<%BANNER%>

An Examination of Strategic Group Membership and Technology in the Nursing Home Industry

University of Florida Institutional Repository
Permanent Link: http://ufdc.ufl.edu/UFE0022511/00001

Material Information

Title: An Examination of Strategic Group Membership and Technology in the Nursing Home Industry
Physical Description: 1 online resource (130 p.)
Language: english
Creator: Laberge, Alexandre
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: cost, financial, nursing, porter, quality, strategic, technology
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: Purpose: The purpose of the study was to examine if strategic groups in the nursing home industry can be determined by examining how facilities apply their technology and how facilities commit their resources. The study also determined if these groups can be defined using Porter's generic strategies and whether the different groups have varying strategic-performance relationships. Methodology: The study began by performing a factor analysis using scope (technology) and resource commitment variables from the Minimum Data Set (MDS) and the Online Survey Certification and Reporting (OSCAR) system. The newly defined variables were used in a cluster analysis to identify the different clusters. The clusters were classified using Porter?s generic strategy model. Once the strategic groups were identified, the study analyzed the strategic-performance relationships using negative binomial, ordinary least squares regression (OLS), generalized linear model (GLM) with gamma distribution and ordered logit analytical tools. Quality dependent variables included quality of care deficiencies, long term care (LTC) pressure sore prevalence and post acute pressure ulcer incidence, LTC activities of daily living (ADL) decline, post acute walking improvement, LTC bowel and bladder decline and LTC restraint. Financial measures included cost per resident day, operating margin and total margin. Results: The study identified four strategic groups (differentiator, cost leader, rehab focus, and specialty care focus) and a lack of strategy group. Differentiators, rehab focus and specialty care focus groups had less quality of care deficiencies than the lack of strategy group and cost leaders. Differentiators generally had better long term care quality outcomes compared to other strategic groups and the lack of strategy group. Cost leaders, differentiators and lack of strategy groups had lower cost than the rehab focus and specialty care focus groups. Cost leaders, differentiators and rehab focus group had better financial performance than the specialty care focus group and lack of strategy group. Conclusions: The study indicates that using technology and resources commitments is an effective way of identifying strategies groups. The study also found that some strategic groups have better financial performance than other strategic groups or than nursing homes that have no strategy at all. Nursing homes that provide a high level of technology had better quality, without increased cost or sacrificing financial performance. The study also suggests that nursing homes that service residents with high cost conditions may not be adequately reimbursed.
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 Alexandre Laberge.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Weech-Maldonado, Robert.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0022511:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022511/00001

Material Information

Title: An Examination of Strategic Group Membership and Technology in the Nursing Home Industry
Physical Description: 1 online resource (130 p.)
Language: english
Creator: Laberge, Alexandre
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: cost, financial, nursing, porter, quality, strategic, technology
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: Purpose: The purpose of the study was to examine if strategic groups in the nursing home industry can be determined by examining how facilities apply their technology and how facilities commit their resources. The study also determined if these groups can be defined using Porter's generic strategies and whether the different groups have varying strategic-performance relationships. Methodology: The study began by performing a factor analysis using scope (technology) and resource commitment variables from the Minimum Data Set (MDS) and the Online Survey Certification and Reporting (OSCAR) system. The newly defined variables were used in a cluster analysis to identify the different clusters. The clusters were classified using Porter?s generic strategy model. Once the strategic groups were identified, the study analyzed the strategic-performance relationships using negative binomial, ordinary least squares regression (OLS), generalized linear model (GLM) with gamma distribution and ordered logit analytical tools. Quality dependent variables included quality of care deficiencies, long term care (LTC) pressure sore prevalence and post acute pressure ulcer incidence, LTC activities of daily living (ADL) decline, post acute walking improvement, LTC bowel and bladder decline and LTC restraint. Financial measures included cost per resident day, operating margin and total margin. Results: The study identified four strategic groups (differentiator, cost leader, rehab focus, and specialty care focus) and a lack of strategy group. Differentiators, rehab focus and specialty care focus groups had less quality of care deficiencies than the lack of strategy group and cost leaders. Differentiators generally had better long term care quality outcomes compared to other strategic groups and the lack of strategy group. Cost leaders, differentiators and lack of strategy groups had lower cost than the rehab focus and specialty care focus groups. Cost leaders, differentiators and rehab focus group had better financial performance than the specialty care focus group and lack of strategy group. Conclusions: The study indicates that using technology and resources commitments is an effective way of identifying strategies groups. The study also found that some strategic groups have better financial performance than other strategic groups or than nursing homes that have no strategy at all. Nursing homes that provide a high level of technology had better quality, without increased cost or sacrificing financial performance. The study also suggests that nursing homes that service residents with high cost conditions may not be adequately reimbursed.
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 Alexandre Laberge.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Weech-Maldonado, Robert.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0022511:00001


