarding having a flabby figure, while stretch marks were reported by 70% of the women as not being a general concern in the postpartum period. In general, the women i n the sample cited weight, flabby figure and returning their figures to normal as frequent concerns in the postpartum period. Given the extent that postpartum weight retention is suggested to pose long-term effects on a womans health and well -being (e.g. weight increases later on in life) (Walker, 2007), PNC providers should address this issue during the delivery of PNC, Many women are unaware of the postpartum consequences of not managing weight during pregnancy. Since the body rarely immediately return s to its preconception shape following childbirth, many women tend to be

PAGE 41

41 41 surprised by the extent to which they retain weight after childbirth (Stein & Fairburn, 1996; Wood Baker, Carter, Cohen, & Brownell, 1999). For example, Fairburn & Welch (1990) found in their sample that 38% of the women had no intention of trying to lose weight as they felt their weight would return to normal. It is important for weight gain to be monitored during pregnancy in order to determine if it is within recommended ranges. PN C providers should educate their patients on how to lose weight following delivery in addition to educating women on how much weight to gain during pregnancy (Keppel & Taffel, 1993). Thus, it is important that pregnant women be given information related to food, nutrition, and weight during pregnancy. Olafsdottir et al. (2006) recommend that women should be given guidelines about weight gain and lifestyle modifications during pregnancy. Along with weight gain guidelines, pregnant women should also receive a dvice on changes with respect to eating patterns and habits, as well as changes in body shape and image that that are common among pregnant women. It has been suggested that women who are overweight and/or obese may overestimate their prenatal physical act ivity (Lichtman et al., 1992); Hence, if PNC providers work closely with their pregnant patients, monitor physical activity, and engage in discussion about weight, nutrition, and physical activity at PNC visits, this may, in turn prepare women (e.g., less distress) for managing changes in body shape and weight in the postpartum period (e.g., women who are encouraged to engage in healthy behaviors and do so during pregnancy may carry on these behaviors into the postpartum period). Relation Between Pregnancy, Weight, and Postpartum Distress Russell (1974) found that worrying about loss of figure was among the top five concerns experienced by women in the postpartum period and thus, this can significantly affect a woman during this time. Harris, Ellison, & Cle ment (1999) found that women tend to experience more dissatisfaction with their body during the postpartum period than they were in the preconception

PAGE 42

42 42 period. Four reasons may explain these feelings: 1) It is possible that mothers may experience a stronger drive towards thinness in the postpartum period compared to the preconception period, 2) Mother may perceive that they are heavier postpartum than they were before pregnancy; this may be due to an increase in caloric consumption that often accompanies pregnancy and may carry over into the postpartum period, 3) Mothers may actual be heavier in the postpartum period than they were before pregnancy, and 4) Mothers may put their preconception figures on a pedestal, hence, having a stronger desire to return to t hose figures after childbirth. Thus, because many women worry about returning to the normal weight and body shape during the postpartum period, they may possess negative attitudes and feelings towards their bodies (Strang & Sullivan, 1985). Strang & Sulli van (1985) found that the women in the sample experienced more negative feelings towards their body image during pregnancy than they did in the preconception period. However, in the postpartum period, the women experienced more positive feelings towards th eir body image than they did in the last trimester of their pregnancy. Since body image is comprised of many components (e.g., physical appearance such as weight and skin, posture, sense of fashion, etc.), it can affect a womans personality, self -image, identity, and behaviors and determine they way she responds to input from others and society (e.g., media, positive feedback, respectively). Changes in body images can be a reflection of what society and others define as beautiful and/or acceptable. Women are highly influenced by the feedback received from others and what society tells us is the paradigm of being fit and attractive; many American women connect beauty and attractiveness with success (Moore, 1978). It is important to note that pregnant women tend to react stronger to feedback received about their body image more so than non -pregnant women, as many feel the need for the endorsement from society regarding their weight gain and body shape changes in order to feel

PAGE 43

43 43 that she is successfully experien cing the pregnancy and is capable of becoming a mother (Moore, 1978). The standards in our society today for weight and body size and shape do not allow women to feel proud of their pregnant and postpartum bodies (Jenkin & Tiggemann, 1997). Therefore, it i s important that weight and body image issues be addressed in the pregnancy and the postpartum periods. Women tend to feel that the physical effects of pregnancy on the body (e.g., weight gain, breast changes) can bring about changes in self -esteem and bod y image; for some women, these changes are negative (e.g., negative body image, lowered self -esteem), and for others, these changes are positive (e.g., interpreting the changes as successful nurturing of the fetus, resulting in increases in self -esteem) (K line, Martin, & Deyo, 1998). However, in the postpartum period, many women desire to get their body back (p.845) as they feel that pregnancy and childbirth is a time when control cannot be taken over the physical changes in the body. Thus, it is the PNC providers who can assist women engaging in proper nutrition and exercise habits, which may potentially result in a more positive self -esteem and body image. Since 1) weight is a prominent concern with respect to a womens well -being, and 2) weight change i s a significant feature of pregnancy, it is suggested that changes in weight and body shape may play an important role in the development of PPD symptoms (Cameron et al., 1996). The attitudinal and behavioral changes with regards to weight that often take place during the course of pregnancy tend to be positive, but nonetheless, concerns about eating habits and weight continue to exist and may even extend into the postpartum period. In fact, a woman entering the postpartum period is vulnerable to experience concerns about her weight, because many women tend to retain more weight, and thus, weigh more than they did in the preconception period. Thus, much of the concerns may be more relentless that they were during the preconception period, and, many women do not necessarily attribute weight gain in a positive manner as they

PAGE 44

44 44 did during the pregnancy. However, womens responses toward weight gain during pregnancy may differ depending on acceptance of the role of motherhood that will be undertaken following child birth, and perceptions of physical changes of the body as indications of fetal growth (Lacey & Smith, 1987). These weight concerns may contribute to anxiety or depressive symptoms in the postpartum period (Carter, Baker, & Brownell, 2000), which may result in more weight loss attempts. Wood Baker et al. (1999) found that few women reported attempts to lose weight during pregnancy in comparing the preconception, pregnancy, and postpartum periods; the most weight loss attempts were made in the postpartum peri ods. For these women that engage in healthful behaviors during the postpartum period, it is important to be cognizant of the factors that motivate them to engage in exercise and weight management following childbirth (Keller, Allan, & Tinkle, 2005). Howeve r, Harris, Ellison, & Clement (1999) found that time for exercise and fitness was compromised among many of the women due to the demands of motherhood. Thus, it is suggested that women adopt a healthy diet/fitness regimen during pregnancy (with the guidanc e of their PNC provider) that can be easily transitioned into the postpartum period. It is vital that health promotion activities that begin during pregnancy and continue into the postpartum period target sources of chronic stress (e.g., BMI) and work towa rds supporting womens self -esteem, as they may cultivate positive mental health during both periods (Hall, Kotch, Browne, & Rayens, 1996). In addition, health promotion activities and interventions should seek to determine eating attitudes and behaviors o f all women as both can be influential on postpartum distress. This is given that pregnancy is a time when 1) weight and body shape changes are expected, and 2) many women feel that pregnancy is a license not to have weight concerns (Stein & Fairburn, 1996; Fairburn & Welch, 1990, p. 158), this can affect the eating attitudes and/or eating behaviors of women during pregnancy.

PAGE 45

45 45 Expected PNC Content Versus Actual PNC Content It is important to note that the content of PNC is yet to be standardized and the e xtent to which prenatal health behavior advice is given to women during their PNC is not consistent among all PNC rendered (Kogan, Kotelchuck, Alexander, & Johnson, 1994). Even though the IOM, ACOG, and the U.S. Public Health Service Expert Content on Prenatal Care have issued guidelines and recommendations on PNC content with respect to nutrition education and guidance, there remains differences in the PNC that is actually delivered. A number of studies have compared and contrasted the content PNC that is actually delivered versus the PNC content that is recommended (without considering a birth outcome in the study). To illustrate the differences that exist among the PNC rendered across the U.S., Appendix A (pp.137 155) summarizes the literature that has be en conducted on the content of PNC using nationally representative samples of women. According to Appendix A, there is a considerable amount of variability that exists with respect to the care that is actually delivered. However, the literature in Appendix A suggests that a significant portion of PNC providers in the U.S. are discussing nutritionally related content in the delivery of PNC (e.g., weight gain, exercise, proper nutrition, etc.). In addition to the literature that exemplifies that PNC providers are carrying out nutrition and wellness within the delivery of PNC, there is a collection of literature addressing the effectiveness of PNC for pregnancy outcomes as well as the potential impact on postpartum outcomes. Effectiveness of PNC The efficacy o f the content of PNC has been addressed in the literature, though not sufficiently (Alexander & Kotelchuck, 2001). In addressing the effectiveness of PNC, PNC is often publicized as a health care service that is necessary for improving pregnancy outcomes a mong women in the United States (Alexander & Kotelchuck, 2001). The benefits of receiving

PAGE 46

46 46 early and continuing PNC in the U.S. has been advertised as critical to promoting healthy pregnancy outcomes (Alexander & Kotelchuck, 2001). Some studies have shown t hat early initiation of PNC (in the first trimester) results in improved pregnancy outcomes as opposed to later or no PNC at all (Daniels et al., 2006). Accurately measuring PNC utilization is vital for determining the need for health services, monitoring trends in health care utilization, and determining associations between PNC and pregnancy outcomes (Kogan et al., 1998). PNC also seems to result in healthier pregnancies even for women who have no disease because PNC appears to associate well with the pre vention of adverse pregnancy outcomes (Rosen, 1989). For example, if conditions such as obesity are not addressed in the pre -conception period, as evidence indicates that preconception care can improve pregnancy outcomes (Atrash et al., 2008; Frieder, Dun lop, Culpepper, & Bernstein, 2008), it is recommended that PNC should address risk factors such as obesity as its association with pregnancy complications has been shown (e.g., preeclampsia, respiratory problems, cesarean section, fetal death, etc.) ( LaCou rsiere et al., 2005; Saravanakumar et al., 2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005; Mahmood, 2009; Baeten et al., 2001; Cnattingius et al., 1998). It may also be beneficial for postpartum health to be addressed during PNC deli very. Prenatal Care (PNC) and Postpartum Outcomes Alexander & Kotelchuck (2001) suggest that experiences in PNC that occur through education and support services may positively impact the postpartum health of the mother and the infant, including health status, health behaviors, and health care utilization. However, it is important to note that some women may not feel the need to see their PNC provider for a postpartum check up. This may be due to women believing that unless they experience adverse symptoms in the postpartum period that warrant a postpartum check up, there is no need for a postpartum evaluation of their health. This is unlike the prenatal period when women are

PAGE 47

47 47 concerned about the health of their fetus and doing what is necessary to reduce t he likelihood of pregnancy related complications. More research is needed on the relationship between PNC and postpartum behaviors, particularly postpartum depression (PPD) (Alexander & Kotelchuck, 2001). Though there is an absence of literature addressi ng the impact of PNC on PPD in the U.S., the association between PNC and PPD has been previously demonstrated outside the U.S., with more PNC visits inversely associated with the onset of PPD among highrisk women (El -Kak, Chaaya, Campbell, & Kaddour, 2004). Though the sample of women in this study were Lebanese, this study demonstrates the importance of PNC in preventing adverse postpartum outcomes (e.g., PPD), and warrants attention for the impact of PNC on PPD in the U.S. Since this study suggests an im pact of PNC on a postpartum outcome (PPD), and thus, the experiences that a woman undergoes through her PNC may affect the outcomes she experiences in the postpartum period (Alexander & Kotelchuck, 2001). If a woman seeks PNC and consequently complies with the advice/recommendations given to her by her PNC provider), this may help prevent adverse postpartum outcomes. In addition, postpartum care should be aligned with the PNC in addressing similar issues that can impact weight and health during the postpart um period. This would help women obtain access to professional assistance in addressing obesity issues postpartum. Theoretical Framework For this study, since PNC as an intervention through which PNC providers can educate help address any concerns with a woman about her weight, and provide guidance to patients, this delivery of health care will be seen as informational support. Homan & Korenbrot (1998) looked at support delivered at each of five PNC ambulatory practice settings (community clinic, health department, public hospital clinic, private hospital clinic, private physician office).

PAGE 48

48 48 Support service delivery was defined as psychosocial, health education, and nutrition (using the USPHS recommendations) and the authors concluded that a woman who ha s her needs addressed during PNC (psychological support, nutrition, or health information) has a higher likelihood of experiencing a positive birth outcome (poor obstetric outcomes was defined as experiencing preterm birth, having a low birth weight infant fetal death, ectopic pregnancy, or spontaneous abortion) than a woman who does not have any of her needs addressed during PNC. Obesity: Research has shown that an association exists between pre pregnancy weight and pregnancy weight gain, as well as postpartum weight retention. Pre -pregnancy BMI was chosen as the measure for BMI because it is known to be the strongest predictor of future obesity, including excess perinatal weight gain and future weight gain (Krummel, 2007; Gore, Brown, & Smith West, 2003), and is also suggested to be connected to postpartum BMI distress (Walker, 1998). Prenatal Care (PNC): The intervention of interest in this study will be PNC (health education and guidance on healthy behaviors during pregnancy). In general, PNC includes 1) risk assessment (e.g., medical and psychosocial history, physical examination, laboratory tests), 2) health promotion activities (e.g., counseling to promote healthy behaviors and providing general knowledge about pregnancy and parenting such as physiological and emotional changes, symptoms of preterm labor, fetal growth and development) and 3) a proposed pregnancy plan that is tailored to each womans needs (USPHSEPCPNC, 1989). The following nutritional intervention is recommended for the PNC provider du ring the delivery of care (Institute of Medicine, 1990): 1 ) Encourage the women to achieve a healthy, balanced diet to support adequate weight gain (e.g., IOM recommended ranges) 2 ) Evaluate a womans dietary habits (e.g., food history, food frequency questionna ire)

PAGE 49

49 49 3 ) Calculate the womans pre -pregnancy BMI 4 ) Estimate the womans gestational age 5 ) Conduct a routine dietary assessment to determine dietary needs 6 ) Be aware and knowledgeable about the attitudes and feelings that many pregnant women hold regarding body image and perceptions, especially closer to childbirth and in the postpartum period, as they tend to be negative during these times (Moore, 1978; Stein & Fairburn, 1996). It is the PNC providers that can assist women in viewing pregnancy and its associated phy sical changes as ones that are normal and beautiful (Moore, 1978). Postpartum depression symptoms: If a woman engages in healthy nutrition behaviors during pregnancy, her likelihood for PPD symptoms will be reduced.

PAGE 50

50 50 Figure 2 1. Theoretical framework Prenatal care (informational support): Education and guidance on weight gain, health behaviors Obesity (pre pregnancy BMI) PPD

PAGE 51

51 51 CHAPTER 3 METHODS Data Overview: Pregnancy Risk Assessment Monitoring System (PRAMS) This study used data from the 2004 and 2005 Pregnancy Risk Assessment Monitoring System (PRAMS) (CDC, 2007). This is a continuing population-b ased survey maintained by the Centers for Disease Control (CDC) that collects data on maternal behaviors, experiences, and characteristics in the pre -pregnancy, pregnancy, and postpartum period among randomly selected woman who delivered a live infant. Sta rted in 1987 because the incidence of low birth weight infants had changed very little during the previous 20 years, and because the infant mortality rate was not decreasing as fast as it had in previous years, this database provides state -specific data. Currently, 30 states participate in PRAMS. States can use these data to measure the performance of health programs and to obtain information on maternal experiences within their respective populations, which can contribute to improving maternal and child he alth. These data can also be used to 1) identify the women and infants who are at the highest risk for maternal and child health problems, 2) monitor their health status, and 3) assess their progress in efforts to improve the health of these mothers and their infants. Data Collection Procedures Each state that participates samples 1,300 to 3,400 women annually. Women are initially contacted through mail, and those who do not respond are contacted and interviewed via telephone. The questionnaire includes tw o components: the core questionnaire and the standard questionnaire. The core questionnaire contains 56 questions which all states include in their survey, and the standard questionnaire contains 185questions from which states can choose which to include in their survey (questions are options for survey inclusion). The standard

PAGE 52

52 52 questionnaire allows the collection of data to best meet the needs of each state. Data collection methods and instruments are standardized across states and occur as follows (CDC, 2007): 1 ) A preletter is sent to the mother to inform her that she will be receiving a PRAMS questionnaire 2 ) The initial mail questionnaire packet is sent to all the mothers randomly chosen for the sample 3 7 days after the preletter is sent 3 ) The tickler is the n ame given to the thank you note and a reminder to complete the questionnaire. This document is sent 7 10 days after the initial packet 4 ) The second mail questionnaire packet is sent to all the mothers who do not respond 7 14 days after the tickler is sent ou t 5 ) The third mail questionnaire packet is sent to all the mothers who do not respond 7 14 days after the second questionnaire is sent out 6 ) A telephone follow up occurs for all the mothers who do not respond 7 14 days after the third questionnaire is sent out (with up to 15 phone attempts in efforts to reach the mother). During this follow up, interviewers may coordinate times with the mother to administer the questionnaire over the phone. Thus, PRAMS involves mixed methodology: the self administered questionnaire that is mailed out, or a phone administered questionnaire in the event the mother does not complete and return the questionnaire prior to the phone follow up. The questionnaire packet includes the following: a cover letter that 1) explains PRAMS and why the mother was chosen to participate, 2) provides directions for completion of the survey, 3) describes potential incentives/rewards, and 4) provides a contact number to address further questions. In 2004, the cover letter was divided into two componen ts: an introductory portion, and a document of informed consent. These mailings are sent to the mothers approximately 2 4 months after they have given birth, and the data collection process lasts for approximately 60 95 days. The states included in the fin al data set that is released include those who achieved a 70% response rate or higher. Data collection is attempted every month, by randomly selecting a sample of approximately 100250 women from the current birth certificates, and then sending out the

PAGE 53

53 53 ma ilings to the mothers chosen. Since the birth certificates are linked to the mothers responses, variables on the birth certificate can also be accessed for analysis. Some groups of women, who comprise high risk populations, are sampled at a higher rate so that a sufficient quantity of data are available (CDC, 2007). The oversampling of subpopulations of interest is accomplished by the stratification of the PRAMS sample (Shulman, Gilbert, & Lansky, 2006). For the 2004 and 2005 years of PRAMS data used in th is study, the stratification variables included: birthweight, Medicaid, maternal age, geographic area, maternal race/ethnicity, county density, and smoking status (CDC, 2009). Weighting of Data Data are weighted to account for characteristics of the mothe rs that may influence the response rates (e.g., single marital status). These nonresponse weights assume that those mothers who did not respond would have given similar answers to questions as mothers who did respond to the questionnaire. For each cell, there are at least 25 respondents. For this study, the following weights were set prior to the analysis: a finite population correction factor (fpc), a strata weight, and a final analysis weight comprised of a non response weight, a non-coverage weight, an d a sampling weight. Thus, the final analysis weight in -part, corrects for groups (e.g., obese pre -pregnancy BMI) that have higher than normal response rates. Weights are applied so that each states data are representative of the women who gave birth in t hat state. Rationale for Using PRAMS Data The Pregnancy Risk Assessment Monitoring System ( PRAMS) is a population -based data set that incorporates random sampling techniques from 30 states, 16 of which are included in the main analysis. For this study, a large sample size was used that was representative of the different regions around the U.S (e.g. Southeast, Northeast, Midwest, etc.) and hence, the different populations that reside in these regions. In addition, PRAMS includes a comprehensive

PAGE 54

54 54 list of variables regarding maternal health and behaviors du ring the pre -conception, antenatal, and postpartum periods. There exist no other national databases that include the significant amount of variables that PRAMS includes. Measures/Procedures Postpartum Depression ( D ependent Variable) PRAMS data include 12 measures of PPD. However, participating states have the option of choosing whether to use measures of PPD, if any (PPD questions are part of the standard questionnaire). For the main analysis, two measures of PPD are used: 1 Since your new baby was born, how often have you felt down, depressed, or hopeless? Always Often Sometimes Rarely Never 2 Since your new baby was born, how often have you had little interest or little pleasure in doing things? Always Often Sometimes Rarely Never The states that incl ude these PPD questions include: Alaska, Colorado, Georgia, Hawaii, Illinois, Maine, Minnesota, North Carolina, Nebraska, New Mexico, Oregon, Rhode Island, South Carolina, Utah, Vermont, and Washington. Other questions that pertain to PPD in PRAMS include whether a woman received a professional diagnosis and/or whether she sought treatment for her PPD. However, these questions were excluded from the primary analysis because the extent to which a woman is either 1) diagnosed with PPD and/or 2) received treat ment for her PPD symptoms may depend on the proactive nature of the provider from whom she 1) sought

PAGE 55

55 55 PNC from and or 2) sought her postpartum check up visit from (as they may be two different providers and each may or may not be cognizant of or look out fo r PPD symptoms). The main analysis consisted of a logistic regression model, which included the 16 states from which data were requested. Control variables for the main analysis came from the PRAMS Core Questionnaire only, since all the states are mandated to include the core questions in their state surveys (see page 5 9 for a list of the control variables). In coding PPD for the logistic regression model as either 1 or 0, a scoring system similar to the Patient Health Questionnaire was used. The Pati ent Health Questionnaire (PHQ) is a threepage, patient self administered, criteria -based instrument that was created to diagnose mental disorders (e.g., major depressive disorder, panic disorder, other anxiety disorders, etc.) (Kroenke, Spitzer, & William s, 2001). Not only have the sensitivity and specificity been established (e.g., 68 95%, 84 95% respectively) (Kroenke et al., 2001), as well as criterion validity, construct validity, test retest reliability, and internal reliability, but the condensed 9 i tem module that is used to diagnose major depression and determine its severity is based on the criteria that are used in the DSM IV to diagnose depressive disorders. Stemming from the PHQ 9, the PHQ 2 was extracted to offer clinicians a more concise measu re of depression diagnosis and severity to accommodate the busy clinical settings that exist in todays health care system (Kroenke, Spitzer, & Williams, 2003). The following two items comprise the PHQ 2 (Kroenke et al., 2003, p.1285): Over the last 2 we eks, how often have you been bothered by any of the following problems? a ) Little interest or pleasure in doing things b ) Feeling down, depressed, or hopeless

PAGE 56

56 56 The response and scoring system are as follows: Not at all (score = 0), several days (score = 1), more than half the days (score = 2), nearly everyday (score = 3) (p. 1285). Since a score is given for each of the two question, and the two scores are added to obtain one total score, the highest possible score than can be given is a 6, and the lo west possible score that can be given is a 0. A score of 3 or greater indicates major depressive disorder. The sensitivity and specificity (83% and 92%, respectively) for scores greater than or equal to three, as well as the criterion and construct validit ies of the entire PHQ 2 have been demonstrated (Kroenke et al., 2003). Since the questions on the PHQ 2 mirror the PRAMS PPD measures that were used for this study, a similar scoring system was used to determine PPD symptoms. However, since the PPD measure consists of a 5 item response scale, and the PHQ 2 consists of a 4item response scale, the sometimes and rarely responses were scored as 1. Thus, the following point system was assigned to the PPD responses: 3 = Always 2 = Often 1 = Sometimes 1 = Rarely 0 = N ever For the purpose of this study, scores from 0 2 were coded as 0 (no PPD symptoms), and scores from 3 6 will be coded as 1 (PPD symptoms), similar to the coding of the PHQ 2. In addition to the logistic regression model, an ordinal logistic regress ion was also estimated in order to test the association of PNC on PPD when PPD is analyzed as a non -dichotomous outcome variable. For the ordinal logistic regression model, since there are two questions to represent the dependent variable (PPD), the PHQ 2 scoring system was also used in the analyses. *Note: the scores for PPD in the logistic regression represent the total score after adding the scores for both PPD questions.