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 E20110216_AAAAIQ INGEST_TIME 2011-02-17T02:47:08Z PACKAGE UFE0022511_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 9648 DFID F20110216_AACAPU ORIGIN DEPOSITOR PATH laberge_a_Page_094thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
69a56cc028ddea05469a81e0a19020f8
SHA-1
e839c392fbf4ac159b2f400fb98eef24c299a534
9358 F20110216_AACAQI laberge_a_Page_044thm.jpg
b172c848ded994a4bf5f510078a0bbf0
594d5ab3596b37a5c863cf635f9aad8eb4de836b
28216 F20110216_AACAPV laberge_a_Page_116.QC.jpg
a04fe3ab1e02ef0c2d60703034524b77
6d828ef35303754326875639cede559836c4a296
2074 F20110216_AACAQJ laberge_a_Page_011thm.jpg
a560c4646448e622100bf6d723ae0ed5
eccfb4de23dfd67d6d92e7f1866fdf2a43ed5de1
9477 F20110216_AACAPW laberge_a_Page_057thm.jpg
18863676388946848baf7d04e5bb1e1d
f23987affdcced06b88f8941ceceb2b7943ffb5f
8452 F20110216_AACAQK laberge_a_Page_037thm.jpg
6506d1c6d371a9652c7e916391f59306
ce8e5f50d212d1075dafefb6c18e8f3a408b4604
37515 F20110216_AACAPX laberge_a_Page_074.QC.jpg
4f3ba13c366f30801ef5ac21b91112b3
12477ad2670375400409ad6ba7b5677a014f2ce6
8810 F20110216_AACAQL laberge_a_Page_040thm.jpg
24885a7966255179ff2fcacd9dea28df
89d2e9463f8e44a65971a350ca9c2a8206b322b7
10930 F20110216_AACARA laberge_a_Page_123.QC.jpg
46f4d7b54bf72974e20d611639003e4a
7ddb7b48d8b387c1390136f4afa7df8f1112c73e
25955 F20110216_AACAPY laberge_a_Page_036.QC.jpg
4e4a0b4cc62075999554e555a15cd631
dfdb25a68aac66a36cb2b8cd02b8d81127266199
535 F20110216_AACAQM laberge_a_Page_002thm.jpg
b9d7e77a218d76bd41deed64dfd1454b
cff4830d88e58439b7d87dc0e819c895ee56d50a
9526 F20110216_AACARB laberge_a_Page_096thm.jpg
b4edf822070e6ef5c04d1da2bc7dc9a3
e4da86fe23daf5a15704beff63855e1a59f04a0a
13281 F20110216_AACAPZ laberge_a_Page_105.QC.jpg
d54160662d5ef8a1af0d5ee46330363a
96e806adf64eb90d82583497e5a97186f09f3911
8072 F20110216_AACAQN laberge_a_Page_063thm.jpg
1f446f3f87adcec46b9916716322fd74
1b8f0e7281f513ec1c8cfb437adcde22a4f630e5
9833 F20110216_AACARC laberge_a_Page_061thm.jpg
239142068dbea611f02cde27abf02dee
16e358d986457d99d62a64d7999ef18888593cd0
8476 F20110216_AACAQO laberge_a_Page_020thm.jpg
4ae03a568c23a96176f8cbdb5df8c570
6303b7c0feb15d5b78f46c93f11cb40ef13e4f9c
9023 F20110216_AACARD laberge_a_Page_049thm.jpg
d6580cfb8cb61b57601bbf057dfa3dd6
dbc2c8e41d9a3556c40c266a2f455e39cf0416cf
6996 F20110216_AACAQP laberge_a_Page_110thm.jpg
672989c5f5ea38416944ebda009ae827
f9bbfd4af309ae0494ce2dcfd74570d6876d7fef
7961 F20110216_AACARE laberge_a_Page_109thm.jpg
f9658daa3774eda0ea9aa76d6973b0ae
58615cfaa66d8d7df3aadbefea8f391babb75970
34707 F20110216_AACAQQ laberge_a_Page_065.QC.jpg
fc3ec1eeaabd31b466ca4c4392dec8d7
985b1aaaafb3030db957f215c12ea488bc696100
8756 F20110216_AACARF laberge_a_Page_014thm.jpg
1cb9032da8fb051049740dd9bf354421
94a20c1e217ca0e3bed8a98a2a6b6a5c56929c51
36654 F20110216_AACAQR laberge_a_Page_091.QC.jpg
4a1361009bd999ab667fd8a5b4a9cd12
65e40f67256fd07c809596811ffde96cbb9412ff
13434 F20110216_AACARG laberge_a_Page_103.QC.jpg
262e36b35929b1abf59c1fbfbfa1ceda
263fe19ce30799276f874bb2382811ec05284c53
9138 F20110216_AACAQS laberge_a_Page_129thm.jpg
13a456f7034ca767150879e9134fe911
4635703620bceb8d9b71865443eaaf64cbf62e4c
33736 F20110216_AACARH laberge_a_Page_009.QC.jpg
9d586276ae3d9825788d8b008ac67010
5c96bfc66e3790e8c66a1a20f6ae3d5c28bec1fd
29911 F20110216_AACAQT laberge_a_Page_081.QC.jpg
1f834bc7a4607f700f6fdfd9bde36713
46de82de5d35f6ac116d4145cfdcbdb79c1151cc
38260 F20110216_AACARI laberge_a_Page_128.QC.jpg
dd36790939b223ca30f079a86fb112e2
055f5c12498deb0f91bf5d9ed4e2066ad806e336
28162 F20110216_AACAQU laberge_a_Page_107.QC.jpg
67e9cf2da43f1628e2061fa5a610264d
bb48c160ffab676697241c825ea31605f534ab39
21180 F20110216_AACARJ laberge_a_Page_106.QC.jpg
2fc1bd9433d2e80d73ae2ec4f83079fb
aee456307a96619c9337895c0254e5b77c05a35b
36846 F20110216_AACAQV laberge_a_Page_093.QC.jpg
e0aad4366912c3437dec219adfae2279
3b6f7fc5e78372f5559ea4201d1112ec547e1e75
8714 F20110216_AACARK laberge_a_Page_028thm.jpg
204d2e3144f5187e06812c7cf483fbbf
ee974eff0f09979f42d3751f054bf8fe61c0af5d
F20110216_AACAQW laberge_a_Page_078thm.jpg
6dd52b715d62538824464d83a93a732f
3fa10e2f897e6977e3cb11224960ee6ce4c27174
6467 F20110216_AACARL laberge_a_Page_120thm.jpg
4c603e8e8d92d1afba053c9b82db5bbd
c77e42a33312ac145cb64e0b1f391089846c8680
9290 F20110216_AACAQX laberge_a_Page_080thm.jpg
36dbabcc110927719675e131eb4f120f
60f592866650023ee194400cb1e14b8e0f6c6273
4781 F20110216_AACASA laberge_a_Page_104thm.jpg
1bd06c79f96563d8f58f57a38939ef65
950cf74b6ab79cd2b85d1c8f44d29e1bc48943f5
8604 F20110216_AACARM laberge_a_Page_026thm.jpg
1c1093838ffa1b1a3da26a3eeaa39ccd
016f5eff3851d6298be0f646d661a53692ca44c5
9581 F20110216_AACAQY laberge_a_Page_067thm.jpg
cfef47429ce03c7b3334d63a66ac4396
5dbc9fe8ef4268424002119a4bf4adad7d00b61e
9501 F20110216_AACASB laberge_a_Page_075thm.jpg
0f6e624c1370bab5c27af6a979e297a6
97c1214184e936febe7cc86074814ea28c34ae80
32618 F20110216_AACARN laberge_a_Page_016.QC.jpg
489c1b1ab1c99b1cc2657b886107da2c
8acb2b45ccc326558ffb97306d8c23aab0f1ee6e
8945 F20110216_AACAQZ laberge_a_Page_019thm.jpg
cdeec72195419a85dfb061bad34f0778
e854860d6006310463f10a11863d1a8ee5004a8b
36294 F20110216_AACASC laberge_a_Page_014.QC.jpg
0f6f7761c298f26e733e309b173cf465
3867b5d1e7da8f049b8e0172418010fbf40886bb
36362 F20110216_AACARO laberge_a_Page_085.QC.jpg
f5cd01a8fd52990fd8b057205fd6f3db
066b098c89effc92ce6d1c4da18b175f9758b475
36096 F20110216_AACASD laberge_a_Page_075.QC.jpg
eb856ddd57b845b6d46788226390ef34
f30b70dc506ae98b9641a1f34aaa9a0a9d6fc46f
9278 F20110216_AACARP laberge_a_Page_059thm.jpg
9a692b340535c7861fd7f5eaa50a5859
162bba15f1fc3c473721d3ce0dcee9dbaa3bcdf8
35178 F20110216_AACASE laberge_a_Page_090.QC.jpg
f9d2b9a0f0b500fafceadafb8703fbd7
a7399a9ffeed1ec186618577b8419418c0fac6e9
28212 F20110216_AACARQ laberge_a_Page_120.QC.jpg
80898bbe8edf9b503abd4ea60ff2c8b2
c7212f69be82b2e21c97c7abd5add63e67656a17
11103 F20110216_AACASF laberge_a_Page_004.QC.jpg
8da859a53e4b25b4701798cdfe06e746
c0c34f616c20687752d6ac5fb12fe7888da80eb6
18305 F20110216_AACARR laberge_a_Page_022.QC.jpg
3e83db5850d2c9da5f5070aff9a02f1f
78dc7f97c3f7e30de4cfea3ef4b372d8b63859f6
7033 F20110216_AACASG laberge_a_Page_111thm.jpg
7bc799809709d4eb9a191d6948fa1b1b
ed8822e493918fc88bd477f890b33cdc64ecab1b
35525 F20110216_AACARS laberge_a_Page_044.QC.jpg
8d31d06d7265a94624c0a5cfe05205a6
7461c438ffacb80333604dd771baab406ae23aa0
8767 F20110216_AACASH laberge_a_Page_083thm.jpg
e51ad528429ad8c18f44f9308aac1eac
a5c54b07ba7c94df8f08a492613b2956dcc40641
36826 F20110216_AACART laberge_a_Page_080.QC.jpg
4fe8ae1e00ca91cada0f317e2b61d156
d79a126cee3665214d9e0895c4a6e0de48375b85
9046 F20110216_AACASI laberge_a_Page_064thm.jpg
86df514f84417a477f0edce6f77de763
04eb236438e9a9e418a70e9610adb53feb7d63ba
5262 F20110216_AACARU laberge_a_Page_099thm.jpg
dd80fbcd558a3031f3de5318ee88546a
10740dffe857109cc764b504b0c78f8d4a2b2548
38462 F20110216_AACASJ laberge_a_Page_032.QC.jpg
7f294acd1514e9d686c96017cb8fd192
dd561598e403411839bfc5d97ff807923f5a1121
9559 F20110216_AACARV laberge_a_Page_085thm.jpg
9bedb36de24cfc68b13395199ba75b46
1ed2beda00dee9afd156c797228f732c05dc4565
9466 F20110216_AACASK laberge_a_Page_092thm.jpg
7eb709643b08383669d806b70d1ea97a
c26471a0ee070dc4a6d2eb93b592f634abaa261c
34440 F20110216_AACARW laberge_a_Page_048.QC.jpg
19b9444da39f789a8392472605fac015
e02431c43ede1d1cc0a6a82c9f6447ade8e04d31
36753 F20110216_AACATA laberge_a_Page_067.QC.jpg
0c8824c271f95faf705d6a9bbfe2b899
505aed5f31f3e535ca005d3bc3b162ea9e7a193a
F20110216_AACASL laberge_a_Page_053thm.jpg
f1a11a3d7c28902e685d892d40cf17cc
953b72a1a7f5bf17b4c4f2d510ba29c40d68347f
9266 F20110216_AACARX laberge_a_Page_046thm.jpg
7bb6fa0e23f83e51355552b71b8b9d44
e1026b637359666aef7c0eb6100fca3ec2a08c2b
29815 F20110216_AACASM laberge_a_Page_112.QC.jpg
ee2dacd4fb31971ec92359ea58e1a6b2
0029a9b97d34eaefe2c6640becbfa6217b3282ec
5536 F20110216_AACARY laberge_a_Page_013thm.jpg
f4571960897fe9d09bc01cdf8c2c11e2
fa670fcb56432ab72a4bb1ce9b501b6893eeb695
4128 F20110216_AACATB laberge_a_Page_130thm.jpg
8cfc27bb33a65ac67ab2ff71b245a1c8
94373d77884f6ea6fcec1e8cbf5fc7770987bd20
4627 F20110216_AACASN laberge_a_Page_100thm.jpg
3ea7003d71425c4e3c76a283cf225750
faa64b494f232deeaa6acd4ee7514cfbfa879a21
9544 F20110216_AACARZ laberge_a_Page_126thm.jpg
eff6f19199fd80edd8f624ba5bb2878f
2d8c962ab257bbf8b923c64874be365f0bb32517
211119 F20110216_AACATC UFE0022511_00001.xml FULL
76d5fea0b3c788f2e04ec05523383797
7cf0ef67299631a87178f69010756a1dd7fe01e7
33772 F20110216_AACASO laberge_a_Page_052.QC.jpg
7674dcad60bc6a8f91ee4f03360dd780
9998e4ea3c2919b581e8701959bcc7a661be9f0a
1571 F20110216_AACATD laberge_a_Page_003.QC.jpg
7cbd000bf51c7377b1d27260781720fd
a7d17b37e876e468715cc48bd0060fc90c18ea5f
8126 F20110216_AACASP laberge_a_Page_023thm.jpg
b73041fb8dfcb93033f90f2298ad2b20
abae51579f18af6c4d2566921aa183521f68c575
9719 F20110216_AACATE laberge_a_Page_007.QC.jpg
09d52e43825b7488cb786380bc9c24ab
939aed1bcf086d51dde84fc44a3e32969814cfa9
29622 F20110216_AACASQ laberge_a_Page_063.QC.jpg
0b6d2df5097ee721bd4579843643664f
6b275d1c11e8c21bfd8e0cc70073d7ae1eb21467
26229 F20110216_AACATF laberge_a_Page_008.QC.jpg
68eabfd6894c2a2e9cb50a14f57f2065
ae6459c76d1115dacb324d496f00782b3d07bc9d
7164 F20110216_AACASR laberge_a_Page_036thm.jpg
cf5b31fde0e070f99922d04fe3dd7df2
d62b29de86e56dbf9ea60360d1c30e93d6c5d33e
8108 F20110216_AACATG laberge_a_Page_011.QC.jpg
1af70850724bc5abf22762173017c182
683fc8d1492e78b80859a22b4164dc6e8e5ec766
35818 F20110216_AACASS laberge_a_Page_079.QC.jpg
2fd358ec71a0747c65dc6a6e2193ebc9
f9fb393138811c68b72a63caf0710c13253a7c12
35027 F20110216_AACATH laberge_a_Page_015.QC.jpg
3129fc85dda8291025c5abfdfee0ab8e
544217e92058305a1f8c1bc25b6befad48de223a
5768 F20110216_AACAST laberge_a_Page_106thm.jpg
3cf2c339e867e61a8035b17ead6edf29
45078fac3f98284630420106a4b618116beb505f
33174 F20110216_AACATI laberge_a_Page_018.QC.jpg
bd01649fcae8ad13ef188e6226648f0e
6a1a769b59c07b05c1f25d09b8a6c58805fd7881
2030 F20110216_AACASU laberge_a_Page_087thm.jpg
dae0ccf66c95e0077cdf9476a038260e
e7086fb0210b72d57287af65e0cd06fac2f64ebe
35973 F20110216_AACATJ laberge_a_Page_019.QC.jpg
b3fb6f4561bc4663de82ef0aa52dca0a
92d3d9a680427acc9339c11fd66f3a1a81c89a46
4913 F20110216_AACASV laberge_a_Page_022thm.jpg
d51c20d5d33981c0d7598f5546ca3707
7e154099101d0841999c60072266e0f94dbd6d43
33446 F20110216_AACATK laberge_a_Page_020.QC.jpg
f3bb34f1f7402a8395be4355678fa314
4b10b42fa5d210e8feba98fc8e9847a6cb6e6367
6743 F20110216_AACASW laberge_a_Page_114thm.jpg
42a32386966a2569248c9088ee40ac21
bc37bce45b49bcb0f2eb0c7febdd72be97a9cfe5
6586 F20110216_AACAUA laberge_a_Page_101.QC.jpg
23e3139155d2592f8c10606c1ae0463a
84b90bc7c62225759cbf5e4014fb65689dc05881
34164 F20110216_AACATL laberge_a_Page_037.QC.jpg
2934965cf38c601ea71e9e1b9f0debd3
b3f6cd28fdf7ba6fba51766fe6ce02c4d868c66f
9040 F20110216_AACASX laberge_a_Page_027thm.jpg
9edadce30546f8e36f08a133b2bcb2ed
fe1a8221d925f8c5c63141868363903376d69786
29717 F20110216_AACAUB laberge_a_Page_111.QC.jpg
9ec9523d6b82da4dbe00a4f0f33eea46
9a29504278660bf81b92485c8b818f91eace5d6f
32380 F20110216_AACATM laberge_a_Page_047.QC.jpg
d4eb3b6ce6edd6f0775b233a8dd53e73
35ef0bb06da095dc67eb397ed7259ae8b548c7c5
9295 F20110216_AACASY laberge_a_Page_098thm.jpg
92db2be71ae0c79a91845d75bb58c49b
a1b27649bdb77b593e1905b9878b9e4c39a04091
32465 F20110216_AACATN laberge_a_Page_055.QC.jpg
b53e23c6b33fa407786d88654b9e167f
06e24425b4a2f8b4a7e65e7edc713ea7477d64db
3560 F20110216_AACASZ laberge_a_Page_103thm.jpg
f8c38d9ee3412c076d15626e2ba1a2c3
2bf2b6b01f0c08934f2df23bcfabbd1860e082e3
30955 F20110216_AACAUC laberge_a_Page_113.QC.jpg
a1af97bd0a6c94eda7ff91ecdf1e519e
68dd10686cc8d5d0f1d8d7fdc4aebea23dd4304c
32753 F20110216_AACATO laberge_a_Page_056.QC.jpg
713ad995a3c483753d028c5c23970aaf
159f363ad0591228b4d0e9cae31071cc26f1302b
28409 F20110216_AACAUD laberge_a_Page_115.QC.jpg
a93014c5dce39b836801a31f732371ba
fce02ba0a79a5e0ebd6074eaf333168f56df7cc4
36574 F20110216_AACATP laberge_a_Page_057.QC.jpg
26d6957c0e916691d89e57fd077506da
f08747357281948b0ca1198a2896d5f1dc320993
28523 F20110216_AACAUE laberge_a_Page_121.QC.jpg
e462e8a214d86482dcfae7e34dbfd711
2de85ae94a45a1e2a9c23db6ba53abafd5bef0f2
37775 F20110216_AACATQ laberge_a_Page_060.QC.jpg
0c3d663646c42ccc89e0f524dfce2cb3
24a22149c9506e4e416088dbbed904f71b77d0f2
2143 F20110216_AACAUF laberge_a_Page_001thm.jpg
173544cdbf4a810b9fda8062dcf2410d
fcb36c8925dfba76b32e1d1c4258feb7cae90d87
34063 F20110216_AACATR laberge_a_Page_064.QC.jpg
175c961e5e9671329375e45b45e22552
948b2af411155966419b46fc8ddc8585ad91ea5c
649 F20110216_AACAUG laberge_a_Page_003thm.jpg
96b098c717abe426d2c60068f7e834e7
2e3247da7d978a941634f1e65a0680818dd027b3
36849 F20110216_AACATS laberge_a_Page_082.QC.jpg
897576b08c1c7a2377e5a4d74f53a774
3c41fc11f87161a78a78267ecc4e34ee53e1a2d3
2672 F20110216_AACAUH laberge_a_Page_004thm.jpg
542ab75bc97826ca4d2a99bfd29f551b
aa68be2d8ccf4196a07730840ded82b640ebe232
36395 F20110216_AACATT laberge_a_Page_088.QC.jpg
7ff5d2ef0d7f647db8de310f55d815b4
2e01bd66df7b6a64d0b196a99507bcb31550867a
2391 F20110216_AACAUI laberge_a_Page_007thm.jpg
44e69352cc60a8244d3f248afebff187
a5e2a4d06ff14af5b41a7c954aa2f0c95ea980fd
38918 F20110216_AACATU laberge_a_Page_089.QC.jpg
7c05ee167a4a89ef23b6bc5c986a9051
1ce0cb78f3bcbfa195ebf2fe11fa4c82cb51ebef
7689 F20110216_AACAUJ laberge_a_Page_012thm.jpg
def409706c89521805fb4f994dbf821d
ea7a91c22f07c07bdcce399f44c0bbf54c93dc27
37730 F20110216_AACATV laberge_a_Page_092.QC.jpg
25e9cee0d08f5debd80b525cd2b78ddf
e4fbce5a969a8a1a3ad20fd89a65f2956cd21b94
8377 F20110216_AACAUK laberge_a_Page_018thm.jpg
081addb6738e3a451e2faa68a1164a88
44e971769772d731196eb8bcc28b40abd13bb5b7
38735 F20110216_AACATW laberge_a_Page_094.QC.jpg
da99f6a8e4d19959f2035e44920e1c17
eccfd6f3d8430993e9b71a1913ff923130f388b0
9211 F20110216_AACAVA laberge_a_Page_071thm.jpg
c09fef24e1ff09c241b746ec39bfa0b7
f9725cb69f85e68500469c9f2803e2113a6cbc1c
8540 F20110216_AACAUL laberge_a_Page_021thm.jpg
1c41c21c9408ed44eaf5f7ae65dd4255
597fa660b34a4aa9126ca2fcf632431e91200696
35758 F20110216_AACATX laberge_a_Page_098.QC.jpg
4d3bbeae8c6064b5f8c92f6d845a3af0
c900b3b1e173f2389a99bc1c477fced9ac614dad
8656 F20110216_AACAVB laberge_a_Page_072thm.jpg
2a84fe89590c78d3e05d11e7082dd081
e2bbb105cfb918d7514d6d72467a59da6e4ef646
8923 F20110216_AACAUM laberge_a_Page_024thm.jpg
9f99cd7445de183d69e1f07e589e1efa
1de05d22519b1409a123e2dffe0249f972357762
20077 F20110216_AACATY laberge_a_Page_099.QC.jpg
d4a43837e2323a7dd648f754800006fe
41f0733d899e254984f7fc8317a2b78a65e8dcc7
9268 F20110216_AACAVC laberge_a_Page_073thm.jpg
c8ea5e3ea617d9ba7b6b6568a1949718
b58616066d106e023a0682d3a2d46488b2db601b
9064 F20110216_AACAUN laberge_a_Page_029thm.jpg
a49242013c261bce619281ac5cbe97eb
f9d0be56534c8698c31529dc2ec4455605e26a1e
18988 F20110216_AACATZ laberge_a_Page_100.QC.jpg
6580cfd7709538b8aec847d3ec31484c
031b86cda3878bbfd7be6ebb2afc291d8e012ec0
8606 F20110216_AACAUO laberge_a_Page_031thm.jpg
bb36ebb2bdb45e8b5ebc1c375a9385c5
234f7af7107cc6ec04012fefcd5abb9daaf9c307
9222 F20110216_AACAVD laberge_a_Page_076thm.jpg
11e36f12452b8e7eb67520ae9107b3a8
b1e541299b46c5d1a0957e6730b95f9d324cc337
9066 F20110216_AACAUP laberge_a_Page_032thm.jpg
0021145ac08e82a0e24871a3d5fdd4da
475e021929e6b10eeb06efc333b35152862f4432
10081 F20110216_AACAVE laberge_a_Page_077thm.jpg
d89cba24940a5583a0ca966bd2af49c5
bfa1dec0dd2c5013e3f10db3fdcc44b037fa87b9
8850 F20110216_AACAUQ laberge_a_Page_034thm.jpg
8a5b2cfb61c40937ed591a37c308ce8b
8c116fe37bd596bf7306e9ba9263b641c1de1abe
F20110216_AACAVF laberge_a_Page_079thm.jpg
be63c9201f0f6068c3bbf5a65869e1d9
f80baa97a540b533d106696171b043902b4d0a3b
8399 F20110216_AACAUR laberge_a_Page_039thm.jpg
3d7e77cd089763a4bfad2373a30a1b0e
860dbf65328c4e3c485e86c39650991bc1f2e43b
9309 F20110216_AACAVG laberge_a_Page_082thm.jpg
b795c2467e1066dce87822771eeec2bc
581b7ec51bf721567aa6d5c28e886f10971ad936
8284 F20110216_AACAUS laberge_a_Page_041thm.jpg
8fbce951475dc38ba5e16df93c9792bd
e4dc2163188252e127c69cb013e34531a5f2c6d1
9437 F20110216_AACAVH laberge_a_Page_091thm.jpg
ae794ef2fd132ba04317f70a1936f7e6
30249b33b008a70e8ddcb10e764decd0dab1543a
1740 F20110216_AACAUT laberge_a_Page_042thm.jpg
0675ba645b5952613b95d710f4c5d830
8365b2de8aeb609534f80407ef4cbce9b7e9d7c3
9333 F20110216_AACAVI laberge_a_Page_093thm.jpg
925f0d10247181cf2a39c98fb3a67a16
6d2885291883db37a91fb0a2ae82d737eaf174d1
8748 F20110216_AACAUU laberge_a_Page_043thm.jpg
1d2216a03ff11ff2d9ad52f1c5cfd39a
cfff9c471d40d709fbdb24a1cd62c9361d9fd60b
9550 F20110216_AACAVJ laberge_a_Page_095thm.jpg
d39f85f825ee373fd6951c877be48dd2
c59ba6f912fac3f2ceac468d02aa73f1d9f2c56f
8899 F20110216_AACAUV laberge_a_Page_052thm.jpg
1d3870d4573581df6c70ab58796c21a2
e7d28356a08e6b7cd38658ba6b7ea10788f6326f
9312 F20110216_AACAVK laberge_a_Page_097thm.jpg
f27274e53ffc995f7a5e88609b0e5cc0
f550a84303779dc79dfb6c7d4d6de9aa7a245623
8758 F20110216_AACAUW laberge_a_Page_055thm.jpg
5b81d12701f1706c4e25297ce1c67b45
36582704cf47d93410eab61bd5158689dd9cf131
2085 F20110216_AACAVL laberge_a_Page_101thm.jpg
2adc240147149b2f7b366049e6d9bf83
1327c1ad4ad08986f86edc770760e10f8b07eddb
8611 F20110216_AACAUX laberge_a_Page_056thm.jpg
5c80a2a4c50d2e235476edfaa881c4b6
4216d89695242c7553f42b224502d0dd6f6511b8
3765 F20110216_AACAVM laberge_a_Page_105thm.jpg
df9dd5aa0671ed09c329794d067bec9c
218c32be07f00bf2b64f569da0b59810db69ad16
9644 F20110216_AACAUY laberge_a_Page_058thm.jpg
20ac97e0eb5ff848c704ab44a14df1e6
dab927be66419da75f5ed632f4685706c256155c
6652 F20110216_AACAVN laberge_a_Page_118thm.jpg
9464948d152bd92aafaa34472f1df353
c513492b3f44f34adc1e6094fc6628a120942fdb
9280 F20110216_AACAUZ laberge_a_Page_062thm.jpg
9dd92e64a4fe2d63de2931ad47ff2f61
004c11a6625c7b046ddafa8a701390804fd63cd5
6707 F20110216_AACAVO laberge_a_Page_121thm.jpg
f072f3ca14d5ce33d3a395482a9c77ec
c364c42babb26b6caa86da6f96b2b458d3671f71
6119 F20110216_AACAVP laberge_a_Page_122thm.jpg
3085c5daf51f3bbaaf84b742909f48a2
9e0b08b9a1821f92083f39823731a451712eab38
2787 F20110216_AACAVQ laberge_a_Page_123thm.jpg
668243a19cf58d793a7f86774130133f
67f0014ae5643bbcb094c0721358218705ec6e97
10095 F20110216_AACAVR laberge_a_Page_127thm.jpg
07e7ecc92717d876d14822e3a0a6f33e
6bccbe8254219854b4c83bf085ad503ef3cf0599
70581 F20110216_AABZNA laberge_a_Page_084.pro
cd555f608d9939c7f9604826cbc0095d
5bc9c0ff84d4fe344d3495355fd9d30bec028086
35781 F20110216_AABZNB laberge_a_Page_126.QC.jpg
63152798102d92ff13aff8918bd93067
e357e631f792093838989440f87babb920620853
2294 F20110216_AABZMN laberge_a_Page_075.txt
77e3e4933305f51ff66713655c9bded1
3825d50f6f70ab916b62cd31950b580e441671b5
92803 F20110216_AABZNC laberge_a_Page_070.jpg
91b85c708615bcc0a697519dc19ca9ec
e2e61afba4e7dd4352d721d42921d2dae9abaa4c
21021 F20110216_AABZMO laberge_a_Page_101.jpg
b35f10994d1823ad22d9482e96d8565b
5ecf60695303da0a88fecf711b312d09cffc62bd
8075 F20110216_AABZND laberge_a_Page_081thm.jpg
19f52d0b661fa2a98ae0eba6079985b9
61eb57c357695b73ace91b675ef6fa06902a7130
8425398 F20110216_AABZMP laberge_a_Page_103.tif
757734fd6cf6697b426f8966a34678f7
9e9725fc5431840268a547e01e19235224ff9ab9
27829 F20110216_AABZNE laberge_a_Page_030.QC.jpg
6e10f9f8634f6d56426d093e1700470d
b4551d66b3901fb22935dad226b9350a896fee9f
113280 F20110216_AABZMQ laberge_a_Page_096.jpg
c7db101e3a5cb5a2378b9466d6d2c7de
040013f928c7bb2745a08106d4b747872307fcd9
8423998 F20110216_AABZNF laberge_a_Page_011.tif
03072db9e35bafd056c261ee2d81416b
657a080886cc135ee631fc2d498145b0ce9bff6c
100279 F20110216_AABZMR laberge_a_Page_117.jpg
b74842eee0ee7b6e05785f59b34c3480
ddbf717cc4b58b5ad4544e5b17e4ca444328433c
99049 F20110216_AABZNG laberge_a_Page_115.jpg
ca2174acefe66128ef218e0caaed0149
1aeff31516ae256a99825ad0c14d6af7124e3b7b
4962 F20110216_AABZMS laberge_a_Page_102thm.jpg
81bdec7639b4d0312d66acfc14f5bf5a
6247e24982f94d41dd0a0beb2258c3a7161dc1a2
102475 F20110216_AABZNH laberge_a_Page_047.jpg
913a4b2a564f06bbc20779af833a2519
7e70e04e08623308d56d65d55106892c0bb04b66
9215 F20110216_AABZMT laberge_a_Page_065thm.jpg
a34c852cf150d70f2f17bd8c74ead8e4
95799280aea006897ebc91a493a2e35d765723e3
2103 F20110216_AABZNI laberge_a_Page_031.txt
5b7621e9b22d7f8e8719ef3eb67cb706
0ca8ce14219a1bf81e5ee0f588ca2730d791d494
F20110216_AABZMU laberge_a_Page_003.tif
2f40751cdaff0945e7e1e5aed66469fe
e2dcb399e51625e2d2cf4717ec01089dc1cccbc3
62285 F20110216_AABZNJ laberge_a_Page_100.jpg
4a4ee5fa32bdad86ecc85a20aea9ead2
b6f10914501b4f01199d380b290b4c731e7287dd
1051943 F20110216_AABZMV laberge_a_Page_074.jp2
d109bafee33e089816b2a3bc53044a67
ec31a8de7b2e98be89f5791f6cbfec40520e30cf
2248 F20110216_AABZNK laberge_a_Page_010thm.jpg
d3124d17e006ee7be6396b9bcac016e2
718d60a3dcdfc53b8f7c9243602eec8e611fe2fa
55293 F20110216_AABZMW laberge_a_Page_065.pro
4e5a8bd1622032973a784e97e8c84c4c
60200e06a07e451f45c459a80ad63c3974cade45
1051907 F20110216_AABZNL laberge_a_Page_035.jp2
b5f6c76b8c1e4df13c24d1a60b4499a0
79e9355f635903ca7bb97cfc62db9415637be119
53035 F20110216_AACAAA laberge_a_Page_045.pro
6e46d9f70b227e0a665b7ac7b1776dfb
0780428ced1e0bd5aab5e808317992c2e768b9a4
36256 F20110216_AABZMX laberge_a_Page_078.QC.jpg
7b0cf585c8856e24abed546a749d9641
9a7d374fe1294b9b94874e6e1397bf577a35f898
50856 F20110216_AABZOA laberge_a_Page_020.pro
826b397b560e2759e09df472ac401d41
ce504aec8bb3a2c8875bd3cd0aa29b349121a9a1
53906 F20110216_AABZNM laberge_a_Page_048.pro
694a7bc7180dcc563e38fef1abcf3695
3e51a2e5c46035d33689661973d3ec34bd133419
55972 F20110216_AACAAB laberge_a_Page_046.pro
8335979752c8bf361dcebab9f22106a4
a36a27295686d8e5f83f6a3285ce2b22b9107a2d
28506 F20110216_AABZMY laberge_a_Page_117.QC.jpg
922a755dab251dd82a2792270e46721e
ab595747750b350545defda7b58b12c04c251ca1
18001 F20110216_AABZOB laberge_a_Page_104.QC.jpg
0f197533ae3ab6e144877e3bcf328529
fbe56cd6401d9cb8332ae98bced2ab438b4e8504
F20110216_AABZNN laberge_a_Page_052.tif
3525b4409967b978dd9ac0778a509f6f
726ec814ed45763a5ad2059eb43a0693eefef3c1
52170 F20110216_AACAAC laberge_a_Page_047.pro
12a07cb6926f4f204376f14f0a6eebde
62ee2a0947a5763602f04c79bf38c0cdb2d93e72
36896 F20110216_AABZMZ laberge_a_Page_095.QC.jpg
a5a7c3f54cb3ac7acd2cffd0a14ca7e6
14ae8b23e3ac0b110be74b67ebb2e8a0b103bd80
F20110216_AABZNO laberge_a_Page_006.tif
9f0a1e93b5dda0f9d561e2eb68c26296
5c001d7dfdbcf00b931ac148fa770dfdc29a4ea8
56474 F20110216_AACAAD laberge_a_Page_049.pro
db720720b9191bac4433f44fca3fa19c
680268c56fecc065c550ac7dd72947f24ffd50e4
1051976 F20110216_AABZOC laberge_a_Page_078.jp2
d40de2ea853593847f4bcb24324364da
0a94ea140c3aa0a8c16926c86fa780b1bb79c0a7
F20110216_AABZNP laberge_a_Page_026.jp2
ab27df697dd3f4aeaee2ef62cd8b9032
eb4ca7747c4fb5ed016b9383180e1ffc1ae06925
57498 F20110216_AACAAE laberge_a_Page_050.pro
ad030be75b32603b848f54aa90613642
cf45f58099cc1624916e0e7f266b6553e784a283
111038 F20110216_AABZOD laberge_a_Page_029.jpg
41574dedd90c3141339054c63ef467d9
b0134dda66c84d336c54f8c7dbd6162089ad44c5
8953 F20110216_AABZNQ laberge_a_Page_015thm.jpg
8280d1bfe5ec07469dee063e95d655c3
cf1e9f15a22fd62b79f7fc56a5ef9d49fa1cfe43
56863 F20110216_AACAAF laberge_a_Page_051.pro
494c49c8124faf883615165231362715
0bb80bc8a5089ae2e64469ef54d1e86583e1a854
2420 F20110216_AABZOE laberge_a_Page_058.txt
1380a55c318a225c6c516c49c0d59d56
4a169f7d9f92e45c0a283b0cc214ebc988105bc9
757191 F20110216_AABZNR laberge_a_Page_013.jp2
90016eac7ccd140231137289a7f20f6b
e2aba0a567447c9e48b4aae5fd2083d5469144a9
52732 F20110216_AACAAG laberge_a_Page_052.pro
1d343424c2b492cf751e9b6a0e1216cd
f63dee381d7c23db6b111e0743b68d1ed34a0bf7
5410 F20110216_AABZOF laberge_a_Page_104.txt
7a257687d2f2901b41a8780f5e9cc424
62cb4ef1b438a5da1ebf27da28952ddc4eea7e5c
610 F20110216_AABZNS laberge_a_Page_004.txt
e5cd109959cf4221bcf4604e3541c33b
35aa123aaa9d8c5442236b6084bc0e2882c85f26
52825 F20110216_AACAAH laberge_a_Page_053.pro
569ab0c64de57f886841881429ee3dbc
c9ef41857d8298e66acca33804f7bc21520d5230
59086 F20110216_AABZOG laberge_a_Page_092.pro
dfe366a965f5a596adff8d58d286d124
33fb63429d1f2c4dfd5592558d7c5de61e5ddc64
100899 F20110216_AABZNT laberge_a_Page_005.pro
1a416a411bf10db51aee79763936604a
2f6dffa29850e2af222c8d458aab0980f3425738
56450 F20110216_AACAAI laberge_a_Page_054.pro
e01231eedcb741a4ad95fdd9d708724e
641eb5453b2bbeec07f757d2b00558d6312c1be2
18631 F20110216_AABZOH laberge_a_Page_011.pro
4feceb93ab5244fb30c2ca60f5bf19dc
626c47a324aae68c0652e4e765847d8d3fba7e3f
2864 F20110216_AABZNU laberge_a_Page_009.txt
80f9fe54053ae2163ecbcb66800d4f42
c5ba925fc2fccf9e0698782f39428d58505cbf63
50047 F20110216_AACAAJ laberge_a_Page_055.pro
12cf1a67a5566deb26050f5d42cfd90a
66bf41ed49aac0c2dbe61993733fe3e01748977c
105520 F20110216_AABZOI laberge_a_Page_015.jpg
225793c65049e649859374e0d40b31c1
e4965d31494c2fa2c1ad604156c0447bf64e7fc0
2244 F20110216_AABZNV laberge_a_Page_097.txt
0c145983742da0679180e2c65870ab0c
2e21e19bca4e01b46b630abdb67a782958f1071f
50152 F20110216_AACAAK laberge_a_Page_056.pro
ee390d456720ff57bc3c2222b485cfb8
f7e4b066e706f825138a4d83e664bdab22a772b9
49161 F20110216_AABZOJ laberge_a_Page_041.pro
aec148bbb4104291cefc86cc9f897591
bd2cd5d9315ec3db1cdd7f43003583a164cd6f62
F20110216_AABZNW laberge_a_Page_023.tif
cca6d5d2012b8649e3d2226757abddd7
4b2cb4fc96cd7b48534f42dee6be0e542ddc6a3d
57648 F20110216_AACAAL laberge_a_Page_057.pro
1622b93beb03342e5a49f03334d31775
b4db799b93e84d83f278a7ce5c6951bc7b1ab736
255967 F20110216_AABZOK laberge_a_Page_011.jp2
293adfa1881257165384d6288116c273
adc23d146d99e9c8c94c7759581c1702a6a5e9b8
36340 F20110216_AABZNX laberge_a_Page_097.QC.jpg
97c7d181e7ab70d6459658f38579fc1a
bdd3945961ac222b2b5e5124b2664e5c0cbccaa4
60297 F20110216_AACAAM laberge_a_Page_058.pro
0b996bb854071f2fd9975901a0e2cd9c
0c2b090da9060f28c74b4837d470eab8839b5566
4113 F20110216_AABZOL laberge_a_Page_002.jpg
84aec9b03d324bf95ed8f2bd1eef9948
9f8e4228316355f3d342ec0f800e547958b66b66
57855 F20110216_AACABA laberge_a_Page_074.pro
0ab0d60b24ff5b211df14076db22653b
9bb9f79241ef2468dfec3625fd1cc54c17027a86
108706 F20110216_AABZNY laberge_a_Page_073.jpg
f9d60e33dc9bb842a1bafeb958761828
62be2703a2d4491ff327d7c08d979f95b35bff6c
55030 F20110216_AACAAN laberge_a_Page_059.pro
a9193fcab78155e38625b87f697140f8
92303dd0abaffa000976150e020c6674fab98085
F20110216_AABZPB laberge_a_Page_001.tif
e6c5ff17193b0596288a90d3ee4d090c
af91be11d3ce793266f45deedf854eb4566f2b0e
1051949 F20110216_AABZOM laberge_a_Page_057.jp2
3053fb3b24e5479c867b47cf4c78dbf5
9cadb784c359e446852b5aacf6e3d12bab49efc2
56961 F20110216_AACABB laberge_a_Page_075.pro
d8d56f523641012ed43a632974b8e480
32e26c1bd1e804cc7b9afa1b8c90e750d8955749
6503 F20110216_AABZNZ laberge_a_Page_119thm.jpg
f9f1795688f651fa5eac535711fa7fa3
fc540ede04c6134b2de08a818a6f7b6a40442b14
F20110216_AABZPC laberge_a_Page_002.tif
4670969c7a19f8d4b52286a964d49b0f
09c9f3db52dffa508c8ef18cbe58e79e7cd3b326
54702 F20110216_AABZON laberge_a_Page_069.pro
8d0c086bb67b2dd5617e53e1a317b790
ffc2fc8da964cc7aec0ca52b3737ef73280b1c98
74945 F20110216_AACABC laberge_a_Page_076.pro
c421d453d344eeface532fd6a80194fa
a8f7fe62cc34745e541da03ecf813662207168f3
58061 F20110216_AACAAO laberge_a_Page_060.pro
34fe25304988ada3d79aab7c34c757a7
1b98452e9d3b50612b903a0c60ca12933deb6a72
36509 F20110216_AABZOO laberge_a_Page_086.QC.jpg
623517fe2f359003d4d25c1c304d54d9
4e0a56f0ebfb4b01e39ba45efd2c8b3b452673d1
60795 F20110216_AACABD laberge_a_Page_077.pro
1b4255fc573f27009ef51498f62516d9
55c0f4229ff92ac81df81b7c4bb96f37f4c7f502
61450 F20110216_AACAAP laberge_a_Page_061.pro
0be8010d019f600c3b7c83f33021f5e6
23ee729b80bc7578cfd3a8a1ff3f15a197c74f37
F20110216_AABZPD laberge_a_Page_004.tif
fe2f603e00f3a0d0703febee761e9a6b
66a15cea2ad389b56dcae25e011d3a8a557e6c06
56676 F20110216_AABZOP laberge_a_Page_098.pro
6bcd1428ea2d0b5fce9b02f7c4ad59a3
a94de177417cb0b4b37c2b3fbc3607fda4e473b7
76273 F20110216_AACABE laberge_a_Page_078.pro
f9e9514a6d72b4a47efb0011d70c843c
fd9f7a96bafb221771ace5882621cb5591726b2c
56242 F20110216_AACAAQ laberge_a_Page_062.pro
72aa95cddb73218729a69e81909ecf69
56f7e2bbfa93ad8e5c2821b8aaf2e707113a627a
F20110216_AABZPE laberge_a_Page_005.tif
aa500baa22fe76ce63745fd8f3b14f3e
1e9e200ab2bbf0ef1b0209427637a5c0c1043fa0
61475 F20110216_AABZOQ laberge_a_Page_112.pro
ff9435d76a6eb5c9618ad4a16f04527d
ca6a53165e6506abfb4e41790565739533f844dc
69757 F20110216_AACABF laberge_a_Page_079.pro
a1af301c38777bfc2b6aaabb734370b4
0a27b64c6c5ced5f99fed55d4e8dcd6b600b95f2
47441 F20110216_AACAAR laberge_a_Page_063.pro
ae6ca0f6d12241fa75c825598818f0f3
f91eb1123042da3b99bbcdd794a69a720fb37744
F20110216_AABZPF laberge_a_Page_007.tif
e844f70547de20a3554d7931ef650ab2
1c43fe8a5893ca65b8287de84de68f994b832a6c
203447 F20110216_AABZOR laberge_a_Page_087.jp2
060f4ba965abdb08df8bb379d5fd5289
ae116a88f6ba4b7c834b8d3a0ebbff27fd19ac42
57999 F20110216_AACABG laberge_a_Page_080.pro
571997423b77d89676b5b43bf710bad1
b1b32bd27320fb0b727c97558fd7fbe09d9c9cd6
54619 F20110216_AACAAS laberge_a_Page_064.pro
791e8af53cb5c24f5296417a6cdb3054
9c10a536dee5a4e659ba1e5e918baed6d6186ae9
F20110216_AABZPG laberge_a_Page_008.tif
d798b550ff8571d4cc9389198419445d
a9e7befddeeae5acdb6ec29d94e456ee29f18343
9717 F20110216_AABZOS laberge_a_Page_089thm.jpg
5cbcbc9835bc244f2fbbb6c9934b8727
1a0bcd2ad804e6d0efe032b95968d1f877330c66
34387 F20110216_AACABH laberge_a_Page_081.pro
bc50a9985cc94614ea09c4f4da6acacf
701735886b5ef98022569086fc20d8566ee9a9e2
60207 F20110216_AACAAT laberge_a_Page_066.pro
0675a381b3f0c52c7e0fa1a6ccc319ac
0b2d8749478275df7d5698a1d9be44a9015284c6
F20110216_AABZPH laberge_a_Page_009.tif
d2dd45f564311f73c22c6422d7c707e0
020d74ce0f0b5e4f80258b45a8e0a2c3e9de1654
35891 F20110216_AABZOT laberge_a_Page_096.QC.jpg
1249f886a6aa10cc0db8b8ace93e3243
e5ff355b4fc5eadb40d894b389e4a40603599e31
71316 F20110216_AACABI laberge_a_Page_082.pro
ee8cc9a5e7b2319b2211228d08721a89
8752ba427f6ad10e59e6431d089c1426cfeb5b21
57583 F20110216_AACAAU laberge_a_Page_067.pro
d909a785db6be08ba3f581a3d38482c6
a071986caf5822ff9c0df123413e0bc81bdbe33b
F20110216_AABZPI laberge_a_Page_010.tif
f09417f03e9b88f02a089a729f5af9d2
2e8694bc7ff1c8078ca06b8d4791a17d298c9e03
39025 F20110216_AABZOU laberge_a_Page_003.jp2
1b0150afe785445e9711a7312da20900
6918a165fcd19f09680ef6b3ee4a64aeaab9ffc9
69316 F20110216_AACABJ laberge_a_Page_083.pro
59c3f8a62707066061281100c69dc3d6
e2ce1917d370648f86b34b36529322eac791d8d0
60175 F20110216_AACAAV laberge_a_Page_068.pro
d70af494f84ce8338f2a5e589b1b7121
da097c829da44a2a360ff3e6605c046936a5803c
F20110216_AABZPJ laberge_a_Page_012.tif
d968756310069a173c7dc511e151638d
cddeae67c4e92519638ba32027547e6fe8334b45
F20110216_AABZOV laberge_a_Page_104.tif
3408fb641124b52761c6768d123cad2d
0e6402a4060b7a06c46b33fb191d56648609fbbe
57327 F20110216_AACABK laberge_a_Page_085.pro
47e2ac2864f837ff5d5ac099cfc872ba
40c71964c6d819e52dc898bea392a5d7575d4ef7
45474 F20110216_AACAAW laberge_a_Page_070.pro
7bed77d0fd616bc83a2ba98c7e5c9e7c
747e079d15ef258f0866f54606aa565da1dca74c
F20110216_AABZPK laberge_a_Page_013.tif
4666bc903de2fae6eadc2e27384e920a
aa021f8f7673300ec4970de9d57151660c9c19d0
F20110216_AABZOW laberge_a_Page_066.tif
426d7fc1d51cb9b8b38a927f3fa40a15
7a29d402d98be1327d910848092eabbe6d944458
60166 F20110216_AACABL laberge_a_Page_086.pro
f22a640e41043785e7685734ede49fb0
e56c09b1a9de8ab38a11113503a1304548ef9098
F20110216_AABZQA laberge_a_Page_030.tif
c9248cdc9150a374d7f98d9ca2f41b57
87113d95aab290937d7bbfe257d1fc13caa2ff38
56575 F20110216_AACAAX laberge_a_Page_071.pro
d66d06790457d5ac1ef1d243a2f816ed
b4d32208246be5398b6f239e3b0e429cda83a744
F20110216_AABZPL laberge_a_Page_014.tif
298a5ddf7709608b06b6d42e723d3bd9
00bd76744a7bb826154eae8e55310e5841653f2d
69192 F20110216_AACACA laberge_a_Page_103.pro
0d7488fabb708d5b9ac36e3e43d7c2fb
0fda73b4a0b19f1f954527544250aca711e57948
74107 F20110216_AABZOX laberge_a_Page_105.pro
7e88ea2baea2f78a16543a59ca1d628c
c5e5f10f1e86bdc923f1025df413931c164061ab
9391 F20110216_AACABM laberge_a_Page_087.pro
05bb45400715601b2753ffe332c6fd62
d2f719febf54155d2b4f9882b37332de1263c184
F20110216_AABZQB laberge_a_Page_031.tif
120f62f0dd942e7d110616939d7d2e97
d724fa6317ddc3f94aeb655d2fc115d242e61e5b
54740 F20110216_AACAAY laberge_a_Page_072.pro
c68db48534aaa61b541b0e6ecb5f1177
343090b2ef337ff7913783a3ebdaaf3dfe6f7c87
F20110216_AABZPM laberge_a_Page_015.tif
b97186ff74c64e3b56729b3de246c0e4
3c05f09c6947048ddd28371a1cc64a45488abafb
100106 F20110216_AACACB laberge_a_Page_104.pro
d7c6b252e2bb83c312956e5a7a8a20e6
81f9ba733c13ab77d5fe2ba164f94b58e5994e13
154234 F20110216_AABZOY UFE0022511_00001.mets
e93d40ae18424e0cdaeba3f294047d50
3c13ebd4e722b1ad4ac9bb2742a469801ed582eb
56066 F20110216_AACABN laberge_a_Page_088.pro
d7bc477c1b6ca20da292a62cdac68d94
279d396420c21f272dbae9cc23e955542c9cb6ad
F20110216_AABZQC laberge_a_Page_032.tif
81bfb10b9cf170e6fadf68ca33fd738a
4112f061ee8992d02038425b88dca5b7a7891bb6
54350 F20110216_AACAAZ laberge_a_Page_073.pro
4de6360f44b44bc06f782db01a007d15
a43aa7e51a0442bde7b4d34b1557920c71b27951
F20110216_AABZPN laberge_a_Page_016.tif
94989a861b12b5761eb272b945112794
10a8d7697d9b97aa04ef2b2e6fdf3800d4cdc6b9
31060 F20110216_AACACC laberge_a_Page_106.pro
a875237afd2c2d95ec0f7373a80b7acb
c6efe4f69ce0df88feb81c8d6ec56e80967586e5
61104 F20110216_AACABO laberge_a_Page_089.pro
24407de83935ad25faf59e84f0db99d9
3d874fcd51a87afd876ae1ac8483903c59ca04b2
F20110216_AABZQD laberge_a_Page_033.tif
da3e3534eb0d5bfb5b07705c8556d285
0f057af14014c76715aa6803a03add3b9f151aae
F20110216_AABZPO laberge_a_Page_017.tif
e73070166128dc5da71ae2212d794211
4f2f759832da034794b15b736bdb084a676bdb50
44922 F20110216_AACACD laberge_a_Page_107.pro
4619127facc1866686b7c55f17479d7d
ff41da7661a3ee2ca81e335ebe6a1f5d47852642
F20110216_AABZPP laberge_a_Page_018.tif
3772da471e648af7bbfa15bac1e4bee2
bd742da799b6fca0c7a92d159d4b5d93c1263b48
76566 F20110216_AACACE laberge_a_Page_108.pro
d7620810c695ef91b77eb23e47c02436
61121d9a17ba5052f5eb7f43384aea7c153bf39c
56416 F20110216_AACABP laberge_a_Page_090.pro
554b5a629f54049d7e2ee2ec349850c1
be7eb0cf304677381e195516b9626ba7efd61975
F20110216_AABZPQ laberge_a_Page_019.tif
0512148faf66672abdb35b342106305f
f5d14e154a43b221259d3c863045612fcb3f3d05
75234 F20110216_AACACF laberge_a_Page_109.pro
fdfb19a2e77faeb27550fb8491196238
202de4e09bdfa1c87383751ee1e49ed537d3b9a2
57981 F20110216_AACABQ laberge_a_Page_091.pro
ad04dd1ae50b8b4b9f9c64e6333d626f
e815a45606273c0da9888d92065077b5eeb01643
F20110216_AABZQE laberge_a_Page_034.tif
2e472103e3c95fb991a809b254c4ba7f
6e705c77daeae10c063acb22f539c9020bc6487a
F20110216_AABZPR laberge_a_Page_020.tif
cd6848011beef2504685a0aec3d11002
212be8c158bf054518e091465b7278881ee90aac
69606 F20110216_AACACG laberge_a_Page_110.pro
4c18cd161586a48e34c9988f418bfc8a
d9acb1d9362015e2678fa7feb1c919a0420b7dcd
57168 F20110216_AACABR laberge_a_Page_093.pro
ff3653a08815779ef908e34130d410aa
38fe2cc7fe9c1d6b4d3c18e2a579b503df566797
F20110216_AABZQF laberge_a_Page_035.tif
a26373d07adeb20926f483e8f9b9d589
1fefc16e3999d6b0cdfd4596e882a4a98ed9f3e6
F20110216_AABZPS laberge_a_Page_021.tif
6fcfcf4fc97e32611aae29c130da8d72
aecee42b288da20b6eb79931284c1f1cfbcc09a3
59230 F20110216_AACACH laberge_a_Page_111.pro
cc5473ec8055b84a7d5b1797423216b8
ec56c1c23aee39b707be159d8fb216c9679adcd0
60210 F20110216_AACABS laberge_a_Page_094.pro
36d509cc666c7c94945938e1bff4ef24
319983f9a5cb4f7a7854c676fb5cb830dc795201
F20110216_AABZQG laberge_a_Page_036.tif
cf41b6a11fb10e1ff9786d8d9c403e04
cce774d2a13028723def85f09d16e2cf49c9e31d
F20110216_AABZPT laberge_a_Page_022.tif
f54baf9362b19605812827450be13da3
bf5e9e5f0f01f942ab16cc8798feec34122298d4
64238 F20110216_AACACI laberge_a_Page_113.pro
a15059ea4d410fdc65df109d7b52e968
6cb9f90a6e586d334e362f55682b9d4bb783ee53
58847 F20110216_AACABT laberge_a_Page_095.pro
7d2f74be1e647da639aee90df1455adb
94957bd0e55ee41cabf5d24b9ee5bd257a2f7a9e
F20110216_AABZQH laberge_a_Page_037.tif
f76ba331bca4e0497c984fdf04d19e70
70982db3e6ed868b9ff007586f1eb655d2dadf8f
F20110216_AABZPU laberge_a_Page_024.tif
73c305c89567287c588bdfd874db90ea
3c000a464269a009e9fe14890dd45ca2aca0133b
63026 F20110216_AACACJ laberge_a_Page_114.pro
a21677a1741269e8e95e11301f673119
3a78ef999fd0e790ba91c9609352890bab9a80ef
58270 F20110216_AACABU laberge_a_Page_096.pro
086a431f84c2a605a7ec377d54019a3c
ca616f0182b8f25948c1f7dd00f5d552954daf9d
F20110216_AABZQI laberge_a_Page_038.tif
b710e97be17f4f0bf2f3dfc2d427ce7c
52ee449504a410f88c0f3e3767f22e6a856b05d3
F20110216_AABZPV laberge_a_Page_025.tif
6f6e0fc68b0d7b6a5b28f057b336b05a
4130a4854504ff8c8496c243daf700f57bfae34c
59452 F20110216_AACACK laberge_a_Page_115.pro
ae4adde13c8903a41e68a05e0d9255f4
b7581ad59da9d9df6054dd3b8fb15b8ab9d6e0e3
56972 F20110216_AACABV laberge_a_Page_097.pro
9df11a96787296738c14d1579a03c3f3
62e1bba0e1a353eff07d6c5b27e96df502f5ede5
F20110216_AABZQJ laberge_a_Page_039.tif
59c7bf8453c5839fc159292b5875ae55
99d54ea8691f57a799f3de6cb3f6ba3e5346a319
F20110216_AABZPW laberge_a_Page_026.tif
5177eed1c035eac03f8c24aa90927a3b
9279307455a38fcffeefa86ed8c8764db9ae2156
61129 F20110216_AACACL laberge_a_Page_116.pro
fcd0309ccacc55e6cc3ccdd9fcf9d95f
1295dd5515e48a3bc02dceb9938186680eff6321
30271 F20110216_AACABW laberge_a_Page_099.pro
e9ab80a892c334621b2959787ca48ab2
92abba4028a670e99c8533d3925e641be24715fb
F20110216_AABZQK laberge_a_Page_040.tif
234cb0d6ffb06e02077d47ef54d7ef7c
3319ee2041d408910ad1793a3d271842ab5414d8
27353 F20110216_AACADA laberge_a_Page_001.jpg
181893f9154a48fd00ea3c87fc17f5d0
8b40928e43ac4345034f6eabdb12de30592d1c84
F20110216_AABZPX laberge_a_Page_027.tif
a5b14a53004657a23cb461fd9cac387d
7074fa41772978e0f57a67fd3fc5d03c609b8073
63002 F20110216_AACACM laberge_a_Page_117.pro
3e75275ed23bfe2a0b70ed8e98cc7fa6
dc13f5a56b2c74ea08a560aa1ead134e3d9d63fa
F20110216_AABZRA laberge_a_Page_057.tif
997eac6b1b1606a5fe4b7e94fa025a07
30da30183e3e9fc4dc206d49c7bab416510bd29a
36352 F20110216_AACABX laberge_a_Page_100.pro
f5871be8d06a38eb4a5d61b31c375ca5
a126b54605fffb069d4e3f644983e6ff677cb16d
F20110216_AABZQL laberge_a_Page_041.tif
2b59aef41b4c5ff6748b40c71b870274
51a2fd92d238ae6237274686bc572a94de762dec
5856 F20110216_AACADB laberge_a_Page_003.jpg
aba6ba962856f112cd281a5c9f85ff92
a11c13ed7a841b52b7a566ffa89d728eb99a3f68
F20110216_AABZPY laberge_a_Page_028.tif
37531a8c8e48f5558b0602a3b8d45483
327a0a3ecc1cc4b87c9ceda53e4101420f903e49
69205 F20110216_AACACN laberge_a_Page_118.pro
b28d2116ad6986dc0a05d5e12e573302
a30ee542511e1290a17d9afb8cbb4b38f9369900
F20110216_AABZRB laberge_a_Page_058.tif
b4180cba02f0ef30da3ad39f183536b0
c2816502e1bdbb46ba7d9978b13cc34cdffdb4d0
5773 F20110216_AACABY laberge_a_Page_101.pro
9af0799204ad11e59ee38aa3244dea7f
c6f9add597a7e7f6b77cd707770715830f18d437
F20110216_AABZQM laberge_a_Page_042.tif
a7e50af685b78c3ee9fb90afc61f79f6
d0fa08f327c196a1b5586b9d23ae9e2fc367f481
33435 F20110216_AACADC laberge_a_Page_004.jpg
7faa9afbde478ca7d8c7cf5eb3c6c76c
537f0513de0982d92fdf29473afecf0d9459077d
F20110216_AABZPZ laberge_a_Page_029.tif
477d2b8eee496c32bea6e8420fbfccd8
678e8a799734e1a74676ef588e5dae4768cba398
57966 F20110216_AACACO laberge_a_Page_119.pro
4e806aec240af2df61eec89de631af64
114271608285666e4c287f9914b8a93c3574d668
F20110216_AABZRC laberge_a_Page_059.tif
1b43e571c33d5faeccf7ab642b1d1bdd
51e40f1508e11ab5131caa36f505cbcb7a4ba419
104161 F20110216_AACABZ laberge_a_Page_102.pro
38266e703f8b7739b235586ef606749e
097920bcf209d73ba70cda7d7b674072c6cd571e
F20110216_AABZQN laberge_a_Page_043.tif
46ba56cf9eab1b582b538f57232ceb52
8a182176339ce5c96a2a710586eb53cd8023b6ef
91272 F20110216_AACADD laberge_a_Page_005.jpg
32910d2b62b3b5019e429f4c0a2af856
00cc3f3e769ec13b3f3f59cae88c8781eaa7dc9c
63626 F20110216_AACACP laberge_a_Page_120.pro
b600b0c1aa402f334fc5435a9fa9628b
6c8141fb760e0eca38e2150c446c9b17a7e8fe30
F20110216_AABZRD laberge_a_Page_060.tif
f12c765a1b0b627121e5c3a2f9a079bb
f7375a739d2172a37e36e6f2387a2d068140b9ab
F20110216_AABZQO laberge_a_Page_044.tif
355ef9eb6ab3bc960cb2dea47acd452e
bbc031c2b51016c1f98361f2553b9556702f66b4
123504 F20110216_AACADE laberge_a_Page_006.jpg
dc62dffb7e11fb062f56447526f19b4a
f44cd5290750f348f1bdb14a0cb7acfaa7e3550b
F20110216_AABZRE laberge_a_Page_061.tif
9bd885d5196a6269131be108cfdd5501
2eeee097dc52fcafbc1c35af6a643b084e8cac0e
F20110216_AABZQP laberge_a_Page_045.tif
c0b92b561b54d89079ad946764f783f2
9d7bfbccf9936c1e228c9777b7ae3c512150e823
47048 F20110216_AACADF laberge_a_Page_007.jpg
f2a11839aaf33001b616a1eaa0cf107a
cc47b4be13e27bfbfb1843150bfe1150a2ac23c2
62185 F20110216_AACACQ laberge_a_Page_121.pro
095b6bb3ab28188d672a2f2aec95888e
511fc3249e1347f66bb5b9d4aab7af326c5779e5
F20110216_AABZQQ laberge_a_Page_046.tif
cc048f07b003077c4d0ea37cc5d910d7
898536308976eb40c7bf56e7ed90d6a72acbe617
89021 F20110216_AACADG laberge_a_Page_008.jpg
8de619e3fc718c0d5ab5114adf7f74d7
be3be3df07edc4ce2897d413a5cd3e5affc2812e
59185 F20110216_AACACR laberge_a_Page_122.pro
3fef2e85500a26a4ce48a3e995b2e3e8
0a3b9c23b3fde3125219620aa2584f034cb72289
F20110216_AABZRF laberge_a_Page_062.tif
78046837e616d73fed2c8e8db70e2ccb
3c601633affc3ce0602dc593c072142a926c2a0e
F20110216_AABZQR laberge_a_Page_047.tif
b1979619bf2221d7b6220f1489d9560c
bcbeca43aa6cd00b075d80956aad58f0948c4e61
110206 F20110216_AACADH laberge_a_Page_009.jpg
0b528dec801f754dfa9774e96d114f53
37d454f03b7bb21232d096ba703028e85f3b9fec
22116 F20110216_AACACS laberge_a_Page_123.pro
93a328eee8963ef13664419d6b159ed7
3edc30abec1ea9ef2c4601c7b20d000b45abd254
F20110216_AABZRG laberge_a_Page_063.tif
daee3bffb6f16a30684b399a0d467daa
6abb0ed02c32b99b0e1f9b5cc7ff92dd7b49f8c7
F20110216_AABZQS laberge_a_Page_048.tif
ad8dfaec5cfeee3f0c386ba9e0197918
2414eced2dee6e5c9b79a1c4da8f15c1df00ff49
28431 F20110216_AACADI laberge_a_Page_010.jpg
83aa8db6605b123195bef65425df77d2
7f96da6fd812ae4727c471726623a9745cb0d369
27710 F20110216_AACACT laberge_a_Page_124.pro
e018432bb6e8cad812140ec6fdafe1c3
898049f9e98a2245180e83011869d8f59f97e037
F20110216_AABZRH laberge_a_Page_064.tif
cf8c9957eeb1c1c136d83be3d297721f
bb9637e89474e6b803a34304b855fa28009f634b
F20110216_AABZQT laberge_a_Page_049.tif
0c222770a0627bd8bfd06a9c9a53f063
d39c97a848014408170ebcce35318a380ae5f5cc
26073 F20110216_AACADJ laberge_a_Page_011.jpg
f7ff0c9db5d0ff595cfa8f71e930fb87
1f8df05523a733d029458492ab5076ad93f1e17f
63351 F20110216_AACACU laberge_a_Page_125.pro
208bceccde7b39ba59b3af0fb639ae57
9b05b5453fd78d29e33247d25668bea03ee44616
F20110216_AABZRI laberge_a_Page_065.tif
d8c57338b287eb13f557b6fdd8d2eb23
da51297cd241640747ea6d5ff1118ae99ad9301e
F20110216_AABZQU laberge_a_Page_050.tif
a43787cdad1f14fbbedb4ee9e04adc17
82d27f77dcfcf54322772a48474a90951bd20ba3
101560 F20110216_AACADK laberge_a_Page_012.jpg
0083ad9380176ab9c3b24d3fbc639ebd
9e530ddddfdd8cfb8adb19e932d5dee2fdccd7fc
65872 F20110216_AACACV laberge_a_Page_126.pro
afb924c5e0d3720ae18ce5601b7cf1e3
955956fab063d614c5f5b95f2c257ca674615820
F20110216_AABZRJ laberge_a_Page_067.tif
40874125864e1566c62c93784220b603
c4b7b73df1f86379eb7e71f28797f31f41fd9d6a
F20110216_AABZQV laberge_a_Page_051.tif
65956bcb112133f70c87959e873329b6
c468d024c411b8055ce0a1f6761846cd108a49b4
69175 F20110216_AACADL laberge_a_Page_013.jpg
a5b111fd73a0dec7ef2f9e46b6817de7
271c03fc378ed4b4c1172822535f0c90ecd80ed3
73510 F20110216_AACACW laberge_a_Page_127.pro
6b1f760536efbae631f0997111eeaaa6
81b7f2ad703a540588f849de6cc84b155068336b
F20110216_AABZRK laberge_a_Page_068.tif
7bdafe10a1d89fd2bb0b4ba1afa90b3d
609708df9e74b1e967e43c441ebb4b6528c7ff15
F20110216_AABZQW laberge_a_Page_053.tif
52128d189470d55132a822480dc21bee
ef034413e4c7ab38669769ea75897c3685458c29
111160 F20110216_AACADM laberge_a_Page_014.jpg
723c7809ec68836b467d2c0133170ece
290a7d74a30a4f596d834457a33df3e3a85e7661
F20110216_AABZSA laberge_a_Page_084.tif
1504acb1bb4f93901fb6beae474a5089
3b5017771ddca48082ba1bf6080957679b671e77
70424 F20110216_AACACX laberge_a_Page_128.pro
81491548ddcdc0cecbf226831c1906ad
c0b99ceb0871cef5d9d8dd0dd6f75f19cfc10c3a
F20110216_AABZRL laberge_a_Page_069.tif
325ac2701e58726769ece0de2f7fd93e
aafc1778c298297890eafda836e8fb249952f17f
83925 F20110216_AACAEA laberge_a_Page_030.jpg
a231b7eb630b476b33fd7f3cfe981380
1d0bb3951e9b547334857a3d79f1a953410e73a6
F20110216_AABZQX laberge_a_Page_054.tif
c414ca724fb9394562628694acce2662
47da6735bdf46396d07e42aee39f870934638973
98667 F20110216_AACADN laberge_a_Page_016.jpg
4a3083f59d0a0d3273aa123dd2d70c9c
aafe3bfa19c5dd482766b5b0e2d3d0d8bd1e461f
F20110216_AABZSB laberge_a_Page_085.tif
6616754e25ea0fe50fa9683ba794cbf5
358c63031220aec996ffccddf75fc3e54ba99687
68898 F20110216_AACACY laberge_a_Page_129.pro
afb6f884420a3ae22ce075a8af532358
e5a3d08541d3a8e8ff314838cb782cd6bb991edc
F20110216_AABZRM laberge_a_Page_070.tif
56bd7513e6147dd783b8277d8ddd3e1b
0b4482b08fb5a2d600536af413d628140d183528
106324 F20110216_AACAEB laberge_a_Page_031.jpg
b27eb1f2d1e98c6d5d6668533fedda41
5aca4ac8bade9cdc34d20b0809d3c8512537897b
F20110216_AABZQY laberge_a_Page_055.tif
35bb4dd302b60b7b49bf515ed9e5aed9
60033986c897e43b0c6de03f6afcd33b0ac59bdd
102839 F20110216_AACADO laberge_a_Page_017.jpg
e133c61d9f34a676b6513b1bf0f10c2a
01bb7afb348708ebe4181712f3de822ca5d3dce6
F20110216_AABZSC laberge_a_Page_086.tif
31eaf8a8eb24b59e8efb81df4df168f1
74889d08df39b3343423cc4277c745c71d081a8d
22350 F20110216_AACACZ laberge_a_Page_130.pro
851e670fec50451830338b41d801bea9
cab08edfd96409114de796421c16e18bc8d7a461
F20110216_AABZRN laberge_a_Page_071.tif
622142bd9290bce7160302ccf5bb203c
137e3797b79623eb73e2948e72f7a8a01f294485
115006 F20110216_AACAEC laberge_a_Page_032.jpg
fc0d1a0920ae9e86f0f17b90a0299b72
dec045c62f99d8b366cffca6639bb30c79ee4cb4
F20110216_AABZQZ laberge_a_Page_056.tif
aa7b6390690e46c9f7ef5823a513cb8b
ed7d8f114161e93e8bb6ea1c5607fa266f0641ce
102632 F20110216_AACADP laberge_a_Page_018.jpg
92d7d5f9e4a3637e0ff1d05fd1d3037c
3e602dba222d3b9306ddd783b9cb629b3ea23fbf
F20110216_AABZSD laberge_a_Page_087.tif
ad9f590f6e5cfeb94afd4f4d82270d3b
6589620b7dd74364432cc7269c81df6b2a13eb66
F20110216_AABZRO laberge_a_Page_072.tif
6ca608d99d3cb18ed9489b9468c7f50c
6c8378ff522eac7d8f3f3603f6721d05e1dca49c
113182 F20110216_AACAED laberge_a_Page_033.jpg
4b2b6886e6c860f59fa4489bcadd4020
8a6b037cd0c7f2076d5acfd13b5f24fe8fdd841d
110201 F20110216_AACADQ laberge_a_Page_019.jpg
bbc93e704b0422eedff466016d79bd2a
e5ba5f63c4bdf64cb1385926fee73e385cfba093
F20110216_AABZSE laberge_a_Page_088.tif
23fb5c5de249645ccc991d51a7824f87
faacc90f63ab1583bd545d34f289f03977aab509
F20110216_AABZRP laberge_a_Page_073.tif
8a6e426d912f58549e59bc89279816b0
1bed48323dec3981a85cfe2750bad164b7dc5d71
107421 F20110216_AACAEE laberge_a_Page_034.jpg
8de2db20058d77371320a0aee39717b8
b200af703f2bc6456ed5bdcbda53b838bd43dfe5
F20110216_AABZSF laberge_a_Page_089.tif
c50264069efb635adea22a6a390b8642
f2b7bfff1b691e1c481acf71f5a074ef422d5028
F20110216_AABZRQ laberge_a_Page_074.tif
64e7de017f7642aee44f517a665ef187
752d327703d985e96773cf68cb0a061391e9d4bf
101401 F20110216_AACAEF laberge_a_Page_035.jpg
1b6aa09b31a040aa03e8a106f1724a05
56050a76ea753b2ab072d03874da5e3708b58409
103760 F20110216_AACADR laberge_a_Page_020.jpg
b44e617df1c12ff24eab080d051bf4b2
8f170f5373aa097a6acd86cc652dbb2ccc813ce4
F20110216_AABZRR laberge_a_Page_075.tif
409afc09a73dfa6c3b157e6471ccd79f
2e2a504a39df29de44f971c2871fcdedaa5d1fbb
77680 F20110216_AACAEG laberge_a_Page_036.jpg
045fcdd08cc0a3b9701c4369669b673d
c0527bb93f76fb1e2435feb9700fafbee478253c
103583 F20110216_AACADS laberge_a_Page_021.jpg
318c1df4ed02e30d4f25986bbb5e9469
397ab3c5f9a7cda2080121ad9f1cfa2c770fd57e
F20110216_AABZSG laberge_a_Page_090.tif
e98597ccd1249f60b7d7ee1172a035ba
d895358ee8e00bff3510821fdcdd58b8cc56c84e
F20110216_AABZRS laberge_a_Page_076.tif
6e12dd24e431034491424b52a4d86d9e
c3cca8cf87543ad64dcbab16134162c47a0a64aa
100343 F20110216_AACAEH laberge_a_Page_037.jpg
2ad27fa310946170e719d914fcf0cd42
cc96ab5b129b0585c8cd1d35bee546174f03979d
54361 F20110216_AACADT laberge_a_Page_022.jpg
3a498bad7394497739d856bef2284a86
bf47b4a00ec6b23e86698d399c4a3d470d32527d
F20110216_AABZSH laberge_a_Page_091.tif
f43fcb0676756a075c3798bf288ab7d2
3af044b2801b707de94ddc9a77dc6ce606ad7efb
F20110216_AABZRT laberge_a_Page_077.tif
cfb6c7e75d3cec0d135b6e72598d4de0
0d360c6726d2f97ee7207b4c1556e251a0f419ca
99759 F20110216_AACAEI laberge_a_Page_038.jpg
bec4c1425656051266f889c376343bab
215ed9573f3b6499b501317919a13c8c6794c569
F20110216_AABZSI laberge_a_Page_092.tif
4b173b7c2f27112924be3547b48d29c3
948f7983ca7f83e00fdce690a083bde0cf0746ce
F20110216_AABZRU laberge_a_Page_078.tif
85000f4cf84207ab7bfd70bda785f0e8
bec8e673779b0f5c70e90e2bf3e0beb7c5db65c7
102433 F20110216_AACAEJ laberge_a_Page_039.jpg
2c824016090272d7de839eb2bf3211ca
fb7f81fc7bb79711f54569f58cf3cbe5ed5a9f89
101635 F20110216_AACADU laberge_a_Page_023.jpg
67f12efa98456fb3e47d5f64b697a044
7bb7686e69900d0833896455d29d213d2dd38871
F20110216_AABZSJ laberge_a_Page_093.tif
140c73552dde24465b47d43ee7f2900f
61c255acbf7f7df41b65ab735b3c9396558b6eef
F20110216_AABZRV laberge_a_Page_079.tif
81ea11c46999d155e00646ddfbcbaa78
d7b272d4c4e7c742d7deb74eb8dcca40e0389101
109401 F20110216_AACAEK laberge_a_Page_040.jpg
8052b3248e01a689ba946fbe867d1886
e6e5974025a933a02c3c4a3f9a8060a0e1493de4
111821 F20110216_AACADV laberge_a_Page_024.jpg
491f58f807a17ebda403de64d62ac3f6
5a1a721abb6f773e22f60271a26a79725f3e730e
F20110216_AABZSK laberge_a_Page_094.tif
ac5872f8da0d0c29c01c49168ff7bd95
ebca0994649e4ea37fa2108afec51c09dced9ccb
F20110216_AABZRW laberge_a_Page_080.tif
08f1089c6a56553f3d9012a4a0c761d0
9f6a2bc3f82469b9699c8315efe7918274c5dad4
98718 F20110216_AACAEL laberge_a_Page_041.jpg
3aff9d8218a1e61eb195404ed80fa5ba
e01534b70dfad21497257b43812ad1af8db78738
104046 F20110216_AACADW laberge_a_Page_025.jpg
4d622da937129da2bf6e1cd5252c8de2
799342b298542fb962107ded46f4e1dc9b9b1abe
F20110216_AABZSL laberge_a_Page_095.tif
94f1f7d9577920495e33e65a6ecc77be
591df8916509dc642c169f8889df4c26ac659880
115958 F20110216_AACAFA laberge_a_Page_057.jpg
3c11723d75788e2f7d4402b1ab79237b
c88c2636d983530c81e75818a08cbbd90c228df2
F20110216_AABZRX laberge_a_Page_081.tif
67be35e7211fef8f6fa809b5b941165d
83c9074b8f8403eea5e56106082158ec6ccc6d96
16963 F20110216_AACAEM laberge_a_Page_042.jpg
1861d329b6cc0e7dce41fb27d45fe894
d55f525c519c3ff630260589014394fc1e266401
F20110216_AABZTA laberge_a_Page_112.tif
54810498b1737ade3634fdf08be69115
8c18a63b2c056483605d6dea739396afbc32a157
108745 F20110216_AACADX laberge_a_Page_026.jpg
8f6bd6eb776e4c0bc0cc2d10d197b4b8
9471daedd822d7ddbcf5eedbf8b2550d56a0d2e8
F20110216_AABZSM laberge_a_Page_096.tif
4d27bbdace85d2649276b2383fe14843
678e58378142edee3bb97cc52ceb4594e650499d
118161 F20110216_AACAFB laberge_a_Page_058.jpg
23d5885d2cf8f7aad327c651a1d9331d
125bf0d0fded14c9ae6d5c0b78d221682c9857ce
F20110216_AABZRY laberge_a_Page_082.tif
4209e209546f5279ff0226ba83bb1f25
8a77dcdf800c4e82140656077ef3b73c8c37e8b3
109157 F20110216_AACAEN laberge_a_Page_043.jpg
b83a65289fa0f9fc9d989dff85a22844
3f1f51395f757e2ad816f6f0c31ec20adc2cfba6
F20110216_AABZTB laberge_a_Page_113.tif
c8cb9eec1d9b535309171c697acd629d
65f2a64bb93f6164777550421ef7525322ef3f10
110616 F20110216_AACADY laberge_a_Page_027.jpg
9cb4354947f881465bc8b4d0c59e02e0
d43292a7293e6e2531765c28e2f5c0a73cf42ca8
F20110216_AABZSN laberge_a_Page_097.tif
b6c4c9ddcc744dddb6d77e1bc079818e
3759e96766a2a97b71329ea64e0a3494a06cd620
109922 F20110216_AACAFC laberge_a_Page_059.jpg
8b1090830dd6e1d664e624551ba32fa8
95ed885bb386427e68786d41c3bc72ea05e36371
F20110216_AABZRZ laberge_a_Page_083.tif
7e73fbe1f2a7a87b3bda5b26438a3855
cc23c521846b36a11a9a3a7acaf38937c505c5e4
110642 F20110216_AACAEO laberge_a_Page_044.jpg
059a6be761feba409fb52ae7cd902fd2
ad0a843f0e93dc581953bfd00944333c88f51465
F20110216_AABZTC laberge_a_Page_114.tif
aca26073e1b5381037dead3e95073690
a4b5a88bc0f2c16b53e9537b1715170a72c8bf47
114282 F20110216_AACADZ laberge_a_Page_028.jpg
025109009ec8ca44f2e647a312f330f2
6763984e9226cbbd430008db0465a3887e993aed
F20110216_AABZSO laberge_a_Page_098.tif
f8eb50a3e0f2a8e2610b69eca3b1e4b7
fd83a480bfd4659512de53b1ba36b317ffff410a
115751 F20110216_AACAFD laberge_a_Page_060.jpg
2c990d6aab853893780be6206c0d0673
f02bf3ca7ce06942b3e3d86670077c669c6b33bf
106153 F20110216_AACAEP laberge_a_Page_045.jpg
1a243449a8326b7f2f1b156f6c2d0ad3
ba58fe75f88691a41f3bb295ed0a61313ffdb52a
F20110216_AABZTD laberge_a_Page_115.tif
1e09617962fe8c564ec3c2504981d549
13cc4e5ac287b768a6e3aebbdd1f23c8ba6ae673
F20110216_AABZSP laberge_a_Page_099.tif
2b8b49cd651e1cb656d5e7b04be15180
4e8b165fa60f74399ab5e112d44e6d9013ff5a88
121102 F20110216_AACAFE laberge_a_Page_061.jpg
30356469b5d2c4364a585fbc8a174184
8035a072690b00340d994256ad69f98a88446d39
110451 F20110216_AACAEQ laberge_a_Page_046.jpg
ab59da7c3c7ba1c3f675390903bf2174
486d1573a70f9fdfb21c95796834a04e5c39ab72
F20110216_AABZTE laberge_a_Page_116.tif
c8ff6c4bdabdd333395efd62e45d83db
31a90cfce90afe9651f7f6813e7feebc8a46ac00
F20110216_AABZSQ laberge_a_Page_100.tif
336cfb7c31286e026113be8026128cb3
08a3535d64d9f72b207e3a6ac57304eec94ebe62
112424 F20110216_AACAFF laberge_a_Page_062.jpg
838e360784e1989262f003881dd1e295
a088c50c8ec8a99882f688a446efc46197659198
107613 F20110216_AACAER laberge_a_Page_048.jpg
0abd8fd5257aae060a5155194797b52e
09499c4ede0e128259c08c56e54880c11656913f
F20110216_AABZTF laberge_a_Page_117.tif
2debfbe704c37537f75c6f1e38367505
bb276e91dd0543f756b1e68a1b7268cf6fb9104e
F20110216_AABZSR laberge_a_Page_101.tif
d91fd56496d15847dedb3f336ac09076
04e2a1d8b7c8d073229b8326814302ff0b0732e8
93803 F20110216_AACAFG laberge_a_Page_063.jpg
1115af12e24ee3144867b007c8424fcf
a82c42ffede4a07c91fecda23e928e3d579e4c53
F20110216_AABZTG laberge_a_Page_118.tif
a3f7b155bc257b8eee0325352094548c
f55db300a586129f1f9b7a8915739ae276665ba8
F20110216_AABZSS laberge_a_Page_102.tif
b7551b1d03a638b22341d543974d45f1
e235310eab4356d0d6753ebe3016a0272c959334
107185 F20110216_AACAFH laberge_a_Page_064.jpg
a54d46371efd71b11df4266c9a1a4618
e346b0d66a114cfb4cfefe88714eae2490e3ab55
111199 F20110216_AACAES laberge_a_Page_049.jpg
3fa8eff51a517e9a5b6e1cb0be870445
0afdd7ee5ae7f1a25d169cc4fd06af0cf88328c7
F20110216_AABZST laberge_a_Page_105.tif
bbc7a2eb6e21a9cda3cde84468069851
df4f63596a701b662f00c8b7a115d079959e76ef
110345 F20110216_AACAFI laberge_a_Page_065.jpg
e5cae7474d5e80d8fb152f436bbdb981
a353050b1da8c0222cf7dbfaef971b8fe04d7cd8
112099 F20110216_AACAET laberge_a_Page_050.jpg
f7ce1ffe5d3b27548d1249e12f6d96bd
725df1d1b7f152717c81ff76ccf62da2803433a5
F20110216_AABZTH laberge_a_Page_119.tif
4d594cbca80ca45c6d85fff1f064abd9
81bdcb5118295cc18c16c76e15f32b04e6964e35
F20110216_AABZSU laberge_a_Page_106.tif
8e05d1865f88e2a32be3c232b591150d
8575e6bbff436752468b6ff2e202369295801d24
120998 F20110216_AACAFJ laberge_a_Page_066.jpg
2e6dea00688f9922b6f3a72c30e5e3da
c8d6355915b3c7197a520ca560377dba9ea7313d
110467 F20110216_AACAEU laberge_a_Page_051.jpg
33dd6eff7beef22ad056a2a458c5a6f1
c37b5e6b48a53b03a07954f44618f256e7f43d28
F20110216_AABZTI laberge_a_Page_120.tif
e1f176f717da13394a308051036c9332
ae30c4f66d7a780c6f395958d7ed7f4388a31c80
F20110216_AABZSV laberge_a_Page_107.tif
e8f90432cfa5cead66104455996baf48
02e3dd388d5d0a59c8269f69289464b3bf8ee044
116290 F20110216_AACAFK laberge_a_Page_067.jpg
f9619bd66c1cc1ef59d9a039d32e99d4
36b21bebeaa89de1e160038a1687d95c4063d266
104869 F20110216_AACAEV laberge_a_Page_052.jpg
535b11dcf472a07b9edcb7bd782e0060
f7676b87e5bbbcc9e2ca8d8706732f1787806739
F20110216_AABZTJ laberge_a_Page_121.tif
1bcc33e0cb5e138874792b3c31eac3f3
7932fa9030530be0a1e032b9168c7cb88b5c2ebe
F20110216_AABZSW laberge_a_Page_108.tif
bccca6e975560d92e14df23eea788ef7
30af7fb46763bd7c8a2cac24756c21785da9248c
119255 F20110216_AACAFL laberge_a_Page_068.jpg
26a8c1b93924f042059394dfa73d065d
c50d5732597c50dc07b1089bfa8f9594dccf38be
104443 F20110216_AACAEW laberge_a_Page_053.jpg
18c868a9ffdfd2e2b912471961942324
cd0120c00c6724f05c70f31d4609cedba4adac2c
F20110216_AABZTK laberge_a_Page_122.tif
cfd0cd88d40d5cf2b5909013a5d1db78
2ac826ac580fd465bef223bac05c065632728f56
F20110216_AABZSX laberge_a_Page_109.tif
48b3d18fcb2d2da0bead260d3e973406
0e4502f9eba6d5c51c140fa1ad3c2482f236fb3b
110171 F20110216_AACAFM laberge_a_Page_069.jpg
10dfbc97a91e8f008cb8b21528e46292
d43333008318a8050e34656a896e98c9ae81d9b7
627 F20110216_AABZUA laberge_a_Page_010.txt
91e272d15ad375db2da310c4cdc92364
fe38782009759d529953e4346e26ef7e7d260c83
113443 F20110216_AACAEX laberge_a_Page_054.jpg
80d1b73c38dca3fa84123e746e6218d4
d24c8c0aea26f98bfb45617ef357489ef564c460
F20110216_AABZTL laberge_a_Page_123.tif
e5e65a3993607dae06d1638e87964bf1
45d7ffb86fbb5f61c11d0ab1634834ebc75eac40
114035 F20110216_AACAGA laberge_a_Page_085.jpg
b963851da7e2c002b62e2bf744312811
0526734eab46954054f2ade900da7dbce370c1a8
F20110216_AABZSY laberge_a_Page_110.tif
e7dd28f30dabb986118c32b12635581c
03f99bdc4799183b08c97e2af1fd60227a679612
111181 F20110216_AACAFN laberge_a_Page_071.jpg
54435a8faeb25eed4a914b512ed64d4f
d3388be0292f50377b22b8aab4782e4b838c261d
783 F20110216_AABZUB laberge_a_Page_011.txt
058ed934bc4c6ea57f3cc2f25b7009d2
786b5d0104d4fd8c2ad1b95e5937bcc06e615e60
100311 F20110216_AACAEY laberge_a_Page_055.jpg
c80c45adb4b81e1c5b1ad5f837a2c5f3
b099b9e55a5302b6cae0fd999b74292aa57165ff
F20110216_AABZTM laberge_a_Page_124.tif
b9cd1db4915c90ecab42a6f0ee21e078
398a657acde3db6eadfe083fb071de849397c459
116534 F20110216_AACAGB laberge_a_Page_086.jpg
936f135f97e2106b0e2a50c92d06eaa3
435b9e28e7fd42372768673f9961434f3d73ba5d
F20110216_AABZSZ laberge_a_Page_111.tif
fa7fc945e82446defaf38c6dc5247a3a
71265f4d49d2565e0b94aff31ee44f933fc86573
108531 F20110216_AACAFO laberge_a_Page_072.jpg
5a38069b4ccc93ce987b71f21eed0832
351c21be7ffd9d0c15277306532847b644ddbc42
2063 F20110216_AABZUC laberge_a_Page_012.txt
68f0dbb9fd74f2f89431b8cfba74abc6
f3399c1444bd7864d59655db1652f3374062388c
102222 F20110216_AACAEZ laberge_a_Page_056.jpg
3cd97567972a69b23ebb8393ba1e77f6
3e9e348430d2bad0198966cbbc2c46305bbb29a7
F20110216_AABZTN laberge_a_Page_125.tif
9d11846d8672af832595ba993345aa58
fd6817cb113c930a81bcf6aafa224cf247de0448
21803 F20110216_AACAGC laberge_a_Page_087.jpg
a96060a3dcd47a556f2b283e27a8b807
ef63a345e62efea981781dcff46d248ec124f9a9
114840 F20110216_AACAFP laberge_a_Page_074.jpg
76aa14028bfacff1fc8109e434a0d72f
6d908c40956ce927e7c42aad5e78c10844559fe6
1337 F20110216_AABZUD laberge_a_Page_013.txt
ab752b553431fd8933728fa3aa8bb062
2a321816fd76bc16db6fff06b9f78ed4ddf4bd0c
F20110216_AABZTO laberge_a_Page_126.tif
be4159ed81cc7debbcf16e39a2ffa1b7
831995c4ed5fa31b5e5acb78e67ef7347365e99c
112955 F20110216_AACAGD laberge_a_Page_088.jpg
aa4f81f2e357ab6b16aba1f6a1350a73
366adaafd1312d8185774a4105b03fff7023f285
114895 F20110216_AACAFQ laberge_a_Page_075.jpg
d38ca27f50668fa2e095cd84381208b6
2f9f03f074bcaca46385aa88d58c642c65937f9b
2208 F20110216_AABZUE laberge_a_Page_014.txt
8e8fed2eb94ce485e811caca3e6fedc3
3bfc72bab1fd2a484a3ec40b4671e7920d197d56
F20110216_AABZTP laberge_a_Page_127.tif
4f7732c68798d163b8f13a9248b57f5c
72f1a14bc066c0605b4fbdadc1ea7a91b459e9da
122124 F20110216_AACAGE laberge_a_Page_089.jpg
34f436b6308106030496d8d03abc116a
55d5f92f8e66a792d5fdd6e7245ec070f28f676c
128250 F20110216_AACAFR laberge_a_Page_076.jpg
e7dd0071f9cd8e0ca77a986f81c3ed2f
e0874b6e6c4e3b8b54d341cee3e4e7990b618f61
2076 F20110216_AABZUF laberge_a_Page_015.txt
91d8a1e48e322a7b455494fdb4250438
2410d6d6d70898390696e82fba034d58931a432d
F20110216_AABZTQ laberge_a_Page_128.tif
b003907cef61d5319b59dabd7b5d4a84
17349be14962cbe7bb5bd4a40073cd4e467dac23
110858 F20110216_AACAGF laberge_a_Page_090.jpg
ba8fa54cd77c9462d3762eb6302f61dc
496fe643670a943ca037353bfe0a1b633133e465
120528 F20110216_AACAFS laberge_a_Page_077.jpg
26b0063c73d1dfdbfef4b4d262504470
4c84839fbeda106a9a00374e9057a16a23c8486a
1952 F20110216_AABZUG laberge_a_Page_016.txt
d6c2252708df50654c0e3967c101d0fc
41b254b03f61e2fb54ac3012cc6d1cd5d421bec3
F20110216_AABZTR laberge_a_Page_129.tif
f3fa65b92b270014c7448c3ccca12dea
078bc8b219197dc2c214ea3dc4b64725cc7c79ff
115175 F20110216_AACAGG laberge_a_Page_091.jpg
ff9480aa6ca73dbb8e71b688e4fef402
02c5d00690eb6f54445435435d1be8acc9d2c4d3
2025 F20110216_AABZUH laberge_a_Page_017.txt
4642343d1f6ddd50ee37a9dae938a6a8
87ab512b15862991c5053e75e58a0896f9d35437
F20110216_AABZTS laberge_a_Page_130.tif
f6ea543ade17fffb872de68ce415cdb3
bd3c6979da21c915278c63e9ee2390692230b4ff
118053 F20110216_AACAGH laberge_a_Page_092.jpg
18a0218d6063fcf1012189a2002c7d60
16be05e04a6e096a1a2e7330bdb912081eeef370
120279 F20110216_AACAFT laberge_a_Page_078.jpg
00cf29f37bb0c37c7c65f4cb415d4b61
0a8a37b31dd8650e7fcf60fdda49000a922cff92
498 F20110216_AABZTT laberge_a_Page_001.txt
76ec51f5e1ccf70d4720b839235ef972
425b39dd36da216766394f8a543554c318de7c45
114232 F20110216_AACAGI laberge_a_Page_093.jpg
5e99ac241ffb43cf1c85d0b9276cf4eb
5d06fbff0a8102a17d714e5f94d1079f85dd9565
118384 F20110216_AACAFU laberge_a_Page_079.jpg
9517f8b8dfbe2643cd1ab7c339425260
c8a26f0b8ae1f53e97a147c7ae51fb60c2804ed9
1987 F20110216_AABZUI laberge_a_Page_018.txt
4cd7ab830e70c9cd3039fadaf63a104b
e36bab1786c1db931c2f4f66b883d8a4ff3b1294
93 F20110216_AABZTU laberge_a_Page_002.txt
e8c3d5d61e6a3815d4b86eb0ae3f6cfc
6ea34762c10deb5f404eb277f57e461067f63b87
120022 F20110216_AACAGJ laberge_a_Page_094.jpg
acbf77debf6bca3dcdc7330c217ac701
81acc8b12cd5d639ba38277c0bb42a093a1e4b76
114297 F20110216_AACAFV laberge_a_Page_080.jpg
c02224c3e2c7c38ed8a083689745a5ec
2975b153050612e536128f9e5a5a940f723ed72e
2131 F20110216_AABZUJ laberge_a_Page_019.txt
91d8d1fde0f2e6f6682e4a7da5bd28fe
af9ef1612dc62222f1838a1fcd4d4a3f6a58a997
124 F20110216_AABZTV laberge_a_Page_003.txt
46944422287d3da74a4301572a506df2
10f0f7f65dbb7adbd35b2dbee5105d8beafd7a70
115657 F20110216_AACAGK laberge_a_Page_095.jpg
64f08941cd038ed94cc8fd87bf2be1a0
dfde96163efb77c65b1a1359d6c025c94a2d9aa9
91457 F20110216_AACAFW laberge_a_Page_081.jpg
72a34d176ec843df23b3af6490b540e5
0d6789351bfd75d3d05b7f3858d0ed2bf260c617
2029 F20110216_AABZUK laberge_a_Page_020.txt
d853349dc631088e5c4ea41ed109fc85
cc8cd9c7286dc78939e55949185e6b0ca4019631
4134 F20110216_AABZTW laberge_a_Page_005.txt
ce94d32616bb6454c8ecf6b77f127548
57677d2a42c5588f330489df07b9545179cb5585
113379 F20110216_AACAGL laberge_a_Page_097.jpg
34f12a69663714e50e3026fb24a86a2f
327b390166c0a42e46cdcc5a4cf625615e208fe9
2026 F20110216_AABZVA laberge_a_Page_037.txt
c303b4389897139b55cbdd49112c2102
fa0c04aab4339b80241568e110842a7f678f4e78
117990 F20110216_AACAFX laberge_a_Page_082.jpg
3284623259085e71ffacefa8c3e6bd05
7af329b1853ee76f32e088125f2a81f8c6a36d19
2064 F20110216_AABZUL laberge_a_Page_021.txt
f9d58e3c2e07afef4abf5e119b1b6c83
16c822a483e19d69290781e4b41328c7d81e26f0
100788 F20110216_AACAHA laberge_a_Page_114.jpg
f0d646f8f77d1cb5ea3248c7df2ad502
2c13e2b313df33de64ce31eb81df97d5fdb29204
5193 F20110216_AABZTX laberge_a_Page_006.txt
ee0212fe326647d4508b00a6d6d976ce
475a0585848a1b0d90df9a11e240967a30e4774e
113269 F20110216_AACAGM laberge_a_Page_098.jpg
ab30ecf64b5cddbdad5619079ab550d3
5bc8e38ec3845982fb6c1931a55ab5e4c799c149
111746 F20110216_AACAFY laberge_a_Page_083.jpg
78f73953efdd9277f5d8178d56d857ad
713c7ea49245594137457faae7d69d2f0e6ac238
1121 F20110216_AABZUM laberge_a_Page_022.txt
14d0da058a1ac5ee539172d837995ea2
14e56b4a90d76b6a3049b13083006214d13bcc65
98471 F20110216_AACAHB laberge_a_Page_116.jpg
8e2f72ddb695b071fd299cfd2c4e5c86
4a9753304809e8ed511c47d6a329275cc7f0bd4b
2289 F20110216_AABZTY laberge_a_Page_007.txt
1223ac7e1324db2e3c7e2604e4d7878c
3785b7f585d25fd77ea8097b06fa12bdaac08a02
62922 F20110216_AACAGN laberge_a_Page_099.jpg
48ee6c3dc792ead8f3919ed07fe1d31f
ea9dda188c28fe99d28bfc112ff703f2a03345e2
1973 F20110216_AABZVB laberge_a_Page_038.txt
5893c0844ff0bf5af0bcbe569664de89
f7dae230e70819c6bfbd53bb612e6ecc963c8536
120171 F20110216_AACAFZ laberge_a_Page_084.jpg
3f534186d0220088f47c741c58a3c9b8
51c4d2f2a96ffc638ed92b2b30e598c2a1c551f7
2088 F20110216_AABZUN laberge_a_Page_023.txt
6d661af85768eeb2c4eb538f82d244e0
7cc88de62ecbdcf883c3ad0388cd3878acebc139
98516 F20110216_AACAHC laberge_a_Page_118.jpg
cee98633106b5188720f580090ebad5a
a8e7af320196c6b6847a9f493e5ce155c085fd62
2740 F20110216_AABZTZ laberge_a_Page_008.txt
b0cdc612621aa55f4adb82e39e4d6790
69b7b84d90754f86e67cda09be7c073cabee45c9
73569 F20110216_AACAGO laberge_a_Page_102.jpg
2aeb405fb320cbd93ad829b7e445f146
9382c9afeb6aea926319519080effd4d243bc0e7
F20110216_AABZVC laberge_a_Page_039.txt
903381827196e6b27058beced1ea76db
c4b0a79c94932bcbe87fa1fc7704a649f2c51195
2201 F20110216_AABZUO laberge_a_Page_024.txt
7b635677d62f2d604bc482ef5bf3e2cb
2122745f814b33e71988eb99b9ed4bf468e28fcc
98281 F20110216_AACAHD laberge_a_Page_119.jpg
4a28961e588212f09705d523504368b8
41e3fb7e546cf3d7668282fa27f70187f44f1c68
50423 F20110216_AACAGP laberge_a_Page_103.jpg
087bd2027b6f02543fb04ac02553ec17
1504e5b318cf5bccf3d6933c3659e070be646ec5
2163 F20110216_AABZVD laberge_a_Page_040.txt
ad02a5551b3d8c82f2135b0e874fde09
64ae7c5164300b8f681de0991ba16dd53346153c
1988 F20110216_AABZUP laberge_a_Page_025.txt
62e517ed9bd049a8a0030a688c6a6beb
fef00afbfb800eb4749bd6b571d4644df1cf5e0e
97906 F20110216_AACAHE laberge_a_Page_120.jpg
c5a873d7e0bf88a94b7bd05296652c86
0ecf35c5a42b5c6cd6aa01461cee4e108fbf096c
67922 F20110216_AACAGQ laberge_a_Page_104.jpg
9c5fa4a8772425dc3913f260bbcd50e8
2d185e39dd4b759d272a76c68afcacc36e73e9b2
2036 F20110216_AABZVE laberge_a_Page_041.txt
90ac21cddec1af2e87d6b2817c4fc789
54769ee95908f160d26068658da38abf83660e31
2153 F20110216_AABZUQ laberge_a_Page_026.txt
2b4d00bc52292f9d5d39d09d7f74e356
ab84e21121386d932a8a84cf743ac841d2dd23ef
100560 F20110216_AACAHF laberge_a_Page_121.jpg
ea6b6fc45f1ff11f49523ad0e700dbe7
3b730ca1d52ee5a151b729153b6aaf848c3c91c1
49580 F20110216_AACAGR laberge_a_Page_105.jpg
6d5083b3814e687aaaf300f676fed0d3
700fcaccc61770e9315b8bb800d22d16cce22b06
296 F20110216_AABZVF laberge_a_Page_042.txt
ae83106efdda76e87253f849c8af7013
4eb874cdc3c6f903a9468490e6055545a24d04b5
2184 F20110216_AABZUR laberge_a_Page_027.txt
be08d249b983784e4cdb3b4fced4b5e1
456010a6f37744f85feda9f40971cfae55ca9988
87429 F20110216_AACAHG laberge_a_Page_122.jpg
50bf4918022c27df51a0040f1e9cd9eb
f536c92dde932acc0d95ffb02193e5040fdb69f2
64880 F20110216_AACAGS laberge_a_Page_106.jpg
dc2d5babe8bcb71694b87503724b360d
9f327ebbef0283d0f72bb38f51df88b2aa5960d2
F20110216_AABZVG laberge_a_Page_043.txt
f94fc5224e858c8220863ffa10f232c2
80a6ea008704fe65f8c000f77d98aba8de10cfc2
2260 F20110216_AABZUS laberge_a_Page_028.txt
12407d123e9f30ae15c0e8b04d35c196
4418902b3a10c723a7e3daf30f25b0d05f3d60c6
36195 F20110216_AACAHH laberge_a_Page_123.jpg
df34e4b83fc52ad0076ff3dc541037a4
6194f068368e8e21b633fcbd10cead8fe3dfaa69
98640 F20110216_AACAGT laberge_a_Page_107.jpg
6222568e34d95ef2194d0ed527f5144c
7a1f776c81373db66fb028773bd09b5d295d7ca0
2247 F20110216_AABZVH laberge_a_Page_044.txt
a03e52d8e27ee3cc8fe0aafb4327f087
d769c82bec0d4cdce8facfb94f44f14a4a6264db
2203 F20110216_AABZUT laberge_a_Page_029.txt
491d376683995c551e07d61ee8edfa00
d9ac6a57fed21a36a22883030f04b095a5ffe9e0
32194 F20110216_AACAHI laberge_a_Page_124.jpg
94bad27fdfe7c1230de3305777e5bf90
44b873f0c8ddcf37ac6433714f2ac2783847b9db
2134 F20110216_AABZVI laberge_a_Page_045.txt
af7de32d8f3fb9468226e681ca615abb
181190414e3359572ee5c6a07f4da877ca7afa31
1831 F20110216_AABZUU laberge_a_Page_030.txt
b48bd0e4a6cc2493de05e968d6db21b3
bedd1812daf9f24882dfccf3109889d0ae79dbca
117541 F20110216_AACAHJ laberge_a_Page_125.jpg
70523fe72a751e372d8d4d71d215bc8c
7f1bec1564addfa2f912700c6180c2fbf4db9362
106964 F20110216_AACAGU laberge_a_Page_108.jpg
34e4297388fc088c6367a5f5e6b4745a
2025c72652b5a896f48768af36d7f9bd8430ba10
2242 F20110216_AABZUV laberge_a_Page_032.txt
a201feedc42403e97186d41cfbed6471
dab7f6a8876235bff00780d2339695960e622017
119834 F20110216_AACAHK laberge_a_Page_126.jpg
bd1d0e912ecae8f008ab1a603e86a18a
95d2847141bcb3de766890de08b8f2bb66064d30
113161 F20110216_AACAGV laberge_a_Page_109.jpg
d0e4609a236c5ed51d6d3093e6161aac
475768f3207c6d59041541f2f85a9d1e99941d97
2209 F20110216_AABZVJ laberge_a_Page_046.txt
c8566f4218a0d3661e4a40fc8ff8a967
712b6ead0efdc49a56b4ffdea105e94255d1512d
2222 F20110216_AABZUW laberge_a_Page_033.txt
5e507940d71c34b1eb884ff7874ab432
4041dde1ac75fc455b23439c4992f34937031ad9
137078 F20110216_AACAHL laberge_a_Page_127.jpg
26c764d40e91518233dd17298b57dd16
573ccc61369a947c59b207b9e504793648bab391
99613 F20110216_AACAGW laberge_a_Page_110.jpg
e95393fe07b719e3b8f3d5394b3cf6a0
c1a6d217b0555759ef2ca95d1a1255fbcec66bfd
2123 F20110216_AABZVK laberge_a_Page_047.txt
b875fecdff1ab4aedf265e46b8cbe8b7
f563ef19f83b3ff2c2c296b3002f00d95b444ebe
1051979 F20110216_AACAIA laberge_a_Page_015.jp2
eadc54175d41fdee32b98631ce477625
c58270bce6407a328df3aa60a162e78116c584ce
2126 F20110216_AABZUX laberge_a_Page_034.txt
91c7bdf470fc07a18ba0d27ecd254abc
5c58d66d0fc7497f232b6e64bffae2c23ea24f7d
128529 F20110216_AACAHM laberge_a_Page_128.jpg
196b350ff834e2d61cf27a43f7c6b71f
dd8c3a351d9a5ba36e8a1978eb4f2a156187c119
2213 F20110216_AABZWA laberge_a_Page_064.txt
a10aaa485dba95e888bbe0b63a149c7f
814d98eeae747657e5ce4c141ad48d1b77421d5f
100335 F20110216_AACAGX laberge_a_Page_111.jpg
02f735c403b63736d77078bf9ec74dcd
1c587d541f888727b4ffaa595342cecdb3b8102f
2129 F20110216_AABZVL laberge_a_Page_048.txt
49381193bf74353e06bed4986e3f561a
713295f0f139ebb4a5e0da0ff723320d6b6505e9
1051936 F20110216_AACAIB laberge_a_Page_016.jp2
e312ecdce2dba7172687c0c96c2a7479
fdd248be353bcc4f8e17412c8032720bc00c0af5
2035 F20110216_AABZUY laberge_a_Page_035.txt
67d4f229662b1d16ceb5dc68b449532c
393c42004bd391c9d5f5bda9ab7e140f72d16496
127220 F20110216_AACAHN laberge_a_Page_129.jpg
aa21cbc225432b2f08b178303d76e4e9
fe4828c3ef41dcc6c7ccb2b9ca8b458831cd5773
2178 F20110216_AABZWB laberge_a_Page_065.txt
87afa58febfb836a2d521bd2d5a6ed04
bbfd273e966c932053939cfd41865b307c6e99b3
100222 F20110216_AACAGY laberge_a_Page_112.jpg
5aca8d52aa80a1c941d802bcb655fcf1
ca3763c85790d5ac1b2fdca059c13b0669e1bbf5
2221 F20110216_AABZVM laberge_a_Page_049.txt
5b3fb072b49bcfe43f63c311bb64a4d2
32472856bd031dab3eca1446db6d6bf2db3a850a
1051986 F20110216_AACAIC laberge_a_Page_017.jp2
1e3e19d55f1480dfe70f0a1fde65e6bb
368074c8295e3ffbafde8bcbff117c765b597ca7
1741 F20110216_AABZUZ laberge_a_Page_036.txt
ec26b97a36af2ddb5478c7a4d5af81de
da00dcfa53b979747d20d43106d6c9c3a73c8990
49379 F20110216_AACAHO laberge_a_Page_130.jpg
3b1a01026fb7d85fe8bfd37f44bca472
35dc721ac84e3396c96a34c85ba9f2b13dc049e3
2435 F20110216_AABZWC laberge_a_Page_066.txt
9479038dc525d05a63a517c92095f689
439889506fb8428d2033d000ded9ed8b6d0e3209
100979 F20110216_AACAGZ laberge_a_Page_113.jpg
2afc6f05c99802576e3720f9fdf6061f
4fe2d1935049c685dccd26476430078fae973a8e
2261 F20110216_AABZVN laberge_a_Page_050.txt
f8b88b48bc69631015ea7a7cee182444
cb705b15424d21c5b14b214e5395b60d1d8da3d9
1051974 F20110216_AACAID laberge_a_Page_018.jp2
5769626e9c52809c5c4ce7f7606aa85b
3c2e422f5529db7c35e7224cd2e570659040ae87
255986 F20110216_AACAHP laberge_a_Page_001.jp2
2f01771291fb916d91e982d17374dfe1
b076d307b80885e63c046c0b12d6d6c9a5b7d7c7
2302 F20110216_AABZWD laberge_a_Page_067.txt
e8101c5d84a1bb6586a7ed3f5befb6bb
7b8309782ec10beab714ff8b094476881a852f8c
2240 F20110216_AABZVO laberge_a_Page_051.txt
126a0d3197a3a070e0b43ed28d0622f7
5a53a248e3ce67d21913ff0ae088eb24fd78e68c
1051957 F20110216_AACAIE laberge_a_Page_019.jp2
3fc2ee1cb46be7f329ef28fb2cc27ea6
137049aec2a0a8eee74d4c9c334c21ada5556444
23685 F20110216_AACAHQ laberge_a_Page_002.jp2
54d52463637bff37143682fbd217f109
4bf22741ebbddafc94b5cc2562f551835ef5342e
2395 F20110216_AABZWE laberge_a_Page_068.txt
dd71eeb179c28fa618ee8797e902652a
8e5cf9fe616e52fb9822dfbf26c12cabcb5fb636
2156 F20110216_AABZVP laberge_a_Page_052.txt
f82cfd9e47cbb2d7f467fe2085745d19
42fa555757a4bf85e904c91756dfebbceef74be3
1051941 F20110216_AACAIF laberge_a_Page_020.jp2
120cd11509b3302eb7a93768a266bf84
dd7506f7709fd64c2aa806339d774e3e101906a5
332548 F20110216_AACAHR laberge_a_Page_004.jp2
04f77bd013fce31b29ec33c91b317fee
824744e1462be132523319209c81c56eaa51dd29
2187 F20110216_AABZWF laberge_a_Page_069.txt
c6cdeae9dea18082148b30153db37bec
cac7c5a1f2e229a896218faaefe4d0fa9eed8803
2098 F20110216_AABZVQ laberge_a_Page_053.txt
17abc299817cd8a5394e404db7b7de0b
31cadd7228f1fdddd252be753d797f998559dec9
1051900 F20110216_AACAIG laberge_a_Page_021.jp2
f4e1ff05e210041a04042a60c581f7af
fbb7e6346658b4f3ec2384ab128164d661cad369
934614 F20110216_AACAHS laberge_a_Page_005.jp2
92b230b098b8469f9d997f2899e41fbc
aad4a8c658668bac67a3ab8b0ab0644ba0b76bdb
1864 F20110216_AABZWG laberge_a_Page_070.txt
7c782345eb5aca37480b759da37082c8
1620d99c5d2fad6df0c5770f4d7efd6f5c75fb3c
2292 F20110216_AABZVR laberge_a_Page_054.txt
c3ae2a51354c6c1152b4bdce642fdeb1
7e3bc243e8fdb5ba551da6786739d81cc61b6668
585484 F20110216_AACAIH laberge_a_Page_022.jp2
ccae0ff8a82b00f4659c3d3cc618f99e
c6b0c98aed336b33da7cf71f0d7a0db9332a600d
1051985 F20110216_AACAHT laberge_a_Page_006.jp2
6f8595ae7c50094e624aae960b89455c
9597ae0813a256eb47a16317b13ebb93e3a768b3
2278 F20110216_AABZWH laberge_a_Page_071.txt
8ca788a3c5c36297009c63f250abb49e
bafb459cef85ec6d4f278d587007c2959709e545
1994 F20110216_AABZVS laberge_a_Page_055.txt
7a772945e58c47d5ae805d7977d62ed2
c05dbb7924df3d59e398e917783fecdaff2e79f3
1051951 F20110216_AACAII laberge_a_Page_023.jp2
0149416d2c89d5e1e0c059adada38d7a
9b6aa4ba1ca686829b6dde55fa40d3718ad75d79
445406 F20110216_AACAHU laberge_a_Page_007.jp2
b1c025762ce2aff96e7328b06a310a95
50e6ee7123e99268d8eb66932a281f0c3314e57f
2159 F20110216_AABZWI laberge_a_Page_072.txt
43dc01fcc3edcb1f7caad9ad8c770f9e
5d6a6a4258751a6c2a46c36cfe3446b0dd062225
F20110216_AABZVT laberge_a_Page_056.txt
f3945e3091b620a7159547f199d46f83
165b53d5f67cdf25c7415939d5f5bfa29be554bd
1051984 F20110216_AACAIJ laberge_a_Page_024.jp2
b3e64daf36ce56000e3b4c42a59d2252
c4eabb261393096500cfce48980eac8c3c91e6b6
F20110216_AABZWJ laberge_a_Page_073.txt
2e0942f0edf10a193f5052b5df2fcf8f
af6c4b25c83f506d912e4797959511d327739ca3
2269 F20110216_AABZVU laberge_a_Page_057.txt
14282107efaaf9f77de9ec9022eaa02c
7e231d6fc869d368d2349787eb983f9afc49e9c6
1051910 F20110216_AACAIK laberge_a_Page_025.jp2
ca5103b8b05967b91dfbed3bf09dc0de
2037387793d604d6c7859e45215c0e222cbdc6b7
965482 F20110216_AACAHV laberge_a_Page_008.jp2
cac22b5967f97f307308ba3c03b2777f
4a57077eb182773164c730e75a834ed1067e0987
2177 F20110216_AABZVV laberge_a_Page_059.txt
5d4c2264d8f1592ffeb098cf70dc4758
4b29c879ec327f9b044db5035a9597b1df4935e0
F20110216_AACAIL laberge_a_Page_027.jp2
0ae6133d3871e47d25bfa1354c23107d
31394651f5ca56fbd2ff70f2fd5501445bf99d4d
F20110216_AACAHW laberge_a_Page_009.jp2
a8cf2eb8895485cf95593b159e488c7d
0a4fbddc101a4696cd45d58be964d4d5d986c6ca
2392 F20110216_AABZWK laberge_a_Page_074.txt
d97813674e4c84aaf65b21b61785af68
28d349bdeebb75e33d3c38ecea845b03b6ccc3ed
2286 F20110216_AABZVW laberge_a_Page_060.txt
65bc5361f7591531fe06d490a4e15c5a
f6c7d6fe44e8ed1314c4e21d5822b4d442b15323
F20110216_AACAIM laberge_a_Page_028.jp2
66d258b364c7dfa6bc14beaf4908280b
69383356771dbf596ddd62bc70ec17b0524aa1b4
2328 F20110216_AABZXA laberge_a_Page_091.txt
3775957ab51c8755fcfae985cb4a65d9
fa0abd209282bc361e19bbe1b68f0d400bf1a11d
282332 F20110216_AACAHX laberge_a_Page_010.jp2
d62916dd6d9959d06a375f3ade1a9098
d4f988be096dea4bde88f8c57a1e1f813e4bc6dc
3087 F20110216_AABZWL laberge_a_Page_076.txt
7d4a4de33e93e5b9a60fcb1ef04d1237
25f388ddd74037dfca1b33a9305ecad6d8e46949
1051978 F20110216_AACAJA laberge_a_Page_043.jp2
414e5e1cd90e94a2e9d31c3479b00fbd
24766274b9ac614ce75c7fb5eca7c79b4d58d175
2451 F20110216_AABZVX laberge_a_Page_061.txt
238bcd4bd94300379652eccd4011a8b3
4726444b263da1b1ec5a42c8fdd2e95af510f97f
1051980 F20110216_AACAIN laberge_a_Page_029.jp2
77df21a807aa7010be338384eb81b99f
19c495aaf3b25bcaa6c6a9d3ebbaa7168cdd82b9
2310 F20110216_AABZXB laberge_a_Page_092.txt
564c9897d0bf87034acc9edcd771393d
3d8d488aa6585aa78ce41b7702f5be57c73731e9
1051920 F20110216_AACAHY laberge_a_Page_012.jp2
10456e181f7e52ab4590d1d4c72af939
c7f657de7cc0dfd544f30e5255b1f79bc8bf9c5b
2430 F20110216_AABZWM laberge_a_Page_077.txt
515a5b20d168054258499ef0dcecf51b
02052c44f614ede7fb972679fd8ffd66304a61a7
1051969 F20110216_AACAJB laberge_a_Page_044.jp2
12cd8c7bdb1233e0b10160a24127d726
794f1c67e22f4b5af55c468f8b90b322e9d6c4dd
2282 F20110216_AABZVY laberge_a_Page_062.txt
15d182aa917db7659e7ade16172e8395
960db0696d1ebf82d40902791f502c9a4ee2afd6
891171 F20110216_AACAIO laberge_a_Page_030.jp2
547d52119208cab726589afb6c1ab969
bafcc5c7bd0b28e265080b19a949139b91359a51
F20110216_AABZXC laberge_a_Page_093.txt
4075efb4f45f9d516436ad32aa7e0402
7e1608a835ff9c48750395c9d45e70eeff9de8e7
1051959 F20110216_AACAHZ laberge_a_Page_014.jp2
540507fe7c3bbc3c1e6d98c90d4aab90
bdafe605e1cee6cf83ea23e48a2843bd46eb18c8
3307 F20110216_AABZWN laberge_a_Page_078.txt
3d5eefd4dd0484f4e64bee42bfd3b388
b2939b56961a1d539d684073bc75878f5e032406
F20110216_AACAJC laberge_a_Page_045.jp2
3bf8d17dd62b6d89427722195b00fcca
aa3f36854edc4cab3ed75c50ddcb09743ca81394
1915 F20110216_AABZVZ laberge_a_Page_063.txt
1a0f36ea2bec3bf0192c7a1dae50c9c4
217d97d0f3e176667d80fcbe65e7392f4d0b56b2
1051965 F20110216_AACAIP laberge_a_Page_031.jp2
f7f8bb1b2559617e46612f6fb45dc9a4
d35cf364bddce5fed2cc9fa5da20363745cc47d4
2403 F20110216_AABZXD laberge_a_Page_094.txt
980323fa8efdf87f8e9f758b0ac12ddf
ade8ac5f36567cb2c38ee5db276c20de3544f65f
3082 F20110216_AABZWO laberge_a_Page_079.txt
333aba2417e652cc7380d02d0c23cb8c
3c7d26b61cff1142004078e1b1e388d8d14294e1
1051981 F20110216_AACAJD laberge_a_Page_046.jp2
7cab847d9f712bf91e94fbace519e44b
8a8084778946eb5eeb801a0d16698e4a28d6a68f
F20110216_AACAIQ laberge_a_Page_032.jp2
ce1abcd759da338635f4ed2444feba78
661eb7886d498b4d0554685fcbb784998e71fc36
2367 F20110216_AABZXE laberge_a_Page_095.txt
e96f7271b5406b85ccf6c37a67fa78fe
ee68567b37474c6d304d7152f00f72114dc8b9fb
2287 F20110216_AABZWP laberge_a_Page_080.txt
b31ad0df0453ab65189542a08ac559cf
b1e73271ba98737ae12bbd5235c4b13c14ec67d6
F20110216_AACAJE laberge_a_Page_047.jp2
fd9402c47589bf7891f340a73731fd79
327a69d0e0572afa14819d6feeea125c4c0f7e1d
1051956 F20110216_AACAIR laberge_a_Page_033.jp2
351ff06eecb1c6d7d6777a9e927c5ab6
2b4110e64b9294ab3cf7b269fa63016d4e617957
2348 F20110216_AABZXF laberge_a_Page_096.txt
aa98bbf0a64c9dc39c7619cc74a7972c
16fb2c5b48964c63d9bf5b589b0697f33f83aef4
1394 F20110216_AABZWQ laberge_a_Page_081.txt
60edbe518374ac4bff5c292f409bbe1e
c46fb08b31fc62aa16bee60db6bd38ceb56d2bd6
1051953 F20110216_AACAJF laberge_a_Page_048.jp2
3cc203c87c920eb156c10d05f880a688
9f1154a25aa1736785279a0c23d42af0a7b75d40
F20110216_AABZXG laberge_a_Page_098.txt
981949f7c7a3984708644b75387494e3
788157f4a558560b9e629cf44e3fea934f3fd09f
2916 F20110216_AABZWR laberge_a_Page_082.txt
a2b814311206f6281df5cf12d775c0ee
35592321c103c062a0c3ea4aed9f9056a2a99046
1051975 F20110216_AACAJG laberge_a_Page_049.jp2
a25a6c12b5e5fca0b5a94de2eb3c3269
8e16dcbacba9f56294550c579a6c84f297bf132b
F20110216_AACAIS laberge_a_Page_034.jp2
7cd2de3801426e6b5af764a49b0814ff
a1eaa27049fa144c086c1e740571731974d2cc4c
1211 F20110216_AABZXH laberge_a_Page_099.txt
1b7fd9ebb0e211f9ad8c794c2eaeb140
2117afa0ccd4935052774ecd957c44fcdee1c078
2914 F20110216_AABZWS laberge_a_Page_083.txt
c819bf682e1c645f628f99446e3ff760
b9bf6f686218fe47b18dd2f5fd5d0d9483e724a1
F20110216_AACAJH laberge_a_Page_050.jp2
8ca316833d7364b0304e30f64b24f0f7
63b2e96a05b7385cdb2027462915511dcb8d907e
809999 F20110216_AACAIT laberge_a_Page_036.jp2
1ab727ff65ef781f474cd07ad6b605ae
8c90bdd1c6babcce2556b82f41597081f9604369
2193 F20110216_AABZXI laberge_a_Page_100.txt
5535939f78bec17c4f0681d92ca32339
21857cdd9809f8d172379d4c4cafea025c63d6c4
2976 F20110216_AABZWT laberge_a_Page_084.txt
38fbc669ce4ee664b3552d4d8ee909e2
158dd4f8c07dff9c7a32acee77c74c5e2a4a76fb
F20110216_AACAJI laberge_a_Page_051.jp2
e8c2187f8ac43d3deb7250a2969a2325
d3bb2426d1b95151cfbcb9c563d36bc68e7b22d2
1051977 F20110216_AACAIU laberge_a_Page_037.jp2
9e0f927c040ee202dbea456d0d9bac66
262b63e8313e56f73e6e55c2f6ca9beea3a34ab4
440 F20110216_AABZXJ laberge_a_Page_101.txt
5ba8a81dedc0dba8c5ca296a7b25da4b
1766a8cb2efe35388f639c3543d50443632c9f4d
2521 F20110216_AABZWU laberge_a_Page_085.txt
77d525d271034ef7bbeeb73275ccf23f
de37e3c3fa87eb53ceaaa5994f0d583f82193111
1051902 F20110216_AACAJJ laberge_a_Page_052.jp2
e4c1f80aeb205d3387170cacd1c09147
2cc6d291d3aebd99ee4c48b662589be40d7ae418
1051968 F20110216_AACAIV laberge_a_Page_038.jp2
1325e45b358ef0bf6225355d7b7c6f97
56901718a6f05def2e4fb166278b1fb31ae02a81
5529 F20110216_AABZXK laberge_a_Page_102.txt
81170d7771fc59617184383b643707e4
8c902fc6bd9cdc8b427edb005ccec8585ed696d2
2648 F20110216_AABZWV laberge_a_Page_086.txt
eaa684d95d52f95599499db55bf57e84
8bf23c8bb34f3f2773fff0c45ce79c2176753ddf
F20110216_AACAJK laberge_a_Page_053.jp2
e8262e7fd13e735c243b6068d2402d7a
9533d6d731c187673b0a87f3b6b9fe676bfc68c6
477 F20110216_AABZWW laberge_a_Page_087.txt
b0ec49458800eb6fd0c8bd7824a6f696
e3187718a1000fd74012ee15d12763ddf92fcf51
F20110216_AACAJL laberge_a_Page_054.jp2
fd53bd16a9adad50643c89ffacb3512b
1b69a4f7e0a38d1cf04f18954730e03aef7a7d73
1051946 F20110216_AACAIW laberge_a_Page_039.jp2
495296b061f15c7996740614fcc80c79
3b6e74866faa4aed389c06307e76bc775819bcbd
3358 F20110216_AABZXL laberge_a_Page_103.txt
aac47b27ceb5ee59bf941ff0912ce578
59b1a370720d765804158379d015e0ed1a82f197
1022339 F20110216_AACAKA laberge_a_Page_070.jp2
079bdadc0ec4f461ccf7a86556700ebf
8eb9d7c7eafacd9cd6651b0df85691a4a7baca95
2280 F20110216_AABZWX laberge_a_Page_088.txt
b869bda1ae0c1860394dcff60edf889f
69c054fddb7774c12036be7cf8377cedfd4258f2
F20110216_AACAJM laberge_a_Page_055.jp2
a4a19d033f956fc087e6570b61a3b690
f811597f3aecdaf38fc6617bb1edb7211c51fe92
3552 F20110216_AABZYA laberge_a_Page_119.txt
5670015d60bf855ce7c5a9837a100701
ec9d54657dc0d971d6d787721412160d4e7671ad
F20110216_AACAIX laberge_a_Page_040.jp2
74439edcda8a8371f18eb312e5a136d5
be83328002e089273acff6124f8ea61bd883554d
3910 F20110216_AABZXM laberge_a_Page_105.txt
c2f7b52cd5d768ad560152ca573af877
18df2695389532a843cf900b314a9346e054a4fc
1051935 F20110216_AACAKB laberge_a_Page_071.jp2
bd03cbb361484d7f02ecf6af5af892f9
4538c238f660181beefa27a5773c0d71b973d3db
2444 F20110216_AABZWY laberge_a_Page_089.txt
b6797dc7d421c5ba583f73e5a54f76c0
7021ab1bb57fab3132edcd555bd665e44284b909
1051944 F20110216_AACAJN laberge_a_Page_056.jp2
08484930fe5dd9a3afde8368560b3ef0
73cc0ffeab59695ac853a223d5a53a5757888607
3673 F20110216_AABZYB laberge_a_Page_120.txt
c01f33bab0741ca92d8f19cf4f2552d1
b57d8159cadd5d02af6e33a91390141e0c024c73
1051972 F20110216_AACAIY laberge_a_Page_041.jp2
6ad3278fca5ea9f7829ae967b1560b6c
2a7735dedc3bb79b70f6a267773fd7cf609681e3
1595 F20110216_AABZXN laberge_a_Page_106.txt
9572e27ced48221782ef324fcb4db9c3
dfc55782c877370c50ed89de89ffbb6f1624a2af
F20110216_AACAKC laberge_a_Page_072.jp2
3daad91d263b6cd478a7c4b0ca72683d
dd9a08c68cf9a53bf1bc9802846e379996e9d887
2215 F20110216_AABZWZ laberge_a_Page_090.txt
b37e3ba500f0aefd9a7235a371b14d59
ef6de78ad8180d76ba3177ca3a563465ab42f1b8
1051971 F20110216_AACAJO laberge_a_Page_058.jp2
0227f07503449c11c6d9e4096d8d8e27
93b665dec0828231f312ba48162cd523afa9d92c
3720 F20110216_AABZYC laberge_a_Page_121.txt
376f622590b5042475761ffbed5a807b
e1a18a48521b598e4641167f234ce7f693114e32
162360 F20110216_AACAIZ laberge_a_Page_042.jp2
5cff9dfa8f893b83830864cba411346e
2a59ef1953d984f93a45502e7bf33cdc33bcd436
1913 F20110216_AABZXO laberge_a_Page_107.txt
37d1490b211d2fa057d7197c0da0f565
0b5e52dada6ad2fff9e8dca8cc547f84d8075165
1051964 F20110216_AACAKD laberge_a_Page_073.jp2
2fe8c39103a949cfdbd41d95e56f7451
7f6a068584e674a2f65b4e0fe0dcbfd218e82af7
F20110216_AACAJP laberge_a_Page_059.jp2
a78c64dda29e739d5c4faca6a10fb528
2bb304c7a5c8ed9046a571b85ffd82b5e764538a
3223 F20110216_AABZYD laberge_a_Page_122.txt
5e86faf03f19cfda596983aa33210af3
4d9a6ca212fa24c0e23c9f1469d49ca7d47bff61
3111 F20110216_AABZXP laberge_a_Page_108.txt
ec2f1f34ce44d8881f87eda4b5983616
69309ff35b0e384c9caa0f07a9738af2a1867262
1051955 F20110216_AACAKE laberge_a_Page_075.jp2
390d3735323e128b5921d5e3a80f4400
34ab8ba8750f22b747676eac500b753e7db0b7e5
F20110216_AACAJQ laberge_a_Page_060.jp2
79515ac069cecc944fe5de1102cc7ce9
11f317767fe650c3c605d71e02168f7c464be834
1197 F20110216_AABZYE laberge_a_Page_123.txt
056ddf8ee688e1c3af9dd44a9c72149b
7209f57af0bf012f741c67acc673ddfa640d9af4
4001 F20110216_AABZXQ laberge_a_Page_109.txt
1457f4516d10599445c5157eaa43267b
8a807e0235f278197170d517eb8127efc27f661e
F20110216_AACAKF laberge_a_Page_076.jp2
5a5e96b8bdce27c0b5f1f71c55e903db
753bc84fafea229e180fa40fa89c554df414274b
F20110216_AACAJR laberge_a_Page_061.jp2
de77028fa214e45b55fe41d80a1f172b
da2e0ca884ed88ff2b0df82e81c0af44b4359ba3
1370 F20110216_AABZYF laberge_a_Page_124.txt
6328b53adf23da6d4bba8b8c170d187d
432a3d3c19779fef9bd4aa0a9479e897b0feaedf
3787 F20110216_AABZXR laberge_a_Page_110.txt
6fe77b66980b0737a181d876ae2d71d4
7ec7cfac97f6c09e1bb5d61c25b34ff097589e94
1051983 F20110216_AACAKG laberge_a_Page_077.jp2
5e83aaf64a5190a006647f977f46e4a8
fb48f3e8359353ecf73d62cb394a8f67f1d2c733
F20110216_AACAJS laberge_a_Page_062.jp2
f27ee53363f775475fdc85ad695c594c
e36dc48ca8f598b5735373a860b74d9ddfbf7e49
2460 F20110216_AABZYG laberge_a_Page_125.txt
6defb3b2dd1a717897b1f00d87adffb4
cda81015bf1d8eab8fdb3c36e45f29cfdb6a8b3e
3047 F20110216_AABZXS laberge_a_Page_111.txt
82c14e3e5c12fc84ca1788835f23d4b8
05a9ff91c72ada6e43486f950af61cfc191a2402
F20110216_AACAKH laberge_a_Page_079.jp2
deb7e8aac88dde70ea9a0633df8c6f9b
ef7a0f395eca6774f9067c48aabfe7a63ad9ff8a
1039670 F20110216_AACAJT laberge_a_Page_063.jp2
9540c56948c3287c4e03ed5a6e1a45f0
36c81655ae02680697db0697c2c6cd47784e9156
2571 F20110216_AABZYH laberge_a_Page_126.txt
0e3f918c55eec859c1bc33fb5ca07365
81537180e287dc32eef4b6fb44bf0ab88988daec
3323 F20110216_AABZXT laberge_a_Page_112.txt
2a2e9ee2af75873d904fca64e2b195b6
3b9bea94c03b62f6897953bf4c15bf02a17520ed
F20110216_AACAKI laberge_a_Page_080.jp2
2548bb7a0a323c0968cb7297c215b30f
c4350676375d983e470fc44885409734192d48eb
1051887 F20110216_AACAJU laberge_a_Page_064.jp2
ecb9084fed56a565b3a66d30ae6370cd
931640d64f7966327eaf52811e21974dca5393f2
2847 F20110216_AABZYI laberge_a_Page_127.txt
981c06464c3406ffc70765381b0255c3
719f4e172b6121897d16929500a7861a88ca8b60
3449 F20110216_AABZXU laberge_a_Page_113.txt
4429210404b9141f411e9c70fa3dd85e
98db3aba2c9cb09c66c036a6827f7d158259e794
897579 F20110216_AACAKJ laberge_a_Page_081.jp2
d182f2fa3542c97c10d6ca17342b682f
e03928e48376b0ad69dc21ea671907f9f8eb93cd
F20110216_AACAJV laberge_a_Page_065.jp2
78e08892c4e2c471c16c30d36324e952
5cbe6516f9c0adc9c19e14760f5a90ea0249cea1
2727 F20110216_AABZYJ laberge_a_Page_128.txt
51b1d14a8a4a20c6ae701d919012f22b
d4914c1026a3c15e27873fef272b7c224175958f
3591 F20110216_AABZXV laberge_a_Page_114.txt
725b412257682c6a55c337f9747d9072
bda1a9e02136d7e9903d832fb826aef28a05b9a9
F20110216_AACAKK laberge_a_Page_082.jp2
3cef7efa4fb5ad56b9bb577f7ab1768c
c97e0657c936bd656f299a66130e2e1893e80c6c
F20110216_AACAJW laberge_a_Page_066.jp2
b053217f4be336e0e969dd26e5168213
4cd6be0611172ad7f3b8c38f025008d9a66b67c0
2677 F20110216_AABZYK laberge_a_Page_129.txt
bf3e07de40dbf36b03eee961383ed3ad
65577b7b62d4ec7f484edee4611eabbde85a8c55
3655 F20110216_AABZXW laberge_a_Page_115.txt
9d29092586f183153403fdcb7d8407dc
4303fd6ab9b2d69e47c2cdb08791f79536a1a815
1051958 F20110216_AACAKL laberge_a_Page_083.jp2
a24c5be654fa84fe84fecea44b29651f
73eae009a2efb50d67215da47844ba0d7a32808a
931 F20110216_AABZYL laberge_a_Page_130.txt
bbdfeddc8f271bf7d81a65b6edea0b72
33d6ecd19f1bfb4bf5db622be4b2aa7cbf09391c
3611 F20110216_AABZXX laberge_a_Page_116.txt
68416a36453d026cbc4e2e2bbc8c6f28
41aa2563f43ecce49615cdcdbe8fa2951695d529
1051948 F20110216_AACAKM laberge_a_Page_084.jp2
39972008e275d5d1ea1b9384a22ab957
f70cbb6d25f8b81de19497296d2b7ddac54add4f
51036 F20110216_AABZZA laberge_a_Page_017.pro
7190432d75db308f577bedfae59f0110
e025c05933c839c1363824b9352c94d77bde76a7
F20110216_AACAJX laberge_a_Page_067.jp2
f5d8ee0cb8de329444afab007fbf415d
d6e36c1c77d52a7b7deb31f614c21a90da19334d
680440 F20110216_AACALA laberge_a_Page_099.jp2
563ceb05d3e38889f1f572ff84bee7e4
483a701f944998f46496fc0da97f7617137b2dce
3600 F20110216_AABZXY laberge_a_Page_117.txt
7f0f8a89508db2c98e6b61a153ec184a
dbd5d25f55f275b2606c7565d88bd08bc3206f02
F20110216_AACAKN laberge_a_Page_085.jp2
10cccd10cc394ca7f53f06c3985d3d36
7628f33ea34030048f43aa0de0abb4a647b65f6c
50463 F20110216_AABZZB laberge_a_Page_018.pro
3f5bd125a0993c9c4b26142b99cadfe3
ad99d61100d5140aa3ff39b10e169c0bd2b52072
F20110216_AACAJY laberge_a_Page_068.jp2
7a93a059c80433cb202231c207077574
a294132882630e9f19ca024e7133d666d34e7dfe
8529 F20110216_AABZYM laberge_a_Page_001.pro
04c44e9a3fa9b0eeb0d21d64ed236002
0205cc23124f0226a5784c3e306ae2e73329bb1e
714287 F20110216_AACALB laberge_a_Page_100.jp2
a2865860d02df694c92ac2b9279d7a85
2e81ed41fa084e807fab503c1a78268f734b4b91
1051927 F20110216_AACAKO laberge_a_Page_086.jp2
f02aea66e46e9dc25ede439db5408080
aa44db429d1e979a8469c1626de3cb55add1d629
53501 F20110216_AABZZC laberge_a_Page_019.pro
fb804721b9d22c19acc5299bf2cdf573
add282ffea7a22085d500d1d5fbde021ffef28dd
F20110216_AACAJZ laberge_a_Page_069.jp2
51b03ace18fa905eadea33566c1dc944
f3a61cd9e704fe871c033c4298efb5f8aa01e0e4
913 F20110216_AABZYN laberge_a_Page_002.pro
868ccb0019d8fe0cacfa0a05a92f9c2e
c2279f36a5994ba1799809cd3246b11c785ea54e
163388 F20110216_AACALC laberge_a_Page_101.jp2
e18b3c69cd3cba110021fe91ba0a92b0
901cbc1be0061aaad3550a1d7a8affb4312298bf
3792 F20110216_AABZXZ laberge_a_Page_118.txt
46b92dbc4b18d1e81a87434ee4bdf875
40e729157173c2b18c917e3f73b09205ebeee1a9
F20110216_AACAKP laberge_a_Page_088.jp2
f214f56b318939ea77f5b2f3c0ab3172
531cfd82888ff0381878a8be4e48e09cda9bae46
51476 F20110216_AABZZD laberge_a_Page_021.pro
3f1b73143923777a4dcfbb50027e6b5d
17462e437025e1ca0b57d13fd2ebcca596197815
1717 F20110216_AABZYO laberge_a_Page_003.pro
1efcd359bb174aae07905151110ec75a
d943b614002134de56e52ca3b406567e28bf2c7f
F20110216_AACALD laberge_a_Page_102.jp2
62b1dd05cf0d46780bbbc9d60d3bd0b7
a70955e78cada0d15a49adb29f501ed1e9dcbd58
F20110216_AACAKQ laberge_a_Page_089.jp2
fdca594c7823fd936bf5a99be8d8470b
2e017bfcbaca53367d454b8a9c5d12e6f4f1e473
25391 F20110216_AABZZE laberge_a_Page_022.pro
dd3ce76785f9074fffaeeb26ea971800
c104aa832eebf9370b5d5dc3742fe1256de67879
14391 F20110216_AABZYP laberge_a_Page_004.pro
1ba6b082987f86c4a7107f1b5b2991e1
7d2023e48681476d6e991c31ca855e958b919f25
912761 F20110216_AACALE laberge_a_Page_103.jp2
baac08ce4a45056acb3e788e8ac2251d
cf5048f3eda52d521f59f99318cb3f3007d53684
1051982 F20110216_AACAKR laberge_a_Page_090.jp2
a0e763fa76c590420fcda8514bb5c5b3
df56a3988fc33cbc3e9f6a6d358a4690e8946412
50255 F20110216_AABZZF laberge_a_Page_023.pro
2b9497eb882e6928d888c99b9a2a06df
7417b02af3f9cc9b2b84deb5c21c2a88a86cfbcd
126376 F20110216_AABZYQ laberge_a_Page_006.pro
30aef05b0d8407a338022dde0021133a
2ca6418e66745b4267b78a47084582f6b80f6fdd
F20110216_AACALF laberge_a_Page_104.jp2
3bf192c5229aeaf5b5369b5f32e502f1
4251671a1766bde84ce367fbd6d4f09468c335f4
1051921 F20110216_AACAKS laberge_a_Page_091.jp2
8d0ae6069458d49154a455eb32560195
3378621fa251d38c54a02e3f8ba6962fafe75854
55939 F20110216_AABZZG laberge_a_Page_024.pro
2f32eb80cd5ffbeabff8e2404a8083b2
abf67f10ce43ef51cebe93cdcf09ed7a5c39a8b3
56982 F20110216_AABZYR laberge_a_Page_007.pro
9be33dfacbd5c99b3589c94333cc44b6
ce6bd2707107009ccbf2007184aabf9fca7afdc2
880314 F20110216_AACALG laberge_a_Page_105.jp2
70c5a70a9bdfb0333d3dea70cd686a34
7f7cb960258d4f9ac9eba34ea325b91303be0b62
F20110216_AACAKT laberge_a_Page_092.jp2
7654efb21cdc9d880c7e46bc01471ee1
94442e28609d0a9729c6499bc5094a2b0c641bfe
50327 F20110216_AABZZH laberge_a_Page_025.pro
2d1f3ef038fa0f809040f7ebcbde5877
02b6da7ee95b2f0dd4a9683d94f243b9362a77bd
68554 F20110216_AABZYS laberge_a_Page_008.pro
8df4a1a6d369f7b6a9b6455c097569be
13e2659a5f309c8becc34970d914a6c4708865e3
629381 F20110216_AACALH laberge_a_Page_106.jp2
03c4761f0a581b6aab6b4ca9e931b37a
add60bb2a340dfac6aa506b2431cfa434f75ae44
F20110216_AACAKU laberge_a_Page_093.jp2
fef7c97c76f491baf5127fd35a4f3f1e
1e0899a99a16c83369a15a2e5272520f93c04768
53845 F20110216_AABZZI laberge_a_Page_026.pro
2576db01c176a8c7a286792fc77b9dea
edd551f8c41ce66cdeb0fb0bb0a0c54df53e08e2
72211 F20110216_AABZYT laberge_a_Page_009.pro
b9a0762b1a9d0c9de4c014c00ab958dd
0b9ae8a949ce96fa3eda22887e359c0fc70cba8d
1049005 F20110216_AACALI laberge_a_Page_107.jp2
e91df29115ab8a9d9f7e7124436b8c40
1ccba2fe11d86b109e014ae6e31068f8675726b5
F20110216_AACAKV laberge_a_Page_094.jp2
1b9454de9edbb1b4dc874b09ef4c2d53
5d0c81ff2de3ceb7bd51f9dcc3c05c86d906a58e
55658 F20110216_AABZZJ laberge_a_Page_027.pro
658a5340a91858d166213c31ab33f2b8
fdbeafd29617f5262088754c19aba4101bbf0f60
15604 F20110216_AABZYU laberge_a_Page_010.pro
516ab2a0018620061a5d2f2716e30920
689f7a876d47c9b0519e3faa10597cfa21b18c3d
1051954 F20110216_AACALJ laberge_a_Page_108.jp2
681d06e75d181617c1688476766ed37c
8300f0bb41a570b8ba0fb7db55e9b62375effa7a
F20110216_AACAKW laberge_a_Page_095.jp2
ba495cc8e4b61e2c72863bc09b160e83
6e26a01464b25344b70ba9e6a024c12a94f61f7e
57063 F20110216_AABZZK laberge_a_Page_028.pro
392e6e530cb37a2b899009f20054df76
8e500807634d6e052b9f40a6ab889f3379f8fe63
47511 F20110216_AABZYV laberge_a_Page_012.pro
d2b845be4d844e2c809430bded0011de
b16c6c9f893ce9c089daa953f87c5d9d07479ee2
F20110216_AACALK laberge_a_Page_109.jp2
11fae9aacf1ee2359fef41f6334862b2
dd12320039e7aeb447ed5e77cb5ac846f1d192ae
1051967 F20110216_AACAKX laberge_a_Page_096.jp2
f2b7a3ee5cc6c2961dcf2975685bfee2
c8b33ec9916dda784b9914d8b829bcbc59bd73a7
56011 F20110216_AABZZL laberge_a_Page_029.pro
8116863433e3a4e99fe03f18e178fbe7
fdd7e44006a9b96c861df703396fd4459e81330c
33477 F20110216_AABZYW laberge_a_Page_013.pro
0bec8c9f8cee3d4b5faf81231e6a56cb
4149d83ba4470dfa3e48e2a3efac71e63241c2f9
1008861 F20110216_AACALL laberge_a_Page_110.jp2
a608aac91d8653999f3140a8c23efe10
27b0bc96ace36aca6ef9ef5af377412c19031209
42673 F20110216_AABZZM laberge_a_Page_030.pro
5691be25d834b08b5a04e41a8ebd4bf8
9a2ba7b04d47ee5cd0f31e3efe7e3a5cebc5255c
F20110216_AACAMA laberge_a_Page_125.jp2
01c9f5219ba031aec5be27b0cafcfc78
663db2ab595ebec28c3e7ed9d632df3538e46839
54196 F20110216_AABZYX laberge_a_Page_014.pro
f89a2892fcb2743f7e309bd8eec1a8df
7f1c1212bbdf26b0a23560058dcde9e8d3775970
1012279 F20110216_AACALM laberge_a_Page_111.jp2
643a1e3468cd09cd74905ab93ab79bc3
195128667e143dc2f75c15c4fd1e1e4c079080b0
1051963 F20110216_AACAKY laberge_a_Page_097.jp2
9d053f7a7bf3639de9eeea87e4d367d7
7f9501c4993c4ecffb756884cc35dd8bfc7f41fc
F20110216_AACAMB laberge_a_Page_126.jp2
ab6a36369d048e62852c048093ce54dd
3931f26b97ac1c1f8c1a283751b98f1b1bb62c51
52491 F20110216_AABZYY laberge_a_Page_015.pro
55ef8a14249d59a48b5e770f3026ea16
397738c088e476b6501965a6b26adbe1f94a4408
1006498 F20110216_AACALN laberge_a_Page_112.jp2
dbcf0ef657bb7c524978c605ac53e1b9
cb5b9e96899b1c9f74467c01450dec8939454098
F20110216_AACAKZ laberge_a_Page_098.jp2
82eef54ce452830ca04319ff7d284dea
c2539e3aacf2774d00ced2929d35c07e7a0bba2c
52803 F20110216_AABZZN laberge_a_Page_031.pro
46e1d283e6920f86af66eafd83e4d174
715939a9df970bf8ba170b1e9d2d0943ea1d53f0
1051940 F20110216_AACAMC laberge_a_Page_127.jp2
fd73f4596d84a06d380a0b2c37510f4a
b94b0f448d9fbe29cbfc3208c921b5029f81b8c4
49306 F20110216_AABZYZ laberge_a_Page_016.pro
0982bc3b9b664615149fd260f9f46f30
329af2e75742c3db4a27d627d658267a455799cd
1020003 F20110216_AACALO laberge_a_Page_113.jp2
b13ea70617bb9766a3558c1936a188df
07705905c2b9ee942643a3df07520a2a8003669a
57274 F20110216_AABZZO laberge_a_Page_032.pro
8060e2b92cc6934535a46549a257058b
41693803b897e3a29b43bfd4d58bcf0022d10be0
1051970 F20110216_AACAMD laberge_a_Page_128.jp2
4edb09e3e3c9970e31b0a80a1c78001f
ad4c5b4d141d85f9cbe355938886044d204bad7a
1042567 F20110216_AACALP laberge_a_Page_114.jp2
dee38387bb3441551b92c81eca77c331
d2d8eeab821797a064ea0ef026dbb1159a63b4e3
56611 F20110216_AABZZP laberge_a_Page_033.pro
bfeea4fc34426ccdeb06ea152e3fee9d
678ed240286ea6da72e0d264b57184f6ba9e5cd2
F20110216_AACAME laberge_a_Page_129.jp2
20c1803c260c87300316be3e260f5778
d784c4b11c0d13fae1faa55b9da1d1a7d8efc76f
1029356 F20110216_AACALQ laberge_a_Page_115.jp2
25dacc2c244b964483198ed93d333172
a79ed4eb82e0712e8bc58c3ca97f9359f7be459f
53323 F20110216_AABZZQ laberge_a_Page_034.pro
1b46546a75de0862a792227f00264caf
1c02d93d8ccb66b40b26309b9f1d78c6c6769ce6
516697 F20110216_AACAMF laberge_a_Page_130.jp2
c74a406fc15de4c90f5c6b0127c80a25
6a343bf8fd8bf68fd237ea9ee47999418fcd0745
1023018 F20110216_AACALR laberge_a_Page_116.jp2
a2a69762c8557285face4b1e7cd5d064
42af246f5a31071d5df4cefab1d172a98c8355fc
51480 F20110216_AABZZR laberge_a_Page_035.pro
4ce97ce11c7ad392bdbb675a2d0b4d2d
b999fe006f243033d2e7852086fb0feec2dc3b7c
1027133 F20110216_AACAMG laberge_a.pdf
91a5c96640804e0e68e8200113409ee1
d0e0286238f194e24b08d3608a9005b85b7952bb
1043898 F20110216_AACALS laberge_a_Page_117.jp2
06889caaab4d8b70c7384a33fd06f002
cb58fde5631a84e779f0c096b43c3344bb8985c2
35579 F20110216_AABZZS laberge_a_Page_036.pro
b135439a8ee74cb64c3a7d7dc02e70e5
1db02fcd273abd7ddd6b9b66c179bba26a500e74
35685 F20110216_AACAMH laberge_a_Page_071.QC.jpg
09eaea5e388f9e88d912317ac86d3439
0e1a41501f72f363ca4492acbc069372d372e643
1020374 F20110216_AACALT laberge_a_Page_118.jp2
e4c1128dc559d72f26123f08dcfb592d
caa4eff15291ef685c83a711f8a6c6fb4356a6d2
50910 F20110216_AABZZT laberge_a_Page_037.pro
f5cf70df0c274dd8208938dd15300c01
46a654b1a1255bd352d460735b90093eedc18ad9
33692 F20110216_AACAMI laberge_a_Page_053.QC.jpg
f6634392e27019420d3c845304616260
98f798100b24beaffe140f7cd46f14bbf90c2e18
1018682 F20110216_AACALU laberge_a_Page_119.jp2
ebd830eaaca8578f403c91927588932c
879e0327ddf9bca03224903b68456b397b798cf7
49992 F20110216_AABZZU laberge_a_Page_038.pro
633bf888bf0001cfdb949ebc430d7ef1
554b025710435051e98f11e2f65a2ceeab1efd05
33973 F20110216_AACAMJ laberge_a_Page_072.QC.jpg
b2cce1bcbd850042eb2c2df592be3e63
472a4c7c1e5108d36e563d505fc4e86ac23ee223
1016839 F20110216_AACALV laberge_a_Page_120.jp2
8088a9e4fd46df9fcc3725686c409c2b
5c6143d168c1a9b5ff663f3e35a7789f8de6c07f
50361 F20110216_AABZZV laberge_a_Page_039.pro
553803e46a0c958d64cb87baf678b797
519cf657662c3ac0b108f7f2d064893914f7d881
8177 F20110216_AACAMK laberge_a_Page_009thm.jpg
c072b28211fa79496fff2f0e1e8161f7
0a31b94e14434d4437696b6c5f83e880def4695b
1034623 F20110216_AACALW laberge_a_Page_121.jp2
452fed639ed1ad7058e54a97d67f3d15
7f901ef6c8eaac83a7cf1976b25cd3f918871f1c
54729 F20110216_AABZZW laberge_a_Page_040.pro
de23e79e1a5aad2daffa874e6cfd65f2
ac59d96b2671921e780670fbc76327c3b82b6f0d
33484 F20110216_AACAML laberge_a_Page_021.QC.jpg
52f9ca6aae39f1626c4cb904831bb830
dac8af09ab2ca422ec2a5b3bd48bc62e27351fc7
889744 F20110216_AACALX laberge_a_Page_122.jp2
f7275f9ea0e4e525337a0d06a8c4b543
95085bb59d65066e39c4a4c176e690d68cdb9eae
22824 F20110216_AACANA laberge_a_Page_013.QC.jpg
b749b7eb58ed42c54e1aab6aebab3498
b88422c36194d8e79f788338580b1982bb814b8c
6988 F20110216_AABZZX laberge_a_Page_042.pro
d02ec347f36f6ec016a4fc28c9bdeea0
e4db34de0bbe244b337fefc2424b20c4bfbe5ec9
33715 F20110216_AACAMM laberge_a_Page_073.QC.jpg
fe68949b8131deebe89788a0da67d025
6c727ce5dbfaa81fd99a89dd243565bafaa5751c
344413 F20110216_AACALY laberge_a_Page_123.jp2
8e25ac3b69886566e71e9bcd8324cc96
40f0a731b8871f4929620901488fc9a601aa6205
9005 F20110216_AACANB laberge_a_Page_048thm.jpg
b3dc54e126650817177edc8fd07e1164
ca1ac61e5bf7122a64a1774d734da2444be2ddac
52897 F20110216_AABZZY laberge_a_Page_043.pro
746da62a922427f14fae56082c6bb2f7
b2d91e6a875e2d1d226727743ca3ffe9a9bde34d
8064 F20110216_AACAMN laberge_a_Page_070thm.jpg
de784a3f96b706d19d2661a70e09e5eb
dc2a07bdcf16bbdc6ac28c512be4b5b4e988a9cb
7118 F20110216_AACANC laberge_a_Page_113thm.jpg
7c66f06c72c1a639de050443170ec182
ed24143fb4cb110f7d5d141698a8ac3883dec673
54995 F20110216_AABZZZ laberge_a_Page_044.pro
50a9b024f86b6aaf27af6ee30fa50733
90b74d2f2d9fe0c133fb4e82fe8c84efedca87cc
37936 F20110216_AACAMO laberge_a_Page_058.QC.jpg
ccd05743622f86093f8b2fd5af332b8a
c7b66efaed756e126ee089caeda69f945ae994c3
480493 F20110216_AACALZ laberge_a_Page_124.jp2
642a9e1e0bcd947e24cdff8157ea716c
d0975eaa454d2c157c05fd2ae21f210b48be6e16
9772 F20110216_AACAND laberge_a_Page_128thm.jpg
f47cbc0f0c6c0fda02ed19a4106202b7
bc4c3541f9e92d2708185f2270079899302c8f12
33941 F20110216_AACAMP laberge_a_Page_125.QC.jpg
d40a7024ababde8650c0a8c3e917b346
fff030f851cd0d910c5c748a4d7a6dc7c0fa6048
33185 F20110216_AACANE laberge_a_Page_035.QC.jpg
b89e1066baba0120817324328eaec078
39f0f4a8e81f759948c9b8b9cfeedd43860680cc
6591 F20110216_AACAMQ laberge_a_Page_087.QC.jpg
ac1d59b45eb18ec75968f88a4d13d9f5
8aec7c186ee5e27a9b28bc35b15f9457aad47240
36967 F20110216_AACANF laberge_a_Page_129.QC.jpg
37e988f22af9efd2514b34eb55709776
aa0a06f7ab7fa294fb9261a84c2048a3d149c70a
35116 F20110216_AACAMR laberge_a_Page_059.QC.jpg
a246eebc9bc31b9380f91348a98dd759
3a39f349d8a9ffc5289480575a8e6a4863bb0a8b
8995 F20110216_AACANG laberge_a_Page_125thm.jpg
fa20f9c808c31a388544659871db2dd6
bd162529ce98be25b3690b0ac9ce1b5f34eb4cee
9142 F20110216_AACAMS laberge_a_Page_050thm.jpg
ed4922371f2339f1f9d8a2c3e645d61b
46e3bc39f0d41ddc8c1323d3d586a0c6514e3641
35935 F20110216_AACANH laberge_a_Page_040.QC.jpg
778a3704f79ec72b5a1a4b01d3b08eaa
b98e02a3b4b93eb8a147081f09f2aa4adb1cac11
35801 F20110216_AACAMT laberge_a_Page_050.QC.jpg
6644b22b0fd17cbb7131c646bb7bf533
43df46beb5568692d5c54c0504d671b330712397
36239 F20110216_AACANI laberge_a_Page_026.QC.jpg
b499906e6de70d4e8ae6dc0a856f58d7
081f7a66e0d803b0f1801549c0c96eabc95f2394
7697 F20110216_AACAMU laberge_a_Page_016thm.jpg
a33a460e43677e23e5dd2332f97b037b
da42ebb35a3887a05a55cad32eb21dfcaa01be2e
36584 F20110216_AACANJ laberge_a_Page_027.QC.jpg
12a6a576ebbecda402b96ab0f60826ea
bdc154efb6553adea7ebcf05417bfed1379e3bbf
8434 F20110216_AACAMV laberge_a_Page_001.QC.jpg
6ba79fd4b50078095ca9ac49acaeee07
d897f388c81a87df225dce711134f4d09ad2b7dc
36749 F20110216_AACANK laberge_a_Page_054.QC.jpg
eec6f538d261957271bec0611e687181
5f5fc6fe1a6b1da2fd637e47b25bcd3ecd175e88
34542 F20110216_AACAMW laberge_a_Page_039.QC.jpg
f16a3649b6bc0ceafeab9d7a54cc3846
afe549877c164fe9ddaa62ef8ea7e48bb50613a5
33399 F20110216_AACANL laberge_a_Page_023.QC.jpg
401c40320e6fd16174291e5a4c6894bc
4606b69aa1ecea726a79983fc4233daa88a8517e
34208 F20110216_AACAMX laberge_a_Page_045.QC.jpg
c4d1ee811d94de8cf9b7577e9aa6f3d3
35b61f34ba5299fb99367ead10238c21ed33aec0
35606 F20110216_AACANM laberge_a_Page_031.QC.jpg
48d88594362fc0d2cf7602e834460363
3405de0b1b18ea426cccd31d0b96ed289d7ce3cb
1185 F20110216_AACAMY laberge_a_Page_002.QC.jpg
77bac6609369bdc02d8a6b33caa5988c
ebb75dfd0043018eedaaec9224d71f06febb38c4
9110 F20110216_AACAOA laberge_a_Page_069thm.jpg
5612ddf93e9a63ba0d0cc34d8fc08184
04fef67d1eb8c1a73618cb300fc73606ab8a3833
25378 F20110216_AACANN laberge_a_Page_006.QC.jpg
1431049d76be085e1f0719680b36e34c
332ea72baf7e7f6cb15ef1bb81f17f46ba88856c
4799 F20110216_AACAMZ laberge_a_Page_005thm.jpg
e69e53ca0f2dec59067e93c38a875b79
a8a62a1694a96222ac4de165f8fa8d985fb65caf
8164 F20110216_AACAOB laberge_a_Page_025thm.jpg
1a660ad809369060a53d8a803f86a2f5
ddd5fffd548cb61a7ec69144d1cb836735bc5970
36840 F20110216_AACANO laberge_a_Page_084.QC.jpg
837ac5a63801b26804addcb66360f97d
5b32d8b805301079ce6b7fc2d685d7f2f0190f17
9065 F20110216_AACAOC laberge_a_Page_010.QC.jpg
0e1973cbe02ccafb4a18b218874928d3
98b50f1587006306e7588e1b8696f97fd4e0662d
9547 F20110216_AACAOD laberge_a_Page_054thm.jpg
8eb46e1f42d857dbe39d327552f14e3c
e9da61ad83cb165375f8e40a76c207585c9b294a
9277 F20110216_AACANP laberge_a_Page_045thm.jpg
89c0cec9301467186c7f232416beb072
8ae5072496e883b498c3ebc254b8fdd49478a83b
33923 F20110216_AACAOE laberge_a_Page_025.QC.jpg
0ff4f98ebd7a7642d5a16eb0b7308913
942549391b6bab89885b19df4b1065c922bf3d0e
36165 F20110216_AACANQ laberge_a_Page_062.QC.jpg
c56f762c9c3ed15928bc01d3be89b67e
abbdc1a51655228eb6147408b24e598df0593861
8401 F20110216_AACAOF laberge_a_Page_017thm.jpg
1bcbdde84c8b24607f999fd724dffe09
cdaf22709ede0fae2b188315aca0ee20bc968067
37674 F20110216_AACANR laberge_a_Page_028.QC.jpg
2f494fa40995f24a63abc7d866f7d742
544ff45d9f770bab0db5d6a9de2b4374d4e3dd87
19545 F20110216_AACAOG laberge_a_Page_005.QC.jpg
7560e65b336589b78e8874ba99519aef
7bbcad83afe6e5086f6398ecb080c8c6177015e1
7025 F20110216_AACANS laberge_a_Page_112thm.jpg
a3d6b2ff5c27c28c12febf35443e3cd4
ac0bb89a1a83d8cef18e5a34b895439896f72ac1
36266 F20110216_AACAOH laberge_a_Page_024.QC.jpg
4ca6f3f23f5fdd1fc2a36139cbf6409e
2bdf6893bba5429729e804709ade73a1d8011838
F20110216_AACANT laberge_a_Page_083.QC.jpg
712a76352fcc01ee7116a7c2605b0815
15ce5172f394868066787f4f75146ae65918f13a
26075 F20110216_AACAOI laberge_a_Page_108.QC.jpg
08177cdcab7c5dd1b6220816e4ae4b8a
1d3bdfe940b3e9f38cac711bd92218131098e9a0
6632 F20110216_AACANU laberge_a_Page_116thm.jpg
cece222f66081edd712e01b67e9d7a0f
1dfd688385ce9853a0e1a7d7074f6bc54e27b456
28548 F20110216_AACAOJ laberge_a_Page_114.QC.jpg
5d5151aaba8bbfddaaaadcc3108bd010
a7c2dd4bb442046d31fc7aef5c0081840f5c442d
9354 F20110216_AACANV laberge_a_Page_074thm.jpg
e4e7d26bfdca5f7f6aae49aebb9501bc
bdd2091d5aa5db7df3f4794573efe85ab7a81564
38012 F20110216_AACAOK laberge_a_Page_068.QC.jpg
6398290c58187a6ca5345a4b8ee86203
ee1f6e4d5b4ee2408a5dc8cb2b48cff94849c106
28271 F20110216_AACANW laberge_a_Page_118.QC.jpg
83b6510d2746b4c01667d830f076c0f1
87889475a7c4c18d655f2e3128bb0575ef46d2b2
32493 F20110216_AACAOL laberge_a_Page_012.QC.jpg
94a4664b954a4d8ae21057766948b720
eec90737a67f8ad5fe8148aadcffb34e74510dad
39326 F20110216_AACANX laberge_a_Page_127.QC.jpg
dfff7799c1c9185beb4c6e737bb64070
a0ad47b887e98d22c94469dbeb9032fb1fb146bb
25032 F20110216_AACAPA laberge_a_Page_122.QC.jpg
cc0464dea7529b9bbd64b61fad13f460
eafb0baf228fa929ca8b6eef3fb1ab3d1ffe1dc7
15976 F20110216_AACAOM laberge_a_Page_130.QC.jpg
d48942a0b22f47f614f58a5e08f38c5b
f32d593d890756acba419fc8f245c9aefc31bdc4
6732 F20110216_AACANY laberge_a_Page_117thm.jpg
62885b02e86d86612be5404586b94594
10677bcc6330d3169707742db90dd7a35a55d434
7640 F20110216_AACAPB laberge_a_Page_030thm.jpg
57f659268c7afa15ff49b03f088e7f0d
4d00a4a6227af36165ed819a52ee83242b21af0b
9697 F20110216_AACAON laberge_a_Page_060thm.jpg
da93dd99777cb205b6ce72d03770a59b
5a6a07fc42cbe039e3a01e57116bfc6b5c9a3c95
35293 F20110216_AACANZ laberge_a_Page_043.QC.jpg
25f851c5dea984f71afa15d414c3e5bc
a5413586b386e1141a4690c363a064943232bdaf
38452 F20110216_AACAPC laberge_a_Page_033.QC.jpg
be10f812128e07a3c0706d26a8da4eba
dda9bf9c24f4793cbf49df57bf437a826b94cd2c
33167 F20110216_AACAOO laberge_a_Page_017.QC.jpg
ba6b5f970c2a0a5ba23e981d31b9b661
337162bd7f19370d3fb14d593aa29b852f241afa
35749 F20110216_AACAPD laberge_a_Page_051.QC.jpg
0fb3b4d80fb72309323c63f17c1cf853
006e1535fb36020d9a1cd18893e2f4032d3c3fa6
35002 F20110216_AACAOP laberge_a_Page_034.QC.jpg
4cd6bf1f5b6bc43892aa56423234f731
c930bbf33685ed7b45184848bc8923be2c687243
35843 F20110216_AACAPE laberge_a_Page_049.QC.jpg
5dee5ba85d77afd4946fabe7fbbee9a4
d6c07726989227455ab51a6da8faa7f0da713348
28209 F20110216_AACAOQ laberge_a_Page_119.QC.jpg
f09c392bb0fe8776fc6e34b689e1dc9b
cb54cc0d91a00aa1d3865722243d4d66928015a4
38766 F20110216_AACAPF laberge_a_Page_066.QC.jpg
82cb54698d3b4024e40c9f75e55d49f5
82b7013f25dbef7b33c8b10e36d949d7c33cc8b0
35462 F20110216_AACAOR laberge_a_Page_069.QC.jpg
d4f845beb5f8040488c728a295ca171c
41820e11f77248a3f166bbd26018eec568c113f4
5507 F20110216_AACAPG laberge_a_Page_108thm.jpg
9c0c1a690e7f2b136b7c0d24f906c0bd
39fef96caf136f77b928c320515bf6aa1250d25b
2894 F20110216_AACAOS laberge_a_Page_124thm.jpg
893c353c9225709575c34f0a2999f0b7
01c62af103dd2314a4b797066de37cab22137227
6374 F20110216_AACAPH laberge_a_Page_107thm.jpg
026c0dacd76c26e5fc22aa38f6b82774
453fd3afcbed12a23257c86edafca80a11a368d9
30132 F20110216_AACAOT laberge_a_Page_110.QC.jpg
bbe8eb1693921baed6c5ea4e1ffa85f5
69f6be90bc726a3467553f9f780445392c3fb7bd
6565 F20110216_AACAPI laberge_a_Page_008thm.jpg
409d42d58fa869aab6db196c81c0bd11
3a135bdb7b1f2800eaa508f6671cfcc74f7680d7
9627 F20110216_AACAOU laberge_a_Page_124.QC.jpg
d0bb0e4474dafe6f5d9e4f946682b2f8
42fb3a6a1bd497249955b41fa38cecb9f57aae94
9171 F20110216_AACAPJ laberge_a_Page_084thm.jpg
c5b79702adbee800fd0540ac533b2279
53c2a1e5af3fdf88fcac29f67768b8a99cc59fc9
9253 F20110216_AACAOV laberge_a_Page_086thm.jpg
bfd7cd61a36da213ab3d2c4eed4a375e
20aa692620b5f0a8c70ca1ea5ace6896b4008c47
8983 F20110216_AACAPK laberge_a_Page_051thm.jpg
68c5e28630e7eaaef83bb4db33bbb697
2e65ac71fe66fe23f452094bd31e4f0660109d8e
9492 F20110216_AACAOW laberge_a_Page_090thm.jpg
d059f017a68479d6466c129ccab3b43f
5a0ee4b6ada5285801021a4ed905a7c5d2cc4e6c
5547 F20110216_AACAPL laberge_a_Page_042.QC.jpg
476f67483b81e833b567be044df24aa7
0ab3934e0b3e186b2fb29883e47872a687883b86
35367 F20110216_AACAOX laberge_a_Page_046.QC.jpg
1c409a2bb09720442643e2272d23dfcf
abbfe20d296410bc46c802bae09590c99168a186
6975 F20110216_AACAPM laberge_a_Page_115thm.jpg
fde5a2f922081c6750cbeb71694f79df
2d60106a05fd1155c3993d90b2408792efd3599d
33322 F20110216_AACAOY laberge_a_Page_041.QC.jpg
5a9117e8ba629a91ed2dacfa0ab73af7
c2a483b5d2e14cd79b799507d7bbd025d23c32fb
31250 F20110216_AACAQA laberge_a_Page_070.QC.jpg
53ad9979316f87ab8af5ba2d1366851f
eada1a84defddae77aef725124b2fc8fac149d0a
9845 F20110216_AACAPN laberge_a_Page_068thm.jpg
e9011ffee4c8a23cc227f2aca833d757
b65e590082e00760bd74c0b40603180ee01667f0
9050 F20110216_AACAOZ laberge_a_Page_033thm.jpg
310244e5ba98a4c9e7b470f668ae037c
4b14d8fdcac58e2d8e9f6affd026c19f37cc0d12
5958 F20110216_AACAQB laberge_a_Page_006thm.jpg
0a0c3c47871139c6ba917b1d125a06a6
528788ea31aefcda6b114afeed5cdd4d3e12c8e5
33184 F20110216_AACAPO laberge_a_Page_109.QC.jpg
736ba870b4611c1d352e4e6bad22c33f
e140bd51f7b3d88bf31c94455a1e8a8d2d3f2904
8783 F20110216_AACAQC laberge_a_Page_088thm.jpg
f85a1b58f27ff028c5be2988090e575a
251688e4e602c04580a760ea7fc4456c2827eb7b
36543 F20110216_AACAPP laberge_a_Page_029.QC.jpg
f7e459ec78ce8eea282f817046975aec
3f69215b50752e709a475f40c991539ea1cfc2fd
8550 F20110216_AACAQD laberge_a_Page_047thm.jpg
4b49ea23a03fd154bde259bd963e15f4
249b384e2c747f0204a8d44e78b8f09b0b175ddc
19872 F20110216_AACAPQ laberge_a_Page_102.QC.jpg
ad84267d441650110e7043c735ae27ec
86316bc0cc8ec2babc49fee927c9107591678dba
8416 F20110216_AACAQE laberge_a_Page_038thm.jpg
7148f778198b5df1d71e65e8f8e8be8f
d3e2c21d792bca94f67ee47d827f2e478b02997c
38776 F20110216_AACAPR laberge_a_Page_077.QC.jpg
7d2edee908ee86124a6e3e8b9807409f
2ce5ed2d43dd6a4f21f2577f8a489dbbd44b0c18
37860 F20110216_AACAQF laberge_a_Page_076.QC.jpg
671878cdf079318231493990a5723d8a
0f6387853d45dc13d65c1dcf8bb388454801c392
9781 F20110216_AACAPS laberge_a_Page_066thm.jpg
584522e0b32875d84b616b92b3ec2a9c
18efb6d59edb6b8b84c59d82bd044e58cacb4562
38258 F20110216_AACAQG laberge_a_Page_061.QC.jpg
f873392c811ab00a8cc003ae7bbef22b
6c3830f49bb886cfbd65cf4a5340b668b3887a20
8483 F20110216_AACAPT laberge_a_Page_035thm.jpg
9c31f91dcde6b9f7c0f35191bc0ddb01
fa8d3a377e7abe0c5b9b281cb44b7385aaa77e6d
34649 F20110216_AACAQH laberge_a_Page_038.QC.jpg
7facc35de8b8f1898e629b7033592270
8b58fe8f5e20516f114de355a619063712d535b6