PAGE 57

57 57 Obesity (Main Independent Variable) BMI is calculated using the following formula : Weight in pounds BMI = --------------------------X 703 (conversion factor) Square of height in inches (3 1) However, since maternal pre -pregnancy BMI is a variable included in the PRAM S data, there was no need to separately calculate and categorize a womans pre -pregnancy BMI. Table 3 1 shows the BMI categories, one of which a womans pre -pregnancy BMI is classified into. Prenatal Care Utilization (Moderating Variable) The Adequacy of Prenatal Care Utilization (APNCU) is an index measure of PNC that considers both the adequacy of PNC initiation, as well as the adequacy of the number of PNC services received (Kotelchuck, 1994). The benefits for using this index are as follows: 1) the APN CU offers a suitable index for assessing the degree of prenatal care utilization once care is initiated, 2) this index also separates the initiation of care from compliance with the number of visits as recommended by the American College of Gynecologists once care is initiated, 3) this index can include women who did not receive any PNC, 4) this index is valuable for research that is aimed towards enhancing PNC, 5) this index has a separate category for high -risk pregnancies (adequate plus care), and 6) thi s index, among others, provides the most serious depiction of prenatal care utilization (Alexander & Kotelchuck, 1996). The APNCU is calculated based on the time of PNC initiation and the number of PNC visits, adjusting for gestational age. Gestational age refers to the period from the first day of the woman's last menstrual cycle to the date of the babys birth (National Institutes of Health, 2007). Adjusting for gestational age is important, as it may be associated with the duration of PNC as well as prob lems experienced during pregnancy/postpartum. The index is defined as the ratio of observed (actual) prenatal care visits (determined from the birth certificate)/expected number of PNC visits (based on

PAGE 58

58 58 recommendations from the American College of Obstetric ians and Gynecologists). The percentage obtained is then categorized into one four PNC categories (see Table 3 5) (Kotelchuck, 1994). The APNCU Index was calculated in this study by the following methods: 1 ) Gestational age (included in PRAMS as number of weeks) was recoded based on the number of expected visits that should occur based on a womans gestational age. The rule of thumb is that there should be one PNC visit each month until 28 weeks (or seven months), which then changes to a PNC visit once every 2 wee ks until 36 weeks (or nine months) with a PNC visit weekly until a woman gives birth (Kotelchuck, 1994). The month that a woman gave birth was not counted as a PNC visit because it was unclear whether the woman received her PNC visit for that month or not before she gave birth. Table 3 3 lists the number of expected visits dependent on the gestation age. 2 ) Next, month of PNC initiation was recoded into number of missed visits. Table 3 4 lists the number of missed visits depending on when a woman initiated he r first PNC visit. Missed visits were calculated based on the maximum number of PNC visits a woman could miss given the month she initiated her PNC. 3 ) Since steps 1 and 2 comprise the denominator, a variable was next calculated subtracting the number of mis sed visits from the number of expected visits This variable was labeled indexdenominator 4 ) The index was then calculated by dividing the indexdenominator variable into the actual number of PNC visits variable (measured in PRAMS). 5 ) Table 3 5 shows how each index was cate gorized into one of the APNCU categories. Table 3 6 presents the characteristics of the APNCU Index categories. 6 ) Since some women many have received the quantity of expected visits even though they initiated PNC late, all women who initiated care after the fourt h month who were not initially coded into inadequate care were recoded into inadequate care. The month of PNC initiation takes precedence over the quantity of visits when determining the APNCU (Kotelchuck, 1994). Thus, women who initiated PNC after the fo urth month were filtered out of the sample, and recoded to inadequate care. Finally, the filter was removed to allow those women to remain as part of the sample; however, now with the correct APNCU. Table 3 7 contains the frequencies of women before recodi ng occurred, while Table 3 8 contains the frequencies of women who were incorrectly coded, and Table 3 9 contains the frequencies after recoding was accomplished. After creating the APNCU, dummy variables were first created for each of the main effects (e ach pre -pregnancy BMI and each PNC utilization category). Then, dummy variables, labeled

PAGE 59

59 59 as interaction terms, were created for each combination of pre -pregnancy BMI and PNC (9 groups) to detect any differences among any of the combinations. Normal pre -pre gnancy BMI and adequate PNC were the reference groups. 1 ) Obese/Inadequate care 2 ) Overweight/Inadequate care 3 ) Underweight/Inadequate care 4 ) Obese/Intermediate care 5 ) Overweight/Intermediate care 6 ) Underweight/Intermediate care 7 ) Obese/Adequate-plus care 8 ) Overweight/Ade quate -plus care 9 ) Underweight/Adequate -plus care Control Variables Control variables included maternal age, maternal education, maternal income (12 months before), maternal race, maternal ethnicity, birthweight, gender of infant, vaginal delivery, alcohol and smoking behaviors during pregnancy, participation in Women, Infants, and Children, breastfeeding practice, if the infant ended up in the ICU, pregnancy intention, how PNC was paid for (representing insurance), and maternal morbidities. All control vari ables were categorical except for maternal age. Analysis All analyses were conducted using Stata v.10. Both univariate and bivariate analyses were conducted, with chi -square tests used to test significance between groups. Significant association of PPD an d maternal age was tested using a t test, while ANOVA was used to test the association of pre -pregnancy BMI and maternal age, and PNC utilization and maternal age. The primary multivariate analysis held PPD as a dichotomous variable via a logistic regress ion (logit) model, while a secondary multivariate analyses held PPD as a dichotomous variable via a logistic regression (logit) model, and manually risk adjusted for highrisk pregnancies after

PAGE 60

60 60 removing observations that met any of the criteria for a high -risk pregnancy. Thus, the sample analyzed in the secondary logistic regression model included healthy pregnancies only. Primary Risk Adjusted Logistic Regression Specific aim 1 : What is the association of pre -pregnancy BMI with subsequent development of postpartum depression (PPD) symptoms? Hypothesis: Women, who were obese before pregnancy will have the highest likelihood of PPD, followed by overweight women, then underweight women. Women who had a pre pregnancy BMI of normal were the reference group. A dummy variable was created for each BMI category and the reference group for this analysis was normal pre -pregnancy BMI. Women who had a pre -pregnancy BMI of obese are predicted to have the highest likelihood for PPD. The s model, and in accordance with the hypothesis, were obese pre pregnancy BMI and underweight pre -pregnancy BMI which were expected to have the most positive (meaning the likelihood will be the highest for women in this group) and the least positive (the PP D likelihood will be the lowest for women in this group) association with PPD, respectively, relative to normal pre -pregnancy BMI (the reference group). The odds ratios were hat the women who had a pre -pregnancy BMI of obese, relative to women who had a pre -pregnancy BMI of normal, would have the highest likelihood for PPD symptoms and that the pre -pregnancy BMI of overweight and underweight groups would have the second to hig hest and lowest likelihoods for PPD symptoms, respectively. Since this is a logistic regression, PPD was analyzed as a dichotomous variable, using the scoring system similar to the Patient Health Questionnaire 2 [for a complete description of the scoring s ystem, see pp.6061: Postpartum depression (dependent variable) ]. The model specification was as follows:

PAGE 61

61 61 logit = log ) 0 ( ) 1 ( Y P Y P = + 1(underweight) + 2(overweight) + 3(obese)+ kXk (control variables) + (3 2) Specific aim 2: Does PNC moderate the relationship between pre -pregnancy BMI and PPD? Hypothesis: This aim tested whether the association of pre -pregnancy BMI with PPD symptoms varied with the level of PNC. The association between pre -pregnancy BMI and PPD symptoms w as expected to remain after estimating PNC as a moderating variable. This means that women who had a pre pregnancy BMI of obese would continue to have the highest likelihood for PPD symptoms, and women who had a pre pregnancy BMI of underweight would have the lowest likelihood, relative to women who had a pre -pregnancy BMI of normal. However, for each pre -pregnancy BMI group, it was predicted that relative to women who received adequate PNC, the highest likelihood for PPD symptoms would decrease as follows: inadequate PNC, adequate plus PNC, intermediate PNC. Each prepregnancy BMI category had its own reference group: women in the BMI group who received adequate care. For example, obese inadequate, obese intermediate, and obese adequate -plus were compared to obese adequate women. The model specification for the main logistic regression was as follows: logit = log ) 0 ( ) 1 ( Y P Y P = + 1(obese/adequate plus care) + 2(overweight/adequate plus care) + 3(underweight adequate plus care) + 4(obese/inte rmediate care) + 5(overweight/intermediate care) + 6(underweight/intermediate care) + 7(obese/inadequate care) + 8(overweight/inadequate care) + 9(underweight/inadequate care) + kXk(control variables) + (3 3) Secondary Risk Adjusted Logistic Regress ion Adequate plus, sometimes referred to as intensive care, is defined as PNC that is initiated by the fourth month (inclusive), and/or a woman received 110% or more of the expected

PAGE 62

62 62 number of PNC visits (Kotelchuck, 1994). A number of women in the adequ ate plus PNC category are considered to be a high risk pregnancy group of women who receive more PNC than the standards as established by the American College of Obstetricians and Gynecologists (Kotelchuck, 1994). Compared to the other categories of PNC in the APNCU: inadequate, intermediate, and adequate PNC, a woman may receive adequate plus PNC if she possesses health risks such as medical conditions that can put her at a greater risk for complications and/or adverse pregnancy outcomes; hence, terming th e pregnancy as high risk (Chism, 1997; Kotelchuck, 1994). Thus, the increased number of PNC visits in this category is meant to allow for additional monitoring of a womans pregnancy by her PNC provider(s). However, some adequate plus PNC is delivered to women who seek extra PNC due to other unobserved characteristics unrelated to risk (e.g., motivation to seek more PNC than necessary because of a womans desires to do so). Since the APNCU index does not risk adjust, but rather, remains a conventional index that looks at the quantity and initiation of PNC (Kotelchuck, 1994), this study sought to risk adjust for women who received adequate plus PNC in two ways: 1) risk adjusting by controlling for high risk characteristics, and 2) risk adjusting by removing observations with high risk characteristics (e.g., women with preterm labor) from the sample. Risk adjustment for the second model was accomplished by removing observations that met one or more of the following criteria during pregnancy: 1 ) Birthweight less than 2,500 grams 2 ) Women less than 18 years of age or greater than 40 years of age 3 ) Diabetes before pregnancy 4 ) Incompetent cervix 5 ) Preterm labor 6 ) Placenta previa or placenta abruptio 7 ) Bedrest 8 ) Car crash injury 9 ) Blood transfusion

PAGE 63

63 63 10) Medical risk factors 11) Hospitalized d uring pregnancy Thus, after removing these observations from the sample, it was assumed that the observations in the sample constituted pregnancies with standard, common characteristics (e.g., nausea, infant birthweight of 2,500 g or greater, etc.). A sub analysis was estimated for the first risk adjustment model, holding adequate plus PNC as the dependent variable and subsequently controlling for the significant variables in the main model. A further description of the subanalysis for the first model is provided in Appendix B (pp. 156 187). A sensitivity analysis was estimated for the second risk adjustment model. The specific aims, hypotheses, and specification for this model remained the same as the primary risk adjusted logistic regression model. Wa ld Test Finally, to test the equality of the interaction term coefficients within each BMI category, which would test the moderating effect of PNC on the BMI and PPD association, Wald tests were performed for each model that addressed the second specific a im. Two types of Wald tests were run: 1) to determine if the coefficients were equal to one another, and 2) to determine if the coefficients were equal to 0. Model Fit To test the model fit of the primary and secondary logistic regression models, the Del measure was calculated. This goodness of fit measure takes the scope and precision of the model into account. Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Merging Strata The PSU in this study was the woman herself. Since STATA woul d not analyze strata with only one PSU, frequencies were run on the strata weight variable and four strata were identified

PAGE 64

64 64 with less than ten PSUs (1 strata with 1 PSU, 2 stratas with 2 PSUs, and 1 strata with 9 PSUs). These strata were then merged into th e strata with the largest number of PSUs, after which the analysis was accomplished. Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Dropped Cases Since the loss of an infant can be traumatic for a mother, women who reported having an i nfant that did not live were removed from the analysis ( N= 1,032). Pregnancy Risk Assessment Monitoring System (PRAMS) Data Changes: Imputed Data Since the income variable included in the analysis had 3,346 missing values (more missing values than most of the other variables included in the analysis), the data were imputed via the conditional means approach using the following steps: 1 ) An OLS regression was estimated on demographic variables included in the multivariate analyses. 2 ) Predicated values were cal culated after running the OLS regression 3 ) The missing values were replaced with the predicted values. However, prior to imputation, because some states include different types of questions with respect to the same income category, data were collapsed to reflect four income categories. Table 3 10 shows the raw income variables categories and Table 3 11 shows the collapsed income variable before it was imputed. In addition, the variables representing how PNC was paid for, and the maternal morbidities were i mputed using a conditional means approach. This was accomplished due to the large number of missing values for some of the variables in these categories.

PAGE 65

65 65 Table 3 1. Specific aims, dependent, and independent variables Specific aim Dependent variable Indep endent variable(s) 1 Postpartum depression Pre pregnancy BMI 2 Postpartum depression Pre pregnancy body mass index (BMI), prenatal care utilization (PNC), pre pregnancy BMI/PNC interaction terms (moderator) Table 3 2. Classification of body mass index (BMI) BMI Weight status > 18.5 Underweight 18.5 24.9 Normal 25.0 29.9 Overweight >/= 30.0 Obese (National Institutes of Health, 1998) Table 3 3. Step 1: Gestational age calculation into expected number of visits for the APNCU Index Gestational a ge (in weeks) Expected number of visits 18 20 4 21 24 5 25 28 6 29 30 7 31 32 8 33 34 9 35 36 10 37 11 38 12 39 13 40 14 41 15 42 16 43 17 Table 3 4. Step 2: Month of PNC initiation calculation into number of missed visits for the APNCU Index Month of PNC initiation Number of missed visits 1 st month 0 2 nd month 1 3 rd month 2 4 th month 3 5 th month 4 6 th month 5 7 th month 6 8 th month 8 9 th month 10

PAGE 66

66 66 Table 3 5. Step 3: Categorization of index into categories for the APNCU Index APNCU group Included indices Inadequate Lowest thru 49.99 Intermediate 50.00 thru 79.99 Adequate 80.00 thru 109.99 Adequate plus 110.00 thru highest Table 3 6. Characteristics of the APNCU Index groups APNCU group Characteristics Inadequate care Initiated after the 4th month; under 50% of expected visits were received; can be divided to include those who did not receive PNC Intermediate care Initiated by the fourth month; between 50 79% of expected visits were received. Adequate care Initiated b y the fourth month; 80 109% of expected visits were received Adequate plus care Initiated by the fourth month; 110% or more of expected visits were received (Kotelchuck, 1994) Table 3 7. Adequacy of Prenatal Care Utilization (APNCU) Index frequencies be fore recoding APNCU index Frequency Percentage Inadequate 1,743 3.6% Intermediate 7,631 15.6% Adequate 20,678 42.2% Adequate plus 18,952 38.7% Total 49,004 100% Table 3 8. Adequacy of Prenatal Care Utilization (APNCU) Index frequencies of incorrec t codings (observations incorrectly coded into other PNC utilization categories that were recoded into inadequate PNC utilization based on the month of initiation) APNCU index Frequency Percentage Intermediate 1,680 42.4% Adequate 1,237 31.2% Adequat e plus 1,047 26.4% Total # of observations to be recoded into inadequate PNC 3,964 100% Table 3 9. Step 4: Adequacy of Prenatal Care Utilization (APNCU) Index frequencies after recoding all cases from Table 3 8 into inadequate PNC utilization APNC U I ndex Frequency Percentage Inadequate 5,707 11.6% Intermediate 6,584 13.4% Adequate 19,441 39.7% Adequate plus 17,272 35.2% Total 49,004 100%

PAGE 67

67 67 Table 3 10. Coding for raw income variable categories before collapsing categories Code Category 1 Les s than 10,000 2 $10,000 to $14,999 3 $15,000 to $19,999 4 $20,000 to $24,999 5 $25,000 to $34,999 6 $35,000 to $49,999 7 $50,000 or more 8 Less than $8,000 9 $8,000 to $9,999 10 $50,000 to $74,999 11 $75,000 or more Table 3 11. Coding for colla psed income variable categories Code Category 1 Less than 10,000 2 $10,000 to $24,999 3 $25,000 to $49,999 4 $50,000 or greater Changes made ** 2 Expanded to include income up to $24,999 ** 3 and 4 Collapsed into category 2 ** 5 Category number ch anged to 3 and expanded to include income up to $49,999 ** 6 Collapsed into category 3 ** 7 Category number changed to 4 **8 Collapsed into category 1 **9 Collapsed into category 1 **10 Collapsed into category 4 **11 Collapsed into category 4

PAGE 68

68 68 CHA PTER 4 RESULTS In this chapter, 1) the PRAMS sample is described through univariate analyses, and 2) bivariate analyses indicating percentages and chi -square significance for a) the dependent variable (PPD), b) each of the primary independent variables of interest, pre -pregnancy BMI and PNC utilization, and c) adequate plus PNC. Then, the results from the multivariate analyses are discussed via the primary risk adjusted logistic regression model and the secondary risk adjusted logistic regression, which rem oved highrisk pregnancies from the sample. Univariate Analyses The description of the sample is presented in Table 4 1 for the categorical variables, and Table 4 2 for the one continuous variable in this study, maternal age. Table 4 1 describes 1) the dis tribution of the dependent variable (PPD symptoms), 2) the distribution of the main independent variables (pre -pregnancy BMI and PNC utilization), and 3) the distribution of the 39 categorical control variables included in the analyses. Control variables w ere organized in the table by main variables, demographic control variables, insurance control variables, pregnancy and delivery control variables, high -risk maternal morbidity control variables, and non high risk maternal morbidity control vari ables. The latter two labels were added since the secondary analysis sought to distinguish high -risk from healthy pregnancies. Table 4 2 presents the mean, standard deviation, maximum and minimum values for maternal age. Basic demographic frequencies for the entire sample show that the largest percentage of women in the sample were of White race (63.64%), about half of the women received a secondary education or less (49.45%), and 63.29% of the women were married. Finally, the income distribution consi sted of about 20% each for women who received less than $10,000 and women who

PAGE 69

69 69 received between $25,000 to $49,999 (22.80%), and about 30% each for women who received between $10,000 to $24,999 (28.18%) and women who received $50,000 or greater (28.69%). P PD symptom frequencies for the sample analyzed in the primary risk adjusted logistic regression (including all pregnancies), and the sample analyzed in the secondary risk adjusted logistic regression (including healthy pregnancies only) were provided in Ta ble 4 1 to compare the distribution of PPD symptoms in a sample including women comprising healthy and highrisk pregnancies, versus a sample that removed highrisk pregnancies and included healthy pregnancies only. For PPD symptoms, the frequencies showed that among 45,285 women included in the primary risk adjusted logistic regression model, the prevalence of postpartum blues was 84.6%, while the prevalence of PPD symptoms in the sample was approximately 15.4%. Removing the highrisk pregnancies from the analysis, to include only healthy pregnancies with common pregnancy characteristics (e.g., nausea) showed that among the 15,443 women included in this sample, the prevalence of postpartum blues was higher, at 88.5% and the prevalence of PPD symptoms was lower, at 11.5%. Frequencies for the primary main effect independent variable, pre -pregnancy BMI, showed a high number of women who had an obese pre -pregnancy BMI (n=10,270; 22.15%). This group of women comprised the second highest group in the sample following the reference group, women with a normal pre -pregnancy BMI, who comprised about half of the sample (n=23,834; 51.40%). The number and percent of women who had an obese pre -pregnancy BMI was nearly twice as great as those of the group of women w ho were the lowest in number: women who had an overweight pre pregnancy BMI (n=5,888; 12.70%). Finally, the number of women who had an underweight pre -pregnancy BMI was not considerably higher than the number of women who had an overweight pre -pregnancy BM I (n=6,379; 13.76%). For the secondary main effect

PAGE 70

70 70 (moderating) independent variable, PNC utilization, the highest percentage was also comprised of the reference group, women who utilized an adequate quantity of PNC (n=19,258; 40.03%). However, the secon d highest number was comprised of women who utilized an adequate plus quantity of PNC (n=16,748; 34.82%). This group of women nearly tripled the number of women who utilized an inadequate quantity of PNC (n= 5,568; 11.58%) Finally, the number of women who utilized an intermediate quantity of PNC was not considerably different from the number of women who utilized an inadequate quantity of PNC (n= 6,529; 13.57%). Due to the high number of women who utilized adequate plus PNC, it was speculat ed that reasons for this high number could be attributed to either 1) high risk status, or 2) a womans own desire to seek more PNC than necessary (as established in medical guidelines). In attempts to distinguish pregnancies that were high risk versus pre gnancies that experienced normal, standard pregnancy related troubles (e.g., nausea), 17 maternal morbidity control variables were included in this study. The morbidity variables representative of highrisk and removed, in-part, from the secondary sample ( to include healthy pregnancies only) included nine variables: 1) Diabetes before pregnancy, 2) incompetent cervix, 3) preterm labor, 4) placenta previa or abruptio, 5) bedrest, 6) car crash injury, 7) blood transfusion, 8) having general medical risk facto rs, and 9) hospitalization during pregnancy. The distribution of the highrisk morbidity variables was as follows: 1) four of the variables were comprised of less than 5% of the sample (less than 5% of the women answered yes to having had the morbidity during pregnancy), with the lowest percentage representing women who had a blood transfusion during pregnancy (n=677; 1.31%), 2) one variable was comprised of a percentage between 5 10%, and 3) the remaining four variables were comprised of percentages betw een 10 40%, with the highest percentage representing women who had general medical risk factors during pregnancy

PAGE 71

71 71 (n=18,209; 35.29%). Since the largest percentage of the highrisk morbidity variables represented a general measure (women who answered yes t o having had medical risk factors during pregnancy), the highest percentage representing a single, defined morbidity was comprised of about 25% of the sample: women who had preterm labor (n=13,595; 26.35%). Among the non highrisk morbidity variables, the distribution occurred as follows: 1) one variable was comprised of less than 10% of the sample, with the lowest percentage representing women who had gestational diabetes (n=4,786; 9.28%) 2) four variables were comprised of percentages between 10 20%, and 3) the remaining two variables were comprised of percentages between 20 40%, with labor/delivery complications comprising the highest percentage (n=18,131; 35.14%). Similar to the high-risk morbidity variables, since the largest percentage for the non high risk morbidity variables represented a general measure (women who answered yes to having had labor/delivery complications), the highest percentage representing a single, defined morbidity involved a very common morbidity in pregnant women: nausea, compr ising about 30% of the sample (n=15,104; 29.27%). To further examine characteristics of women who utilized adequate plus PNC versus women who utilized other quantities of PNC, beyond frequencies, chi -square analyses, presented in the next section, were carried out. Bivariate Analyses Chi -square analyses, presented in Tables 4 3, 45, and 4 6 respectively, were performed to examine relationships between 1) the dependent variable (PPD) and each of the following: a) pre pregnancy BMI, b) PNC utilization, and c) each of the control variables, 2) the primary main effect independent variable (pre pregnancy BMI) and each of the control variables, and 3) the secondary main effect independent variable (the moderating variable, PNC utilization) and each of the cont rol variables. Also, since there were many women who utilized adequate plus PNC, the highest quantity of PNC, chi -square analyses, presented in Appendix B (pp. 156187), were

PAGE 72

72 72 performed with adequate plus PNC to further describe characteristics of women who utilized this quantity of PNC. All chi -square analyses were carried out in order to test the specific aims, and to determine the significance of the relationships between the variables of interest (PPD symptoms, pre -pregnancy BMI, and PNC utilization) and the control variables. Table 4 4 presents the results from the t -test for the continuous variable, maternal age, with PPD symptoms. Since PPD symptoms was the dependent variable in this study, chi -square analyses were estimated to examine the relationsh ip of characteristics among women who were categorized as experiencing PPD symptoms versus those who were categorized as not experiencing PPD symptoms. Table 4 3 shows that there was a significant difference in percentages of PPD symptoms across the four p re -pregnancy BMI categories (p<0.05), with normal pre -pregnancy BMI displaying the highest percentage of PPD symptoms, followed by women who had an obese pre pregnancy BMI. Overweight pre pregnancy BMI displayed the lowest percentage of PPD symptoms, while the percentage of women who had an underweight pre -pregnancy BMI was not considerably higher than the percentage of women who had an overweight pre -pregnancy BMI. In fact, the percentage of women who had an obese pre pregnancy BMI with PPD symptoms was tw ice as much as the percentage of women who had an overweight pre -pregnancy BMI. As for PNC utilization, there was a significant difference in percentages of PPD symptoms among the four PNC utilization categories. Women who utilized adequate plus PNC displa yed the highest percentage for PPD symptoms, followed by women who utilized adequate PNC. Women who utilized intermediate PNC displayed the lowest percentage for PPD symptoms, while the percentage of women who utilized inadequate PNC was not considerably h igher. Overall, chi square results in comparing chi -square statistics for 41 variables showed a significant difference

PAGE 73

73 73 in percentages for PPD symptoms across all the variables (p<0.05), except for three variables that did not show any significant differenc e (PNC paid by: Native American Health Services, gender of the infant, and labor abnormalities), and one variable that was significantly different at a p<0.10 (labor/delivery complications). Since pre -pregnancy BMI was the primary independent variable in t his study, chi -square analyses were estimated to examine the relationship of characteristics among women from all four pre -pregnancy BMI categories. Table 4 5 shows that there was a significant difference for prenatal care (PNC) utilization frequencies amo ng all pre -pregnancy BMI groups (p<0.05). The percentages of PNC utilization for each pre -pregnancy BMI group were not considerably different from the percentages presented in the univariate analysis for PNC utilization. It was only for obese pre pregnancy BMI that the percentage of women who utilized adequate plus PNC slightly increased (about 4% higher) than the average percentage for the other pre pregnancy BMI groups. Further addressing bivariate results for pre -pregnancy BMI, since women who had an ob ese pre -pregnancy BMI were the focus of this study, it is especially worthy to note the bivariate characteristics of women who had an obese pre -pregnancy BMI. For example, in looking at the frequencies of the 17 maternal morbidities presented in Table 4 5 with pre pregnancy BMI, given the 14 morbidities in which there was a significant difference, except for two variables: preterm labor and placental problems (placenta previa or abruptio), women who had an obese pre -pregnancy BMI always comprised the highes t percentage of women who experienced those morbidities when comparing the percentages for all four pre pregnancy BMI groups. The highest percentage for preterm labor and placental problems was for women who had an underweight pre pregnancy BMI. To further note relevant bivariate statistics, women who

PAGE 74

74 74 had an underweight pre pregnancy BMI also had the highest percentage (26.6%) of low birth weight babies (between 1,500 to 2,499 grams), while women who had an obese pre -pregnancy BMI had the highest percentage (7.95%) of very low birth weight babies (less than 1,500 grams), compared to women from the three other pre -pregnancy BMI groups. Overall, chi -square results in comparing chi -square statistics among women from all prepregnancy BMI groups for 40 variabl es showed a significant difference among all the characteristics (p<0.05), except for six variables that did not show any significant difference (PNC paid by the military, gender of the infant, weight gain talk during pregnancy, car crash injury, blood tra nsfusion, and labor/delivery complications). Since PNC utilization was the secondary independent variable (the moderating variable) in this study, chi -square analyses were estimated to examine the relationship of characteristics among women from all four P NC utilization categories (Table 4 6). Overall, chi -square results among all women in PNC utilization groups for 39 variables showed a significant difference among all the characteristics (p<0.05), except for one variable that did not show any significant difference (gender of the infant). Due to a large number of women who utilized adequate plus PNC, chi -square statistics, presented in Appendix B (pp.156 187), were estimated to examine the relationship of characteristics among women who were categorized as utilizing adequate plus PNC versus those who were categorized as utilizing other quantities of PNC. Comparing pre -pregnancy body mass indices (BMIs) among women who did not receive adequate plus PNC versus women who utilized adequate plus PNC, there wa s a larger percentage of women in each pre -pregnancy BMI group for women who did not receive adequate plus PNC, except for women who had an obese pre pregnancy BMI; hence, 24.6% of women who utilized adequate plus PNC had an obese pre -

PAGE 75

75 75 pregnancy BMI, while 20.6% of women who utilized other quantities of PNC had an obese pre pregnancy BMI. Finally, in comparing t he morbidity variables among women who utilized adequate plus PNC, overall results showed that the women who received adequate plus PNC were medic ally and obstetrically high risk. These results provided a basis for controlling for highrisk characteristics in the multivariate analyses. A number of multivariate models were estimated in addition to 1) the univariate statistics that gave a general over view of the sample characteristics, many of which were noteworthy, and 2) the bivariate statistics that gave an indication of frequencies and significance between each variable of interest: PPD symptoms, pre -pregnancy BMI, PNC utilization, and adequate plu s PNC, and the other variables included in this study (e.g., control variables). Since the chi -square analyses demonstrated significant differences between women with PPD symptoms versus women without PPD symptoms, among all four pre -pregnancy BMI groups a nd all four PNC utilization groups (the two independent variables of interest), a variety of multivariate analyses were conducted to determine if these relationships would remain. The multivariate analyses, which included a variety of logistic regression m odels, were estimated to 1) determine if significance with the variables of interest (e.g., PPD symptoms, pre pregnancy BMI) would be demonstrated, thus supporting the specific aims and hypotheses, and 2) find the best model fit for the specific aims and h ypotheses proposed for this study. Multivariate Analyses A variety of logistic regression (logit) models were estimated in attempts to demonstrate 1) an association between pre -pregnancy BMI and PPD symptoms, and 2) a moderating effect of PNC in the association between pre -pregnancy BMI and PPD symptoms. Logistic regression models for both specific aims were estimated since this study sought to detect a moderating

PAGE 76

76 76 effect of PNC utilization as opposed to a mediating effect. Six of these models are present ed in the main analyses, and the other ten models are presented in B (pp.156 187). All odds ratios were calculated using a 95% confidence interval. Primary Risk Adjusted Logistic Regression Analysis Baseline model Table 4 7 presents the results from the ba seline logit model that was estimated for pre pregnancy BMI and PNC utilization only, without the control variables and interaction effects. This model showed significance for obese pre pregnancy BMI, suggesting an association between women from this pre -p regnancy BMI category only, and PPD symptoms. Compared to women who had a pre -pregnancy BMI of normal, women who had a pre -pregnancy BMI of obese had 15% greater odds for PPD symptoms (OR=1.15, p=0.02). All PNC utilization categories were significant in t he baseline model, suggesting a general association between PNC utilization and PPD symptoms. Compared to women who utilized adequate PNC, women who utilized inadequate PNC had 84% greater odds (OR=1.84, p<0.0001) for PPD symptoms, while women who utilized intermediate PNC had approximately one -fifth greater odds (OR=1.19, p=0.02), and finally, women who utilized the highest quantity of PNC, adequate plus, had 28% greater odds for PPD symptoms (OR=1.28, p<0.0001). Specific aim 1 The model presented in Tabl e 4 8 sought to detect an association between the main effects of pre -pregnancy BMI and PPD symptoms, while taking into account the control variables identified in the chi-square analyses. However, unlike the baseline model, only borderline significance wa s found for underweight pre pregnancy BMI. In contrast to the hypothesis, this group had lower odds for PPD symptoms compared to women who had a normal pre pregnancy BMI (OR=0.87, p=0.08). Thus, specific aim 1 was not supported. In addition, the

PAGE 77

77 77 associatio n between PNC utilization and PPD symptoms was not found in this controlled analysis. Specific aim 2 The model presented in Table 4 9 sought to detect a moderating effect of PNC in the association between pre -pregnancy BMI and PPD by including interaction terms between each combination of pre pregnancy BMI and PNC utilization category. This all -inclusive model included the main effects, interaction effects, and control variables. However, despite the significance seen for underweight pre pregnancy BMI in the previous model, this effect disappeared as significance was not seen for any of these main effects. No moderating effect of PNC was apparent for any of the pre pregnancy BMI groups either. Figure 4 2 presents a graph of the odds ratios for the interact ion effects within each pre -pregnancy BMI group. Secondary Risk Adjusted Logistic Regression: Subpopulation With Healthy Pregnancies Another approach was taken to test the specific aims. The logistic regression models carried out under this approach consisted of women who had normal, healthy pregnancy characteristics. For a list of the characteristics that constituted high -risk characteristics and were removed from these analyses, please refer to pp.68 70. The risk adjustment process was accomplished by removing high risk pregnancies from the analysis. Baseline model This model, presented in Table 4 11, showed significance in the baseline model among women who utilized inadequate PNC. Compared to women who utilized adequate PNC, they had 97% greater odd s (OR=1.97, p<0.0001) for PPD symptoms (Table 415). Compared to the baseline model for the entire population (Table 4 7), the significant relationship of obese pre pregnancy BMI with PPD symptoms, and the significant relationships of intermediate and adeq uate plus PNC utilization with PPD symptoms disappeared.

PAGE 78

78 78 Specific aim 1 The presented in Table 4 12, which included the main effects and the control variables, did not show any significance among the main effects. Thus, after adding the control variables to the baseline model for the logistic regression that removed highrisk pregnancies, the significance of inadequate PNC also disappeared. Specific aim 2 The all -inclusive model, presented in Table 4 13, showed borderline significance among women who had a pre -pregnancy BMI of obese and utilized inadequate PNC. However, contrary to what was expected, women who had a pre -pregnancy BMI of obese and utilized inadequate PNC had roughly half the odds (OR=0.51, p<0.08) for PPD symptoms compared to women who had a pre -pregnancy BMI of obese and utilized adequate PNC. Wald Test Tables 4 10 and 4 14, present the results from the main risk adjusted logistic regression and the risk adjusted logistic regression that removed the high risk pregnancies (respectively). Th e results of the Wald tests indicated that the PNC categories within each pre-pregnancy BMI group were not different from each other; hence, no moderating effect of PNC was seen after performing these tests. Model Fit The Del measures calculated for the pr imary and secondary logistic regression models were 0.03 and 0.02, respectively. The Del values of both models indicated a poor model fit in the ability to predict PPD symptoms.

PAGE 79

79 79 Table 4 1. Univariate statistics for all categorical variables included in the bivariate and multivariate analyses Categorical variables N Frequency (%) Main variables Primary analysis dependent variable: Postpartum depressive symptoms (risk adjusted logistic regression including all pregnancies) No Yes 45,285 38,320 (84. 62%) 6,965 (15.38%) Secondary analysis dependent variable: Postpartum depressive symptoms (risk adjusted logistic regression including healthy pregnancies only) No Yes 15,443 13,662 (88.5%) 1,781 (11.5%) Main independent variable: Pre pregnanc y body mass index (BMI) Underweight Normal Overweight Obese 46,371 6,379 (13.76%) 23,834 (51.40%) 5,888 (12.70%) 10,270 (22.15%) Main independent variable: Adequacy of Prenatal Care Utilization (APNCU) Index Inadequate Intermediate Adequ ate Adequate p lus 48,103 5,568 (11.58%) 6,529 (13.57%) 19,258 (40.03%) 16,748 (34.82%) Demographic control variables Maternal r ace : White No Yes 49,119 17,858 (36.36%) 31,261 (63.64%) Maternal r ace : Black No Yes 49,11 9 41,400 (84.29%) 7,719 (15.71%) Maternal r ace : Other No Yes 49,119 38,980 (79.36%) 10,139 (20.64%) Hispanic Not Hispanic Hispanic 48,972 39,792 (81.55%) 9,000 (18.45%) Maternal education 0 8 years 9 11 years 12 years 1315 years 16+ yea rs 50,765 2,266 (4.46%) 7,294 (14.37%) 15,542 (30.62%) 11,871 (23.38%) 13,792 (27.17%)

PAGE 80

80 80 Table 4 1. Continued Categorical variables N Frequency (%) Income (12 months prior) Less than $10,000 $10,000 to $24,999 $25,000 to $49,999 $50,000 or more 51,600 10,489 (20.33%) 14,542 (28.18%) 11,765 (22.80%) 14,804 (28.69%) Marital status Married Other 51,564 32,635 (63.29%) 18,929 (36.71%) Insurance control variables PNC paid by income No Yes 51,600 41,349 (80.13%) 10,251 (19.87%) PNC paid by insurance/HMO No Yes 51,600 25,598 (49.61%) 26,002 (50.39%) PNC paid by Medicaid No Yes 51,600 30,395 (58.91%) 21,205 (39.16%) PNC paid by military No Yes 51,600 50,488 (97.84%) 1,112 (2.16%) PNC paid by Native American Health Services No Y es 51,600 51,097 (99.03%) 503 (0.009%) Pregnancy and delivery control variables Birthweight Less than 1,500 g 1,500 g 2,499 g 2,500 g or greater 51,559 2,754 (5.34%) 10,920 (21.18%) 37,885 (73.48%) Smoking during pregnancy No Yes 51,147 45,569 (8 9.09%) 5,578 (10.91%) Vaginal delivery No Yes 51,544 16,286 (31.60%) 35,258 (68.40%) Gender of infant Male Female 51,599 26,061 (50.51%) 25,538 (49.49%) Infant in the intensive care unit (ICU) No Yes 50,758 40,259 (79.32%) 10 ,499 (20.68%)

PAGE 81

81 81 Table 4 1. Continued Categorical variables N Frequency (%) Pregnancy intention No Yes 50,893 26,143 (51.37%) 24,750 (48.63%) Breastfed (ever) No Yes 50,408 9,454 (18.75%) 40,954 (81.25%) Alcohol c onsumption in the last three months of pregnancy No Yes 50,603 47,291 (93.45%) 3,312 (6.55%) Women, Infants, and Children (WIC) during pregnancy No Yes 50,878 27,037 (53.14%) 23,841 (46.86%) Subpopulation variable ( A ppendix B): Weight gain talk during pregnancy No Yes 9,377 2,088 (22.27%) 7,289 (77.73%) High risk maternal morbidity control variables Diabetes before pregnancy No Yes 51,600 50,505 (97.88%) 1,095 (2.12%) Incompetent cervix No Yes 51,600 50,640 (98.14%) 960 (1.86%) Preterm labor No Yes 51,600 38,005 (73.65%) 13,595 (26.35%) Placenta previa or placenta abruptio No Yes 51,600 48,142 (93.30%) 3,458 (6.70%) Bedrest No Yes 51,600 40,099 (77.71%) 11,501 (22.29%) Car crash injury No Yes 51,600 50,698 (98.25%) 902 (1.75%) Blood transfusion No Yes 51,600 50,923 (98.69%) 677 (1.31%) Medical risk factors No Yes 51,600 33,391 (64.71%) 18,209 (35.29%)

PAGE 82

82 82 Table 4 1. Continued Categorical variables N Frequency (%) Hospitalized during pregnancy No Yes 51,600 41,638 (80.69%) 9,962 (19.31%) Non high risk maternal morbidity control variables Gestational diabetes No Yes 51,600 46,814 (90.73%) 4,786 (9.28%) Kidney/bladder infection No Yes 51,600 42,178 (81.74%) 9,422 (18.26%) Nausea No Yes 51,600 36,496 (70.73%) 15,104 (29.27%) High blood pressure No Yes 51,600 43,739 (84.77%) 7,861 (15.23%) Vaginal bleeding No Yes 51,600 42,682 (82.70%) 8,918 (17.28%) Premature rupture of membrane (PROM) No Yes 51,600 46,282 (89.69%) 5,318 (10.31%) L abor abnormalities No Yes 51,600 40,934 (79.33%) 10,666 (20.67%) Labor/delivery complications No Yes 51,600 33,469 (64.86%) 18,131 (35.14%) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent var iables were pre -pregnancy body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. Table 4 2. Univariate statistics (continuous variable ) Variable (continuous) N Mean Standard deviation Minimum Maximum Maternal age 51,596 27.46 6.195 12 55 The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variable was maternal age. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.

PAGE 83

83 Table 4 3. Chi -square analyses comparing 41 characteristics among women with postpartum depressive (PPD) symptoms versus women without postpartum depres sive (PPD) symptoms Categorical variables Dependent variable: N o PPD symptoms (Frequency, %) n for no PPD symptoms Dependent variable: Y es PPD symptoms (Frequency, %) n for yes PPD symptoms N P value Main variables Main independent variable: Pre pregnancy body mass index (BMI) Underweight Normal Overweight Obese 4,663 (13.5%) 17,921 (52.1%) 4,352 (12.6%) 7,491 (21.8%) 34,427 903 (14.4%) 2,979 (47.4%) 806 (12.8%) 1,603 (25.5%) 6,291 40,718 <0.0001* Main independent variable: Adequacy of Prenatal Care Utilization Index (APNCU) Inadequate Intermediate Adequate Adequate plus 3,841 (10.7%) 4,848 (13.5%) 14,641 (40.9%) 12,472 (34.8%) 35,802 1,026 (16.0%) 913 (14.2%) 2,218 (34.6%) 2,256 (35.2%) 6,413 42,215 <0.0001* Demograp hic control variables Maternal r ace : White No Yes 12,733 (35.2%) 23,430 (64.8%) 36,163 3,039 (45.6%) 3,672 (55.1%) 6,666 42,829 <0.0001* Maternal r ace : Black No Yes 31,220 (86.3%) 4,943 (13.7%) 36,163 5,299 (79.5%) 1,367 (20.5%) 6,666 42,829 <0.0001* Maternal r ace : Ot her No Yes 28,373 (78.5%) 7,790 (21.5%) 36,163 4,994 (74.9%) 1,672 (25.1%) 6,666 42,829 <0.0001*

PAGE 84

84 Table 4 3. Continued Categorical variables Dependent variable: N o PPD symptoms (Frequency, %) n for no PPD symptoms Dependent variable: Y es PPD symptoms (Frequency, %) n for yes PPD symptoms N P value Hispanic Not Hispanic Hispanic 29,905 (83.2%) 6,052 (16.8%) 35,957 5,377 (81.3%) 1,237 (18.7%) 6,614 42,571 <0.0001* Maternal education 0 8 years 9 11 years 12 years 1315 years 16+ years 1,486 (3.94%) 4,824 (12.8%) 11,252 (29.9%) 9,021 (23.9%) 11,112 (29.5%) 37,695 347 (5.08%) 1,492 (21.8%) 2,497 (36.5%) 1,559 (22.8%) 941 (13.8%) 6,836 44,531 <0.0001* Income (12 months prior) Less than $10,000 $10,000 to $24,999 $25,000 to $49,999 $50,000 or more 6,9 13 (18.0%) 10,258 (26.8%) 9,177 (23.9%) 11,972 (31.2%) 38,320 2,316 (33.3%) 2,289 (32.9%) 1,392 (20.0%) 968 (13.9%) 6,965 45,285 <0.0001* Marital status Married Other 25,517 (66.6%) 12,775 (33.4%) 38,292 3,425 (49.2%) 3,536 (50.8%) 6,961 45,253 <0.0001 Insurance control variables PNC paid by income No Yes 30,259 (79.0%) 8,061 (21.0%) 38,320 5,739 (82.4%) 1,226 (17.6%) 6,965 45,285 <0.0001* PNC paid by insurance/HMO No Yes 17,779 (46.4%) 20,541 (53.6%) 38,320 4,528 (65.0%) 2 ,437 (35.0%) 6,965 45,285 <0.0001* PNC paid by Medicaid No Yes 23,928 (62.4%) 14,392 (37.6%) 38,320 3,108 (44.6%) 3,857 (55.4%) 6,965 45,285 <0.0001* PNC paid by military No Yes 37,335 (97.4%) 985 (2.57%) 38,320 6,849 (98.3%) 116 (1. 67%) 6,965 45,285 <0.0001*

PAGE 85

85 Table 4 3. Continued Categorical variables Dependent variable: No PPD symptoms (Frequency, %) n for no PPD symptoms Dependent variable: Yes PPD symptoms (Frequency, %) n for yes PPD symptoms N P value PNC paid by Native Ameri can Health Services No Yes 37,921 (98.9%) 399 (1.04%) 38,320 6,880 (98.8%) 85 (1.22%) 6,965 45,285 0.18 Pregnancy and delivery control variables Birthweight Less than 1,500 g 1,500 g to 2,499 g 2,500 g or greater 1,914 (5.00%) 7,702 (20.1%) 28,678 (74.9%) 38,294 526 (7.57%) 1,589 (22.9%) 4,837 (69.6%) 6,952 45,246 <0.0001* Smoking during pregnancy No Yes 34,126 (98.9%) 3,828 (10.1%) 37,954 5,647 (81.8%) 1,259 (18.2%) 6,906 44,860 <0.0001* Vaginal delivery No Yes 11,933 (31. 2%) 26,343 (68.8%) 38,276 2,263 (32.5%) 4,693 (67.5%) 6,956 45,232 0.025* Gender of infant Male Female 19,370 (50.5%) 18,949 (49.5%) 38,319 3,552 (51.0%) 3,413 (49.0%) 6,965 45,284 0.49 Infant in the intensive care unit (ICU) No Yes 30,591 (80.6%) 7 ,375 (19.4%) 37,966 5,150 (75.3%) 1,693 (24.7) 6,843 44,809 <0.0001* Pregnancy intention No Yes 18,514 (48.9%) 19,335 (51.1%) 37,849 4,440 (64.7%) 2,419 (35.3%) 6,859 44,708 <0.0001* Breastfed (ever) No Yes 6,589 (17.4%) 31,189 (82.6%) 37,778 1,564 (23.3%) 5,147 (76.7%) 6,711 44,489 <0.0001*

PAGE 86

86 Table 4 3. Continued Categorical variables Dependent variable: No PPD symptoms (Frequency, %) n for no PPD symptoms Dependent variable: Yes PPD symptoms (Frequency, %) n for yes PPD symptoms N P value Alcoho l consumption in the last three months of pregnancy No Yes 35,359 (93.7%) 2,390 (6.33%) 37,749 6,361 (93.0%) 480 (7.00%) 6,841 44,590 0.034* Women, Infants, and Children (WIC) during pregnancy No Yes 21,370 (56.4%) 16,551 (43.6%) 37,921 2,641 (38 .4%) 4,233 (61.6%) 6,874 44,795 <0.0001* Subpopulation variable (Appendix B): Weight gain talk during pregnancy No Yes 1,084 (21.7%) 3,906 (78.3%) 4,990 224 (27.9%) 580 (72.1%) 804 5,794 <0.0001* High risk maternal morbidity control variables Diabetes before pregnancy No Yes 37,576 (98.0%) 744 (1.94%) 38,320 6,746 (96.9%) 219 (3.14%) 6,965 45,285 <0.0001* Incompetent cervix No Yes 37,661 (98.3%) 659 (1.72%) 38,320 6,794 (97.5%) 171 (2.46%) 6,965 45,285 <0.0001* Preterm labor No Yes 28,714 (74.9%) 9,606 (25.1%) 38,320 4,510 (64.8%) 2,455 (35.2%) 6,965 45,285 <0.0001* Placenta previa or placenta abruption No Yes 35,818 (93.5%) 2,502 (6.53%) 38,320 6,409 (92.0%) 556 (7.98%) 6,965 45,285 <0.0001* Bedrest No Yes 30,153 (78 .7%) 8,167 (21.3%) 38,320 4,918 (70.6%) 2,047 (29.3%) 6,965 45,285 <0.0001*

PAGE 87

87 Table 4 3. Continued Categorical variables Dependent variable: No PPD symptoms (Frequency, %) n for no PPD symptoms Dependent variable: Yes PPD symptoms (Frequency, %) n for ye s PPD symptoms N P value Car crash injury No Yes 37,686 (98.4%) 634 (1.65%) 38,320 6,797 (97.6%) 168 (2.41%) 6,965 45,285 <0.0001* Blood transfusion No Yes 37,895 (98.9%) 425 (1.11%) 38,320 6,812 (97.8%) 153 (2.20%) 6,965 45,285 <0.0001* Medical r isk factors No Yes 25,029 (65.3%) 13,291 (34.7%) 38,320 4,304 (61.8%) 2,661 (38.2%) 6,965 45,285 <0.0001* Hospitalized during pregnancy No Yes 31,156 (81.3%) 7,164 (18.7%) 38,320 5,269 (75.6%) 1,696 (24.4%) 6,965 45,285 <0.0001* High risk maternal morbidity control variables Gestational diabetes No Yes 34,890 (91.0%) 3,430 (8.95%) 38,320 6,210 (89.2%) 755 (10.8%) 6,965 45,285 <0.0001* Kidney/bladder infection No Yes 31,839 (83.1%) 6,481 (16.9%) 38,320 5,172 (74.3%) 1,793 (25.7%) 6,965 45,285 <0.0001* Nausea No Yes 27,787 (72.5%) 10,533 (27.4%) 38,320 4,137 (59.4%) 2,828 (40.6%) 6,965 45,285 <0.0001* High blood pressure No Yes 32,478 (84.8%) 5,842 (15.2%) 38,320 5,760 (82.7%) 1,205 (17.3%) 6,965 45,285 <0.0001*

PAGE 88

88 Table 4 3. Contin ued Categorical variables Dependent variable: No PPD symptoms (Frequency, %) n for no PPD symptoms Dependent variable: Yes PPD symptoms (Frequency, %) n for yes PPD symptoms N P value Vaginal bleeding No Yes 31,937 (83.3%) 6,383 (16.7%) 38,320 5,486 ( 14.3%) 1,479 (21.2%) 6,965 45,285 <0.0001* Premature rupture of membrane (PROM) No Yes 34,536 (90.1%) 3,784 (9.87%) 38,320 6,119 (87.9%) 846 (12.1%) 6,965 45,285 <0.0001* Labor abnormalities No Yes 29,741 (77.6%) 8,579 (22.4%) 38,320 5,356 (76.9%) 1,609 (23.1%) 6,965 45,285 0.19 Labor/delivery complications No Yes 25,010 (65.3%) 13,310 (34.7%) 38,320 4,463 (64.1%) 2,502 (35.9%) 6,965 45,285 0.06** The dependent variable for this table was postpartum depression (PPD), while the main independe nt variables were pre -pregnancy body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included all pregnancies and the yea rs of PRAMS data collection were for 2004 & 2005. Table 4 4. Maternal age (continuous variable) a nd postpartum depression (PPD) symptoms t -test results Group (PPD symptoms) Observations Mean Standard error Standard deviation 95% Confidence interval (lower, upper) No 38,317 27.65 .0313 6.12 (27.58, 27.71) Yes 6,964 25.95 .0749 6.25 (25.81, 26.10) Combined 45,281 27.38 .0290 6.17 (27.33, 27.44) Difference -------1.692 .0799 -------(1.535, 1.849) The dependent variable for this table was postpartum depression (PPD), while the main independent variable was maternal age. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.

PAGE 89

89 Table 4 5. Chi -square analyses comparing 40 characteristics among women from four body mass index (BMI) groups Categorical variable Dependent variable: U nderweight pre pregnancy BMI n Dependent variable: Normal pre pregnancy BMI n Dependent variable: Overweight pre pregnancy BMI n Dependent variable: Obese pre pregnancy BMI n N P value Main variable Adequacy of Prenatal Care Utilization Index (APNCU) Inadequate Intermediate Adequate Adequate plus 747 (12.5%) 798 (13.4%) 2,374 (39.8%) 2,045 (34.3%) 5,964 2,460 (11.1%) 3,243 (14.6%) 9,179 (41.2%) 7,384 (33.2%) 22,266 642 (11.7%) 757 (13.8%) 2,216 (40.3%) 1,880 (34.2%) 5,495 1,058 ( 11.1%) 1,190 (12.5%) 3,579 (37.6%) 3,686 (38.7%) 9,513 43,238 <0.0001* Demographic control variables Maternal r ace : White No Yes 2,349 (37.0%) 3,993 (63.0%) 6,342 7,976 (33.7%) 15,711 (66.3%) 23,687 2,223 (38.0%) 3,624 (62.6%) 5,847 4,236 (41.6%) 5,952 (58.4%) 10,188 46,064 <0.0001* Maternal r ace : Black No Yes 5,507 (86.8%) 835 (13.2%) 6,342 20,525 (86.7%) 3,162 (13.3%) 23,687 4,764 (81.5%) 1,083 (18.5%) 5,847 7,909 (77.6%) 2,279 (22.4%) 10,188 46,064 <0.0001* Maternal r ace : Other No Yes 4,828 (76.1%) 1,51 4 (23.9%) 6,342 18,873 (79.7%) 4,814 (20.3%) 23,687 4,707 (80.5%) 1,140 (19.5%) 5,847 8,231 (80.8%) 1,957 (19.2%) 10,188 46,064 <0.0001* Hispanic Not Hispanic Hispanic 5,458 (86.8%) 829 (13.2%) 6,287 19,895 (84.6%) 3,618 (15.4%) 23,513 4,717 (18.4%) 1,076 (18.6%) 5,793 8,511 (83.9%) 1,629 (16.1%) 10,140 45,733 <0.0001* Maternal education 0 8 years 9 11 years 12 years 13 15 years 16+ years 187 (3.00%) 1,080 (17.3%) 1,944 (31.1%) 1,309 (20.9%) 1,735 (27.7%) 6,255 639 (2.70%) 3,075 (13.1%) 6,711 (28 .6%) 5,374 (22.9%) 7,659 (32.6%) 23,458 226 (3.90%) 803 (13.7%) 1,835 (31.7%) 1,499 (25.9%) 1,427 (24.6%) 5,790 347 (3.43%) 1,357 (13.4%) 3,570 (35.3%) 2,898 (28.7%) 1,939 (19.2%) 10,111 45,614 <0.0001*

PAGE 90

90 Table 4 5. Continued Categorical variable Depende nt variable: Underweight pre pregnancy BMI n Dependent variable: Normal pre pregnancy BMI n Dependent variable: Overweight pre pregnancy BMI n Dependent variable: Obese pre pregnancy BMI n N P value Income (12 months prior) Less than $10,000 $10,000 to $2 4,999 $25,000 to $49,999 $50,000 or more 1,485 (23.3%) 1,899 (29.8%) 1,248 (19.6%) 1,747 (27.4%) 6,379 4,402 (18.5%) 6,105 (25.6%) 5,229 (21.9%) 8,098 (34.0%) 23,834 1,150 (19.5%) 1,691 (28.7%) 1,418 (24.1%) 1,629 (27.7%) 5,888 2,166 (21.1%) 3,103 (30.2%) 2,684 (26.1%) 2,317 (22.6%) 10,270 46,371 <0.0001* Marital status Married Other 3,836 (60.2%) 2,540 (39.8%) 6,376 15,630 (65.6%) 8,202 (34.4%) 23,832 3,737 (63.5%) 2,147 (36.5%) 5,884 6,345 (61.8%) 3,917 (38.2%) 10,262 46,354 <0.0001* Insuran ce control variables PNC paid by income No Yes 5,170 (81.0%) 1,209 (19.0%) 6,379 18,964 (79.6%) 4,870 (20.4%) 23,834 4,747 (80.6%) 1,141 (19.4%) 5,888 8,272 (80.5%) 1,998 (19.5%) 10,270 46,371 0.02* PNC paid by insurance/HMO No Yes 3,364 (52.7%) 3,015 (47.3%) 6,379 10,677 (44.8%) 13,157 (55.2%) 23,834 2,917 (49.5%) 2,971 (50.5%) 5,888 5,365 (52.2%) 4,905 (47.8%) 10,270 46,371 <0.0001* PNC paid by Medicaid No Yes 3,647 (57.2%) 2,732 (42.8%) 6,379 15,148 (63.6%) 8,686 (36.4%) 23,834 3,418 (58.1%) 2,470 (41.9%) 5,888 5,441 (53.0%) 4,829 (47.0%) 10,270 46,371 <0.0001* PNC paid by military No Yes 6,237 (97.8%) 142 (2.23%) 6,379 23,237 (97.5%) 597 (2.50%) 23,834 5,750 (97.7%) 138 (2.34%) 5,888 10,053 (97.9%) 217 (2.11%) 10,270 46,371 0.15 PNC paid by Native American Health Services No Yes 6,343 (99.4%) 36 (0.56%) 6,379 23,615 (99.1%) 219 (0.92%) 23,834 5,801 (98.5%) 87 (1.48%) 5,888 10,129 (98.6%) 141 (1.37%) 10,270 46,371 < 0.0001*