AN EXAMINATION OF STRATEGIC GROUP MEMBERSHIP AND TECHNOLOGY IN
THE NURSING HOME INDUSTRY























By

ALEXANDRE LABERGE


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 Alexandre Laberge



































This document is dedicated to the four women in my life









ACKNOWLEDGMENTS

There are a number of people who provided me with the opportunity to achieve this goal.

To begin with, I would like to say thanks my mother who not only supported and encouraged me

to learn but also encouraged me to finish what I started. I would like to thank my father who

inspired me to think outside of the box. I would like to thank Dr. Weech-Maldonado for his

guidance, generosity and tremendous patience. And finally I would to thank my wife, whose

unconditional love, support and understanding made it all possible.









TABLE OF CONTENTS

page

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

LIST O F TA BLE S ........................................................................................................... ..... 8

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

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

CHAPTER

1 INTRODUCTION ............... .......................................................... 14

What Are Nursing Homes and What Are Their Roles in the Health Care System? .............15
Long Term Care Resident Reimbursement ............................................................16
M e d icaid ............................................................................... 16
P riv a te P a y ................................................................................................................. 1 8
M medicare ................................................................................................... ............. 18
Post Acute Care Reim bursem ent.......................................................................... 19
N using H om e E nvironm ent............................................................................ .............. 20
The Importance of Focusing on Strategy...................................................... ..................21
P u rp o se ..............................................................................................2 2
Research Questions ........................................................ .. ......... ............... 22

2 CONCEPTUAL FRAMEWORK............................................................ ...............23

S trateg ic G rou p s ........................................................................... 2 3
N ursing H om e T technology ........................................................................... ....................26
N using H om e R sources ............................................................................. .................... 28
Scope and R source C om m itm ent............................................................................ .....31
Scope U sing T technology .......................................................................... ..............32
R source C om m itm ent ....................................................................... .... ...................33
P orter's G generic Strategies ....................................................................................... .......... 34
Strategic Group Model Using Scope (Technology) Commitment, Resources
Commitment and Porter's Generic Strategy Theory.........................................................36
Differentiator ...................................................... ................... ............... 37
C ost L leadership ............. ................................................................. ..............37
B e st C o st .................................................................................................................... 3 8
F o c u s ....................................................................... 3 8
L ack of Strategy ................................................................... 39
M easu rin g P perform an ce .................................................................................................... 3 9
N ew C o n trib u tio n s ............................................................................................................ 4 1
H y p o th e sis ..........................................................................4 1









3 METHODOLOGY ................................. ......... ....................... 43

D ata ......... ....... ... ............. ... ............................................................... 4 3
P o p u latio n ........... ..... ... ............. ... ...........................................................4 4
Overview of M ethodology............................................. .. .... .. ..................... .....45
Hypothesis #1: Nursing homes that have a strategy can be categorized into one of four
strategic groups: Differentiator, cost leadership, focus or best cost. ................................45
S c o p e ..........................................................................4 5
Nursing Technology .................................... ..... .. ...... ............... 45
R ehab T technology ................ .. ............................................................. ..... 52
R source C om m itm ent .......................... ............................................... ............... 54
Hypothesis #1 Analysis ................................... ..... .. ...... ............... 56
Factor A analysis .................................................................................... .................... 57
Factor M atrix ..................................................................................... ..................... 58
N um ber-of-F actors P problem ......................................... .............................................59
Rotation of the Factor Structure ........................................................ ............. 59
Internal Consistency of M measures ............................................................................60
C lu ste r A n a ly sis ......................................................................................................... 6 0
Strategic G groups D defined ........................................................................ .................. 63
D eterm ine the total variance value........................................................................ 63
C lu ster v alu e ............. ................................................................. ............. 63
C lu ster rank score ............. ..................................................... ........ ........ 64
A analysis of V ariance .......... ............................... .. ..... .. .. .......... .. .... ..... .... .......... .... 65
Hypothesis #2: Nursing Homes in the Differentiator Strategic Group will have the
H highest Q quality. ..................................................................66
Dependent Variables: ............................................66
Independent V ariables ..................................... ......... ...... .... .. ........ .... 67
Control Variables.......... ........ ............. .. ... .... ..................67
A n aly sis ............................................... ..... .. ..... ........ ............... 6 8
Hypothesis #3: Nursing Homes in the Focus Strategic Group will provide Higher
Quality in the area that they are focused on compared to other Strategic Groups..............69
D dependent V ariables ......................... ......... .. .. ........... ... ...... 69
Independent Variables ............. ................................... .... ......... 69
Control Variables.......... ........ ............. .. ... .... ..................70
A n aly sis ......................................... .. .. .. ............ ....... .. ............... 7 0
Hypothesis #4: Nursing homes in the Cost Leadership Strategic Group will have the
low est costs. ..................................................................70
D dependent V ariables ......................... ......... .. .. ........... ... ...... 70
Independent V ariables .................................................. ........ ..............70
C o n tro l V ariab le s....................................................................................................... 7 1
A n aly sis ..................... .... ..... .. .. ............... ........ ...... ... .. .............. 7 1
Hypothesis #5: Differentiation, Cost Leadership, Focus and Best Cost Strategic Groups
will have better Financial Performance than Nursing Homes that Lack a Strategy ...........71
D dependent V ariables ......................... ......... .. .. ........... ... ...... 71
Independent V ariables ............. ...... ............................................... ........ 72
Control Variables.......... ........ ............. .. ... .... ..................72
A n a ly sis ................................................................................7 2


6










4 R E SU L T S ....................................................... 74

H y p o th e sis # 1 ................................................................................................................... 7 4
H y p o th e sis # 2 ................................................................................................................... 8 2
H y p o th e sis # 3 ................................................................................................................... 8 3
H y p o th e sis # 4 ................................................................................................................... 8 4
H y p o th e sis # 5 ................................................................................................................... 8 5

5 D ISC U S SIO N ............................................................................... 88

Q u a lity ................................................................................................................... 9 0
C o st ........... ....................................................................................9 3
Financial Performance ..................................................................... ........ 94
M an ag erial Im p location s ................................................................................................... 9 5
P o licy Im p lic atio n s ........................................................................................................... 9 6
C o n clu sio n ................... ...................9...................8..........
L im itatio n s ................... ...................9...................8..........

APPENDIX: BACKGROUND TABLES ....................................................... 100

L IST O F R E F E R E N C E S ....................................................................................................12 5

BIOGRAPHICAL SKETCH ................................................................... ... ......... 130






























7









LIST OF TABLES


Table page

4-1 Internal consistency of the composite variables ..................................... ....... ........... 76

4-2 Individual variables used in cluster analysis that were not included in a factor
v a riab le ................... .......... ............................................. ................ 7 6

4-3 Rank scores scope and resource variables .............................................. ............... 78

4-4 Rank scores of structure and market factors ..................................... ........ ............... 79

4-5 Strategic group quality of care with Differentiator / nursing care focus cluster as the
reference group ............................................................................ 82

4-6 Strategic group quality of care with rehab focus cluster as the reference group ..............83

4-7 Strategic group quality of care with Specialty care focus cluster as the reference
g ro u p ............................................................................................. 8 4

4-8 Strategic groups and costs with cost leader as reference group regression results ............85

4-9 Strategic groups and predicted costs.......... ..... ......... ................... 85

4-10 Odds Ratio of Nursing Hom e Strategy ........................................ ......................... 86

4-11 Probability of nursing homes being in the highest operating margin tier .......................86

4-12 Probability of nursing homes being in the highest total margin tier..............................87

A-i Eigenvalues of scope (technology) factor analysis................................... ........100

A-2 Factor analysis with 10 factors of scope (technology) variables.............. ................ 102

A-3 Factor analysis with 10 factors of scope (technology) variables.............. ................ 104

A-4 Eigenvalues of scope (technology) factor analysis................................... ........106

A-5 Factor analysis of resource staffing intensity with three factors.................... ........ 106

A-6 W hard's minimum variance cluster analysis................................ ........................ 107

A-7 Cluster History ............................. ..................... ... ........ 107

A-8 Determination of rank scores table ......................... ............................ ........... .... 108

A-9 Descriptives of dependent variables, independent variables and control variables.........109









A-10 Negative binomial of quality of care deficiencies with differentiators as reference
g ro u p ................... ........................................................... ................ 1 1 0

A-11 OLS regressions of root transformed bladder decline with differentiators as ref.
g ro u p ................... ........................................................... ................ 1 1 0

A-12 OLS regressions of root transformed bowel decline with differentiators as ref. group... 111

A-13 OLS regressions of root transformed ADL decline with differentiators as ref group .... 111

A-14 OLS regressions of log transformed restraints with differentiators as ref group............ 112

A-15 OLS regressions of root transformed LTC ulcers with differentiators as ref. group....... 112

A-16 GLM with gamma distribution with PA walking improve with differentiators as ref.
g ro u p ................... ........................................................... ................ 1 1 3

A-17 OLS regressions of root transformed PA skin ulcers with differentiators as ref group.. 113

A-18 Negative binomial of quality of care deficiencies with rehab focus as reference group .114

A-19 OLS regressions of root transformed bladder decline with rehab focus as ref group..... 114

A-20 OLS regressions of root transformed bowel decline with rehab focus as ref group....... 115

A-21 OLS regressions of root transformed ADL decline with rehab focus as ref. group ........ 115

A-22 OLS regressions of log transformed restraint with rehab focus as ref. group ...............116

A-23 OLS regressions of root transformed skin ulcer with rehab focus as ref group ...........16

A-24 GLM of PA walking improve with gamma distribution with rehab focus as ref group .117

A-25 OLS regressions of root transformed PA skin ulcers with rehab focus as ref. group...... 117

A-26 Negative binomial of quality of care deficiencies with specialty care as reference
g rou p .............. ............................. .. .............................. ........................................1 18

A-27 OLS regressions of root transformed bladder decline with specialty care focus as ref.
g rou p .............. ............................. .. .............................. ........................................1 18

A-28 OLS regressions of root transformed bowel decline with specialty care as ref group ... 119

A-29 OLS regressions of root transformed ADL decline with specialty care as ref. group..... 119

A-30 OLS regressions of root transformed restraint with specialty care as ref. group............120

A-31 OLS regressions of root transformed LTC skin ulcer with specialty care as ref. group.. 120









A-32 GLM with gamma distribution PA walking improve with specialty care as ref group..121

A-33 OLS regressions of root transformed PA skin ulcers with specialty care as ref. group ..121

A-34 GLM regressions of nursing home cost with gamma distribution............................122

A-35 Odds of nursing home of being in the highest operating margin tier ............................122

A-36 Odds of nursing home of being in the highest total margin tier ..............................123

A-37 Correlations of independent variables ........ .......... ...... .... ............... 124









LIST OF FIGURES


Figure page

2-1 T he nursing hom e process ........................................................................ ...................30

2-2 Strategic Group Model Based on Scope and Resource commitment.............................36

4-1 Strategic group model based on scope and resource commitment in the nursing
hom e indu story ....................................................... .................. 8 1

A-i Scree plot of scope (technology) factor analysis .................................... ............... 101

A-2 Scree plot of resource staffing intensity from factor analysis............... ... ...............106









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

AN EXAMINATION OF STRATEGIC GROUP MEMBERSHIP AND TECHNOLOGY IN
THE NURSING HOME INDUSTRY

By

Alexandre Laberge

May 2009

Chair: Robert Weech-Maldonado PhD
Major: Health Services Research, Management and Policy

Purpose: The purpose of the study was to examine if strategic groups in the nursing home

industry can be determined by examining how facilities apply their technology and how facilities

commit their resources. The study also determined if these groups can be defined using Porter's

generic strategies and whether the different groups have varying strategic-performance

relationships.

Methodology: The study began by performing a factor analysis using scope (technology)

and resource commitment variables from the Minimum Data Set (MDS) and the Online Survey

Certification and Reporting (OSCAR) system. The newly defined variables were used in a

cluster analysis to identify the different clusters. The clusters were classified using Porter's

generic strategy model. Once the strategic groups were identified, the study analyzed the

strategic-performance relationships using negative binomial, ordinary least squares regression

(OLS), generalized linear model (GLM) with gamma distribution and ordered logit analytical

tools. Quality dependent variables included quality of care deficiencies, long term care (LTC)

pressure sore prevalence and post acute pressure ulcer incidence, LTC activities of daily living

(ADL) decline, post acute walking improvement, LTC bowel and bladder decline and LTC

restraint. Financial measures included cost per resident day, operating margin and total margin.









Results: The study identified four strategic groups (differentiator, cost leader, rehab focus,

and specialty care focus) and a lack of strategy group. Differentiators, rehab focus and specialty

care focus groups had less quality of care deficiencies than the lack of strategy group and cost

leaders. Differentiators generally had better long term care quality outcomes compared to other

strategic groups and the lack of strategy group. Cost leaders, differentiators and lack of strategy

groups had lower cost than the rehab focus and specialty care focus groups. Cost leaders,

differentiators and rehab focus group had better financial performance than the specialty care

focus group and lack of strategy group.

Conclusions: The study indicates that using technology and resources commitments is an

effective way of identifying strategies groups. The study also found that some strategic groups

have better financial performance than other strategic groups or than nursing homes that have no

strategy at all. Nursing homes that provide a high level of technology had better quality, without

increased cost or sacrificing financial performance. The study also suggests that nursing homes

that service residents with high cost conditions may not be adequately reimbursed.









CHAPTER 1
INTRODUCTION

Nursing homes, like other health care organizations face many challenges. They have a

limited ability to dictate their price and to control their costs. Nursing homes also face higher

competition from the rapid rise in home health agencies and assisted living facilities. To deal

with these challenges, nursing homes must devise and implement effective strategies if they are

to remain profitable. These strategies can vary from focusing on providing quality service,

controlling cost or identifying niche markets. Although there are over 16,000 nursing homes in

the United States, the number of different strategies to choose from is small. It is therefore

possible that nursing homes can be categorized into different groups that have similar strategies.

Nursing homes are an important component of the health care industry consisting of 8% of

the United States health care expenditures (Centers for Medicare and Medicaid 2007). The

nursing home industry provides care for 1.6 million of the frailest people in the United States

(Center for Disease Control and Prevention 2007). A study by the U.S. Department of Health and

Human Services says that people who reach the age of 65 have a 40 percent chance of entering a

nursing home. Of those who enter a nursing home, 10 percent of them will remain there five

years or more (Centers for Medicare and Medicaid 2007).

Expenditures for nursing home care skyrocketed in the years following the implementation

of the hospital prospective payment system of 1983. The industry grew from 50 billion dollars in

1990 to 90 billion dollars in 1997 (Centers for Medicare and Medicaid 2007). However, this

changed after the implementation of the Balanced Budget Act of 1997 (BBA 1997) where the

industry only grew 10% between the years 1999 and 2004. The BBA cuts in reimbursement were

so severe that a number of nursing homes went bankrupt including some of the nation's largest

nursing home chains (GAO 2000). Since the implementation of the BBA, the remaining nursing









homes have been faced with considerable challenges that include increased government

regulation (GAO 2001, Walshe et al. 2002), increased litigation (Johnson et al. 2003) and

increased competition (Weech-Maldonado et al. 2003).

In order to survive, nursing home managers need to identify effective strategies so that

they can continue to provide a very valuable service to a population with few options. Some of

those strategies include providing high quality care, keeping costs as low as possible and

focusing on a specific market segment of the industry. These different types of strategy do not

only relate to better financial performance, but can also influence quality.

To fully understand these different types of strategies, it is necessary to have a clear grasp

of what nursing homes are, to have insight on what their role is in the health care industry and to

comprehend how they function in an uncertain environment.

What Are Nursing Homes and What Are Their Roles in the Health Care System?

A nursing home is a place for people who do not need to be in a hospital but cannot be

cared for at home. Most nursing homes have nursing aides and skilled nurses on hand 24 hours a

day (National Institutes of Health 2008). The stay in a nursing home can be permanent or can be

temporary.

Nursing homes have two major functions. The first function is to provide long term care to

those individuals who require more care than can be provided by their families or can be

provided by any other type of placement. The desired outcome for the long term care resident is

that during their stay, their health status remains at least at the same level as it was on admission.

It is understood, however, that the longer the resident remains in a facility, the more likely they

will experience a decline in their health status. The decline occurs as a result of the normal

progression of the pre existing medical conditions that typically precipitated the admission to the









facility in the first place. The desired outcome for long term care residents is to maintain their

health status for as long as possible.

The second major function of a nursing home is to provide short term care (also known as

post acute or sub acute) for those residents that require skilled services after a qualified hospital

stay. To qualify for a post acute stay, residents must have either a skilled nursing care need (e.g.

IV antibiotics) or a rehab care need based on the recommendations of the hospital therapists.

The goal of post acute care is to improve the resident's health status so that the resident can be

discharged home or to the most appropriate level of care. The discharge options are quite broad

and can range from patients going home by themselves to becoming long term residents in a

nursing home. Discharge from a skilled nursing rehabilitation facility (also known as a post acute

or subacute facility) can occur when the patient reaches their highest level of function or the stay

is no longer covered by the payer. The desired outcome for a rehabilitation resident is a better

health status.

Long Term Care Resident Reimbursement

Reimbursement for long term care is primarily provided by Medicaid or by the residents

themselves (private pay). A small proportion of residents are covered by other payers like the

Department of Veteran Affairs (VA) and private insurance. Although Medicare does not pay for

a long term care resident's stay in a nursing home, Medicare does cover some ancillary services

like physical or occupational therapy.

Medicaid

Medicaid is the major payer of nursing home long-term care services. Over 50% of nursing

homes revenues and 70% of nursing home beds are covered by Medicaid (Rhoades et al. 2000).

Swan et al. 2000 describe five rate setting methodologies that state Medicaid programs use to









pay nursing homes: retrospective, prospective class, prospective facility-specific, adjusted, and

combination.

Retrospective payments freely adjust rates to current costs. In contrast, prospective

payment rates are not fully adjusted to cost during the costs year. Rates are usually set based on

prior year costs. Prospective class rates are the same for all nursing homes in the state while

prospective facility-specific are set based on the costs for each facility. Adjusted reimbursement

is similar to retrospective reimbursement however adjusted systems allow interim rates to

increase during a year but not to fully reflect costs. Finally, there are also combination

reimbursement methods which are a mixture of retrospective and prospective payment systems

(Swan et al. 2000). As of 1997, only one state used retrospective reimbursement, 25 used

prospective, 21 used adjusted and 3 used combinations (Feng et al. 2006).

Since the late 1990s, most state Medicaid programs have used a prospective per diem

system to reimburse nursing homes for all the cost that is incurred in providing the care to the

resident in the facility (Feng et al. 2008). With this reimbursement methodology, a nursing home

that provides a lot of nursing services will not necessarily be compensated more than a facility

that provides fewer nursing services.

An increasing number of states have modified their reimbursement methodology by

making case mix adjustments to their per diem rate. This allows nursing homes to collect higher

rates for residents that have higher acuity and that are likely to cost the facility more to care for

them. The number of states that use case mix reimbursement increased from 19 in 1991 to 35 in

2004 (Feng et al. 2006).

The best-known and most widely used case mix methodology is the Resource Utilization

Groups system (RUGs), currently in its third version (Fries et al. 1994). This system classifies









residents into homogeneous categories based on their estimated resource utilization. Associated

with each of these categories is a case-mix index, which represents, at least relatively, the time,

or cost, of caring for the average resident in the group. A higher case-mix index indicates a

greater degree of complexity and, consequently, a greater need for input resources. Under this

system, nursing homes with a higher case-mix index, on average, would be reimbursed a higher

rate (Zinn et al. 2008).

Private Pay

Private paying residents reimburse approximately 18% of the beds (GAO 2000). Private

pay reimbursement is commonly based on a per diem rate, which is on average 1.4 times the

amount that Medicaid reimburses (Grabowski 2004).

Medicaid only pays for those residents who do not have enough funds to pay for the

nursing home stay themselves. Residents who have some assets will pay the private pay rates

until their personal funds are exhausted. Once the resident has spent down their assets, they

usually qualify for Medicaid.

Medicare

There are some processes of care provided to the long term care residents that are paid for

by Medicare (Wodchis et al. 2004). Physical, occupational and speech therapy are three services

that can be provided to long term care residents by nursing homes. These services are reimbursed

over and above the Medicaid per diem rate or the private pay rate. Therapy services are paid for

under Medicare part B and is considered an outpatient service. Although the price is set by the

Center for Medicaid and Medicare (CMS) however the amount of therapy services provided was

not limited until 2005. In 2005, the CMS implemented a cap on therapy services that limited

residents to a maximum annual benefit of 1700 dollars of physical and speech therapy combined

and 1700 dollar cap on occupational therapy. Initially passed by congress in 1997,









implementation of the caps was delayed by a number of moratoriums. Even in 2005, after the

caps were implemented, nursing homes were able to circumvent the caps if a resident had

therapy needs that were deemed medical necessary (APTA 2008).

Post Acute Care Reimbursement

Reimbursement for post-acute care is primarily reimbursed for by Medicare; however it

can also be covered by private insurance and the VA. Approximately 12% of the beds in nursing

homes are occupied by rehabilitation patients paid for by Medicare (GAO 2000). These patients

usually have experienced a decline in their health status and require rehabilitation before they

can return either home or to the next appropriate level of care.

Medicare reimbursement uses a prospective payment system. Similar to Medicaid case mix

reimbursement, residents are classified using the Resource Utilization Groups (RUG). These

groups are commonly determined by the amount of therapy and nursing services that the resident

requires. For example, a facility will receive a higher daily rate if the resident gets more therapy.

Each rehab RUG level has a minimum amount of therapy minutes that have to be provided

within 7 days prior to a given assessment reference date. For example, for a patient to be

classified as an ultra high rehab RUG level, they would have to receive 720 minutes of therapy

during the 7 day assessment period. Patients are assessed a maximum of 5 times depending on

the amount of time they remain at the facility for rehab. The assessments include the 5-day, 14

day, 30 day, 60 day and the 90 day assessment. The 5 day assessment identifies the RUG level

for day 1 to 14. The 14 day assessment identifies the RUG level for day 15-30. Each subsequent

assessment identifies the RUG level for the following period. Although the each resident is

eligible for up to 100 days of rehab per medical incident, most patients are discharged from

subacute services well before the 100 day limit because they go home or are no longer making









progress. The difference in reimbursement rate between the RUG levels is significant (MDS

2007).

Higher rehab RUG reimbursement levels are associated with higher therapy utilization

(and higher reimbursement rates) but are not necessarily associated with higher acuity (MedPAC

2007). For example, a post acute patient with a knee replacement that only requires supervision

with his ADLs may be classified at an ultra high RUG level because the patient can tolerate high

volume of therapy, not because the patient has higher acuity. In contrast, a post acute patient that

is dependent and has low tolerance to activity will only classify at a high or medium RUG level.

Based on acuity scores, the second patient would be considered more acute but the facility would

be compensated less. Therefore, nursing homes that have a higher proportion of residents in the

highest RUG level will have higher revenue but they will not necessarily have higher comparable

costs. MedPAC (2007) found that nursing homes increased the proportion of their residents

receiving the two highest paying RUG levels (ultra and very) by 50% between the years 2000

and 2005.

Nursing Home Environment

Similar to most health care organizations, nursing homes exist in a turbulent, complex and

uncertain environment. Since the Balanced Budget Act 1997 (BBA), the nursing home industry

has been dealing with a number of cost control mechanisms imposed by federal and state

government payers. This problem affected the whole industry since government payers like

Medicare, Medicaid and the VA account for over 62% of the nursing home industries' revenues

(MedPAC 2003). Public payers often dictate price of a resident's per diem stay to the nursing

home. By removing price from the equation, nursing homes must find other ways to compete for

residents that provide the nursing home with the highest revenue and that have the lowest cost.









Nursing homes also face increased competition from growing industries like assisted living

facilities and home health agencies that tend to draw the less dependent (and more profitable)

private pay residents. This ultimately leaves nursing homes with the residents who have higher

medical acuity, greater functional dependence and have higher care costs (Weech-Maldonado et

al. 2003).

With the limited ability to control their revenue and with the increased competition created

by substitutes, nursing homes find themselves in a precarious position. For nursing homes to

thrive (or even just survive), it is necessary for them to identify strategies that allow them to

achieve competitive advantage.

The Importance of Focusing on Strategy

Government payers and interest groups are interested in persuading nursing homes to

provide high quality of care. One way to encourage facilities to provide better quality of care is

to demonstrate that better quality results in better financial performance. Some recent studies

have determined a positive relationship between financial performance and quality of care

(Weech-Maldonado et al. 2003, 2008).

Even though nursing homes may have the incentive to provide better quality, it is also

necessary for the nursing home to determine the best strategies to be able to provide high quality

care. Without incentives and a strategy, a nursing home will have some difficulty achieving these

desired outcomes.

Managers are interested in achieving above normal profits. Strategies are plans related to

the mission and the vision of an organization that provide leadership with the means to achieve

high performance. Like any industry, nursing homes attempt to gain competitive advantage by

selecting and implementing an effective strategy that the competitors are not able to reproduce

(Barney 1991). Because the number of truly distinct strategies available in any industry is likely









to be small, managers decide not so much how to be unique, but rather which group of

competitors their strategies should be similar to (Marlin et al. 2002). The strategic group model

is one approach that can help managers and researchers identify the best strategy.

Purpose

The purpose of the study is to examine if strategic group structure of the nursing home

industry can be determined by how facilities focus their technology and how facilities commit

their resources. The study will also determine if these groups can be defined using Porter's

generic strategies and will examine whether the different groups have different strategic-

performance relationships.

Research Questions

What strategic groups exist in the nursing home industry?

How does nursing home strategic group membership affect quality of care?

How does nursing home strategic group membership affect costs?

How does nursing home strategic group membership affect financial performance?









CHAPTER 2
CONCEPTUAL FRAMEWORK

This study will begin by defining strategic group theory, Porters' generic strategy theory,

technology and Cool and Schendel's scope and resource commitment theory. The study will then

explain how these theories are integrated to generate a model that uses Porter's generic strategies

to define strategic groups in the nursing home. Finally, the study will define how the different

strategic groups can have different strategic-performance relationships as it pertains to quality,

cost and financial performance.

Strategic Groups

The concept of strategic groups was introduced by Hunt (1972) in his thesis that

examined the appliance industry. He discovered that there was less competitive rivalry than what

industry concentration ratios suggested. He attributed this to the existence of subgroups within

the industry that effectively reduce the number of competitors in each market. Caves and Porter

(1977) expanded the theory by defining strategic groups as a set of firms that face similar threats

and opportunities that are different from the threats and opportunities faced by other firms in the

industry. Strategic group theory stipulates that within an industry, there could be groupings of

organizations that have very different strategies and yet may still have good financial

performance. This theory was a shift in thinking from the industrial organizational theory where

it was generally believed that there was only one right way of doing things (Leask 2007).

Analyzing strategic groups gives insight to different competitors' approaches to the marketplace

(Harrigan 1985). The number of strategic groups within an industry corresponds to the number

of unique strategies within that industry (Marlin et al. 1999). Strategic groups are persistent

strategic characteristics of an industry, which are protected by mobility barriers (Porter 1980).









Movement between groups is limited by mobility barriers, which are associated with the

cost of moving from one strategic group to another (Porter 1980). Conceptually, mobility

barriers behave similar to barriers to entry in Porter's five forces analysis framework (Barney

1991). This framework describes entry of new organizations to the market place as threats to the

profitability of the incumbent organizations. By establishing barriers to entry, an organization

can block out new competitors from entering the market, thereby maintaining their above normal

margins (Caves and Porter 1977). The existence of mobility barriers can allow two groups within

the same industry to exist without directly competing with each other (Leask 2007). In the

nursing home industry, facilities that are a part of the strategic groups concerned with efficiency

and keeping operating cost as low as possible are different from those nursing homes that

attempt to differentiate themselves by providing better care (Marlin 1999). Although these

strategies are very different, it is possible that nursing homes in either one of these strategic

group types will have good financial performance.

For an organization to move from one strategic group to another, it would be necessary for

them to change their business strategy and put themselves at risk of not recouping the resources

they invest (Porter 1980). These costs offer protection to group members by discouraging entry

of rivals into the group (Porter 1980). An example would be an organization from a cost

leadership strategic group trying to move to a differentiating strategic group. This organization

would have to make some considerable investment and they would also have to change the

organizational culture from being cost focus to providing more and better services.

The cost of movement between strategic groups can vary. For a facility to move from cost

leadership to focus may require significantly less resources than moving from cost leadership to

differentiator. Therefore the higher the cost to change strategic groups, the higher the mobility









barriers. Higher barriers offer higher insularity to the members of the groups, thus limiting

competition and group membership (Marlin et al. 2002). The height of the mobility barrier is

expected to be directly related to the performance of the strategic group (Porter 1980).

Since the introduction of the strategic group concept by Hunt (1972) and the additional

contributions by Caves and Porter (1978), there have been a significant number of empirical

studies that have examined strategic groups and their impact on firm performance. Studies that

have examined the beer (Tremblay 1985), banking (Mehra 1996), airline industry (Kling et

al.1988), and the pharmaceutical industry (Figenbaum et al. 1992, Leask 2007) have all found

the existence of strategic groups with strategic-performance relationships. Direct care health care

industries have also been examined using strategic group theory in the hospital (Marlin et al.

2002, Ketchen et al. 2003) and the nursing home industry (Castle 2003, Marlin et al. 1999, Zinn

et al. 1994).

The three studies that have examined strategic groups in the nursing home industry all

have found significant difference between the groups and have also found some strategy-

performance relationships. Castle 2003 used a survey that asked the leadership of the nursing

homes to self characterize themselves using Miles and Snow's categories (1990): prospectors,

analyzers, reactors or defenders. He found that the Miles and Snow's prospectors had higher

levels of financial performance and higher quality. Zinn et al. 1994 and Marlin et al. 1999

attempted to determine if there was a relationship between strategic group membership and

nursing home performance. Both studies used scope and resource commitment as defined by

Cool and Shendel (1988) to determine their clusters and found the existence of succinct strategic

groups in the nursing home industry.









Nursing Home Technology

Hulin and Roznowski (1985) defined technology as the physical and knowledge processes

by which materials in some forms are transformed into outputs. In the nursing home industry,

technology generally refers to the care processes provided by the staff to the residents. In order

to provide these processes, an organization requires sufficient resources. For example, in the

nursing home industry, a facility that fails to have adequate staffing levels (the primary resource

in the nursing home industry) is not going to be able to adequately provide the care processes

(i.e. technology) that are necessary to have good patient care outcomes.

The Centers for Medicare and Medicaid (CMS) statute 483.25 dictate that each resident

must receive from the facility the necessary care and services to attain or maintain the highest

practicable physical, mental, and psychosocial well-being, in accordance with the comprehensive

assessment and plan of care (CMS 2008).

Nursing homes use a number of different technologies that allows them to address the

multiple needs of their residents. Each of these technologies addresses a specific care need.

There are two broad categories of nursing home technology: nursing technology and rehab

technology

Nursing technology: Nursing technology includes all of the care processes that are

provided by the nursing staff. These processes of care are required so that the resident can

maintain or improve their health status. Nursing provides routine care such as bathing, providing

medication and repositioning the residents. Nursing technology also includes less routine

services like skin ulcer care or catheter placement. Most routine services require that the nursing

home establish a program in order for the processes to be provided properly. For example, even

though a routine technology like restorative ambulation may only require one nursing assistant to

walk with the resident, a restorative program must be established. The program requires that









there is an MD order, that the goals are generated by the physical therapist and the program

requires some oversight to ensure that the service is provided. The responsibility of the

organizational features of a restorative program is placed on the licensed nurse.

Failure to provide these processes of care can result in a decline in the resident's health

status. One example of nursing technology is pressure sore prevention. If pressure sore

preventative processes are not provided, a resident will be more likely experience a decline in

their health status because they can develop a pressure sore. Having a patient on a turning

schedule every 2 hours or providing them with a pressure relieving device reduces the likelihood

of a pressure sore will occur. Turning schedules relieve pressure and allow blood to flow to the

skin. Pressure relieving devices distribute the pressure over a greater surface area and lower the

pressure on the high-risk breakdown areas like the coccyx (CMS 2008). A nursing home that

puts emphasis on nursing technology may be trying to get better nursing outcomes. More

processes of care and better outcomes are indicators of better quality (Schnell et al. 2004). Most

of nursing home technology is provided by the nurses. A facility focused on better quality may

be trying to differentiate themselves from their competitors. By differentiating themselves from

their competitors, a facility may get a greater proportion of private pay residents and/or higher

profit margin Medicare residents. In addition, better outcomes can also lead to lower costs

(Weech-Maldonado et al. 2003).

Rehabilitation technology: Rehabilitation technology can be provided to either post acute

or long term care residents. These processes of care consist primarily of therapy services and are

geared to improve the resident's health status. Rehabilitation technology is not considered

routine because of its dynamic nature. Therapy services vary based on the needs of the residents.

Some residents may have difficulty with transferring from the chair to the toilet while others









have difficulty walking. Both these issues would be identified and addressed by physical therapy

through an assessment and then would be treated with a specifically program tailored to the

individual residents needs. Even the treatment protocol can change over time. Initially the

resident may require treatment to increase their strength and balance so they are able to stand.

Once the patient is able to stand, physical therapy will then focus on balance and walking.

Therapy assessment can only be performed by a licensed therapist while therapy treatments can

only be performed by either a licensed therapist or licensed therapy assistant. Nursing homes

have an incentive to use rehab technology over nursing technology since rehab technology

provides better revenue streams. MedPAC 2007 has warned that the current Medicare

reimbursement system favors over utilization of therapy. Although it may be fiscally

advantageous to provide rehabilitation, it only serves as a few of the processes of care that a

resident receives on any given day. The therapist actually only plays a small role in the overall

care of residents when compared to role of the nurses.

Nursing Home Resources

According to the resource based view of the firm, performance differences across firms can

be attributed to the variance in the firms' resources and capabilities (Hitt et al. 1990). Resources

that are valuable, unique and difficult to imitate can provide the basis for firms' competitive

advantages (Amit & Schoemaker, 1993; Barney 1991). Firms employ both tangible resources

(such as buildings and financial resources) and intangible resources (like knowledge and brand

equity) in the development and implementation of strategies. Human capital has long been

argued as a critical resource in most firms (Pfeffer, 1994). In the nursing home setting, the

tangible resources like human capital and intangible resources like staff expertise are especially

important because most of the nursing homes expenses are the nurses and therapist salaries. In

addition, a facilities' technology is dependent of the amount and quality of their health care staff.









Human capital provides the physical care and possesses the knowledge of the facility.

Knowledge is one of the most critical resources that a firm possesses (Hitt et al. 1991).

The nursing home requires adequate resources if they are going to be able to provide their

technology effectively. The correlation between staffing levels and technology provided is

expected to be high. The amount of nursing staff and the skill of the nursing staff will influence

the quality outcomes (Harrington et al. 2000, Weech-Maldonado et al. 2003) and will likely

influence the quality of the technology that the facility provides.

Nursing resource: The total nursing staff levels of a nursing home is measured by adding

all the registered nurses (RN) per resident, the licensed practical nurses (LPN) per resident and

the certified nursing assistant (CNA) per resident. Although the duties for each of these staffing

types are well defined and are intended to be separate, there remains is a certain degree of

overlap. As a result some facilities may chose to substitute higher skilled staff for lower skilled

staff (E.g. CNA for RN or LPN for RN). Using lower skilled staff allows facilities to keep costs

down. However, nursing homes must have an appropriate amount skilled staff in order to provide

quality of care. Facilities who substitute staffing may lose their savings to the cost of poor

outcome. For example, preventing a pressure ulcer from occurring through good care practices is

going to cost less then treating a newly acquired ulcer.

Rehab resource: The total rehab staffing level of a nursing home is measured by adding

all the physical therapists (PT), physical therapy assistant (PTA), occupational therapist (OT),

certified occupational therapy assistant (COTA) per resident and speech pathologist per resident.

The duties for each of these staffing types for each discipline are less well defined than nursing

because therapy assistants can provide almost all the treatments that a therapist can provide. As a

result facilities may chose to substitute more skilled staff for lesser skilled staff.









Figure #1 illustrates the relationship between the technology provided by the facility and

the resources needed to provide that technology. A nursing home's primary function is to

provide care to all the facility's residents. The required inputs not only include the facilities

tangible and the intangible resources but also include the residents' health status on admission.

The technology includes all the nursing homes' care processes that use the facilities tangible and

intangible resources. Staffing levels are the tangible resources that provide the physical care.

The intangible resources include all the knowledge that the nursing home staff possesses and is

measured by the amount of the facility's staff is licensed (Weech-Maldonado et al. 2003).

Similar to inputs, the outputs also use the measure of the health status of the residents. However,

the measure only occurs after the nursing home care processes have been applied.




INPUTS Technology OUTPUTS


Tangible Resources



Resident Health Nursing Home
Status Care Processes Resident Health
S (Technology) Status


Intangible Resources


Figure 2-1. The nursing home process

One factor that is unique for long term care is that because the resident is not likely to

leave the facility, the desired output is the sustained health and well being of the resident. The

measure of quality is how well the nursing home maintains (or improves) a resident's health and

well being during their stay in the nursing home. This is not equally true for the post acute









resident since Medicare only pays for patient's stay as long as the patient's health status

improves. Therefore the desired output for the post acute patient is a better health status in hope

that the resident can be discharged home.

Scope and Resource Commitment

Early writings postulated that business level strategy, the level relevant to strategic groups,

consists minimally of two sets of activities: (1) business scope commitments and (2) resource

commitments (Ansoff 1965; Katz 1970; Hofer and Schendel 1978; Day 1984; Cool and Schendel

1988).

Scope commitments are those decisions that lead to the selection of market segments to

compete and to the types of services to offer (Cool and Schendel 1988, Zinn et al. 1994). While

applicable across diverse industries, the variables chosen to operationalize these strategic

dimensions should be industry specific, reflecting the bases for competition in the industry under

study (Cool and Schendel 1988). Resource commitment includes business level deployment of

resources to those functional areas that are necessary to achieve and maintain competitive

advantage (Cool and Schendel 1988).

There have been some nursing home studies that have utilized scope and resource

commitment as the parameters to identify whether strategic groups existed in the nursing home

industry. Zinn et al. 1994 used scope commitment measures of percent Medicare, percent

Medicaid, percent independent capacity, case mix, average length of stay and percent over 85.

Zinn et al. 1994 used resource commitment measures of size, private pay rate, and occupancy,

RN per resident and staff per resident. Using cluster analysis, they found that facilities with the

highest private pay residents had the best patient outcomes and the facilities with the highest

percent Medicare beds had better financial performance. Marlin et al. 1999 used Medicaid

utilization, Medicare utilization, insurance/health maintenance organization (HMO) utilization,









private-pay utilization, and Veteran's Administration utilization, average length of stay, average

patient age and case mix. For resource commitment, Marlin et al. 1999 used percentage of

nursing costs, percentage of ancillary costs, occupancy rate, semiprivate room rate, number of

beds, registered nurses (RNs) per resident, and staff per resident. The study also used cluster

analysis and found that the group with the highest private pay utilization combined with high

Medicare utilization generally performed better along financial and quality indicators.

Scope Using Technology

Prior studies of strategic groups in nursing homes have used the range of market segments

as the measure of scope commitment (Zinn et al. 1994, Marlin et al. 1999). Zinn et al. 1994 and

Marlin et al. 1999 argued that to compete effectively, nursing homes need to be responsive to

differences in payer demand characteristics because the payer mix of a nursing home is the major

basis of segmentation in the industry (Zinn et al. 1994, Marlin et al. 1999). However, using payer

demand characteristics may not be the best way of determining the strategic groups because the

payer mix of a nursing home may be more the result of good strategy. For example, many

facilities may have a strategy to acquire Medicare and private pay residents but not be successful.

Whether a facility has a high proportion Medicare and private pay beds may be independent of

the nursing homes' attempt to acquire these types of residents.

Structure Conduct Performance theory is a model used to link elements of the market

structure to business conduct and performance in industrial economics. Structure refers to market

structure defined mainly by the concentration of market share in the market. Conduct refers to

the behavior of firms whether competitive or collusive (pricing and production, production, goals

of firms, promotion) (Britton et al. 1992). Performance is mainly defined by the consequences of

market power. Based on the Structure Conduct Performance paradigm, payer demand

characteristics would be considered performance measures rather than conduct measures. In the









nursing home industry, measures of this type of performance include percent of beds filled by

Medicare patients and percent of beds filled by private pay residents. Both Medicare and private

pay clientele pay higher reimbursement rates than other payers like Medicaid. Market shares of

Medicare and/or private pay residents are the result of a strategy but do not give any information

about what strategy was used by the nursing home to accomplish that performance.

Unlike the aforementioned performance measures, technology focus and resource

deployment decisions are made by the business level management and are under the direct

control of management. Decisions to increase or decrease the amount of nursing technologies

provided (services like restorative care or skin ulcer prevention) can be made independent of the

environment that the facility is located in. However the resulting quality of care caused by

providing these services can influence the performance of the facility as measured by the ability

of the facility to draw higher reimbursing Medicare or private pay residents.

In the nursing home industry, technology includes all the processes of care that are

provided to the residents by the facility's healthcare staff. Nursing homes use their technology to

transform their inputs into outputs. The nursing home industry is different from other industries

because of the way technology is applied and the way the product is measured.

Resource Commitment

Resource commitment refers to the commitment of resources to functional areas that are

needed to gain and maintain competitive advantage in targeted market segments. For nursing

home administrators, resource commitment decisions should be reflected by labor, price and

capacity decisions (Zinn et al. 1994). Nursing homes have some control over their labor but they

have limited control over their products' price. Government reimbursement accounts for over

62% of their revenue (MedPAC 2007). After the implementation of the Balanced Budget Act of

1997, Medicare dictated the price that they are going to pay for the nursing home stay. Similarly









most states have changed their Medicaid reimbursement schematic in the direction of a

prospective payment system. Only about 15% of nursing home beds are private pay where a

facility can set their price. Capacity decisions can be impacted by economies of scale where the

decision to be larger can allow the nursing home to benefit from the efficiencies created by more

resources derived from size. Weech-Maldonado et al. 2008 found that larger facilities have better

financial performance. Industrial organization studies have found that firms following similar

strategies are likely to be of comparable scale (Porter 1980). In addition, labor and capacity can

also be related. With widespread nursing and therapist shortages, some facilities may not be able

to increase the capacity due to labor limitations.

Porter's Generic Strategies

Besides strategic group models, this study will examine the strategic-performance

relationship using a conceptual scheme that indicate the strategies that are available to firms.

Porter (1980) introduced the concept that there are certain generic types of strategies that can be

used by firms to outperform other organizations in the industry. Porter's generic strategies are

perhaps the most widely used and heavily researched. Firms have different types of

environmentally determined functional demands and thus choose different strategies in an effort

to gain competitive advantage (Porter 1980).

Originally Porter (1980) argued that a firm could choose one of three strategies to compete

at the business level. One of those strategies is differentiation where a nursing home

differentiates itself from other nursing homes in the market. One way facilities can differentiate

themselves from their competitors is providing a high quality product. In the nursing home

industry this equates to better care. By providing a high quality product, businesses are able to

shield themselves from reductions in price-cost margins (Porter, 1980). Weech-Maldonado et al.









(2003) study found that there was a positive relationship between quality of care and financial

performance in nursing homes.

A nursing home can also use cost leadership in which the nursing home attempts to have

relative lower costs. One would expect that nursing homes that follow the cost leadership

paradigm would be more efficient at using their staff

The third strategy is a focus strategy, in which the firm concentrates on a particular group

of customers, geographic markets, or product line segments. In the case of the nursing home

industry, some facilities may choose to focus on rehabilitation because of the higher revenues per

bed that Medicare pays.

Porter (1980) considered firms that attempted to apply both a differentiating strategy and a

cost leader strategy at the same time as 'stuck in the middle' or 'muddlers'. Stuck in the middle

or muddlers are considered those organization who have no coherent strategy. These

organizations did not satisfy the demands of either cost leaders or differentiators (Marlin et al.

2002). In contrast, Hill et al. 1988 and Wright et al. 1987 have posited that some nursing homes

can be successful at differentiating themselves from their competitors while controlling their

costs at the same time. Marlin et al. 2002 coined this type of nursing home as 'best costs' and

found that the best cost strategy existed in the hospital industry. Porter has acknowledged that on

rare occasions, firms can be successful with both differentiating strategy and a cost leader

strategy (Murray et al.1988).

Marlin et al. (2002) used strategic groups as defined by Porter's theory of competitive

strategy in the hospital industry. Marlin et al. 2002 used an objective classification procedure to

classify four strategic groups: differentiation, low cost, best costs and the muddlers. The

variables for the cluster analysis were selected by a panel of experts. Those variables selected to










capture cost leadership were measures that capture different types of cost like salaries and labor

hours. The variables used to capture the differentiation group were types of services. Marlin et al.

(2002) found that hospital strategic groups could be defined using Porter's generic strategies

theory and that different strategic groups had different structure-performance relationships.

Strategic Group Model Using Scope (Technology) Commitment, Resources Commitment
and Porter's Generic Strategy Theory

Figure #2 is a strategic group model of the nursing homes industry. The horizontal axis

represents the scope commitment of the nursing home as measured by the amount of technology

provided by the facility.

Upper Scope Boundary

High F\

SFocus


Lack of Strategy I Differentiator
Resource -1

R esorcF Best Cost

Coaster 1, Lower Resource Boundary


Low
Low High
Scope


Figure 2-2. Strategic Group Model Based on Scope and Resource commitment

The vertical axis represents the resources that the facilities commit to provide its

technology. The circles represent clusters as defined by Porter's strategies (Differentiator, Cost

Leadership, Best Cost and Focus). Those facilities that do not have a strategy are defined as

being groups that lack strategy









The center of the circle depicts the centroid of the cluster. To be in the cluster, a nursing

home is closer to the centroid of its cluster than the centroid of any other cluster in the industry.

All nursing homes within the groups are located within the outer boundaries of each cluster. A

nursing home is not able to be in two separate clusters.

The upper scope boundary is a vertical line that passes through the centroid of the cluster

that has the highest scope (technology) value. The lower resource boundary is a horizontal line

that passes through the centroid that has the lowest resources. There are nursing homes that fall

beyond both these boundaries; however the centroids of the clusters cannot fall beyond these

boundaries.

Differentiator

For a nursing home to differentiate itself from its competitors, it must provide superior

processes of care. By taking on the strategy of providing more service than the competitors in

their local markets, these nursing homes are more likely to attract residents to their facility

resulting in higher occupancy rates and choose residents that can provide them with higher

margins. A differentiating nursing home would provide a high volume of service relative to the

average nursing home in the market. In Figure #2, the differentiating strategic group would be

expected to be in the upper right quadrant of the graph. This is because the differentiating group

is defined as the highest provider of service (technology). If a high volume of service is

provided, it would be expected that facilities in the differentiation strategic groups would also

have to commit a greater amount of resources. The centroid of the cluster demarks the upper

scope boundary.

Cost Leadership

Some nursing homes will be focused on keeping their costs low. Since staffing is the

significant cost driver in the nursing home industry, keeping lower staffing levels is one obvious









method that facilities can use to control cost. RNs cost significantly more than CNAs and LPNs

and therefore by manipulating staffing levels to a lower skill mix, cost leadership nursing homes

are expected to keep their costs low.

In Figure #2, the cost leadership group would be expected to be located in the bottom left

quadrant of the graph. This is because the cost leadership group has the lowest use of resources.

It is expected that because of the lower availability of resources, cost leadership strategic groups

will provide less processes of care. The centroid of the cluster demarks the lower resource

boundary.

Best Cost

Some nursing homes may seem like a hybrid of the cost leadership group and the

differentiation group relative to the lack of strategy group. The best cost group represents

efficient organizations. In Figure #2, the best cost group would be expected to have it's centroid

between the cost leaders and the differentiators. More specifically, the best cost group is

expected to provide fewer services than the differentiating group but use fewer resources. It

would also be expected to provide more services than the cost leadership group using more

resources.

Focus

Like any industry, it is possible for a nursing home to focus their strategy on a smaller

segment of the industry. By allocating resources to that area, it may be possible to perform well

financially. In Figure #2, the centroid of the focus group could be located anywhere in the graph

within the upper scope boundary and the lowest resource boundary. However it is likely that the

focus group would provide more services than the cost leadership group because it would be

expected that it is focused on providing a specific type of service like rehab. There is more than

one type of technology that a nursing home can focus on providing. It is therefore possible that









within the nursing home industry, there exists more than one focus group. For example, a nursing

home could focus their attention on providing nursing care while a different one could focus

their attention on providing rehabilitation services.

Lack of Strategy

Facilities that use high levels of resources but provide low levels of technology fall into

the lack of strategy group. This group will also not have any focus of services and special

allocation of resources.

Measuring Performance

Financial performance metrics are the most commonly used when comparing different

strategic groups. Some of the more common measures of performance include total margin and

operating margin. Total margin is the total revenue divided by the total expense. This is an

important measure because any organization that is unable to sustain positive margins will cease

to exist. The total margin not only includes the operating costs and revenue, but other costs

(capital costs and interest) and other revenue (like endowments and donations). Operating

margin only includes operating revenue and operating cost. Operating margins are important

because they measure efficiency. The higher the operating margin, the more profitable a

company's core business. Operating margin is related to the day-to-day decisions that managers

make such as pricing strategy, prices for raw materials, and / or labor costs. Operating margin is

also a measure of managerial flexibility and competency of the firm (Gapenski 2000).

In the health care industry (including nursing homes), patient costs is also widely used as

a financial metric (Weech-Maldonado et al. 2003, Shen et al. 2007). Since public payers like

Medicare and Medicaid have been placing significant pressure on a facilities' ability to generate

revenue, it has been necessary for facilities to control their costs in order to achieve positive

margins.









In most industries, success is most commonly defined by financial metrics; however,

Venkatraman et al. 1986 have suggested that other measures such as market share, new product

introduction and product quality can be equally important.

One measurement that has recently gained popularity in business is the balanced

scorecard (BSC) approach. A major characteristic of a BSC is that it combines long-range

strategic financial goals with day-to day operations. Kaplan and Norton (2001) stated that the

customer's perspective and internal operations are important areas that should be considered

along with the financial perspective.

Taking the customer's perspective, the ability of a nursing home to provide quality care is

crucial if a facility is going to survive over the long term. Poor quality can result in higher costs

(Weech-Maldonado et al. 2003) as well as lower revenue streams due to lower occupancy rates.

Poor quality can also result in moratorium on new admissions and even a facility's closure.

Quality in nursing homes has also generated concern for policymakers and the public ever since

the release of the Institute of Medicine (IOM) report in 1986. It is therefore very important that

quality indicators be used in conjunction with financial variables when measuring performance.

There are three key reasons why measures other than financial performance are needed in

the nursing home industry. First, the government programs like Medicare and Medicaid were

established to pay for the care of the program's beneficiaries. Although enough funds must be

allocated to keep companies viable, the purpose of the program is not to give organizations

above normal profits. Second, not all health care organizations are for profit. For example, nearly

one third of nursing homes in the US are either government or not for profit facilities (Harrington

et al. 2001) and it is important to consider what health care organizations do. The presence of

not-for-profit in the industry is another reason why performance objectives other than profit









maximization are needed (Scanlon, 1980). Finally, health care organizations (like nursing

homes) are often charged with the responsibility of providing care for one of the most vulnerable

populations. The direness of poor quality health care is so great that using financial performance

as the sole measure of success (like in other industries) is just not adequate. For the

aforementioned reasons, the measurement success in the health care industry has to be based on

quality of services provided, cost of that service and financial performance.

New Contributions

To date, no study has examined if strategic groups can be identified by the technology

choices made by the nursing home management. The purpose of this study is to determine if

strategic groups can be classified using Porter's generic strategies based on how nursing home

management allocate their resources and how they select their technology. The study will also

determine if these groups are succinct and if they differ with respect to quality, financial

performance and cost. What and how much of technology a nursing home administrator chooses

to provide is an important strategy choice that can have significant repercussions. The study is

different from prior studies because it uses the industry's technology as the scope variables to

determine the strategic groups and will define those groups using Porter's generic strategies.

Hypothesis

Hypothesis #1: Nursing homes that have a strategy can be categorized into one of four

strategic groups: Differentiator, cost leadership, focus or best cost.

Hypothesis #2: Nursing homes in the differentiator strategic group will provide higher

quality care than other strategic groups.

Hypothesis #3: Nursing homes in focus strategic group will provide higher quality in the

area that they are focused on than other strategic groups.









Hypothesis #4: Nursing homes in the cost leadership strategic group will have lower

costs than other strategic groups.

Hypothesis #5: Nursing homes that lack a strategy will have lower financial performance

than nursing homes that are in a strategic group.







CHAPTER 3
METHODOLOGY

Data

This study combined the On-line Survey Certification of Automated Records (OSCAR),

the Minimum Data Set (MDS), the Area Resource File (ARF) and the Medicare Cost Reports.

The OSCAR dataset is a facility level dataset that is routinely collected through the Medicare

and Medicaid certification process conducted by state licensure and certification agencies. This

data is updated annually as part of the recertification survey. OSCAR provides a snapshot about

the characteristics and care processes occurring within the nursing homes at the time of the

survey. It includes variables like staffing levels, ownership status, number of residents, facility

acuity level as well as quality outcome measures like state survey care deficiencies and pressure

sore prevalence.

The Omnibus Budget Reconciliation Act of 1987 (OBRA) was the first major legislative

step to improve the quality of care in nursing homes. It established the Resident Assessment

Instrument (RAI) which was implemented in 1991. The RAI is the basis of the Minimum Data

Set (MDS) which is a dataset that records information about nursing home residents. The MDS

includes demographic information, health status as well as amount and types of services that a

resident receives while they reside at the nursing home. Each long term care resident is assessed

when first admitted to a nursing home, then each quarter thereafter. The post acute resident is

assessed on the 5th day, the 14th day, the 30th day, 60th day and 90th day (for as long they remain

in the facility). The CMS MDS is the most comprehensive dataset available for examining

quality outcomes in the long-term care setting. OSCAR and Quality Indicators (QI) generated

from the MDS are used by the CMS as part of its Nursing Home Compare website.

The MDS was acquired using a reuse agreement from the CMS. Data is currently located

behind a firewall in a fully secured server of the College of Public Health and Health Professions







building operated by the IT department at the University of Florida. It was originally acquired by

University of South Florida.

The Area Resource File (ARF) contains county level data on supply and demand factors,

such as composition of health care workforce, and socioeconomic and demographic

characteristics of the population.

The Medicare Cost Report data is a public access relational dataset that can be acquired

from the CMS via the CMS website. All CMS certified nursing homes that have Medicare beds

are required to submit the Medicare cost report on an annual basis. MedPAC has repeatedly been

using this dataset to report the financial information of nursing homes to congress over the past 7

years (MedPAC 2007). This dataset provides financial data like total patient revenue, total

patient costs, and total revenue.

Population

This study included all nursing homes in the United States that had Center for Medicaid

and Medicare certification and that were included in the OSCAR, MDS and Medicare cost report

datasets in 2004. There were 16000+ nursing homes in the United States in 2004. However, a

total of 4500 facilities were omitted because they were either hospital based (2000 facilities) or

government based (700 facilities), did not have Medicare beds (1200 facilities) or were the only

nursing home in a county and would not have any market competition (500 facilities). Hospital

based facilities were omitted because they may behave differently from free standing facilities

(Weech-Maldonado et al. 2003, Weech-Maldonado et al. 2004). For example, hospitals may

place patients in their own subacute beds that they cannot place in other nursing homes due to

the level of acuity of the patient or because the patient lacks insurance. Government facilities

may behave differently than non-government facilities because of their different financing

structure. They are less influenced by market forces because of funding that they may receive

directly from public sources. Finally, some facilities that may be included in the OSCAR may

44







not have any Medicare residents or even Medicare beds available and therefore they would not

have to file a Medicare cost report. Therefore, this study is limited to nursing homes that are non

government, free standing and have Medicare certified beds. The total number of facilities in the

study is approximately 11601 facilities in the year 2004.

Overview of Methodology

To identify the strategic groups and the strategy-performance relationships of the

strategic groups, the study performed a factor analysis on the scope and resource commitment

variables. The resulting variables were used in a cluster analysis to identify the different clusters.

Once clusters were identified, ordinary least squares regression (OLS), negative binomial and

ordered logit was used to determine if the different groups had different financial and quality

strategic-performance relationships.

Hypothesis #1: Nursing homes that have a strategy can be categorized into one of four
strategic groups: Differentiator, cost leadership, focus or best cost.

Scope

Similar to Zinn et al. (1994) and Marlin et al. (1999), this study used the scope

commitment and resource commitment (also known as resource deployment) dimensions to

identify the strategic groups. The scope of this study included technologies used by nursing

homes to take care of the residents. The technology of a nursing home can be classified into two

main categories: nursing technology and rehab technology.

Nursing Technology

Nursing technology includes many different types of processes. The overall care of the residents

can be divided into several different component technologies that are specifically designed to

address the varied needs of the nursing home resident. Each of these technologies has a specific

purpose and if not applied properly will have an undesired outcome. Some of the essential

technologies include skin care technology, restorative technology, incontinence reducing

technologies and nutritional technologies.







Skin care technology is designed to prevent pressure ulcers from occurring and to treat

pressure ulcers that already exist. Failure to apply preventative skin care technology properly can

result in pressure ulcers. Failure to treat already existing pressure ulcers can result in pressure

ulcer worsening. Restorative nursing technologies engage the resident in functional activities.

Residents functional abilities can decline in the nursing home setting and restorative technology

can be used to help maintain their functional levels. Incontinence reducing technologies assist

residents to maintain or regain their continence through toileting programs or bladder retraining

programs. Nutritional technologies help residence maintain a healthy weight. In the nursing

home setting, residents are at risk of either gaining weight or losing weight depending on their

medical condition. Weight management programs and therapeutic diets assist with proper

nutrition while the use of plate guards and other devices ensures that residents adequately feed

themselves.

Skin care technology: All processes of care that involve either preventative skin care or

treatment skin care are considered skin care technology.

Pressure relieving seat: This is a continuous facility level variable that represents the

proportion of residents in the nursing home that are given a pressure relieving seat. Pressure

relieving seats are an important intervention when trying to prevent a new pressure sore from

occurring or to prevent an already acquired pressure sore from worsening. Includes gel, air (e.g.,

Roho), or other cushioning placed on a chair or wheelchair. Include pressure relieving, pressure

reducing, and pressure redistributing devices.

Pressure relieving bed: This is a continuous facility level variable that represents the

proportion of residents in the nursing home that are given a pressure relieving bed. Pressure

relieving beds are an important intervention to prevent new pressure sores from occurring or to

prevent an already acquired pressure sore from worsening. Includes air fluidized, low air loss







therapy beds, flotation, water, or bubble mattress or pad placed on the bed. Include pressure

relieving, pressure reducing, and pressure redistributing devices.

Turning /Repositioning: This is a continuous facility level variable that represents the

proportion of residents in the nursing home that are placed on a turning-repositioning schedule.

Turning-repositioning schedules are an important intervention to prevent a new pressure sore

from occurring or to prevent an already acquired pressure sore from worsening. It includes a

continuous, consistent program for changing the resident's position and realigning the body.

"Program" is defined as "a specific approach that is organized, planned, documented, monitored,

and evaluated."

Ulcer care: This is a continuous facility level variable that represents the proportion of

residents in the nursing home that are receiving ulcer care treatment. Residents who get ulcer

care already have ulcers.

Surgical care: This is a continuous facility level variable that represents the proportion of

residents in the nursing home that are receiving surgical care treatment. Residents who get

surgical care already have ulcers.

Ointment: This is a continuous facility level variable that represents the proportion of

residents in the nursing home that are receiving ointment care. It includes ointments or

medications used to treat a skin condition (e.g., cortisone, antifungal preparations,

chemotherapeutic agents, etc.).

Dressing care: This is a continuous facility level variable that represents the proportion of

residents in the nursing home that are receiving dressing care. Residents who get dressing care

already have ulcers.

N\ki nutrition program care: This is a continuous facility level variable that represents the

proportion of residents in the nursing home that are on a skin nutrition program. Dietary

measures received by the resident for the purpose of preventing or treating specific skin

47







conditions e.g., wheat-free diet to prevent allergic dermatitis, high calorie diet with added

supplements to prevent skin breakdown, high protein supplements for wound healing. Vitamins

and minerals, such as Vitamin C and Zinc, which are used to mange a potential or active skin

problem, should be coded here.

Other skin care: This is a continuous facility level variable that represents the proportion

of residents in the nursing home that are receiving other skin care. Residents who get other skin

care already have ulcers.

Restorative technology: All processes of care provided by the nursing staff that involves

having the patient engage in an activity that is directed towards accomplishing an ADL goal that

will result in improved health status.

Restorative ambulation: The restorative ambulation variable is facility level and is

continuous. It represents the average number of days that a resident receives restorative

ambulation while remaining in the facility. It is generated by dividing the resident level

restorative variable that represents the number of days of ambulation provided in the 7 days

before the assessment date by the total number of residents in the facility. Restorative ambulation

has the nursing staff walk the residents in order to improve the distance that the resident can

ambulate. It is a routine service and does not require the skill of the therapist. However the goals

of the program need to be generated by the therapist.

Restorative Passive Range of Motion (PROM): The restorative PROM variable is facility

level and is continuous. It represents the average number of days that a resident receives

restorative PROM while remaining in the facility. It is generated by dividing the resident level

restorative variable that represents the number of days of PROM provided in the 7 days before

the assessment date by the total number of residents in the facility. Restorative PROM has the

nursing staff range the residents in order to improve the PROM of the resident. It is a routine







service and does not require the skill of the therapist. However the goals of the program need to

be generated by the therapist.

Restorative Active Range ofMotion (AROM): The restorative AROM variable is facility

level and is continuous. It represents the average number of days that a resident receives

restorative AROM while remaining in the facility. It is generated by dividing the resident level

restorative variable that represents the number of days of AROM provided in the 7 days before

the assessment date by the total number of residents in the facility. Restorative AROM has the

nursing staff range the residents in order to improve the AROM of the resident. It is a routine

service and does not require the skill of the therapist. However the goals of the program need to

be generated by the therapist.

Restorative transfers: The restorative transfer variable is facility level and is continuous. It

represents the average number of days that a resident receives restorative transfers while

remaining in the facility. It is generated by dividing the resident level restorative variable that

represents the number of days of transfers provided in the 7 days before the assessment date by

the total number of residents in the facility. Restorative transfers have the nursing staff transfer

the residents in order to improve the resident's ability to transfer from one position to another. It

is a routine service and does not require the skill of the therapist. However the goals of the

program need to be generated by the therapist.

Restorative bed mobility: The restorative bed mobility variable is facility level and is

continuous. It represents the average number of days that a resident receives restorative bed

mobility while remaining in the facility. It is generated by dividing the resident level restorative

variable that represents the number of days of bed mobility provided in the 7 days before the

assessment date by the total number of residents in the facility. Restorative bed mobility has the

nursing staff transfer the residents in order to improve the resident's ability to transfer from one







position to another in bed. It is a routine service and does not require the skill of the therapist.

However the goals of the program need to be generated by the therapist.

Restorative dressing: The restorative dressing variable is facility level and is continuous. It

represents the average number of days that a resident receives restorative dressing while

remaining in the facility. It is generated by dividing the resident level restorative variable that

represents the number of days of dressing provided in the 7 days before the assessment date by

the total number of residents in the facility. Restorative dressing have the nursing staff help dress

the residents in order to improve the resident's ability to dress by themselves. It is a routine

service and does not require the skill of the therapist. However the goals of the program need to

be generated by the therapist.

Restorative eating: The restorative eating variable is facility level and is continuous. It

represents the average number of days that a resident receives restorative eating while remaining

in the facility. It is generated by dividing the resident level restorative variable that represents the

number of days of eating provided in the 7 days before the assessment date by the total number

of residents in the facility. Restorative eating have the nursing staff help the residents with eating

in order to improve the resident's ability to eat by themselves. It is a routine service and does not

require the skill of the therapist. However the goals of the program need to be generated by the

therapist.

Restorative splinting: The restorative splinting variable is facility level and is continuous.

It represents the average number of days that a resident receives restorative splinting while

remaining in the facility. It is generated by dividing the resident level restorative variable that

represents the number of days of splinting provided in the 7 days before the assessment date by

the total number of residents in the facility. Restorative splinting have the nursing staff assist

with the donning and doffing of splints in order to improve the resident's ability to splint







themselves. It is a routine service and does not require the skill of the therapist. However the

goals of the program need to be generated by the therapist.

Restorative amputee intervention: The restorative amputee intervention variable is facility

level and is continuous. It represents the average number of days that a resident receives

restorative amputee intervention while remaining in the facility. It is generated by dividing the

resident level restorative variable that represents the number of days of amputee intervention

provided in the 7 days before the assessment date by the total number of residents in the facility.

Restorative amputee intervention have the nursing staff assist with the donning and doffing of

prosthetic in order to improve the resident's ability to don and doff their prosthetic by

themselves. It is a routine service and does not require the skill of the therapist. However the

goals of the program need to be generated by the therapist.

Restorative communication: The restorative communication variable is facility level and is

continuous. It represents the average number of days that a resident receives restorative

communication while remaining in the facility. It is generated by dividing the resident level

restorative variable that represents the number of days of communication provided in the 7 days

before the assessment date by the total number of residents in the facility. Restorative

communication has the nursing staff assist with the communication in order to improve the

resident's ability to communicate. It is a routine service and does not require the skill of the

therapist. However the goals of the program need to be generated by the therapist.

Restorative other treatment: The restorative other treatment variable is facility level and is

continuous. It represents the average number of days that a resident receives other restorative

while remaining in the facility. It is generated by dividing the resident level restorative variable

that represents the number of days of other provided in the 7 days before the assessment date by

the total number of residents in the facility.







Incontinence reducing technologies: All processes of care provided by the nursing staff

intended for the patient engage in a program that is directed towards accomplishing continence.

Toiletingprogram: This is a continuous facility level variable that represents the

proportion of residents in the nursing home that have a toilet program. Toileting programs

include such items like toileting schedules.

Bladder retraining: This is a continuous facility level variable that represents the

proportion of residents in the nursing home that have a bladder retraining program. Bladder

training programs include such items like toileting schedules.

Nutrition technology: All processes of care provided by the nursing staff intended for the

patient engage in a program that is directed towards maintaining a healthy weight.

Weight management program: This is a continuous facility level variable that represents

the proportion of residents in the nursing home that have a weight management program.

Resident is receiving a program of which the documented purpose and goal are to facilitate

weight gain or loss (e.g., double portions; high calorie supplements; reduced calories; 10 grams

fat).

Therapeutic diet: This is a continuous facility level variable that measures the proportion

of residents that are on a therapy diet. A diet ordered to manage problematic health conditions.

Examples include calorie-specific, low-salt, low-fat lactose, no added sugar, and supplements

during meals.

Plate guard: This is a continuous facility level variable that represents the proportion of

residents in the nursing home having a plateguard. This includes any type of specialized, altered,

or adaptive equipment to facilitate the resident's involvement in self performance of eating.

Rehab Technology

Rehab technology is utilized to improve the resident's health status. It is provided by

licensed therapy staff. There are different types of therapist that have different functions. These

52







processes of care include physical therapy for mobility issues and occupational therapy for ADL

issues other than mobility.

Long term care rehabilitation: All processes of care provided by the therapy staff that

involves rehabilitation services to the long term care residents.

Mean PT minutes long term care per resident: Facility level variable that represents the

average number of minutes of physical therapy per resident provided in the 7 days prior to the

quarterly assessment date. The numerator is the number of minutes of PT and the denominator is

the number of residents in the nursing home. This is a MDS resident level variable that is raised

to the facility level.

Mean OTminutes long term care per resident: Facility level variable that represents the

average number of minutes of occupational therapy per resident provided in the 7 days prior to

the quarterly assessment date. The numerator is the number of minutes of OT and the

denominator is the number of residents in the nursing home. This is a MDS resident level

variable that is raised to the facility level.

Post acute rehabilitation: All processes of care provided by the therapy staff that involves

rehabilitation services to the long term care residents.

Mean PT minutes post acute per resident: Facility level variable that represents the

average number of minutes of Physical Therapy the 7 days prior to the 14 day assessment date.

The numerator is the number of minutes of PT and the denominator is the number of residents in

the nursing home. This is a MDS resident level variable that is raised to the facility level.

Mean OTminutes post acute per resident: Facility level variable that represents the

average number of minutes of Occupational Therapy provided in the 7 days prior to the 14 day

assessment date. The numerator is the number of minutes of PT and the denominator is the

number of residents in the nursing home. This is a MDS resident level variable that is raised to

the facility level.







Percent Medicare beds: Facility level variable that represents the percent of the total

number of resident beds reimbursed by Medicare. Medicare only reimburses for resident who are

at the facility for short term stay. This variable is used because it is necessary to account for the

proportion of the patients that are receiving rehabilitation at the facility. It is generated by

dividing number of Medicare residents by the total number of beds for that facility.

Resource Commitment

Resource commitment refers to the commitment of resources to the different technology

areas that are needed to gain and maintain competitive advantage in targeted market segments

(Zinn et al. 1994). The primary resource is staffing levels (the nurses and the therapist) since it is

the health care staff that provides the technology. Therefore, organizations manipulate their

staffing levels to control costs and ensure quality in an attempt to perform well financially.

This study divided the resource commitment into two separate subtypes because they are

mutually exclusive of each other and do not overlap.

Resource commitment nursing: Resource commitment nursing subtype includes all the

licensed and non licensed nursing staff. They are responsible for all the nursing technologies

RNFTE / Total Residents: Registered Nurse (RN) per resident is a continuous variable

representing the ratio of the total number of RN FTEs divided by the total number of residents in

the facility. RNs are involved in establishing processes and supervising other nursing staff.

LPNFTE /Total Residents: Licensed practical nurse (LPN) per resident is a continuous

variable representing the ratio of the total number of LPN FTEs divided by the total number of

residents in the facility. Both measures are from the OSCAR dataset. LPNs' are able to provide

some skilled nursing processes of care like dispensing medication. They have to be under the

supervision of a nurse.

CNA FTE / Total Residents: Certified Nursing Assistant (CNA) per resident is a

continuous variable representing the ratio of the total number of CNA FTEs divided by the total

54







number of residents in the facility. Both measures are from the OSCAR dataset. CNAs' are

mostly involved with the direct care of patients.

Percent Licensed: Percent nursing staff that are licensed is a continuous staffing mix

variables that measures the proportion of nursing staff that have a license. Higher percent

licensed will give the nursing home a higher capacity to have knowledge and intangible

resources.

Percent of nursing staff that are RNs: Percent of nursing staff that are RNs is a continuous

staffing mix variable that measures the proportion of the nursing staff that are RNs. Higher RN

staffing mixes are advantageous since RNs can use their professional skills and tacit knowledge

to improve the quality of patient care through their involvement in establishing processes and

supervising other nursing staff (Weech-Maldonado et al. 2003).

Resource commitment rehab: Resource commitment rehab includes all therapists and

therapy assistants. Only licensed rehab staff is included because the use of rehab aides for

therapy treatments is minimal due to state and federal practice guidelines.

PTFTE/ Total Residents: Physical Therapist (PT) per resident is a continuous variable

represents the ratio total number of PT FTEs divided by the total number of residents in the

facility. PTs evaluate the patient and oversee the plan of care. They supervise the physical

therapy assistants

PTA FTE/ Total Residents: Physical Therapy Assistants (PTA) per resident is a continuous

variable represents the ratio total number of PTA FTEs divided by the total number of residents

in the facility. PTAs provide the treatment to the patients under the supervision of the physical

therapist.

Percent PT: Percent of the physical therapy staff that are PTs is a continuous staffing mix

variable. Higher PT staffing mixes are advantageous since PTs can use their professional skills







and tacit knowledge to improve the quality of patient care through their involvement in

establishing processes and supervising other physical therapy staff

OTFTE /Total Residents: Occupational Therapist (OT) per resident is a continuous

variable represents the ratio total number of OT FTEs divided by the total number of residents in

the facility. OT evaluates the patient and oversees the plan of care. They supervise the Certified

Occupational Therapy Assistant

COTA FTE /Total Residents: Certified Occupational Therapy Assistant (COTA) per

resident is a continuous variable represents the ratio total number of COTA FTEs divided by the

total number of residents in the facility. COTAs provide the treatment to the patients under the

supervision of the occupational therapist.

Percent OT: Percent of the occupational therapy staff that are OTs is a continuous staffing

mix variable. Higher OT staffing mixes are advantageous since OTs can use their professional

skills and tacit knowledge to improve the quality of patient care through their involvement in

establishing processes and supervising other occupational therapy staff

Wages: Nursing homes average wages is a continuous variable derived by dividing the

total salary expense by the total number of working hours reported in the Medicare costs reports.

Wages are considered resources measures because nursing homes utilize the wages to control

their cost.

Hypothesis #1 Analysis

Since there was a total of 68 variables of interest (technology and resource variables), it

was necessary to reduce the number of variables into composite variables by using factor

analysis. The internal consistency of the newly created composite variables was tested using

Cronbach's (1951) alpha methodology. Some variables that were not included in the composite

scores were still included in the cluster analysis as individual variables.







The composite (or index) and the appropriate individual variables were transformed into

cluster variables using z-scores. Wards minimum variance method was used to identify the

number of clusters and k-means cluster analysis to identify the different strategic groups. The

analysis of variance and covariance (ANOVA) statistical test with Tukey's comparison test was

used for each variable of interest to determine if the strategic groups were significantly different

from each other. The Multivariate analysis of variance and covariance (MANOVA) test with

Wilk's Lambda was used to determine if the overall clusters were significantly different from

each other. Each variable was given a weighted rank score for each cluster. The weighted rank

score was then aggregated by type of commitment (scope or resource). Using the aggregated

weighted rank score, the clusters were classified based on Porter's generic strategies as defined

in the technology, resources and strategic group model presented in Figure #2.

Factor Analysis

Factor analysis is a variable reducing technique. It is a family of procedures designed to

remove the redundancy from a set of highly correlated variables and representing the variables in

a smaller set of derived variables or 'factors' (Kachigan 1991). Factor analysis can be seen as the

grouping of similar variables (Kachigan 1991). These groupings of variables (or factors) can be

used to describe a particular construct. For example, in the nursing home industry, two of the

treatments used in an attempt to decrease pressure sores include placing a pressure relieving

device in a chair and placing a pressure relieving device in a bed. It would be expected that these

two processes of care are going to be highly correlated since both processes are similar in nature.

The study used exploratory factor analysis to identify scope commitment and resource

commitment variables that are correlated to each other. These variables were combined into new

index variables that were a measure of what was considered a technology or resource factor. In

the above example, if the factor analysis determined that the two pressure relieving device







variables are correlated and need to be combined, the new variable may be considered a measure

of the construct 'Pressure Sore Preventative Interventions'.

The study used the interaction between the scope and the resource dimensions to classify

the strategic groups. It was therefore necessary to perform separate exploratory factor analysis

for scope commitment and for resource commitment. Similarly, the staffing intensity and staffing

mix variables were also analyzed separately. This is because it is necessary to separate the

intensity and the skill mix of the staff to be able to identify the relationship between licensed and

non licensed staff. Staffing mix variables cannot be used as a substitute for the staffing intensity

(i.e. nurse per resident) variables because they do not explain the volume of nursing care

provided. For example, in the case of nursing, CNA's can be negatively correlated with the RNs

in one cluster while being positively correlated in another cluster. As a result, the three separate

factor analysis include technology variables, staffing intensity variables and skill mix variables.

Variables included in the cluster analysis but t not in the factor analysis were wages and

percent Medicare beds. Although average wage paid is a resource commitment measure and is

related to the staffing mix variables (higher RN mix represents higher average wages), it was not

included in any exploratory factor analysis. Wages are used by nursing homes as an incentive for

hiring new staff and retaining their staff and therefore were included separately in the cluster

analysis. Percent Medicare beds variable is related to scope because it indicates the proportion of

the facility residents getting rehab. It was included in the cluster analysis independently to

identify the total volume of therapy services provided. For example, a facility that provided a lot

of minutes of therapy to a small number of residents could be misclassified as being a rehab

facility if the total amount of residents receiving therapy were not accounted for.

Factor Matrix

In the factor matrix, the columns represent the derived factors while the rows are the

variables. The cells represent the factor loadings which represents the degree that the variables

58







relate to the derived factor. Factor loadings are used to reveal the extent that the variables

contribute to the meaning of the variable. Factor loading are the correlation coefficients between

the variables and the factors. A factor loading score of greater than 0.4 was the inclusion

criterion to be considered a part of that factor (Kachigan 1991).

Number-of-Factors Problem

One concern when using factor analysis is how many factors are retained. Each factor

accounts for a certain amount of variance in the data. Typically, the first factor will extract the

most variance; the second factor will extract the second most and so on. Once the variance of

each successive factor extracts is known, it is possible to determine how many factors can be

retained. There are two primary methods that can be used to determine the amount of factors

retained:

1) Eigenvalue is a measure of the variation and represents the equivalent number of

variables that the factor represents. The Kaiser criterion is a commonly used standard that

includes only factors with eigenvalues greater than 1 (Kachigan 1991). The rationale for using

such a threshold is that unless a factor extracts at least as much variation as the equivalent of one

original variable, it should be dropped. This criterion was proposed by Kaiser (1960), and is

probably the one most widely used standard in determining how many factors should be used.

2) Scree plot is another method that can be used to identify the number of factors. It is also

possible to identify how many factors to retain by plotting the incremental variance accounted

for by each successive factor. The idea is that the tail of the curve represents mostly random error

variance and therefore the factor solution is to include all factors that exist just prior to the

leveling off of the curve (Kachigan 1991).

Rotation of the Factor Structure

We could plot the factor loadings shown above in a scatter plot. In that plot, each variable

is represented as a point. In this plot we could rotate the axes in any direction without changing







the relative locations of the points to each other; however, the actual coordinates of the points,

that is, the factor loadings would of course change (Kachigan 1991).

Rotational strategies: There are various rotational strategies that have been proposed. The

goal of all of these strategies is to obtain a clear pattern of loadings, that is, factors that are

somehow clearly marked by high loadings for some variables and low loadings for others.

Fabrigar et al. 1999 describe two commonly used types of rotational strategies: orthogonal

and oblique. Although orthogonal is conceptually simpler, factors have to be uncorrelated.

Oblique rotation on the other hand does not have this limitation. Since the resource factors have

the potential of being correlated to the technology that they are used for (for example, therapy

resources will be correlated with rehab technology), this study used the oblique rotation.

Internal Consistency of Measures

It was necessary to determine whether the variables of the generated factors have internal

consistency. Cronbach alpha tested the correlation between the variables of the construct. The

higher the alpha value, the closer the variables are correlated (Cronbach 1951). This value should

be above 0.7 (Allen 2002). If the alpha was lower than 0.7, the factor was dropped because the

variables did not correlate well together. To determine how much impact the individual variables

had on the overall alpha, multiple tests will provide a score of the resulting alpha. Any variable

whose exclusion causes the alpha score to rise above 0.7 were be dropped. Similarly, any

variable whose inclusion causes the alpha score to drop below 0.7 were dropped.

Cluster Analysis

Cluster analysis is an exploratory data analysis tool which aims at sorting different objects

into groups in a way that the degree of association between two objects is maximal if they belong

to the same group and minimal otherwise (Kachigan 1991). Computationally, cluster analysis

could be thought of as analysis of variance (ANOVA) "in reverse." The program starts with k

random clusters, and then move objects between those clusters with the goal to 1) minimize







variability within clusters and 2) maximize variability between clusters. In other words, the

similarity rules apply maximally to the members of one cluster and minimally to members

belonging to the rest of the clusters. Cluster analysis can also be described as a set of techniques

for partitioning a set of objects into relatively homogenous subsets based on inner objects

similarities. The process begins by measuring a set number of objects (nursing homes) on each of

the variables of interest. A measure of difference between each pair of nursing homes is

obtained. Porter's competitive strategy typology was used to define the strategic groups of

nursing homes. Similar to Marlin et al. 1999 and Zinn et al. 1994, this study used the scope and

resource commitment measures in the cluster analysis.

Prior studies argue that due to the large interstate variation in legislation, they could only

use one state for their analysis (Zinn et al. 1994, Marlin et al. 1999). Although it is true that

states have different governing laws, different reimbursement systems and different survey

methodologies, limiting the analysis to one state limits the generalizability of the results. Rather

than limiting to one state, this study standardized all variables using z-score based on the mean

of the nursing home market (county). Regulation and reimbursement may vary not only by state

but also by county. Similarly income and demographics can vary by county. All these factors

may influence the nursing home's strategy on how to use their resources and how much

technology to provide. Therefore it is necessary to compare nursing homes to others in their

county. Z-score or normal score is a dimensionless quantity derived by subtracting the

population mean from an individual raw score and then dividing the difference by the population

standard deviation. In this study, the county mean and the county standard deviation were used.

This study used an inductive methodology to determine the strategic group clusters. K-

means is a common method used to determine clusters using inductive reasoning (Ketchen 1996,

Harrigan 1981). A significant number of studies have used k-means to identify the strategic

groups (Marlin et al. 1999, Ford et al. 1998, Zinn et al. 1994, Short et al. 2002, Dess et al. 1984).

61







K-means cluster analysis was specifically designed for large datasets (Rudolph et al. 1991).

Since this study began with over 11500 observations, K means cluster analysis was the most

appropriate. K-means cluster analysis requires the researcher to establish the number of clusters

apriori. This presents a challenge because it is not known if the number of clusters selected a

priori is the most well defined clusters. Using theory to determine the number of clusters is

considered to be a valid method to identify the k-means apriori (Ketchen et al. 1996). However,

Ketchen et al. (1994) also recommended using a quantitative method in conjunction with the

inductive method to increase the robustness of the number of clusters selected. One such method

is the Ward's minimum variance method which has been used by numerous strategic group

studies (Marlin et al. 1999, Ford et al. 1998, Zinn et al. 1994, Short et al. 2002).

One problem with the Ward's minimum variance method is that it functions poorly with

very large datasets and it is very susceptible to outliers. FASTCLUS was used as technique to

reduce the variables into a smaller number of clusters that would be more manageable for the

Ward's minimum variance method (SAS 1990). FASTCLUS uses k-means cluster analysis to

reduce the number of observations to a level that the Wards minimum variance can manage. In

this study, the initial FASTCLUS reduced the dataset to 100 cluster/observations. The study also

removed outlier observations that were greater than five standard deviations.

Ward's minimum variance method was used for determining the number of clusters to be

used. The study used the following decision criteria to identify the optimal cluster solutions

(SAS 1990).

1. a local peak in the cubic clustering criterion

2. a local peak in the pseudo F statistic,

3. a small value of the pseudo t2 statistic and a larger pseudo t2 statistic for the next cluster
solution level,

4. an additional cluster increasing the overall fit by less than 5 percent, and

5. The clusters obtained explaining at least 65 percent of the overall variance.

62







These decision criteria are consistent with those used in prior strategic group research (i.e.,

Fiegenbaum and Thomas 1990; Mehra 1996; Marlin 1999).

The number of clusters identified by the Wards minimum variance method was used in the

k-means cluster analysis using the original dataset.

Once clusters were identified, the means for each one of the variables used to identify the

cluster. These mean values were used to determine which of the Porter's generic strategies the

cluster could be categorized into.

Strategic Groups Defined

As a result of using z scores for the cluster analysis, the variable values had a mean score

of zero. In order for the model in Figure #2 to be utilized, it was necessary to convert the values

into a positive scale. Therefore the clusters were ranked from highest to lowest from 0 to the total

number of clusters that were determined in the cluster analysis. For example, if there were a total

of 3 clusters identified in the analysis, then the ranking would range from 0 to 3.

The actual distance between clusters for each variable was not equal and therefore it was

necessary to weight the rank score relative to the total variance of the variable. The ranking

protocol of each variable was determined as follows:

Determine the total variance value

The total variance for each variable was determined by subtracting the lowest mean value

from the cluster with the highest mean value. For example, if for a given variable, the cluster

with the lowest values was -1.5 and the cluster with the highest value was +1.5. Therefore, the

total variance value was 3.

Cluster value

For each variable, a cluster value was determined by subtracting the cluster with the

lowest value from each individual cluster variable's value. Using the above example of the







lowest value being -1.5 and a third cluster variable having a value -0.5, the cluster value of this

cluster value would be 1 (-0.5-(-1.5)).

Cluster rank score

The cluster value was divided by the total variance value. This ratio was then multiplied

by the total number of clusters. Using the above information and assuming only three clusters,

the cluster rank score would be 1/3 multiplied by 3 for a rank score of 1. For this case, the rank

scores would be 0 for the lowest rank, 1 for the middle rank and 3 for the top rank. Table A-8 in

the appendix includes all the total variance and cluster values for each variable.

All resource variables and all scope variables were summed to generate a cumulative

resource utilization score and cumulative scope variable score each cluster. These total values

were then plotted on the model as illustrated in Figure 2. Each cluster was classified into

strategies using the parameters defined below.

Differentiator: For a nursing home to differentiate itself from the other facilities in its

market, it would provide the greatest amount care. It is expected that a differentiating nursing

home would be highly ranked in the skin care technology, restorative nursing technology, other

care technology such as bladder training or weight program, and highest rehab care relative to

the average nursing home in their county. To be considered a differentiator, the cluster would

have to have the highest cumulative scope score.

Cost leadership: Some nursing homes will be focused on keeping their costs low. Since

staffing is the significant cost driver in the nursing home industry, strategies to have lower

staffing levels can be used as a measure of controlling cost. Therefore it was expected that the

cost leadership group would have the lowest cumulative rank scores for therapy resources,

nursing resources and wages.

Best cost: Some nursing homes may seem like a hybrid of the cost leadership group and

the differentiator group relative to the lack of strategy group. It was expected that this group

64







would have higher cumulative rank scores for resources and scope when compared to the cost

leadership group and would have lower cumulative resources and scope scores than the

differentiator. In Figure #2, the centroid best cost cluster would have fallen in the area between

the boundaries set by the cost leadership's centroid and the differentiating cluster's centroid.

Focus: Like any industry, it is possible for a nursing home to focus their strategy on a

smaller market segment. The focus group can be identified by providing a high amount of

technology and resources in a specific area like rehab services. For example, if a cluster had

high rank scores in post acute rehab, long term care rehab and therapy resources, but low scores

in every other category; they would be considered a rehab focus group. There was the possibility

that other focus groups also existed.

Lack of strategy: A facility that uses more resources but provides less technology than a

cost leader was considered as a facility that lacks a strategy.

Analysis of Variance

This study used Analysis of Variance (ANOVA) to determine if the component variables

of the newly identified strategic groups are significantly different from each other. The study

also used Tukey's method to make multiple pair wise comparisons of the means for each

variable. The ANOVA used the F test statistic which is the ratio of the between error estimates

divided by the within error estimates. If Ho is false, the between estimates tend to overestimate

the variance, so it tends to be larger than the within estimates. This can make the F statistic

significantly greater than 1.0 (Agresti and Finlay 1997). The Ho Hypothesis is that the mean of

each variable of interest (for example RNs per resident) was the same for each strategic group;

that is Differentiator = Cost Leadership = Best Cost = Focus = Lack of Strategy. To determine

whether the entire clusters are significant from each other, the study also used the MANOVA test

with Wilks' Lambda. Similar to ANOVA, this statistic evaluates the hypothesis that the







population means are equal across groups; however, this method uses a multivariate approach. It

was used in Zinn et al. 1994 study that examined strategic groups.

Hypothesis #2: Nursing Homes in the Differentiator Strategic Group will have the Highest
Quality.

Dependent Variables:

There were a total of 8 dependent variables that were quality measures; five of which

relate to long term care quality, two of which measure post acute quality and one measure of the

overall quality of the facility. The long term care (LTC) and post acute (PA) measures were

quality indicator (QI) variables; they included LTC bladder decline in continence, LTC bowel

decline in continence, LTC 4 point decline in ADL function, the LTC prevalence of the pressure

sores, LTC restraint prevalence, PA walking improvement, and PA pressure sore incidence. The

bladder and bowel decline variables are measures of the proportion of residents who had become

incontinent. The ADL decline in function is a risk adjusted ratio of the proportion of the

residents who experienced a 4 point ADL decline (out of 16 points) relative to the total number

of residents in the nursing home. The restraint prevalence is the proportion of residents that are

restrained at the facility. The pressure sores prevalence is a risk adjusted ratio of the proportion

of residents who have a pressure sores relative to the total number of residents in the nursing

home. The walking improvement variable measures the change in ambulation status between the

5 day assessment and the 14 day assessment. Pressure sores incidence variable is a measure of

the amount of post acute residents who developed a pressure sore while remaining in the nursing

home relative to the total number of post acute residents in the nursing home. The quality

indicators were continuous variables calculated using the MDS based on the methodology as

described by Abt Associates 2004. These quality measures are included on the Nursing Home

Compare website and are currently used by social workers and consumers to differentiate

between good and bad facilities. The eighth quality variable was the total number of quality of

care deficiencies. This is an overall measure of the facility's quality. Quality of care deficiencies
66







is a count variable that represents the number of citations issued by state survey agencies and is

located in the OSCAR dataset. Deficiencies are categorized according to level of seriousness

from letter "A" to "L." Deficiencies categorized "F" and higher denote those with varying

degrees of pervasion, which have the potential to, or actually cause harm or immediate jeopardy

to residents (O'Neill et al. 2004). It also is on the Nursing Home Compare Website and is used

by social workers and consumers to differentiate between good and poor facilities.

Independent Variables

The independent variables of interest were dichotomous variable (0 or 1) for each

strategic group generated by the cluster analysis. The expected strategic group variables include:

differentiator strategy, cost leadership strategy, best cost strategy, focus strategy, and lack of

strategy for those facilities that did not have any strategy.

Control Variables

The control variables in the study included both organizational and market factors. They

included size, ownership, chain affiliation, market competition, occupancy, payer mix and case

mix. Nursing home size is measured using the number of beds. Larger size provides more

resources and gives an advantage when trying to recruit staff. Similar to size, chain affiliation

gives more resources to draw upon which can also improve the facility efficiency. Ownership

and chain affiliation in nursing homes has been associated with lower quality of care (Harrington

et al. 2001). The percent Medicare and percent private pay beds are outcomes of effective

strategy because both give the facility an ability to increase their revenue streams and

subsequently provide the facility with more resources to provide quality of care. Market

competition and the occupancy of the facility can influence the behavior of a facility as they

attempt to attract more profitable residents. A facility's case mix is the measure of the acuity of

the nursing home's residents. Higher acuity of residents may be more costly and more difficult to

care for and may influence quality measures like pressure ulcers and ADL decline that is beyond

67







the control of the facility. Several studies have utilized the Acuindex as the case mix variable

(Johnson et al. 2004, Laberge et al. 2005, Weech-Maldonado et al. 2006). The Acuindex is an

acuity variable located in the OSCAR dataset. It is represented by different factors including

mobility issues like how much assistance the resident requires with functional activities like

ambulation or transfers, or nursing issues like proportion of residents receiving suctioning or

proportion of residents receiving intravenous therapy. This study also included other measures of

acuity that included ADL index, tube feeds, respiratory care, suction, tracheotomy care, IVs, and

injections. These measures were included because different strategic groups may have different

types of acuity. For example, a strategic group may have a high proportion of the residents that

need IVs, respiratory care and need injections but who are independent with their ADLs.

Although the Acuindex is a good overall measure of acuity, it cannot capture the intricacies of

the different groups that could be related to the different strategies. These different nursing

acuity measures were not used in the cluster analysis because they represent the needs of the

residents rather than the technology that was provided to them. For example, knowing that half

the residents need respiratory care is not the same as a measure that represents the amount of

treatment or number of different kinds of treatment provided. Although they were not used in

determining the different groups, these variables are needed to be included as controls because

they can influence how the technology of the facility is provided.

Analysis

The study intended to use the ordinary least squares (OLS) regression to analyze seven of

the quality of care variables (AREG function on STATA) (LTC bladder decline in continence,

LTC bowel decline in continence, LTC 4 point decline in ADL function, the LTC prevalence of

the pressure sores, LTC restraint prevalence, PA walking improvement, and PA pressure sore

incidence). Prior to the analysis, it was necessary to evaluate the dependent variables in order to

ensure that they have normal distribution. If any of the variables were not normally distributed,

68







the dependent variables were transformed by using the log function. If the initial log

transformation did not result in proper skewness and kurtosis, other transformations were

attempted. Other possible transformations included the cubed root, the root, the square and the

cube. All of them would be attempted until a transformed variable was identified that had a

skewness and kurtosis closest to 0 and 3 respectively. If the skewness was not able to be resolved

and the variable seemed to have a gamma distribution, the study used the generalized linear

model with gamma distribution and a log link. After the OLS analysis, it was necessary to

determine if the regressions pass the OLS linearity assumptions. The linearity assumption was

tested through observation of the residual-predicted values plots and the Hosmer-Lemeshow test.

The quality of care deficiencies is a count variable and was analyzed using the negative

binomial equation. The study used White's heteroskedacity-consistent covariance matrix (1980) to

correct for a heteroskedascity. It used county level fixed effects. Dependent variable outliers of 5

standard deviations and above were dropped from the dataset.

Qual = V + &giGroupl + &g2Groupl + .....+&gnGroupN + &cControl + &feFE + E

Qual represents one of the seven quality dependent variables.V represents the constant

value, while &gn represents the effect of each strategic group on the quality variable. &gc

represents the effect of the control variables on the quality variable. &fe represents the effect of

the county level fixed effects. E represents the random error.

Hypothesis #3: Nursing Homes in the Focus Strategic Group will provide Higher Quality
in the area that they are focused on compared to other Strategic Groups.

Dependent Variables

Hypothesis #3 used the same quality of care dependent variables as Hypothesis #2.

Independent Variables

As in hypothesis #2, the independent variables of interest included a dichotomous

variable (0 or 1) for each strategic group generated by the cluster analysis. However, unlike







hypothesis #2 where the differentiator was used as the reference group; this hypothesis used the

focus strategic group variable as the reference group.

Control Variables

Organizational characteristics that were used as controls included size, for-profit status,

chain affiliation, payer mix, and case mix as described in hypothesis #2.

Analysis

As in hypothesis #2, the planned analysis for the QIs was the OLS with county fixed

effects (AREG function on STATA. The study used White's heteroskedacity-consistent covariance

matrix (1980) to correct for a heteroskedascity. It used county level fixed effects. Dependent

variable outliers of 5 standard deviations and above were dropped from the dataset.

Focus Qual = V + &giGroupl + &g2Groupl + .. +&gnGroupN + &cControl + &feFE + E

Qual represents one of the seven quality dependent variables.V represents the constant

value, &gn represents the effect of each strategic group on the quality variable. &gc represents the

effect of the control variables on the quality variable. &fe represents the effect of the county level

fixed effects. E represents the random error.

Hypothesis #4: Nursing homes in the Cost Leadership Strategic Group will have the lowest
costs.

Dependent Variables

The patient cost equation used operating expense per resident day. It is determined by

dividing the operating revenue by the total amount of residents and dividing it by 365 days.

Independent Variables

The independent variables of interest included a dichotomous variable (0 or 1) for each

strategic group generated by the cluster analysis. Since the focus of the cost leader was to have

the lowest cost, the cost leadership strategic group was used as the reference group.







Control Variables

Organizational characteristics that will be analyzed are size, for-profit status, chain

affiliation, payer mix, market competition and case mix as described in hypothesis #2.

Analysis

Similar to health care expenditure measures, the cost per patient per day tends to be

skewed to the right with some facilities having significantly higher cost than the mean. Similar to

prior studies dealing with health care cost, the study used the generalized linear model (GLM)

with gamma distribution and a log link (Fleishman et al. 2006, Pagano et al. 2005). The study

then determined the predicted value using the smearing method. The study used county level

fixed effects. The study used White's heteroskedacity-consistent covariance matrix (1980) to correct

for any heteroskedascity as found in STATA (2001). It used county level fixed effects. Dependent

variables outliers of 5 standard deviations and above were dropped from the dataset.

FlnCost = V + &giGroupl + &g2Groupl + .....+&gGroupN + &cControl + &feFE + s,

FlnCost represents one of the seven quality dependent variables.V represents the constant

value, &gn represents the effect of each strategic group on the quality variable. &gc represents the

effect of the control variables on the quality variable. &fe represents the effect of the county level

fixed effects. E represents the random error.

Hypothesis #5: Differentiation, Cost Leadership, Focus and Best Cost Strategic Groups will
have better Financial Performance than Nursing Homes that Lack a Strategy

Dependent Variables

There were two variables used to examine financial performance. The first of the two

financial performance variables used, the operating margin, is calculated using the operating

revenue minus the operating expense then divided by the operating revenue. Operating revenue

and expense were direct measures of doing business and therefore the operating margins may be

more sensitive to nursing home strategies. The second financial performance variable included

total profit margin. Profit margin not only includes the operating revenue and operating expense,
71







but also includes the non patient revenues like charitable donations and non patient expenses like

capital costs. Although profit margin is not as sensitive as operating margins to changes in the

nursing homes' strategies, it is still important because it does represent the overall picture of the

nursing homes financial performance. Financial performance is important because it strongly

influences the nursing homes behavior. If strategies that produces higher quality result in better

financial performance, then customers' needs and providers' goals will be aligned. However, if a

strategy provides higher quality of care but does not provide the facility with better financial

performance, alternative strategies by the nursing homes will likely be considered.

Independent Variables

The independent variables of interest included a dichotomous variable (0 or 1) for each

strategic group generated by the cluster analysis. The lack of strategy group was used as the

reference group since it was expected that nursing homes that had a strategy would have better

financial performance than those that did not have a strategy.

Control Variables

Organizational characteristics that were utilized as controls included size, for-profit status,

chain affiliation, Medicaid and case mix as described in Hypothesis #3. Unlike the other

hypothesis, hypothesis #5 only used Medicaid as a control variable. Some nursing homes are

trying to acquire the higher reimbursing Medicare as part of their strategy. By using the percent

Medicare variable as a control variable, it would not be possible to determine the benefit of

having a strategy directed towards acquiring rehab patients.

Analysis

Although the operating margin and the total margin were both continuous variables,

neither of them is normally distributed and therefore Ordinary Least Squares regression is not

appropriate. By the nature of the mathematical equation that is used to determine them, their

range is from negative infinity to positive one. This is because both equations use the revenue of







the facility in the numerator and the denominator and both equations subtract the costs of the

facility from the revenue in the numerator. Therefore the highest value that the operating and

total profit margins can be is one ((revenue minus no costs) / revenue = 1). Since the variable is

not normally distributed, and cannot be transformed in a parsimonious way, a different analytical

tool is necessary.

In addition, the focus of the study is to determine if the total and the operating margins

vary by strategic groups. As a result, determining the odds of being a higher financial performer

is appropriate. Therefore, the study used an ordered logit with odds ratio analysis to determine

the odds that the strategic group will be in the highest group. Weech-Maldonado et al. 2008 used

the ordered logit model with odds ratio in their study that examined quality and financial

performance. The study determined the predicted probabilities of a strategic group being in each

financial tier for each strategic group. By providing the predicted probabilities, it was possible to

identify which strategies have the best financial performance for operating margin and total

margin. The study divided the operating margin and the total margin variables into three

financial performance tier: low (bottom 10%), medium (middle 80%), and high (upper 10%).

Similarly, since different counties can have significantly different market forces between

them, it was necessary to control for these county differences by using county fixed effects.

Predicted values using logit were used to identify predicted probabilities by tier for each

strategic group.

Ologit (B) = V + &giGroupl + &g2Groupl + .....+&gGroupN + &cControl + &feFE + E

FlnCost represents the cost dependent variable.V represents the constant value, &gn

represents the effect of each strategic group on the quality variable. &gc represents the effect of

the control variables on the quality variable. &fe represents the effect of the county level fixed

effects. E represents the random error







CHAPTER 4
RESULTS

Hypothesis #1

Identification of Different Factors: The study completed three separate factor analyses to

reduce the total number of variables. The newly generated composite variables represented valid

constructs of nursing home's technology and resource commitments. Initially there were 57

technology variables and 11 resource variables that were used in the factor analysis.

Factor analysis #1 (Technology): The first factor analysis included only the variables that

measured the technology of the nursing homes. Of the 57 reported variables, there were 20 factors

that were derived reporting an eigenvalue greater than 1 (table A-i in the appendix), with ten of

those factors having eigenvalues between 1.0 and 1.54. Although the factors just above one would

qualify using the Kaiser (1960) criteria, the amount of variation removed by each subsequent

factor after 10 was marginally inadequate. Therefore it was necessary to use the scree plot to

identify the ideal number of clusters. The scree plot indicated a point of inflection between 9 and

11 factors (Figure A-i in the appendix). Although the eigenvalues could justify the use of 20

factors, the scree plot indicates that the change in the amount of variance contributed by using

additional factors beyond 10 appeared to be non- significant. Therefore, the factor analysis for

technology only used the 10 factors.

All variables with a factor loading above 0.4 were initially included in the each of the 10

factor variables (Table A-2 Appendix A). There were six variables that were included in more

than one factor. Since each of these six variables can only be in one factor, it was necessary to

determine the most appropriate factor for each variable. The six variables included long term care

pressure relieving chairs, long term care pressure relieving beds, long term care and post acute

restorative PROM and AROM. The long term care pressure relieving chairs and beds had loadings

that were similar in both factors (0.6). One of the factors also included post acute pressure

relieving devices. Therefore this factor generated a clear construct of pressure relieving devices.
74







The ROM variables were included in the restorative long term care and restorative post acute

measures. The ROM variables were also included in a factor that included splinting; however this

factor had a Chronbach's alpha score that was below the 0.7 level and therefore was dropped. The

end result was 8 technology factors with variables that had factor loadings over 0.4 (Table A-3

Appendix A).

Factor Analysis #2 (Resource): The staffing intensity factor analysis identified three

factors with eigenvalues greater than one (Table A-4 Appendix A). The scree plot appeared to

demonstrate a point of inflection between 2 and 3. The three factors included licensed therapist

(PT and OT), therapy assistants (PTA and COTA) and nursing staffing (including RN, LPN, and

CNA) (Appendix A table 4).

Factor Analysis #3 (Resource): For the staffing mix, the eigenvalues were all below 1.

Since each of the factors removed less variance than one variable, it was more appropriate to

include each of the staffing mix variables individually rather than as a composite.

The factor analysis identified 10 technology factors and 3 resource factors. These factors

were tested for internal consistency using the Cronbach's alpha method. As illustrated in Table 4-

1, 8 technology factors and 2 resource factors had Cronbach's alphas over the 0.7 threshold. These

factors included long term care skin care, post acute care skin care, long term care restorative

process, post acute restorative processes, relief devices, therapy minutes, toileting program,

weight management program, therapist, therapy assistants. The restorative splinting technology

variable was included in the factor that was associated with restorative AROM and PROM yet

failed to have Cronbach's alpha score greater than 0.7. The technology variables therapy diet and

plate guard made up the final factor. This factor was dropped because they failed to have an alpha

score greater than 0.7.

Nursing (RN, LPN and CNA) on the other hand did have an alpha score of 0.69; however it

was decided not to use a total staffing intensity factor because a nursing home could, as a strategy,

75








choose to substitute a licensed staff member with a non licensed staff in an attempt to keep costs

low. By generating an index that included all three nursing variables, the ability of the cluster

analysis to identify nursing homes with different types of staffing intensity strategies would be

compromised. It was therefore deemed necessary to include the RN, LPN and CNA variable in the

cluster analysis as individual variables.

Table 4-1. Internal consistency of the composite variables
Cluster variable Alpha Variables used
Long term care skin Long term care ointment, dressing application process, surgical wound care, ulcer
processes of care 0.82 care, skin nutrition and turning repositioning program
Post acute skin processes
of care 0.72 Post acute ulcer care, dressing application process,
Long term care restorative Long term care restorative dressing, restorative transfers, restorative bed mobility,
processes 0.86 restorative PROM, Restorative AROM, restorative ambulation and restorative eating

Post acute restorative Post acute restorative dressing, restorative transfers, restorative bed mobility,
processes 0.91 restorative PROM, Restorative AROM, restorative ambulation and restorative eating
Long term care pressure relieving bed and chair, Post acute pressure relieving bed
Relief Devices 0.81 and chair
Therapy minutes 0.93 Long term care and Post acute physical and occupational therapy
Toileting program 0.88 Long term care and Post acute toileting program
Weight management prg 0.88 Long term care and post acute weight management program
Therapist 0.91 Total physical and occupational therapist per resident
Therapy assistants 0.77 Total physical and occupational therapy assistants per resident

Table 4-2 lists the nursing variables and all the individual variables utilized in the cluster analysis.

Other variables included the percent of nurses that were RNs, percent of nurses that were

Table 4-2. Individual variables used in cluster analysis that were not included in a factor variable.
Individual Variables Definition

RN Registered Nurse Per Resident
LPN Licensed Practical Nurse Per Resident
CNA Certified Nurse Assistant Per Resident
RN Percent Percent of Nurses are Registered nurse
License Nurse Percent Percent of Nurses are Licensed
PT Percent Percent of Physical Therapist are licensed
OT Percent Percent of Occupational Therapist are licensed
Wages Facility direct care average wage
Percent Medicare Percent residents who are post acute
licensed, percent of physical therapist, and the percent occupational therapist. As described in the

methodology, wages and percent Medicare beds were also used as individual variables.







Cluster Analysis: The 10 factor variables and the 9 individual variables were standardized

using z scores that were based on the county mean. The dataset were then trimmed for outliers

using the 5 standard deviations for each variable. This resulted in the reduction of the number of

nursing homes used in the cluster analysis from 11601 to 9726. The newly generated variables

were used in the Ward's Minimum Variance Cluster Analysis and the k-means cluster analysis.

Ward's Minimum Variance Cluster Analysis identified 5 clusters as the best number of clusters.

This was based on a local peak in the pseudo F statistic, a small value of the pseudo t2 statistic

followed by a larger pseudo t2 statistic for the next cluster solution level, and that adding an

additional cluster increased the overall fit by less than 5 percent (Table A-6 and A-7 Appendix A).

No other cluster arrangement met the three criteria. The data was then analyzed using the k-means

with an a priori of 5 clusters.

Clusters: Table 4-3 represents the results of the cluster analysis. Using the ranking protocol

described in the methodology (Table A-8 Appendix), the different clusters were defined using the

rank score ranging from 0 to 5. 0 represented the lowest and 5 represented the highest value for

both the resource variables and the technology variables. For each cluster, the cumulative rank

score was calculated by summing the values for the scope variables and summing the resource

variables. These cumulative rank scores were plotted on a graph to generate model as indicated in

Figure 4-1. The scope score was represented by the horizontal axis and the resource score was

represented by the vertical axis. Table 4-4 provided some descriptive structural, market and case

mix variables of each strategic group.

The ANOVA test identified that the clusters were significantly different from each other for

each variable. In addition, the Tukey's test compared each pair of clusters for every variable and

found that there were significant differences between each pair 79% of the time. Finally, the

MANOVA test criteria (Wilks' Lambda) found that the entire clusters were significantly different

from each other (p <.000). Descriptives of all variables is located in table A-9 in the appendix.