PAGE 91

91 Table 4 5. Continued Categorical variable Dependent variable: Underweight pre pregnancy BMI n Dependent variable: Normal pre pregnancy BMI n Dependent variable: Overweight pre pregnancy BMI n Dependent variable: Obese pre pregnancy BMI n N P va lue Pregnancy and delivery control variables Birthweight Less than 1,500 g 1,500 g to 2,499 g 2,500 g or greater 318 (4.99%) 1,697 (26.6%) 4,358 (68.4%) 6,373 1,075 (4.51%) 5,052 (21.2%) 17,692 (74.3%) 23,819 338 (5.74%) 1,127 (19.2%) 4,415 (75.1%) 5,880 816 (7.95%) 2,000 (19.5%) 7,449 (72.6%) 10,265 46,337 <0.0001* Smoking during pregnancy No Yes 5,522 (87.3%) 805 (12.7%) 6,327 21,258 (90.0) 2,365 (10.0%) 23,623 5,216 (89.4%) 621 (10.6%) 5,837 9,000 (88.4%) 1,176 (11.6%) 10, 176 45,963 <0.0001* Vaginal delivery No Yes 1,586 (24.9%) 4,787 (75.1%) 6,373 6,847 (28.8%) 16,960 (71.2%) 23,807 1,946 (33.1%) 3,934 (66.9%) 5,880 4,381 (42.7%) 5,877 (57.3%) 10,258 46,318 <0.0001* Gender of infant Male Female 3,161 (49.6%) 3,218 ( 50.4%) 6,379 12,019 (50.4%) 11,814 (49.6%) 23,833 3,014 (51.2%) 2,874 (48.8%) 5,888 5,195 (50.6%) 5,075 (49.4%) 10,270 46,370 0.332 Infant in the intensive care unit (ICU) No Yes 5,002 (79.9%) 1,258 (20.1%) 6,260 18,922 (80.6%) 4,564 (19.4%) 23,4 86 4,605 (79.3%) 1,205 (20.7%) 5,810 7,633 (75.4%) 2,494 (24.6%) 10,127 45,683 <0.0001* Pregnancy intention No Yes 3,451 (54.9%) 2,841 (45.1%) 6,292 11,634 (49.4%) 11,913 (50.6%) 23,547 3,051 (52.4%) 2,769 (47.6%) 5,820 5,566 (54.9%) 4,572 (45.1 %) 10,138 45,797 <0.0001* Breastfed (ever) No Yes 1,219 (19.6%) 4,986 (80.4%) 6,205 3,885 (16.6%) 19,449 (83.4%) 23,334 1,161 (20.1%) 4,606 (79.9%) 5,767 2,315 (23.1%) 7,707 (76.9%) 10,022 45,328 <0.0001*

PAGE 92

92 Table 4 5. Continued Categorical variable De pendent variable: Underweight pre pregnancy BMI N Dependent variable: Normal pre pregnancy BMI n Dependent variable: Overweight pre pregnancy BMI n Dependent variable: Obese pre pregnancy BMI n N P value Alcohol consumption in the last three months of p regnancy No Yes 5,837 (93.3%) 422 (6.74%) 6,259 21,631 (92.5%) 1,750 (7.48%) 23,381 5,437 (94.1%) 342 (5.91%) 5,779 9,657 (95.5%) 454 (4.49%) 10,111 45,530 0.034* Women, Infants, and Children (WIC) during pregnancy No Yes 3,364 (53.5%) 2,927 (46.5%) 6,291 13,885 (59.1%) 9,627 (40.9%) 23,512 3,002 (51.6%) 2,811 (49.4%) 5,813 4,629 (45.6%) 5,516 (54.4%) 10,145 45,761 <0.0001* Subpopulation variable (Appendix B): Weight gain talk during pregnancy No Yes 213 (21.1%) 795 (78.9% ) 1,008 830 (23.4%) 2,713 (76.6%) 3,543 199 (23.8%) 636 (76.2%) 835 328 (24.0%) 1,037 (76.0%) 1,365 6,751 0.359 High risk maternal morbidity control variables Diabetes before pregnancy No Yes 6,327 (99.2%) 52 (0.82%) 6,379 23,5 31 (98.7%) 303 (1.27%) 23,834 5,752 (97.7%) 136 (2.30%) 5,888 9,808 (95.5%) 462 (4.50%) 10,270 46,371 <0.0001* Incompetent cervix No Yes 6,268 (98.3%) 111 (1.74%) 6,379 23,440 (98.3%) 394 (1.65%) 23,834 5,762 (97.9%) 126 (2.14%) 5,888 10,023 (97 .6%) 247 (2.40%) 10,270 46,371 <0.0001* Preterm labor No Yes 4,462 (69.9%) 1,917 (30.1%) 6,379 17,673 (74.2%) 6,161 (25.8%) 23,834 4,352 (73.9%) 1,536 (26.1%) 5,888 7,476 (72.8%) 2,794 (27.2%) 10,270 46,371 <0.0001*

PAGE 93

93 Table 4 5. Continued Categorical variable Dependent variable: Underweight pre pregnancy BMI n Dependent variable: Normal pre pregnancy BMI n Dependent variable: Overweight pre pregnancy BMI n Dependent variable: Obese pre pregnancy BMI n N P value Placenta previa or placenta abruption No Yes 5,872 (92.1%) 507 (7.94%) 6,379 22,261 (93.4%) 1,573 (6.60%) 23,834 5,521 (93.8%) 367 (6.23%) 5,888 9,542 (92.9%) 728 (7.10%) 10,270 46,371 <0.0001* Bedrest No Yes 4,978 (78.0%) 1,401 (22.0%) 6,379 18,914 (79.4%) 4,920 (20.6%) 23,834 4, 509 (76.6%) 1,379 (23.4%) 5,888 7,372 (71.8%) 2,898 (28.2%) 10,270 46,371 <0.0001* Car crash injury No Yes 6,261 (98.2%) 118 (1.85%) 6,379 23,435 (98.3%) 399 (1.67%) 23,834 5,785 (98.3%) 103 (1.75%) 5,888 10,070 (98.1%) 200 (1.95%) 10,270 46,371 0.34 Blood transfusion No Yes 6,278 (98.4%) 101 (1.58%) 6,379 23,516 (98.7%) 318 (1.33%) 23,834 5,819 (98.3%) 69 (1.17%) 5,888 10,145 (98.8%) 125 (1.22%) 10,270 46,371 0.15 Medical risk factors No Yes 4,366 (68.4%) 2,013 (31.6%) 6,379 16,083 (67.5%) 7,751 (32.5%) 23,834 3,750 (63.7%) 2,138 (36.3%) 5,888 5,776 (56.2%) 4,494 (43.8%) 10,270 46,371 <0.0001* Hospitalized during pregnancy No Yes 5,122 (80.3%) 1,257 (19.7%) 6,379 19,436 (81.5%) 4,398 (18.5%) 23,834 4,761 (80.9%) 1,127 (19.1%) 5,888 7,881 (76.7%) 2,389 (23.3%) 10,270 46,371 <0.0001* Non high risk maternal morbidity control variables Gestational diabetes No Yes 6,023 (94.4%) 356 (5.58%) 6,379 22,191 (93.1%) 1,643 (6.89%) 23,834 5,244 (89.1%) 644 (10.9%) 5,888 8,641 (84.1%) 1,629 (15.9%) 10,270 46,371 <0.0001* Kidney/bladder infection No Yes 5,166 (81.0%) 1,213 (19.0%) 6,379 19,781 (83.0%) 4,053 (17.0%) 23,834 4,770 (81.0%) 1,118 (19.0%) 5,888 8,110 (79.0%) 2,160 (21.0%) 10,270 46,371 <0.0001*

PAGE 94

94 Tabl e 4 5. Continued Categorical variable Dependent variable: Underweight pre pregnancy BMI n Dependent variable: Normal pre pregnancy BMI n Dependent variable: Overweight pre pregnancy BMI n Dependent variable: Obese pre pregnancy BMI n N P value Nausea No Y es 4,504 (70.6%) 1,875 (29.4) 6,379 17,330 (72.7%) 6,504 (27.3%) 23,834 4,101 (69.7%) 1,787 (30.3%) 5,888 6,880 (67.0%) 3,390 (33.3%) 10,270 46,371 <0.0001* High blood pressure No Yes 5,848 (91.7%) 531 (8.32%) 6,379 20,768 (87.1%) 3,066 (12.9%) 23, 834 4,850 (82.4%) 1,038 (17.6%) 5,888 7,655 (74.5%) 2,615 (25.5%) 10,270 46,371 <0.0001* Vaginal bleeding No Yes 5,277 (82.7%) 1,102 (17.3%) 6,379 19,844 (83.3%) 3,990 (16.7%) 23,834 4,849 (82.4%) 1,039 (17.6%) 5,888 8,305 (80.9%) 1,965 (19.1%) 10, 270 46,371 <0.0001* Premature rupture of membrane (PROM) No Yes 5,683 (89.1%) 696 (10.9%) 6,379 21,465 (90.1%) 2,369 (9.94%) 23,834 5,289 (89.8%) 599 (10.2%) 5,888 9,135 (88.9%) 1,135 (11.1%) 10,270 46,371 0.007* Labor/delivery complications No Yes 4,181 (65.5%) 2,198 (34.5%) 6,379 15,554 (65.3%) 8,280 (34.7%) 23,834 3,814 (64.8%) 2,074 (35.2%) 5,888 6,614 (64.4%) 3,656 (35.6%) 10,270 46,371 0.36 Labor abnormalities No Yes 5,197 (81.5%) 1,182 (18.5%) 6,379 19,288 (80.9%) 4,546 (19.1% ) 23,834 4,667 (79.3%) 1,221 (20.7%) 5,888 7,888 (76.8%) 2,382 (23.2%) 10,270 46,371 <0.0001* The dependent variable for this table was pre pregnancy body mass index (BMI), while the main independent variables were pre prenatal care (PNC) utilization a nd control variables The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.

PAGE 95

95 Table 4 6. Chi -square analyses comparing 39 characteristics among women from four prenatal care (PNC) utilization gr oups Categorical variable Dependent variable: Inadequate PNC utilization n Dependent variable: Intermediate PNC utilization n Dependent variable: Adequate PNC utilization n Dependent variable: Adequate plus PNC utilization n N P value Demographic control variables Maternal r ace : White No Yes 2,579 (47.5%) 2,846 (52.5%) 5,425 2,652 (41.6%) 3,717 (58.4%) 6,369 6,057 (33.1%) 12,248 (66.9%) 18,305 4,912 (31.3%) 10,789 (68.7%) 15,701 45,800 <0.0001* Maternal r ace : Black No Yes 4,336 (79.9%) 1,089 (20.1%) 5,425 5,412 (85.0%) 957 (15.0%) 6,369 15,922 (87.0%) 2,383 (13.0%) 18,305 13,093 (83.4%) 2,608 (16.6%) 15,701 45,800 <0.0001* Maternal r ace : Other No Yes 3,935 (72.5%) 1,490 (27.5%) 5,425 4,674 (73.4%) 1,695 (26.6%) 6,369 14,631 (79.9%) 3,674 (20.1%) 18,305 13, 397 (85.3%) 2,304 (14.7%) 15,701 45,800 <0.0001* Hispanic Not Hispanic Hispanic 3,974 (73.6%) 1,429 (26.4%) 5,403 4,956 (78.5%) 1,361 (21.5%) 6,317 14,986 (82.3%) 3,220 (17.7%) 18,206 13,322 (85.4%) 2,280 (14.6%) 15,602 45,528 <0.0001* Maternal educa tion 0 8 years 9 11 years 12 years 1315 years 16+ years 462 (8.48%) 1,443 (26.5%) 1,937 (35.6%) 1,027 (18.9%) 579 (10.6%) 5,448 357 (5.56%) 985 (15.4%) 1,965 (30.6%) 1,494 (23.3%) 1,618 (25.2%) 6,419 719 (3.67%) 2,282 (11.6%) 5,523 (28.2%) 4,547 (23.2% ) 5,935 (30.3%) 19,600 530 (3.20%) 2,013 (12.2%) 5,048 (30.5%) 4,046 (24.4%) 4,928 (29.8%) 16,565 47,438 <0.0001* Income (12 months prior) Less than $10,000 $10,000 to $24,999 $25,000 to $49,999 $50,000 or more 2,023 (36.3%) 2,096 (37.6%) 900 (16.2%) 549 (9.86%) 5,568 1,403 (21.5%) 1,918 (29.4%) 1,473 (22.6%) 1,735 (26.7%) 6,529 3,237 (16.8%) 4,973 (25.8%) 4,754 (24.7%) 6,294 (32.7%) 19,258 2,944 (17.6%) 4,495 (26.8%) 3,887 (23.2%) 5,422 (32.4%) 16,748 48,103 <0.0001* Marital status Married Other 2, 362 (42.5%) 3,202 (57.5%) 5,564 4,083 (62.5%) 2,445 (37.5%) 6,528 13,125 (68.2%) 6,128 (31.8%) 19,253 11,142 (66.6%) 5,588 (33.4%) 16,730 48,075 <0.0001* Insurance control variables PNC paid by income No Yes 4,590 (82.4%) 978 (17.6% ) 5,568 5,299 (81.2%) 1,230 (18.8%) 6,529 15,239 (79.1%) 4,019 (20.9%) 19,258 13,294 (79.4%) 3,454 (3454%) 16,748 48,103 <0.0001* PNC paid by insurance/HMO No Yes 4,168 (74.9%) 1,400 (25.1%) 5,568 3,478 (53.3%) 3,051 (46.7%) 6,529 8,500 (44. 1%) 10,758 (55.9%) 19,258 7,418 (44.3%) 9,330 (55.7%) 16,748 48,103 <0.0001*

PAGE 96

96 Table 4 6. Continued Categorical variable Dependent variable: Inadequate PNC utilization n Dependent variable: Intermediate PNC utilization n Dependent variable: Adequate PNC u tilization n Dependent variable: Adequate plus PNC utilization n N P value PNC paid by Medicaid No Yes 2,326 (41.8%) 3,242 (58.2%) 5,568 3,800 (58.2%) 2,729 (41.8%) 6,529 12,308 (63.9%) 6,950 (36.1%) 19,258 10,143 (60.6%) 6,605 (39.4%) 16,748 48,103 <0.0001* PNC paid by military No Yes 5,489 (98.6%) 79 (1.42%) 5,568 6,247 (95.7%) 282 (4.32%) 6,529 18,874 (98.0%) 474 (2.46%) 19,258 16,526 (98.7) 222 (1.33%) 16,748 48,103 <0.0001* PNC paid by Native American Health Services No Yes 5,472 (98.3%) 96 (1.72%) 5,568 6,410 (98.2%) 119 (1.82%) 6,529 19,096 (99.2%) 162 (0.84%) 19,258 16,649 (99.4%) 99 (0.59%) 16,748 48,103 <0.0001* Pregnancy and delivery control variables Birthweight Less than 1,500 g 1,500 g to 2 ,499 g 2,500 g or greater 304 (5.46%) 1,178 (21.2%) 4,082 (73.4%) 5,564 157 (2.41%) 846 (13.0%) 5,526 (84.6%) 6,529 403 (2.09%) 2,531 (13.2%) 16,320 (84.8%) 19,254 1,622 (9.68%) 5,664 (33.8%) 9,462 (56.5%) 16,748 48,095 <0.0001* Smoking during pregnan cy No Yes 4,558 (82.8%) 950 (17.2%) 5,508 5,840 (90.1%) 644 (9.93%) 6,484 17,420 (90.9%) 1,750 (9.13%) 19,170 14,793 (88.8%) 1,875 (11.2%) 16,668 47,830 <0.0001* Vaginal delivery No Yes 1,538 (27.6%) 4,029 (72.4%) 5,567 1,708 (26.2%) 4,819 (7 3.8%) 6,527 5,335 (27.7%) 13,911 (72.3%) 19,246 6,542 (39.1%) 10,195 (60.9%) 16,737 48,077 <0.0001* Gender of infant Male Female 2,785 (50.0%) 2,783 (50.0%) 5,568 3,360 (51.5%) 3,169 (48.5%) 6,529 9,758 (50.7%) 9,500 (49.3%) 19,258 8,437 (50.4%) 8,3 10 (49.6%) 16,747 48,102 0.332 Infant in the intensive care unit (ICU) No Yes 4,267 (78.9%) 1,141 (21.1%) 5,408 5,502 (85.8%) 909 (14.2%) 6,411 16,633 (87.4%) 2,395 (12.6%) 19,028 11,308 (68.6%) 5,185 (31.4%) 16,493 47,340 <0.0001* Pregnancy inte ntion No Yes 3,870 (70.6%) 1,614 (29.4%) 5,484 3,382 (53.4%) 3,056 (48.2%) 6,338 9,098 (47.9%) 9,903 (52.1%) 19,001 7,822 (47.3%) 8,701 (52.7%) 16,523 47,446 <0.0001* Breastfed (ever) No Yes 1,337 (25.2%) 3,961 (74.8%) 5,298 1,066 (16.7%) 5,299 (83. 3%) 6,365 3,148 (16.6%) 15,788 (83.4%) 18,936 3,256 (19.8%) 13,174 (80.2%) 16,430 47,029 <0.0001*

PAGE 97

97 Table 4 6. Continued Categorical variable Dependent variable: Inadequate PNC utilization n Dependent variable: Intermediate PNC utilization n Dependent va riable: Adequate PNC utilization n Dependent variable: Adequate plus PNC utilization n N P value Alcohol consumption in the last three months of pregnancy No Yes 5,074 (93.6%) 345 (6.37%) 5,419 5,932 (93.0%) 449 (7.0%) 6,381 17,616 (93.0%) 1,319 (6.97%) 18,935 15,486 (94.2%) 960 (5.84%) 16,446 47,181 0.034* Women, Infants, and Children (WIC) during pregnancy No Yes 2,137 (39.2%) 3,314 (60.8%) 5,451 3,313 (51.5%) 3,126 (48.5%) 6,439 10,953 (57.6%) 8,064 (42.4%) 19,017 9,066 (54.8%) 7,4 67 (45.2%) 16, 533 47,440 <0.0001* Subpopulation variable (Appendix B): Weight gain talk during pregnancy No Yes 225 (27.1%) 605 (72.9%) 830 176 (25.4%) 518 (74.6%) 694 887 (22.4%) 3,082 (77.7%) 3,969 691 (20.5%) 2,675 (79.5%) 3,366 8,859 <0.0 001* High risk maternal morbidity control variables Diabetes before pregnancy No Yes 5,427 (97.5%) 141 (2.53%) 5,568 6,446 (98.7%) 83 (1.27%) 6,529 18,986 (98.6%) 272 (1.41%) 19,258 16,245 (97.0%) 503 (3.00%) 16,748 48,103 <0.0001* Incom petent cervix No Yes 5,468 (98.2%) 100 (1.80%) 5,568 6,448 (98.8%) 81 (1.24%) 6,529 19,044 (98.9%) 214 (1.11%) 19,258 16,271 (97.2%) 477 (2.84%) 16,748 48,103 <0.0001* Preterm labor No Yes 4,159 (74.7%) 1,409 (25.3%) 5,568 5,334 (81.7%) 1,195 (18.3%) 6,529 15,482 (80.4%) 3,776 (19.6%) 19,258 10,528 (62.9%) 6,220 (37.1%) 16,748 48,103 <0.0001* Placenta previa or placenta abruption No Yes 5,281 (94.8%) 287 (5.15%) 5,568 6,242 (95.6%) 287 (4.4%) 6,529 18,289 (95.0%) 969 (5.03%) 19,258 15,11 1 (90.2%) 1,637 (9.77%) 16,748 48,103 <0.0001* Bedrest No Yes 4,467 (80.2%) 1,101 (19.8%) 5,568 5,526 (84.6%) 1,003 (15.4%) 6,529 16,144 (83.8%) 3,114 (16.2%) 19,258 11,295 (67.4%) 5,453 (32.6%) 16,748 48,103 <0.0001* Car crash injury No Yes 5,464 (98.1%) 104 (1.87%) 5,568 6,435 (98.6%) 94 (1.44%) 6,529 18,948 (98.4%) 310 (1.61%) 19,258 16,431 (98.1%) 317 (1.89%) 16,748 48,103 0.04* Blood transfusion No Yes 5,469 (98.2%) 99 (1.78%) 5,568 6,460 (98.9%) 69 (1.06%) 6,529 19,093 (99.1%) 165 (0.0 9%) 19,258 16,466 (98.3%) 282 (1.68%) 16,748 48,103 <0.0001*

PAGE 98

98 Table 4 6. Continued Categorical variable Dependent variable: Inadequate PNC utilization n Dependent variable: Intermediate PNC utilization n Dependent variable: Adequate PNC utilization n Dep endent variable: Adequate plus PNC utilization n N P value Medical risk factors No Yes 3,464 (62.2%) 2,104 (37.8%) 5,568 4,636 (71.0%) 1,893 (29.0%) 6,529 13,217 (68.6%) 6,041 (31.4%) 19,258 9,553 (57.0%) 7,195 (43%) 16,748 48,103 <0.0001* Hospitali zed during pregnancy No Yes 4,582 (82.3%) 986 (17.7%) 5,568 5,733 (87.8%) 796 (12.2%) 6,529 16,932 (87.9%) 2,326 (12.1%) 19,258 11,641 (69.5%) 5,107 (30.5%) 16,748 48,103 <0.0001* Non high risk maternal morbidity control variables Gest ational diabetes No Yes 5,130 (92.1%) 438 (7.87%) 5,568 6,051 (92.7%) 478 (7.32%) 6,529 17,774 (92.3%) 1,484 (7.71%) 19,258 14,739 (88.0%) 2,009 (12.0%) 16,748 48,103 <0.0001* Kidney/bladder infection No Yes 4,360 (78.3%) 1,208 (21.7%) 5,568 5,477 (83.9%) 1,052 (16.1%) 6,529 16,131 (83.8%) 3,127 (16.2%) 19,258 13,405 (80.0%) 3,343 (20.0%) 16,748 48,103 <0.0001* Nausea No Yes 3,940 (70.8%) 1,628 (29.2%) 5,568 4,761 (72.9%) 1,768 (27.1%) 6,529 14,024 (72.8%) 5,234 (27.2%) 19,258 11,378 (67.9%) 5,370 (32.1%) 16,748 48,103 <0.0001* High blood pressure No Yes 4,778 (85.8%) 790 (14.2%) 5,568 5,850 (89.6%) 679 (10.4%) 6,529 17,108 (88.8%) 2,150 (11.2%) 19,258 13,027 (77.8%) 3,721 (22.2%) 16,748 48,103 <0.0001* Vaginal bleeding No Yes 4,771 (85.9%) 797 (14.3%) 5,568 5,641 (86.3%) 888 (13.6%) 6,529 16,435 (85.3%) 2,823 (14.7%) 19,258 12,944 (77.3%) 3,804 (22.7%) 16,748 48,103 <0.0001* Premature rupture of membrane (PROM) No Yes 5,033 (90.4%) 535 (9.60%) 5,568 6,097 (93.4%) 432 (6.61%) 6,529 18,147 (94.2%) 1,111 (5.77%) 19,258 3,949 (83.3%) 2,799 (16.7%) 16,748 48,103 <0.0001* Labor abnormalities No Yes 4,385 (78.8%) 1,183 (21.2%) 5,568 5,193 (79.5%) 1,336 (20.5%) 6,529 15,800 (82.0%) 3,458 (18.0%) 19,258 13,543 (80.9%) 3,205 (19.1%) 16,748 48,103 <0.0001* Labor/delivery complications No Yes 3,507 (63.0%) 2,061 (37.0%) 5,568 4,401 (67.4%) 2,128 (32.6%) 6,529 12,593 (65.4%) 6,665 (34.6%) 19,258 10,298 (61.5%) 6,450 (38.5%) 16,748 48,103 <0.0001* The dependent variable f or this table was prenatal care (PNC) utilization, while the main independent variables were pre -pregnancy body mass index (BMI) and the control variables. The population for this table included all pregnancies and the years of PRAM S data collection were f or 2004 & 2005.

PAGE 99

99 Table 4 7. Primary baseline logistic regression with the main effect independent variables Dependent variable: Postpartum (PPD) depression symptoms Odds ratio P value 95% Confidence interval (lower, upper) Main effect independent variabl e: Pre pregnancy BMI Underweight 1.047 0.54 (0.905, 1.212) Overweight 1.05 0.52 (0.903, 1.224) Obese 1.15 0.02* (1.023, 1.302) Main effect independent variable: PNC utilization Inadequate 1.84 <0.0001* (1.589, 2.142) Intermediate 1.19 0.02* (1 .025, 1.380) Adequate plus 1.28 <0.0001* (1.141, 1.438) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were pre -pregnancy body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval (CI). Table 4 8. Specific aim 1: Primary risk adjusted logistic regression with the main effect independent variables and control variables Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) Main effect independent variable: Pre pregnancy BMI Normal (reference) 1.00 ----------------------Underweight 0.87 0.08** (0.735, 1.018) Overweight 0.94 0.46 (0.798, 1.107) Obese 0.91 0.15 (0.792, 1.036) Main effect independent variable: PNC utilization Adequate (reference) 1.00 -----------------------Inadequate 1.10 0 .26 (0.931, 1.309) Intermediate 1.08 0.33 (0.923, 1.273) Adequate plus 1.08 0.23 (0.952, 1.232) Demographic control variables Maternal race: White (reference) 1.00 -----------------------Maternal race: Black 1.36 <0.0001* (1.159, 1.592) Matern al race: Other 1.51 <0.0001* (1.305, 1.741) Hispanic ethnicity 0.997 0.97 (0.848, 1.172) Maternal education 0.91 0.003* (0.851, 0.969) Maternal age 0.99 0.14 (0.981, 1.003) Higher income: $50,000 or more (reference) 1.00 -----------------------Ver y low income: Less than $10,000 2.14 <0.0001* (1.689, 2.712)

PAGE 100

100 Table 4 8. Continued Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) Low income: $10,000 $24,999 1.68 <0.0001* (1.366, 2.065 ) Moderate income: $25,000 $49,999 1.51 <0.0001* (1.269, 1.785) Marital status 1.01 0.86 (0.882, 1.163) Insurance control variables PNC paid by income (reference) 1.00 -----------------------PNC paid by insurance/HMO 0.90 0.18 (0.764, 1.051) P NC paid by Medicaid 1.09 0.30 (0.926, 1.280) PNC paid by military 0.89 0.56 (0.592, 1.325) PNC paid by Native American/Alaskan HS 0.73 0.06** (0.518, 1.016) Pregnancy and delivery control variables Birthweight 1.05 0.45 (0.927, 1.184) Smoking durin g pregnancy 0.79 0.004* (0.676, 0.930) Vaginal delivery 0.91 0.11 (0.805, 1.021) Gender of infant 1.06 0.28 (0.954, 1.177) Infant in the intensive care unit (ICU) 1.27 0.006* (1.073, 1.504) Pregnancy intention 0.82 0.001* (0.722, 0.919) Breastfed 0.93 0.31 (0.818, 1.065) Alcohol during pregnancy 1.27 0.02* (1.033, 1.569) Women, Infants, and Children during pregnancy 1.05 0.47 (0.912, 1.220) High risk maternal morbidity control variables Diabetes before pregnancy 1.36 0.12 (0.928, 1.980) Cervix sewn shut (incompetent) 0.95 0.80 (0.629, 1.427) Preterm labor 1.44 <0.0001* (1.262, 1.641) Placenta previa or placenta abruptio 1.04 0.76 (0.826, 1.297) Bedrest during pregnancy 1.16 0.047* (1.002, 1.333) Medical risk factors 1.12 0.07** (0.992, 1.24 7) Hospitalized during pregnancy 1.06 0.50 (0.900, 1.241) Gestational diabetes 1.18 0.08** (0.979, 1.424) Vaginal bleeding 1.27 0.001* (1.100, 1.455) Kidney/bladder infection 1.46 <0.0001* (1.282, 1.656) High blood pressure 0.95 0.54 (0.814, 1.114) Premature rupture of membrane (PROM) 0.69 0.001* (0.549, 0.862) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were pre -pregnancy body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a double asterisk corresponds to a 90% confidence interval (CI).