77









Table 4-3. Rank scores scope and resource variables
Variables Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5
Differentiator Cost Rehab Specialty Lack of
TLqA d F FQSt-t


ea er ocus care ocu y


Average Wages 1.985
RN per Resident 1.70*
LPN per Resident 0.993,4
CNA per Resident 4.362,3,4
Therapist per Resident 1.00*
Therapy assistant per resident 1.19*
Percent Licensed PT 2.542,3,5
Percent Licensed OT 2.612'3'5
Percent RN 1.15*
Percent Licensed Nurse 0.41*

Total Resource 17.91*

LTC Skin Process 3.23*
Acute Skin Process 4.682,4,5
LTC Restorative Process 5.00*
Acute Restorative Process 5.00*
ReliefDevices 4.112A45
Toilet Program 4.96*
Weight Program 5.003,5
Therapy Minutes 0.00*
Percent Medicare 0.103,4

Total Scope 32.1*
*Significantly different from all clusters,
Superscript is significantly different than identified groups


0.813,4
0.00*
1.073,4
3.59*
0.00*
2.48*
0.00*
0.00*
0.00*
0.071,3,4

8.06*

0.60*
1.56*
0.101,3,4
0.141,3,4
1.672 4A5
1.49*
3.733,5
1.601 3,5
0.153,4

11.03*


4.042,5
2.72*
5.00*
5.00*
5.00*
5.00*
1.58*
1.94*
1.42*
1.76*

33.49*

5.00*
5.002,4,5
0.89*
0.00*
5.001,4,5
0.111,2
0.00,2,4
5.00*
5.00*

26.10*


5.001,2,5
5.00*
0.00*
0.00*
1.75*
1.62*
2.442,3,5
2.602,3,5
5.00
5.00*

28.32*

2.07*
3.03*
0.34*
0.33*
1.5124,5
0.091,2
2.533
1.681,3,5
0.74*

12.32


0.001,2
0.40*
0.863,4
4.052,3,4
2.23*
0.00*
5.00*
5.00*
0.25*
0.001,3,4

17.78*

0.00*
0.00*
0.00,3,4
0.141,3,4
0.00*
0.001 2
1.031,2
1.13*
0.003,4

2.29


The descriptives values were not adjusted for the county level fixed effects and therefore were not

used in the determination of the clusters.

The plotted clusters in Figure 4-2 were then categorized using Porter's generic strategies.

Based on the definition described in the methodology, Cluster #1 was initially classified as

Differentiator because it's scope score was the highest (technology cumulative rank score).

However this group was ranked the lowest in all the categories that was related to rehab (i.e.

physical therapy minutes and percent Medicare beds). Therefore this cluster may be more

appropriately classified as a nursing care focus group rather than a differentiator. Although these

nursing homes provided a high degree of technology, they did not use the highest amount of

resources. They tended to have median level wages, median level total RN staffing, and below

median level of percent licensed therapist. They had low staffing mix as a result of their high


78


=13.8"**
:744***
:121***
=367***
:1110***
:1492***
=1180***
:906***
=1538***
=833***

=186***

:186***
:58.35***
=688***
=1987***
=39.8***
:19.7***
=8.62***
=386***
=918***

=59***








CNA staffing levels. This group also tended to have high acuity primarily driven by residents who

were more dependent and patients who had tube feeds.

Table 4-4. Rank scores of structure and market factors
Variables
Cluster 4 Cluster 5
Cluster 1 Cluster 2 Cluster 3 Specialty Lack of
Differentiator Cost Leader Rehab Focus care Strategy
N=990 N=2752 N=1816 N=1634 N=2534

Percent for profit 71.2% 75.7% 79.7% 76.9% 76.2% Chi2= 27
Percent chain affiliated 51.5% 60.0% 66.4% 63.0% 58.2% Chi2= 70
Market competition 1.43 0.00 0.71 5.00 4.29 F=0.17
Occupancy 5.00 3,4 4.53 3,4 0.00 1,2,5 1.98 1,2 3.163 F=12.15
Size 0.67 2 5.00* 1.15 2,5 1.86 2,5 0.002,3,4 F=59.23
Case Mix 4.57 2,4,5 1.06 1,3 5.00 2,4,5 2.18 1,3,5 0.00 1,3,4 F=25.95
DL index 4.73 2,4,5 1.14 1,3 5.00 2,4,5 2.08 1,3,5 0.00 1,3,4 F=21.45
Percent tube feed 4.22 2,5 1.19 1,3,4 4.51 2,5 5.00 2,5 0.00 1,3,4 F=9.65
Percent tracheotomy 3.485 1.49 4 3.15 5 5.00 2,5 0.00 1,3,4 F=14.63
Percent requiring suctioning 2.96 2,5 0.711,3,4 3.392,5 5.00 2,5 0.00 1,3,4 F=15.63
Percent requiring respiratory care 1.28 3 0.50 3,4 5.00* 2.67 2,3,5 0.00 3,4 F=42.13
Percent requiring IV 0.64 3,4 0.74 3,4 5.00* 1.87* 0.00 3,4 F=54.7
Percent requiring injection 0.983 0.78 3 5.00* 2.193,5 0.00 3,4 F=33.18
*Significantly different from all clusters,
Superscript is significantly different than identified groups

Cluster #2 was classified as the cost leader because it used the least resources. The only nursing

resource that the cost leader was not the lowest on was the CNA staffing. The lowest RN staffing

coupled with the median CNA staffing resulted in the lowest staffing mix. Cost leadership groups

also had the lowest amount of therapist with an above median level of therapy assistant. This

resulted in the low therapy staffing mix. Cost leaders were significantly larger that other nursing

homes and tended to have lower acuity. These were all consistent with what would be expected

with a cost leader strategic group.

Cluster #3 was classified as the rehabfocus cluster. It had higher wages and the highest

staffing intensity. It had an above median level of RN staffing and above median percent licensed

staff. Rehab focus nursing homes provided the highest skin services and therapy. They had the

highest use of resources and provided a significant amount of service second only to the

differentiators. This cluster cannot be classified as best cost because although the nursing homes







provided more service than the cost leadership group, they also used more resources that the

differentiators. The rehab focus group had the highest acuity in most categories.

Cluster #4 was classified as specialty care focus group. It had the highest overall wages,

overall RN staffing and RN proportion when compared to the other clusters. This cluster had skin

care services and weight program, and overall case mix acuity in the middle range. This cluster

did not use the least resources, or provide the most processes and therefore could not be classified

as either cost leader or differentiator. It could not be considered a best cost strategic group because

although this cluster provided a greater amount of care than the cost leader cluster, it used more

resources than the differentiating cluster. Although this group did not have the highest overall

acuity level, it did have the highest amount of residents with specialty care needs like tube feeds

and tracheotomies. However, these specialty care services have little effect on the overall case mix

score (acuindex) because these types of services account for only 20% of the acuindex variable

(while the ADL factors account for 80% of the variable).

Cluster #5 was classified as the lack of strategy cluster. This cluster of nursing homes

appeared to lack any specific strategy. They had patients with the lowest acuity, and they provide

by far less services than other groups while their staffing levels still remained at a median level.

All the clusters were plotted on the strategic group model (Figure 4-1) using the cumulative

technology and cumulative resource scores from table 4-3. Hypothesis #1 was partially supported.

The results indicated that there exist four distinct strategies in the nursing home industry, however

these groups were different then what was originally hypothesized. Based on the Figure #3, there

was no best cost group; instead there was a fifth group that was classified as a specialty care focus

group. Figure #3 depicts the existence of a differentiator group based on the methodology,

however closer examination of the individual processes of care indicates a nursing care focus

strategy.








Nu rsing ho ie strategic groups


Figure 4-1. Strategic group model based on scope and resource commitment in the nursing home
industry

Normal distribution: All of the eight quality dependent variables had excessive skewness

and non-normal kurtosis levels and therefore had to be transformed. After being transformed using

the either the root or log function, the skewness and the kurtosis of the quality valuables were long

term care (LTC) bladder decline (-0.03, 3.62), LTC bowel decline (0.15, 3.19), LTC 4 pt ADL

decline (0.13, 3.77), LTC restraints (0.18, 2.52), LTC pressure ulcer prevalence (-0.09, 3.81), Post

Acute (PA) walking improvement (0.315, 4.15) and PA pressure ulcer incidence (-0.78, 4.32). The

PA measures both failed the normal distribution assumption even after transformation. Both post

acute variables were therefore analyzed using the generalized linear model with gamma

distribution.

Linearity: By observing the predicted value and residual plot, all six quality variables that

were used in the OLS regression model appeared to pass the linearity assumption. The study also

confirmed these results with a negative Hosmer-Lemeshow test (non-significant) for all the

dependent variables.


40
35

~25
820

S15
0
0


10 15 20
Scope (Technology)








All the above quality variables with skewness under 4 were analyzed using OLS with

county level fixed effects. The quality variables that had skewness over 4 were initially analyzed

by using the generalized linear model (glm) with gamma distribution and a log link.

Unfortunately, the glm was not able to find a solution when using post acute pressure ulcer

incidence quality variable. Therefore the PA pressure ulcer incidence was analyzed using OLS

with fixed effects.

Hypothesis #2

Hypothesis #2 was partially supported. Table 4-5 (full regression results are located in tables

A-9 to A-16 in the appendix) illustrates that the differentiator strategic group had either the same

or better long term care quality than all other strategic groups in all measures except long term

care pressure ulcers. Differentiators had lower 4 point ADL decline than cost leaders and those

nursing homes that lacked a strategy. Differentiators also had fewer pressures sores than nursing

homes that had a rehab focus or had a specialty care focus. However, with respect to the post

acute measures, differentiators performed poorly. The differentiator strategic group generally had

worse outcomes with less walking improvement than all other groups. Differentiators had higher

incidence of post acute pressure sore than lack of strategy and cost leaders.

Table 4-5. Strategic group quality of care with Differentiator / nursing care focus cluster as the
reference group
Dependent Variables Cost Rehab focus Specialty care Lack of Strategy

Quality of care deficiencies 1.06 (0.09)* 0.99 (0.86) 1.00 (0.97) 1.12 (0.00)*
Long Term care measures

Bladder Decline 0.006 (0.04)* 0.012 (0.00)* 0.009 (0.01)* 0.005 (0.12)
Bowel Decline 0.008 (0.02)* 0.009 (0.03)* 0.012 (0.00)* 0.005 (0.16)
ADL 4 pt decline 0.008 (0.00)* 0.003 (0.39) 0.005 (0.10)* 0.004 (0.13)
Restraint use 0.011 (0.01) 0.004 (0.46) 0.001 (0.19) 0.001 (0.19)
Pressure Ulcers -0.001 (0.66) 0.012 (0.00)* 0.004 (0.14) -0.004 (0.09)
Post acute measures
Ambulation improvement 0.031 (0.10)y 0.050 (0.02)y 0.037 (0.08)y 0.060 (O.OO)y
Pressure Ulcers incidence -0.045 (0.02)y -0.015 (0.46) -0.027(0.20) -0.076 (O.OO)y
Quality of care results are incidence rate ratio
Square root transformation LTC bladder decline, bowel decline pressure ulcer prevalence, LTC 4 point ADL decline, restraints, Post acute pressure
sore incidence are OLS regression coefficients. Post acute walking improvement was glm with gamma distribution. Represents significant values
in the desired direction. y Represents significant values in the negative direction.








Hypothesis #3

This hypothesis was partially supported. As described above, nursing homes that focus on

providing nursing care had good quality. However, the other to focus groups did not have similar

results. The rehab focus group had the highest rehab staff, provided a highest amount of therapy

minutes, provided the highest amount of skin care and had the highest proportion of post acute

beds. This group was expected to have the best post acute outcomes. The second focus groups had

a high proportion of their nurses that were RNs and had the highest proportion of patients with

specialty care needs like tracheotomies, tube feeds and patients who required suctioning. This

group was expected to have better overall care than those nursing homes that lacked strategy and

better quality outcomes than the cost leaders.

Table 4-6 (full regression results are available in table A-18 to A-25 in Appendix A)

illustrates that nursing homes in the rehab focus group had lower ADL 4 pt decline and lower

restraint use than the cost leaders. In addition, the rehab focus had lower quality of care

deficiencies than the cost leaders and the lack of strategy group. This group only had better PA

outcomes than differentiators. This group generally had poorer LTC quality of care

Table 4-6. Strategic group quality of care with rehab focus cluster as the reference group
Differentiator /
Dependent Variables Cost Nursing care focus Specialty care Lack of Strategy
Quality of care deficiencies 1.07 (0.03)* 1.01 (0.86) 1.01 (0.80) 1.13 (0.00)*
Long Term care measures
Bladder Decline -0.006 (0.05)y -0.012 (O.OO)y -0.003 (0.39) -0.007 (0.02)y
Bowel Decline -0.001 (0.77) -0.009 (0.03)y 0.002 (0.52) -0.004 (0.21)
ADL 4 pt decline 0.006 (0.03)* -0.003 (0.39) 0.002 (0.39) 0.002 (0.53)
Restraint use 0.011 (0.01)* 0.004 (0.46) 0.006 (0.19) 0.006 (0.22)
Pressure Ulcers -0.013 (O.OO)y -0.012 (O.OO)y -0.008 (O.OO)y -0.016 (O.OO)y

Post acute measures
Ambulation improvement -0.019 (0.28) -0.050 (0.02)* -0.014 (0.47) 0.014 (0.57)
Pressure Ulcers incidence -0.030 (0.06)y 0.015 (0.46) -0.012 (0.49) -0.061 (O.OO)y
Quality of care results are incidence rate ratio
Square root transformation LTC bladder decline, bowel decline pressure ulcer prevalence, LTC 4 point ADL decline, restraints, Post acute pressure
sore incidence are OLS regression coefficients. Post acute walking improvement was glm with gamma distribution. Represents significant
values in the desired direction. y represents significant values in the negative direction.








especially in comparison to the differentiator and the lack of strategy group. This group had higher

LTC prevalence and higher PA incidence of pressure ulcers than cost leader and lack of strategy

group.

Table 4-7 (full regression results are available in table A-26 to A-33 in Appendix A)

illustrates that nursing homes in the specialty care focus group had lower quality of care

deficiencies than the cost leaders and the lack of strategy group. They had less pressure ulcer

prevalence than facilities in the rehab focus group. The specialty care focus group had higher

prevalence of pressure ulcers than cost leader and lack of strategy. It had higher bladder decline,

bowel decline, ADL 4 point decline than differentiators. It had lower restraint use than cost

leaders. With respect to post acute incidence of pressure sores, the results were mixed with

specialty care focus having a higher incidence of pressure ulcers than lack of strategy group. And

a higher prevalence of long term care pressure sores that cost leaders and lack of strategy.

Table 4-7. Strategic group quality of care with Specialty care focus cluster as the reference group
Differentiator /
Dependent Variables Cost Nursing care focus Specialty care Lack of Strategy
Quality of care deficiencies 1.06 (0.05)* 0.99 (0.80) 1.00 (0.97) 1.12 (0.00)*
Long term care measures
Bladder Decline -0.003 (0.26) 0.003 (0.39) -0.009 (0.01)y -0.004 (0.12)
Bowel Decline -0.003 (0.27) -0.002 (0.52) -0.012 (O.OO)y -0.006 (0.03)y
ADL 4 pt decline 0.003 (0.16) -0.002 (0.39) -0.005 (0.10)y -0.001 (0.78)
Restraint use 0.005 (0.22) -0.006 (0.19) -0.002 (0.68) -0.001 (0.88)
Pressure Ulcers -0.005 (0.01)y 0.008 (0.00)* -0.004 (0.14) -0.008 (O.OO)y
Post acute measures
Ambulation improvement -0.005(0.74) 0.014 (0.47) -0.037 (0.08)* 0.024 (0.15)
Pressure Ulcers incidence -0.018 (0.25) 0.012 (0.49) 0.027 (0.20) -0.049 (O.OO)y
*Quality of care results are incidence rate ratio
Square root transformation LTC bladder decline, bowel decline pressure ulcer prevalence, LTC 4 point ADL decline, restraints,
Post acute pressure sore incidence are OLS regression coefficients. Post acute walking improvement was glm with gamma
distribution. Represents significant values in the desired direction. y Represents significant values in the negative
direction.

Hypothesis #4

Hypothesis #4 was not supported. Table 4-8 illustrates that cost leaders nursing homes had

significantly lower costs that specialty care and rehab focus groups, but no significant difference

between cost leaders and differentiators or with those facilities that lack a strategy altogether.







Table 4-9 provides the cost predicted values for each strategic group. The predicted values

are the patient costs per resident day. Specialty care focus and Rehab focus groups had

significantly higher costs than the cost leader by 9 and 13 dollars per resident day respectively.

Table 4-8. Strategic groups and costs with cost leader as reference group regression results
Lack of Differentiation / Specialty Care
Strategy Nursing Care Focus Rehab Focus

Costs per resident day -0.006 (0.60) 0.002 (0.90) 0.039 (0.00)* 0.080 (0.00)*
Represents significant values compared to cost leadership cluster

Table 4-9. Strategic groups and predicted costs
Differentiator Cost Rehab Specialty Lack
/ Nursing Leader Focus Care of Strategy
Care Focus Focus
Costs per resident day 149.15 149.06 161.65* 155.03* 148.12

In dollars

Hypothesis #5

The initial analysis that included all the case mix variables was unable to converge. It was

therefore necessary for the study to use acuindex variable rather than the 7 individual case mix

variables. By dropping the 7 case-mix variables, the ologit analyses were able to converge (A-34

and A-35 in Appendix A).

The ologit for both financial performance measures (table A-34 and A-35 Appendix A) had

variables that were completely determined. 53 of the observations for operating margin and 57

observations for total margin observations were completely determined. There was possibility this

was the result of multicollinearity between some of the county level fixed effects. There was over

1500 county level fixed effects variables with many of them having a very low number of nursing

homes within the county. A correlation matrix was generated to determine if the completely

determined variables were either the independent variables of interest or one of the control

variables other than the county fixed effects variables. The correlation matrix (table A-36

appendix A) illustrated that none of the independent or control variables were completely

determined. As a result of the high number of county level fixed effects, the determined variables

85







made up approximately 5% of all the county fixed effects and therefore likely have minimal

impact the overall model.

Hypothesis #5 was partially supported. Table 4-10 highlights that differentiator had a 20%

greater odds of being in the highest tier for operating margin and a 31% greater odds of being in

the highest tier for total margin relative to those nursing homes that lacked a strategy. Cost leaders

had 25% greater odds of being in the highest tier for operating margin and 22% greater odds of

being in the highest tier for total margin relative to those nursing homes that lacked a strategy.

Rehab focus had 46% greater odds of being in the highest tier for operating margin and 27%

greater odds of being in the highest tier for total margin relative to those nursing homes that

lacked a strategy. Differentiators, cost leaders and rehab focus had better total margins than the

specialty care focus strategy but only the rehab focus group and the cost leadership group had

better operating margins than the specialty care focus group. The specialty care focus group did

not have significantly higher odds of being in the highest tier with respect to the nursing homes

that lacked strategy.

Table 4-10. Odds Ratio of Nursing Home Strategy
Differentiation Specialty F test
Dependent Cost care Rehab
Variable Leader Focus Focus
Operating Margin 1.25 (0.01)* 1.20 (0.09)* 1.04(0.66) 1.46(0.00)* F=9.7 (0.05)
Total Margin 1.22 (0.01)* 1.31 (0.01)* 0.98 (0.80) 1.27 (0.01)* F=10.0 (0.03)
* Significantly different from lack of strategy

Table 4-11 and 4-12 illustrate the predicted probabilities of the operating and total margins.

Both demonstrate that the nursing homes with differentiation, cost leadership and rehab focus

strategy perform better than those nursing homes that lack strategy.

Table 4-11. Probability of nursing homes being in the highest operating margin tier
Specialty
Dependent Differentiator5 Cost Rehab care Lack
Variable Leader4'5 Focus4'5 Focus2'3 of Strategy1'2'3
Low 0.099 0.096 0.085 0.110 0.111
Mid 0.799 0.799 0.797 0.798 0.797
High 0.102 0.105 0.118 0.092 0.090
#Significant p>0.10
86











Table 4-12. Probability of nursing homes being in the highest total margin tier
Specialty


Dependent
Variable
Low
Mid
High
# Significant p>0.05


0.089
0.799
0.112


Cost
Leader4,5
0.094
0.799
0.106


Rehab
Focus4,5
0.094
0.799
0.106


care
Focus1,2,3
0.112
0.798
0.090


Lack
of Strategy1,2,3
0.109
0.799
0.092


Differentiator '







CHAPTER 5
DISCUSSION

This study concurred with prior research because it demonstrated the existence of strategic

groups in the nursing home industry. However, this study used a different conceptual framework

to identify them. While earlier studies by Zinn et al. 1994 and Marlin et al. 2002 used resources

and payer mix, this study used the business level decisions of resource allocation and technology

provision. One advantage of using technology over payer mix variables to identify strategic

groups is the applicability of the results for managers. Management has control over how much

technology to provide and how much resource to use. How a nursing home decides to balance

these two factors is an important part of that facility's overall strategy. The results from this study

demonstrate that different strategic groups provide their technology and utilize their resources in a

different way.

Using technology and resource allocation to identify the clusters makes it possible to define

the groups using Porter's generic strategies. There are no nursing home studies that have

attempted to define strategic groups by Porter's generic strategies. However, Marlin et al. 2002

did find the existence of Porter's strategies in the hospital industry. Unlike this study, Marlin et al

2002 explicitly used variables that defined the differentiator strategic group. These variables

included the total number of hospital units, the number high technology units and ratio of

intensive care beds to total beds. They did not use any measures that reflect how the care was

provided. Although a hospital could have a high number of different specialty units, it did not

necessarily provide a high degree of technology within those units. Each specialty unit could have

been poorly staffed and the patients could have received a very low level of services. Marlin et al.

2002 study was examining the governance level of decision making as opposed to the business

level decisions. Business level operations are significantly impacted by the day to day decisions of

how to provide service and how to use resources.







The study identified 5 of the strategic groups listed in the hypothesis (nursing care focus

(differentiator), cost leader, rehab focus, specialty care focus and lack of strategy). The study did

not identify the best cost group and did not identify a well defined differentiator group. Instead,

the study identified three focus groups; rehab focus, specialty care focus and a nursing care group.

The rehab group from this study appeared similar to the high Medicare facilities in the prior

nursing home strategic group studies. This group not only had a high proportion of Medicare beds,

but also provided a high level of therapy minutes and skin care. The specialty care focus group

had a high RN intensity and had a high concentration of residents with specific care needs.

Residents with specialty care needs like tracheotomy care or suctioning require the nursing home

that has a capacity to deliver a high level of skilled services. These conditions are also only

minimally covered by Medicare in the post acute setting and have long been considered

underserved by MedPAC (MedPAC 2003, 2007). Beginning in 2004, nursing homes could get

more reimbursement for residents with these extended needs. This change in reimbursement only

started in 2004 and it may be possible that the facilities had not been able to fully take advantage

of it.

One unexpected finding was that the defined differentiator strategic group provided the least

therapy services. The fact that this group is defined as a differentiator when it provides the lowest

amount of therapy brings up the question of the appropriateness of using differentiator as this

group's definition. The group clearly provided the most nursing care technology and therefore was

reclassified as the nursing care focus strategic group. Whether this group is considered a

differentiator or a nursing care focus strategic group, this group still represents nursing homes that

are trying to separate themselves from their competition by providing a high level of technology

and therefore they can still be defined by using Porter's generic strategies.

Out of all the clusters identified, the cost leader proved to be the best defined. The cost

leader group had the lowest staffing levels and lower staffing mixes. Lower staffing mixes would

89







indicate a cost leader because it indicates that the nursing home is substituting CNAs for RNs. In

addition, on observation of the structural features of this group, the cost leaders also tended to be

larger and have lower acuity. It is clear that this group's strategy was to keep cost under control by

providing a low amount of service.

Finally, the lack of strategy group was identified because it used a median amount of

resources with almost no technology. In addition, the nursing homes that lacked strategy appeared

not to have any specialty services.

By using technology and resource deployment, this study found that strategic groups not

only exist, but can be classified using Porter's generic strategies. The next question is whether

strategic-performance relationships exist between the different strategic groups and quality, costs

and financial performance.

Quality

This study identified that the nursing care focus (differentiator), specialty care focus and

rehab focus strategic groups had lower quality of care deficiencies than the nursing homes that

were in the cost leader strategic group or the nursing homes that lacked strategy. This finding

provides some evidence that focusing on providing quality either by providing technology or

having better staffing mixes can result in some better outcomes. However, only quality of care

deficiencies had consistent results through these three strategic groups. Quality of care

deficiencies is a general measure of the facilities' quality and can relate to anything from staff that

is inadequately trained to poor care practices. The Nursing Home Compare website lists and

describes all of facilities deficiencies from the most recent state survey. It is a good general

measure of quality and has been used in past research (Harrington et al. 2000, 2001). One

limitation of deficiencies is that it does not highlight specific care outcomes. Other measures like

the QIs that were derived by Abt are specifically designed to identify the specific quality issues

like pressure ulcer prevalence and ADL decline. Knowing the specific quality areas would







hopefully help managers address the problems directly. In this study the results of these measures

were mixed.

When compared to other groups, the nursing care focus strategic group generally provided

better care to their long term care residents. The nursing care focus, however, also had poor post

acute outcomes. The fact that this group provided the least amount of minutes of therapy, had a

lower proportion of post acute patients, and had the poorest post acute quality provides further

support to classify this cluster as a nursing technology focus group rather than an overall

differentiator group. These results suggest that there may be a relationship between a strategic

group focus and the quality of care provided.

The other two focus strategic groups did not fair as well. The rehab focus strategic group

had a high level of rehab staffing and provided high amount of rehab technology. The specialty

care focus had a high level of RN staffing and provided care to residents with specific care needs.

It would be expected that the rehab focus group would have the better post acute care

outcomes than other groups. However, rehab focus strategic group had a higher incidence of

pressure sores and did not have better walking improvement than the specialty nurse, cost leaders

or nursing homes that lacked strategy. This was surprising since the rehab focus did not have

better outcomes even though they provided significantly more therapy. There are a number of

possible explanations.

The rehab focus strategic group had higher acuity than other strategic groups. Although the

quality measures used are risk adjusted and the case mix of the facility was controlled for, there

may be factors that may not be captured. Mukamel et al. 2008 has determined that the current

quality measures including pressure ulcers would benefit with a greater degree of risk adjustment.

The rehab focus group has a high proportion of Medicare patients and may have the expertise to

handle patients with more co morbidities that may not be captured by the risk adjustments of the

outcome variables. Similarly, different strategic groups had different types of acuity measures.

91







Some of the case mix variables may have greater influence on the quality outcomes than others.

The rehab focus groups may be 'penalized' by taking on residents that are more prone to pressure

ulcers and whose risk factors are not captured by the risk adjustments of the quality measures or

control variables used in the analysis.

In the case of the walking improvement, the rehab focus group provided nearly 25% more

minutes then any other group and yet did not attain better outcomes than the specialty care focus,

cost leaders or lack of strategy groups. There are a few reasons that may explain why the rehab

focus group does not have better walking improvement outcomes. First, rehab focus nursing

homes may be downgrading the progress of their residents. Rehab focus strategic groups are

accustomed to providing care to rehab patients and may be more rigorous in the application of

walking improvement scales. A second reason is that there may be an upper therapy minute

threshold where adding more minutes of therapy does not translate into better outcomes. The

rehab group provided significantly more minutes of therapy when compared to the specialty care

focus, the cost leader groups and the lack of strategy group yet did not have better therapy

outcomes. The fact that differentiators provide considerably less therapy and have worst outcomes

when compared to all other groups supports the notion that there is at least a minimal threshold

level of minutes of therapy that needs to be achieved in order for basic outcomes to occur.

Another possibility is that since revenue streams are enhanced by providing higher intensity of

therapy for a longer period of time, these facilities may be more conservative with their 14 day

assessments in an effort to increase the length of stay. By having a lower 14 day assessment value,

the facility could be able to justify keeping the resident at the facility for a longer period time.

Over the past 7 years, MedPAC have issued statements of concern that the current post acute

reimbursement schematic promotes therapy overutilization. More research is needed to determine

if the lack of post acute quality outcomes difference between the rehab group and the cost leader,







lack of strategy and the specialty care groups is the result of some therapy intensity upper

boundary, or if the rehab focus groups are manipulating from the current reimbursement system.

It would be expected that the specialty care focus strategic group would have better quality

of care outcomes in the areas where high RN staffing levels are needed. Generally, besides the

quality of care deficiencies and in comparison to the differentiator, the specialty care group results

were somewhat mixed. The fact that differentiators had better outcomes is not surprising since the

outcome measures are more closely related to the technology provided by the differentiators.

Similar to the differentiator and the rehab focus group, the specialty care group had a higher

prevalence of pressure ulcers with their long term care residents than the cost leader and the lack

of strategy. This was contrary to Weech-Maldonado et al. 2004 that identified RN staffing

intensity and staffing mix as related to quality outcomes like lower pressure ulcers. One

explanation for the high pressure sore prevalence could be endogeneity. Nursing homes with a

high prevalence of pressure sores may employ more RNs so they can treat those residents

effectively.

Cost

Rehab focus and specialty care focus group have significantly higher costs than the other

groups. This is not surprising since both groups have a higher amount of licensed staff. What was

unexpected is that the nursing care focus group (differentiator group) had similar costs to the cost

leader group. These results are unexpected because nursing care focus group had high acuity and

provided the most technology at the same cost as the cost leaders and those nursing homes that

lacked a strategy. Past studies have found that better quality translates to lower cost in nursing

homes (Weech-Maldonado et al. 2003). In the case of this study, there was a relationship between

the nursing care focus group and better long term care outcomes. Although the nursing care focus

group did not have the lowest staffing costs, it may compensate by saving money by providing

more effective and efficient care.







Financial Performance

The study found that rehab focus, nursing care focus group and cost leaders had better

financial performance than the nursing homes that lacked a strategy and the specialty care focus

group. The poor financial performance of the specialty care focus group could be related to higher

costs of having more RNs without the benefit of generating higher revenue streams. Unlike the

rehab group where the more therapy results in more revenue, the specialty care focus groups may

have limited ability to increase revenue. Medicare and some Medicaid programs can (depending

on the state) provide more payment for nursing care; however the reimbursement is provided

based on the utilization of services and not based on the staffing levels of the facility. In addition,

these groups provided a care to patients with highly specific and costly conditions. MedPAC 2007

has expressed concern that the current Medicare reimbursement system does not cover the costs of

care adequately to patients with needs over than therapy. Rehab focus nursing homes would be

able to generate the greatest amount of revenues since they had the most Medicare rehab patients.

This would explain why even though the rehab focus group had some of the highest costs, they

still managed to be one of the best financial performers. These results are providing some support

with prior research that found that nursing homes with higher Medicare residents and private pay

residents had the best financial performance (Zinn et al. 1994, Marlin et al. 1999). However, in

this study there were other groups that also performed well. Differentiators managed to have good

financial performance without a high degree of rehab. One explanation is that by 2004, most states

through Medicaid used case mix reimbursement that would compensate facilities that provided

certain types of nursing technology (e.g. restorative care). Nursing homes that provide services

that can generate revenue and lower costs via better quality can increase their odds of achieving

better financial performance.

The hypothesis of this study stated that nursing homes that lacked a strategy would not

perform as well as nursing homes that had a strategy. The results suggest that this is true with one

94







exception. The specialty care focus group appeared to be a poor strategy with respect to financial

performance. This outcome presents the possibility that not all strategies are created equal.

The strategic groups rehab focus, cost leader and nursing care focus group (differentiator)

all had better total margins than the other nursing homes with no significant difference between

these strategies. One aspect of strategic group theory is that it is possible that there are multiple

strategies within the same industry that are effective but not necessarily better than each other.

This study supports this aspect of strategic group theory.

Porter's (1980) original theory described three effective strategies: cost leaders,

differentiators and focus. This study provides support for these strategies even if the line between

the differentiator and the focus strategy is blurred. Porter's generic strategies can help nursing

homes identify strategies that allow them to achieve good financial performance.

Managerial Implications

Management is interested in performance. Providing managers with a way of identifying

effective strategies that can improve their facilities performance should be well received. Cool and

Schendel (1987) defined strategic groups as sets of firms competing within an industry on the

basis of similar combinations of scope and resource deployment commitments. Marlin et al. 2002

argued that for health care administrators, scope decisions refer to the selection of market

segments to compete in and the types of services to offer in these market segments. This study

differed from prior research because it examined if the business level decision of technology

commitment could be used to identify different strategies and whether those strategies would have

different performance outcomes.

For managers that find themselves in facilities that would be considered to have poor or no

strategy, this study provides some interesting findings. Strategic group theory posits that there

exist mobility barriers that restrict a facility from moving from one strategic to another. The size

of the investment needed for a nursing home to move from one strategic groups dictate the height

95







of the barrier. This study partially supports that theory. Certainly a nursing home would have to

incur significant cost to enter the rehab focus group thereby protecting this group with a relatively

high mobility barrier. However, the mobility barrier to the differentiator may not be as high since

these groups staffing levels, wages and cost are considerably low.

Since it appears that nursing homes that provide higher levels of technology can have low

costs and be profitable, managers with low performing strategy can improve their performance

providing more care with the staff that they have.

The study found that business level decisions of technology and resource utilization can be

used to identify strategies that lead to greater financial performance and lower cost. The study

identified cost leaders, rehab focus and differentiators groups as having significantly better

financial performance than the RN focus strategy and nursing homes that had no strategy. The

rehab focus group had the highest proportion of Medicare beds and other payer beds. Although

this supports prior studies that found that those facilities that had higher Medicare beds and

private pay beds perform well financially (Marlin et al. 1999 and Zinn et al. 1994), it found that

this strategy was not significantly better than the differentiation or the cost leadership. Therefore,

managers may be able to choose from one of these three alternatives and still have their facility

perform well financially.

One of the primary functions of managers is to control costs. With differentiators able to

provide a high degree of care at the same costs as the cost leaders raises the question of whether

there are cost savings in keeping staffing levels and staffing mixes low. Further research is needed

to know how the differentiators were able to provide all the services yet have similar costs.

Policy Implications

State and federal governments have a considerable interest in nursing homes since they

provide over 60% of the funding. In return for their investment, they expect quality care for the

residents. Therefore, it is in the government's best interest to know if there are some strategies that

96







are better than others with respect to ensuring quality of care. This study has found some evidence

that nursing homes with strategies directed toward providing care have better quality of care than

those nursing homes focused on cost savings. The study also found some evidence that nursing

homes that provide a high level of nursing care technology to their long term care residents have

better quality of care outcomes for that population. The study also identified a group of nursing

homes that provide care to patient with needs other than rehabilitation that may not be receiving

adequate funding. The specialty care group provides service to a population that is very

vulnerable. Future policies should be directed to provide incentives that increase the amount of the

important processes of care provided and not just address structural factors like staffing.

Another factor is that is of interest to government agencies like CMS is that facilities that

provided a significantly higher amount of therapy did not have better post acute outcomes with

respect to walking improvement. At the same time, the strategic group that provided significantly

less therapy than the other groups had worse walking improvement. From these results, it appears

that there is a minimum amount of therapy required for adequate improvement, but there is limit

beyond which more therapy does not lead to any more significant benefit. With the high costs of

post acute care which is almost solely paid for by the government, understanding how much

therapy is sufficient is be very valuable. Further research is necessary to determine the minimum

amount of therapy that is necessary.

Similar to managers, policymakers are interested to know that a strategy of providing a

high degree of service does not necessarily costs more. The fact that some facilities are able to

provide a high level of service and yet keep their cost down can guide policymakers to provide

incentives that focus on care provided as well as staffing. Further research is needed examining

the relationship between processes of care and the outcomes that they would be expected to

influence.







Conclusion

The study indicates that strategic groups exist in the nursing home industry that can be

defined by Porter's generic strategy. The study also demonstrates that strategic groups can be

determined using the business level strategies of what technology to provide and how to deploy

resources. With respect to quality, nursing homes that focused on providing technology (rehab or

nursing) or had higher RN staffing levels had less quality deficiencies than facilities that did not

have a strategy at all. Only the strategic group that provided a broad array of nursing technology

had better quality in the area that the technology was focused. The study found that nursing homes

that provide a high level of nursing technology do not necessarily have higher costs. The study

further demonstrate that Porter's cost leader, nursing care focus and rehab focus strategies have

better financial performance than those facilities that have poor strategy or do not have a strategy

at all.

Limitations

There are a number of limitations to this study. First, the results of the study apply only to

free standing, non government facilities that have Medicare cost report and are located in a county

that has at least 2 nursing homes. Facilities that only had private pay residents or facilities that

only had Medicaid residents were not included in the sample. This limits the generalizability to

the 9700 nursing homes that have all these characteristics.

Another limitation includes the possible endogeneity between pressure ulcer quality

measures and the technology variable for skin technology. The skin technology variable includes

both preventative and treatment processes. As a result, especially in the case of the long term care

skin technology variable, nursing homes with a high prevalence of pressure ulcers are forced to

provide treatments. This fact increases the likelihood of an endogenous relationship between the

quality variable and the strategic groups that provide a high level of long term care skin processes

and may explain the why the rehab focus group had the highest prevalence and incidence of







pressure sores and also provided the most skin care. Another concern about the post acute pressure

ulcer incidence variable was that it did not have a normal distribution and an adequate

transformation was not achieved. When the GLM analysis was attempted, the analysis could not

converge. Therefore the results of this measure may not be reliable.

Another limitation of the study was the magnitude of the impact of some of the dependent

variables. 4 point ADL decline and post acute improvement in walking had very small regression

coefficients relative to the constant. Although these variables presented statistical significance, the

small size of the effect may limit the relevance.

The study used cross-sectional data and therefore caution should be taken when trying to

infer a causal relationship. Finally, the datasets included in the study are administrative data.

Although the nursing homes can be held accountable for the accuracy of the data, some

inconsistencies can occur depending on the individual who is entering the data. For example, the

MDS is entered by at least one nurse in each of the 16000 nursing homes in the United States.









APPENDIX
BACKGROUND TABLES

Table A-1. Eigenvalues of scope (technology) factor analysis


Eigenvalue
7.03
4.89
3.76
2.68
2.29
2.21
1.80
1.78
1.54
1.51
1.46
1.37
1.34
1.28
1.21
1.17
1.12
1.06
1.04
1.01
0.95
0.85
0.81
0.77
0.70
0.69
0.63
0.63
0.62
0.60
0.54
0.50
0.47
0.43
0.43
0.41
0.39
0.38
0.36
0.33
0.32
0.30
0.25
0.24
0.21
0.20
0.19
0.18
0.18
0.17


Difference
2.134
1.138
1.074
0.391
0.080
0.416
0.017
0.241
0.031
0.043
0.098
0.024
0.067
0.061
0.048
0.047
0.057
0.022
0.027
0.064
0.101
0.037
0.044
0.066
0.014
0.053
0.009
0.006
0.022
0.060
0.033
0.032
0.038
0.008
0.021
0.021
0.006
0.024
0.026
0.009
0.018
0.050
0.017
0.023
0.011
0.011
0.008
0.005
0.005
0.008


Proportion
0.126
0.087
0.067
0.048
0.041
0.040
0.032
0.032
0.028
0.027
0.026
0.024
0.024
0.023
0.022
0.021
0.020
0.019
0.019
0.018
0.017
0.015
0.015
0.014
0.013
0.012
0.011
0.011
0.011
0.011
0.010
0.009
0.008
0.008
0.008
0.007
0.007
0.007
0.006
0.006
0.006
0.005
0.005
0.004
0.004
0.004
0.003
0.003
0.003
0.003


Cumulative
0.126
0.213
0.280
0.328
0.369
0.408
0.440
0.472
0.500
0.527
0.553
0.577
0.601
0.624
0.646
0.666
0.686
0.705
0.724
0.742
0.759
0.774
0.789
0.802
0.815
0.827
0.839
0.850
0.861
0.871
0.881
0.890
0.898
0.906
0.914
0.921
0.928
0.935
0.941
0.947
0.953
0.958
0.963
0.967
0.971
0.974
0.978
0.981
0.984
0.987













Scree plot


8.00
7.00
6.00
5.00
S-*- Series1
4.00 3
3.00
2.00
1.00 -
0.00
0 10 20 30 40 50 60


Figure A-1. Scree plot of scope (technology) factor analysis











Table A-2. Factor analysis with 10 factors of scope (technology) variables
Rotated Factor Pattern
Factor Factor2 Factor3 Factor4 Factor5 Factor6 Factor? Factor 8 Factor9 FactorlO
Long term care plateguard -0.01 0.13 0.07 0.02 -0.04 0.01 0.05 0.02 0.00 0.76
Long term care Weight program -0.01 -0.01 0.04 0.03 0.02 -0.03 0.03 0.02 0.92 0.03
Long term care Therapy Diet 0.01 0.17 0.07 -0.03 0.03 0.05 0.13 -0.05 -0.02 0.70
Long term care Bladder training -0.03 0.07 0.08 -0.11 0.02 0.05 0.12 0.30 -0.05 -0.23
Long term care toilet program 0.07 -0.05 0.02 0.20 -0.06 -0.02 -0.06 0.79 0.05 0.22
Long term care no skin treatment 0.00 0.09 0.00 -0.71 -0.24 -0.05 0.00 -0.11 -0.01 -0.06
Long term care other skin tx -0.03 0.63 0.14 0.24 0.05 -0.05 0.01 0.08 0.00 0.22
Long term care ointment -0.06 0.70 0.18 0.02 0.13 0.02 0.06 -0.02 -0.02 0.16
Long term care pressure relief bed -0.05 0.57 0.18 0.59 -0.19 0.04 0.02 -0.11 -0.05 0.04
Long term care pressure relief chair 0.00 0.61 0.12 0.51 -0.20 -0.05 0.06 0.00 -0.02 0.08
Long term care dressing application -0.06 0.77 0.11 -0.07 0.16 0.13 0.05 -0.18 -0.01 0.06
Long term care surgical wound care 0.06 0.40 -0.09 -0.07 0.08 0.35 0.06 -0.07 0.01 0.01
Long term care ulcer care -0.03 0.72 0.07 -0.06 0.15 0.11 0.03 -0.20 0.00 0.01
Long term care skin nutrition program 0.02 0.60 0.08 0.04 0.21 -0.09 -0.08 0.12 0.09 -0.06
Long term care turning repositioning -0.02 0.69 0.14 0.18 0.03 0.04 0.00 0.19 -0.03 -0.03
Long term care OT minutes 0.03 0.34 -0.11 -0.05 -0.11 0.75 0.05 -0.09 0.10 -0.02
S Long term care PT minutes 0.03 0.38 -0.10 -0.05 -0.10 0.74 0.04 -0.09 0.09 0.01
I Long term care other restorative care -0.07 0.15 0.47 0.01 -0.04 0.02 0.21 0.08 -0.02 -0.17
Long term care restorative communication 0.17 0.01 0.39 0.04 0.04 -0.01 -0.13 -0.07 0.01 0.02
Long term care restorative amputation -0.04 -0.03 0.38 0.05 -0.03 -0.02 0.05 0.01 0.00 -0.19
Long term care restorative eating 0.17 0.12 0.72 0.00 0.01 0.06 0.05 -0.01 0.03 0.08
Long term care restorative dressing 0.30 0.11 0.74 -0.02 0.01 0.02 0.03 0.03 0.03 0.13
Long term care restorative transfers 0.31 0.14 0.73 0.00 0.00 0.01 0.06 0.01 0.00 0.09
Long term care restorative bed mobility 0.39 0.06 0.64 0.00 0.03 0.00 -0.03 0.01 0.03 -0.01
Long term care restorative splinting 0.02 -0.03 0.18 0.05 0.23 0.14 0.61 -0.03 0.00 -0.03
Long term care restorative Passive ROM 0.12 0.31 0.42 0.03 -0.03 -0.03 0.59 0.03 0.00 0.17
Long term care restorative Active ROM 0.10 0.28 0.47 0.01 -0.09 0.01 0.51 0.05 -0.01 0.17
Long term care restorative ambulation 0.05 0.33 0.62 0.02 -0.07 0.03 0.35 0.07 -0.02 0.22
Post acute plateguard 0.03 0.02 -0.01 0.06 -0.03 -0.14 0.12 0.12 0.10 0.24
Post acute Weight program -0.02 0.01 0.03 0.03 0.05 -0.03 0.03 0.01 0.93 0.02
Post acute Therapy Diet -0.01 -0.04 -0.04 0.04 0.10 -0.15 0.29 0.02 0.05 0.10
Post acute Bladder training -0.01 -0.04 0.06 -0.05 0.14 0.05 0.15 0.39 -0.07 -0.20
Post acute toilet program 0.11 -0.06 -0.05 0.18 -0.03 -0.07 -0.03 0.80 0.09 0.12











Table A-2. Continued
Post acute no skin treatment 0.01 -0.08 -0.04 -0.79
Post acute other skin tx 0.03 0.19 -0.01 0.32
Post acute ointment -0.03 0.16 0.02 0.04
Post acute pressure relief bed 0.01 0.04 0.03 0.84
Post acute pressure relief chair 0.10 0.14 -0.04 0.65
Post acute dressing application -0.04 0.07 0.01 0.03
Post acute surgical wound care 0.01 -0.06 0.00 0.12
Post acute ulcer care 0.00 0.07 -0.05 0.01
Post acute skin nutrition program 0.06 0.27 -0.03 0.09
Post acute turning repositioning 0.07 0.07 -0.06 0.26
Post acute OT minutes -0.14 -0.13 0.14 0.06
Post acute PT minutes -0.10 -0.19 0.14 0.08
Post acute other restorative care 0.23 0.04 0.11 -0.02
Post acute restorative communication 0.49 -0.03 0.14 0.01
Post acute restorative amputation 0.31 -0.02 -0.05 0.03
Post acute restorative eating 0.65 -0.03 0.29 0.03
Post acute restorative dressing 0.86 -0.01 0.26 0.00
Post acute restorative transfers 0.90 -0.02 0.19 0.01
Post acute restorative bed mobility 0.88 -0.03 0.15 0.01
S Post acute restorative splinting 0.30 -0.04 0.01 0.03
Post acute restorative Passive ROM 0.58 -0.02 -0.04 0.00
Post acute restorative Active ROM 0.65 0.01 0.01 0.01
Post acute restorative ambulation 0.79 0.02 0.10 0.01
Values over 0.4 are highlighted and represent the initial variables used to define the factor


-0.31
0.31
0.66
0.01
-0.06
0.77
0.42
0.71
0.52
0.39
0.05
0.10
-0.03
0.02
0.04
0.05
-0.03
-0.03
0.00
0.09
0.03
-0.07
-0.08


-0.03 -0.02
-0.06 0.00
-0.02 0.12
0.05 0.03
-0.09 0.06
0.06 0.07
0.15 -0.07
-0.03 0.15
-0.13 0.00
0.01 0.02
0.69 -0.10
0.67 -0.08
-0.02 0.18
-0.01 -0.12
0.01 0.12
0.00 -0.01
-0.01 -0.01
-0.03 0.02
-0.01 0.00
-0.03 0.52
-0.10 0.54
-0.08 0.46
-0.08 0.15


-0.09 -0.02
0.20 0.05
0.03 0.02
-0.04 -0.01
0.13 0.06
-0.07 -0.03
0.08 -0.12
-0.13 0.05
0.17 0.14
0.35 0.02
0.06 -0.10
0.11 -0.12
0.07 -0.01
-0.07 0.00
-0.01 -0.01
0.02 0.01
0.04 0.01
0.05 -0.01
0.03 0.00
0.05 0.01
0.02 0.02
0.05 0.01
0.10 -0.01


-0.08
0.15
0.00
-0.06
-0.02
0.09
0.27
-0.13
-0.16
-0.18
-0.02
0.02
-0.19
-0.01
0.00
0.02
0.04
0.02
0.00
0.01
-0.03
-0.03
0.05













Table A-3. Factor analysis with 10 factors of scope (technology) variables
Rotated Factor Pattern
Post acute LTC skin LTC Pressure Post acute Therapy Toilet Weight
restorative care restorative devices skin care minutes program program
Long term care plateguard -0.01 0.13 0.07 0.02 -0.04 0.01 0.02 0.00
Long term care Weight program -0.01 -0.01 0.04 0.03 0.02 -0.03 0.02 0.92
Long term care Therapy Diet 0.01 0.17 0.07 -0.03 0.03 0.05 -0.05 -0.02
Long term care Bladder training -0.03 0.07 0.08 -0.11 0.02 0.05 0.30 -0.05
Long term care toilet program 0.07 -0.05 0.02 0.20 -0.06 -0.02 0.79 0.05
Long term care no skin treatment 0.00 0.09 0.00 -0.71 -0.24 -0.05 -0.11 -0.01
Long term care other skin tx -0.03 0.63 0.14 0.24 0.05 -0.05 0.08 0.00
Long term care ointment -0.06 0.70 0.18 0.02 0.13 0.02 -0.02 -0.02
Long term care pressure relief bed -0.05 0.57 0.18 0.59 -0.19 0.04 -0.11 -0.05
Long term care pressure relief chair 0.00 0.61 0.12 0.51 -0.20 -0.05 0.00 -0.02
Long term care dressing application -0.06 0.77 0.11 -0.07 0.16 0.13 -0.18 -0.01
Long term care surgical wound care 0.06 0.40 -0.09 -0.07 0.08 0.35 -0.07 0.01
Long term care ulcer care -0.03 0.72 0.07 -0.06 0.15 0.11 -0.20 0.00
Long term care skin nutrition program 0.02 0.60 0.08 0.04 0.21 -0.09 0.12 0.09
S Long term care turning repositioning -0.02 0.69 0.14 0.18 0.03 0.04 0.19 -0.03
Long term care OT minutes 0.03 0.34 -0.11 -0.05 -0.11 0.75 -0.09 0.10
Long term care PT minutes 0.03 0.38 -0.10 -0.05 -0.10 0.74 -0.09 0.09
Long term care other restorative care -0.07 0.15 0.47 0.01 -0.04 0.02 0.08 -0.02
Long term care restorative communication 0.17 0.01 0.39 0.04 0.04 -0.01 -0.07 0.01
Long term care restorative amputation -0.04 -0.03 0.38 0.05 -0.03 -0.02 0.01 0.00
Long term care restorative eating 0.17 0.12 0.72 0.00 0.01 0.06 -0.01 0.03
Long term care restorative dressing 0.30 0.11 0.74 -0.02 0.01 0.02 0.03 0.03
Long term care restorative transfers 0.31 0.14 0.73 0.00 0.00 0.01 0.01 0.00
Long term care restorative bed mobility 0.39 0.06 0.64 0.00 0.03 0.00 0.01 0.03
Long term care restorative splinting 0.02 -0.03 0.18 0.05 0.23 0.14 -0.03 0.00
Long term care restorative Passive ROM 0.12 0.31 0.42 0.03 -0.03 -0.03 0.03 0.00
Long term care restorative Active ROM 0.10 0.28 0.47 0.01 -0.09 0.01 0.05 -0.01
Long term care restorative ambulation 0.05 0.33 0.62 0.02 -0.07 0.03 0.07 -0.02
Post acute plateguard 0.03 0.02 -0.01 0.06 -0.03 -0.14 0.12 0.10
Post acute Weight program -0.02 0.01 0.03 0.03 0.05 -0.03 0.01 0.93
Post acute Therapy Diet -0.01 -0.04 -0.04 0.04 0.10 -0.15 0.02 0.05
Post acute Bladder training -0.01 -0.04 0.06 -0.05 0.14 0.05 0.39 -0.07
Post acute toilet program 0.11 -0.06 -0.05 0.18 -0.03 -0.07 0.80 0.09










Post acute LTC skin LTC Pressure Post acute Therapy Toilet Weight
restorative care restorative devices skin care minutes program program
Table A-3. Continued
Post acute no skin treatment 0.01 -0.08 -0.04 -0.79 -0.31 -0.03 -0.09 -0.02
Post acute other skin tx 0.03 0.19 -0.01 0.32 0.31 -0.06 0.20 0.05
Post acute ointment -0.03 0.16 0.02 0.04 0.66 -0.02 0.03 0.02
Post acute pressure relief bed 0.01 0.04 0.03 0.84 0.01 0.05 -0.04 -0.01
Post acute pressure relief chair 0.10 0.14 -0.04 0.65 -0.06 -0.09 0.13 0.06
Post acute dressing application -0.04 0.07 0.01 0.03 0.77 0.06 -0.07 -0.03
Post acute surgical wound care 0.01 -0.06 0.00 0.12 0.42 0.15 0.08 -0.12
Post acute ulcer care 0.00 0.07 -0.05 0.01 0.71 -0.03 -0.13 0.05
Post acute skin nutrition program 0.06 0.27 -0.03 0.09 0.52 -0.13 0.17 0.14
Post acute turning repositioning 0.07 0.07 -0.06 0.26 0.39 0.01 0.35 0.02
Post acute OT minutes -0.14 -0.13 0.14 0.06 0.05 0.69 0.06 -0.10
Post acute PT minutes -0.10 -0.19 0.14 0.08 0.10 0.67 0.11 -0.12
Post acute other restorative care 0.23 0.04 0.11 -0.02 -0.03 -0.02 0.07 -0.01
Post acute restorative communication 0.49 -0.03 0.14 0.01 0.02 -0.01 -0.07 0.00
Post acute restorative amputation 0.31 -0.02 -0.05 0.03 0.04 0.01 -0.01 -0.01
Post acute restorative eating 0.65 -0.03 0.29 0.03 0.05 0.00 0.02 0.01
Post acute restorative dressing 0.86 -0.01 0.26 0.00 -0.03 -0.01 0.04 0.01
S Post acute restorative transfers 0.90 -0.02 0.19 0.01 -0.03 -0.03 0.05 -0.01
Post acute restorative bed mobility 0.88 -0.03 0.15 0.01 0.00 -0.01 0.03 0.00
Post acute restorative splinting 0.30 -0.04 0.01 0.03 0.09 -0.03 0.05 0.01
Post acute restorative Passive ROM 0.58 -0.02 -0.04 0.00 0.03 -0.10 0.02 0.02
Post acute restorative Active ROM 0.65 0.01 0.01 0.01 -0.07 -0.08 0.05 0.01
Post acute restorative ambulation 0.79 0.02 0.10 0.01 -0.08 -0.08 0.10 -0.01
Values over 0.4 are highlighted and represent the initial variables used to define the fact










Table A-4. Eigenvalues of scope (technology) factor analysis


1
2
3
4
5
6
7



3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00


Eigenvalues of the Correlation Matrix:
Eigenvalue Difference Proportion
3.16 1.77 0.45
1.39 0.32 0.20
1.08 0.43 0.15
0.65 0.30 0.09
0.35 0.10 0.05
0.24 0.11 0.03
0.13 0.02


Cumulative
0.45
0.65
0.80
0.90
0.95
0.98
1


0 1 2 3 4 5 6 7 8


Figure A-2. Scree plot of resource staffing intensity from factor analysis


Table A-5. Factor analysis of resource staffing intensity with three factors
Total Total Total
therapist therapy nursing
assistants
Total PT per Resident 0.94987 0.13329 0.09479


Total OT per Resident
Total PTA per Resident
Total OTA per Resident
RN per Resident
LPN per Resident
CNA per Resident
Values over 0.4 are higl


0.93244
-0.02059
0.39775
0.0587
0.15186


0.10457
0.92192
0.78854
0.03395
0.56287


0.13311
0.05157
0.21303
0.79055
0.61385


0.13473 0.15559 0.86025
lighted and represent the initial variables used to define the factor


-*- Seriesi


7












Table A-6. Ward's minimum variance cluster analysis
The CLUSTER Procedure
Ward's Minimum Variance Cluster Analysis
Eigenvalues of the Covariance Matrix
Eigenvalue Difference Proportion Cumulative


6.24393174
5.61295630
4.08166266
3.49904251
1.82518078
1.42486178
1.15088786
1.00157383
0.54994229
0.42184551
0.40626272
0.33925671
0.23580760
0.14780646
0.12732224
0.10034939
0.08968920
0.06519883
0.03024354
0.02011316
0.00687351


0.63097545
1.53129364
0.58262014
1.67386173
0.40031900
0.27397392
0.14931403
0.45163154
0.12809678
0.01558279
0.06700601
0.10344911
0.08800114
0.02048422
0.02697285
0.01066020
0.02449037
0.03495529
0.01013037
0.01323966


0.2280
0.2050
0.1491
0.1278
0.0667
0.0520
0.0420
0.0366
0.0201
0.0154
0.0148
0.0124
0.0086
0.0054
0.0047
0.0037
0.0033
0.0024
0.0011
0.0007
0.0003


Root-Mean-Square Total-Sample Standard Deviation
Root-Mean-Square Distance Between Observations


0.2280
0.4330
0.5821
0.7099
0.7766
0.8286
0.8706
0.9072
0.9273
0.9427
0.9575
0.9699
0.9785
0.9839
0.9886
0.9923
0.9955
0.9979
0.9990
0.9997
1.0000
= 1.141862
= 7.400109


Table A-7. Cluster History
NCL ------Clusters Joined------


CL22
CL12
CL15
CL17
CL9
CL22
CL6
CL4
CL3
CL2


FREQ SPRSQ RSQ ERSQ


CL29
CL13
CL31
CL11
CL15
CL23
CL7
CL10
CL8
CL5


0.0250
0.0258
0.0286
0.0330
0.0466
0.0550
0.0887
0.0942
0.1260
0.1472


.645
.619
.591
.558
.511
.456
.367
.273
.147
.000


.605
.583
.558
.530
.497
.458
.400
.320
.188
.000


CCC PSF PST2 e
2.70 13.1 4.3
2.42 13.4 5.4
2.13 13.8 5.2
1.79 14.3 3.8
0.78 14.4 9.1
-.12 14.7 8.4
-1.7 13.7 13.6
-2.3 13.5 11.7
-2.2 12.6 13.1
0.00 12.









Table A-8. Determination of rank scores table
RN focus n=1634 Rehab Focus n=1816 Differentiator n=990 Cost leader n=2752 Lack of strategy n=2554
V LV TV R V LVTR V LV TV RV LV V LV TV R

Zwages 0.15 -0.10 0.25 5.0 0.10 -0.10 0.20 4.0 0.00 -0.10 0.10 2.0 -0.06 -0.10 0.04 0.8 -0.10 -0.10 0.00 0.0
ZRNperres 1.16 -0.56 1.72 5.0 0.38 -0.56 0.94 2.7 0.02 -0.56 0.58 1.7 -0.56 -0.56 0.00 0.0 -0.42 -0.56 0.14 0.4
ZLPNperres -0.26 -0.26 0.00 0.0 0.57 -0.26 0.83 5.0 -0.10 -0.26 0.16 1.0 -0.08 -0.26 0.18 1.1 -0.12 -0.26 0.14 0.9
ZCNAperres -1.02 -1.02 0.00 0.0 0.46 -1.02 1.48 5.0 0.27 -1.02 1.29 4.4 0.04 -1.02 1.06 3.6 0.18 -1.02 1.20 4.0
Therapy -0.07 -0.81 0.74 1.8 1.31 -0.81 2.12 5.0 -0.39 -0.81 0.42 1.0 -0.81 -0.81 0.00 0.0 0.14 -0.81 0.95 2.2
Therapy assist -0.19 -0.91 0.72 1.6 1.33 -0.91 2.24 5.0 -0.38 -0.91 0.53 1.2 0.20 -0.91 1.11 2.5 -0.91 -0.91 0.00 0.0
ZPT percent 0.06 -0.79 0.85 2.4 -0.24 -0.79 0.55 1.6 0.10 -0.79 0.89 2.5 -0.79 -0.79 0.00 0.0 0.95 -0.79 1.74 5.0
ZOT percent 0.07 -0.76 0.83 2.6 -0.14 -0.76 0.62 1.9 0.07 -0.76 0.83 2.6 -0.76 -0.76 0.00 0.0 0.84 -0.76 1.60 5.0
ZRN percent 1.67 -0.58 2.25 5.0 0.06 -0.58 0.64 1.4 -0.06 -0.58 0.52 1.2 -0.58 -0.58 0.00 0.0 -0.47 -0.58 0.11 0.3
ZLicense perc. 1.47 -0.48 1.95 5.0 0.21 -0.48 0.69 1.8 -0.32 -0.48 0.16 0.4 -0.45 -0.48 0.03 0.1 -0.48 -0.48 0.00 0.0

Total Resource 28 34 18 8 18
zltcskinpr-s 0.05 -0.31 0.36 2.1 0.56 -0.31 0.87 5.0 0.25 -0.31 0.56 3.2 -0.20 -0.31 0.11 0.6 -0.31 -0.31 0.00 0.0
Zrestorative 14d -0.21 -0.43 0.22 0.3 -0.43 -0.43 0.00 0.0 2.93 -0.43 3.36 5.0 -0.34 -0.43 0.09 0.1 -0.34 -0.43 0.09 0.1
Zrestorative Q -0.17 -0.31 0.14 0.3 0.05 -0.31 0.36 0.9 1.73 -0.31 2.04 5.0 -0.27 -0.31 0.04 0.1 -0.31 -0.31 0.00 0.0
Zrelief device -0.05 -0.17 0.12 1.5 0.24 -0.17 0.41 5.0 0.17 -0.17 0.34 4.1 -0.03 -0.17 0.14 1.7 -0.17 -0.17 0.00 0.0
Therapy min -0.08 -0.57 0.49 1.7 0.88 -0.57 1.45 5.0 -0.57 -0.57 0.00 0.0 -0.11 -0.57 0.46 1.6 -0.24 -0.57 0.33 1.1
Zacute skin care 0.07 -0.23 0.30 3.0 0.26 -0.23 0.49 5.0 0.23 -0.23 0.46 4.7 -0.08 -0.23 0.15 1.6 -0.23 -0.23 0.00 0.0
Ztoiletprogram -0.06 -0.07 0.01 0.1 -0.06 -0.07 0.01 0.1 0.29 -0.07 0.36 5.0 0.04 -0.07 0.11 1.5 -0.07 -0.07 0.00 0.0
Weight prog. 0.01 -0.10 0.11 2.5 -0.10 -0.10 0.00 0.0 0.12 -0.10 0.22 5.0 0.06 -0.10 0.16 3.7 -0.05 -0.10 0.05 1.0
Percent Mcare -0.13 -0.39 0.26 0.7 1.37 -0.39 1.76 5.0 -0.36 -0.39 0.03 0.0 -0.34 -0.39 0.05 0.2 -0.39 -0.39 0.00 0.1
Total scope 12 _26 32 11 2.3
V is the actual Z score value
LV is the lowest value for the variable among clusters
TV is the total variance is the difference between the highest value among the clusters and the lowest values among clusters
R is the final rank from 0 to 5. Determined ((V-LV)/TV*5)









Table A-9. Descriptives of dependent variables, independent variables and control variables
Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5
Differentiator Cost Leader Rehab focus Specialty Lack of
N=990 N=2752 N=1816 care strategy
N=1634 N=2534
Avg Wages 12.61 12.52 12.99 12.76 12.45
RN per Resident 0.06 0.04 0.07 0.08 0.05
LPN per Resident 0.14 0.14 0.17 0.14 0.14
CNA per Resident 0.44 0.42 0.47 0.31 0.44
Therapist per Res. 0.02 0.01 0.04 0.02 0.02
Therapist asst per Res. 0.01 0.02 0.03 0.01 0.01
Percent Licensed PT 0.62 0.39 0.54 0.6 0.84
Percent Licensed OT 0.65 0.41 0.6 0.64 0.86
Percent RN 0.09 0.07 0.1 0.17 0.08
LTC Skin Process 0.83 0.72 0.87 0.78 0.69
Acute Skin Process 0.91 0.83 0.91 0.86 0.81
Relief device 1.95 1.8 1.98 1.79 1.72
LTC Restorative Process 4.28 1.8 2.03 1.88 1.67
Acute Restorative Process 3.69 0.47 0.39 0.58 0.44
Therapy minutes 324.7 357.8 439.6 357.3 349.6
Toilet Program 0.68 0.59 0.58 0.57 0.57
Weight Program 0.34 0.32 0.28 0.31 0.29
Percent Medicare 10.7 10.82 22.2 11.88 10.47
Percent Other 23.39 21.7 26.56 24.68 22.65
Percent Medicaid 65.91 67.48 51.24 63.45 66.88
Acuindex 10.39 10.2 10.45 10.25 10.17
ADL index 10.13 10.02 10.23 10.03 9.98
Percent tube feed 0.07 0.06 0.06 0.06 0.06
Percent tracheotomy 0.01 0.01 0.01 0.01 0
Percent requiring suctioning 0.01 0.01 0.01 0.01 0.01
Percent requiring respiratory care 0.11 0.1 0.12 0.11 0.1
Percent requiring IV 0.01 0.01 0.02 0.01 0.01
Percent requiring injection 0.15 0.15 0.17 0.16 0.15
Quality care deficiency 1.782 1.928 1.749 1.777 1.929
Bladder decline 0.186 0.193 0.209 0.192 0.187
Bowel decline 0.169 0.176 0.188 0.176 0.168
ADL 4 point decline 0.165 0.172 0.17 0.169 0.17
Restraint 0.076 0.079 0.075 0.076 0.077
Long term care pressure ulcer 0.08 0.076 0.093 0.081 0.075
Post acute walking improvement 0.232 0.238 0.246 0.24 0.246
Post acute pressure ulcer 0.187 0.173 0.177 0.174 0.169
Ownership 0.71 0.76 0.8 0.77 0.76
Market Competition 0.13 0.14 0.14 0.16 0.14
Chain 0.52 0.6 0.66 0.63 0.58
Total beds 114 130 116 116 111









Table A-10. Negative binomial of quality of care deficiencies with differentiators as reference group
IRR Std. Err. z P>z
Cost leaders 1.059 0.036 1.72 0.09
Rehab focus 0.993 0.038 -0.18 0.86
Specialty care focus 1.001 0.037 0.04 0.97
Lack of strategy 1.124 0.038 3.43 0.00
Ownership 1.099 0.029 3.59 0.00
Percent Medicaid 1.005 0.001 7.69 0.00
Percent Medicare 1.005 0.001 3.75 0.00
Market competition 0.773 0.367 -0.54 0.59
Occupancy 0.830 0.065 -2.38 0.02
Total beds 1.003 0.000 14.74 0.00
Chain 1.024 0.021 1.18 0.24
Acuity index 1.040 0.013 3.20 0.00
ADL index 0.952 0.029 -1.63 0.10
Tube feed care 1.009 0.237 0.04 0.97
Respiratory care 0.972 0.157 -0.18 0.86
Suctioning care 0.811 0.454 -0.37 0.71
IV therapy care 3.826 1.304 3.94 0.00
Tracheostomy care 1.142 0.745 0.20 0.84
Need injection 1.189 0.141 1.46 0.14
Ifips_2 1.206 0.470 0.48 0.63

Table A-11. OLS regressions of root transformed bladder decline with differentiators as ref group
Coef Std. Err. t P>t
Cost leaders 0.006 0.003 2.03 0.04
Rehab focus 0.012 0.004 3.27 0.00
Specialty care focus 0.009 0.004 2.68 0.01
Lack of strategy 0.005 0.003 1.56 0.12
Ownership 0.006 0.002 2.46 0.01
Percent Medicaid 0.000 0.000 -5.35 0.00
Percent Medicare 0.001 0.000 3.84 0.00
Market competition 0.077 0.064 1.21 0.23
Occupancy 0.014 0.008 1.70 0.09
Total beds 0.000 0.000 3.66 0.00
Chain 0.004 0.002 1.97 0.05
Acuity index 0.009 0.001 6.66 0.00
ADL index 0.021 0.003 6.68 0.00
Tube feed care -0.061 0.027 -2.24 0.03
Respiratory care -0.040 0.014 -2.99 0.00
Suctioning care 0.097 0.067 1.44 0.15
IV therapy care 0.004 0.040 0.09 0.93
Tracheostomy care -0.245 0.074 -3.32 0.00
Need injection??? 0.004 0.012 0.33 0.74
cons 0.241 0.017 13.91 0.00









Table A-12. OLS regressions of root transformed bowel decline with differentiators as ref. group
Coef. Std. Err. t P>t
Cost leaders 0.008 0.004 2.31 0.02
Rehab focus 0.009 0.004 2.22 0.03
Specialty care focus 0.012 0.004 2.92 0.00
Lack of strategy 0.005 0.004 1.41 0.16
Ownership 0.005 0.003 1.83 0.07
Percent Medicaid 0.000 0.000 -3.76 0.00
Percent Medicare 0.001 0.000 3.82 0.00
Market competition 0.040 0.050 0.79 0.43
Occupancy 0.013 0.009 1.49 0.14
Total beds 0.000 0.000 2.94 0.00
Chain 0.004 0.002 1.79 0.07
Acuity index 0.011 0.001 7.26 0.00
ADL index 0.025 0.003 7.43 0.00
Tube feed care -0.039 0.029 -1.34 0.18
Respiratory care -0.040 0.015 -2.74 0.01
Suctioning care 0.014 0.077 0.18 0.86
IV therapy care 0.016 0.044 0.36 0.72
Tracheostomy care -0.101 0.085 -1.19 0.23
Need injection 0.020 0.013 1.55 0.12
cons 0.176 0.017 10.11 0.00


Table A-13. OLS regressions


Cost leaders
Rehab focus
Specialty care focus
Lack of strategy
Ownership
Percent Medicaid
Percent Medicare
Market competition
Occupancy
Total beds
Chain
Acuity index
ADL index
Tube feed care
Respiratory care
Suctioning care
IV therapy care
Tracheostomy care
Need injection
cons


of root transformed
Coef.
0.008
0.003
0.005
0.004
0.001
0.000
0.000
0.083
0.006
0.000
-0.001
0.002
-0.003
-0.028
-0.005
-0.054
-0.013
-0.072
-0.008
0.374


ADL decline with
Std. Err.
0.003
0.003
0.003
0.003
0.002
0.000
0.000
0.053
0.007
0.000
0.002
0.001
0.003
0.026
0.013
0.057
0.033
0.066
0.011
0.016


differentiators as ref
t
2.94
0.86
1.67
1.50
0.50
0.10
2.97
1.57
0.84
-0.71
-0.72
1.46
-1.01
-1.10
-0.42
-0.95
-0.39
-1.10
-0.69
23.87


group
P>t
0.00
0.39
0.10
0.13
0.62
0.92
0.00
0.12
0.40
0.48
0.47
0.14
0.31
0.27
0.67
0.34
0.70
0.27
0.49
0.00









Table A-14. OLS regressions of log transformed restraints with differentiators as ref group
Coef. Std. Err. t P>t
Cost leaders 0.011 0.004 2.48 0.01
Rehab focus 0.004 0.006 0.74 0.46
Specialty care focus 0.006 0.005 1.32 0.19
Lack of strategy 0.006 0.005 1.24 0.22
Ownership 0.020 0.004 5.50 0.00
Percent Medicaid 0.000 0.000 4.06 0.00
Percent Medicare 0.000 0.000 1.16 0.25
Market competition 0.053 0.070 0.76 0.45
Cost leaders -0.010 0.012 -0.84 0.40
Total beds 0.000 0.000 7.90 0.00
Chain -0.007 0.003 -2.24 0.03
Acuity index 0.011 0.002 6.07 0.00
ADL index 0.016 0.004 3.80 0.00
Tube feed care -0.049 0.041 -1.20 0.23
Respiratory care -0.012 0.022 -0.54 0.59
Suctioning care -0.138 0.101 -1.37 0.17
IV therapy care 0.028 0.054 0.51 0.61
Tracheostomy care 0.060 0.110 0.54 0.59
Need injection -0.055 0.018 -3.12 0.00
cons -0.004 0.023 -0.19 0.85


Table A-15. OLS regressions


Cost leaders
Rehab focus
Specialty care focus
Lack of strategy
Ownership
Percent Medicaid
Percent Medicare
Market competition
Occupancy
Total beds
Chain
Acuity index
ADL index
Tube feed care
Respiratory care
Suctioning care
IV therapy care
Tracheostomy care
Need injection
cons


of root transformed LTC ulcers
Coef. Std. Err.
-0.001 0.002
0.012 0.003
0.004 0.003
-0.004 0.002
0.006 0.002
0.000 0.000
0.001 0.000
-0.001 0.029
-0.035 0.006
0.000 0.000
-0.002 0.002
0.008 0.001
0.007 0.002
0.109 0.024
0.031 0.011
0.045 0.060
0.150 0.032
-0.031 0.065
0.046 0.010
0.151 0.012


with differentiators as ref.
t
-0.45
4.19
1.47
-1.71
3.32
-0.71
6.94
-0.04
-5.55
3.31
-1.57
8.42
3.11
4.6
2.74
0.74
4.7
-0.47
4.74
12.84


group
P>t
0.66
0.00
0.14
0.09
0.00
0.48
0.00
0.97
0.00
0.00
0.12
0.00
0.00
0.00
0.01
0.46
0.00
0.64
0.00
0.00









Table A-16. GLM with gamma distribution with PA walking improve with differentiators as ref group
Coef. Std. Err. z P>z
Cost leaders 0.031 0.019 1.65 0.10
Rehab focus 0.050 0.022 2.32 0.02
Specialty care focus 0.037 0.021 1.76 0.08
Lack of strategy 0.060 0.019 3.17 0.00
Ownership -0.087 0.014 -6.04 0.00
Percent Medicaid -0.004 0.000 -10.18 0.00
Percent Medicare -0.002 0.001 -2.81 0.01
Market competition -0.425 0.248 -1.71 0.09
Occupancy -0.160 0.043 -3.73 0.00
Total beds 0.000 0.000 -0.53 0.60
Chain -0.032 0.012 -2.74 0.01
Acuity index -0.028 0.007 -4.07 0.00
ADL index -0.110 0.017 -6.49 0.00
Tube feed care -0.666 0.143 -4.66 0.00
Respiratory care 0.139 0.081 1.71 0.09
Suctioning care 0.444 0.419 1.06 0.29
IV therapy care 0.240 0.213 1.13 0.26
Tracheostomy care -0.238 0.449 -0.53 0.60
Need injection 0.038 0.068 0.56 0.58
cons 0.38 0.283 0.77 0.44


Table A-17. OLS regressions of root transformed PA skin ulcers with
Coef. Std. Err.
Cost leaders -0.045 0.020
Rehab focus -0.015 0.021
Specialty care focus -0.027 0.021
Lack of strategy -0.076 0.020
Ownership 0.034 0.014
Percent Medicaid 0.001 0.000
Percent Medicare 0.001 0.001
Market competition -0.031 0.308
Occupancy 0.014 0.046
Total beds 0.000 0.000
Chain -0.021 0.012
Acuity index 0.034 0.007
ADL index 0.027 0.019
Tube feed care 0.568 0.156
Respiratory care 0.088 0.082
Suctioning care 0.535 0.413
IV therapy care 0.270 0.194
Tracheostomy care -0.160 0.431
Need injection 0.183 0.069
cons -2.513 0.098


differentiators as
t
-2.30
-0.74
-1.28
-3.81
2.38
3.39
0.70
-0.10
0.31
5.08
-1.76
4.88
1.42
3.64
1.07
1.30
1.40
-0.37
2.63
-25.77


ref. group
P>t
0.02
0.46
0.20
0.00
0.02
0.00
0.48
0.92
0.75
0.00
0.08
0.00
0.16
0.00
0.28
0.20
0.16
0.71
0.01
0.00









Table A-18. Negative binomial of quality of care deficiencies with rehab focus as reference group
IRR Std. Err. z P>z
Cost leader 1.067 0.032 2.14 0.03
Differentiator 1.007 0.038 0.18 0.86
Specialty care focus 1.008 0.033 0.25 0.80
Lack of strategy 1.132 0.035 4.03 0.00
Ownership 1.099 0.029 3.59 0.00
Percent Medicaid 1.005 0.001 7.69 0.00
Percent Medicare 1.005 0.001 3.75 0.00
Market competition 0.773 0.367 -0.54 0.59
Occupancy 0.830 0.065 -2.38 0.02
Total beds 1.003 0.000 14.74 0.00
Chain 1.024 0.021 1.18 0.24
Acuity index 1.040 0.013 3.20 0.00
ADL index 0.952 0.029 -1.63 0.10
Tube feed care 1.009 0.237 0.04 0.97
Respiratory care 0.972 0.157 -0.18 0.86
Suctioning care 0.811 0.454 -0.37 0.71
IV therapy care 3.826 1.304 3.94 0.00
Tracheostomy care 1.142 0.745 0.20 0.84
Need injection 1.189 0.141 1.46 0.14
Ifips_2 1.206 0.470 0.48 0.63
Ifips_3 1.025 0.198 0.13 0.90


Table A-19. OLS regressions of root
Coef.
Cost leader -0.006
Differentiator -0.012
Specialty care focus -0.003
Lack of strategy -0.007
Ownership 0.006
Percent Medicaid 0.000
Percent Medicare 0.001
Market competition 0.077
Occupancy 0.014
Total beds 0.000
Chain 0.004
Acuity index 0.009
ADL index 0.021
Tube feed care -0.061
Respiratory care -0.040
Suctioning care 0.097
IV therapy care 0.004
Tracheostomy care -0.245
Need injection 0.004
cons 0.253


transformed bladder
Std. Err.
0.003
0.004
0.003
0.003
0.002
0.000
0.000
0.064
0.008
0.000
0.002
0.001
0.003
0.027
0.014
0.067
0.040
0.074
0.012
0.017


decline with rehab
t
-1.97
-3.27
-0.86
-2.32
2.46
-5.35
3.84
1.21
1.70
3.66
1.97
6.66
6.68
-2.24
-2.99
1.44
0.09
-3.32
0.33
14.72


focus as ref.


group
P>t
0.05
0.00
0.39
0.02
0.01
0.00
0.00
0.23
0.09
0.00
0.05
0.00
0.00
0.03
0.00
0.15
0.93
0.00
0.74
0.00









Table A-20. OLS regressions of root transformed bowel
Coef. Std. Err.
Cost leader -0.001 0.003
Differentiator -0.009 0.004
Specialty care focus 0.002 0.003
Lack of strategy -0.004 0.003
Ownership 0.005 0.003
Percent Medicaid 0.000 0.000
Percent Medicare 0.001 0.000
Market competition 0.040 0.050
Occupancy 0.013 0.009
Total beds 0.000 0.000
Chain 0.004 0.002
Acuity index 0.011 0.001
ADL index 0.025 0.003
Tube feed care -0.039 0.029
Respiratory care -0.040 0.015
Suctioning care 0.014 0.077
IV therapy care 0.016 0.044
Tracheostomy care -0.101 0.085
Need injection 0.020 0.013
cons 0.186 0.017


decline with rehab focus as ref group
t P>t
-0.29 0.77
-2.22 0.03
0.64 0.52
-1.26 0.21
1.83 0.07
-3.76 0.00
3.82 0.00
0.79 0.43
1.49 0.14
2.94 0.00
1.79 0.07
7.26 0.00
7.43 0.00
-1.34 0.18
-2.74 0.01
0.18 0.86
0.36 0.72
-1.19 0.23
1.55 0.12
10.67 0.00


Table A-21. OLS regressions of root transformed ADL
Coef. Std. Err.
Cost leader 0.006 0.003
Differentiator -0.003 0.003
Specialty care focus 0.002 0.003
Lack of strategy 0.002 0.003
Ownership 0.001 0.002
Percent Medicaid 0.000 0.000
Percent Medicare 0.000 0.000
Market competition 0.083 0.053
Occupancy 0.006 0.007
Total beds 0.000 0.000
Chain -0.001 0.002
Acuity index 0.002 0.001
ADL index -0.003 0.003
Tube feed care -0.028 0.026
Respiratory care -0.005 0.013
Suctioning care -0.054 0.057
IV therapy care -0.013 0.033
Tracheostomy care -0.072 0.066
Need injection -0.008 0.011
cons 0.377 0.016


decline with
t
2.18
-0.86
0.86
0.63
0.50
0.10
2.97
1.57
0.84
-0.71
-0.72
1.46
-1.01
-1.10
-0.42
-0.95
-0.39
-1.10
-0.69
23.89


rehab focus as ref. group
P>t
0.03
0.39
0.39
0.53
0.62
0.92
0.00
0.12
0.40
0.48
0.47
0.14
0.31
0.27
0.67
0.34
0.70
0.27
0.49
0.00









Table A-22. OLS regressions of log transformed restraint with rehab focus as ref. group
Coef. Std. Err. t P>t
Cost leader 0.011 0.004 2.48 0.01
Differentiator 0.004 0.006 0.74 0.46
Specialty care focus 0.006 0.005 1.32 0.19
Lack of strategy 0.006 0.005 1.24 0.22
Ownership 0.020 0.004 5.50 0.00
Percent Medicaid 0.000 0.000 4.06 0.00
Percent Medicare 0.000 0.000 1.16 0.25
Market competition 0.053 0.070 0.76 0.45
Occupancy -0.010 0.012 -0.84 0.40
Total beds 0.000 0.000 7.90 0.00
Chain -0.007 0.003 -2.24 0.03
Acuity index 0.011 0.002 6.07 0.00
ADL index 0.016 0.004 3.80 0.00
Tube feed care -0.049 0.041 -1.20 0.23
Respiratory care -0.012 0.022 -0.54 0.59
Suctioning care -0.138 0.101 -1.37 0.17
IV therapy care 0.028 0.054 0.51 0.61
Tracheostomy care 0.060 0.110 0.54 0.59
Need injection -0.055 0.018 -3.12 0.00
cons -0.004 0.023 -0.19 0.85


Table A-23. OLS regressions of root
Coef.
Cost leader -0.013
Differentiator -0.012
Specialty care focus -0.008
Lack of strategy -0.016
Ownership 0.006
Percent Medicaid 0.000
Percent Medicare 0.001
Market competition -0.001
Occupancy -0.035
Total beds 0.000
Chain -0.002
Acuity index 0.008
ADL index 0.007
Tube feed care 0.109
Respiratory care 0.031
Suctioning care 0.045
IV therapy care 0.150
Tracheostomy care -0.031
Need injection 0.046
cons 0.163


transformed skin ulcer with
Std. Err.
0.002
0.003
0.002
0.002
0.002
0.000
0.000
0.029
0.006
0.000
0.002
0.001
0.002
0.024
0.011
0.060
0.032
0.065
0.010
0.012


rehab focus
t
-5.69
-4.19
-3.34
-6.89
3.32
-0.71
6.94
-0.04
-5.55
3.31
-1.57
8.42
3.11
4.60
2.74
0.74
4.70
-0.47
4.74
13.90


as ref group
P>t
0.00
0.00
0.00
0.00
0.00
0.48
0.00
0.97
0.00
0.00
0.12
0.00
0.00
0.00
0.01
0.46
0.00
0.64
0.00
0.00









Table A-24. GLM of PA walking improve with gamma distribution with rehab focus as ref group
Coef. Std. Err. z P>z
Cost leader -0.019 0.018 -1.09 0.28
Differentiator -0.050 0.022 -2.32 0.02
Specialty care focus -0.014 0.019 -0.73 0.47
Lack of strategy 0.010 0.018 0.57 0.57
Ownership -0.087 0.014 -6.04 0.00
Percent Medicaid -0.004 0.000 -10.18 0.00
Percent Medicare -0.002 0.001 -2.81 0.01
Market competition -0.425 0.248 -1.71 0.09
Occupancy -0.160 0.043 -3.73 0.00
Total beds 0.000 0.000 -0.53 0.60
Chain -0.032 0.012 -2.74 0.01
Acuity index -0.028 0.007 -4.07 0.00
ADL index -0.110 0.017 -6.49 0.00
Tube feed care -0.666 0.143 -4.66 0.00
Respiratory care 0.139 0.081 1.71 0.09
Suctioning care 0.444 0.419 1.06 0.29
IV therapy care 0.240 0.213 1.13 0.26
Tracheostomy care -0.238 0.449 -0.53 0.60
Need injection 0.038 0.068 0.56 0.58
Ifips_2 0.218 0.283 0.77 0.44
Ifips_3 0.209 0.188 1.11 0.27


Table A-25. OLS regressions of root
Coef.
Cost leader -0.030
Differentiator 0.015
Specialty care focus -0.012
Lack of strategy -0.061
Ownership 0.034
Percent Medicaid 0.001
Percent Medicare 0.001
Market competition -0.031
Occupancy 0.014
Total beds 0.000
Chain -0.021
Acuity index 0.034
ADL index 0.027
Tube feed care 0.568
Respiratory care 0.088
Suctioning care 0.535
IV therapy care 0.270
Tracheotomy care -0.160
Need injection 0.183
cons -2.528


transformed PA skin
Std. Err.
0.016
0.021
0.017
0.016
0.014
0.000
0.001
0.308
0.046
0.000
0.012
0.007
0.019
0.156
0.082
0.413
0.194
0.431
0.069
0.097


ulcers with rehab
t
-1.91
0.74
-0.69
-3.80
2.38
3.39
0.70
-0.10
0.31
5.08
-1.76
4.88
1.42
3.64
1.07
1.30
1.40
-0.37
2.63
-26.06


focus as ref.


group
P>t
0.06
0.46
0.49
0.00
0.02
0.00
0.48
0.92
0.75
0.00
0.08
0.00
0.16
0.00
0.28
0.20
0.16
0.71
0.01
0.00









Table A-26. Negative binomial of quality of care deficiencies with specialty care as reference group
IRR Std. Err. z P>z
Cost leader 1.058 0.030 2.00 0.05
Rehab focus 0.992 0.033 -0.25 0.80
Differentiator 0.999 0.037 -0.04 0.97
Lack of strategy 1.123 0.032 4.02 0.00
Ownership 1.099 0.029 3.59 0.00
Percent Medicaid 1.005 0.001 7.69 0.00
Percent Medicare 1.005 0.001 3.75 0.00
Market competition 0.773 0.367 -0.54 0.59
Occupancy 0.830 0.065 -2.38 0.02
Total beds 1.003 0.000 14.74 0.00
Chain 1.024 0.021 1.18 0.24
Acuity index 1.040 0.013 3.20 0.00
ADL index 0.952 0.029 -1.63 0.10
Tube feed care 1.009 0.237 0.04 0.97
Respiratory care 0.972 0.157 -0.18 0.86
Suctioning care 0.811 0.454 -0.37 0.71
IV therapy care 3.826 1.304 3.94 0.00
Tracheostomy care 1.142 0.745 0.20 0.84
Need injection 1.189 0.141 1.46 0.14
Ifips_2 1.206 0.470 0.48 0.63

Table A-27. OLS regressions of root transformed bladder decline with specialty care focus as ref
group
Coef. Std. Err. t P>t
Cost leader -0.003 0.003 -1.13 0.26
Rehab focus 0.003 0.003 0.86 0.39
Differentiator -0.009 0.004 -2.68 0.01
Lack of strategy -0.004 0.003 -1.55 0.12
Ownership 0.006 0.002 2.46 0.01
Percent Medicaid 0.000 0.000 -5.35 0.00
Percent Medicare 0.001 0.000 3.84 0.00
Market competition 0.077 0.064 1.21 0.23
Occupancy 0.014 0.008 1.70 0.09
Total beds 0.000 0.000 3.66 0.00
Chain 0.004 0.002 1.97 0.05
Acuity index 0.009 0.001 6.66 0.00
ADL index 0.021 0.003 6.68 0.00
Tube feed care -0.061 0.027 -2.24 0.03
Respiratory care -0.040 0.014 -2.99 0.00
Suctioning care 0.097 0.067 1.44 0.15
IV therapy care 0.004 0.040 0.09 0.93
Tracheostomy care -0.245 0.074 -3.32 0.00
Need injection 0.004 0.012 0.33 0.74
cons 0.251 0.017 14.71 0.00









Table A-28. OLS regressions of root transformed bowel
Coef. Std. Err.
Cost leader -0.003 0.003
Rehab focus -0.002 0.003
Differentiator -0.012 0.004
Lack of strategy -0.006 0.003
Ownership 0.005 0.003
Percent Medicaid 0.000 0.000
Percent Medicare 0.001 0.000
Market competition 0.040 0.050
Occupancy 0.013 0.009
Total beds 0.000 0.000
Chain 0.004 0.002
Acuity index 0.011 0.001
ADL index 0.025 0.003
Tube feed care -0.039 0.029
Respiratory care -0.040 0.015
Suctioning care 0.014 0.077
IV therapy care 0.016 0.044
Tracheostomy care -0.101 0.085
Need injection 0.020 0.013
cons 0.188 0.017


decline with specialty
t
-1.10
-0.64
-2.92
-2.18
1.83
-3.76
3.82
0.79
1.49
2.94
1.79
7.26
7.43
-1.34
-2.74
0.18
0.36
-1.19
1.55
10.99


care as ref. group
P>t
0.27
0.52
0.00
0.03
0.07
0.00
0.00
0.43
0.14
0.00
0.07
0.00
0.00
0.18
0.01
0.86
0.72
0.23
0.12
0.00


Table A-29. OLS regressions of root transformed ADL decline with
Coef. Std. Err.
Cost leader 0.003 0.002
Rehab focus -0.002 0.003
Differentiator -0.005 0.003
Lack of strategy -0.001 0.002
Ownership 0.001 0.002
Percent Medicaid 0.000 0.000
Percent Medicare 0.000 0.000
Market competition 0.083 0.053
Occupancy 0.006 0.007
Total beds 0.000 0.000
Chain -0.001 0.002
Acuity index 0.002 0.001
ADL index -0.003 0.003
Tube feed care -0.028 0.026
Respiratory care -0.005 0.013
Suctioning care -0.054 0.057
IV therapy care -0.013 0.033
Tracheostomy care -0.072 0.066
Need injection -0.008 0.011
cons 0.379 0.016


specialty
t
1.39
-0.86
-1.67
-0.28
0.50
0.10
2.97
1.57
0.84
-0.71
-0.72
1.46
-1.01
-1.10
-0.42
-0.95
-0.39
-1.10
-0.69
24.46


care as ref. group
P>t
0.16
0.39
0.10
0.78
0.62
0.92
0.00
0.12
0.40
0.48
0.47
0.14
0.31
0.27
0.67
0.34
0.70
0.27
0.49
0.00









Table A-30. OLS regressions of root transformed restraint with specialty care as ref. group
Coef. Std. Err. t P>t
Cost leader 0.005 0.004 1.22 0.22
Rehab focus -0.006 0.005 -1.32 0.19
Differentiator -0.002 0.005 -0.41 0.68
Lack of strategy -0.001 0.004 -0.15 0.88
Ownership 0.020 0.004 5.50 0.00
Percent Medicaid 0.000 0.000 4.06 0.00
Percent Medicare 0.000 0.000 1.16 0.25
Market competition 0.053 0.070 0.76 0.45
Occupancy -0.010 0.012 -0.84 0.40
Total beds 0.000 0.000 7.90 0.00
Chain -0.007 0.003 -2.24 0.03
Acuity index 0.011 0.002 6.07 0.00
ADL index 0.016 0.004 3.80 0.00
Tube feed care -0.049 0.041 -1.20 0.23
Respiratory care -0.012 0.022 -0.54 0.59
Suctioning care -0.138 0.101 -1.37 0.17
IV therapy care 0.028 0.054 0.51 0.61
Tracheostomy care 0.060 0.110 0.54 0.59
Need injection -0.055 0.018 -3.12 0.00
cons 0.002 0.023 0.08 0.93


Table A-31. OLS regressions of root transformed LTC skin
Coef. Std. Err.
Cost leader -0.005 0.002
Rehab focus 0.008 0.002
Differentiator -0.004 0.003
Lack of strategy -0.008 0.002
Ownership 0.006 0.002
Percent Medicaid 0.000 0.000
Percent Medicare 0.001 0.000
Market competition -0.001 0.029
Occupancy -0.035 0.006
Total beds 0.000 0.000
Chain -0.002 0.002
Acuity index 0.008 0.001
ADL index 0.007 0.002
Tube feed care 0.109 0.024
Respiratory care 0.031 0.011
Suctioning care 0.045 0.060
IV therapy care 0.150 0.032
Tracheostomy care -0.031 0.065
Need injection 0.046 0.010
cons 0.155 0.012


ulcer with specialty
t
-2.50
3.34
-1.47
-3.97
3.32
-0.71
6.94
-0.04
-5.55
3.31
-1.57
8.42
3.11
4.60
2.74
0.74
4.70
-0.47
4.74
13.38


care as ref. group
P>t
0.01
0.00
0.14
0.00
0.00
0.48
0.00
0.97
0.00
0.00
0.12
0.00
0.00
0.00
0.01
0.46
0.00
0.64
0.00
0.00









Table A-32. GLM with gamma distribution PA walking improve with specialty care as ref. group
Coef. Std. Err. Z P>z
Cost leader -0.005 0.016 -0.33 0.74
Rehab focus 0.014 0.019 0.73 0.47
Differentiator -0.037 0.021 -1.76 0.08
Lack of strategy 0.024 0.016 1.45 0.15
Ownership -0.087 0.014 -6.04 0.00
Percent Medicaid -0.004 0.000 -10.18 0.00
Percent Medicare -0.002 0.001 -2.81 0.01
Market competition -0.425 0.248 -1.71 0.09
Occupancy -0.160 0.043 -3.73 0.00
Total beds 0.000 0.000 -0.53 0.60
Chain -0.032 0.012 -2.74 0.01
Acuity index -0.028 0.007 -4.07 0.00
ADL index -0.110 0.017 -6.49 0.00
Tube feed care -0.666 0.143 -4.66 0.00
Respiratory care 0.139 0.081 1.71 0.09
Suctioning care 0.444 0.419 1.06 0.29
IV therapy care 0.240 0.213 1.13 0.26
Tracheostomy care -0.238 0.449 -0.53 0.60
Need injection 0.038 0.068 0.56 0.58
Ifips_2 0.218 0.283 0.77 0.44
cons -0.35 0.188 1.11 0.27


Table A-33. OLS regressions of root
Coef.
Cost leader -0.018
Rehab focus 0.012
Differentiator 0.027
Lack of strategy -0.049
Ownership 0.034
Percent Medicaid 0.001
Percent Medicare 0.001
Market competition -0.031
Occupancy 0.014
Total beds 0.000
Chain -0.021
Acuity index 0.034
ADL index 0.027
Tube feed care 0.568
Respiratory care 0.088
Suctioning care 0.535
IV therapy care 0.270
Tracheostomy care -0.160
Need injection 0.183
cons -2.540


transformed PA skin
Std. Err.
0.016
0.017
0.021
0.016
0.014
0.000
0.001
0.308
0.046
0.000
0.012
0.007
0.019
0.156
0.082
0.413
0.194
0.431
0.069
0.096


ulcers with specialty
T
-1.15
0.69
1.28
-3.00
2.38
3.39
0.70
-0.10
0.31
5.08
-1.76
4.88
1.42
3.64
1.07
1.30
1.40
-0.37
2.63
-26.36


care as ref. group
P>t
0.25
0.49
0.20
0.00
0.02
0.00
0.48
0.92
0.75
0.00
0.08
0.00
0.16
0.00
0.28
0.20
0.16
0.71
0.01
0.00










Table A-34. GLM regressions of nursing home cost with gamma distribution
Coef. Std. Err. z P>z
Differentiator 0.002 0.016 0.12 0.90
Rehab focus 0.080 0.015 5.42 0.00
Specialty care focus 0.039 0.014 2.91 0.00
Lack of strategy -0.006 0.012 -0.54 0.59
Ownership -0.108 0.012 -8.87 0.00
Percent Medicaid -0.009 0.000 -27.82 0.00
Percent Medicare 0.001 0.001 1.49 0.13
Market competition -0.188 0.222 -0.85 0.39
Occupancy -0.904 0.035 -26.03 0.00
Total beds -0.001 0.000 -11.94 0.00
Chain -0.043 0.010 -4.38 0.00
Acuity index -0.012 0.006 -2.09 0.03
ADL index 0.001 0.014 0.07 0.94
Tube feed care 0.012 0.120 0.10 0.92
Respiratory care 0.159 0.069 2.29 0.02
Suctioning care 0.330 0.334 0.99 0.32
IV therapy care 0.056 0.175 0.32 0.75
Tracheotomy care 0.636 0.360 1.77 0.07
Need injection 0.093 0.056 1.66 0.09
Ifips_2 -0.407 0.235 -1.73 0.08
cons 7.526 0.150 49.92 0.00

Table A-35. Odds of nursing home of being in the highest operating margin tier
Odds Ratio Std. Err. Z P>z
Cost leader 1.25 0.10 2.79 0.01
Rehab focus 1.46 0.13 4.12 0.00
Specialty care focus 1.04 0.09 0.44 0.66
Differentiator 1.20 0.13 1.71 0.09
Ownership 4.65 0.37 19.15 0.00
Percent Medicaid 1.00 0.00 2.76 0.01
Acuindex 1.04 0.03 1.27 0.21
Total beds 1.00 0.00 1.94 0.05
Market competition 0.92 1.23 -0.06 0.95
Chain Affiliation 1.39 0.09 4.92 0.00
Ifips_2 131.45 180.57 3.55 0.00
Ifips_3 153.83 145.15 5.34 0.00
Note: 53 observations completely determined. Standard errors questionable









Table A-36. Odds of nursing home of being in the highest total margin tier
Odds Ratio Std. Err. z P>z
Cost leader 1.22 0.09 2.58 0.01
Rehab focus 1.27 0.11 2.74 0.01
Specialty care focus 0.98 0.09 -0.25 0.80
Differentiator 1.31 0.14 2.6 0.01
Ownership 1.64 0.12 6.62 0.00
Percent Medicaid 1.06 0.03 1.98 0.05
Acuindex 1.00 0.00 1.29 0.20
Total beds 0.92 1.25 -0.06 0.95
Market competition 1.04 0.07 0.58 0.56
Chain Affiliation 10.53 15.18 1.63 0.10
Ifips_2 18.39 18.84 2.84 0.00
Note: 57 observations completely determined. Standard errors questionable









Table A-37. Correlations of independent variables


Cost RN Rehab Lack of Pct Pct
Leader Focus Focus Differentiate Strategy Ownership Case mix Size Occupancy Medicaid Medicare Competition System


Cost Leader
RN Focus
Rehab Focus
Differentiator
Lack of Strategy
Ownership
Case mix
Size
Occupancy
Pct Medicaid
Pct Medicare
Competition
0 System


-0.282 1.000
-0.301 -0.215 1.000
-0.212 -0.151 -0.161 1.000
-0.373 -0.267 -0.284 -0.200
-0.010 0.006 0.038 -0.041
-0.032 -0.005 0.069 0.033
0.126 -0.017 -0.020 -0.026
0.033 -0.046 -0.041 0.034
0.124 0.000 -0.288 0.041
-0.136 -0.051 0.434 -0.077
-0.010 0.071 -0.025 -0.034
-0.004 0.024 0.059 -0.061


1.000
-0.001 1.000
-0.047 0.062 1.000
-0.079 -0.029 0.082 1.000
0.019 -0.084 0.106 -0.027 1.000
0.100 0.231 0.024 0.160 0.048 1.000
-0.149 0.101 0.098 -0.001 -0.039 -0.477 1.000
-0.004 0.023 -0.095 -0.178 -0.105 0.088 -0.089
-0.026 0.171 0.015 -0.066 -0.046 0.038 0.101


1.000
0.093










LIST OF REFERENCES


Agresti A, Finlay B, Statistical methods for social sciences, third edition, New Jersey 1997

Allen, M.J., & Yen, W. M. (2002). Introduction to Measurement Theory. Long Grove, IL: Waveland Press.

American Physical Therapy Association History of Therapy caps Available at:
http://www.apta.org/AM/Template.cfm?Section=Media&TEMPLATE=/CM/ContentDisplay.cfm&CONTENT
ID=47095 Accessed March 30th, 2008

Amit, R., & Schoemaker, P. f. H. 1993. Strategic assets and organizational rent. Strategic Management
Journal, 14: 33-4B

Ansoff, H. Igor, Corporate Strategy, McGraw-Hill, New York, 1965

Abt Associates (2004), National Nursing Home Quality Measures User's Manual, Available at:
http://www.cms.hhs.gov/NursingHomeQualityInits/downloads/NHQIQMUsersManual.pdf Accessed March
30th, 2008

Barney J B. 1991. Firm resources and sustained competitive advantage. Journal ofManagement, 17: 99-129.

Britton, L. C.; Clark, T. A. R.; Ball, D. F..Modify or Extend? The Application of the Structure Conduct
Performance Approach to Service Industries. By: Service Industries Journal, Jan92, Vol. 12 Issue 1, p34-43,
10p;

Center for Medicare and Medicaid 2007. National Health Expenditures by type of service and source of funds,
CY 1960-2006. Available at:
http://www.cms.hhs.gov/NationalHealthExpendData/02-NationalHealthAccountsHistorical.asp#TopOfPage
Accessed March 30th, 2008

Center for Disease Control. 2007. NCHS faststats Nursing home care. Available at: http://www.cdc.gov/ Cohen
nchs/fastats/nursingh.htm Accessed March 30th, 2007

Center for Medicare and Medicaid 2007 Medicare Long Term Care Available at:
http://www.medicare.gov/longtermcare/static/home.asp Accessed March 30th, 2008

Cool K, Schendel, D STRATEGIC GROUP FORMATION AND PERFORMANCE: THE CASE OF THE
U.S. PHARMACEUTICAL INDUSTRY, 1963-1982. Management Science; Sep87, Vol. 33 Issue 9, p1102-
1124, 23p

Castle NG. Strategic groups and outcomes in nursing facilities. Health Care Manage Rev. 2003 Jul-
Sep;28(3):217-27.

Cool, K., Schendel, D., (1988) Performance differences among strategic group members. Strategic
Management Journal, 9 (3), pp 207-223

Caves, R., Porter, M., (1977), From entry barriers to mobility barriers: conjectural decisions and contrived
deterrence to new competition. Quarterly Journal of Economics p.241-261

Centers for Medicare and Medicaid: RAI MDS Users Manual Available at:










http://www.cms.hhs.gov/NursingHomeQualityInits/downloads/MDS20rai 1202ch3.pdf Accessed March
30th, 2008

Centers for Medicare and Medicaid : State Operations Manual Appendix PP Guidance to Surveyors for Long
Term Care Facilities Available at:
http://www.cms.hhs.gov/manuals/Downloads/soml07ap_pp_guidelinesltcf.pdf Accessed March 30th, 2008

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297-334

Day, G., Strategic Market Planning, West Publishing Co., St. Paul, 1984

Dess, G., Davis, P., (1984), Porter's (1980) Generic Strategies as Determinants of Strategic Group Membership
and Organizational Performance. Academy of Management Journal; Sep84, Vol. 27 Issue 3, p467-488, 22p, 5
charts

Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory
factor analysis in psychological research. Psychological Methods, 4, 272-299.

Feng Z, Grabowski DC, Intrator O, Mor V. The effect of state medicaid case-mix payment on nursing home
resident acuity.Health Serv Res. 2006 Aug;41 (4 Pt 1):1337.

Feng Z, Grabowski DC, Intrator O, Zinn J, Mor V.Medicaid payment rates, case-mix reimbursement, and
nursing home staffing--1996-2004. Med Care. 2008 Jan;46(1):33-40.

Fiegenbaum, A; Thomas, H., INDUSTRY AND STRATEGIC GROUP DYNAMICS: COMPETITIVE
STRATEGY IN THE INSURANCE INDUSTRY, 1970-84. Journal of Management Studies, Jan93, Vol. 30
Issue 1, p69-105, 37p

Fleishman JA, Cohen JW, Manning WG, Kosinski M. Using the SF-12 health status measure to improve
predictions of medical expenditures. Med Care. 2006 May;44(5 Suppl):I54-63.

Ford EW, Duncan JD, Ginter PM., The Structure of State Health Agencies: A Strategic Analysis. MCR&R
60:1 (March 2003)

Fries BE, Schneider DP, Foley WJ, et al. Refining a case-mix measure for nursing homes: Resource Utilization
Groups (RUG-III). Med Care.1994;32:668-685.

Gapenski, L., Understanding Healthcare Financial Management, Third Edition, 2000 Health Administration
Press

GAO Report to Congressional Requesters (July 2001), Nursing Home Quality: Prevalence of Serious
Problems, While Declining Reinforces Importance of Enhanced Oversight,
http://www.gao.gov/new.items/d03561 .pdf

Grabowski DC, Castle NG. Nursing homes with persistent high and low quality. Med Care Res Rev. 2004
Mar;61(1):89-115.

Grabowski DC. A longitudinal study of Medicaid payment, private-pay price and nursing home quality. Int J
Health Care Finance Econ. 2004;4:5-26.










Harrington, C., Zimmerman, D., Karon, S.L. Robinson, J. & Beutel, P.(2000). Nursing home staffing and its
relationship to deficiencies. Journal of Gerontology: Social Sciences, 55B, S278-S287

Harrington, C., Kovner, C., Mezey, M., Kayser-Jones, J., Burger, S., Mohler, M., Burke, S. & Zimmerman, D.
(2000) Experts recommend minimum nurse staffing standards for nursing homes in the United States. The
Gerontologist: 40, 5-16

Harrington C, Woolhandler S, Mullan J, Carrillo H, Himmelstein DU. Does investor ownership of nursing
homes compromise the quality of care? Am J Public Health. 2001 Sep;91(9): 1452-5.

Hill, C.W. Differentiation Versus Low Cost or Differentiation and Low Cost: A Contingency Framework.
Academy of Management Review 13 (1988): 401-12

Hitt M., Bierman L.,Shimizu K., Kochhar L. Direct and Moderating Effects of Human Capital on Strategy and
Performance in Professional Service Firms: A Resource-Based Perspective Michael A. Hitt, Leonard Bierman,
Katsuhiko Shimizu and Rahul Kochhar The Academy of Management Journal, Vol. 44, No. 1 (Feb., 2001), pp.
13-28

Hofer, C. Schendel, D., Strategy Formulation: Analytical Concepts, West Publishin, St. Paul, 1978.

Harrigan, C An Application of Clustering for Strategic Group Analysis. Strategic Management Journal; Jan-
Mar85, Vol. 6 Issue 1, p55-73, 19p

Hulin, C., Roznowski, M., (1985) Organizational Technologies: Effect on organizational characteristics and
individuals responses." Research in Organizational behavior, vol7, p.39-85

Hunt, M.S. "Competition in the Major Home Appliance Industry, 1960-1970." Unpublished Doctoral
dissertation, Harvard, 1972.

Instititute of Medicine, 1986. Improving the quality of care of nursing homes. Washington DC: National
Academy Press

Johnson CE, Dobalian A, Burkhard J, Hedgecock DK, Harman J. Predicting lawsuits against nursing homes in
the United States, 1997-2001. Health Serv Res. 2004 Dec;39(6 Pt 1):1713-31.

Kachigan S.K., Multivariate Statistical analysis: A conceptual introduction, Radius, New York, 1991

Kaiser, H. F. (1960). The application of electronic computers to factor analysis. Educational and Psychological
Measurements, 20, 141-51

Kaplan, and Norton. 2001. Transforming the balanced scorecard from performance measurement to strategic
management: Part I. Accounting Horizons (March): 87-104

Katz, R., Cases and Concepts in Corporate Strategy, Prentice-Hall, Englewood Cliffs, N.J. 1970

Ketchen DJ, Shook CJ. The Application of Cluster Analysis In Strategic Management Research: An Analysis
and Critique. Strategic Management Journal 17 (1996): 441-58

Kling, J A.; Smith, K A. Identifying Strategic Groups in the U.S. Airline Industry: An Application of the Porter
Model. Transportation Journal, Winter95, Vol. 35 Issue 2, p26-34, 9p










Laberge A, Weech-Maldonado R, Johnson CE, Jia H, Dewald L.Outsourcing veterans for long-term care:
comparison of community and state veterans' nursing homes. J Health Hum Serv Adm. 2008 Spring;30(4):441-
67.

Leask, G.; Parker, D. Strategic groups, competitive groups and performance within the U.K. pharmaceutical
industry: Improving our understanding of the competitive process. Strategic Management Journal, Jul2007,
Vol. 28 Issue 7, p723-745

Marlin D, Sun M, Huonker JW. Strategic groups and performance in the nursing home industry: a
reexamination.Med Care Res Rev. 1999 Jun;56(2):156-76

Marlin D, Huonker JW, Sun M. An examination of the relationship between strategic group membership and
hospital performance. Health Care Manage Rev. 2002 Fall;27(4):18-29

Mukamel DB, Glance LG, Li Y, Weimer DL, Spector WD, Zinn JS, Mosqueda L. Does risk adjustment of the
CMS quality measures for nursing homes matter? Med Care. 2008 May;46(5):532-41.

Mehra, Ajay. RESOURCE AND MARKET BASED DETERMINANTS OF PERFORMANCE IN THE U.S.
BANKING INDUSTRY. Strategic Management Journal, Apr96, Vol. 17 Issue 4, p307-322

Murray, A.L. "A Contingency view of porter's generic strategies." Academy of Management Review 13
(1988): 390-400

National Institute of Health, http://www.nlm.nih.gov/medlineplus/nursinghomes.html, accessed March 20th,
2008

O'Neill C, Harrington C, Kitchener M, Saliba D. Quality of care in nursing homes: an analysis of relationships
among profit, quality, and ownership. Med Care. 2003 Dec;41(12): 1318-30.

Pagano E, Bo S, Petrinco M, Rosato R, Merletti F, Gregori D. Factors affecting hospitalization costs in Type 2
diabetic patients. J Diabetes Complications. 2009 Jan-Feb;23(1):1-6. Epub 2008 Apr 16.

Pfeffer, J. 1994. Competitive advantage through people. Boston: Harvard Business School Press

Porter, M.E., "Competitive Strategy, New York: Free press, 1980

Rhoades J, Sommers J., Expenses and Sources of Payment for Nursing Home Residents 1996. Rockville, MD:
Agency for Healthcare Research and Quality; 2000. MEPS Research Findings No 13. AHRQ Pub. No 01-0010

SAS manual 1990, SAS Technical Report A-108

Scanlon, W. 1980. "A theory of the nursing home market." Inquiry 17 (1):25-41.

Schnelle JF. Determining the relationship between staffing and quality. Gerontologist. Feb 2004;44(1): 10-12

Shen YC, Eggleston K, Lau J, Schmid CH.Hospital ownership and financial performance: what explains the
different findings in the empirical literature? Inquiry. 2007 Spring;44(1):41-68

Short JC, Palmer TB, Ketchen DJ Jr. Resource-based and strategic group influences on hospital performance.
Health Care Manage Rev. 2002 Fall;27(4):7-17.










STATA. STATA reference manual release 7. College station: STATA Press; 2001


Swan JH, Harrington C, Clemefia W, Pickard RB, Studer L, deWit SK. Med Care Res Rev. 2000
Sep;57(3):361-78 Medicaid nursing facility reimbursement methods: 1979-1997

The Medicare Payment Advisory Commission 2007 Report to the Congress: Medicare Payment Policy (March
2007) Available at: http://www.MedPAC.gov/documents/Mar07_EntireReport.pdf Accessed March 30th, 2007

The Medicare Payment Advisory Commission 2003 Report to the Congress: Medicare Payment Policy (March
2007) Available at: http://www.MedPAC.gov/documents/Mar03_EntireReport.pdf Accessed March 30th, 2007

Tremblay, V. J., Strategic groups and the demand for beer. Journal of Industrial Economics, Dec85, Vol. 34
Issue 2, p183, 16p

United States General Accounting Office 2000 NURSING HOMES: Aggregate Medicare Payments Are
Adequate Despite Bankruptcies, GAO/T-HEHS-00-192

Venkatraman N., Vasudevan R. Measurement of Business Performance in Strategy Research: A Comparison of
Approaches The Academy of Management Review, Vol. 11, No. 4. (Oct., 1986), pp. 801-814.

Walshe K, Harrington C. Regulation of nursing facilities in the United States: an analysis of resources and
performance of state survey agencies. Gerontologist. 2002 Aug;42(4):475-87.

Weech-Maldonado R, Neff G, Mor V. Does quality of care lead to better financial performance?: the case of
the nursing home industry. Health Care Manage Rev. 2003 Jul-Sep;28(3):201-16.

Weech-Maldonado, R., Meret-Hanke, L., Neff, M., and Mor, V. (2004). Nurse staffing patterns and quality of care
in nursing homes. Health Care Management Review, 29(2), 1-10.

Weech-Maldonado, R., Laberge, A., Pradham, R., Johnson, C., Yang, Z., Hyers, K., Nursing home quality and
financial performance. Working Paper

White H. A. Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for
Heteroskedasticity Econometrica 1980;48 (4):817-38

Wodchis WP, Fries BE, Pollack H. Payer incentives and physical rehabilitation therapy for nonelderly
institutional long-term care residents: evidence from Michigan and Ontario. Arch Phys Med Rehabil. 2004
Feb;85(2):210-7.

Wright P., A Refinement of Porter's Strategies Strategic Management Journal 8 (1987) 93-101

Zinn JS, Aaronson WE, Rosko MD. Strategic groups, performance, and strategic response in the nursing home
industry. Health Serv Res. 1994 Jun;29(2): 187-205.

Zinn J, Feng Z, Mor V, Intrator O, Grabowski D. Restructuring in response to case mix reimbursement in
nursing homes: a contingency approach. Health Care Manage Rev. 2008 Apr-Jun;33(2):113-23.










BIOGRAPHICAL SKETCH

Alexandre Laberge completed his honors bachelors of physical therapy at the University of

Western Ontario in London, Ontario in 1995. He worked as a physical therapist in a variety of settings

including hospitals, skilled nursing facilities, assisted living facilities and in the home health settings.

He completed his Masters of Business Administration in information technology at Goldey Beacom

College in 2002. He completed his Ph.D. in health services research, management and policy in May in

2009. He worked in skilled nursing facilities for over 6 years as a therapist as well as a rehab manager.

In the 4.5 years that he worked as a research assistant, he worked closely with Dr Weech-Maldonado

on a variety of projects that examined the relationship between cost, quality and financial performance

in nursing homes.





PAGE 1

AN EXAMINATION OF STRA TEGIC GROUP MEMBERSHIP AND TECHNOLOGY IN THE NURSING HOME INDUSTRY By ALEXANDRE LABERGE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Alexandre Laberge 2

PAGE 3

This document is dedicated to the four women in my life 3

PAGE 4

ACKNOWLEDGMENTS There are a number of people who provided me with the opportunity to achieve this goal. To begin with, I would like to say thanks my mother who not only supported and encouraged me to learn but also encouraged me to finish what I started. I would like to thank my father who inspired me to think outside of the box. I would like to thank Dr. Weech-Maldonado for his guidance, generosity and tremendous patience. And finally I would to thank my wife, whose unconditional love, support and unde rstanding made it all possible. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................8 LIST OF FIGURES .......................................................................................................................11 ABSTRACT ...................................................................................................................................12 CHAPTER 1 INTRODUCTION................................................................................................................. .14 What Are Nursing Homes and What Are Their Roles in the Health Care System? ...............15 Long Term Care Resident Reimbursement .....................................................................16 Medicaid ..........................................................................................................................16 Private Pay .......................................................................................................................18 Medicare ..........................................................................................................................18 Post Acute Care Reimbursement ............................................................................................19 Nursing Home Environment ...................................................................................................20 The Importance of Focusing on Strategy ................................................................................21 Purpose ...................................................................................................................................22 Research Questions .................................................................................................................22 2 CONCEPTUAL FRAMEWORK...........................................................................................23 Strategic Groups .....................................................................................................................23 Nursing Home Technology .....................................................................................................26 Nursing Home Resources .......................................................................................................28 Scope and Resource Commitment ..........................................................................................31 Scope Using Technology .................................................................................................32 Resource Commitment ....................................................................................................33 Porters Generic Strategies .....................................................................................................34 Strategic Group Model Using Scope (T echnology) Commitment, Resources Commitment and Porters Generic Strategy Theory ...........................................................36 Differentiator ...................................................................................................................37 Cost Leadership ...............................................................................................................37 Best Cost ..........................................................................................................................38 Focus ................................................................................................................................38 Lack of Strategy ..............................................................................................................39 Measuring Performance ..........................................................................................................39 New Contributions ..................................................................................................................41 Hypothesis ..............................................................................................................................41 5

PAGE 6

3 METHODOLOGY.................................................................................................................4 3 Data .........................................................................................................................................43 Population ...............................................................................................................................44 Overview of Methodology ......................................................................................................45 Hypothesis #1: Nursing homes that have a strategy can be categorized into one of four strategic groups: Differentiator, cost leadership, focus or best cost. ..................................45 Scope ...............................................................................................................................45 Nursing Technology ........................................................................................................45 Rehab Technology ...........................................................................................................52 Resource Commitment ....................................................................................................54 Hypothesis #1 Analysis ..........................................................................................................56 Factor Analysis ................................................................................................................57 Factor Matrix ...................................................................................................................58 Number-of-Factors Problem ............................................................................................59 Rotation of the Factor Structure ......................................................................................59 Internal Consistency of Measures ...................................................................................60 Cluster Analysis ...............................................................................................................60 Strategic Groups Defined ................................................................................................63 Determine the total variance value ...........................................................................63 Cluster value .............................................................................................................63 Cluster rank score .....................................................................................................64 Analysis of Variance .......................................................................................................65 Hypothesis #2: Nursing Home s in the Differentiator Strategic Group will have the Highest Quality. ..................................................................................................................66 Dependent Variables: ......................................................................................................66 Independent Variables .....................................................................................................67 Control Variables .............................................................................................................67 Analysis ...........................................................................................................................68 Hypothesis #3: Nursing Homes in the Focus Strategic Group will provide Higher Quality in the area that they are focuse d on compared to other Strategic Groups. .............69 Dependent Variables .......................................................................................................69 Independent Variables .....................................................................................................69 Control Variables .............................................................................................................70 Analysis ...........................................................................................................................70 Hypothesis #4: Nursing homes in the Cost Leadership Strategic Group will have the lowest costs. ........................................................................................................................70 Dependent Variables .......................................................................................................70 Independent Variables .....................................................................................................70 Control Variables .............................................................................................................71 Analysis ...........................................................................................................................71 Hypothesis #5: Differentiation, Co st Leadership, Focus and Best Cost Strategic Groups will have better Financial Performance than Nursing Homes that Lack a Strategy ...........71 Dependent Variables .......................................................................................................71 Independent Variables .....................................................................................................72 Control Variables .............................................................................................................72 Analysis ...........................................................................................................................72 6

PAGE 7

4 RESULTS...................................................................................................................... .........74 Hypothesis #1 .........................................................................................................................74 Hypothesis #2 .........................................................................................................................82 Hypothesis #3 .........................................................................................................................83 Hypothesis #4 .........................................................................................................................84 Hypothesis #5 .........................................................................................................................85 5 DISCUSSION................................................................................................................... ......88 Quality ....................................................................................................................................90 Cost .........................................................................................................................................93 Financial Performance ............................................................................................................94 Managerial Implications .........................................................................................................95 Policy Implications .................................................................................................................96 Conclusion ..............................................................................................................................98 Limitations ..............................................................................................................................98 APPENDIX: BACKGROUND TABLES................................................................................100 LIST OF REFERENCES .............................................................................................................125 BIOGRAPHICAL SKETCH .......................................................................................................130 7

PAGE 8

LIST OF TABLES Table page 4-1 Internal consistency of the composite variables ................................................................76 4-2 Individual variables used in cluster analysis that we re not included in a factor variable. ..............................................................................................................................76 4-3 Rank scores scope and resource variables .........................................................................78 4-4 Rank scores of struct ure and market factors ......................................................................79 4-5 Strategic group quality of care with Differentiator / nursi ng care focus cluster as the reference group ..................................................................................................................82 4-6 Strategic group quality of care with re hab focus cluster as the reference group...............83 4-7 Strategic group quality of care with Speci alty care focus clus ter as the reference group ..................................................................................................................................84 4-8 Strategic groups and costs with cost l eader as reference group regression results ............85 4-9 Strategic groups and predicted costs ..................................................................................85 4-10 Odds Ratio of Nursing Home Strategy ..............................................................................86 4-11 Probability of nursing homes being in the highest operating margin tier ..........................86 4-12 Probability of nursing homes bei ng in the highest total margin tier ..................................87 A-1 Eigenvalues of scope (technology) factor analysis ..........................................................100 A-2 Factor analysis with 10 factor s of scope (technology) variables .....................................102 A-3 Factor analysis with 10 factor s of scope (technology) variables .....................................104 A-4 Eigenvalues of scope (technology) factor analysis ..........................................................106 A-5 Factor analysis of resource staffing intensity with three factors ......................................106 A-6 Ward's minimum variance cluster analysis ......................................................................107 A-7 Cluster History .................................................................................................................107 A-8 Determination of rank scores table ..................................................................................108 A-9 Descriptives of dependent variables, independent variables and control variables .........109 8

PAGE 9

A-10 Negative binomial of quality of care defi ciencies with differentiators as reference group ................................................................................................................................110 A-11 OLS regressions of root transformed bladder decline with differentiators as ref. group ................................................................................................................................110 A-12 OLS regressions of root tr ansformed bowel decline with differentiators as ref. group ...111 A-13 OLS regressions of root transformed ADL decline with differentiators as ref. group ....111 A-14 OLS regressions of log transformed rest raints with differentiators as ref. group ............112 A-15 OLS regressions of root transformed LTC ulcers with di fferentiators as ref. group .......112 A-16 GLM with gamma distribution with PA walking improve with differentiators as ref. group ................................................................................................................................113 A-17 OLS regressions of root tr ansformed PA skin ulcers with differentiators as ref. group ..113 A-18 Negative binomial of quality of care defi ciencies with rehab focus as reference group .114 A-19 OLS regressions of root transformed bladder decline with rehab focus as ref. group .....114 A-20 OLS regressions of root transformed bowel decline with rehab focus as ref. group .......115 A-21 OLS regressions of root transformed ADL decline with rehab focus as ref. group ........115 A-22 OLS regressions of log transformed restraint with rehab focus as ref. group .................116 A-23 OLS regressions of root transformed skin ulcer with rehab focus as ref. group .............116 A-24 GLM of PA walking improve with gamma distribution with rehab focus as ref. group .117 A-25 OLS regressions of root tr ansformed PA skin ulcers with rehab focus as ref. group ......117 A-26 Negative binomial of quality of care defi ciencies with specialty care as reference group ................................................................................................................................118 A-27 OLS regressions of root tr ansformed bladder decline with specialty care focus as ref. group ................................................................................................................................118 A-28 OLS regressions of root transformed bowel decline with specialty care as ref. group ...119 A-29 OLS regressions of root transformed ADL decline with specialty care as ref. group .....119 A-30 OLS regressions of root transformed restraint with sp ecialty care as ref. group .............120 A-31 OLS regressions of root tr ansformed LTC skin ulcer with specialty care as ref. group ..120 9

PAGE 10

A-32 GLM with gamma distribution PA walking improve with specialty care as ref. group ..121 A-33 OLS regressions of root tr ansformed PA skin ulcers with specialty care as ref. group ..121 A-34 GLM regressions of nursing hom e cost with gamma distribution ...................................122 A-35 Odds of nursing home of being in the highest operating margin tier ..............................122 A-36 Odds of nursing home of being in the highest total margin tier ......................................123 A-37 Correlations of independent variables .............................................................................124 10

PAGE 11

LIST OF FIGURES Figure page 2-1 The nursing home process .................................................................................................30 2-2 Strategic Group Model Based on Scope and Resource commitment ................................36 4-1 Strategic group model based on scope and resource commitment in the nursing home industry.....................................................................................................................81 A-1 Scree plot of scope (technology) factor analysis ............................................................101 A-2 Scree plot of resource staffi ng intensity from factor analysis .........................................106 11

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AN EXAMINATION OF STRA TEGIC GROUP MEMBERSHIP AND TECHNOLOGY IN THE NURSING HOME INDUSTRY By Alexandre Laberge May 2009 Chair: Robert Weech-Maldonado PhD Major: Health Services Research, Management and Policy Purpose: The purpose of the study was to examine if strategic groups in the nursing home industry can be determined by examining how faci lities apply their technology and how facilities commit their resources. The study also determined if these groups can be defined using Porters generic strategies and whether the different groups have varying strategic-performance relationships. Methodology : The study began by performing a factor analysis using scope (technology) and resource commitment variables from the Minimum Data Set (MDS) and the Online Survey Certification and Reporting (OSCAR) system. The newly defined variables were used in a cluster analysis to identify the different cluste rs. The clusters were classified using Porters generic strategy model. Once the strategic group s were identified, the study analyzed the strategic-performance relationships using negativ e binomial, ordinary least squares regression (OLS), generalized linear model (GLM) with ga mma distribution and ordered logit analytical tools. Quality dependent variab les included quality of care defi ciencies, long term care (LTC) pressure sore prevalence and post acute pressure ulcer incidence, LTC ac tivities of daily living (ADL) decline, post acute walking improveme nt, LTC bowel and bladder decline and LTC restraint. Financial measures in cluded cost per resident day, oper ating margin and total margin. 12

PAGE 13

13 Results: The study identified four strategic groups (differentiator, cost leader, rehab focus, and specialty care focus) and a lack of strate gy group. Differentiators, rehab focus and specialty care focus groups had less quality of care deficien cies than the lack of strategy group and cost leaders. Differentiators generally had better l ong term care quality outcomes compared to other strategic groups and the lack of strategy group. Cost leaders, diffe rentiators and lack of strategy groups had lower cost than the rehab focus a nd specialty care focus groups. Cost leaders, differentiators and rehab focus group had better financial performance than the specialty care focus group and lack of strategy group. Conclusions: The study indicates that using technology and resour ces commitments is an effective way of identifying strategies groups. The study also found that some strategic groups have better financial performance than other st rategic groups or than nursing homes that have no strategy at all. Nursing homes that provide a hi gh level of technology had better quality, without increased cost or sacrificing fi nancial performance. The study also suggests that nursing homes that service residents with high cost conditions may not be adequately reimbursed.

PAGE 14

CHAPTER 1 INTRODUCTION Nursing homes, like other health care orga nizations face many challenges. They have a limited ability to dictate their pr ice and to control their costs. Nursing homes also face higher competition from the rapid rise in home health ag encies and assisted living facilities. To deal with these challenges, nursing homes must devise and implement effective strategies if they are to remain profitable. These strategies can va ry from focusing on providing quality service, controlling cost or identifying niche market s. Although there are ove r 16,000 nursing homes in the United States, the number of different strategi es to choose from is sm all. It is therefore possible that nursing homes can be categorized into different groups that have similar strategies. Nursing homes are an important component of the health care industry consisting of 8% of the United States health care expenditures (Cen ters for Medicare and Medicaid 2007). The nursing home industry provides care for 1.6 million of the frailest people in the United States (Center for Disease Control and Prevention 2007). A study by the U.S. Department of Health and Human Services says that people who reach the age of 65 have a 40 percent chance of entering a nursing home. Of those who enter a nursing home, 10 percent of them will remain there five years or more (Centers for Medicare and Medicaid 2007). Expenditures for nursing home care skyrockete d in the years following the implementation of the hospital prospective payment system of 19 83. The industry grew from 50 billion dollars in 1990 to 90 billion dollars in 1997 (Centers for Medicare and Medicaid 2007). However, this changed after the implementation of the Bala nced Budget Act of 1997 (BBA 1997) where the industry only grew 10% between the years 1999 and 2004. The BBA cuts in reimbursement were so severe that a number of nursing homes went bankrupt including some of the nations largest nursing home chains (GAO 2000). Since the implem entation of the BBA, the remaining nursing 14

PAGE 15

homes have been faced with considerable challenges that include increased government regulation (GAO 2001, Walshe et al. 2002), incr eased litigation (Johnson et al. 2003) and increased competition (Weech-Maldonado et al. 2003). In order to survive, nursing home managers ne ed to identify effective strategies so that they can continue to provide a very valuable service to a population with few options. Some of those strategies include providing high quality care, keeping costs as low as possible and focusing on a specific market segment of the indu stry. These different types of strategy do not only relate to better financ ial performance, but can also influence quality. To fully understand these different types of stra tegies, it is necessary to have a clear grasp of what nursing homes are, to have insight on what th eir role is in the health care industry and to comprehend how they function in an uncertain environment. What Are Nursing Homes and What Are Their Roles in the Health Care System? A nursing home is a place for people who do not need to be in a hospital but cannot be cared for at home. Most nursing homes have nursing aides and skilled nurses on hand 24 hours a day (National Institutes of Health 2008). The stay in a nursing home can be permanent or can be temporary. Nursing homes have two major functions. The firs t function is to provide long term care to those individuals who require more care than can be provided by their families or can be provided by any other type of placement. The desired outcome for the long term care resident is that during their stay, their health status remains at least at the same level as it was on admission. It is understood, however, that the longer the resident remains in a facility, the more likely they will experience a decline in their health status The decline occurs as a result of the normal progression of the pre existing medical conditions that typically precipitated the admission to the 15

PAGE 16

facility in the first place. The desired outcome for long term care residents is to maintain their health status for as long as possible. The second major function of a nursing home is to provide short term care (also known as post acute or sub acute) for those residents that require skilled se rvices after a qualified hospital stay. To qualify for a post acute st ay, residents must have either a skilled nursing care need (e.g. IV antibiotics) or a rehab care need based on th e recommendations of the hospital therapists. The goal of post acute care is to improve the residents health stat us so that the resident can be discharged home or to the most appropriate leve l of care. The discharge options are quite broad and can range from patients going home by themse lves to becoming long term residents in a nursing home. Discharge from a skilled nursing reha bilitation facility (als o known as a post acute or subacute facility) can occur wh en the patient reaches their highest level of function or the stay is no longer covered by the payer. The desired out come for a rehabilitation resident is a better health status. Long Term Care Resident Reimbursement Reimbursement for long term care is primarily provided by Medicaid or by the residents themselves (private pay). A small proportion of residents are covered by other payers like the Department of Veteran Affairs (VA) and private insurance. A lthough Medicare does not pay for a long term care residents stay in a nursing home, Medicare does cover some ancillary services like physical or occupational therapy. Medicaid Medicaid is the major payer of nursing home long-term care services. Over 50% of nursing homes revenues and 70% of nursing home beds are covered by Medicaid (Rhoades et al. 2000). Swan et al. 2000 describe five rate setting methodologies that state Medicaid programs use to 16

PAGE 17

pay nursing homes: retrospective, prospective clas s, prospective facility-s pecific, adjusted, and combination. Retrospective payments freely adjust rates to current costs. In contrast, prospective payment rates are not fully adjusted to cost duri ng the costs year. Rates are usually set based on prior year costs. Prospective cl ass rates are the same for all nursing homes in the state while prospective facility-specific are set based on the costs for each f acility. Adjusted reimbursement is similar to retrospective reimbursement however adjusted systems allow interim rates to increase during a year but not to fully reflect costs. Finally, there are also combination reimbursement methods which are a mixture of retrospective and prospective payment systems (Swan et al. 2000). As of 1997, only one state used retrospective reimbursement, 25 used prospective, 21 used adjusted and 3 us ed combinations (Feng et al. 2006). Since the late 1990s, most state Medicaid prog rams have used a prospective per diem system to reimburse nursing homes for all the cost that is incurred in providing the care to the resident in the facility (Feng et al. 2008). With this reimbursement methodology, a nursing home that provides a lot of nursing services will not ne cessarily be compensated more than a facility that provides fewer nursing services. An increasing number of states have m odified their reimbursement methodology by making case mix adjustments to their per diem ra te. This allows nursing homes to collect higher rates for residents that have higher acuity and that are likely to cost the facility more to care for them. The number of states that use case mix reimbursement increased from 19 in 1991 to 35 in 2004 (Feng et al. 2006). The best-known and most widely used case mix methodology is the Resource Utilization Groups system (RUGs), currently in its third version (Fries et al. 1994). This system classifies 17

PAGE 18

residents into homogeneous categ ories based on their estimated re source utilization. Associated with each of these categories is a case-mix index, which represents, at least relatively, the time, or cost, of caring for the average resident in the group. A higher case-mix index indicates a greater degree of complexity and, consequently, a greater need for input resources. Under this system, nursing homes with a higher case-mix in dex, on average, would be reimbursed a higher rate (Zinn et al. 2008). Private Pay Private paying residents reimburse approxima tely 18% of the beds (GAO 2000). Private pay reimbursement is commonly based on a per diem rate, which is on average 1.4 times the amount that Medicaid reimburses (Grabowski 2004). Medicaid only pays for those residents w ho do not have enough funds to pay for the nursing home stay themselves. Residents who have some assets will pay the private pay rates until their personal funds are exhausted. Once the resident has spent down their assets, they usually qualify for Medicaid. Medicare There are some processes of care provided to th e long term care residents that are paid for by Medicare (Wodchis et al. 2004). Physical, occupational and speech therapy are three services that can be provided to long term care residents by nursing homes. These services are reimbursed over and above the Medicaid per diem rate or the private pay rate. Therapy services are paid for under Medicare part B and is cons idered an outpatient service. Although the price is set by the Center for Medicaid and Medica re (CMS) however the amount of therapy services provided was not limited until 2005. In 2005, the CMS implemented a cap on therapy services that limited residents to a maximum annual benefit of 1700 dolla rs of physical and speech therapy combined and 1700 dollar cap on occupational thera py. Initially passed by congress in 1997, 18

PAGE 19

implementation of the caps was delayed by a num ber of moratoriums. Even in 2005, after the caps were implemented, nursing ho mes were able to circumvent the caps if a resident had therapy needs that were deemed medical necessary (APTA 2008). Post Acute Care Reimbursement Reimbursement for post-acute care is prim arily reimbursed for by Medicare; however it can also be covered by private insurance and th e VA. Approximately 12% of the beds in nursing homes are occupied by rehabilitation patients paid for by Medicare (GAO 2000). These patients usually have experienced a declin e in their health status and re quire rehabilitation before they can return either home or to the next appropriate level of care. Medicare reimbursement uses a prospective pa yment system. Similar to Medicaid case mix reimbursement, residents are classified usi ng the Resource Utilizati on Groups (RUG). These groups are commonly determined by the amount of th erapy and nursing services that the resident requires. For example, a facility will receive a higher daily rate if the resident gets more therapy. Each rehab RUG level has a minimum amount of therapy minutes that have to be provided within 7 days prior to a given assessment refe rence date. For example, for a patient to be classified as an ultra high reha b RUG level, they would have to receive 720 minutes of therapy during the 7 day assessment period. Patients ar e assessed a maximum of 5 times depending on the amount of time they remain at the facility for rehab. The assessments include the 5-day, 14 day, 30 day, 60 day and the 90 day assessment. The 5 day assessment identifies the RUG level for day 1 to 14. The 14 day assessment identifies the RUG level for day 15-30. Each subsequent assessment identifies the RUG level for the fo llowing period. Although the each resident is eligible for up to 100 days of rehab per medical incident, most patients are discharged from subacute services well before the 100 day lim it because they go home or are no longer making 19

PAGE 20

progress. The difference in reimbursement rate between the RUG levels is significant (MDS 2007). Higher rehab RUG reimbursement levels are a ssociated with higher therapy utilization (and higher reimbursement rates) but are not necessarily associat ed with higher acuity (MedPAC 2007). For example, a post acute patient with a kn ee replacement that only requires supervision with his ADLs may be classified at an ultra hi gh RUG level because the patient can tolerate high volume of therapy, not because the patient has higher acuity. In contrast, a post acute patient that is dependent and has low tolerance to activity will only classify at a high or medium RUG level. Based on acuity scores, the second patient would be considered more acute but the facility would be compensated less. Therefore, nursing homes th at have a higher proportion of residents in the highest RUG level will have highe r revenue but they will not nece ssarily have higher comparable costs. MedPAC (2007) found that nursing homes increased the proportion of their residents receiving the two highest paying RUG levels (ultra and very) by 50% between the years 2000 and 2005. Nursing Home Environment Similar to most health care organizations, nursing homes exist in a turbulent, complex and uncertain environment. Since the Balanced Budget Act 1997 (BBA), the nursing home industry has been dealing with a number of cost c ontrol mechanisms imposed by federal and state government payers. This problem affected th e whole industry since government payers like Medicare, Medicaid and the VA account for over 62% of the nursing home industries revenues (MedPAC 2003). Public payers ofte n dictate price of a residents per diem stay to the nursing home. By removing price from the equation, nursi ng homes must find other ways to compete for residents that provide th e nursing home with the highest revenue and that have the lowest cost. 20

PAGE 21

Nursing homes also face increased competition from growing industries like assisted living facilities and home health agenci es that tend to draw the less dependent (and more profitable) private pay residents. This ultimately leaves nursing homes with the re sidents who have higher medical acuity, greater functional dependence and have higher care costs (Weech-Maldonado et al. 2003). With the limited ability to control their reve nue and with the increased competition created by substitutes, nursing homes find themselves in a precarious position. For nursing homes to thrive (or even just survive), it is necessary fo r them to identify strategies that allow them to achieve competitive advantage. The Importance of Focusing on Strategy Government payers and interest groups ar e interested in persuading nursing homes to provide high quality of care. One way to encourage facilities to provide better quality of care is to demonstrate that better quality results in be tter financial performance. Some recent studies have determined a positive re lationship between financial pe rformance and quality of care (Weech-Maldonado et al. 2003, 2008). Even though nursing homes may have the incenti ve to provide better quality, it is also necessary for the nursing home to determine the best strategies to be able to provide high quality care. Without incentives and a strategy, a nursing home will have some difficulty achieving these desired outcomes. Managers are interested in achieving above normal profits. Stra tegies are plans related to the mission and the vision of an organization that provide leadership with the means to achieve high performance. Like any industry, nursing homes attempt to gain competitive advantage by selecting and implementing an effective strategy that the competitors ar e not able to reproduce (Barney 1991). Because the number of truly distinct strategies available in any industry is likely 21

PAGE 22

22 to be small, managers decide not so much how to be unique, but rather which group of competitors their strategies should be similar to (Marlin et al. 2002). The strategic group model is one approach that can help managers and researchers identify the best strategy. Purpose The purpose of the study is to examine if strategic group struct ure of the nursing home industry can be determined by how facilities focus their technology a nd how facilities commit their resources. The study will also determine if these groups can be defined using Porters generic strategies and will examine whether th e different groups have different strategicperformance relationships. Research Questions What strategic groups exist in the nursing home industry? How does nursing home strategic group memb ership affect quality of care? How does nursing home strategic group membership affect costs? How does nursing home strategic group member ship affect financial performance?