PAGE 101

101 Table 4 9. Specific aim 2: Primary risk adjusted logistic regression with the main effect independent variables, interaction effect variables, and control variables Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval ( lower, upper) Main effect independent variable: Pre pregnancy BMI Normal (reference) 1.00 -----------------------Underweight 0.84 0.18 (0.656, 1.081) Overweight 0.86 0.25 (0.656, 1.116) Obese 0.92 0.43 (0.746, 1.133) Main effect independent v ariable: PNC utilization Adequate (reference) 1.00 -----------------------Inadequate 1.11 0.40 (0.874, 1.400) Intermediate 1.05 0.67 (0.836, 1.324) Adequate plus 1.06 0.55 (0.882, 1.266) Interaction effect variables: Pre pregnancy BMI/PNC utili zation Obese BMI/Adequate PNC (reference) 1.00 -----------------------Obese BMI/Inadequate PNC 0.71 0.11 (0.459, 1.083) Obese BMI/Intermediate PNC 1.20 0.36 (0.813, 1.770) Obese BMI/Adequate plus PNC 0.99 0.97 (0.735, 1.343) Overweight BMI/Adequate PNC (reference) 1.00 -----------------------Overweight BMI/Inadequate PNC 1.40 0.19 (0.848, 2.318) Overweight BMI/Intermediate PNC 0.95 0.83 (0.588, 1.528) Overweight BMI/Adequate plus PNC 1.18 0.42 (0.793, 1.743) Underweight BMI/Adequate PNC (reference) 1.00 -----------------------Underweight BMI/Inadequate PNC 1.13 0.63 (0.697, 1.822) Underweight BMI/Intermediate PNC 0.98 0.92 (0.587, 1.620) Underweight BMI/A dequate plus PNC 1.03 0.86 (0.703, 1.522) Demographic control variables Maternal race: White (reference) 1.00 -----------------------Maternal race: Black 1.36 <0.0001* (1.159, 1.591) Maternal race: Other 1.50 <0.0001* (1.303, 1.737) Hispanic ethnicity 0.997 0.97 (0.848, 1.162) Maternal education 0.91 0.003* (0.850, 0.968) Maternal age 0.99 0.17 (0.981, 1.003) Higher income: $50,000 or more (reference) 1.00 -----------------------Very low income: Less than $10,000 2.16 <0.0001* (1.702, 2.736) Low income: $10,000 $24,999 1.69 <0.0001* (1.371, 2.073) Moderate income: $25,000 $49,999 1.50 <0.0001* (1.267, 1.784) Marital status 1.01 0.86 (0.882, 1.163) Insurance control variables PNC paid by income (reference) 1.00 ----------------------

PAGE 102

102 Table 4 9. Continued Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) PNC paid by insurance/HMO 0.90 0.18 (0.764, 1.051) PNC paid by Medicaid 1.09 0.29 (0.928, 1.283) PNC pai d by military 0.89 0.58 (0.596, 1.335) PNC paid by Native American/Alaskan HS 0.73 0.07** (0.521, 1.022) Pregnancy and delivery control variables Birthweight 1.05 0.44 (0.929, 1.186) Smoking during pregnancy 0.79 0.004* (0.677, 0.930) Vaginal deliv ery 0.91 0.11 (0.806, 1.022) Gender of infant 1.06 0.28 (0.954, 1.177) Infant in the intensive care unit (ICU) 1.28 0.004* (1.080, 1.514) Pregnancy intention 0.82 0.001* (0.724, 0.922) Breastfed 0.93 0.31 (0.818, 1.065) Alcohol during pregnancy 1.27 0 .03* (1.028, 1.564) Women, Infants, and Children during pregnancy 1.05 0.49 (0.910, 1.218) High risk maternal morbidity control variables Diabetes before pregnancy 1.35 0.12 (0.929, 1.975) Cervix sewn shut (incompetent) 0.95 0.81 (0.631, 1.432) Pre term labor 1.44 <0.0001* (1.261, 1.640) Placenta previa or placenta abruptio 1.03 0.80 (0.822, 1.291) Bedrest during pregnancy 1.16 0.04* (1.006, 1.337) Medical risk factors 1.11 0.07** (0.992, 1.247) Hospitalized during pregnancy 1.05 0.55 (0.894, 1. 235) Gestational diabetes 1.18 0.08** (0.981, 1.433) Vaginal bleeding 1.27 0.001* (1.103, 1.459) Kidney/bladder infection 1.46 <0.0001* (1.282, 1.655) High blood pressure 0.96 0.61 (0.817, 1.118) Premature rupture of membrane (PROM) 0.69 0.001* (0.55 3, 0.869) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were pre -pregnancy body mass index (BMI), prenatal care (PNC) utilization, and the pre -pregnancy BMI/PNC utilization interaction effect variables. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a double asterisk corresponds to a 90% confidence interval (CI).

PAGE 103

103 Table 4 10. Wald tests for pre -pregnancy BMI/PNC interaction terms: Primary risk adjusted logistic regression Interaction effect variables P value Interaction effect variables tested equal to each other within a pre pregnancy BMI group Obese BMI/Ina dequate PNC Obese BMI/Intermediate PNC = 0 Obese BMI/Inadequate PNC Obese BMI/Adequate plus PNC = 0 0.11 Overweight BMI/Inadequate PNC Overweight BMI/Intermediate PNC = 0 Overweight BMI/Inadequate PNC Overweight BMI/Adequate plus PNC = 0 0.42 Und erweight BMI/Inadequate PNC Underweight BMI/Intermediate PNC = 0 Underweight BMI/Inadequate PNC Underweight BMI/Adequate plus PNC = 0 0.89 Each interaction effect variable tested equal to 0 Obese BMI/Inadequate PNC = 0 Obese BMI/Intermediate PNC = 0 Obese BMI/Adequate plus PNC = 0 0.21 Overweight BMI/Inadequate PNC = 0 Overweight BMI/Intermediate PNC = 0 Overweight BMI/Adequate plus PNC = 0 0.48 Underweight BMI/Inadequate PNC = 0 Underweight BMI/Intermediate PNC = 0 Underweight BMI/Adequate plus PN C = 0 0.96 The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were the pre -pregnancy BMI/PNC utilization interaction effect variables. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. Table 4 11. Secondary baseline logistic regression (healthy pregnancies only) with the main effect independent variables Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P va lue 95% Confidence interval (lower, upper) Main effect independent variable: Pre pregnancy BMI Underweight 0.97 0.82 (0.744, 1.263) Overweight 1.13 0.36 (0.870, 1.475) Obese 1.12 0.30 (0.900, 1.403) Main effect independent variable: PNC utilization Inadequate 1.97 <0.0001* (1.529, 2.541) Intermediate 1.05 0.68 (0.827, 1.336) Adequate plus 1.11 0.37 (0.885, 1.388) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were pre -pregnancy body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included healthy pregnancies only, and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a double asterisk corresponds to a 90% confidence interval (CI).

PAGE 104

104 Table 4 12. Specific aim 1: Secondary risk adjusted logistic regression (healthy pregnancies only) with the main effect independent variables, interaction effect variables, and control variables Depe ndent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) Main effect independent variable: Pre pregnancy BMI Normal (reference) 1.00 -----------------------Underweight 0.89 0.44 (0.672, 1.18 7) Overweight 1.00 0.98 (0.746, 1.329) Obese 0.94 0.62 (0.737, 1.199) Main effect independent variable: PNC utilization Adequate (reference) 1.00 -----------------------Inadequate 1.17 0.30 (0.872, 1.562) Intermediate 0.94 0.67 (0.729, 1.223) Adequate plus 1.03 0.83 (0.805, 1.312) Demographic control variables Maternal race: White (reference) 1.00 -----------------------Maternal race: Black 1.32 0.07** (0.977, 1.778) Maternal race: Other 1.81 <0.0001* (1.426, 2.300) Hispanic ethni city 1.08 0.58 (0.820, 1.424) Maternal education 0.94 0.30 (0.843, 1.055) Maternal age 0.99 0.23 (0.965, 1.009) Higher income: $50,000 or more (reference) 1.00 -----------------------Very low income: Less than $10,000 2.29 <0.0001* (1.523, 3.445) Low income: $10,000 $24,999 1.52 0.02* (1.080, 2.147) Moderate income: $25,000 $49,999 1.36 0.03* (1.031, 1.802) Marital status 1.01 0.91 (0.780, 1.287) Insurance control variables PNC paid by income (reference) 1.00 -----------------------PNC paid by insurance/HMO 0.99 0.93 (0.749, 1.302) PNC paid by Medicaid 1.34 0.046* (1.005, 1.774) PNC paid by military 1.15 0.66 (0.617, 2.142) PNC paid by Native American/Alaskan HS 1.33 0.51 (0.575, 3.066) Pregnancy and delivery control variables Sm oking during pregnancy 0.82 0.23 (0.591, 1.135) Vaginal delivery 0.86 0.19 (0.693, 1.075) Gender of infant 1.16 0.12 (0.963, 1.386) Infant in the intensive care unit (ICU) 1.22 0.30 (0.839, 1.764) Pregnancy intention 0.86 0.15 (0.698, 1.056) Breastfed 1.09 0.50 (0.850, 1.397) Alcohol during pregnancy 1.14 0.47 (0.804, 1.610) Women, Infants, and Children during pregnancy 0.94 0.62 (0.730, 1.205)

PAGE 105

105 Table 4 12. Continued Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% C onfidence interval (lower, upper) Non high risk maternal morbidity control variables Gestational diabetes 1.49 0.09** (0.940, 2.376) Vaginal bleeding 0.80 0.15 (0.541, 1.096) Kidney/bladder infection 1.63 <0.0001* (1.267, 2.092) High blood pressur e 0.83 0.43 (0.528, 1.314) Nausea 1.57 <0.0001* (1.259, 1.948) Premature rupture of membrane (PROM) 1.58 <0.0001* (1.259, 1.948) Labor abnormalities 1.32 0.01* (1.057, 1.640) Labor/delivery complications 0.94 0.60 (0.756, 1.177) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were pre -pregnancy body mass index (BMI) and prenatal care (PNC) utilization. The population for this table included healthy pregnancies only, and the years of P RAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a double asterisk corresponds to a 90% confidence interval (CI). Table 4 13. Specific aim 2: Secondary risk adjusted logistic regression (healthy pregnanci es only) with the main effect independent variables, interaction effect variables, and control variables Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) Main effect independent variable: Pre pregnancy BMI Normal (reference) 1.00 -----------------------Underweight 0.90 0.61 (0.606, 1.343) Overweight 0.81 0.33 (0.531, 1.236) Obese 1.15 0.42 (0.819, 1.612) Main effect independent variable: PNC utilization Adequate (reference ) 1.00 -----------------------Inadequate 1.25 0.26 (0.847, 1.836) Intermediate 0.98 0.92 (0.692, 1.394) Adequate plus 1.03 0.85 (0.743, 1.435) Interaction effect variables: Pre pregnancy BMI/PNC utilization Obese BMI/Adequate PNC (reference) 1.00 -----------------------Obese BMI/Inadequate PNC 0.51 0.08** (0.243, 1.090) Obese BMI/Intermediate PNC 0.87 0.67 (0.456, 1.654) Obese BMI/Adequate Plus PNC 0.61 0.11 (0.331, 1.126) Overweight BMI/Adequate PNC (re ference) 1.00 -----------------------

PAGE 106

106 Table 4 13. Continued Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) Overweight BMI/Inadequate PNC 1.70 0.21 (0.738, 3.902) Overwe ight BMI/Intermediate PNC 1.03 0.93 (0.465, 2.299) Overweight BMI/Adequate Plus PNC 1.73 0.16 (0.810, 3.698) Underweight BMI/Adequate PNC (reference) 1.00 -----------------------Underweight BMI/Inadequate PNC 0.88 0.70 (0.373, 2.062) Underweight BMI/Intermediate PNC 0.89 0.78 (0.386, 2.046) Underweight BMI/Adequate Plus PNC 1.13 0.75 (0.538, 2.367) Demographic control variables Maternal race: White (reference) 1.00 -----------------------Maternal race: B lack 1.31 0.08** (0.969, 1.770) Maternal race: Other 1.82 <0.0001* (1.430, 2.307) Hispanic ethnicity 1.08 0.57 (0.823, 1.430) Maternal education 0.95 0.33 (0.846, 1.058) Maternal age 0.99 0.26 (0.966, 1.009) Higher income: $50,000 or more (reference) 1.00 -----------------------Very low income: Less than $10,000 2.30 <0.0001* (1.527, 3.451) Low income: $10,000 $24,999 1.53 0.02* (1.082, 2.153) Moderate income: $25,000 $49,999 1.35 0.03* (1.023, 1.789) Marital status 1.03 0.82 (0.810, 1.303) In surance control variables PNC paid by income (reference) 1.00 -----------------------PNC paid by insurance/HMO 0.99 0.94 (0.750, 1.304) PNC paid by Medicaid 1.34 0.04* (1.010, 1.785) PNC paid by military 1.18 0.61 (0.633, 2.184) PNC paid by Nat ive American/Alaskan HS 1.27 0.59 (0.535, 2.994) Pregnancy and delivery control variables Smoking during pregnancy 0.82 0.23 (0.594, 1.134) Vaginal delivery 0.86 0.19 (0.693, 1.076) Gender of infant 1.16 0.12 (0.965, 1.389) Intensive care unit 1.23 0.27 (0.849, 1.780) Pregnancy intention 0.86 0.15 (0.700, 1.058) Breastfed 1.09 0.48 (0.852, 1.403) Alcohol during pregnancy 1.12 0.52 (0.790, 1.591) Women, Infants, and Children during pregnancy 0.94 0.61 (0.730, 1.203) Non high risk maternal morbid ity control variables Gestational diabetes 1.48 0.096** (0.933, 2.355) Vaginal bleeding 0.76 0.14 (0.536, 1.088) Kidney/bladder infection 1.63 <0.0001* (1.266, 2.090) High blood pressure 0.83 0.43 (0.525, 1.315) Nausea 1.57 <0.0001* (1.261, 1.952) Premature rupture of membrane (PROM) 1.65 0.49 (0.393, 6.955)

PAGE 107

107 Table 4 13. Continued Dependent variable: Postpartum depression (PPD) symptoms Odds ratio P value 95% Confidence interval (lower, upper) Labor abnormalities 1.31 0.02* (0.052, 0.491) Lab or/delivery complications 0.93 0.53 (0.746, 1.163) The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were pre -pregnancy body mass index (BMI), prenatal care (PNC) utilization, and the pre -pregnancy BMI/PNC utilization interaction effect variables. The population for this table included healthy pregnancies only, and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval and a double aste risk corresponds to a 90% confidence interval (CI) Table 4 14. Wald tests for pre -pregnancy BMI/PNC interaction terms: Secondary risk adjusted logistic regression (healthy pregnancies only) Interaction effect variables P value Interaction effect variable s tested equal to each other within a pre pregnancy BMI group Obese BMI/Inadequate PNC Obese BMI/Intermediate PNC = 0 Obese BMI/Inadequate PNC Obese BMI/Adequate plus PNC = 0 0.46 Overweight BMI/Inadequate PNC Overweight BMI/Intermediate PNC = 0 O verweight BMI/Inadequate PNC Overweight BMI/Adequate plus PNC = 0 0.47 Underweight BMI/Inadequate PNC Underweight BMI/Intermediate PNC = 0 Underweight BMI/Inadequate PNC Underweight BMI/Adequate plus PNC = 0 0.77 Each interaction effect variable te sted equal to 0 Obese BMI/Inadequate PNC = 0 Obese BMI/Intermediate PNC = 0 Obese BMI/Adequate plus PNC = 0 0.22 Overweight BMI/Inadequate PNC = 0 Overweight BMI/Intermediate PNC = 0 Overweight BMI/Adequate plus PNC = 0 0.37 Underweight BMI/Inadequate PNC = 0 Underweight BMI/Intermediate PNC = 0 Underweight BMI/Adequate plus PNC = 0 0.92 The dependent variable for this table was postpartum depression (PPD) symptoms, while the main independent variables were the pre -pregnancy BMI/PNC utilization interac tion effect variables. The population for this table included healthy pregnancies only, and the years of PRAMS data collection were for 2004 & 2005.

PAGE 108

108 Figure 4 1. Primary risk adjusted logistic regression with postpartum depression (PPD) symptom odds rati os for each interaction effect variable (Note: Odds ratios for PPD symptoms are presented on the vertical axis; interaction groups are presented on the horizontal axis). 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6Underweight BMI/Inadequate PNC Underweight BMI/Intermediate PNC Underweight BMI/Adequate Plus PNC Overweight BMI/Inadequate PNC Overweight BMI/Intermediate PNC Overweight BMI/Adequate Plus PNC Obese BMI/Inadequate PNC Obese BMI/Intermediate PNC Obese BMI/Adequate Plus PNC

PAGE 109

109 CHAPTER 5 DISCUSSION This study sought to determine the role that PNC utilization p lays in the relationship between pre -pregnancy BMI and postpartum depression (PPD) symptoms among a sample of 51,600 women in the United States. These women represented 16 states across the nation: Alaska, Colorado, Georgia, Hawaii, Illinois, Maine, Minnes ota, North Carolina, Nebraska, New Mexico, Oregon, Rhode Island, South Carolina, Utah, Vermont, and Washington. Even though a consistent moderating effect of PNC was not seen throughout the association between pre pregnancy BMI and PPD symptoms, patterns that warrant attention and provide insight into this model with pre -pregnancy BMI, PNC utilization, and PPD symptoms were seen among the univariate analyses, the bivariate analyses, and the multivariate analyses Univariate Analyses The frequencies reporte d in the univariate analyses showed that there are characteristics of the sample that call for discussion. With regards to the frequencies of the dependent variable, PPD symptoms, removing the highrisk pregnancies from the analysis did not change the prev alence drastically. In comparing the frequencies of PPD symptoms with previous literature, the prevalence of postpartum blues was a little higher than what has been reported in previous literature, approximately 5080%, while the prevalence of PPD symptoms was within the average range for PPD symptoms as reported in previous literature for non -psychotic PPD, or approximately 10 15% ( Miller, 2002; Evins & Theofrastous, 1997; Negus Jolley & Betrus, 2007). However, in comparing PPD frequencies between this s tudy and previous studies, it should be noted that the PPD measure used for this study was self -reported PPD symptoms whereas many of the frequencies reported in previous literature reflect those obtained from screening instruments that were used to diagno se PPD (not self -reported). However, given that

PAGE 110

110 the measure was self reported, the methodology used to categorize observations into PPD symptoms versus no PPD symptoms mirrored an instrument that has demonstrated sensitivity and specificity, criterion an d construct validities for diagnosing major depressive disorder (Kroenke et al., 2003). Also, with regard to these PPD frequencies, it is important to note that the method of categorizing PPD symptoms (e.g., blues versus actual depressive symptoms) for t he primary risk adjusted logistic regression model and the secondary risk adjusted logistic regression model that removed high -risk pregnancies may have included both women who self reported non -psychotic PPD symptoms and women who reported PPD psychosis s ymptoms. Scores of 3 or greater in the PHQ 2 are not meant to diagnose the severity of depression, but are rather used to screen for depression (Kroenke et al., 2003). Therefore, the prevalence for nonpsychotic PPD symptoms in this study may have been les s than 15.4% and 11.5% (as reported in this study) for the risk adjusted primary and secondary logistic regression models, respectively. Frequencies for the main effect independent variables showed some results that call for discussion. Though frequencies for pre -pregnancy BMI showed that over half of the women in the sample were classified as having had a normal pre -pregnancy BMI (the high number of women in this group was expected), the next highest percentage in the sample was for women who had an obese pre -pregnancy BMI (n =10,270). Some groups of women, who comprise highrisk populations, are sampled at a higher rate, via stratification variables in PRAMS, so that a sufficient quantity of data are available (CDC, 2007). Since obesity is a risk -factor t hat has been associated with outcomes such as complications during pregnancy (e.g., preeclampsia, respiratory problems, etc.) ( LaCoursiere et al., 2005; Saravanakumar et al., 2006; Cedergren, 2004), in this study, obesity was considered to be a general ris k factor for PPD symptoms. However, since the stratification variables in PRAMS for the 2004 and 2005 years of data (see p.

PAGE 111

111 53 for a list of the stratification variables) did not include pre pregnancy BMI, this group was not oversampled. If response rates are higher than normal for groups of women in PRAMS, they are automatically adjusted for by the analysis weight that was incorporated into the PRAMS dataset and used in the analyses for this study. However, in comparing these percentages with two previous studies that used PRAMS to look at pre -pregnancy BMI, Kim, Dietz, England, Morrow, & Callaghan (2007), who looked at 20022003 PRAMS data, showed the following ranges for pre pregnancy BMI among 9 states: underweight (13%16%), normal (45%-54%), overweight (11% 14%), and obese (18%26%), while DAngelo et al. (2007) showed the following ranges for pre -pregnancy BMI among 26 states using 2004 PRAMS data: underweight (10%17%), normal (reference group), overweight (11%15%), and obese (15%26%). Thus, even though the PRAMS dataset used for this study was for 2004 and 2005, the percentages of pre pregnancy BMI in this study were within average ranges compared to previous studies. However, t he percentages of this study differ (in descending order) from the resul ts demonstrated by LaCoursiere et al. (2006), who also sampled women through the PRAMS 20002001 years of data (in the state of Utah only). Results from their study showed that the highest percentage of women had a normal pre -pregnancy BMI (who represented about half of their sample), women who had an underweight pre -pregnancy BMI were the second largest in percentage (about 20% of the sample), women who had an obese pre -pregnancy BMI comprised the next group at 16%, and then women who had an overweight pre -pregnancy BMI was the lowest percentage in the sample (about 11%). For PNC utilization, it was expected that the highest percentage would be comprised of women who utilized an adequate quantity of PNC. However, the next highest percentage, which includ ed high risk pregnancies, was comprised of women who utilized an adequate

PAGE 112

112 plus quantity of PNC. Even though adequacy of PNC was previously a stratification variable in PRAMS (e.g., 1990 and 1991 PRAMS data, 1995 PRAMS data) (Goodwin et al., 2006, Cente rs for Disease Control, 1995), it was not one of the stratification variables in the 2004 and 2005 years of data from which oversampling occurred. Thus, as mentioned for women who had an obese pre -pregnancy BMI, the high response rate for women in this PNC utilization group was adjusted for by the analysis weight that was incorporated into the PRAMS dataset and used in the analyses for this study. Comparing these percentages with trends in PNC over time, according to Kogan et al. (1998), who used national b irth records, the percentage of utilization (according to the APNCU Index) occurred as follows: there was a decrease in inadequate PNC utilization from 12% in 1981 to 8.9% in 1995, there was a decrease in intermediate PNC utilization from 23.2% in 1981 to 17.2% in 1995, there was a slight decrease in adequate PNC utilization from 45.1% in 1981 to 43.9% in 1995, and there was a significant increase in adequate plus PNC (intensive) from 18.4% in 1981 to 28.8% in 1995. Thus, given the 10-year gap between Kogan et al. (1998) and the use of the APNCU Index, calculated from the PNC information provided on the birth certificates and linked with the PRAMS data, it seems that there were slight decreases in intermediate and adequate PNC utilization, as slight increase s in inadequate and adequate plus PNC utilization with the data used for this study. To further explain characteristics that may, in part, have contributed to the large number of adequate plus PNC observations, a variety of maternal morbidities that affec ted women in the sample were investigated, many of which provide insight into the risk-status of a large portion of the women in the sample. The univariate results for the maternal morbidities showed that there were a variety of potential health risks (mat ernal morbidities) that may have prompted the delivery of adequate plus PNC, as it appears that there were a number of women in the sample

PAGE 113

113 who possessed high risk characteristics. One thing that is unclear, however, is whether women who answered yes to h aving medical risk factors during pregnancy in the PRAMS questionnaire had risk factors that actually warranted utilization of adequate plus PNC. Bivariate Analyses Given that there was a high number of women (as shown in the univariate results) who had a normal pre -pregnancy BMI, this may explain that the highest percentage of women who experienced PPD symptoms was also women from this BMI category; however, the result showing that one -fourth of women with PPD symptoms had an obese pre pregnancy BMI, wi th this result being statistically significant (p<0.0001*) provides evidence for a significant difference in percentages for PPD symptoms across all pre -pregnancy BMI categories. The chi square results presented for pre -pregnancy BMI and PPD symptoms mostl y countered the results for pre -pregnancy BMI and PPD symptoms by LaCoursiere et al. (2006). This study showed a significant difference in percentages for PPD symptoms across pre -pregnancy BMI categories in the following order: women who had a normal pre -p regnancy BMI, who had the highest percentage of PPD symptoms, followed by women who had an obese pre pregnancy BMI, then followed by women who had an underweight pre -pregnancy BMI, and finally, women who had an overweight pre pregnancy BMI, who had the low est percentage of PPD symptoms. LaCoursiere et al. (2006) showed the highest percentage of PPD symptoms among women who had an obese pre -pregnancy BMI, followed by women who had an underweight pre -pregnancy BMI, then followed by women who had an overweight pre -pregnancy BMI, and finally, women who had a normal pre pregnancy BMI, who had the lowest percentage of PPD symptoms. The results for women without PPD symptoms for this study matched the results for women with PPD symptoms (in order), but were shown a s different by LaCoursiere et al. (2006) as highest for women who had a normal pre -pregnancy BMI, followed by women who had an underweight

PAGE 114

114 pre pregnancy BMI, then followed by women who had an obese pre -pregnancy BMI, and finally, women who had an overweigh t pre -pregnancy BMI, who had the lowest percentage. So surprisingly, and contrary to what was hypothesized for this study, these results demonstrated that women who had an overweight pre pregnancy BMI reported the lowest percentage of self reported PPD sy mptoms, whereas LaCoursiere et al. (2006) demonstrated that women who had a normal pre -pregnancy BMI reported the lowest percentage of self -reported moderate or greater PPD symptoms. It was not surprising that there was a significant difference in per centages for PPD symptoms across the four PNC utilization groups. It is worthy to note that the highest percentage for PPD symptoms was among women who utilized adequate plus PNC. Though the outcome was a birth outcome and not a postpartum outcome, Kotelch uck (1994) revealed a U -shaped association between PNC utilization and low birthweight rates, with women who utilized inadequate PNC and adequate plus PNC having the highest rates of low birthweight babies. Though the adequate plus groups were on the highest end in both this study and Kotelchuck (1994), on the low end, inadequate was higher than intermediate in this study, but not as high so as to form a U -shape as was the case for Kotelchuck (1994); rather, the shap e in this study was a J -shape Cons idering the variables in which there were no significant differences in percentages for PPD symptoms, it was surprising that gender of the infant did not demonstrate a significant difference. This is considering that women receive pressures, especially i n Asian cultures, to give birth to a boy, and, giving birth to a girl, when a boy is expected, has been found to increase the risk for PPD with this risk increasing as the number of female children increase (Chan, Levy, Chung, & Lee, 2002, Dindar & Erdogan 2007; Patel, Rodrigues, & DeSouza, 2002; Rahman,

PAGE 115

115 Iqbal, & Harrington, 2003). Given that the women in the sample comprised a variety of races/ethnicities, it was expected that gender of the infant would demonstrate a significant difference in percentage s for PPD symptoms between women who gave birth to a female versus a male infant. With regards to the chi -square analyses on pre -pregnancy BMI, the lack of a significance difference in percentages for weight gain discussion during PNC across the four pre -pregnancy BMI groups was surprising as significant difference was expected for this variable. This is considering that 1) obesity has been shown in a number of studies as a risk -factor for complications during pregnancy and/or adverse pregnancy outcomes L aCoursiere et al., 2005; Saravanakumar et al., 2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005; Mahmood, 2009; Baeten et al., 2001; Cnattingius et al., 1998), 2) it has been found that women who were overweight or obese prior to pregn ancy were more likely to experience excessive weight gain during pregnancy (Lederman et al., 2002; Olafsdottir et al., 2006), and 3) women who have an underweight pre -pregnancy BMI have a higher odds for delivering a preterm infant (Siega Riz, Adair, & Ho bel, 1996). However, it is recommended that 1) clinicians support and encourage women during PNC delivery in gaining the appropriate amount of weight during pregnancy (Lederman, 2001), and 2) young obese women who are planning a pregnancy be cautioned of t he possible complications during pregnancy and/or at birth (Dietl, 2005). Further considering that obesity has been previously shown as a risk-factor for complications during pregnancy and/or adverse pregnancy outcomes LaCoursiere et al., 2005; Saravanakum ar et al., 2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005; Mahmood, 2009; Baeten et al., 2001; Cnattingius et al., 1998), it was expected that labor/delivery complications would be significantly different in percentages across the fo ur pre -pregnancy BMI groups. However, in

PAGE 116

116 comparing this result with results from the multivariate analyses, significance was not demonstrated for labor/delivery complications as a highrisk variable (p. 167) with the logit model holding adequate plus PNC a s the dependent variable, and for the secondary logistic regression models (that removed the high risk pregnancies form the analyses), this variable was not significant for either specific aim 1 or specific aim 2. Thus, consistency remained with the result s of this variable. Though this may or may not have been related to PPD symptoms in this study, the chi square analyses addressing pre -pregnancy BMI and the maternal morbidities (Table 4 5) suggest that the majority of these morbidities affected women who had an obese pre -pregnancy BMI the most (compared to the three other pre pregnancy BMI groups). Interestingly enough, the bivariate results showed that the highest percentage of preterm labor was among women who had an underweight pre -pregnancy BMI, with these women also having the highest percentage of low birth weight babies (between 1,500 to 2,499 grams). This may be expected considering a consequence of preterm birth for the infant may be low birth weight. Surprisingly, women who had an obese pre -pregn ancy BMI had the highest percentage of very low birth weight babies (less than 1,500 grams), compared to women from the three other pre pregnancy BMI groups. This result is surprising considering that previous studies have found higher/excessive pre preg nancy weight to be associated with giving birth to a macrosomic infant (Baeten et al., 2001; Rosenberg, Garbers, Chavkin, & Chiasson, 2003; Cedergren, 2004). Though it is unclear in this sample whether giving birth to a very low birth weight baby caused en ough postpartum distress (e.g., depression) for a woman that was at a non psychotic severity, previous studies have suggested an association between low birth weight and psychological distress that may call for emotional support (Singer et al., 1999; Kerst ing et al., 2004).