PAGE 23

CHAPTER 2 CONCEPTUAL FRAMEWORK This study will begin by defining strategic grou p theory, Porters generic strategy theory, technology and Cool and Schendels scope and re source commitment theory. The study will then explain how these theories are integrated to genera te a model that uses Porters generic strategies to define strategic groups in the nursing home. Finally, the study will define how the different strategic groups can have differe nt strategic-performance relations hips as it pertains to quality, cost and financial performance. Strategic Groups The concept of strategic groups was introduced by Hunt (1972) in his thesis that examined the appliance industry. He discovered that there was less competitive rivalry than what industry concentration ratio s suggested. He attributed this to the existence of subgroups within the industry that effectively reduce the number of competitors in each market. Caves and Porter (1977) expanded the theory by defini ng strategic groups as a set of fi rms that face similar threats and opportunities that are different from the threats and opportunities faced by other firms in the industry. Strategic group theory stipulates that within an industry, there could be groupings of organizations that have very different stra tegies and yet may still have good financial performance. This theory was a shift in thinki ng from the industrial organizational theory where it was generally believed that there was only one right way of doing things (Leask 2007). Analyzing strategic groups gives insight to different competitors approaches to the marketplace (Harrigan 1985). The number of strategic group s within an industry corresponds to the number of unique strategies within that industry (Marlin et al. 1999). St rategic groups are persistent strategic characteristics of an industry, which are protected by mobility barriers (Porter 1980). 23

PAGE 24

Movement between groups is limited by mobility barriers, which are associated with the cost of moving from one strategic group to another (Porter 1980). Conceptually, mobility barriers behave similar to barrier s to entry in Porters five fo rces analysis framework (Barney 1991). This framework describes entr y of new organizations to the ma rket place as threats to the profitability of the incumbent organizations. By establishing barriers to entry, an organization can block out new competitors from entering the market, thereby maintaining their above normal margins (Caves and Porter 1977). The existence of mobility barriers can a llow two groups within the same industry to exist without directly co mpeting with each other (Leask 2007). In the nursing home industry, facilities that are a part of the strategic gr oups concerned with efficiency and keeping operating cost as low as possible are different from those nursing homes that attempt to differentiate themselves by provi ding better care (Marlin 1999). Although these strategies are very different, it is possible that nursing homes in either one of these strategic group types will have good financial performance. For an organization to move from one strategic group to anot her, it would be necessary for them to change their business strategy and put th emselves at risk of not recouping the resources they invest (Porter 1980). These costs offer pr otection to group member s by discouraging entry of rivals into the group (Porter 1980). An ex ample would be an organization from a cost leadership strategic group trying to move to a differentiating strategic group. This organization would have to make some considerable investment and they would also have to change the organizational culture from being cost focu s to providing more and better services. The cost of movement between strategic groups can vary. For a facility to move from cost leadership to focus may require significantly le ss resources than moving from cost leadership to differentiator. Therefore the highe r the cost to change strategi c groups, the higher the mobility 24

PAGE 25

barriers. Higher barriers offer higher insularity to the member s of the groups, thus limiting competition and group membership (Marlin et al. 2002). The height of the mobility barrier is expected to be directly rela ted to the performance of th e strategic group (Porter 1980). Since the introduction of the strategic group concept by Hunt (1972) and the additional contributions by Caves and Porter (1978), there have been a significant number of empirical studies that have examined stra tegic groups and their impact on firm performance. Studies that have examined the beer (Tremblay 1985), banki ng (Mehra 1996), air line industry (Kling et al.1988), and the pharmaceutical industry (Figen baum et al. 1992, Leask 2007) have all found the existence of strategic groups with strategicperformance relationships. Direct care health care industries have also been examined using strate gic group theory in the hospital (Marlin et al. 2002, Ketchen et al. 2003) and the nursing hom e industry (Castle 2003, Ma rlin et al. 1999, Zinn et al. 1994). The three studies that have examined strategic groups in the nursing home industry all have found significant difference between the groups and have also found some strategyperformance relationships. Castle 2003 used a surv ey that asked the leadership of the nursing homes to self characterize themselves using Mi les and Snows categorie s (1990): prospectors, analyzers, reactors or defenders. He found that the Miles and Snows prospectors had higher levels of financial performance and higher qua lity. Zinn et al. 1994 a nd Marlin et al. 1999 attempted to determine if th ere was a relationship between strategic group membership and nursing home performance. Both studies used scope and resource commitment as defined by Cool and Shendel (1988) to determine their clusters and found the existence of succinct strategic groups in the nursing home industry. 25

PAGE 26

Nursing Home Technology Hulin and Roznowski (1985) defined technology as the physical and knowledge processes by which materials in some forms are transfor med into outputs. In the nursing home industry, technology generally refers to the care processes provided by the staff to the residents. In order to provide these processes, an organization requires sufficient resources. For example, in the nursing home industry, a facility that fails to have adequate staffi ng levels (the primary resource in the nursing home industry) is not going to be able to adequately provide the care processes (i.e. technology) that are necessary to have good patient care outcomes. The Centers for Medicare and Medicaid (CMS) statute 483.25 dictate that each resident must receive from the facility the necessary care and services to attain or maintain the highest practicable physical, mental, and psychosocial we ll-being, in accordance with the comprehensive assessment and plan of care (CMS 2008). Nursing homes use a number of different techno logies that allows them to address the multiple needs of their residents. Each of th ese technologies addresses a specific care need. There are two broad categories of nursing hom e technology: nursing technology and rehab technology Nursing technology: Nursing technology includes all of the care processes that are provided by the nursing staff. These processes of care are required so that the resident can maintain or improve their health status. Nursing provides routine care such as bathing, providing medication and repositioning the residents. Nursing technology also includes less routine services like skin ulcer care or catheter placement. Most routine services require that the nursing home establish a program in order for the proce sses to be provided properly. For example, even though a routine technology like rest orative ambulation may only requ ire one nursing assistant to walk with the resident, a restorative program must be established. The program requires that 26

PAGE 27

there is an MD order, that the goals are generated by the physical th erapist and the program requires some oversight to ensure that the service is provided. The responsibility of the organizational features of a restorativ e program is placed on the licensed nurse. Failure to provide these processes of care can result in a decline in the residents health status. One example of nursing technology is pr essure sore prevention. If pressure sore preventative processes are not provided, a resident will be more likely experience a decline in their health status because they can develop a pressure sore. Having a patient on a turning schedule every 2 hours or providing them with a pressure relieving device reduces the likelihood of a pressure sore will occur. Turning schedules relieve pressure and allow blood to flow to the skin. Pressure relieving devices distribute the pressure over a great er surface area and lower the pressure on the high-risk br eakdown areas like the coccyx (C MS 2008). A nursing home that puts emphasis on nursing technology may be tr ying to get better nursing outcomes. More processes of care and better outcomes are indicato rs of better quality (Sch nell et al. 2004). Most of nursing home technology is provided by the nu rses. A facility focused on better quality may be trying to differentiate themselves from their competitors. By differentiating themselves from their competitors, a facility may get a greater pr oportion of private pay residents and/or higher profit margin Medicare residents. In addition, be tter outcomes can also lead to lower costs (Weech-Maldonado et al. 2003). Rehabilitation technology: Rehabilitation technology can be provided to either post acute or long term care residents. These processes of care consist primar ily of therapy services and are geared to improve the residents health stat us. Rehabilitation technology is not considered routine because of its dynamic nature. Therapy serv ices vary based on the needs of the residents. Some residents may have difficulty with transfer ring from the chair to the toilet while others 27

PAGE 28

have difficulty walking. Both these issues would be identified and addressed by physical therapy through an assessment and then would be treated with a specifically program tailored to the individual residents needs. Even the treatment protocol can change over time. Initially the resident may require treatment to increase their st rength and balance so they are able to stand. Once the patient is able to stand, physical th erapy will then focus on balance and walking. Therapy assessment can only be performed by a licensed therapist while therapy treatments can only be performed by either a licensed therapist or licensed therapy assi stant. Nursing homes have an incentive to use rehab technology ove r nursing technology since rehab technology provides better revenue streams. MedPAC 2007 has warned that the current Medicare reimbursement system favors over utilizati on of therapy. Although it may be fiscally advantageous to provide rehabili tation, it only serves as a few of the processes of care that a resident receives on any given da y. The therapist actually only pl ays a small role in the overall care of residents when compared to role of the nurses. Nursing Home Resources According to the resource based view of the firm, performance differences across firms can be attributed to the variance in the firms res ources and capabilities (Hitt et al. 1990). Resources that are valuable, unique and difficult to imitate can provide the basis for firms competitive advantages (Amit & Schoemaker, 1993; Barney 1991). Firms employ both tangible resources (such as buildings and financia l resources) and intangible res ources (like knowledge and brand equity) in the development and implementation of strategies. Human capital has long been argued as a critical resource in most firms (Pfeffer, 1994). In the nursing home setting, the tangible resources like human cap ital and intangible resources like staff expertise are especially important because most of the nursing homes expe nses are the nurses and therapist salaries. In addition, a facilities technology is dependent of the amount and quality of their health care staff. 28

PAGE 29

Human capital provides the physical care and po ssesses the knowledge of the facility. Knowledge is one of the most critical resour ces that a firm possesses (Hitt et al. 1991). The nursing home requires adequate resources if they are going to be able to provide their technology effectively. The correlation between staffing levels and technology provided is expected to be high. The amount of nursing staff and the skill of the nursing staff will influence the quality outcomes (Harrington et al. 2000, We ech-Maldonado et al. 2003) and will likely influence the quality of the techno logy that the facility provides. Nursing resource : The total nursing staff levels of a nursing home is measured by adding all the registered nurses (RN) per resident, the licensed practical nurses (LPN) per resident and the certified nursing assi stant (CNA) per resident. Although th e duties for each of these staffing types are well defined and are in tended to be separate, there remains is a certain degree of overlap. As a result some facilities may chose to substitute higher skilled staff for lower skilled staff (E.g. CNA for RN or LPN for RN). Using lower skilled staff a llows facilities to keep costs down. However, nursing homes must ha ve an appropriate amount skilled staff in order to provide quality of care. Facilities who substitute staffi ng may lose their savings to the cost of poor outcome. For example, preventing a pressure ulcer from occurring throug h good care practices is going to cost less then trea ting a newly acquired ulcer. Rehab resource: The total rehab staffing level of a nursing home is measured by adding all the physical therapists (PT) physical therapy assistant (PTA ), occupational therapist (OT), certified occupational therapy assistant (COTA) pe r resident and speech pathologist per resident. The duties for each of these staffing types for each discipline are less well defined than nursing because therapy assistants can provide almost all the treatments that a therapist can provide. As a result facilities may chose to substitute mo re skilled staff for lesser skilled staff. 29

PAGE 30

Figure #1 illustrates the relationship between the technology provided by the facility and the resources needed to provide that tec hnology. A nursing homes pr imary function is to provide care to all the facilitys residents. The required inputs not onl y include th e facilities tangible and the intangible resour ces but also include the reside nts health status on admission. The technology includes all the nursing homes care pr ocesses that use the facilities tangible and intangible resources. Staffing le vels are the tangible resources that provide the physical care. The intangible resources include all the knowledge that the nurs ing home staff possesses and is measured by the amount of the facilitys st aff is licensed (Weech-Maldonado et al. 2003). Similar to inputs, the outputs also use the measure of the health status of the residents. However, the measure only occurs after the nursing home care processes have been applied. Figure 2-1. The nursing home process One factor that is unique for long term care is that because the resident is not likely to leave the facility, the desired output is the sust ained health and well being of the resident. The measure of quality is how well th e nursing home maintains (or improves) a residents health and well being during their stay in the nursing home. This is not equally true for the post acute 30

PAGE 31

resident since Medicare only pays for patients stay as long as the patients health status improves. Therefore the desired ou tput for the post acute patient is a better health status in hope that the resident can be discharged home. Scope and Resource Commitment Early writings postulated that business level strategy, the level relevant to strategic groups, consists minimally of two sets of activities: (1) business scope commitments and (2) resource commitments (Ansoff 1965; Katz 1970; Hofer and Schendel 1978; Day 1984; Cool and Schendel 1988). Scope commitments are those decisions that lead to the selection of market segments to compete and to the types of services to offer (Cool and Schendel 1988, Zinn et al. 1994). While applicable across diverse industries, the variab les chosen to operationalize these strategic dimensions should be industry specific, reflectin g the bases for competition in the industry under study (Cool and Schendel 1988). Resource commitmen t includes business level deployment of resources to those functional ar eas that are necessary to achieve and maintain competitive advantage (Cool and Schendel 1988). There have been some nursing home studies that have utilized scope and resource commitment as the parameters to identify whet her strategic groups existed in the nursing home industry. Zinn et al. 1994 used scope commitme nt measures of percent Medicare, percent Medicaid, percent independent capacity, case mix, average lengt h of stay and percent over 85. Zinn et al. 1994 used resource commitment measur es of size, private pay rate, and occupancy, RN per resident and staff per resi dent. Using cluster analysis, they found that facilities with the highest private pay residents had the best patient outcomes and the facilities with the highest percent Medicare beds had better financial performance. Marlin et al. 1999 used Medicaid utilization, Medicare u tilization, insurance/health maintena nce organization (H MO) utilization, 31

PAGE 32

private-pay utilization, and Vete rans Administration utilization, average length of stay, average patient age and case mix. For resource commitme nt, Marlin et al. 1999 used percentage of nursing costs, percentage of ancillary costs, occupancy rate, semiprivate room rate, number of beds, registered nurses (RNs) per resident, and st aff per resident. The study also used cluster analysis and found that the group with the highest private pay utilization combined with high Medicare utilization generally performed better along financial and quality indicators. Scope Using Technology Prior studies of strategic groups in nursing hom es have used the range of market segments as the measure of scope commitment (Zinn et al. 1994, Marlin et al. 1999 ). Zinn et al. 1994 and Marlin et al. 1999 argued that to compete effectively, nursing ho mes need to be responsive to differences in payer demand characteristics becau se the payer mix of a nursing home is the major basis of segmentation in the indus try (Zinn et al. 1994, Ma rlin et al. 1999). However, using payer demand characteristics may not be the best way of determining the strategic groups because the payer mix of a nursing home may be more the result of good strategy. For example, many facilities may have a strategy to acquire Medicare and private pay residents but not be successful. Whether a facility has a high pr oportion Medicare and private pay beds may be independent of the nursing homes attempt to acqui re these types of residents. Structure Conduct Performance theory is a m odel used to link elements of the market structure to business conduct and performance in i ndustrial economics. Struct ure refers to market structure defined mainly by the concentration of market share in the market. Conduct refers to the behavior of firms whether co mpetitive or collusive (pricing and production, production, goals of firms, promotion) (Britton et al. 1992). Perfor mance is mainly defined by the consequences of market power. Based on the Structure C onduct Performance paradigm, payer demand characteristics would be considered performance measures rather than conduct measures. In the 32

PAGE 33

nursing home industry, measures of this type of performance include percent of beds filled by Medicare patients and percent of beds filled by private pay residents. Both Medicare and private pay clientele pay higher reimburseme nt rates than other payers lik e Medicaid. Market shares of Medicare and/or private pay residents are the result of a stra tegy but do not give any information about what strategy was used by the nursi ng home to accomplish that performance. Unlike the aforementioned performance m easures, technology focus and resource deployment decisions are made by the business level management and are under the direct control of management. Decisions to increase or decrease the amount of nursing technologies provided (services like restorative care or skin ulcer prevention) can be made independent of the environment that the facility is located in. However the resulting quality of care caused by providing these services can influe nce the performance of the facility as measured by the ability of the facility to draw higher reimburs ing Medicare or private pay residents. In the nursing home industry, technology incl udes all the processes of care that are provided to the residents by the facilitys health care staff. Nursing homes use their technology to transform their inputs into output s. The nursing home industry is different from other industries because of the way technology is applied and the way the product is measured. Resource Commitment Resource commitment refers to the commitment of resources to functional areas that are needed to gain and maintain competitive advantage in targeted market segments. For nursing home administrators, resource commitment deci sions should be reflec ted by labor, price and capacity decisions (Zinn et al. 1994 ). Nursing homes have some c ontrol over their labor but they have limited control over their products price. Government reimbursement accounts for over 62% of their revenue (MedPAC 200 7). After the implementation of the Balanced Budget Act of 1997, Medicare dictated the price that they are going to pay for the nursing home stay. Similarly 33

PAGE 34

most states have changed their Medicaid reim bursement schematic in the direction of a prospective payment system. Only about 15% of nursing home beds are private pay where a facility can set their price. Capacity decisions can be impacted by economies of scale where the decision to be larger can allow the nursing home to benefit from the efficiencies created by more resources derived from size. Weech-Maldonado et al 2008 found that larger fa cilities have better financial performance. Industrial organization st udies have found that firms following similar strategies are likely to be of comparable scale (Porter 1980). In addition, labor and capacity can also be related. With widespread nursing and therapist s hortages, some facilities may not be able to increase the capacity due to labor limitations. Porters Generic Strategies Besides strategic group models, this study will examine the strategic-performance relationship using a conceptual sc heme that indicate the strategies that are available to firms. Porter (1980) introduced the concept that there are certain generic t ypes of strategies that can be used by firms to outperform other organizations in the industry. Po rters generic strategies are perhaps the most widely used and heavily researched. Firms have different types of environmentally determined functional demands and t hus choose different strategies in an effort to gain competitive advantage (Porter 1980). Originally Porter (1980) argued that a firm could choose one of three strategies to compete at the business level. One of those strategies is differe ntiation where a nursing home differentiates itself from other nur sing homes in the market. One way facilities can differentiate themselves from their competitors is providi ng a high quality product. In the nursing home industry this equates to better care. By providing a high quality product, businesses are able to shield themselves from reductions in price-co st margins (Porter, 1980). Weech-Maldonado et al. 34

PAGE 35

(2003) study found that there was a positive rela tionship between quality of care and financial performance in nursing homes. A nursing home can also use cost leadership in which the nursing home attempts to have relative lower costs. One would expect that nursing homes that follow the cost leadership paradigm would be more effici ent at using their staff. The third strategy is a focus strategy, in which the firm concentrates on a particular group of customers, geographic markets, or product lin e segments. In the case of the nursing home industry, some facilities may choose to focus on re habilitation because of the higher revenues per bed that Medicare pays. Porter (1980) considered firms that attempte d to apply both a differe ntiating strategy and a cost leader strategy at the same time as stuck in the middle or muddlers. Stuck in the middle or muddlers are considered those organiza tion who have no coherent strategy. These organizations did not satisfy the demands of either cost leaders or differe ntiators (Marlin et al. 2002). In contrast, Hill et al. 1988 and Wright et al. 1987 have pos ited that some nursing homes can be successful at differentiating themselves from their competitors while controlling their costs at the same time. Marlin et al. 2002 coined this type of nursing home as best costs and found that the best cost strategy existed in the hospita l industry. Porter has acknowledged that on rare occasions, firms can be successful with bot h differentiating strategy and a cost leader strategy (Murray et al.1988). Marlin et al. (2002) used strategic groups as defined by Porters theory of competitive strategy in the hospital industry. Ma rlin et al. 2002 used an object ive classification procedure to classify four strategic groups : differentiation, low cost, best costs and the muddlers. The variables for the cluster analysis were selected by a panel of experts. Thos e variables selected to 35

PAGE 36

capture cost leadership were measures that capture different types of cost like salaries and labor hours. The variables used to captu re the differentiation group were t ypes of services. Marlin et al. (2002) found that hospital strategi c groups could be defined using Porters generic strategies theory and that different strategic groups had different structure-performance relationships. Strategic Group Model Using Scope (Technology) Commitment, Resources Commitment and Porters Generic Strategy Theory Figure #2 is a strategic group model of the nursing homes industry. The horizontal axis represents the scope commitment of the nursing home as measured by the amount of technology provided by the facility. Figure 2-2. Strategic Group Model Based on Scope and Resource commitment The vertical axis represents the resources that the facilities commit to provide its technology. The circles represent clusters as defined by Porters stra tegies (Differentiator, Cost Leadership, Best Cost and Focus). Those faciliti es that do not have a strategy are defined as being groups that lack strategy 36

PAGE 37

The center of the circle depicts the centroid of the cluster. To be in the cluster, a nursing home is closer to the centroid of its cluster than the centroid of any other cluster in the industry. All nursing homes within the groups are located within the outer boundaries of each cluster. A nursing home is not able to be in two separate clusters. The upper scope boundary is a vertical line th at passes through the cen troid of the cluster that has the highest scope (technology) value. The lower resour ce boundary is a horizontal line that passes through the centroid th at has the lowest resources. Ther e are nursing homes that fall beyond both these boundaries; however the centroids of the clusters ca nnot fall beyond these boundaries. Differentiator For a nursing home to differentiate itself from its competitors, it must provide superior processes of care. By taking on the strategy of providing more service than the competitors in their local markets, these nursing homes are more likely to attract residents to their facility resulting in higher occupancy ra tes and choose residents that can provide them with higher margins. A differentiating nursing home would pr ovide a high volume of service relative to the average nursing home in the market. In Figure #2, the differentiating strategic group would be expected to be in the upper right quadrant of the graph. This is because the differentiating group is defined as the highest provider of servic e (technology). If a high volume of service is provided, it would be expected that facilities in the differentia tion strategic groups would also have to commit a greater amount of resources. The centroid of the cluster demarks the upper scope boundary. Cost Leadership Some nursing homes will be focused on keepi ng their costs low. Since staffing is the significant cost driver in the nursing home industry, k eeping lower staffing levels is one obvious 37

PAGE 38

method that facilities can use to control cost. RN s cost significantly more than CNAs and LPNs and therefore by manipulating staffing levels to a lower skill mix, cost le adership nursing homes are expected to keep their costs low. In Figure #2, the cost leadership group would be expected to be loca ted in the bottom left quadrant of the graph. This is because the cost le adership group has the lowest use of resources. It is expected that because of the lower availabi lity of resources, cost le adership strategic groups will provide less processes of care. The centroid of the cluster demarks the lower resource boundary. Best Cost Some nursing homes may seem like a hybrid of the cost leadership group and the differentiation group relative to the lack of strategy group. The best cost group represents efficient organizations. In Figure #2, the best cost group would be expected to have its centroid between the cost leaders and th e differentiators. More specifi cally, the best cost group is expected to provide fewer services than the differentiating group but use fewer resources. It would also be expected to provide more servic es than the cost leadership group using more resources. Focus Like any industry, it is possible for a nursing home to focus their strategy on a smaller segment of the industry. By allocating resources to that area, it may be possible to perform well financially. In Figure #2, the centr oid of the focus group could be located anywhere in the graph within the upper scope boundary and the lowest resource boundary. However it is likely that the focus group would provide more services than the cost leadership gr oup because it would be expected that it is focused on pr oviding a specific type of service like rehab. There is more than one type of technology that a nursing home can focus on providing. It is therefore possible that 38

PAGE 39

within the nursing home industry, there exists more than one focus group. For example, a nursing home could focus their atten tion on providing nursing care while a different one could focus their attention on providing rehabilitati on services. Lack of Strategy Facilities that use high levels of resources but provide low levels of technology fall into the lack of strategy group. This group will also not have any focus of services and special allocation of resources. Measuring Performance Financial performance metrics are the most commonly used when comparing different strategic groups. Some of the more common measures of performance include total margin and operating margin. Total margin is the total revenu e divided by the total expense. This is an important measure because any organization that is unable to sustain positive margins will cease to exist. The total margin not only includes th e operating costs and revenue, but other costs (capital costs and interest) and other revenue (like endowments and donations). Operating margin only includes operating revenue and opera ting cost. Operating margins are important because they measure efficiency. The higher the operating margin, the more profitable a company's core business. Operating margin is re lated to the day-to-day decisions that managers make such as pricing strategy, pri ces for raw materials, and / or la bor costs. Operating margin is also a measure of managerial flexibility a nd competency of the firm (Gapenski 2000). In the health care industry (inc luding nursing homes), patient cost s is also widely used as a financial metric (Weech-Maldonado et al. 200 3, Shen et al. 2007). Since public payers like Medicare and Medicaid have been placing significant pressure on a f acilities ability to generate revenue, it has been necessary for facilities to co ntrol their costs in order to achieve positive margins. 39

PAGE 40

In most industries, success is most commonly defined by financial metrics; however, Venkatraman et al. 1986 have suggested that othe r measures such as market share, new product introduction and product quality can be equally important. One measurement that has recently gained popularity in business is the balanced scorecard (BSC) approach. A major characterist ic of a BSC is that it combines long-range strategic financial goals with day-to day operations. Kaplan a nd Norton (2001) stated that the customers perspective and internal operations are important areas that should be considered along with the financ ial perspective. Taking the customers perspective, the ability of a nursing home to provide quality care is crucial if a facility is going to survive over the long term. Poor quality can result in higher costs (Weech-Maldonado et al. 2003) as well as lower reve nue streams due to lower occupancy rates. Poor quality can also result in moratorium on new admissions and even a facilitys closure. Quality in nursing homes has also generated c oncern for policymakers and the public ever since the release of the Institute of Me dicine (IOM) report in 1986. It is therefore very important that quality indicators be used in conjunction with financial variables when measuring performance. There are three key reasons why measures othe r than financial perfor mance are needed in the nursing home industry. First, the government programs like Medicare and Medicaid were established to pay for the care of the program s beneficiaries. Alt hough enough funds must be allocated to keep companies viable, the purpose of the program is not to give organizations above normal profits. Second, not a ll health care organizations are fo r profit. For example, nearly one third of nursing homes in the US are either go vernment or not for profit facilities (Harrington et al. 2001) and it is important to consider wh at health care organiza tions do. The presence of not-for-profit in the industry is another reason why performan ce objectives other than profit 40

PAGE 41

maximization are needed (Scanlon, 1980). Fina lly, health care organi zations (like nursing homes) are often charged with th e responsibility of providing care for one of the most vulnerable populations. The direness of poor quality health car e is so great that usin g financial performance as the sole measure of success (like in othe r industries) is just not adequate. For the aforementioned reasons, the measurement success in the health care industry has to be based on quality of services provided, cost of th at service and financial performance. New Contributions To date, no study has examined if strategic groups can be identifie d by the technology choices made by the nursing home management. The purpose of this study is to determine if strategic groups can be classified using Port ers generic strategies based on how nursing home management allocate their resources and how they select their technology. The study will also determine if these groups are succinct and if they differ with respect to quality, financial performance and cost. What and how much of technology a nursing home administrator chooses to provide is an important stra tegy choice that can ha ve significant repercussions. The study is different from prior studies because it uses the industrys technology as the scope variables to determine the strategic groups and will define th ose groups using Porters generic strategies. Hypothesis Hypothesis #1: Nursing homes th at have a strategy can be categorized into one of four strategic groups: Differentiator, cost leadership, focus or best cost. Hypothesis #2: Nursing homes in the diffe rentiator strategic gr oup will provide higher quality care than othe r strategic groups. Hypothesis #3: Nursing homes in focus stra tegic group will provide higher quality in the area that they are focused on than other strategic groups. 41

PAGE 42

42 Hypothesis #4: Nursing homes in the cost leadership strategic group will have lower costs than other strategic groups. Hypothesis #5: Nursing homes that lack a strategy will have lower financial performance than nursing homes that are in a strategic group.

PAGE 43

43 CHAPTER 3 METHODOLOGY Data This study combined the On-line Survey Ce rtification of Automated Records (OSCAR), the Minimum Data Set (MDS), the Area Resource File (ARF) and the Medicare Cost Reports. The OSCAR dataset is a facility level dataset th at is routinely collect ed through the Medicare and Medicaid certification process conducted by st ate licensure and certifi cation agencies. This data is updated annually as part of the recertif ication survey. OSCAR pr ovides a snapshot about the characteristics and care processes occurri ng within the nursing hom es at the time of the survey. It includes variables like staffing levels, ownership status, number of residents, facility acuity level as well as quality outcome measures lik e state survey care deficiencies and pressure sore prevalence. The Omnibus Budget Reconciliation Act of 1987 (OBRA) was the first major legislative step to improve the quality of care in nursing homes. It established the Resident Assessment Instrument (RAI) which was implemented in 1991. The RAI is the basis of the Minimum Data Set (MDS) which is a dataset that records info rmation about nursing home residents. The MDS includes demographic information, health status as well as amount and types of services that a resident receives while they reside at the nursi ng home. Each long term care resident is assessed when first admitted to a nursing home, then each quarter thereafter. The post acute resident is assessed on the 5th day, the 14th day, the 30th day, 60th day and 90th day (for as long they remain in the facility). The CMS MDS is the most comprehensive dataset available for examining quality outcomes in the long-term care setting. OSCAR and Quality Indicators (QI) generated from the MDS are used by the CMS as part of its Nursing Home Compare website. The MDS was acquired using a reuse agreement from the CMS. Data is currently located behind a firewall in a fully secured server of the College of Public Health and Health Professions

PAGE 44

44 building operated by the IT departme nt at the University of Florida. It was originally acquired by University of South Florida. The Area Resource File (ARF) contains count y level data on supply and demand factors, such as composition of health care workfo rce, and socioeconomic and demographic characteristics of the population. The Medicare Cost Report data is a public a ccess relational dataset that can be acquired from the CMS via the CMS website. All CMS certified nursing homes that have Medicare beds are required to submit the Medicare cost report on an annual basi s. MedPAC has repeatedly been using this dataset to report the financial information of nursing homes to congress over the past 7 years (MedPAC 2007). This dataset provides financ ial data like total patient revenue, total patient costs, and total revenue. Population This study included all nursing homes in the United States that had Center for Medicaid and Medicare certification and that were included in the OSCAR, MDS and Medicare cost report datasets in 2004. There were 16000+ nursing homes in the United States in 2004. However, a total of 4500 facilities were omitte d because they were either hospital based (2000 facilities) or government based (700 facilities), did not have Medicare beds (1200 facilities) or were the only nursing home in a county and would not have any market competition (500 facilities). Hospital based facilities were omitted because they may behave differently from free standing facilities (Weech-Maldonado et al. 2003, Weech-Maldonado et al. 2004). For example, hospitals may place patients in their own subacute beds that th ey cannot place in other nursing homes due to the level of acuity of the patient or because the patient lacks insurance. Government facilities may behave differently than non-government fac ilities because of their different financing structure. They are less influenced by market forces because of funding that they may receive directly from public sources. Finally, some fac ilities that may be included in the OSCAR may

PAGE 45

45 not have any Medicare residents or even Medicare beds availabl e and therefore they would not have to file a Medicare cost re port. Therefore, this study is li mited to nursing homes that are non government, free standing and have Medicare certifie d beds. The total number of facilities in the study is approximately 11601 facilities in the year 2004. Overview of Methodology To identify the strategic groups and the strategy-performance relationships of the strategic groups, the study perf ormed a factor analysis on the scope and resource commitment variables. The resulting variables we re used in a cluster analysis to identify the different clusters. Once clusters were identified, ordinary least s quares regression (OLS), negative binomial and ordered logit was used to determine if the diffe rent groups had different financial and quality strategic-performance relationships. Hypothesis #1: Nursing homes that have a st rategy can be categorized into one of four strategic groups: Differentiator, cost leadership, focus or best cost. Scope Similar to Zinn et al. (1994) and Marlin et al. (1999), this study used the scope commitment and resource commitment (also kno wn as resource deployment) dimensions to identify the strategic groups. The scope of this study included technologies used by nursing homes to take care of the residents. The tec hnology of a nursing home can be classified into two main categories: nursing t echnology and rehab technology. Nursing Technology Nursing technology includes many different types of processes. The overall care of the residents can be divided into several different component technologies that are specifically designed to address the varied needs of the nursing home resident. Each of these technologies has a specific purpose and if not applied properly will have an undesired outcome. Some of the essential technologies include skin car e technology, restorative technol ogy, incontinence reducing technologies and nutritional technologies.

PAGE 46

46 Skin care technology is designed to prevent pr essure ulcers from occurring and to treat pressure ulcers that already ex ist. Failure to apply preventativ e skin care technology properly can result in pressure ulcers. Failure to treat alrea dy existing pressure ulcers can result in pressure ulcer worsening. Restorative nurs ing technologies engage the resi dent in functional activities. Residents functional abilities can decline in the nursing home se tting and restorative technology can be used to help maintain their functional levels Incontinence reducing technologies assist residents to maintain or regain their continence through toileting programs or bladder retraining programs. Nutritional technologies help residen ce maintain a healthy we ight. In the nursing home setting, residents are at risk of either ga ining weight or losing weight depending on their medical condition. Weight management programs and therapeutic diets assist with proper nutrition while the use of plate gua rds and other devices ensures th at residents adequately feed themselves. Skin care technology : All processes of care that involve either preventative skin care or treatment skin care are cons idered skin care technology. Pressure relieving seat : This is a continuous facility le vel variable that represents the proportion of residents in the nursi ng home that are given a pre ssure relieving seat. Pressure relieving seats are an important intervention when trying to prev ent a new pressure sore from occurring or to prevent an alre ady acquired pressure sore from worsening. Includes gel, air (e.g., Roho), or other cushioning placed on a chair or wheelchair. Include pressure relieving, pressure reducing, and pressure re distributing devices. Pressure relieving bed : This is a continuous facility le vel variable that represents the proportion of residents in the nur sing home that are given a pr essure relieving bed. Pressure relieving beds are an important intervention to prevent new pressure sores from occurring or to prevent an already acquired pressure sore from worsening. Includes air fluidized, low air loss

PAGE 47

47 therapy beds, flotation, water, or bubble mattre ss or pad placed on the bed. Include pressure relieving, pressure reducing, and pr essure redistributing devices. Turning / Repositioning : This is a continuous facility le vel variable that represents the proportion of residents in the nursi ng home that are placed on a turning-repositioning schedule. Turning-repositioning schedules are an important intervention to prevent a new pressure sore from occurring or to prevent an already acquire d pressure sore from worsening. It includes a continuous, consistent program for changing th e residents position and realigning the body. Program is defined as a specific approach th at is organized, planned, documented, monitored, and evaluated. Ulcer care: This is a continuous facility level va riable that represents the proportion of residents in the nursing home that are receiving ulcer care trea tment. Residents who get ulcer care already have ulcers. Surgical care: This is a continuous facility level vari able that represents the proportion of residents in the nursing home th at are receiving surgical care treatment. Residents who get surgical care already have ulcers. Ointment: This is a continuous facility level va riable that represents the proportion of residents in the nursing home th at are receiving ointment car e. It includes ointments or medications used to treat a skin conditi on (e.g., cortisone, anti fungal preparations, chemotherapeutic agents, etc.). Dressing care: This is a continuous facility level vari able that represents the proportion of residents in the nursing home th at are receiving dressing care. Residents who get dressing care already have ulcers. Skin nutrition program care: This is a continuous facility le vel variable that represents the proportion of residents in the nursing home that are on a skin nutrition program. Dietary measures received by the resident for the purpo se of preventing or treating specific skin

PAGE 48

48 conditions e.g., wheat-free diet to prevent alle rgic dermatitis, high calorie diet with added supplements to prevent skin breakdown, high protein supplements for wound healing. Vitamins and minerals, such as Vitamin C and Zinc, whic h are used to mange a potential or active skin problem, should be coded here. Other skin care : This is a continuous fac ility level variable that represents the proportion of residents in the nursing home that are receivi ng other skin care. Residents who get other skin care already have ulcers. Restorative technology : All processes of care provided by the nursing staff that involves having the patient engage in an activity that is directed towards accomplishing an ADL goal that will result in improved health status. Restorative ambulation : The restorative ambulation variable is facility level and is continuous. It represents the av erage number of days that a resident receiv es restorative ambulation while remaining in the facility. It is generated by dividi ng the resident level restorative variable that represents the number of days of ambulation provided in the 7 days before the assessment date by the total number of residents in the facility. Restorative ambulation has the nursing staff walk the residents in order to improve the distance that the resident can ambulate. It is a routine service and does not requ ire the skill of the therapist. However the goals of the program need to be generated by the therapist. Restorative Passive Range of Motion (PROM) : The restorative PROM variable is facility level and is continuous. It represents the averag e number of days that a resident receives restorative PROM while remaining in the facilit y. It is generated by divi ding the resident level restorative variable that represents the number of days of PROM provided in the 7 days before the assessment date by the total number of reside nts in the facility. Restorative PROM has the nursing staff range the residents in order to improve the PROM of the resident. It is a routine

PAGE 49

49 service and does not require the skill of the therap ist. However the goals of the program need to be generated by the therapist. Restorative Active R ange of Motion (AROM) : The restorative AROM variable is facility level and is continuous. It represents the averag e number of days that a resident receives restorative AROM while remaining in the facilit y. It is generated by dividing the resident level restorative variable that represents the number of days of AROM provided in the 7 days before the assessment date by the total number of reside nts in the facility. Restorative AROM has the nursing staff range the residents in order to improve the AROM of the resident. It is a routine service and does not require the skill of the therap ist. However the goals of the program need to be generated by the therapist. Restorative transfers : The restorative transfer variable is facility level and is continuous. It represents the average number of days that a resident receives rest orative transfers while remaining in the facility. It is generated by dividing the resident level restorative variable that represents the number of days of transfers provided in the 7 days before the assessment date by the total number of residents in the facility. Re storative transfers have the nursing staff transfer the residents in order to improve the residents ab ility to transfer from one position to another. It is a routine service and does not require the sk ill of the therapist. However the goals of the program need to be gene rated by the therapist. Restorative bed mobility : The restorative bed mobility vari able is facility level and is continuous. It represents the aver age number of days that a resident receives restorative bed mobility while remaining in the facility. It is ge nerated by dividing the resident level restorative variable that represents the number of days of bed mobility provided in the 7 days before the assessment date by the total number of residents in the facility. Restorative bed mobility has the nursing staff transfer the residents in order to improve the reside nts ability to tr ansfer from one

PAGE 50

50 position to another in bed. It is a routine servi ce and does not require the skill of the therapist. However the goals of the program need to be generated by the therapist. Restorative dressing : The restorative dressing variable is facility level and is continuous. It represents the average number of days that a resident receives restorative dressing while remaining in the facility. It is generated by divi ding the resident level restorative variable that represents the number of days of dressing provided in the 7 days before the assessment date by the total number of residents in the facility. Re storative dressing have th e nursing staff help dress the residents in order to improve the residents ability to dress by themselves. It is a routine service and does not require the skill of the therap ist. However the goals of the program need to be generated by the therapist. Restorative eating : The restorative eating variable is facility level and is continuous. It represents the average number of days that a re sident receives restorativ e eating while remaining in the facility. It is generated by dividing the resident level restora tive variable that represents the number of days of eating provide d in the 7 days before the assessment date by the total number of residents in the facility. Rest orative eating have the nursing sta ff help the residents with eating in order to improve the residents ability to eat by themselves. It is a routine service and does not require the skill of the therapist. However the goals of the program need to be generated by the therapist. Restorative splinting : The restorative splinting variable is facility level and is continuous. It represents the average number of days that a resident receives restorative splinting while remaining in the facility. It is generated by dividing the resident level restorative variable that represents the number of days of splinting provided in the 7 days before the assessment date by the total number of residents in the facility. Re storative splinting have the nursing staff assist with the donning and doffing of splints in order to improve the residents ability to splint

PAGE 51

51 themselves. It is a routine service and does not require the skill of the therapist. However the goals of the program need to be generated by the therapist. Restorative amputee intervention : The restorative amputee intervention variable is facility level and is continuous. It represents the averag e number of days that a resident receives restorative amputee intervention while remaining in the facility. It is generated by dividing the resident level restorative variab le that represents the number of days of amputee intervention provided in the 7 days before the assessment date by the total number of residents in the facility. Restorative amputee intervention have the nursing staff assist with th e donning and doffing of prosthetic in order to improve the residents abil ity to don and doff their prosthetic by themselves. It is a routine service and does not require the skill of the therapist. However the goals of the program need to be generated by the therapist. Restorative communication : The restorative communication vari able is facility level and is continuous. It represents the av erage number of days that a resident receiv es restorative communication while remaining in the facility. It is generated by dividing the resident level restorative variable that represents the number of days of communication provided in the 7 days before the assessment date by the total number of residents in the facility. Restorative communication has the nursing staff assist with the communication in order to improve the residents ability to communicate. It is a routin e service and does not re quire the skill of the therapist. However the goals of the program need to be generated by the therapist. Restorative other treatment : The restorative other treatment variable is facility level and is continuous. It represents the aver age number of days that a resident receives other restorative while remaining in the facility. It is generated by dividing the resident level restorative variable that represents the number of days of other provid ed in the 7 days before the assessment date by the total number of residents in the facility.

PAGE 52

52 Incontinence reducing technologies : All processes of care pr ovided by the nursing staff intended for the patient engage in a program that is directed towards accomplishing continence. Toileting program : This is a continuous facility level variable that represents the proportion of residents in the nursing home that have a to ilet program. Toileting programs include such items like toileting schedules. Bladder retraining : This is a continuous facility level variable that represents the proportion of residents in the nur sing home that have a bladde r retraining program. Bladder training programs include such items like toileting schedules. Nutrition technology: All processes of care provided by the nursing staff intended for the patient engage in a program that is dir ected towards maintaining a healthy weight. Weight management program : This is a continuous facility level variable that represents the proportion of residents in the nursing home that have a weight management program. Resident is receiving a program of which the documented purpose and goal are to facilitate weight gain or loss (e.g., double portions; high calorie supplement s; reduced calories; 10 grams fat). Therapeutic diet: This is a continuous facility level variable that measures the proportion of residents that are on a thera py diet. A diet ordered to manage problematic health conditions. Examples include calorie-specific, low-salt, lo w-fat lactose, no added sugar, and supplements during meals. Plate guard : This is a continuous facility level va riable that represents the proportion of residents in the nursing home havi ng a plateguard. This includes a ny type of specialized, altered, or adaptive equipment to facilitate the reside nts involvement in self performance of eating. Rehab Technology Rehab technology is utilized to improve the residents health stat us. It is provided by licensed therapy staff. There are di fferent types of therapist that have different functions. These

PAGE 53

53 processes of care include physical therapy for mobility issues and occupational therapy for ADL issues other than mobility. Long term care rehabilitation: All processes of care provided by the therapy staff that involves rehabilitation services to the long term care residents. Mean PT minutes long te rm care per resident : Facility level variable that represents the average number of minutes of phys ical therapy per resi dent provided in the 7 days prior to the quarterly assessment date. The numer ator is the number of minutes of PT and the denominator is the number of residents in the nurs ing home. This is a MDS resident level variable that is raised to the facility level. Mean OT minutes long te rm care per resident : Facility level variable that represents the average number of minutes of o ccupational therapy per resident pr ovided in the 7 days prior to the quarterly assessment date. The numerator is the number of minut es of OT and the denominator is the number of residents in th e nursing home. This is a MDS resident level variable that is raised to the facility level. Post acute rehabilitation: All processes of care provided by th e therapy staff that involves rehabilitation services to the long term care residents. Mean PT minutes post acute per resident: Facility level variable that represents the average number of minutes of Physical Therapy th e 7 days prior to the 14 day assessment date. The numerator is the number of minutes of PT and the denominato r is the number of residents in the nursing home. This is a MDS resi dent level variable that is raised to the facility level. Mean OT minutes post acute per resident : Facility level variable that represents the average number of minutes of Occupational Therapy provided in the 7 days prior to the 14 day assessment date. The numerator is the number of minutes of PT and th e denominator is the number of residents in the nursing home. This is a MDS resident level variable that is raised to the facility level.

PAGE 54

54 Percent Medicare beds : Facility level variable that represents the percent of the total number of resident beds reimbursed by Medicare. Medicare only reimburses for resident who are at the facility for short term stay. This variable is used because it is necessary to account for the proportion of the patients that ar e receiving rehabilitation at th e facility. It is generated by dividing number of Medicare residents by the total number of beds for that facility. Resource Commitment Resource commitment refers to the commitment of resources to the different technology areas that are needed to gain and maintain competitive advantag e in targeted market segments (Zinn et al. 1994). The primary resour ce is staffing levels (the nurses and the therapist) since it is the health care staff that pr ovides the technology. Therefore, organizations manipulate their staffing levels to control costs and ensure qual ity in an attempt to perform well financially. This study divided the resource commitment into two separate subtypes because they are mutually exclusive of each other and do not overlap. Resource commitment nursing: Resource commitment nursing subtype includes all the licensed and non licensed nursing staff. They are responsible for all the nursing technologies RN FTE / Total Residents : Registered Nurse (RN) per resi dent is a continuous variable representing the ratio of the tota l number of RN FTEs divided by the total number of residents in the facility. RNs are involved in establishing processes and supervisi ng other nursing staff. LPN FTE / Total Residents : Licensed practical nurse (LPN) per resident is a continuous variable representing the ratio of the total numbe r of LPN FTEs divided by the total number of residents in the facility. Both measures are from the OSCAR dataset. LPNs are able to provide some skilled nursing processes of care like dispensing medication. They have to be under the supervision of a nurse. CNA FTE / Total Residents : Certified Nursing Assistant (CNA) per resident is a continuous variable representing the ratio of the total number of CNA FTEs divided by the total

PAGE 55

55 number of residents in the f acility. Both measures are from the OSCAR dataset. CNAs are mostly involved with the direct care of patients. Percent Licensed : Percent nursing staff that are li censed is a continuous staffing mix variables that measures the proportion of nur sing staff that have a license. Higher percent licensed will give the nursing home a higher ca pacity to have know ledge and intangible resources. Percent of nursing staff that are RNs : Percent of nursing staff that are RNs is a continuous staffing mix variable that measur es the proportion of the nursing staff that are RNs. Higher RN staffing mixes are advantageous since RNs can us e their professional skills and tacit knowledge to improve the quality of patient care through their involvement in establishing processes and supervising other nursing sta ff (Weech-Maldonado et al. 2003). Resource commitment rehab : Resource commitment rehab includes all therapists and therapy assistants. Only licensed rehab staff is included because the use of rehab aides for therapy treatments is minimal due to state and federal practice guidelines. PT FTE / Total Residents : Physical Therapist (PT) per resident is a continuous variable represents the ratio total number of PT FTEs divided by the total number of residents in the facility. PTs evaluate the patient and oversee th e plan of care. They supervise the physical therapy assistants PTA FTE / Total Residents : Physical Therapy Assistants (P TA) per resident is a continuous variable represents the ratio total number of PT A FTEs divided by the to tal number of residents in the facility. PTAs provide the treatment to the patients under the supervision of the physical therapist. Percent PT: Percent of the physical th erapy staff that are PTs is a continuous staffing mix variable. Higher PT staffing mixes are advantageous since PTs can use their professional skills

PAGE 56

56 and tacit knowledge to improve the quality of patient care through their involvement in establishing processes and supervis ing other physical therapy staff. OT FTE / Total Residents : Occupational Therapist (OT) per resident is a continuous variable represents the ratio total number of OT FTEs divided by the total number of residents in the facility. OT evaluates the pa tient and oversees the plan of car e. They supervise the Certified Occupational Therapy Assistant COTA FTE / Total Residents : Certified Occupational Ther apy Assistant (COTA) per resident is a continuous variable represents the ratio total numbe r of COTA FTEs divided by the total number of residents in the facility. COTAs provide the treatme nt to the patients under the supervision of the occupational therapist. Percent OT: Percent of the occupational therapy staff that are OTs is a continuous staffing mix variable. Higher OT staffing mixes are adva ntageous since OTs can use their professional skills and tacit knowledge to improve the quality of patient care through their involvement in establishing processes an d supervising other occu pational therapy staff. Wages: Nursing homes average wages is a cont inuous variable derived by dividing the total salary expense by the total number of working hours reported in the Medicare costs reports. Wages are considered resources measures because nursing homes utilize the wages to control their cost. Hypothesis #1 Analysis Since there was a total of 68 va riables of interest (technology and resource variables), it was necessary to reduce the num ber of variables into composite variables by using factor analysis. The internal consistency of the newly created compos ite variables was tested using Cronbachs (1951) alpha methodology. Some variables that were not included in the composite scores were still included in the clus ter analysis as individual variables.

PAGE 57

57 The composite (or index) and the appropriate individual variables were transformed into cluster variables using z-scores. Wards mini mum variance method was used to identify the number of clusters and k-means cluster analysis to identify the different strategic groups. The analysis of variance and covariance (ANOVA) statistical test with Tukeys comparison test was used for each variable of interest to determine if the strategic groups were significantly different from each other. The Multivariate anal ysis of variance and covariance (MANOVA ) test with Wilks Lambda was used to determine if the ove rall clusters were signi ficantly different from each other. Each variable was given a weighted rank score for each cluster. The weighted rank score was then aggregated by type of commitme nt (scope or resource). Using the aggregated weighted rank score, the clusters were classified based on Porters generic strategies as defined in the technology, resources and strate gic group model presented in Figure #2. Factor Analysis Factor analysis is a variable reducing technique. It is a fa mily of procedures designed to remove the redundancy from a set of highly correla ted variables and representing the variables in a smaller set of derived variables or factors (Kachigan 1991). Factor analys is can be seen as the grouping of similar variables (Kachigan 1991). Thes e groupings of variables (or factors) can be used to describe a particular construct. For example, in the nursing home industry, two of the treatments used in an attempt to decrease pre ssure sores include placing a pressure relieving device in a chair and placing a pressure relieving device in a bed. It would be expected that these two processes of care are going to be highly correlat ed since both processes are similar in nature. The study used exploratory factor analysis to identify scope commitment and resource commitment variables that are correlated to each other. These variables were combined into new index variables that were a measure of what was considered a technology or resource factor. In the above example, if the factor analysis dete rmined that the two pressure relieving device

PAGE 58

58 variables are correlated and need to be combine d, the new variable may be considered a measure of the construct Pressure Sore Preventative Interventions. The study used the interaction between the scope and the reso urce dimensions to classify the strategic groups. It was theref ore necessary to perform separate exploratory factor analysis for scope commitment and for resource commitment. Similarly, the staffing intensity and staffing mix variables were also analyzed separately. Th is is because it is necessary to separate the intensity and the skill mix of the staff to be able to identify the relationship between licensed and non licensed staff. Staffing mix vari ables cannot be used as a substi tute for the staffing intensity (i.e. nurse per resident) variables because th ey do not explain the volume of nursing care provided. For example, in the case of nursing, C NAs can be negatively correlated with the RNs in one cluster while being positively correlated in another cluster. As a result, the three separate factor analysis include technology variables, sta ffing intensity variables and skill mix variables. Variables included in the cluste r analysis but t not in the f actor analysis were wages and percent Medicare beds. Although av erage wage paid is a resource commitment measure and is related to the staffing mix variab les (higher RN mix represents higher average wages), it was not included in any exploratory factor analysis. Wages are used by nursing homes as an incentive for hiring new staff and retaining thei r staff and therefore were incl uded separately in the cluster analysis. Percent Medicare beds va riable is related to scope because it indicates the proportion of the facility residents getting re hab. It was included in the clus ter analysis independently to identify the total volume of therapy services prov ided. For example, a facility that provided a lot of minutes of therapy to a small number of re sidents could be misclassi fied as being a rehab facility if the total amount of residents receiving therapy were not accounted for. Factor Matrix In the factor matrix, the columns represen t the derived factors while the rows are the variables. The cells represent the factor loadings which represents the de gree that the variables

PAGE 59

59 relate to the derived factor. Fact or loadings are used to reveal the extent that the variables contribute to the meaning of the variable. Factor loading are the correlation coefficients between the variables and the factors. A factor loadi ng score of greater than 0.4 was the inclusion criterion to be considered a part of that factor (Kachigan 1991). Number-of-Factors Problem One concern when using factor analysis is how many factors are reta ined. Each factor accounts for a certain amount of variance in the data. Typically, the first f actor will extract the most variance; the second factor will extract the second most and so on. Once the variance of each successive factor extracts is known, it is possible to de termine how many factors can be retained. There are two primary methods that can be used to determine the amount of factors retained: 1) Eigenvalue is a measure of the variation and represents the equiva lent number of variables that the factor represents. The Kaiser criterion is a commonl y used standard that includes only factors with eigenvalues greater than 1 (Kachigan 1991). The rationale for using such a threshold is that unless a factor extracts at least as much variation as the equivalent of one original variable, it should be dropped. This criterion was pro posed by Kaiser (1960), and is probably the one most widely used standard in determining how many factors should be used. 2) Scree plot is another method that can be used to id entify the number of factors. It is also possible to identify how many factors to retain by plotting the incremental variance accounted for by each successive factor. The idea is that the ta il of the curve represents mostly random error variance and therefore the factor solution is to in clude all factors that exist just prior to the leveling off of the curve (Kachigan 1991). Rotation of the Factor Structure We could plot the factor loadings shown above in a scatter plot. In that plot, each variable is represented as a point. In this plot we could rotate the axes in any direction without changing

PAGE 60

60 the relative locations of the poi nts to each other; however, the act ual coordinates of the points, that is, the factor loadings would of course change (Kachigan 1991). Rotational strategies: There are various rota tional strategies that have been proposed. The goal of all of these strategies is to obtain a clear pattern of loadings, th at is, factors that are somehow clearly marked by high loadings for so me variables and low loadings for others. Fabrigar et al. 1999 describe two commonly used types of ro tational strategies: orthogonal and oblique. Although orthogonal is conceptually simpler, factors have to be uncorrelated. Oblique rotation on the other hand do es not have this limitation. Si nce the resource factors have the potential of being correlated to the technology that they are used for (for example, therapy resources will be correlated with rehab technol ogy), this study used the oblique rotation. Internal Consistency of Measures It was necessary to determine whether the vari ables of the generated factors have internal consistency. Cronbach alpha tested the correlation between the variables of the construct. The higher the alpha value, the closer the variables are correlated (C ronbach 1951). This value should be above 0.7 (Allen 2002). If th e alpha was lower than 0.7, the factor was dropped because the variables did not correlate well together. To dete rmine how much impact the individual variables had on the overall alpha, multiple tests will provide a score of the resulting alpha. Any variable whose exclusion causes the alpha score to rise above 0.7 were be dropped. Similarly, any variable whose inclusion causes the alpha score to drop below 0.7 were dropped. Cluster Analysis Cluster analysis is an explorat ory data analysis tool which ai ms at sorting different objects into groups in a way that the degree of association between two objects is maximal if they belong to the same group and minimal otherwise (Kach igan 1991). Computationa lly, cluster analysis could be thought of as analysis of variance (ANOVA) "in reverse." The program starts with k random clusters, and then move objects between those clusters with the goal to 1) minimize

PAGE 61

61 variability within clusters and 2) maximize vari ability between clusters In other words, the similarity rules apply maximally to the members of one cluster and minimally to members belonging to the rest of the cluste rs. Cluster analysis can also be described as a set of techniques for partitioning a set of objects into relative ly homogenous subsets based on inner objects similarities. The process begins by measuring a set number of objects (nursing homes) on each of the variables of interest. A measure of difference between each pair of nursing homes is obtained. Porters competitive strategy typology wa s used to define the strategic groups of nursing homes. Similar to Marlin et al. 1999 and Zinn et al. 1994, this study used the scope and resource commitment measures in the cluster analysis. Prior studies argue that due to the large interstate variation in legislation, they could only use one state for their analysis (Zinn et al. 1994, Marlin et al 1999). Although it is true that states have different governing laws, different reimbursement systems and different survey methodologies, limiting the analysis to one state lim its the generalizability of the results. Rather than limiting to one state, this study standardi zed all variables using z-score based on the mean of the nursing home market (c ounty). Regulation and reimburseme nt may vary not only by state but also by county. Similarly income and demogr aphics can vary by county. All these factors may influence the nursing homes strategy on how to use their resources and how much technology to provide. Therefore it is necessary to compare nursing homes to others in their county. Z-score or normal score is a dimensionless quantit y derived by subtracting the population mean from an individu al raw score and then dividi ng the difference by the population standard deviation. In this study, the county mean and the county standard deviation were used. This study used an inductive methodology to determine the strategic group clusters. Kmeans is a common method used to determine cl usters using inductive reasoning (Ketchen 1996, Harrigan 1981). A significant numbe r of studies have used k-me ans to identify the strategic groups (Marlin et al. 1999, Ford et al. 1998, Zinn et al. 1994, Short et al 2002, Dess et al. 1984).

PAGE 62

62 K-means cluster analysis was specifically desi gned for large datasets (Rudolph et al. 1991). Since this study began with over 11500 observati ons, K means cluster analysis was the most appropriate. K-means cluster analys is requires the researcher to es tablish the number of clusters a priori. This presents a challenge because it is not known if the number of clusters selected a priori is the most well defined clusters. Using theory to determine the number of clusters is considered to be a valid method to identify the k-means a priori (Ketchen et al. 1996 ). However, Ketchen et al. (1994) also r ecommended using a quantitative method in conjunction with the inductive method to increase the robustness of th e number of clusters se lected. One such method is the Wards minimum variance method which has been used by numerous strategic group studies (Marlin et al. 1999, Ford et al. 1998, Zinn et al. 1994, S hort et al. 2002). One problem with the Wards minimum varian ce method is that it functions poorly with very large datasets and it is very susceptible to outliers. FASTCLUS wa s used as technique to reduce the variables into a smaller number of cl usters that would be more manageable for the Wards minimum variance method (SAS 1990). FAST CLUS uses k-means cluster analysis to reduce the number of observati ons to a level that the Wards minimum variance can manage. In this study, the initial FASTCLUS reduced the data set to 100 cluster/observations. The study also removed outlier observations that were greater than five standard deviations. Wards minimum variance method was used for determining the number of clusters to be used. The study used the following decision criteria to identify the optim al cluster solutions (SAS 1990). 1. a local peak in the cu bic clustering criterion 2. a local peak in the pseudo F statistic, 3. a small value of the pseudo t2 st atistic and a larger pseudo t2 statistic for the next cluster solution level, 4. an additional cluster increasing the over all fit by less than 5 percent, and 5. The clusters obtained explaining at leas t 65 percent of the overall variance.

PAGE 63

63 These decision criteria are consis tent with those used in prior strategic group research (i.e., Fiegenbaum and Thomas 1990; Mehra 1996; Marlin 1999). The number of clusters identified by the Ward s minimum variance method was used in the k-means cluster analysis using the original dataset. Once clusters were identified, the means for each one of the variables used to identify the cluster. These mean values were used to determ ine which of the Porters generic strategies the cluster could be categorized into. Strategic Groups Defined As a result of using z scores for the cluster analysis, the variable va lues had a mean score of zero. In order for the model in Figure #2 to be utilized, it was necessary to convert the values into a positive scale. Therefore the clusters were ranked from highest to lowe st from 0 to the total number of clusters that were determined in the cluster analysis. For example, if there were a total of 3 clusters identified in the analysis, then the ranking would range from 0 to 3. The actual distance between clusters for each variable was not equal and therefore it was necessary to weight the rank sc ore relative to the total varian ce of the variable. The ranking protocol of each variable was determined as follows: Determine the total variance value The total variance for each variable was determined by subtracting the lowest mean value from the cluster with the highest mean value. Fo r example, if for a give n variable, the cluster with the lowest values was -1.5 and the cluster with the highest value was +1.5. Therefore, the total variance value was 3. Cluster value For each variable, a cluster value was dete rmined by subtracting the cluster with the lowest value from each individual cluster variables value. Us ing the above example of the

PAGE 64

64 lowest value being -1.5 and a third cluster variab le having a value -0.5, the cluster value of this cluster value would be 1 (-0.5-(-1.5)). Cluster rank score The cluster value was divided by the total va riance value. This ratio was then multiplied by the total number of clusters Using the above information and assuming only three clusters, the cluster rank score would be 1/3 multiplied by 3 for a rank score of 1. For this case, the rank scores would be 0 for the lowest rank, 1 for the middle rank and 3 for the top rank. Table A-8 in the appendix includes all th e total variance and cluster values for each variable. All resource variables and all scope variables were summed to generate a cumulative resource utilization score and cumulative scope variable score each cluster. These total values were then plotted on the model as illustrated in Figure 2. Each cluster was classified into strategies using the parameters defined below. Differentiator : For a nursing home to differentiate itself from the other facilities in its market, it would provide the greatest amount care. It is expected that a differentiating nursing home would be highly ranked in the skin car e technology, restorative nursing technology, other care technology such as bladder tr aining or weight program, and highest rehab care relative to the average nursing home in their county. To be c onsidered a differentiator, the cluster would have to have the highest cumulative scope score. Cost leadership : Some nursing homes will be focused on keeping their costs low. Since staffing is the significant cost driver in the nursing home indus try, strategies to have lower staffing levels can be used as a measure of cont rolling cost. Therefore it was expected that the cost leadership group would have the lowest cumulative rank scores for therapy resources, nursing resources and wages. Best cost : Some nursing homes may seem like a hybr id of the cost leadership group and the differentiator group relative to the lack of strategy group. It was expected that this group

PAGE 65

65 would have higher cumulative rank scores for re sources and scope when compared to the cost leadership group and would have lower cumula tive resources and scope scores than the differentiator. In Figure #2, the cen troid best cost cluster would ha ve fallen in the area between the boundaries set by the cost leaderships centr oid and the differentiati ng clusters centroid. Focus : Like any industry, it is possible for a nursing home to focus their strategy on a smaller market segment. The focus group can be identified by pr oviding a high amount of technology and resources in a spec ific area like rehab services. For example, if a cluster had high rank scores in post acute reha b, long term care rehab and ther apy resources, but low scores in every other category; they w ould be considered a rehab focu s group. There was the possibility that other focus groups also existed. Lack of strategy: A facility that uses more resources but provides less technology than a cost leader was considered as a facility that lacks a strategy. Analysis of Variance This study used Analysis of Variance (ANOVA) to determine if the component variables of the newly identified strategic groups are significantly different from each other. The study also used Tukeys method to make multiple pa ir wise comparisons of the means for each variable. The ANOVA used the F test statistic which is the ratio of the between error estimates divided by the within error estimates. If Ho is false, the between estimates tend to overestimate the variance, so it tends to be larger than the within estimates. This can make the F statistic significantly greater than 1.0 (Agresti and Finlay 1997). The Ho Hypothesis is that the mean of each variable of interest (for example RNs per resident) was the same for each strategic group; that is Differentiator = Cost Leadership = Best Cost = Focus = Lack of Strategy. To determine whether the entire clusters are significant from each other, the study also used the MANOVA test with Wilks' Lambda. Similar to ANOVA, this statistic evaluates the hypothesis that the

PAGE 66

66 population means are equal across groups; however, th is method uses a multivariate approach. It was used in Zinn et al. 1994 study that examined strategic groups. Hypothesis #2: Nursing Homes in the Different iator Strategic Group will have the Highest Quality. Dependent Variables: There were a total of 8 dependent variables that were qual ity measures; five of which relate to long term care quality, two of which measure post acute quality and one measure of the overall quality of the facility. The long term care (LTC) and post acute (PA) measures were quality indicator (QI) variables; they include d LTC bladder decline in continence, LTC bowel decline in continence, LTC 4 poi nt decline in ADL function, the LTC prevalence of the pressure sores, LTC restraint prevalence, PA walking impr ovement, and PA pressure sore incidence. The bladder and bowel decline variables are measures of the proportion of residents who had become incontinent. The ADL decline in function is a risk adjusted ratio of the proportion of the residents who experienced a 4 point ADL decline (out of 16 points) relative to the total number of residents in the nursing home. The restraint prevalence is the proportion of residents that are restrained at the facil ity. The pressure sores prevalence is a risk adjusted ratio of the proportion of residents who have a pressure sores relative to the total nu mber of residents in the nursing home. The walking improvement variable measures the change in ambulation status between the 5 day assessment and the 14 day assessment. Pressu re sores incidence variable is a measure of the amount of post acute residents who developed a pressure sore while remaining in the nursing home relative to the total number of post acute residents in the nursing home. The quality indicators were continuous variables calculated using the MDS based on the methodology as described by Abt Associates 2004. These quality measures are included on the Nursing Home Compare website and are currently used by soci al workers and consumers to differentiate between good and bad facilities. The eighth quality variable was th e total number of quality of care deficiencies. This is an overall measure of the facilitys quality. Quality of care deficiencies

PAGE 67

67 is a count variable that represen ts the number of citations issued by state survey agencies and is located in the OSCAR dataset. Deficiencies are categorized acco rding to level of seriousness from letter A to L. Deficiencies categori zed F and higher denote those with varying degrees of pervasion, which have the potential to, or actually ca use harm or immediate jeopardy to residents (ONeill et al. 2004). It also is on the Nursing Home Compare Website and is used by social workers and consumers to differe ntiate between good and poor facilities. Independent Variables The independent variables of interest were dichotomous variable (0 or 1) for each strategic group generated by the cluster analysis. The expected strategic group variables include: differentiator strategy, cost lead ership strategy, best cost stra tegy, focus strategy, and lack of strategy for those facilities th at did not have any strategy. Control Variables The control variables in the study included both organizational and market factors. They included size, ownership, chain affiliation, ma rket competition, occupancy, payer mix and case mix. Nursing home size is measured using the number of beds. Larger size provides more resources and gives an advantage when trying to recruit staff. Similar to size, chain affiliation gives more resources to draw upon which can also improve the facility efficiency. Ownership and chain affiliation in nursing homes has been associated with lower quality of care (Harrington et al. 2001). The percent Medica re and percent private pay be ds are outcomes of effective strategy because both give the facility an ab ility to increase their revenue streams and subsequently provide the facility with more resources to provide quality of care. Market competition and the occupancy of the facility can influence the behavior of a facility as they attempt to attract more profitable residents. A facilitys case mix is the measure of the acuity of the nursing homes residents. Highe r acuity of residents may be more costly and more difficult to care for and may influence quality measures like pressure ulcers and ADL decline that is beyond

PAGE 68

68 the control of the facility. Several studies have utilized the Acuindex as the case mix variable (Johnson et al. 2004, Laberge et al. 2005, Weech-Mal donado et al. 2006). The Acuindex is an acuity variable located in the OSCAR dataset. It is represented by di fferent factors including mobility issues like how much a ssistance the resident requires with functional activities like ambulation or transfers, or nur sing issues like propor tion of residents receiving suctioning or proportion of residents receiving intravenous thera py. This study also included other measures of acuity that included ADL index, tube feeds, respir atory care, suction, tracheotomy care, IVs, and injections. These measures were included because different strategic groups may have different types of acuity. For example, a strategic group may have a high proportion of the residents that need IVs, respiratory care and need injecti ons but who are independent with their ADLs. Although the Acuindex is a good overall measure of acuity, it cannot capture the intricacies of the different groups that could be related to the different stra tegies. These different nursing acuity measures were not used in the cluster an alysis because they represent the needs of the residents rather than the technol ogy that was provided to them. For example, knowing that half the residents need respiratory care is not the same as a measure that represents the amount of treatment or number of different kinds of tr eatment provided. Although th ey were not used in determining the different groups, these variables ar e needed to be included as controls because they can influence how the techno logy of the facility is provided. Analysis The study intended to use the ordinary least squares (OLS) regression to analyze seven of the quality of care variables (AREG function on STATA) (LTC bladder decline in continence, LTC bowel decline in continence, LTC 4 point de cline in ADL function, the LTC prevalence of the pressure sores, LTC restraint prevalence, PA walking improvement, and PA pressure sore incidence). Prior to the analysis, it was necessary to evaluate the dependent variables in order to ensure that they have normal distribution. If any of the variables were not normally distributed,

PAGE 69

69 the dependent variables were transformed by using the log function. If the initial log transformation did not result in proper skewne ss and kurtosis, other transformations were attempted. Other possible transfor mations included the cubed root the root, the square and the cube. All of them would be attempted until a transformed variable was identified that had a skewness and kurtosis closest to 0 and 3 respectively. If the skewness was not able to be resolved and the variable seemed to have a gamma distribution, the study used the generalized linear model with gamma distribution and a log link. After the OLS an alysis, it was necessary to determine if the regressions pass the OLS linearity assumptions. The linearity assumption was tested through observation of the residual-predicted values plots and the Hosmer-Lemeshow test. The quality of care deficienci es is a count variable and wa s analyzed using the negative binomial equation. The study used Whites heteroskedacity-consistent covariance matrix (1980) to correct for a heteroskedascity. It used county level fixed effects. Dependent variable outliers of 5 standard deviations and above were dropped from the dataset. Qual = + g1Group1 + g2Group1 + ..+ gnGroupN + cControl + feFE + Qual represents one of the se ven quality dependent variables. represents the constant value, while gn represents the effect of each strategic group on the quality variable. gc represents the effect of the contro l variables on the quality variable. fe represents the effect of the county level fixed effects. represents the random error. Hypothesis #3: Nursing Homes in the Focus Strategic Group will provide Higher Quality in the area that they are focused on compared to other Strategic Groups. Dependent Variables Hypothesis #3 used the same quality of care dependent variables as Hypothesis #2. Independent Variables As in hypothesis #2, the independent variables of interest included a dichotomous variable (0 or 1) for each strategic group gene rated by the cluster analysis. However, unlike

PAGE 70

70 hypothesis #2 where the differentiator was used as the reference group; this hypothesis used the focus strategic group variable as the reference group. Control Variables Organizational characteristics that were used as controls included size, for-profit status, chain affiliation, payer mix, and case mix as described in hypothesis #2. Analysis As in hypothesis #2, the planned analysis for the QIs was the OLS with county fixed effects (AREG function on STATA. The study used Whites heteroskedacity-c onsistent covariance matrix (1980) to correct for a heteroskedascity. It used county level fixed effects. Dependent variable outliers of 5 standard deviations and above were dropped from the dataset. Focus Qual = + g1Group1 + g2Group1 + .. +gnGroupN + cControl + feFE + Qual represents one of the se ven quality dependent variables. represents the constant value, gn represents the effect of each strategic group on the quality variable. gc represents the effect of the control variab les on the quality variable. fe represents the effect of the county level fixed effects. represents the random error. Hypothesis #4: Nursing homes in the Cost Le adership Strategic Group will have the lowest costs. Dependent Variables The patient cost equation used operating expe nse per resident day. It is determined by dividing the operating revenue by the total amoun t of residents and dividing it by 365 days. Independent Variables The independent variables of interest include d a dichotomous variable (0 or 1) for each strategic group generated by the cluster analysis. Since the focus of the cost leader was to have the lowest cost, the cost leadership strategic group was used as the reference group.

PAGE 71

71 Control Variables Organizational characteristics that will be analyzed are size, for-profit status, chain affiliation, payer mix, market competition and case mix as described in hypothesis #2. Analysis Similar to health care expenditure measures the cost per patient per day tends to be skewed to the right with some facilities having si gnificantly higher cost than the mean. Similar to prior studies dealing with health care cost, th e study used the genera lized linear model (GLM) with gamma distribution and a log link (Fleishman et al. 2006, Pagano et al. 2005). The study then determined the predicted value using the smearing method. The study used c ounty level fixed effects. The study used Whit es heteroskedacity-consistent covariance matrix (19 80) to correct for any heteroskedascity as found in STATA (2001). It used county level fixed effects. Dependent variables outliers of 5 standard deviations and above were dropped from the dataset. lnCost = + g1Group1 + g2Group1 + ..+ gnGroupN + cControl + feFE + lnCost represents one of the seven quality dependent variables. represents the constant value, gn represents the effect of each strategic group on the quality variable. gc represents the effect of the control variab les on the quality variable. fe represents the effect of the county level fixed effects. represents the random error. Hypothesis #5: Differentiation, Cost Leadership, Focus and Best Cost Strategic Groups will have better Financial Performance than Nursing Homes that Lack a Strategy Dependent Variables There were two variables used to examine fi nancial performance. The first of the two financial performance variables used, the operatin g margin, is calculated using the operating revenue minus the operating expense then divide d by the operating revenue. Operating revenue and expense were direct measures of doing business and therefore the operating margins may be more sensitive to nursing home strategies. The second financial performance variable included total profit margin. Profit margin not only includes the operating revenue and operating expense,

PAGE 72

72 but also includes the non patient revenues like charitable donations and non patient expenses like capital costs. Although profit margin is not as sensitive as operating margins to changes in the nursing homes strategies, it is still important becau se it does represent th e overall picture of the nursing homes financial performance. Financial performance is important because it strongly influences the nursing homes behavi or. If strategies that produces higher quality re sult in better financial performance, then customers needs and providers goals will be aligned. However, if a strategy provides higher quality of care but does not provide the facility with better financial performance, alternative strategies by th e nursing homes will likely be considered. Independent Variables The independent variables of interest include d a dichotomous variable (0 or 1) for each strategic group generated by the cluster analysis. The lack of strategy group was used as the reference group since it was expected that nursin g homes that had a strategy would have better financial performance than those that did not have a strategy. Control Variables Organizational characteristics that were utilized as controls included size, for-profit status, chain affiliation, Medicaid and case mix as de scribed in Hypothesis #3. Unlike the other hypothesis, hypothesis #5 only used Medicaid as a control vari able. Some nursing homes are trying to acquire the high er reimbursing Medicare as part of their strategy. By using the percent Medicare variable as a control variable, it would not be po ssible to determine the benefit of having a strategy directed towa rds acquiring rehab patients. Analysis Although the operating margin a nd the total margin were both continuous variables, neither of them is normally distributed and ther efore Ordinary Least Squares regression is not appropriate. By the nature of the mathematical equation that is used to determine them, their range is from negative infinity to positive one. This is because both equations use the revenue of

PAGE 73

73 the facility in the numerator and the denominator and both equations subtract the costs of the facility from the revenue in the numerator. Therefore the highest value that the operating and total profit margins can be is one ((revenue minus no costs) / revenue = 1). Since the variable is not normally distributed, and cannot be transformed in a parsimoni ous way, a different analytical tool is necessary. In addition, the focus of the study is to dete rmine if the total and the operating margins vary by strategic groups. As a result, determining the odds of being a higher financial performer is appropriate. Therefore, the study used an orde red logit with odds ratio analysis to determine the odds that the strategic group will be in the highest group. Weech-Maldonado et al. 2008 used the ordered logit model with odds ratio in their study that examined quality and financial performance. The study determined the predicted probabilities of a strategic group being in each financial tier for each strategic group. By providi ng the predicted probabilities, it was possible to identify which strategies have the best financ ial performance for opera ting margin and total margin. The study divided the operating margin and the total margin variables into three financial performance tier: low (bottom 10%), medium (middle 80%), and high (upper 10%). Similarly, since different counties can have si gnificantly different market forces between them, it was necessary to control for these count y differences by using co unty fixed effects. Predicted values using logit we re used to identify predicted probabilities by tier for each strategic group. Ologit ( ) = + g1Group1 + g2Group1 + ..+ gnGroupN + cControl + feFE + lnCost represents the co st dependent variable. represents the constant value, gn represents the effect of each stra tegic group on the quality variable. gc represents the effect of the control variables on the quality variable. fe represents the effect of the county level fixed effects. represents the random error

PAGE 74

74 CHAPTER 4 RESULTS Hypothesis #1 Identification of Different Factors: The study completed three separate factor analyses to reduce the total number of variab les. The newly generated compos ite variables represented valid constructs of nursing homes technology and resource commitments. Initially there were 57 technology variables and 11 resource variables that were used in the factor analysis. Factor analysis #1 (Technology) : The first factor analysis included only the variables that measured the technology of the nursing homes. Of the 57 reported variables, there were 20 factors that were derived reporting an eige nvalue greater than 1 (table A1 in the appendix), with ten of those factors having eigenvalues between 1.0 a nd 1.54. Although the factors just above one would qualify using the Kaiser (1960) cr iteria, the amount of variati on removed by each subsequent factor after 10 was marginally in adequate. Therefore it was necessa ry to use the scree plot to identify the ideal number of clusters. The scree pl ot indicated a point of inflection between 9 and 11 factors (Figure A-1 in the appendix). Although the eigenvalues could justify the use of 20 factors, the scree plot indicates that the change in the amount of variance contributed by using additional factors beyond 10 appeared to be nonsignificant. Theref ore, the factor analysis for technology only used the 10 factors. All variables with a factor loading above 0.4 were initially included in the each of the 10 factor variables (Table A-2 Appendix A). There we re six variables that were included in more than one factor. Since each of these six variab les can only be in one factor, it was necessary to determine the most appropriate factor for each vari able. The six variables included long term care pressure relieving chairs, long term care pressure relieving beds, long term care and post acute restorative PROM and AROM. The long term care pr essure relieving chairs and beds had loadings that were similar in both factors (0.6). One of the factors also included post acute pressure relieving devices. Therefore this factor generated a clear construc t of pressure re lieving devices.