PAGE 117

117 Further addressing maternal morbidities, depending on the number of morbidities a woman had, as well as the duration and severity of those morbidities, future research should seek to determine the impact of these factors, after strati fying by pre -pregnancy BMI on postpartum distress, if not on the severity of PPD. Though there are many studies to support an association between pregnancy and/or delivery complications and PPD (Josefsson et al., 2002; Leidner, Singer, Sicherman, Francoise & Divon, 2008; Adewuya, Fatoye, Ola, Ijaodola, & Ibigbami, 2005; Campbell & Cohen, 1991), and studies that refute the association between pregnancy and/or delivery complications and PPD ( Nielsen, Videbech, Hedegaard, Dalby, & Secher 2000; Johnstone, Boy ce, Hickey, Morris -Yates, & Harris, 2001), it is recommended that the study of the pathophysiology of PPD include, in-part, pregnancy complications such as bedrest, gestational diabetes, and preeclampsia (Stowe & Nemeroff, 1995). Regarding the chi -square analyses addressing PNC utilization, it was not surprising that there was no significant difference in percentages for gender of the infant across the four PNC utilization groups, considering that there are no previous studies that have confirmed an asso ciation between these two variables. With regards to the chi -square analyses that compared women who utilized adequate plus PNC versus women who utilized other quantities of PNC, the results showed a significant difference in percentages across the four pre -pregnancy BMI. Women who had an obese pre pregnancy BMI were the only group of women with a higher percentage in the group that utilized adequate plus PNC, compared to the group that utilized other quantities of PNC. This result is worthy to note con sidering that 1) there were a large number of women who utilized adequate plus PNC, and 2) the primary population of focus in this study was women who had an obese pre -pregnancy BMI.

PAGE 118

118 Multivariate Analyses: Primary Risk Adjusted Logistic Regression Analysis Baseline Model Baseline results for the main model were somewhat consistent with conclusions made by Carter et al. (2000), and LaCoursiere et al. (2006) in that higher pre pregnancy BMI is associated with a higher likelihood for PPD symptoms; a linear trend was seen in this study in that higher pre pregnancy BMI increased odds for PPD symptoms. However, as for significance, it was only women in the highest pre -pregnancy BMI category, obese, which had significantly greater odds of having PPD symptoms rel ative to women who had a normal pre pregnancy BMI in these models; this result matched what was predicted for women who had an obese pre pregnancy BMI. Looking into the association between PNC and PPD symptoms, the baseline model demonstrated that the hig hest likelihood for PPD symptoms among levels of PNC utilization was for inadequate PNC, followed by adequate plus PNC, and intermediate PNC, relative to adequate PNC. However, after including control variables and risk adjusting for high-risk pregnancies that received adequate plus PNC due to a medical necessity, the statistical significance disappeared for all PNC utilization levels, demonstrating that there is no association between PNC utilization and PPD symptoms. These results were contrary to the one previous study that has looked at quantity of PNC and PPD: El -Kak et al. (2004). The authors demonstrated a linear relationship between PNC and PPD for high risk women, where a higher number of PNC visits were associated with fewer cases of PPD. However, even though El -Kak et al. (2004) controlled for a variety of characteristics (e.g., parity, education, area of residence, employment during pregnancy) as this study also controlled for a variety of characteristics, this study varied in the measures used fo r PNC and PPD (compared to El -Kak et al., 2004). For example, t his study used the APNCU Index (Kotelchuck, 1994) to categorize PNC utilization into inadequate,

PAGE 119

119 intermediate, adequate, and adequate plus, by considering the frequency of PNC visits, gestational age, and the timing of PNC initiation. El -Kak et al. (2004) analyzed each PNC measure separately: the initiation of the first PNC visit was categorized by trimester (1st, 2nd, 3rd), the frequency of PNC visits was categorized into three categories ( 1 4, 5 9 and 10 + visits) and gestational age was categorized into preterm or term. With regards to pregnancy risk -status, they stratified risk status into two categories (low risk versus high risk), whereas this study used two different approaches to ris k adjustment: 1) including high-risk characteristics as control variables, and 2) removing the highrisk pregnancies from the analyses. This study used a 5 tier likert scale (from the PRAMS questionnaire) to categorize PPD symptoms into yes/no, whereas El -Kak et al. (2004) categorized PPD into yes/no based on the occurrence of PPD symptoms for the women in their sample. Thus, it is postulated that the differences seen in the results for both studies can be explained by the difference in the measures used fo r both studies. Regarding previous studies that have used PNC indices to examine adequacy of PNC with birth outcomes, many studies have examined the effectiveness of PNC on low birth weight. Though some studies have shown that adequacy of PNC is not assoc iated with birth weight, several studies, many of which have used indices as measures to represent PNC (Alexander & Kotelchuck, 1996), have shown the benefits of PNC utilization on birth weight (Gortmaker, 1979; Showstack, Budetti, & Minkler, 1984; Quick, Greenlick, & Roghmann, 1981; Mustard & Roos, 1994). Though the moderating effect of PNC utilization on a postpartum outcome, PPD symptoms, was not demonstrated in this study, it is recommended that research look further into the relationship between the ef fects of PNC on postpartum preventive behaviors (Alexander & Kotelchuck, 2001), while perhaps focusing on women who are experiencing different severities of PPD (e.g., postpartum blues versus non-psychotic PPD).

PAGE 120

120 Specific Aim 1 What is the association of pre -pregnancy body mass index (BMI) with subsequent development of postpartum depression (PPD) symptoms? It was predicted in the hypothesis that women who had an obese pre -pregnancy BMI would have the highest odds of PPD symptoms, followed by women who h ad an overweight BMI, and finally women who had an underweight BMI, who would have the lowest odds for PPD symptoms (compared to women who had a pre -pregnancy BMI of normal). The results for this logistic regression showed borderline significance only among women who had an underweight pre pregnancy BMI, with this significance apparent at a 90% confidence interval only. Contrary to the hypothesis, this group of women had lower odds for PPD symptoms compared to women who had a normal pre pregnancy BMI. Thus the significant relationships demonstrated for obese pre pregnancy BMI, and all the PNC utilization groups, all of which displayed significant higher odds for PPD symptoms in the b aseline primary risk adjusted logistic regression, disappeared after addin g all the control variables. A new significant relationship for women who had an underweight pre -pregnancy BMI appeared in that they had lower odds for PPD symptoms compared to women who had a normal pre -pregnancy BMI. This result was contrary to the initi al hypothesized that women who had an underweight pre -pregnancy BMI would have greater odds for PPD symptoms. The significance that appeared for women who had an underweight pre pregnancy BMI only after adding control variables can be attributed to control ling for variables in this model that were associated with underweight pre -pregnancy BMI and positively associated with PPD symptoms. Hence, not controlling for those variables initially in the baseline model did not allow for the borderline significant as sociation between underweight pre pregnancy BMI and PPD symptoms to appear; after controlling for these variables, this significance was revealed. Future research should further study the significant control variables

PAGE 121

121 included in this model to determine wh ich variables (e.g., race, education, income, etc.) are positively associated with PPD symptoms and associated with the appearance of this borderline significance for underweight pre -pregnancy BMI upon its inclusion in the model. To explain the disappearance of the significance for obese pre -pregnancy BMI and all the PNC utilization groups, it might be that one or more of the control variables included in this model that was significantly associated with PPD symptoms was also confounding variable (associat ed with obese pre -pregnancy BMI and the PNC utilization groups in the previous model (but not controlled for). Thus, after adding the control variable(s) in the model, there was no true association of either obese pre -pregnancy BMI or any of the PNC util ization groups with PPD symptoms. Fifteen control variables added in this model were significantly associated with PPD symptoms; it is hypothesized that one or more of these variables were responsible for (but not controlled for) the initial significance s een for the four variables in the baseline model. Future research should seek to determine the control variables that were responsible for the significance seen for each of the four variables in the baseline model (as confounders) Specific Aim 2 Does PNC moderate the relationship between pre -pregnancy BMI and PPD symptoms? It was predicted in the hypothesis that within each pre -pregnancy BMI category, the likelihood a woman will experience PPD symptoms would decrease as PNC increased: Women who utilized inadequate PNC would have the highest odds for PPD symptoms, followed by women who utilized adequate plus PNC, and finally women who utilized intermediate PNC, who would have the lowest odds for PPD symptoms (compared to women who received adequate PNC). The results showed that there were no significant relationships present among the main effects (pre -pregnancy BMI and PNC utilization) or the pre -pregnancy BMI/PNC utilization interaction variables (no moderating effect of PNC). Thus, the significance dem onstrated for

PAGE 122

122 underweight pre -pregnancy BMI disappeared after including the interaction variables in the model. It appeared that interacting the quantity of PNC utilization with women who had an underweight pre -pregnancy BMI removed the previous significant association seen with underweight pre -pregnancy BMI and PPD symptoms. So, there was an overall effect seen for underweight pre -pregnancy BMI previously that was the average effect of all the interaction terms. However, after including the interaction t erms, the specific effect of each interaction term was subsequently removed from that average effect, leaving only the effect of underweight pre pregnancy BMI alone in the main effect variable; hence, although the direction of the effect remained the sam e and contrary to what was initially hypothesized (women who had an underweight pre -pregnancy BMI had lower odds instead for PPD symptoms compared to women who had a normal pre pregnancy BMI), underweight pre -pregnancy BMI was no longer significant. Multi variate Analyses: Secondary Risk Adjusted Logistic Regression (Subpopulation With Healthy Pregnancies) Baseline Model After adjusting for high risk pregnancies in the logistic regression model, the significance seen in the primary logistic regression model for women who had an obese pre pregnancy BMI disappeared. Even though a linear trend was seen in that increasing pre -pregnancy BMI increased odds for PPD symptoms, none of these effects were significant; thus, refuting conclusions made by Carter et al. (2000) and LaCoursiere et al. (2006) in that higher pre pregnancy BMI is associated with a higher likelihood for PPD symptoms. However, in comparing the sample characteristics of this analysis versus those of the studies conducted by Carter et al. (2000) and LaCoursiere et al. (2006) this study used two risk adjustment approaches: 1) controlling for the characteristics associated with high -risk pregnancies and 2)

PAGE 123

123 removing the high -risk pregnancies from the analysis. It was not evident that Carter et al. (2 000) and LaCoursiere et al. (2006) risk adjusted for the women in their samples via their exclusion criteria or in the description of their samples. Thus, in this study, comparing the primary analysis that included all pregnancies versus the secondary anal ysis that included only the healthy pregnancies showed that significance is only demonstrated for women who had an obese pre pregnancy BMI when the high risk pregnancies are also included. Since 1) obesity has been shown in a number of studies as a risk -fa ctor for complications during pregnancy and/or adverse pregnancy outcomes ( LaCoursiere et al., 2005; Saravanakumar et al., 2006; Cedergren, 2004; Rosenberg et al., 2003; Rosenberg et al., 2005; Mahmood, 2009; Baeten et al., 2001; Cnattingius et al., 1998), and 2) previous studies support an association between pregnancy and/or delivery complications and PPD (Josefsson et al., 2002; Leidner et al., 2008; Adewuya et al., 2005; Campbell & Cohen, 1991), it is hypothesized that it might be the high risk pregnanc ies, among women who had an obese pre -pregnancy BMI, that were responsible for the significant association seen in the primary analysis between obese pre -pregnancy BMI and PPD symptoms. The chi -square results for pre -pregnancy BMI suggest that many of the women who had an obese pre -pregnancy BMI experienced highrisk morbidities. For example, among the nine high risk morbidities analyzed in the chi -square analyses, women who had an obese pre pregnancy BMI had the highest percentage for seven of the nine mor bidities. What is especially worth further noting is that the significance in the baseline model disappeared for women who utilized adequate plus PNC after adding control variables in the primary analyses (models for specific aims 1 and 2). These control variables included the high risk maternal morbidities. What was surprising was that unlike the baseline model for the primary analysis, which included all pregnancies and did not control for high risk characteristics,

PAGE 124

124 the baseline model for the secondary a nalysis, which included healthy pregnancies only, did not demonstrate significance for women who utilized adequate plus. Thus, after risk adjusting by 1) controlling for high risk characteristics by including the maternal morbidities as appropriate control variables (primary analyses), and 2) modifying the design of the model by removing the high risk pregnancies from the sample (secondary analyses), the significance for women who utilized adequate plus ceased to appear as it had appeared in a model that di d not include any control variables and included women from all pregnancy risk -statuses. What may explain this result is that the high risk pregnancies were responsible for the significance that previously appeared, and that among all the women in the samp le who utilized adequate plus PNC, it is only the women who are high -risk that are the women who are subsequently at higher odds for PPD symptoms, compared to women who utilized adequate PNC that were not high -risk. Specific Aim 1 What is the associatio n of pre -pregnancy body mass index (BMI) with subsequent development of postpartum depression (PPD) symptoms? Similar to the risk adjusted primary logistic regression model, it was also predicted in the hypothesis for this model that women who had an obe se pre pregnancy BMI would have the highest odds for PPD symptoms, followed by women who had an overweight BMI, and finally women who had an underweight BMI, who would have the lowest odds for PPD symptoms (compared to women who had a pre -pregnancy BMI of normal). Results showed there was no significance among the pre -pregnancy BMI main effects. However, the main effects for PNC utilization showed that significance among women who utilized inadequate PNC disappeared. Similar to the reasoning provided for the disappearance of significance for obese pre-pregnancy BMI, and the PNC utilization main effects between the primary baseline logistic regression and the primary logistic regression that addressed the first specific aim, it might be that one or more

PAGE 125

125 of th e control variables included in this model that was significantly associated with PPD symptoms was also a confounding variable in the previous model (but not controlled for). Thus, after adding the control variable(s) in the model, there was no true assoc iation of inadequate PNC utilization on PPD symptoms. Eight control variables added in this model were significantly associated with PPD symptoms; it is hypothesized that one or more of these variables were responsible for (but not controlled for) the ini tial significance seen in the baseline model. Future research should seek to determine the control variables that were responsible for the significance seen for the four variables in the secondary baseline logistic regression model (as confounders). Speci fic Aim 2 Does PNC moderate the relationship between pre -pregnancy BMI and PPD symptoms? Similar to the risk adjusted primary logistic regression model, it was also predicted in the hypothesis for this model that within each pre -pregnancy BMI category, the likelihood a woman will experience PPD symptoms would decrease as PNC increased. That is, women who utilized inadequate PNC would have the highest odds for PPD symptoms, followed by women who utilized adequate plus PNC, and finally women who utilized i ntermediate PNC, who would have the lowest odds for PPD symptoms (compared to women who received adequate PNC). However, after removing highrisk pregnancies from the sample, the results for this model showed that there was no significance among any of the main effects (pre-pregnancy BMI or PNC utilization). Surprisingly, significance appeared for an interaction variable, but in the opposite direction to what was hypothesized: women who had an obese pre -pregnancy BMI and utilized inadequate PNC had lower od ds for PPD symptoms compared to women who had an obese pre -pregnancy BMI and utilized adequate PNC. Thus, even though the results support that a significant moderating effect for one interaction group only appeared after adding the

PAGE 126

126 interaction terms, the e ffect was in an unexpected direction, opposing what was initially hypothesized. One reason to explain why women who had an obese pre -pregnancy BMI and utilized inadequate PNC had lower odds for PPD symptoms compared to women who had an obese pre -pregnancy and utilized adequate PNC is that women in the former group were 1) healthy, 2) content with their weight and their bodies, or 3) if they feel healthy, they may not see the need to seek PNC (their self -perceived health status is excellent/healthy). Since t his model included healthy pregnancies only, it is possible that women in this group had a high self perception of their health status during pregnancy. Other reasons for utilizing inadequate PNC, previously found in one study, include denial and/or concea lment of pregnancy, or financial reasons, as concluded by Friedman, Heneghan, & Rosenthal (2009). The significance seen for inadequate PNC utilization disappeared in this model after adding the interaction terms. Similar to the reasoning provided for the disappearance of significance for women who had an underweight pre -pregnancy BMI after adding the interaction variables to the model, the significance seen initially in the model with only the main effect variables (pre -pregnancy BMI and PNC utilization) a nd the control variables was an overall, average effect of all the pre -pregnancy BMI inadequate PNC groups. However, after including the interaction terms between each pre pregnancy BMI group and inadequate PNC, the specific effect of each interaction term was subsequently removed from that average effect, leaving only the effect of inadequate PNC alone in the main effect variable; hence, inadequate PNC was no longer significant. Summary of Multivariate Results Overall, the inconsistency of results bet ween the baseline logistic regression models, the primary and secondary logistic regression models addressing the first specific aim, and the primary and secondary logistic regression models addressing the second specific aim showed

PAGE 127

127 that there is no modera ting effect of PNC on the relationship between pre -pregnancy BMI and PPD symptoms. However, the inconsistency of results with respect to some groups of women (e.g., women who had an underweight pre -pregnancy BMI) shed some light, possibly warranting furthe r exploration on these groups of women to determine reasons and additional variables responsible for the inconsistency of results. For example, this study revealed that many medical and obstetric problems are faced by women during pregnancy, and should be further examined. Perhaps further exploring an unexpected result in this study, specifically, lower odds for PPD symptoms among women who had an obese pre pregnancy BMI who sought little to minimal (inadequate) PNC (compared to women who had an obese pre-p regnancy BMI who received adequate PNC), and the extent to which these women are affected by medical and obstetric problems, would prove to be valuable. Also, another population to further investigate would be women who utilized adequate plus PNC who are high risk. This study showed that the significance for having a higher odds for PPD symptoms among women who utilized adequate plus PNC, compared to women who utilized adequate PNC, disappeared after incorporating two different approaches of risk adjustment suggesting that it is the high -risk adequate plus PNC women who are at risk for PPD symptoms. Thus, further research that looks into women who utilize adequate plus PNC may confirm this, and/or provide insight as to why the significance between women who utilized adequate plus PNC and PPD symptoms was present when control variables (including morbidity control variables as predictors of PPD) were not included, and the sample represented by that result included both highrisk women and non high-risk women. Overall, this inconsistency of results did not agree with results from previous studies (e.g., an association between pre -pregnancy BMI and PPD symptoms). Reasons to explain the inconsistency of results within this study may be attributed to the paucity of variables in the data

PAGE 128

128 to address the content of PNC with respect to nutrition and wellness (e.g., excessive weight gain, exercise, etc.) discussions, discussion of PPD during the delivery of PNC, discussion of what to possibly expect regarding weight re tention in the postpartum period (e.g., returning back to pre pregnancy weight), the type and number of PNC provider(s) since different disciplines may be trained to deliver PNC differently to some extent and having multiple providers may affect the conten t and quality of PNC delivered, and a weight gain discussion measure for all the states to include. The one question on weight gain discussion was only available for two states. However, if this question were included in the PRAMS Core Questionnaire (manda ted for all states to ask), the results might have been affected; this is hypothesized considering the results for the logistic regression estimated on a subpopulation of women who received weight gain discussion from their PNC (pp.186187) showed signific ance for three variables (p<0.05), all within the same pre pregnancy BMI group: underweight pre -pregnancy BMI, and two interaction groups for underweight pre -pregnancy BMI: inadequate and intermediate PNC. The PNC questions included in PRAMS referred to to pics such as the discussion of vitamins, if the women got PNC as early as she wanted, how satisfied she was with the staff and the waiting time. Many of the PNC variables that are suggested, and were not a part of the data, may alter the results shown in t his study, possibly leading to more consistency. Though the database was extensive and included a variety of variables, many of which were included as control variables, the lack of variables related to the discussion of delivery of PNC related to nutrition and wellness leaves much room for questions on the standardization of the content of PNC among the 51,600 women included in the study. Appendix A (pp.137155), which includes a collection of literature on the content of PNC, suggests that though many PNC providers are discussing nutrition and wellness matters in the delivery of PNC, there is still a considerable amount of variability that exists in the

PAGE 129

129 delivery of PNC with respect to nutrition and wellness. Hence, this may have had an impact on the result s of this study since there were minimal measures representing this content of PNC. This study included 34 control variables in the primary logistic regression models and 29 control variables in the secondary logistic regression models. The inclusion of a variety of control variables in the model helped to limit the amount of omitted variable bias in this study. Had some of the control variables not been included in the multivariate analyses, further significance may have been demonstrated with the variab les of interest (e.g., interaction terms), but the significance would not have been a true significance. Though the results suggested that the inclusion of many control variables removed the significance demonstrated among some of the variables of intere st in the absence of those control variables (e.g., PNC utilization), it simultaneously assured that overall, there is no true moderating effect of PNC in this sample. Limitations Since PRAMS is a self -reported survey, the likelihood of recall bias (e.g. frequency a woman experienced depressive symptoms, pre -pregnancy weight) remains among the participants. A woman may not remember her pre -pregnancy weight, especially because it was her weight more than nine months ago. If pre pregnancy weight is reported inaccurately, this could cause error in the frequency of women within a pre -pregnancy BMI category, which could in turn bias regression estimates. Second, regarding the reporting of PPD symptom severity, because of the stigmas that exist regarding mental disorders in general, a woman may have a tendency to under report her symptom severity and/or overlook the frequency and severity of her symptoms (e.g. for fear of being termed a bad m other). This may be especially so if societal pressures to be a good m other and a mothers desire to try and do everything she feels is necessary for her baby prompt her to not want to succumb to openly admitting to PPD symptoms (Epperson, 1999). If symptom severity was inaccurately reported, the prevalence of PPD

PAGE 130

130 symptoms w ould be under reported for the purpose of this study. Third, the amount of weight a woman gained during pregnancy was not controlled for, and there were no measures indicating if a woman experienced pregnancy related obesity or eating disorders that some w omen experience during pregnancy. Evidence exists that both are predictors of postpartum distress (Krummel, 2007; Franko & Spurrell, 2000). Not controlling for either of these factors can undermine the true effect that pre -pregnancy BMI may have on PPD sym ptoms. Fourth, it is expected that in a self reported survey, women may have the tendency to either under report or over report their weight. For women who had a pre -pregnancy BMI of obese or overweight, they may have a higher tendency to under report their weight, and for women who had a pre -pregnancy BMI of underweight, they may have a tendency to over report their weight; thus, the weight indicated on the survey may not be reliable for some of the women. Similar to the reasoning explained previously for recall bias, if pre -pregnancy weight is reported inaccurately, this could cause error in the frequency of women within a pre pregnancy BMI category, which could in turn bias regression estimates. Fifth, the APNCU does not include quality or content of PNC in its measure, which could impact the likelihood of PPD. For example, the satisfaction a woman feels during the delivery of her PNC may impact whether she seeks adequate PNC, especially if she feels uncomfortable in discussing certain pregnancy (psychos ocial) issues with her PNC provider and/or she feels that she is unable to reap the benefits of PNC due to the lack of the duration for each visit. Not accounting for the quality of PNC could either underestimate the effectiveness that PNC may have on redu cing the likelihood for PPD symptoms. However, the subanalyses included a model on a subpopulation of women who had weight gain discussed during their pregnancy; this is one measure regarding the content of PNC that was included, though the quality of thi s content was not included. Sixth, there may be other variables that could impact

PAGE 131

131 PPD symptomatology and were not controlled for in this study; hence, resulting in omitted variable bias. However, given the breadth of the database used for the analysis, thi s study attempted to control for the factors that would be most associated with pre -pregnancy BMI, PNC utilization, and PPD symptoms based on the existing literature. Seventh, this study did not control for depression history or depressive episodes during pregnancy, both of which are shown to increase the likelihood of PPD symptoms (Gotlib, 1989; Gotlib, Whiffen, Wallace, & Mount, 1991; Beck, 2001). Having a systematic form of screening both during pregnancy and monitoring women who show symptoms of depress ion during pregnancy may help prevent depression postpartum, especially because many women suffer depression silently during pregnancy (Marcus, Flynn, Blow, & Barry, 2003; Smith et al., 2004). Not controlling for depression history could also undermine the effect of the association between pre pregnancy BMI and PPD in that a woman may have had an even higher likelihood for PPD symptoms because of previous episodes of depression. Eighth, because it was unclear whether the woman included in this study receive d an actual diagnosis of PPD, any conclusions were made with respect to PPD symptomatology, and not PPD itself. There could be a significant difference between a womans perception of her PPD symptoms versus a professional diagnosis of PPD symptoms; a woma n may not recognize the severity of her symptoms (Epperson, 1999). The effect of this could lead to a misconception of the true prevalence of PPD in this study. Finally, and most importantly, this study was observational in nature and not causal. Importan ce of This Study/Implications Though this study determined through multiple models and two different approaches to risk adjustment (e.g., statistically risk adjusting versus truncating the population) that there is no consistent moderating effect of PNC in the association between pre -pregnancy BMI and PPD symptoms, this study provides a great deal of insight regarding PNC delivery, for researchers,

PAGE 132

132 policymakers, and clinicians, much of which involve and can be heightened with collaboration. Further research should look indepth into discussions between pregnant women and a variety of health care providers (e.g., physicians, nurses, midwives, etc.), while also looking into the relationships between women and each of these health care providers to assess inter personal communication that may provide some insight into the BMI/PNC/PPD relationship among different providers. Even though the purpose of this study sought to find that pre -pregnancy BMI is a potential marker for imminent PPD symptoms, the risk adjustme nt processes carried out in this study generated strong evidence that many of the women experienced an array of medical and obstetric problems during pregnancy, many of which were associated with PPD symptoms in this study. Thus, further research should al so look into the content of discussions between patients and providers regarding 1) the identification of problems and the risk factors that prompt providers to deliver suitable interventions, and 2) the extent to which one or more of these medical and/or obstetric problems are associated with PPD symptoms and/or other psychosocial consequences. In general, research focusing on the content of PNC in addition to the quantity of PNC delivered over the recent years could determine if the results are consistent with the existing literature and the protocols to be followed during PNC delivery. Ongoing research should address whether different PNC providers are adhering to PNC guidelines and to what extent they are adhering to them. In todays practice, if it is f ound that there is an inconsistency with the delivery of PNC content regarding weight, nutrition, and wellness, perhaps policymakers should seek to standardize the delivery of PNC through policy initiatives such as periodic accountability. Policy regarding evaluation of PNC in a variety of settings with a variety of

PAGE 133

133 providers may also provide insight into the quality and content of PNC delivered such as the strengths and weaknesses of current PNC. Regarding the discussion of PPD as a disorder that may affe ct women in the year following the birth of the child, research should look into the extent to which PNC providers are discussing this with their patients in the pregnancy period. Also, PNC providers should address postpartum depression and educate patient s during PNC, as many women may be unaware of this disorder and/or when the severity of symptoms necessitates medical attention. With regard to access to health care in the postpartum period, it may be of worth to look into the extent to which woman are un diagnosed with PPD symptoms, if not PPD itself, due to a lack of access to health care that ceases to exist after a womans six -week postpartum check up. It is suggested that preventive care resources tend to be available during pregnancy, but may not be a s readily available during other times outside of pregnancy (Kopelman et al., 2008). Since 1) PPD is underdiagnosed and/or overlooked in the United States (OHara, & Gorman, 2004; Clayton, 2004), 2) the use of screening instruments for PPD remains uncommon in the U.S. (Seehusen, Baldwin, Runkle, & Clark, 2005; Georgiopoulous, Bryan, Wollan, & Yawn, 2001), and 3) the association of PPD screening with higher rates of symptom recognition, diagnosis, and treatment as well as the feasibility and appropriateness of screening has been suggested (Georgiopoulous et al., 1999; Georgiopoulous et al., 2001), policy initiatives should seek to facilitate PPD screening periodically during the first year postpartum, working towards standardizing PPD screening. Policy initia tives should also seek to train health care providers in being cognizant of signs/symptoms of PPD, screening for PPD, and the importance of addressing and screening for PPD during pregnancy, hospitalization for delivery, and in the postpartum period (Seehusen et al., 2005). Further research should also investigate screening for women at risk for PPD

PAGE 134

134 symptoms during PNC as it has been suggested that detection rates for depressive disorders are lower in obstetric settings compared to other primary care settin gs (Smith et al., 2004). Looking at the extent to which PPD screening takes place during PNC delivery would help 1) identify those women who are at risk for PPD due to previous and/or current history of depression, and 2) subsequently provide additional me dical attention to those women who are identified as having a history of depression. The research implications suggested from the results of this study also encourage policy initiatives to help cultivate research for PPD as a plethora of questions remain r egarding PPD in general among women in the United States. This study makes many contributions. First, this study is among the first in the United States that stratified the quantity of PNC among women of different pre -pregnancy BMI groups, while looking a t the effect on the likelihood of PPD symptoms. Secondly, because women from a variety of PPD symptom severities were included in this study, the importance of both screening for PPD during PNC, the delivery and postpartum hospitalization period, and the s ix week postpartum check up are stressed, because all severities of PPD symptoms are prevalent among women in the U.S., according to the data. Thus, PNC, the hospitalization period, and the postpartum check up remain critical points to screen for PPD. Also if a postpartum woman seeks her postpartum check up visit through her PNC provider, this study can affirm the importance of PNC providers in facilitating a healthy relationship with their patients (e.g., tailoring PNC to each womans needs, supporting an environment that is conducive for a woman to openly address her concerns as a pregnant individual by encouraging open discussion, etc.). This may increase the likelihood that patients will seek care from their PNC provider in the postpartum period through a postpartum check up visit. Also, because some relationships were found among BMI and PPD after stratifying by PNC utilization, this study adds to the literature

PAGE 135

135 that stresses the importance of addressing obesity and PPD because both are rising public he alth concerns in the United States and globally. Finally, and most importantly, although the results showed that there is no association between pre -pregnancy BMI and PPD symptoms (after including suitable control variables), and PNC utilization does not generally moderate this relationship, the results uncovered that many of the women were significantly affected by a variety of medical and obstetric problems, many of which were high risk and associated with PPD symptoms. For future research, it is strongl y recommended that the possible association of these problems with PPD symptoms be further investigated. For practice, it is suggested that 1) PNC providers recognize the risk factors for and the prevalence of medical and obstetrical morbidities during pre gnancy, including, but not limited to those featured in this study, 2) identify and diagnose the morbidities that surface in their patients, 3) establish suitable interventions, and finally, 4) follow up on their patients accordingly.

PAGE 137

137 APPENDIX A SUMMARY OF LITERATURE ON PNC CONTENT

PAGE 138

138 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Covington & Rice (1997) To explore the association between patient receipt of recommended prenatal care interventions and infant birth weight Prenatal care initial interventions (weighted and measured), health promotion advice received (eat proper foods, gain weight), birth weight 3,905 African American w omen 1988 National Maternal and Infant Health Survey U.S. Public Health Service Expert Panel on the Content of Prenatal Care 1) Height/weight taken at initial PNC visit a) 98% with a VLBW infant (<1,500 grams) or a moderately low birth weight (1,500 2,49 9 grams) infant b) 97% with a normal birth weight infant (2,500 grams or greater) 2) Receiving advice on proper foods a) 92% with a VLBW infant b) 93% with a MLBW infant c) 90% with a NBW infant 3) Receiving advice on weight gain: a) 64% with a VLBW i nfant b) 65% with a MLBW infant c) 71% with a NBW infant 3) Association between women who did not receive all types of health promotion advice and birth weight: OR: 1.28 to give birth to a VLBW infant (adjusted for LBW risk)

PAGE 139

139 Authors Objectives Indepe ndent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Freda, Andersen, Damus, & Merkatz (1993) To compare the type of information given to women who sought prenatal care in public an d private clinics and the degree to which the women were satisfied with the information they were given during their prenatal care Prenatal care information received, prenatal care delivery site 159 women (80 who received care in a public setting, 79 who r eceived care in a private setting) in Bronx, New York. Questionnaires U.S. Public Health Service Expert Panel on the Content of Prenatal Care (1989) 1) Nutrition information received a) Private settings: 85% of women b) Public settings: 96% of women 2 ) Exercise during pregnancy a) Private settings: 56% of women b) Public settings: 64% of women at public settings reported receiving this information (p=0.3) 3) Patient satisfaction with information received: patients were more likely to experience satisfa ction on any PNC topic if providers initiated discussion

PAGE 140

140 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Kogan, Alexander, Kotelchuck, Nagey, & Jac k (1994) What percent of women reported receiving PHS recommended procedures and health behavior advice and how do they differ based on health insurance, site of care, and sociodemographics? USPHS recommended national guidelines for prenatal care (health behavior advice and prenatal care procedures), reports of receiving different types of prenatal care procedures and health behavior advice 9,932 women, nationally representative 1988 National Maternal and Infant Health Survey U.S. Public Health Services Expert Panel on the Content of Prenatal Care Report (1989) Percentages reported for: 1) Weight/height taken at 1st or 2nd visit: a) Maternal education: 95.3 98.2% b) Household income, 95.8 98.5% c) Marital status: 96.7 98% d) Race/ethnicity: 92.4 98.7% e) Trimester care began: 94.6 98% f) Site of care: 96 98.3% 2) Proper foods advice: a) Maternal education: 87.5 94.2% b) Maternal age: 90.8 93.4% c) Household income: 90.6 94.4% d) Marital status: 91.5 93.3% e) Race/ethnicity: 88.3 94% f) Parity: 91. 4 94.3 g) APNCU: 85.7 94.2% 3) Weight gain advice: a) Maternal education: 64.9 73 b) Maternal age: 64.9 74.7 c) Household income: 68.9 74.4 d) Marital status: 70.9 76.6 e) Race/ethnicity: 62.2 74.1 f) Parity: 64.7 78 g) APNCU: 59.7 74.5

PAGE 141

141 Authors Object ives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Kogan, Alexander, Kotelchuck, & Nagey (1994) To examine the relationship between maternal reports of health behavior advice received and initial prenatal care procedures performed during the first two visits and low birth weight Health behavior advice and initial prenatal care procedures, low birth weight (<2,500 g) 9,394 women (nationally representative) 1988 National M aternal and Infant Health Survey U.S. Public Health Services Expert Panel on the Content of Prenatal Care Report (1989) 1) Health behavior advice: the following results were reported a) 8,670 women who received advice on proper diet: 5.6% gave birth to LB W infants (p=0.06) b) 6,770 women who received advice on weight gain: 5.3% gave birth to LBW infants (p<0.01) 2) For initial prenatal care procedures, among 9,159 women who had their weight recorded, 5.6% gave birth to LBW infants (p=0.03)

PAGE 142

142 Authors Obj ectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Kotelchuck, Kogan, Alexander, & Jack (1997) To assess if site of prenatal care delivery influences the content of prenatal care given to low income women Recommended initial prenatal care procedures, recommended prenatal care advice, site of prenatal care delivery 3,405 low income women 1988 National Maternal and Infant Health Survey U.S. Public Health Service Expert Panel on the Content of Prenatal Care 1) Between 89.6 92.6% of the women reported received advice on proper foods to eat during pregnancy (not significant) and 64.5 79.9% of women reported receiving advice on weight gain during pregnancy (p<0.001) 2) Comp aring the content of PNC at different sites, between 95.1 97.9% of women were measured and weighed at their initial PNC visit (p=0.006), and between 75.4 87.4% of women had their health history taken (p<0.001) 3) Women who received their PNC at a private office were 1.52 times (CI: 1.18 1.95) more likely not to receive all the PNC procedures (e.g., blood pressure taken, height and weight measured, blood work taken, etc.) at their initial visit, and 1.76 times (CI: 1.34 2.32) more likely not to receive all the types of PNC advice (e.g., alcohol and smoking cessation, proper foods to eat, vitamins to take, weight to gain, etc.) as recommended by the U.S. Public Health Service. 4) Women who received care at sites other than a private office, public clinic, an HMO, or a hospital clinic were 1.73 times (CI: 1.04 2.83) more likely not to receive all the PNC procedures at their initial visit

PAGE 143

143 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Libbus & Sable (1991) To examine the relationship between absence of prenatal care educational content and the risk of adverse birth outcomes Six educational prenatal care content areas, 10 risk areas, term low birthweight, a nd preterm low birthweight 1,484 women from three regions in Missouri Data from a previous study on barriers to prenatal care sponsored by the Missouri Department of Health and the Missouri Perinatal Association Though no source is mentioned in choosing the areas of prenatal content, the Institute of Medicine and the U.S. Public Health Service Expert Panel on the Content of Prenatal Care were included in the reference list. 1) 20.4% of the women reported receiving diet counseling 2) 14.2% of the women poss essed a nutritional risk 3) Not receiving diet education was significantly associated with the risk of delivering a preterm low birthweight infant in the bivariate analyses, but not in the multivariate analyses. 4) Adequacy of care (care initiated w/in 1st 4 months of gestation & atleast 8 visits (term infants) or 5 visits (infants born
PAGE 144

144 Auth ors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Sable & Herman (1997) To 1) examine the relationship between the U.S. Public Health Service Expert Panel on the Content of Prenatal Care recommendations and the risk of low birth weight, and 2) to describe the type and frequency of health behavior advice given to a sample of pregnant women. Prenatal care advice, birth weight 2,205 women from the state of Misso uri National Institute of Child Health and Human Development/ Missouri M aternal and Infant Health Surve y U.S. Public Health Service Expert Panel on the Content of Prenatal Care 1) 54.8% of the women received advice on improving diet and nutrition and eatin g proper foods 2) Regarding weight gain, 62.1% of women received this advice during the course of their parental care 3) For receiving advice on diet and nutrition, 31.6% of the women were told to watch their caloric intake and to avoid excessive weight gain 4) Regarding exercise factors, 29.3% of women were told to get more exercise, and 16.7% of women were told to restrict their exercise 5) In looking at birth weight, women who did not report receiving all the seven types of advice during their PNC as recommended by the U.S. Public Health Service Expert Panel were 1.49 times (CI:1.101.88) more likely to give birth to a baby that was of very low birth weight (less than 1,500 grams) than they were to give birth to a baby of normal birth weight

PAGE 145

145 Autho rs Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Baldwin, Raine, Jenkins, Hart, & Rosenblatt (1994) To what extent do obstetric providers follow ACOG guideli nes? Components of first prenatal care visit, number of prenatal care visits, ACOG recommended laboratory tests, prenatal content monitoring (subsequent visits) Providers: 54 urban OB GYNs, 29 rural OB GYNs, 59 urban FPs, 67 rural FPs, 43 urban MWs; 2,357 female patients The Content of Obstetrical Care Study American College of Obstetricians and Gynecologists 1) Pre pregnancy weight was recorded as follows (ANOVA, p
PAGE 146

146 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Conway & Kutinova (2006) To determine the efficacy of prenatal care and the policies designed to improve access to prenatal care (Medicaid) Adequate care index (APNCU=1), prenatal are measures: advice about weight gain, advice about eating, excessive maternal hospitalization, BMI status change (became overweight after concepti on; become underweight after conception) 7,464 observations 1988 National Maternal and Infant Health Survey American College of Obstetricians and Gynecologists 1) Statistical significance was found between receiving advice about eating and excessive mater nal hospitalization (if the mothers length of stay was longer than her infant). No significant associations were found between receiving weight gain advice and excessive maternal hospitalization. 2) Receiving advice about eating was significantly associ ated with having a BMI change to underweight after birth. 3) A significant association was found for receiving advice about weight gain, and a change in BMI status after birth to underweight. 4) Inverse associations were found for receiving advice about weight gain and having a change in BMI sta tus after birth to overweight No significant associations however were found between receiving either advice about eating or weight gain during pregnancy, and a change in BMI status to overweight. 5) Fo r women with adequate PNC (APNCU index), approximately 9394% of the women received advice about eating. For the women who did not receive adequate care, 87 94% of the women received advice about eating. 6) 7075% of women who received adequate care received advice about weight gain during pregnancy. 7) 6172% of women who did not receive adequate care, received advice about weight gain during pregnancy.

PAGE 147

147 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Yu & Jackson (1995) To determine the prevalence nutrition advice received by women who sought prenatal care (self report) Nutrition advice (seven categories), maternal characteristics 9,639 mo thers who gave birth to a live infant, 4,955 mothers who did not give birth to a live infant 1988 National Maternal and Infant Health Survey Though no guidelines were noted in the literature review, the American Academy of Pediatrics, American College of O bstetrics and Gynecology, and the Institute of Medicine were included in the reference list. 1) Among mothers who gave birth to a live infant, 72.8% of White women received advice about weight gain during pregnancy, 70.1% of Black women received advice ab out weight gain during pregnancy, 63% of Asian and Pacific Islander women received advice about weight gain during pregnancy, and 73.8% of Eskimo, Aleut, and American Indian women received advice about weight gain during pregnancy 2) Regarding eating properly, this advice was received by 93% of White women, 92.7% of Black women, 90.2% of Asian and Pacific Islander women, and 89.3% of Eskimo, Aleut, and American Indian women 3) Among mothers who did not give birth to a live infant, 87.3% received advice on eating properly, and 63.8% received advice on weight gain

PAGE 148

148 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Cogswell, Scanlon, Fein, & Schieve (1999 ) To evaluate if weight gain advice given from a health care provider, a womans target gestational weight gain, and actual weight gain are in congruence with the IOM guidelines Advised weight gain, target weight gain, and actual weight gain 2,237 women P renatal questionnaire, neonatal questionnaire Institute of Medicine 1) 27% did not receive weight gain advice during PNC 2) Advice about weight gain and IOM recommendations a) 14% were advised to gain less weight than IOM guidelines b) 22% advised to gai n more weight than IOM guidelines 3) Target weight gain and IOM recommendations a) 19% had a target weight gain less than IOM guidelines b) 22% had a target weight gain higher than IOM guidelines 3) Actual weight gain and IOM recommendations a) 23% actua lly gained less than IOM guidelines b) 42% of women gained more than IOM guidelines 4) Women in the very high pre pregnancy BMI category were 15 times more likely to receive advice to gain more weight than as recommended by the IOM and 0.9 times as like ly to receive advice to gain less weight than as recommended by the IOM 5) Women in the high pre pregnancy BMI category were 31.8 times more likely to receive advice to gain more weight than as recommended by the IOM and 0.1 times as likely to receive advice to gain less weight than as recommended by the IOM 6) W omen in the low pre pregnancy BMI category were 0.5 times as likely to receive advice to gain more weight than as recommended by the IOM and 0.8 times as likely to receive advice to gain less weight than as recommended by the IOM

PAGE 149

149 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Stotland, Haas, Brawarsky, Jackson, Fuentes Afflick, & Escobar (2005) To study the relationship between pre pregnancy BMI, womens target gestational weight gain, and provider weight gain advice Pre pregnancy BMI, womens target gestational weight gain, provider weight gain advice 1,198 women in the state of Californ ia who received PNC at 1) an urban public hospital, 2) an urban community hospital, 3) a university hospital, or 4) 1 of 3 medical centers affiliated with an MCO. Weight gain advice was received from either a physician, nurse, or a nutrition counselor. Pro ject WISH (Women and Infants Starting Healthy) Institute of Medicine 1) Relationship between BMI and pregnancy target weight gain that was below IOM guidelines: a) Low BMI (OR: 0.63) b) Overweight BMI (OR: 0.05) c) Obese BMI (OR: 0.18) 2) Relation ship between BMI and pregnancy target weight gain that was above IOM guidelines: a) Overweight BMI (OR: 3.79) b) Obese BMI (OR: 2.39) 3) Association between provider advice and a womans target weight gain during her pregnancy a) Advice to gain weigh t gain below IOM guidelines and target weight gain below IOM guidelines (OR: 3.17) b) Advice to gain weight above IOM guidelines and target weight gain above IOM guidelines (OR: 3.39) c) No advice and target weight gain below IOM guidelines (OR: 1.72)

PAGE 150

150 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Petersen, Connelly, Martin, & Kupper (2001) To determine 1) the prevalence of preventive health couns eling given during prenatal care, 2) the prevalence of women who are in higher need of counseling about specific health concerns, and 3) if women who are in higher need of counseling are more likely than the women in lower need to have received counseling. Reports of preventive health counseling during prenatal care (e.g., nutrition) 24,620 women from 14 states Pregnancy Risk Assessment Monitoring System The U.S. Preventive Services Task Force Guide to Clinical Preventive Services Between 84 92% of the wome n (depending on the state) received counseling on nutrition during pregnancy

PAGE 151

151 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Levine, Wigren, Chapma n, Kerner, Bergman, & Rivlin (1993) To examine the degree to which primary care physicians in the U.S. report practicing the basic nutritional competencies in the delivery of care. Nutrition related attitude statements, nutrition related behaviors 3,416 pr imary care physicians A demographic survey, an attitude survey, and a behavior survey N/A 1) 75% or more of the physicians agreed or strongly agreed with the following: a) Continuing medical education courses should devote time to nutritional related issue s, b) it is important to have an understanding of food composition and preparation to provide reliable nutritional counseling, c) in many cases, medication could be reduced or eliminated if patients followed a recommended diet, d) nutrition will have an in creasingly important role in the prevention and treatment of disease, e) doctors should spend more time exploring dietary habits during patient evaluation 2) Statements towards which 75% or greater of physicians in the sample disagreed or strongly disagre ed, this included: a) Most doctors are very knowledgeable about nutrition, b) physicians are well prepared to provide nutritional counseling, c) nutrition is important only in certain medical specialties, d) dietary counseling is a waste of time because pe ople dont change their

PAGE 152

152 habits anyway, and e) nutrition education is not the responsibility of the physician. 3) Regarding specific nutritional advice on what physicians usually or always practice, the authors found that a) 61% of the physicians reported advising or teaching their patients about the rationale for dietary modifications, b) 60% of the physicians reported advising or teaching their patients about achievements and maintenance of health habits such as exercise, c) 56% of physicians reported adv ising or teaching their patients about the achievements of desirable weight, d) 55% of physicians reported prescribing to their patients dietary modifications such as sugar or salt intake reductions, weight reduction, e) 54% of physicians reported monitori ng their patients nutrition status and progress in response to treatments recommended, and f) 52% of physicians reported prescribing to their patients exercise depending on their age, physical condition(s), and their health status.

PAGE 153

153 Authors Objectives Independent/ dependent variables Sample characteristics Data Type of prenatal care guidelines Results (statistical tests, p values) Splett, Reinhardt, & Fleming (1994) To 1) identify physicians need and expectations regarding quality nutrition services rendered in prenatal care 2) to rank the characteristics of the services rendered by importance in making nutrition referral decisions, and 3) identify nutrition services physicians would most likely add to their delivery of care. Nutrition care services, availability of services: currently available/likely to add/unlikely to add; physicians rating of nutrition services 130 prenatal care OB GYN physicians Quality Service Management Model Though no guidelines were noted in the literature review, the Insti tute of Medicine was included on the reference list 1) 63.8% of the physicians had an ongoing monitoring of patient weight gain and dietary patterns, while 16.9% of physicians reported that they would likely start practicing the service, and 13.1% of the p hysicians reported that they would unlikely to start practicing the service 2) Regarding initial PNC screening of women to detect their nutritional risk, 61.5% of physicians reported that they provided that service, 24.6% of physicians reported that they would likely start practicing the service, and 10.8% of the physicians reported that they would unlikely start practicing the service 3) Regarding follow up of women who were identified as having nutrition problems during their pregnancy, 42.3% reported t hat they currently provided the service, 43.8% reported that they would likely start practicing the service, and 5.4% of the physicians reported that they would unlikely start practicing the service 4) Regarding nutritional consultation for each woman dur ing her PNC, 40% of the physicians reported providing the

PAGE 154

154 service, 43.1% of the physicians reported that they would likely start practicing the service, and 11.5% of the physicians reported that they would unlikely start practicing the service 5) Nutritio nal assessment and planning of care for women with a high risk pregnancy, was currently conducted by 36.2% of physicians, while 53.8% reported that they would likely start practicing the service, and 6.2% of physicians reported that they would unlikely s tart practicing the service 6) 26.2% of physicians reported that they currently practice postpartum weight counseling, while 61.5% of physicians reported that they would likely start practicing the service, and 6.2% of physicians reported that they would unlikely start practicing the service. 7) 32% of the women engaged in discussions with their physician for seeking advice on nutrition problems, while 12% of the women engaged in discussions with a registered dietitian, 8% sought information from PNC clas ses, 8% sought information from brochures and pamphlets given in the office where the PNC was provided, and 3% sought assistance through WIC

PAGE 155

155 APPENDIX B MULTIVARIATE SUB -ANALYSES In addition to the main logit models (PPD), sub analyses were conducted to further test the moderating effect of PNC in the association between pre -pregnancy BMI and PPD symptoms. One of the subanalysis logit models was estimated by specifying the dependent variable (PPD symptoms) differently (a sensitivity analysis via an ordinal logistic regression model), while another model was estimated with a different dependent variable (adequate plus PNC), and five of the subanalysis logit models comprised of five different subpopulations: one consisting of women who utilized WIC service s, and each of the four remaining models comprising a different subpopulation income group. Adequate Plus Unlike the secondary analysis that risk adjusted for high -risk pregnancies by removing observations that met any of the high risk criteria defined fo r this study, a logistic regression was estimated holding adequate plus as the dependent variable, to determine high risk characteristics (denoted by significance). Seventeen morbidities, among other characteristics (e.g., demographics) were included in th is model to determine factors that could label a pregnancy as high risk. These morbidities included: 1 ) Pre -pregnancy diabetes 2 ) Gestational diabetes 3 ) Vaginal bleeding 4 ) Kidney/bladder infection 5 ) Nausea 6 ) Hospitalization 7 ) Preterm labor 8 ) Premature rupture of membrane s (PROM) 9 ) Placenta abrutio or placenta previa 10) Incompetent cervix (cervix closed) 11) High blood pressure 12) Blood transfusion 13) Car crash injury 14) Bed rest

PAGE 156

156 15) Labor/delivery complications (in general) 16) Pregnancy abnormalities (in general) 17) Medical risk factors (in general) Sensitivity Analysis: Ordinal Logistic Regression To further test the moderating effect of PNC, a sensitivity analysis employed an ordinal logistic regression and kept PPD symptom response as seven categories (scores of 0 6), prior to grouping the scores into yes or no to create a dichotomous variable for the primary logit model. This model was used to estimate PPD symptoms as an ordinal, categorical variable to determine whether the results are sensitive to the way in which the dependent variable is specified. Reference groups remained the same as the primary logistic regression model. With the dependent variable specified into seven categories, it was also predicted that the highest likelihood for increasing in PPD severity was for women who had a pr e -pregnancy BMI of obese and received inadequate care. As shown in the model specification below, each PPD response (assigned a score) had its own j value (while the coefficients of the independent variables for each PPD response remained the same): l ogit = log ) ( 1 ( ) ( j Y P j Y P = j+ 1(obese/adequate plus care) + 2(overweight/adequate plus care) + 3(underweight adequate plus care) + 4(obese/intermediate care) + 5(overweight/intermediate care) + 6(underweight/intermediate care) + 7(obese/in adequate care) + 8(overweight/inadequate care) + 9(underweight/inadequate care) + k Xk(control variables) + (B 1) (Where j represents the PPD score assigned to each woman) Women, Infants, and Children (WIC) A logistic regression was estimated only o n women who received WIC services during their pregnancy. Eligibility for WIC during pregnancy is based on socioeconomic status. This program is a food and nutrition service that primarily targets low -income women, infants, and

PAGE 157

157 children, who may be nutriti onally at risk, in order to provide health care referrals, information on nutrition, and nutritious food to these women. Since this program may be a potential avenue for women to receive advice and guidance on nutrition, weight, and fitness during pregnanc y, a model inclusive of this sub-population of women was estimated to determine if a moderating effect of PNC could be detected among women who also received WIC services in addition to the quantity of PNC received. Income To determine if there is a PNC m oderating effect after stratifying by income category, and to see if PNC is more effective for certain income groups versus others, a logistic regression was estimated for each income sub -population. Each logit model mirrored the main PPD analysis, but inc luded only women from each income group. Income groups were stratified as follows: 1 ) Less than $10,000 2 ) $10,000 to $24,999 3 ) $25,000$49,999 4 ) $50,000 or more Weight Gain Discussion Since discussion of weight gain during PNC, in addition to nutrition and wellne ss during pregnancy, is the premise of the theory posed for this study, a logistic regression model was estimated with a sub -population of women who answered yes to the following question: Did your health care professional discuss how much weight to gai n? The weight gain discussion variable was selected from the PRAMS Standard Questionnaire. This model included the same variables as the main PPD analysis (PNC, BMI, and the control variables). However, because this additional control variable is optional for inclusion in state surveys, the sample size only contained women from Utah and Vermont; thus, reducing the sample size and limiting the external validity.

PAGE 158

158 Results of Sub Analyses Sub analyses were estimated with six logit models, each using a different sub population of women from the sample used originally for the main models (PPD). Table B 1 presents the chi -square results to describe the model of women who received adequate plus PNC versus women who received other quantities of PNC. Table B 2 presen ts the ttest results with maternal age and adequate plus PNC. Table B 3 presents the results from the logit model for adequate plus PNC, including the twelve maternal morbidities that were significant (p<0.1). Table B 4 presents the results of the ordinal logistic regression model inclusive of the main effects and control variables. Similar to the primary logistic regression model that addressed the first specific aim, this model also showed that women who had an underweight pre -pregnancy BMI had lower odd s for PPD symptoms compared to women who had a normal pre -pregnancy BMI. Thus, women who had a normal pre -pregnancy BMI had 11% greater odds for PPD symptoms compared to women who had an underweight pre -pregnancy BMI (OR=0.90, p<0.05). However, in continui ng to comparing this model with the primary logistic regression model addressing the first specific aim, the significance for women who had an underweight pre -pregnancy BMI increased: p<0.10 versus p<0.05, respectively. In addition, the odds of PPD symptom s for women with a normal pre -pregnancy BMI slightly decreased. The first logistic regression (logit) model held adequate plus PNC as the dependent variable in order to determine the maternal morbidities predictive of high risk adequate plus PNC. The next logit model (Table B 5) included a sub-population of women who received services during pregnancy from Women, Infants, and Children (WIC). Results showed that for the main effects in this model, only overweight pre -pregnancy BMI had a significant associat ion with PPD symptoms. However, the odds for PPD symptoms among women who had a pre pregnancy BMI of normal were greater by 64% compared to women who had a pre -pregnancy

PAGE 159

159 BMI of overweight (OR=0.61, p=0.005). This model also demonstrated a moderating effect of PNC for two pre -pregnancy BMI groups of women who received inadequate PNC. Women who had an obese pre -pregnancy BMI and received adequate PNC had 59% greater odds for PPD symptoms compared to women who had an obese pre pregnancy BMI and received inadeq uate PNC (OR=0.63, p=0095). However, the reverse moderating effect of PNC was seen for women who had an overweight pre -pregnancy BMI and received inadequate PNC in that they had 92% greater odds for PPD symptoms compared to women who had an overweight pre -pregnancy BMI and received adequate PNC (OR=1.92, p=0.048). For women who received WIC services, it is suggested that among women who had an obese pre -pregnancy BMI, compared to women who received adequate PNC, those who received inadequate PNC were perhap s healthy and happy with their bodies and/or pregnancy and did not see a need to seek PNC. Also, women who received adequate PNC may have been the women who were more anxious and worried about their pregnancy. However, for women who had an overweight pre -p regnancy BMI, the PNC may have been beneficial along with the WIC services; thus, resulting in the higher odds for PPD symptoms among women who received inadequate PNC and the lower odds for PPD symptoms among women who received adequate PNC. Looking at a sub -population of women who received WIC services in this model appeared to show a moderating effect of PNC to some extent. Further research should address whether receiving PNC and WIC services simultaneously during pregnancy (in which weight, nutrition, and wellness are addressed in both) has a beneficial effect on reducing the likelihood for PPD symptoms. The next four logit models were estimated using a sub-population of women from each income category to see if PNC was more effective for one income category versus another. These models were estimated in efforts to demonstrate a moderating effect of PNC after

PAGE 160

160 stratifying the sample by income category. The first logit model, estimated on women with an income of less than $10,000 (Table B 6), showed a m oderating effect of PNC for women from two pre -pregnancy BMI groups of women. Women who had an obese pre -pregnancy and received adequate PNC had 96% greater odds for PPD symptoms compared to women who had an obese pre pregnancy BMI and received inadequate PNC (OR=0.51, p=0.07). On the other hand, women who had an overweight pre pregnancy BMI and received inadequate PNC had 2.32 times greater odds for PPD symptoms compared to women who had an overweight pre pregnancy BMI and received adequate PNC (OR=2.32, p =0.06). It is interesting to note that these results remain consistent with the results shown for the logit model estimated on women who received WIC services during their pregnancy, because WIC primarily targets low -income women. These results also suppor t further research to focus on women who receive WIC services and PNC services simultaneously, to see if there is additional benefit, compared to women who receive either one or the other. The next logit model, estimated using a sub -population of women with an income between $10,000 and $24,999 (Table B 7) did not show a consistent moderating effect of PNC as shown in the two previous logit models (with sub -populations of WIC and women with an income of less than $10,000). In fact, there was no significa nce for any of the main effects and PPD symptoms, nor the interaction effects and PPD symptoms. Since this model did not show a similar moderating effect as shown in the logit models estimated with WIC women and women with an income of less than $10,000, i t is suggested that perhaps, when considering women who receive WIC services during pregnancy but get minimal to no PNC (inadequate PNC), the moderating effect occurs more readily for women with a very low income as opposed to a low income, as stratified i n this study. It is surprising though that among women who received WIC

PAGE 161

161 services during pregnancy (Table B 5), women with an income less than $10,000 had a higher odds for PPD (OR=1.96, p=0.002) compared to women who had an income of $50,000 or greater, wh ile women with an income between $10,000 and $24,000 (OR=1.48, p=0.07) compared to women who had an income of $50,000 or greater. Next, in the logit model inclusive of women with an incom e between $25,000 and $49,000 ( T able B 8), only inadequate PNC was s ignificant with PPD symptoms in that women from this PNC utilization category had 72% greater odds compared to women who received adequate PNC (OR=1.72, p=0.046). Though a moderating effect of PNC was not seen after interacting each pre -pregnancy BMI group with each PNC utilization category, the significance shown for inadequate PNC among this group of women with an income between $25,000 and $49,999 suggests further studying the relationship between weight and PPD symptoms amongst women from this income gr oup who receive no to minimal PNC. Understanding the reasons that play a role in explaining why some women from this income group seek no to minimal PNC may help explain these results, especially because it appears that PNC has a beneficial effect on women from this income group who received adequate PNC. Finally, a logit model estimated for a sub -population of women with an incom e of $50,000 or greater ( T able B 9) did not show significance between any of the main effects and PPD symptoms, but a moderating effect was seen for one group of women. Women who had a pre pregnancy BMI of underweight and received intermediate PNC had a 2.4 times greater odds for PPD symptoms compared to women who had a pre -pregnancy BMI of underweight and received adequate PNC (OR =2.40, p=0.06). It is interesting to note that despite lack of significance, women from this pre pregnancy BMI category who received inadequate PNC had 28% greater odds for PPD symptoms compared to women who had an underweight pre -pregnancy BMI who

PAGE 162

162 received adequate PNC. These odds are lower than the women who received the next level of PNC quantity (as the reverse was predicted). For women from this income group and from this pre pregnancy BMI category, it may be of worth to ascertain psychosocial reasons to explain why it is that women in between those receive no to minimal PNC, and those who receive adequate PNC, have more than double the odds for PPD symptoms than those who receive no to minimal PNC. In comparing the overall results from the logit models estimated on a sub population of women from each income group, it seems that a moderating effect of PNC was the most worthy to note in the model that included women with an income of less than $10,000 (very low income). Table B 10 presents the results fr om the logistic regression that looked a subpopulation of women who had weight gain discussed during their PNC. The variables PNC paid by military and PNC paid by Native American health services were not included in this model due to collinearity. For example, women who answered yes to receiving payment for PNC from either of these organizations also answered yes to having weight gain discussed by their PNC provider, hence, resulting in collinearity where the variables (not the women) were automatic ally removed from the model by Stata. Looking at the main effects, among women who had weight gain discussed by their PNC provider, women who had a normal pre pregnancy BMI had double the odds of PPD symptoms compared to women who had an underweight pre -pr egnancy BMI (OR=0.48, p=0.02). However, when looking at the interaction effects among women who had an underweight pre -pregnancy BMI, those who received inadequate PNC had about four times greater odds for PPD symptoms (OR=4.10, p=0.01), and women who rece ived intermediate PNC had about 12.8 times greater odds for PPD symptoms (OR=12.79, p=0.002). To explain this inconsistency of results, table 45 shows that among women who had weight gain discussed

PAGE 163

163 during their PNC, 795 of women who had an underweight pre pregnancy BMI had the discussion versus 2,713 women with a normal pre -pregnancy BMI who had the discussion. Thus, when comparing pre -pregnancy BMI groups, only, even with the higher number of women with a normal pre -pregnancy BMI who had weight gain discu ssed, it was beneficial more -so for women who had an underweight pre -pregnancy BMI as these women had lower odds for PPD symptoms. However, when looking only at women who had an underweight pre -pregnancy BMI, there were some women in which weight gain disc ussion was not beneficial; thus, these women at a higher odds for PPD symptoms compared to women in that same pre pregnancy BMI group who received adequate PNC. It is suggested that perhaps there that women who receive inadequate and intermediate levels of PNC may be more sensitive about their weight, which may prompt them not to seek as much PNC as women who had an underweight pre -pregnancy BMI and received adequate PNC. The latter may also be more proactive about gaining the right amount of weight for the ir health of their baby. Hence, a selection effect may be the reason to explain these results. Further research should seek to determine the feelings and attitudes of women who are underweight towards receiving weight gain discussion from their PNC provide r(s).

PAGE 164

164 Table B 1. Chi -square analyses comparing 40 characteristics among women who utilized adequate plus PNC versus women who utilized other quantities of PNC Categorical control variable No for adequate plus PNC (Frequency, %) n (no adequate plus PN C) Yes for adequate plus PNC (Frequency, %) n (yes for adequate plus PNC) N P value Main variable : Pre pregnancy body mass index (BMI) Underweight Normal Overweight Obese 3,919 (13.9%) 14,882 (52.7%) 3,615 (12.8%) 5,827 (20.6%) 28,243 2,045 ( 13.6%) 7,384 (49.2%) 1,880 (12.5%) 3,686 (24.6%) 14,995 43,238 <0.0001* Demographic control variables Maternal race: White No Yes 11,288 (37.5%) 18,811 (62.5%) 30,099 4,912 (31.3%) 10,789 (68.7%) 15,701 45,800 <0.0001* Maternal race: Black No Yes 25,670 (85.3%) 4,429 (14.7%) 30,099 13,093 (83.4%) 2,608 (16.6%) 15,701 45,800 <0.0001* Maternal race: Other No Yes 23,240 (77.2%) 6,859 (22.8%) 30,099 13,397 (85.3%) 2,304 (14.7%) 15,701 45,800 <0.0001* Hispanic Not Hispanic Hispanic 23,916 (79.9%) 6,010 (20.1%) 29,926 13,322 (85.4%) 2,280 (14.6%) 15,602 45,528 <0.0001*

PAGE 165

165 Table B 1. Continued Categorical control variable No for adequate plus PNC (Frequency, %) n (no adequate plus PNC) Yes for adequate plus PNC (Frequency, %) n (yes for ad equate plus PNC) N P value Maternal education 0 8 years 9 11 years 12 years 1315 years 16+ years 1,538 (4.98%) 4,710 (15.3%) 9,425 (30.5%) 7,068 (22.9%) 8,132 (26.3%) 30,873 530 (3.19%) 2,013 (12.2%) 5,048 (30.5%) 4,046 (24.4%) 4,928 (29.7%) 16,565 47, 438 <0.0001* Income (12 months prior) Less than $10,000 $10,000 to $24,999 $25,000 to $49,999 $50,000 or more 6,663 (21.3%) 8,987 (28.7%) 7,127 (22.7%) 8,578 (27.4%) 31,355 2,944 (17.6%) 4,495 (26.8%) 3,887 (23.2%) 5,422 (32.4%) 16,748 48,103 <0.0001* Marital status Married Other 19,570 (62.4%) 11,775 (37.6%) 31,345 11,142 (66.6%) 5,588 (33.4%) 16,730 48,075 <0.0001* Insurance control variables PNC paid by income No Yes 25,128 (80.1%) 6,227 (19.9%) 31,355 13,294 (79.4%) 3,454 (20.6%) 16, 748 48,103 0.047* PNC paid by insurance/HMO No Yes 16,146 (51.5%) 15,209 (48.5%) 31,355 7,418 (44.3%) 9,330 (55.7%) 16,748 48,103 <0.0001*

PAGE 166

166 Table B 1. Continued Categorical control variable No for adequate plus PNC (Frequency, %) n (no adequate plus PNC) Yes for adequate plus PNC (Frequency, %) n (yes for adequate plus PNC) N P value PNC paid by Medicaid No Yes 18,434 (58.8%) 12,921 (41.2%) 31,355 10,143 (60.6%) 6,605 (39.4%) 16,748 48,103 <0.0001* PNC paid by military No Yes 30,520 (97.3%) 835 (2.67%) 31,355 16,526 (98.7%) 222 (1.32%) 16,748 48,103 <0.0001* PNC paid by Native American Health Services No Yes 30,978 (98.8%) 377 (1.20%) 31,355 16,649 (99.4%) 99 (0.59%) 16,748 48,103 <0.0001* Pregnancy and delivery control variables Birthweight <1,500 g 1,500 g to 2,499 g 2,500+ g 864 (27.6%) 4,555 (14.5%) 25,928 (82.7%) 31,347 1,622 (9.68%) 5,664 (33.8%) 9,462 (56.5%) 16,748 48,095 <0.0001* Smoking during pregnancy No Yes 27,818 (89.3%) 3,344 (10.7%) 31,162 14,793 (88.8%) 1,875 (11.2%) 16,668 47,830 0.08** Vaginal delivery No Yes 8,581(27.4%) 22,759 (72.6%) 31,340 6,542 (39.1%) 10,195 (60.9%) 16,737 48,077 0.025*

PAGE 167

167 Table B 1. Continued Categorical control variable No for adequate plus PNC (Frequency, %) n (no adequate plus PNC) Yes for adequate plus PNC (Frequency, %) n (yes for adequate plus PNC) N P value Gender of infant Male Female 15,903 (50.7%) 15,452 (49.6%) 31,355 8,437 (50.4%) 8,310 (49.6%) 16,747 48,102 0.48 Infant in the intensive care unit (ICU) No Yes 26,402 (85.6%) 4,445 (14.4%) 30,847 11,308 (68.6%) 5,185 (31.4%) 16,493 47,340 <0.0001* Pregnancy intention No Yes 16,350 (52.9%) 14,573 (47.1%) 30,923 7,822 (47.3%) 8,701 (52.7%) 16,523 47,446 <0.0001* Breastfed (ever) No Yes 5,551 (18.1%) 25 ,048 (81.9%) 30,599 3,256 (19.8%) 13,174 (80.2%) 16,430 47,029 <0.0001* Alcohol consumption in the last three months of pregnancy No Yes 28,622 (93.1%) 2,113 (6.87%) 30,735 15,486 (94.2%) 960 (5.84%) 16,446 47,181 0.034* Women, Infants, and Chil dren during pregnancy No Yes 16,403 (53.1%) 14,504 (46.9%) 30,907 9,066 (54.8%) 7,467 (45.2%) 16,533 47,440 <0.0001*

PAGE 168

168 Table B 1. Continued Categorical control variable No for adequate plus PNC (Frequency, %) n (no adequate plus PNC) Yes for adequa te plus PNC (Frequency, %) n (yes for adequate plus PNC) N P value Weight gain talk during pregnancy No Yes 1,288 (23.4%) 4,205 (76.6%) 5,493 691 (20.5%) 2,675 (79.5%) 3,366 8,859 0.0001* High risk maternal morbidity control variables Diabete s before pregnancy No Yes 30,859 (98.4%) 496 (1.58%) 31,355 16,245 (97.0%) 503 (3.00%) 16,748 48,103 <0.0001* Incompetent cervix No Yes 30,960 (98.7%) 395 (1.26%) 31,355 16,271 (97.2%) 477 (2.85%) 16,748 48,103 <0.0001* Preterm labor No Yes 24, 975 (79.7%) 6,380 (20.3%) 31,355 10,528 (62.9%) 6,220 (37.1%) 16,748 48,103 <0.0001* Placenta previa or placenta abruptio No Yes 29,812 (95.1%) 1,543 (4.92%) 31,355 15,111 (90.2%) 1,637 (9.77%) 16,748 48,103 <0.0001* Bedrest No Yes 26,137 (83.4%) 5 ,218 (16.6%) 31,355 11,295 (67.4%) 5,453 (32.6%) 16,748 48,103 <0.0001* Car crash injury No Yes 30,847 (98.4%) 508 (1.62%) 31,355 16,431 (98.1%) 317 (1.89%) 16,748 48,103 0.03*

PAGE 169

169 Table B 1. Continued Categorical control variable No for adequate plus PN C (Frequency, %) n (no adequate plus PNC) Yes for adequate plus PNC (Frequency, %) n (yes for adequate plus PNC) N P value Blood transfusion No Yes 31,022 (98.9%) 333 (1.06%) 31,355 16,466 (98.3%) 282 (1.68%) 16,748 48,103 <0.0001* Medical risk facto rs No Yes 21,317 (68.0%) 10,038 (32.0%) 31,355 9,553 (57.0%) 7,195 (43.0%) 16,748 48,103 <0.0001* Hospitalized during pregnancy No Yes 27,247 (86.9%) 4,108 (13.1%) 31,355 11,641 (69.5%) 5,107 (30.5%) 16,748 48,103 <0.0001* Non high risk maternal morbidity control variables Gestational diabetes No Yes 28,955 (92.3%) 2,400 (7.65%) 31,355 14,739 (88.0%) 2,009 (12.0%) 16,748 48,103 <0.0001* Kidney/bladder infection No Yes 25,968 (82.8%) 5,387 (17.2%) 31,355 13,405 (80.0%) 3,343 (20. 0%) 16,748 48,103 <0.0001* Nausea No Yes 22,725 (72.5%) 8,630 (27.5%) 31,355 11,378 (67.9%) 5,370 (32.1%) 16,748 48,103 <0.0001*

PAGE 170

170 Table B 1. Continued Categorical control variable No for adequate plus PNC (Frequency, %) n (no adequate plus PNC) Yes fo r adequate plus PNC (Frequency, %) n (yes for adequate plus PNC) N P value High blood pressure No Yes 27,736 (88.5%) 3,619 (11.5%) 31,355 13,027 (77.8%) 3,721 (22.2%) 16,748 48,103 <0.0001* Vaginal bleeding No Yes 26,847 (85.6%) 4,508 (14.4%) 31,35 5 12,944 (77.3%) 3,804 (22.7%) 16,748 48,103 <0.0001* Premature rupture of membrane (PROM) No Yes 29,277 (93.4%) 2,078 (6.62%) 31,355 13,949 (83.3%) 2,799 (16.7%) 16,748 48,103 <0.0001* Labor abnormalities No Yes 25,378 (80.9%) 5,977 (19.1%) 31, 355 13,543 (80.9%) 3,205 (19.1%) 16,748 48,103 0.84 Labor/delivery complications No Yes 20,501 (65.4%) 10,854 (34.6%) 31,355 10,298 (61.5%) 6,450 (38.5%) 16,748 48,103 <0.0001* The dependent variable for this table was adequate plus PNC, while the main independent variable was pre pregnancy body mass index (BMI). The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval (CI) and a double as terisk corresponds to a 90% confidence interval (CI).

PAGE 171

171 Table B 2. Maternal age (continuous variable) and adequate plus PNC t -test results Group (Adequate plus PNC) Observations Mean Standard error Standard deviation 95% Confidence interval (Lower, Uppe r) No 31353 27.22 .0347 6.14 (27.15, 27.28) Yes 16746 27.96 .0481 6.22 (27.86, 28.05) Combined 48099 27.48 .0282 6.18 (27.42, 27.53) Difference -------.7374 .0591 .8531 .6215 The dependent variable for this table was adequate plus PNC, while the main independent variable was maternal age. The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005.

PAGE 172

172 Table B 3. Logistic regression for adequate plus PNC to determine significant predictors of a dequate plus PNC Dependent variable: Adequate plus PNC Odds ratio P value 95% Confidence interval (Lower, Upper) Main effect independent variable: Pre pregnancy BMI Normal (reference) 1.00 -----------------------Obese 1.09 0.07** (0.994, 1.193) Overweight 0.95 0.33 (0.848, 1.056) Underweight 1.12 0.04* (1.006, 1.252) Demographic control variables Maternal race: White (reference) 1.00 -----------------------Maternal race: Black 0.89 0.049* (0.801, 0.999) Maternal race: Other 0.71 <0 .0001* (0.638, 0.798) Hispanic ethnicity 0.80 <0.0001* (0.720, 0.898) Maternal education 1.05 0.02* (1.007, 1.096) Maternal age 1.01 0.02* (1.001, 1.016) Higher income: $50,000 or more (reference) 1.00 -----------------------Very low income: Less than $10,000 0.74 <0.0001* (0.635, 0.850) Low income: $10,000 $24,999 0.80 <0.0001* (0.705, 0.901) Moderate income: $25,000 $49,999 0.86 0.004* (0.781, 0.953) Marital status 0.92 0.07** (0.830, 1.008) Pregnancy and delivery control variables Birthw eight 0.52 <0.0001* (0.486, 0.565) Smoking during pregnancy 1.02 0.76 (0.989, 1.160) Vaginal delivery 0.85 <0.0001* (0.782, 0.915) Alcohol during pregnancy 0.76 <0.0001* (0.662, 0.875) Women, Infants, and Children during pregnancy 1.21 <0.0001* (1.100, 1.327) Maternal morbidity control variables Diabetes before pregnancy 1.75 <0.0001* (1.345, 2.284) Gestational diabetes 1.35 <0.0001* (1.190, 1.522) Vaginal bleeding 1.13 0.01* (1.030, 1.248) Kidney/bladder infection 1.12 0.02* (1.017, 1.226) Ce rvix sewn shut (incompetent) 1.35 0.048* (1.002, 1.821) High blood pressure during pregnancy 1.36 <0.0001* (1.225, 1.509) Nausea 1.01 0.89 (0.928, 1.090) Preterm labor 1.46 <0.0001* (1.328, 1.595) Premature rupture of membrane (PROM) 1.49 <0.0001* (1.2 72, 1.736) Placenta previa or placenta abruptio 1.16 0.065** (0.991, 1.351) Bedrest 1.47 <0.0001* (1.337, 1.617) Car crash injury 0.81 0.14 (0.609, 1.072) Blood transfusion 0.79 0.18 (0.556, 1.119) Medical risk factors 1.17 <0.0001* (1.081, 1.259) L abor abnormalities 1.07 0.15 (0.977, 1.169) Labor/delivery complications 0.96 0.35 (0.893, 1.041) Hospitalized during pregnancy 1.26 <0.0001* (1.127, 1.411) The dependent variable for this table was adequate plus PNC, while the main independent variable was pre pregnancy body mass index (BMI). The population for this table included all pregnancies and the years of PRAMS data collection were for 2004 & 2005. An asterisk corresponds to a 95% confidence interval (CI) and a double asterisk corresponds to a 90% confidence interval (CI).

PAGE 173

173 Table B 4. Risk adjusted ordinal logistic regression sensitivity analysis with the main effect independent variables, interaction effect variables, and control variables Dependent variable: Postpartum depressive (PPD) sympto ms Odds ratio P value 95% Confidence interval (Lower, Upper) Main effect independent variable: Pre pregnancy BMI Normal BMI (reference) 1.00 -----------------------Underweight 0.94 0.43 (0.820, 1.087) Overweight 0.96 0.60 (0.830, 1.114) Obe se 1.06 0.36 (0.937, 1.198) Main effect independent variable: PNC utilization Adequate (reference) 1.00 -----------------------Inadequate 0.99 0.93 (0.829, 1.188) Intermediate 1.08 0.26 (0.946, 1.229) Adequate plus 1.04 0.52 (0.930, 1.156) Interaction effect variables: Pre pregnancy BMI/PNC utilization Obese BMI/Adequate PNC (reference) 1.00 -----------------------Obese BMI/Inadequate PNC 0.87 0.38 (0.635, 1.188) Obese BMI/Intermediate PNC 0.84 0.20 (0.645, 1.096 ) Obese BMI/Adequate plus PNC 1.01 0.95 (0.833, 1.216) Overweight BMI/Adequate PNC (reference) 1.00 -----------------------Overweight BMI/Inadequate PNC 1.15 0.53 (0.747, 1.764) Overweight BMI/Intermediate PNC 1.10 0.50 (0.836, 1.441) Overweight BMI/Adequate plus PNC 0.998 0.99 (0.782, 1.273) Underweight BMI/Adequate PNC (reference) 1.00 -----------------------Underweight BMI/Inadequate PNC 1.05 0.79 (0.720, 1.540) Underweight