PAGE 75

75 The ROM variables were included in the restor ative long term care and restorative post acute measures. The ROM variables were also included in a factor that included splinting; however this factor had a Chronbachs alpha score that was below the 0.7 leve l and therefore was dropped. The end result was 8 technology factors with variables that had factor loadings over 0.4 (Table A-3 Appendix A). Factor Analysis #2 (Resource) : The staffing intensity factor analysis identified three factors with eigenvalues greater than one (Table A-4 Appendix A) The scree plot appeared to demonstrate a point of inflection between 2 and 3. The three factors included licensed therapist (PT and OT), therapy assistants (PTA and COTA) and nursing staffing (including RN, LPN, and CNA) (Appendix A table 4). Factor Analysis #3 (Resource) : For the staffing mix, the eigenvalues were all below 1. Since each of the factors removed less variance th an one variable, it was more appropriate to include each of the staffing mix variables in dividually rather than as a composite. The factor analysis identifi ed 10 technology factors and 3 re source factors. These factors were tested for internal consistency using the Cronbachs alpha method. As illustrated in Table 41, 8 technology factors and 2 resource factors ha d Cronbachs alphas over the 0.7 threshold. These factors included long term care sk in care, post acute ca re skin care, long term care restorative process, post acute restorative processes, relie f devices, therapy minutes, toileting program, weight management program, therapist, thera py assistants. The restor ative splinti ng technology variable was included in the factor that was a ssociated with restorative AROM and PROM yet failed to have Cronbachs alpha score greater th an 0.7. The technology variab les therapy diet and plate guard made up the final factor. This factor was dropped because they failed to have an alpha score greater than 0.7. Nursing (RN, LPN and CNA) on the other hand did have an alpha scor e of 0.69; however it was decided not to use a total sta ffing intensity factor because a nursing home could, as a strategy,

PAGE 76

76 choose to substitute a licensed staff member with a non licensed staff in an attempt to keep costs low. By generating an index that included all th ree nursing variables, the ability of the cluster analysis to identify nursing homes with different types of staffing intensity strategies would be compromised. It was therefore deemed necessary to include the RN, LPN and CNA variable in the cluster analysis as i ndividual variables. Table 4-1. Internal consistency of the composite variables Cluster variable Alpha Variables used Long term care skin processes of care 0.82 Long term care ointment, dressing application process, surgical wound care, ulcer care, skin nutrition and turning repositioning program Post acute skin processes of care 0.72 Post acute ulcer care, dressing application process, Long term care restorative processes 0.86 Long term care restorative dressing, restor ative transfers, restorative bed mobility, restorative PROM, Restorative AROM, restorative ambulation and restorative eating Post acute restorative processes 0.91 Post acute restorative dressing, restorative transfers, restorative bed mobility, restorative PROM, Restorative AROM, restorative ambulation and restorative eating Relief Devices 0.81 Long term care pressure relieving bed and chair, Post acute pressure relieving bed and chair Therapy minutes 0.93 Long term care and Post acute physical and occupational therapy Toileting program 0.88 Long term care and Post acute toileting program Weight management prg 0.88 Long term care and post acute weight management program Therapist 0.91 Total physical and occupational therapist per resident Therapy assistants 0.77 Total physical and oc cupational therapy assistants per resident Table 4-2 lists the nursing variables and all the individual variables utilized in the cluster analysis. Other variables included the per cent of nurses that were RNs, percent of nurses that were Table 4-2. Individual va riables used in cluster analysis that were not included in a factor variable. Individual Variables Definition RN Registered Nurse Per Resident LPN Licensed Practical Nurse Per Resident CNA Certified Nurse Assistant Per Resident RN Percent Percent of Nurses are Registered nurse License Nurse Percent Percent of Nurses are Licensed PT Percent Percent of Physical Therapist are licensed OT Percent Percent of Occupa tional Therapist are licensed Wages Facility direct care average wage Percent Medicare Percent re sidents who are post acute licensed, percent of physical therapist, and the percent o ccupational therapist. As described in the methodology, wages and percent Medicare beds were also used as individual variables.

PAGE 77

77 Cluster Analysis: The 10 factor variables and the 9 indi vidual variables were standardized using z scores that were based on the county m ean. The dataset were then trimmed for outliers using the 5 standard deviations for each variable. This resulted in the redu ction of the number of nursing homes used in the cluster analysis from 11601 to 9726. The newly generated variables were used in the Wards Minimum Variance Clus ter Analysis and the kmeans cluster analysis. Ward's Minimum Variance Cluster Analysis identified 5 clusters as the best number of clusters. This was based on a local peak in the pseudo F stat istic, a small value of the pseudo t2 statistic followed by a larger pseudo t2 statistic for the ne xt cluster solution leve l, and that adding an additional cluster increased the overall fit by less than 5 percent (Table A-6 and A-7 Appendix A). No other cluster arrangement met the three criteria. The data was then analyzed using the k-means with an a priori of 5 clusters. Clusters: Table 4-3 represents the results of the cl uster analysis. Using the ranking protocol described in the methodology (Table A-8 Appendix), the different cl usters were defined using the rank score ranging from 0 to 5. 0 represented the lowest and 5 represented the highest value for both the resource variables and th e technology variables. For each cluster, the cumulative rank score was calculated by summing the values fo r the scope variables and summing the resource variables. These cumulative rank scores were plo tted on a graph to generate model as indicated in Figure 4-1. The scope score was represented by th e horizontal axis and the resource score was represented by the vertical axis. Table 4-4 provided some descriptive structural, market and case mix variables of each strategic group. The ANOVA test identified that the clusters we re significantly different from each other for each variable. In addition, the Tukeys test compared each pair of clusters for every variable and found that there were significant differences be tween each pair 79% of the time. Finally, the MANOVA test criteria (Wilks' Lambda) found that th e entire clusters were significantly different from each other (p <.000). Descriptives of a ll variables is located in table A-9 in the appendix.

PAGE 78

78 Table 4-3. Rank scores scope and resource variables Variables Cluster 1 Differentiator Cluster 2 Cost Leader Cluster 3 Rehab Focus Cluster 4 Specialty care Focus Cluster 5 Lack of Strategy Average Wages 1.985 0.813,4 4.042,5 5.001,2,5 0.001,2 F=13.8*** RN per Resident 1.70* 0.00* 2.72* 5.00* 0.40* F=744*** LPN per Resident 0.993,4 1.073,4 5.00* 0.00* 0.863,4 F=121*** CNA per Resident 4.362,3,4 3.59* 5.00* 0.00* 4.052,3,4 F=367*** Therapist per Resident 1.00* 0.00* 5.00* 1.75* 2.23* F=1110*** Therapy assistant per resident 1.19* 2.48* 5.00* 1.62* 0.00* F=1492*** Percent Licensed PT 2.542,3,5 0.00* 1.58* 2.442,3,5 5.00* F=1180*** Percent Licensed OT 2.612,3,5 0.00* 1.94* 2.602,3,5 5.00* F=906*** Percent RN 1.15* 0.00* 1.42* 5.00 0.25* F=1538*** Percent Licensed Nurse 0.41* 0.071,3,4 1.76* 5.00* 0.001,3,4 F=833*** Total Resource 17.91* 8.06* 33.49* 28.32* 17.78* F=186*** LTC Skin Process 3.23* 0.60* 5.00* 2.07* 0.00* F=186*** Acute Skin Process 4.682,4,5 1.56* 5.002,4,5 3.03* 0.00* F=58.35*** LTC Restorative Process 5.00* 0.101,3,4 0.89* 0.34* 0.001,3,4 F=688*** Acute Restorative Process 5.00* 0.141,3,4 0.00* 0.33* 0.141,3,4 F=1987*** Relief Devices 4.112,4,5 1.672,4,5 5.001,4,5 1.512,4,5 0.00* F=39.8*** Toilet Program 4.96* 1.49* 0.111,2 0.091,2 0.001,2 F=19.7*** Weight Program 5.003,5 3.733,5 0.001,2,4 2.533 1.031,2 F=8.62*** Therapy Minutes 0.00* 1.601,3,5 5.00* 1.681,3,5 1.13* F=386*** Percent Medicare 0.103,4 0.153,4 5.00* 0.74* 0.003,4 F=918*** Total Scope 32.1* 11.03* 26.10* 12.32 2.29 F=59*** *Significantly different from all clusters, Superscript is significantly di fferent than identified groups The descriptives values were not adjusted for th e county level fixed effects and therefore were not used in the determination of the clusters. The plotted clusters in Figure 4-2 were then categorized using Porters generic strategies. Based on the definition describe d in the methodology, Cluster #1 was initially classified as Differentiator because its scope score was the hi ghest (technology cumulative rank score). However this group was ranked the lowest in all the categories that was related to rehab (i.e. physical therapy minutes and percent Medicare beds). Therefore this cluster may be more appropriately classified as a nursing care focus group rather than a differentiator. Although these nursing homes provided a high degr ee of technology, they did not use the highest amount of resources. They tended to have median level wages, median level total RN staffing, and below median level of percent licensed therapist. They had low staffing mix as a result of their high

PAGE 79

79 CNA staffing levels. This group al so tended to have high acuity pr imarily driven by residents who were more dependent and patients who had tube feeds. Table 4-4. Rank scores of st ructure and market factors Variables Cluster 1 Differentiator N=990 Cluster 2 Cost Leader N=2752 Cluster 3 Rehab Focus N=1816 Cluster 4 Specialty care N=1634 Cluster 5 Lack of Strategy N=2534 Percent for profit 71.2% 75.7% 79.7% 76.9% 76.2% Chi2= 27 Percent chain affiliated 51.5% 60.0% 66.4% 63.0% 58.2% Chi2= 70 Market competition 1.43 0.00 0.71 5.00 4.29 F=0.17 Occupancy 5.00 3,4 4.53 3,4 0.00 1,2,5 1.98 1,2 3.16 3 F=12.15 Size 0.67 2 5.00* 1.15 2,5 1.86 2,5 0.00 2,3,4 F=59.23 Case Mix 4.57 2,4,5 1.06 1,3 5.00 2,4,5 2.18 1,3,5 0.00 1,3,4 F=25.95 DL index 4.73 2,4,5 1.14 1,3 5.00 2,4,5 2.08 1,3,5 0.00 1,3,4 F=21.45 Percent tube feed 4.22 2,5 1.19 1,3,4 4.51 2,5 5.00 2,5 0.00 1,3,4 F=9.65 Percent tracheotomy 3.485 1.49 4 3.15 5 5.00 2,5 0.00 1,3,4 F=14.63 Percent requiring suctioning 2.96 2,5 0.71 1,3,4 3.39 2,5 5.00 2,5 0.00 1,3,4 F=15.63 Percent requiring respiratory care 1.28 3 0.50 3,4 5.00* 2.67 2,3,5 0.00 3,4 F=42.13 Percent requiring IV 0.64 3,4 0.74 3,4 5.00* 1.87* 0.00 3,4 F=54.7 Percent requiring injection 0.98 3 0.78 3 5.00* 2.19 3,5 0.00 3,4 F=33.18 *Significantly different from all clusters, Superscript is significantly di fferent than identified groups Cluster #2 was classified as the cost leader because it used the least resources. The only nursing resource that the cost leader wa s not the lowest on was the CNA staffing. The lowe st RN staffing coupled with the median CNA staffing resulted in the lowest staffing mix. Cost leadership groups also had the lowest amount of th erapist with an above median level of therapy assistant. This resulted in the low therapy staffing mix. Cost le aders were significantly la rger that other nursing homes and tended to have lower acuity. These were all consistent with wh at would be expected with a cost leader strategic group. Cluster #3 was classified as the rehab focus cluster. It had higher wages and the highest staffing intensity. It had an above median level of RN staffing and above median percent licensed staff. Rehab focus nursing homes provided the highe st skin services and therapy. They had the highest use of resources and provided a significant amount of service second only to the differentiators. This cluster cannot be classified as best cost because although the nursing homes

PAGE 80

80 provided more service than the co st leadership group, they also used more resources that the differentiators. The rehab focus group had th e highest acuity in most categories. Cluster #4 was classified as specialty care focus group. It had the highest overall wages, overall RN staffing and RN proportion when compared to the other clusters. This cluster had skin care services and weight program, and overall cas e mix acuity in the middle range. This cluster did not use the least resources, or provide the most processes and therefore could not be classified as either cost leader or differen tiator. It could not be considered a best cost strategic group because although this cluster provided a greater amount of car e than the cost leader cluster, it used more resources than the differentiating cluster. Alt hough this group did not have the highest overall acuity level, it did have the highest amount of resi dents with specialty care needs like tube feeds and tracheotomies. However, these specialty care services have li ttle effect on the overall case mix score (acuindex) because these types of servi ces account for only 20% of the acuindex variable (while the ADL factors account for 80% of the variable). Cluster #5 was classified as the lack of strategy cluster. This cluster of nursing homes appeared to lack any specific st rategy. They had patients with th e lowest acuity, and they provide by far less services than other groups while their st affing levels still remained at a median level. All the clusters were plotted on the strategic group model (Figure 4-1) using the cumulative technology and cumulative resource scores from tabl e 4-3. Hypothesis #1 was partially supported. The results indicated that there exist four distin ct strategies in the nur sing home industry, however these groups were different then what was or iginally hypothesized. Based on the Figure #3, there was no best cost group; instead ther e was a fifth group that was classi fied as a specialty care focus group. Figure #3 depicts the existence of a differentiator group based on the methodology, however closer examination of the individual pr ocesses of care indicates a nursing care focus strategy.

PAGE 81

81 Figure 4-1. Strategic group mode l based on scope and resource commitment in the nursing home industry Normal distribution: All of the eight quality dependent variables had excessive skewness and non-normal kurtosis levels and therefore had to be transformed. After being transformed using the either the root or log func tion, the skewness and the kurtosis of the quality valuables were long term care (LTC) bladder decline (-0.03, 3.62), LTC bowel decline (0.15, 3.19), LTC 4 pt ADL decline (0.13, 3.77), LTC restrain ts (0.18, 2.52), LTC pressure ulce r prevalence (-0.09, 3.81), Post Acute (PA) walking improvement (0.315, 4.15) and PA pressure ulcer incidence (-0.78, 4.32). The PA measures both failed the normal distribution a ssumption even after transformation. Both post acute variables were therefor e analyzed using the generali zed linear model with gamma distribution. Linearity: By observing the predicted value and residual plot, all six quality variables that were used in the OLS regression model appeared to pass the linearity assumption. The study also confirmed these results with a negative Hosmer-Lemeshow test (non-significant) for all the dependent variables.

PAGE 82

82 All the above quality variables with skew ness under 4 were analyzed using OLS with county level fixed effects. The quality variables that had skewness over 4 were initially analyzed by using the generalized linear model (glm ) with gamma distri bution and a log link. Unfortunately, the glm was not able to find a solution when using post acute pressure ulcer incidence quality variable. Therefore the PA pr essure ulcer incidence was analyzed using OLS with fixed effects. Hypothesis #2 Hypothesis #2 was partially supported. Table 4-5 (f ull regression results ar e located in tables A-9 to A-16 in the appendix) illustrates that the differentiator strategic group had either the same or better long term care quality than all other strategic groups in all measures except long term care pressure ulcers. Differentiators had lower 4 point ADL decline than cost leaders and those nursing homes that lacked a stra tegy. Differentiators also had fe wer pressures sores than nursing homes that had a rehab focus or had a specialty care focus. However, with respect to the post acute measures, differentiators performed poorly. The differentiator strategic group generally had worse outcomes with less walking improvement th an all other groups. Differentiators had higher incidence of post acute pressure sore than lack of strategy and cost leaders. Table 4-5. Strategic group quality of care with Di fferentiator / nursing care focus cluster as the reference group Dependent Variables Cost Rehab focu s Specialty care Lack of Strategy Quality of care deficiencies 1.06 (0.09)* 0.99 (0.86) 1.00 (0.97) 1.12 (0.00)* Long Term care measures Bladder Decline 0.006 (0.04)* 0.012 (0.00)* 0.009 (0.01)* 0.005 (0.12) Bowel Decline 0.008 (0.02)* 0.009 (0.03)* 0.012 (0.00)* 0.005 (0.16) ADL 4 pt decline 0.008 (0.00)* 0.003 (0.39) 0.005 (0.10)* 0.004 (0.13) Restraint use 0.011 (0.01) 0.004 (0.46) 0.001 (0.19) 0.001 (0.19) Pressure Ulcers -0.001 (0.66) 0.012 (0.00)* 0.004 (0.14) -0.004 (0.09) Post acute measures Ambulation improvement 0.031 (0.10) 0.050 (0.02) 0.037 (0.08) 0.060 (0.00) Pressure Ulcers incidence -0.045 (0.02) -0.015 (0.46) -0.027(0.20) -0.076 (0.00) Quality of care results ar e incidence rate ratio Square root transformation LTC bladder decline, bowel decline pr essure ulcer prevalence, LTC 4 point ADL decline, restraints, P ost acute pressure sore incidence are OLS regression coefficien ts. Post acute walking improvement was gl m with gamma distribution. Represents s ignificant values in the desired direction. Represents significant values in the negative direction.

PAGE 83

83 Hypothesis #3 This hypothesis was partially supported. As de scribed above, nursing homes that focus on providing nursing care had good quality. However, th e other to focus groups did not have similar results. The rehab focus group had the highest re hab staff, provided a highest amount of therapy minutes, provided the highest amount of skin care and had the highest proportion of post acute beds. This group was expected to have the best post acute outcomes. The second focus groups had a high proportion of their nurses that were RNs and had the highe st proportion of patients with specialty care needs like tracheotomies, tube f eeds and patients who required suctioning. This group was expected to have better overall care than those nursi ng homes that lacked strategy and better quality outcomes than the cost leaders. Table 4-6 (full regression results are avai lable in table A-18 to A-25 in Appendix A) illustrates that nursing homes in the rehab focu s group had lower ADL 4 pt decline and lower restraint use than the cost leaders. In addition, the reha b focus had lower quality of care deficiencies than the cost leaders and the lack of strategy group. This group only had better PA outcomes than differentiators. This group generally had poorer LTC quality of care Table 4-6. Strategic group quality of care with rehab focus cluster as the reference group Dependent Variables Cost Differentiator / Nursing care focus Specialty care Lack of Strategy Quality of care deficiencies 1.07 (0.03)* 1.01 (0.86) 1.01 (0.80) 1.13 (0.00)* Long Term care measures Bladder Decline -0.006 (0.05) -0.012 (0.00) -0.003 (0.39) -0.007 (0.02) Bowel Decline -0.001 (0.77) -0.009 (0.03) 0.002 (0.52) -0.004 (0.21) ADL 4 pt decline 0.006 (0.03)* -0.003 (0.39) 0.002 (0.39) 0.002 (0.53) Restraint use 0.011 (0.01)* 0.004 (0.46) 0.006 (0.19) 0.006 (0.22) Pressure Ulcers -0.013 (0.00) -0.012 (0.00) -0.008 (0.00) -0.016 (0.00) Post acute measures Ambulation improvement -0.019 (0.28) -0.050 (0.02)* -0.014 (0.47) 0.014 (0.57) Pressure Ulcers incidence -0.030 (0.06) 0.015 (0.46) -0.012 (0.49) -0.061 (0.00) Quality of care results ar e incidence rate ratio Square root transformation LTC bladder decline, bowel decline pr essure ulcer prevalence, LTC 4 point ADL decline, restraints, P ost acute pressure sore incidence are OLS regression coefficien ts. Post acute walking improvement was gl m with gamma distribution. Represents significant values in the desired direction. represents significant values in the negative direction.

PAGE 84

84 especially in comparison to the differentiator and the lack of strategy grou p. This group had higher LTC prevalence and higher PA incidence of pressure ulcers than cost lead er and lack of strategy group. Table 4-7 (full regression results are avai lable in table A-26 to A-33 in Appendix A) illustrates that nursing homes in the specialty care focus group had lower quality of care deficiencies than the cost leaders and the lack of strategy group. They had less pressure ulcer prevalence than facilities in the rehab focus group. The specialty care focus group had higher prevalence of pressure ulcers than cost leader and lack of strate gy. It had higher bladder decline, bowel decline, ADL 4 point decline than differen tiators. It had lower restraint use than cost leaders. With respect to post acute incidence of pressure sores, the results were mixed with specialty care focus having a higher incidence of pressure ulcers than l ack of strategy group. And a higher prevalence of long term care pressure so res that cost leaders and lack of strategy. Table 4-7. Strategic group quality of care with Speci alty care focus cluster as the reference group Dependent Variables Cost Differentiator / Nursing care focus Specialty care Lack of Strategy Quality of care deficiencies 1.06 (0.05)* 0.99 (0.80) 1.00 (0.97) 1.12 (0.00)* Long term care measures Bladder Decline -0.003 (0.26) 0.003 (0.39) -0.009 (0.01) -0.004 (0.12) Bowel Decline -0.003 (0.27) -0.002 (0.52) -0.012 (0.00) -0.006 (0.03) ADL 4 pt decline 0.003 (0.16) -0.002 (0.39) -0.005 (0.10) -0.001 (0.78) Restraint use 0.005 (0.22) -0.006 (0.19) -0.002 (0.68) -0.001 (0.88) Pressure Ulcers -0.005 (0.01) 0.008 (0.00)* -0.004 (0.14) -0.008 (0.00) Post acute measures Ambulation improvement -0.005(0.74) 0.014 (0.47) -0.037 (0.08)* 0.024 (0.15) Pressure Ulcers incidence -0.018 (0.25) 0.012 (0.49) 0.027 (0.20) -0.049 (0.00) *Quality of care results are incidence rate ratio Square root transformation LTC bladder dec line, bowel decline pressure ulcer prevalence, LTC 4 point ADL decline, restraints, Post acute pressure sore incidence are OLS regression coeffi cients. Post acute walking im provement was glm with gamma distribution. Represents significant values in the desired direction. Represents significant values in the negative direction. Hypothesis #4 Hypothesis #4 was not supported. Table 4-8 illustra tes that cost leaders nursing homes had significantly lower costs that sp ecialty care and rehab focus groups but no significant difference between cost leaders and differentiators or with those facilities that l ack a strategy altogether.

PAGE 85

85 Table 4-9 provides the cost predicted values for each strategic group. The predicted values are the patient costs per resident day. Sp ecialty care focus and Rehab focus groups had significantly higher costs than the cost leader by 9 and 13 dollars per re sident day respectively. Table 4-8. Strategic groups and costs with cost leader as reference group regression results Lack of Strategy Differentiation / Nursing Care Specialty Care Focus Rehab Focus Costs per resident day -0.006 (0.60) 0.002 (0.90) 0.039 (0.00)* 0.080 (0.00)* Represents significant values compared to cost leadership cluster Table 4-9. Strategic groups and predicted costs Differentiator / Nursing Care Focus Cost Leader Rehab Focus Specialty Care Focus Lack of Strategy Costs per resident day 149.15 149.06 161.65* 155.03* 148.12 In dollars Hypothesis #5 The initial analysis that included all the case mix variables was unable to converge. It was therefore necessary for the study to use acuindex va riable rather than the 7 individual case mix variables. By dropping the 7 case-mix variables, th e ologit analyses were ab le to converge (A-34 and A-35 in Appendix A). The ologit for both financial performance meas ures (table A-34 and A-35 Appendix A) had variables that were completely determined. 53 of the observations for operating margin and 57 observations for total margin observations were co mpletely determined. There was possibility this was the result of multicollinearity between some of the county leve l fixed effects. There was over 1500 county level fixed effects variables with many of them having a very low number of nursing homes within the county. A correlation matrix wa s generated to determine if the completely determined variables were either the independent variables of interest or one of the control variables other than the county fixed effects variables. The correlation matrix (table A-36 appendix A) illustrated that none of the indepe ndent or control variables were completely determined. As a result of the high number of c ounty level fixed effects, the determined variables

PAGE 86

86 made up approximately 5% of all the county fixed effects and therefore likely have minimal impact the overall model. Hypothesis #5 was partially supported. Table 410 highlights that differentiator had a 20% greater odds of being in the highest tier for operating margin and a 31% greater odds of being in the highest tier for total margin relative to thos e nursing homes that lacked a strategy. Cost leaders had 25% greater odds of being in the highest tier for operating ma rgin and 22% greater odds of being in the highest tier for tota l margin relative to those nursing homes that lacked a strategy. Rehab focus had 46% greater odds of being in the highest tier for ope rating margin and 27% greater odds of being in the highest tier for to tal margin relative to those nursing homes that lacked a strategy. Differentiators, cost leaders and rehab focus had better total margins than the specialty care focus strategy but only the rehab focus group and the cost leadership group had better operating margins than the specialty care focus group. The specialty care focus group did not have significantly higher odds of being in th e highest tier with respec t to the nursing homes that lacked strategy. Table 4-10. Odds Ratio of Nursing Home Strategy Dependent Variable Cost Leader Differentiation Specialty care Focus Rehab Focus F test Operating Margin 1.25 (0.01)* 1.20 ( 0.09)* 1.04 (0.66) 1.46 (0.00)* F=9.7 (0.05) Total Margin 1.22 (0.01)* 1.31 (0.01)* 0.98 (0.80) 1.27 (0.01)* F=10.0 (0.03) *Significantly different from lack of strategy Table 4-11 and 4-12 illustrate the predicted pr obabilities of the operating and total margins. Both demonstrate that the nursi ng homes with differentiation, cost leadership and rehab focus strategy perform better than those nu rsing homes that lack strategy. Table 4-11. Probability of nur sing homes being in the hi ghest operating margin tier Dependent Variable Differentiator5 Cost Leader4,5 Rehab Focus4,5 Specialty care Focus2,3 Lack of Strategy1,2,3 Low 0.099 0.096 0.085 0.110 0.111 Mid 0.799 0.799 0.797 0.798 0.797 High 0.102 0.105 0.118 0.092 0.090 # Significant p>0.10

PAGE 87

87 Table 4-12. Probability of nursing homes being in the highest total margin tier Dependent Variable Differentiator4,5 Cost Leader4,5 Rehab Focus4,5 Specialty care Focus1,2,3 Lack of Strategy1,2,3 Low 0.089 0.094 0.094 0.112 0.109 Mid 0.799 0.799 0.799 0.798 0.799 High 0.112 0.106 0.106 0.090 0.092 # Significant p>0.05

PAGE 88

88 CHAPTER 5 DISCUSSION This study concurred with prior research because it demonstrated the ex istence of strategic groups in the nursing home industry. However, this study used a different conceptual framework to identify them. While earlier studies by Zinn et al. 1994 and Marlin et al. 2002 used resources and payer mix, this study used the business leve l decisions of resource allocation and technology provision. One advantage of using technology over payer mix variables to identify strategic groups is the applicability of th e results for managers. Management has control over how much technology to provide and how mu ch resource to use. How a nursing home decides to balance these two factors is an important part of that facility s overall strategy. The results from this study demonstrate that different strate gic groups provide their technology a nd utilize their resources in a different way. Using technology and resource allocation to iden tify the clusters makes it possible to define the groups using Porters generic strategies. There are no nursing home studies that have attempted to define strategic groups by Porters generic strategi es. However, Marlin et al. 2002 did find the existence of Porters strategies in the hosp ital industry. Unlike this study, Marlin et al 2002 explicitly used variables th at defined the differentiator strategic group. These variables included the total number of hos pital units, the number high technology units and ratio of intensive care beds to total beds. They did not use any measures that reflect how the care was provided. Although a hospital could have a high number of different specialty units, it did not necessarily provide a high degree of technology within those units. E ach specialty unit could have been poorly staffed and the patients could have received a very low level of services. Marlin et al. 2002 study was examining the governance level of decision making as opposed to the business level decisions. Business level op erations are significantly impacted by the day to day decisions of how to provide service and how to use resources.

PAGE 89

89 The study identified 5 of the strategic groups listed in the hypothesis (nursing care focus (differentiator), cost leader, rehab focus, specia lty care focus and lack of strategy). The study did not identify the best cost group and did not iden tify a well defined differentiator group. Instead, the study identified three focus groups; rehab focus, specialty care focus a nd a nursing care group. The rehab group from this study appeared similar to the high Medicare fa cilities in the prior nursing home strategic group studies This group not only had a high proportion of Medicare beds, but also provided a high level of therapy minutes and skin care. The specialty care focus group had a high RN intensity and had a high concentration of residents with specific care needs. Residents with specialty care needs like trache otomy care or suctioning require the nursing home that has a capacity to deliver a high level of skilled services. These conditions are also only minimally covered by Medicare in the post acu te setting and have long been considered underserved by MedPAC (MedPAC 2003, 2007). Be ginning in 2004, nursing homes could get more reimbursement for residents with these exte nded needs. This change in reimbursement only started in 2004 and it may be possible that the facilities had not been able to fully take advantage of it. One unexpected finding was that the defined differentiator stra tegic group provided the least therapy services. The fact that this group is defi ned as a differentiator when it provides the lowest amount of therapy brings up the question of the ap propriateness of using differentiator as this groups definition. The group clearl y provided the most nursing car e technology and therefore was reclassified as the nursing care focus strate gic group. Whether this group is considered a differentiator or a nursing care fo cus strategic group, this group still represents nursing homes that are trying to separate themselv es from their competition by pr oviding a high level of technology and therefore they can still be defined by using Porters generic strategies. Out of all the clusters identified, the cost l eader proved to be the best defined. The cost leader group had the lowest staffing levels and lower staffing mixes. Lower staffing mixes would

PAGE 90

90 indicate a cost leader because it indicates that the nursing home is substi tuting CNAs for RNs. In addition, on observation of the structur al features of this group, the cost leaders also tended to be larger and have lower acuity. It is clear that this groups strategy was to keep cost under control by providing a low amount of service. Finally, the lack of strategy group was identified because it used a median amount of resources with almost no technology. In addition, the nursing homes that lacked strategy appeared not to have any specialty services. By using technology and resource deployment, this study found that strategic groups not only exist, but can be classified using Porters generic strategies. The ne xt question is whether strategic-performance relationships exist between the different st rategic groups and quality, costs and financial performance. Quality This study identified that the nursing care focus (differentiator), sp ecialty care focus and rehab focus strategic groups had lower quality of care deficienci es than the nursing homes that were in the cost leader strate gic group or the nursing homes th at lacked strategy. This finding provides some evidence that focusing on providi ng quality either by providing technology or having better staffing mixes can result in some better outcomes. However, only quality of care deficiencies had consistent results through these three stra tegic groups. Quality of care deficiencies is a general measure of the facilities quality and can relate to anything from staff that is inadequately trained to poor care practices The Nursing Home Comp are website lists and describes all of facilities deficiencies from th e most recent state survey. It is a good general measure of quality and has been used in past research (Harri ngton et al. 2000, 2001). One limitation of deficiencies is that it does not hi ghlight specific care outcomes. Other measures like the QIs that were derived by Abt are specifically designed to identify the specific quality issues like pressure ulcer prevalence and ADL declin e. Knowing the specific quality areas would

PAGE 91

91 hopefully help managers address the problems direc tly. In this study the results of these measures were mixed. When compared to other groups, the nursing ca re focus strategic group generally provided better care to their long term car e residents. The nursi ng care focus, however, also had poor post acute outcomes. The fact that this group provided the least amount of minutes of therapy, had a lower proportion of post acute patients, and had the poorest post acute quality provides further support to classify this cluster as a nursi ng technology focus group ra ther than an overall differentiator group. These results suggest that there may be a re lationship between a strategic group focus and the quality of care provided. The other two focus strategic gr oups did not fair as well. Th e rehab focus strategic group had a high level of rehab staffing and provided high amount of rehab technology. The specialty care focus had a high level of RN staffing and provide d care to residents with specific care needs. It would be expected that the rehab focus group would have the bett er post acute care outcomes than other groups. However, rehab fo cus strategic group had a higher incidence of pressure sores and did not have better walking im provement than the specialty nurse, cost leaders or nursing homes that lacked strategy. This was surprising since the rehab focus did not have better outcomes even though they provided significantly more therapy. There are a number of possible explanations. The rehab focus strategic group had higher acuity than other strategi c groups. Although the quality measures used are risk adjusted and the case mix of the facility was controlled for, there may be factors that may not be captured. Mukame l et al. 2008 has determined that the current quality measures including pressure ulcers would bene fit with a greater degree of risk adjustment. The rehab focus group has a high proportion of Medi care patients and may have the expertise to handle patients with more co morbidities that may not be captured by the risk adjustments of the outcome variables. Similarly, different strategi c groups had different ty pes of acuity measures.

PAGE 92

92 Some of the case mix variables may have greater influence on the quality outcomes than others. The rehab focus groups may be penalized by taking on residents that are more prone to pressure ulcers and whose risk factors are not captured by the risk adjustment s of the quality measures or control variables used in the analysis. In the case of the walking improvement, th e rehab focus group provided nearly 25% more minutes then any other group and yet did not attain better outcomes than the specialty care focus, cost leaders or lack of strate gy groups. There are a few reasons that may explain why the rehab focus group does not have better walking impr ovement outcomes. First, rehab focus nursing homes may be downgrading the progress of their residents. Rehab focus strategic groups are accustomed to providing care to rehab patients and may be more rigorous in the application of walking improvement scales. A second reason is that there may be an upper therapy minute threshold where adding more minutes of therapy does not translate into better outcomes. The rehab group provided significantly mo re minutes of therapy when compared to the specialty care focus, the cost leader groups and the lack of strategy group ye t did not have better therapy outcomes. The fact that differentiators provide considerably less therapy and have worst outcomes when compared to all other groups supports the no tion that there is at l east a minimal threshold level of minutes of therapy that needs to be achieve d in order for basic outcomes to occur. Another possibility is that si nce revenue streams are enhanced by providing higher intensity of therapy for a longer period of time, these faciliti es may be more conservative with their 14 day assessments in an effort to increase the length of stay. By having a lower 14 day assessment value, the facility could be able to ju stify keeping the resident at the facility for a longer period time. Over the past 7 years, MedPAC have issued stat ements of concern that the current post acute reimbursement schematic promotes therapy overutiliz ation. More research is needed to determine if the lack of post acute quality outcomes differ ence between the rehab group and the cost leader,

PAGE 93

93 lack of strategy and the specialty care groups is the result of some therapy intensity upper boundary, or if the rehab focus groups are manipulating from the current reimbursement system. It would be expected that the specialty care focus strategic group would have better quality of care outcomes in the areas where high RN staffing levels are needed. Generally, besides the quality of care deficiencies and in comparison to the differentiator, the sp ecialty care group results were somewhat mixed. The fact that differentiators had better outcomes is not surprising since the outcome measures are more closely related to the technology provided by the differentiators. Similar to the differentiator and the rehab fo cus group, the specialty care group had a higher prevalence of pressure ulcers with their long term care residents than the cost leader and the lack of strategy. This was contrary to Weech-Maldon ado et al. 2004 that id entified RN staffing intensity and staffing mix as related to quali ty outcomes like lower pressure ulcers. One explanation for the high pressure sore prevalence could be endogeneity. Nursing homes with a high prevalence of pressure sores may employ mo re RNs so they can treat those residents effectively. Cost Rehab focus and specialty care focus group have significantly higher costs than the other groups. This is not surprising since both groups ha ve a higher amount of licensed staff. What was unexpected is that the nursing ca re focus group (differentiator group) had similar costs to the cost leader group. These results are unexpected be cause nursing care focus group had high acuity and provided the most technol ogy at the same cost as the cost leaders and those nursing homes that lacked a strategy. Past studies have found that be tter quality translates to lower cost in nursing homes (Weech-Maldonado et al. 2003). In the case of this study, there was a relationship between the nursing care focus group and be tter long term care outcomes. Although the nursing care focus group did not have the lowest staffing costs, it may compensate by saving money by providing more effective and efficient care.

PAGE 94

94 Financial Performance The study found that rehab focus, nursing care focus group and cost leaders had better financial performance than the nursing homes that lacked a strategy and the specialty care focus group. The poor financial performance of the specialty care focus group could be related to higher costs of having more RNs without the benefit of generating higher revenue streams. Unlike the rehab group where the more therapy results in mo re revenue, the specialty care focus groups may have limited ability to increase revenue. Medica re and some Medicaid programs can (depending on the state) provide more payment for nursing care; however the reimbursement is provided based on the utilization of services and not based on the staffing levels of the facility. In addition, these groups provided a care to pati ents with highly specific a nd costly conditions. MedPAC 2007 has expressed concern that the cu rrent Medicare reimbursement syst em does not cover the costs of care adequately to patients with needs over than therapy. Rehab focus nursing homes would be able to generate the greatest am ount of revenues since they had th e most Medicare rehab patients. This would explain why even though the rehab fo cus group had some of the highest costs, they still managed to be one of the best financial pe rformers. These results are providing some support with prior research that found that nursing homes with higher Medicare residents and private pay residents had the best financial performance (Zi nn et al. 1994, Marlin et al. 1999). However, in this study there were other groups that also perf ormed well. Differentiators managed to have good financial performance without a high degree of reha b. One explanation is th at by 2004, most states through Medicaid used case mix reimbursement that would compensate facilities that provided certain types of nursing technology (e.g. restorative care). Nursing homes that provide services that can generate revenue and lower costs via be tter quality can increas e their odds of achieving better financial performance. The hypothesis of this study stated that nursi ng homes that lacked a strategy would not perform as well as nursing homes that had a strategy. The results suggest that this is true with one

PAGE 95

95 exception. The specialty care focus group appeared to be a poor strategy with respect to financial performance. This outcome presents the possibility that not all strategies are created equal. The strategic groups rehab focus, cost leader and nursing care focus group (differentiator) all had better total margins than the other nur sing homes with no signifi cant difference between these strategies. One aspect of strategic group theory is that it is possible that there are multiple strategies within the same industry that are effective but not necessarily better than each other. This study supports this aspect of strategic group theory. Porters (1980) original theory described three effective strategies: cost leaders, differentiators and focus. This study provides support for these stra tegies even if the line between the differentiator and the focus strategy is blurre d. Porters generic stra tegies can help nursing homes identify strategies that allow them to achieve good financial performance. Managerial Implications Management is interested in performance. Providing managers with a way of identifying effective strategies that can im prove their facilities performance should be well received. Cool and Schendel (1987) defined strategic groups as sets of firms competing within an industry on the basis of similar combinations of scope and re source deployment commitments. Marlin et al. 2002 argued that for health care administrators, scop e decisions refer to the selection of market segments to compete in and the types of services to offer in these market segments. This study differed from prior research because it examin ed if the business level decision of technology commitment could be used to identify different st rategies and whether those strategies would have different performance outcomes. For managers that find themselves in facilities that would be considered to have poor or no strategy, this study provides some interesting findings. Strategic group theory posits that there exist mobility barriers that restrict a facility from moving from one strategic to another. The size of the investment needed for a nursing home to move from one strategic groups dictate the height

PAGE 96

96 of the barrier. This study partia lly supports that theory. Certainl y a nursing home would have to incur significant cost to enter the rehab focus gr oup thereby protecting this group with a relatively high mobility barrier. However, the mobility barrier to the differentiator may not be as high since these groups staffing levels, wages and cost are considerably low. Since it appears that nursing homes that provide higher levels of technology can have low costs and be profitable, managers with low performing strategy can improve their performance providing more care with the staff that they have. The study found that business level decisions of technology and resource utilization can be used to identify strategies th at lead to greater fi nancial performance and lower cost. The study identified cost leaders, rehab focus and diffe rentiators groups as ha ving significantly better financial performance than the RN focus strate gy and nursing homes that had no strategy. The rehab focus group had the highest proportion of Medicare beds and other payer beds. Although this supports prior studies that found that those facilities that had higher Medicare beds and private pay beds perform well fi nancially (Marlin et al. 1999 a nd Zinn et al. 1994), it found that this strategy was not significantly better than the differentiation or the cost leadership. Therefore, managers may be able to choose from one of thes e three alternatives and still have their facility perform well financially. One of the primary functions of managers is to control costs. With differentiators able to provide a high degree of care at the same costs as the cost leaders raises the question of whether there are cost savings in keeping staffing levels and staffing mixes lo w. Further research is needed to know how the differentiators were able to pr ovide all the services yet have similar costs. Policy Implications State and federal governments have a consid erable interest in nur sing homes since they provide over 60% of the funding. In return for thei r investment, they expect quality care for the residents. Therefore, it is in the governments best interest to know if there are some stra tegies that

PAGE 97

97 are better than others with resp ect to ensuring qualit y of care. This study has found some evidence that nursing homes with strategi es directed toward providing care have better quality of care than those nursing homes focused on cost savings. Th e study also found some evidence that nursing homes that provide a high level of nursing care technology to their long te rm care residents have better quality of care outcomes for that population. The study al so identified a group of nursing homes that provide care to patient with needs ot her than rehabilitation that may not be receiving adequate funding. The specialty care group provi des service to a population that is very vulnerable. Future policies should be directed to provide incentives that increase the amount of the important processes of care provided and not just address structural factors like staffing. Another factor is that is of interest to government agencies like CMS is that facilities that provided a significantly higher amount of therapy did not have bett er post acute outcomes with respect to walking improvement. At the same time, the strategic group that provided significantly less therapy than the other groups had worse walk ing improvement. From these results, it appears that there is a minimum amount of therapy required for adequate improvement, but there is limit beyond which more therapy does not lead to any mo re significant benefit. With the high costs of post acute care which is almost solely pa id for by the government, understanding how much therapy is sufficient is be very valuable. Further research is necessary to determine the minimum amount of therapy that is necessary. Similar to managers, policymakers are inte rested to know that a strategy of providing a high degree of service does not ne cessarily costs more. The fact that some facilities are able to provide a high level of service and yet keep th eir cost down can guide policymakers to provide incentives that focus on care prov ided as well as staffing. Further research is needed examining the relationship between processe s of care and the outcomes that they would be expected to influence.

PAGE 98

98 Conclusion The study indicates that strate gic groups exist in the nursi ng home industry that can be defined by Porters generic strategy. The study also demonstrates that strategic groups can be determined using the business leve l strategies of what technol ogy to provide and how to deploy resources. With respect to quality, nursing hom es that focused on providing technology (rehab or nursing) or had higher RN staffing levels had less quality deficien cies than facilities that did not have a strategy at all. Only the strategic group that provided a broad array of nursing technology had better quality in the area that the technology was focused. Th e study found that nursing homes that provide a high level of nursing technology do not necessarily have higher costs. The study further demonstrate that Porters cost leader, nursing care focus and rehab focus strategies have better financial performance than those facilities that have poor strategy or do not have a strategy at all. Limitations There are a number of limitations to this study. First, the results of the study apply only to free standing, non government faciliti es that have Medicare cost re port and are locat ed in a county that has at least 2 nursing homes. Facilities that only had private pay residents or facilities that only had Medicaid residents were not included in the sample. This limits the generalizability to the 9700 nursing homes that have all these charac teristics. Another limitation includes the possible e ndogeneity between pressure ulcer quality measures and the technology variable for skin technology. The skin technology variable includes both preventative and treatment processes. As a resu lt, especially in the case of the long term care skin technology variable, nursing ho mes with a high prevalence of pr essure ulcers are forced to provide treatments. This fact increases the li kelihood of an endogenous relationship between the quality variable and the strategic groups that pr ovide a high level of long term care skin processes and may explain the why the rehab focus group had the highest preval ence and incidence of

PAGE 99

99 pressure sores and also provided the most skin care. Another con cern about the post acute pressure ulcer incidence variable was that it did not have a normal distribution and an adequate transformation was not achieved. When the GLM an alysis was attempted, the analysis could not converge. Therefore the results of this measure may not be reliable. Another limitation of the study was the magnitude of the impact of some of the dependent variables. 4 point ADL decline a nd post acute improvement in walking had very small regression coefficients relative to the cons tant. Although these variables presen ted statistical significance, the small size of the effect may limit the relevance. The study used cross-sectional data and therefore caution shou ld be taken when trying to infer a causal relationship. Finally, the datase ts included in the study are administrative data. Although the nursing homes can be held accountable for the accuracy of the data, some inconsistencies can occur dependi ng on the individual who is enteri ng the data. For example, the MDS is entered by at least one nurse in each of the 16000 nursing homes in the United States.

PAGE 100

100 APPENDIX BACKGROUND TABLES Table A-1. Eigenvalues of scope (technology) factor analysis Eigenvalue Difference Proportion Cumulative 1 7.03 2.134 0.126 0.126 2 4.89 1.138 0.087 0.213 3 3.76 1.074 0.067 0.280 4 2.68 0.391 0.048 0.328 5 2.29 0.080 0.041 0.369 6 2.21 0.416 0.040 0.408 7 1.80 0.017 0.032 0.440 8 1.78 0.241 0.032 0.472 9 1.54 0.031 0.028 0.500 10 1.51 0.043 0.027 0.527 11 1.46 0.098 0.026 0.553 12 1.37 0.024 0.024 0.577 13 1.34 0.067 0.024 0.601 14 1.28 0.061 0.023 0.624 15 1.21 0.048 0.022 0.646 16 1.17 0.047 0.021 0.666 17 1.12 0.057 0.020 0.686 18 1.06 0.022 0.019 0.705 19 1.04 0.027 0.019 0.724 20 1.01 0.064 0.018 0.742 21 0.95 0.101 0.017 0.759 22 0.85 0.037 0.015 0.774 23 0.81 0.044 0.015 0.789 24 0.77 0.066 0.014 0.802 25 0.70 0.014 0.013 0.815 26 0.69 0.053 0.012 0.827 27 0.63 0.009 0.011 0.839 28 0.63 0.006 0.011 0.850 29 0.62 0.022 0.011 0.861 30 0.60 0.060 0.011 0.871 31 0.54 0.033 0.010 0.881 32 0.50 0.032 0.009 0.890 33 0.47 0.038 0.008 0.898 34 0.43 0.008 0.008 0.906 35 0.43 0.021 0.008 0.914 36 0.41 0.021 0.007 0.921 37 0.39 0.006 0.007 0.928 38 0.38 0.024 0.007 0.935 39 0.36 0.026 0.006 0.941 40 0.33 0.009 0.006 0.947 41 0.32 0.018 0.006 0.953 42 0.30 0.050 0.005 0.958 43 0.25 0.017 0.005 0.963 44 0.24 0.023 0.004 0.967 45 0.21 0.011 0.004 0.971 46 0.20 0.011 0.004 0.974 47 0.19 0.008 0.003 0.978 48 0.18 0.005 0.003 0.981 49 0.18 0.005 0.003 0.984 50 0.17 0.008 0.003 0.987

PAGE 101

101 Scree plot1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1920 21 22 23 24 25 26 27 2829 30 31 32 33 3435 36 37 38 39 404142 43 44 45464748 49505152 53 54 5556 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 01 02 03 04 05 06 0 Series1 Figure A-1. Scree plot of scope (technology) factor analysis

PAGE 102

Table A-2. Factor analysis with 10 f actors of scope (technology) variables Rotated Factor Pattern Factor1 Factor2 Factor3 Factor4 Factor5 Factor6 Factor7 Factor 8 Factor9 Factor10 Long term care plateguard -0.01 0.13 0.07 0.02 -0.04 0.01 0.05 0.02 0.00 0.76 Long term care Weight program -0.01 -0.01 0.04 0.03 0.02 -0.03 0.03 0.02 0.92 0.03 Long term care Therapy Diet 0.01 0.17 0.07 -0.03 0.03 0.05 0.13 -0.05 -0.02 0.70 Long term care Bladder training -0.03 0.07 0.08 -0.11 0.02 0.05 0.12 0.30 -0.05 -0.23 Long term care toilet program 0.07 -0.05 0.02 0.20 -0.06 -0.02 -0.06 0.79 0.05 0.22 Long term care no skin treatment 0.00 0.09 0.00 -0.71 -0.24 -0.05 0.00 -0.11 -0.01 -0.06 Long term care other skin tx -0.03 0.63 0.14 0.24 0.05 -0.05 0.01 0.08 0.00 0.22 Long term care ointment -0.06 0.70 0.18 0.02 0.13 0.02 0.06 -0.02 -0.02 0.16 Long term care pressure relief bed -0.05 0.57 0.18 0.59 -0.19 0.04 0.02 -0.11 -0.05 0.04 Long term care pressure relief chair 0.00 0.61 0.12 0.51 -0.20 -0.05 0.06 0.00 -0.02 0.08 Long term care dressing application -0.06 0.77 0.11 -0.07 0.16 0.13 0.05 -0.18 -0.01 0.06 Long term care surgical wound care 0.06 0.40 -0.09 -0.07 0.08 0.35 0.06 -0.07 0.01 0.01 Long term care ulcer care -0.03 0.72 0.07 -0.06 0.15 0.11 0.03 -0.20 0.00 0.01 Long term care skin nutrition program 0.02 0.60 0.08 0.04 0.21 -0.09 -0.08 0.12 0.09 -0.06 Long term care turning repositioning -0.02 0.69 0.14 0.18 0.03 0.04 0.00 0.19 -0.03 -0.03 Long term care OT minutes 0.03 0.34 -0.11 -0.05 -0.11 0.75 0.05 -0.09 0.10 -0.02 Long term care PT minutes 0.03 0.38 -0.10 -0.05 -0.10 0.74 0.04 -0.09 0.09 0.01 Long term care other restorative care -0.07 0.15 0.47 0.01 -0.04 0.02 0.21 0.08 -0.02 -0.17 Long term care restorative communication 0.17 0.01 0.39 0.04 0.04 -0.01 -0.13 -0.07 0.01 0.02 Long term care restorative amputation -0.04 -0.03 0.38 0.05 -0.03 -0.02 0.05 0.01 0.00 -0.19 Long term care restorative eating 0.17 0.12 0.72 0.00 0.01 0.06 0.05 -0.01 0.03 0.08 Long term care restorative dressing 0.30 0.11 0.74 -0.02 0.01 0.02 0.03 0.03 0.03 0.13 Long term care restorative transfers 0.31 0.14 0.73 0.00 0.00 0.01 0.06 0.01 0.00 0.09 Long term care restorative bed mobility 0.39 0.06 0.64 0.00 0.03 0.00 -0.03 0.01 0.03 -0.01 Long term care restorative splinting 0.02 -0.03 0.18 0.05 0.23 0.14 0.61 -0.03 0.00 -0.03 Long term care restorative Passive ROM 0.12 0.31 0.42 0.03 -0.03 -0.03 0.59 0.03 0.00 0.17 Long term care restorative Active ROM 0.10 0.28 0.47 0.01 -0.09 0.01 0.51 0.05 -0.01 0.17 Long term care restorative ambulation 0.05 0.33 0.62 0.02 -0.07 0.03 0.35 0.07 -0.02 0.22 Post acute plateguard 0.03 0.02 -0.01 0.06 -0.03 -0.14 0.12 0.12 0.10 0.24 Post acute Weight program -0.02 0.01 0.03 0.03 0.05 -0.03 0.03 0.01 0.93 0.02 Post acute Therapy Diet -0.01 -0.04 -0.04 0.04 0.10 -0.15 0.29 0.02 0.05 0.10 Post acute Bladder training -0.01 -0.04 0.06 -0.05 0.14 0.05 0.15 0.39 -0.07 -0.20 Post acute toilet program 0.11 -0.06 -0.05 0.18 -0.03 -0.07 -0.03 0.80 0.09 0.12 102

PAGE 103

Table A-2. Continued Post acute no skin treatment 0.01 -0.08 -0.04 -0.79 -0.31 -0.03 -0.02 -0.09 -0.02 -0.08 Post acute other skin tx 0.03 0.19 -0.01 0.32 0.31 -0.06 0.00 0.20 0.05 0.15 Post acute ointment -0.03 0.16 0.02 0.04 0.66 -0.02 0.12 0.03 0.02 0.00 Post acute pressure relief bed 0.01 0.04 0.03 0.84 0.01 0.05 0.03 -0.04 -0.01 -0.06 Post acute pressure relief chair 0.10 0.14 -0.04 0.65 -0.06 -0.09 0.06 0.13 0.06 -0.02 Post acute dressing application -0.04 0.07 0.01 0.03 0.77 0.06 0.07 -0.07 -0.03 0.09 Post acute surgical wound care 0.01 -0.06 0.00 0.12 0.42 0.15 -0.07 0.08 -0.12 0.27 Post acute ulcer care 0.00 0.07 -0.05 0.01 0.71 -0.03 0.15 -0.13 0.05 -0.13 Post acute skin nutrition program 0.06 0.27 -0.03 0.09 0.52 -0.13 0.00 0.17 0.14 -0.16 Post acute turning repositioning 0.07 0.07 -0.06 0.26 0.39 0.01 0.02 0.35 0.02 -0.18 Post acute OT minutes -0.14 -0.13 0.14 0.06 0.05 0.69 -0.10 0.06 -0.10 -0.02 Post acute PT minutes -0.10 -0.19 0.14 0.08 0.10 0.67 -0.08 0.11 -0.12 0.02 Post acute other restorative care 0.23 0.04 0.11 -0.02 -0.03 -0.02 0.18 0.07 -0.01 -0.19 Post acute restorative communication 0.49 -0.03 0.14 0.01 0.02 -0.01 -0.12 -0.07 0.00 -0.01 Post acute restorative amputation 0.31 -0.02 -0.05 0.03 0.04 0.01 0.12 -0.01 -0.01 0.00 Post acute restorative eating 0.65 -0.03 0.29 0.03 0.05 0.00 -0.01 0.02 0.01 0.02 Post acute restorative dressing 0.86 -0.01 0.26 0.00 -0.03 -0.01 -0.01 0.04 0.01 0.04 Post acute restorative transfers 0.90 -0.02 0.19 0.01 -0.03 -0.03 0.02 0.05 -0.01 0.02 Post acute restorative bed mobility 0.88 -0.03 0.15 0.01 0.00 -0.01 0.00 0.03 0.00 0.00 Post acute restorative splinting 0.30 -0.04 0.01 0.03 0.09 -0.03 0.52 0.05 0.01 0.01 Post acute restorative Passive ROM 0.58 -0.02 -0.04 0.00 0.03 -0.10 0.54 0.02 0.02 -0.03 Post acute restorative Active ROM 0.65 0.01 0.01 0.01 -0.07 -0.08 0.46 0.05 0.01 -0.03 Post acute restorative ambulation 0.79 0.02 0.10 0.01 -0.08 -0.08 0.15 0.10 -0.01 0.05 103 Values over 0.4 are highlighted and represent the initial variables used to define the factor

PAGE 104

Table A-3. Factor analysis with 10 f actors of scope (technology) variables Rotated Factor Pattern Post acute restorative LTC skin care LTC restorative Pressure devices Post acute skin care Therapy minutes Toilet program Weight program Long term care plateguard -0.01 0.13 0.07 0.02 -0.04 0.01 0.02 0.00 Long term care Weight program -0.01 -0.01 0.04 0.03 0.02 -0.03 0.02 0.92 Long term care Therapy Diet 0.01 0.17 0.07 -0.03 0.03 0.05 -0.05 -0.02 Long term care Bladder training -0.03 0.07 0.08 -0.11 0.02 0.05 0.30 -0.05 Long term care toilet program 0.07 -0.05 0.02 0.20 -0.06 -0.02 0.79 0.05 Long term care no skin treatment 0.00 0.09 0.00 -0.71 -0.24 -0.05 -0.11 -0.01 Long term care other skin tx -0.03 0.63 0.14 0.24 0.05 -0.05 0.08 0.00 Long term care ointment -0.06 0.70 0.18 0.02 0.13 0.02 -0.02 -0.02 Long term care pressure relief bed -0.05 0.57 0.18 0.59 -0.19 0.04 -0.11 -0.05 Long term care pressure relief chair 0.00 0.61 0.12 0.51 -0.20 -0.05 0.00 -0.02 Long term care dressing application -0.06 0.77 0.11 -0.07 0.16 0.13 -0.18 -0.01 Long term care surgical wound care 0.06 0.40 -0.09 -0.07 0.08 0.35 -0.07 0.01 Long term care ulcer care -0.03 0.72 0.07 -0.06 0.15 0.11 -0.20 0.00 Long term care skin nutrition program 0.02 0.60 0.08 0.04 0.21 -0.09 0.12 0.09 Long term care turning repositioning -0.02 0.69 0.14 0.18 0.03 0.04 0.19 -0.03 Long term care OT minutes 0.03 0.34 -0.11 -0.05 -0.11 0.75 -0.09 0.10 Long term care PT minutes 0.03 0.38 -0.10 -0.05 -0.10 0.74 -0.09 0.09 Long term care other restorative care -0.07 0.15 0.47 0.01 -0.04 0.02 0.08 -0.02 Long term care restorative communication 0.17 0.01 0.39 0.04 0.04 -0.01 -0.07 0.01 Long term care restorative amputation -0.04 -0.03 0.38 0.05 -0.03 -0.02 0.01 0.00 Long term care restorative eating 0.17 0.12 0.72 0.00 0.01 0.06 -0.01 0.03 Long term care restorative dressing 0.30 0.11 0.74 -0.02 0.01 0.02 0.03 0.03 Long term care restorative transfers 0.31 0.14 0.73 0.00 0.00 0.01 0.01 0.00 Long term care restorative bed mobility 0.39 0.06 0.64 0.00 0.03 0.00 0.01 0.03 Long term care restorative splinting 0.02 -0.03 0.18 0.05 0.23 0.14 -0.03 0.00 Long term care restorative Passive ROM 0.12 0.31 0.42 0.03 -0.03 -0.03 0.03 0.00 Long term care restorative Active ROM 0.10 0.28 0.47 0.01 -0.09 0.01 0.05 -0.01 Long term care restorative ambulation 0.05 0.33 0.62 0.02 -0.07 0.03 0.07 -0.02 Post acute plateguard 0.03 0.02 -0.01 0.06 -0.03 -0.14 0.12 0.10 Post acute Weight program -0.02 0.01 0.03 0.03 0.05 -0.03 0.01 0.93 Post acute Therapy Diet -0.01 -0.04 -0.04 0.04 0.10 -0.15 0.02 0.05 Post acute Bladder training -0.01 -0.04 0.06 -0.05 0.14 0.05 0.39 -0.07 Post acute toilet program 0.11 -0.06 -0.05 0.18 -0.03 -0.07 0.80 0.09 104

PAGE 105

105 Post acute restorative LTC skin care LTC restorative Pressure devices Post acute skin care Therapy minutes Toilet program Weight program Table A-3. Continued Post acute no skin treatment 0.01 -0.08 -0.04 -0.79 -0.31 -0.03 -0.09 -0.02 Post acute other skin tx 0.03 0.19 -0.01 0.32 0.31 -0.06 0.20 0.05 Post acute ointment -0.03 0.16 0.02 0.04 0.66 -0.02 0.03 0.02 Post acute pressure relief bed 0.01 0.04 0.03 0.84 0.01 0.05 -0.04 -0.01 Post acute pressure relief chair 0.10 0.14 -0.04 0.65 -0.06 -0.09 0.13 0.06 Post acute dressing application -0.04 0.07 0.01 0.03 0.77 0.06 -0.07 -0.03 Post acute surgical wound care 0.01 -0.06 0.00 0.12 0.42 0.15 0.08 -0.12 Post acute ulcer care 0.00 0.07 -0.05 0.01 0.71 -0.03 -0.13 0.05 Post acute skin nutrition program 0.06 0.27 -0.03 0.09 0.52 -0.13 0.17 0.14 Post acute turning repositioning 0.07 0.07 -0.06 0.26 0.39 0.01 0.35 0.02 Post acute OT minutes -0.14 -0.13 0.14 0.06 0.05 0.69 0.06 -0.10 Post acute PT minutes -0.10 -0.19 0.14 0.08 0.10 0.67 0.11 -0.12 Post acute other restorative care 0.23 0.04 0.11 -0.02 -0.03 -0.02 0.07 -0.01 Post acute restorative communication 0.49 -0.03 0.14 0.01 0.02 -0.01 -0.07 0.00 Post acute restorative amputation 0.31 -0.02 -0.05 0.03 0.04 0.01 -0.01 -0.01 Post acute restorative eating 0.65 -0.03 0.29 0.03 0.05 0.00 0.02 0.01 Post acute restorative dressing 0.86 -0.01 0.26 0.00 -0.03 -0.01 0.04 0.01 Post acute restorative transfers 0.90 -0.02 0.19 0.01 -0.03 -0.03 0.05 -0.01 Post acute restorative bed mobility 0.88 -0.03 0.15 0.01 0.00 -0.01 0.03 0.00 Post acute restorative splinting 0.30 -0.04 0.01 0.03 0.09 -0.03 0.05 0.01 Post acute restorative Passive ROM 0.58 -0.02 -0.04 0.00 0.03 -0.10 0.02 0.02 Post acute restorative Active ROM 0.65 0.01 0.01 0.01 -0.07 -0.08 0.05 0.01 Post acute restorative ambulation 0.79 0.02 0.10 0.01 -0.08 -0.08 0.10 -0.01 Values over 0.4 are highlighted and represent the initial variables used to define the fact

PAGE 106

Table A-4. Eigenvalues of scope (technology) factor analysis Eigenvalues of the Correlation Matrix: Eigenvalue Difference Proportion Cumulative 1 3.16 1.77 0.45 0.45 2 1.39 0.32 0.20 0.65 3 1.08 0.43 0.15 0.80 4 0.65 0.30 0.09 0.90 5 0.35 0.10 0.05 0.95 6 0.24 0.11 0.03 0.98 7 0.13 0.02 1 1 2 3 4 5 6 7 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 012345678 Series1 Figure A-2. Scree plot of resource sta ffing intensity from factor analysis Table A-5. Factor analysis of resource staffing intensity with three factors Total therapist Total therapy assistants Total nursing Total PT per Resident 0.94987 0.13329 0.09479 Total OT per Resident 0.93244 0.10457 0.13311 Total PTA per Resident -0.02059 0.92192 0.05157 Total OTA per Resident 0.39775 0.78854 0.21303 RN per Resident 0.0587 0.03395 0.79055 LPN per Resident 0.15186 0.56287 0.61385 CNA per Resident 0.13473 0.15559 0.86025 Values over 0.4 are highlighted and represent the initial variables used to define the factor 106

PAGE 107

107 Table A-6. Ward's minimum variance cluster analysis The CLUSTER Procedure Ward's Minimum Variance Cluster Analysis Eigenvalues of the Covariance Matrix Eigenvalue Difference Proportion Cumulative 1 6.24393174 0.63097545 0.2280 0.2280 2 5.61295630 1.53129364 0.2050 0.4330 3 4.08166266 0.58262014 0.1491 0.5821 4 3.49904251 1.67386173 0.1278 0.7099 5 1.82518078 0.40031900 0.0667 0.7766 6 1.42486178 0.27397392 0.0520 0.8286 7 1.15088786 0.14931403 0.0420 0.8706 8 1.00157383 0.45163154 0.0366 0.9072 9 0.54994229 0.12809678 0.0201 0.9273 10 0.42184551 0.01558279 0.0154 0.9427 11 0.40626272 0.06700601 0.0148 0.9575 12 0.33925671 0.10344911 0.0124 0.9699 13 0.23580760 0.08800114 0.0086 0.9785 14 0.14780646 0.02048422 0.0054 0.9839 15 0.12732224 0.02697285 0.0047 0.9886 16 0.10034939 0.01066020 0.0037 0.9923 17 0.08968920 0.02449037 0.0033 0.9955 18 0.06519883 0.03495529 0.0024 0.9979 19 0.03024354 0.01013037 0.0011 0.9990 20 0.02011316 0.01323966 0.0007 0.9997 21 0.00687351 0.0003 1.0000 Root-Mean-Square Total-Sample Standard Deviation = 1.141862 Root-Mean-Square Distance Between Observations = 7.400109 Table A-7. Cluster History NCL ------Clusters Joined-----FREQ SPRSQ RSQ ERSQ CCC PSF PST2 e 10 CL22 CL29 3 0.0250 .645 .605 2.70 13.1 4.3 9 CL12 CL13 37 0.0258 .619 .583 2.42 13.4 5.4 8 CL15 CL31 3 0.0286 .591 .558 2.13 13.8 5.2 7 CL17 CL11 7 0.0330 .558 .530 1.79 14.3 3.8 6 CL9 CL15 57 0.0466 .511 .497 0.78 14.4 9.1 5 CL22 CL23 5 0.0550 .456 .458 -.12 14.7 8.4 4 CL6 CL7 64 0.0887 .367 .400 -1.7 13.7 13.6 3 CL4 CL10 67 0.0942 .273 .320 -2.3 13.5 11.7 2 CL3 CL8 70 0.1260 .147 .188 -2.2 12.6 13.1 1 CL2 CL5 75 0.1472 .000 .000 0.00 12.

PAGE 108

Table A-8. Determination of rank scores table RN focus n=1634 Rehab Focus n=1816 Differentiator n=990 Cost leader n=2752 Lack of strategy n=2554 V LV TV R V LV TV R V LV TV R V LV TV R V LV TV R Zwages 0.15 -0.10 0.25 5.0 0.10 -0.10 0.20 4.0 0.00 -0.10 0.10 2.0 -0.06 -0.10 0.04 0.8 -0.10 -0.10 0.00 0.0 ZRN per res 1.16 -0.56 1.72 5.0 0.38 -0.56 0.94 2.7 0.02 -0.56 0.58 1.7 -0.56 -0.56 0.00 0.0 -0.42 -0.56 0.14 0.4 ZLPN per res -0.26 -0.26 0.00 0.0 0.57 -0.26 0.83 5.0 -0.10 -0.26 0.16 1.0 -0.08 -0.26 0.18 1.1 -0.12 -0.26 0.14 0.9 ZCNA per res -1.02 -1.02 0.00 0.0 0.46 -1.02 1.48 5.0 0.27 -1.02 1.29 4.4 0.04 -1.02 1.06 3.6 0.18 -1.02 1.20 4.0 Ztherapy -0.07 -0.81 0.74 1.8 1.31 -0.81 2.12 5.0 -0.39 -0.81 0.42 1.0 -0.81 -0.81 0.00 0.0 0.14 -0.81 0.95 2.2 Ztherapy assist -0.19 -0.91 0.72 1.6 1.33 -0.91 2.24 5.0 -0.38 -0.91 0.53 1.2 0.20 -0.91 1.11 2.5 -0.91 -0.91 0.00 0.0 ZPT percent 0.06 -0.79 0.85 2.4 -0.24 -0.79 0.55 1.6 0.10 -0.79 0.89 2.5 -0.79 -0.79 0.00 0.0 0.95 -0.79 1.74 5.0 ZOT percent 0.07 -0.76 0.83 2.6 -0.14 -0.76 0.62 1.9 0.07 -0.76 0.83 2.6 -0.76 -0.76 0.00 0.0 0.84 -0.76 1.60 5.0 ZRN percent 1.67 -0.58 2.25 5.0 0.06 -0.58 0.64 1.4 -0.06 -0.58 0.52 1.2 -0.58 -0.58 0.00 0.0 -0.47 -0.58 0.11 0.3 ZLicense perc. 1.47 -0.48 1.95 5.0 0.21 -0.48 0.69 1.8 -0.32 -0.48 0.16 0.4 -0.45 -0.48 0.03 0.1 -0.48 -0.48 0.00 0.0 Total Resource 28 34 18 8 18 zltcskinpr~s 0.05 -0.31 0.36 2.1 0.56 -0.31 0.87 5.0 0.25 -0.31 0.56 3.2 -0.20 -0.31 0.11 0.6 -0.31 -0.31 0.00 0.0 Zrestorative 14d -0.21 -0.43 0.22 0.3 -0.43 -0.43 0.00 0.0 2.93 -0.43 3.36 5.0 -0.34 -0.43 0.09 0.1 -0.34 -0.43 0.09 0.1 Zrestorative Q -0.17 -0.31 0.14 0.3 0.05 -0.31 0.36 0.9 1.73 -0.31 2.04 5.0 -0.27 -0.31 0.04 0.1 -0.31 -0.31 0.00 0.0 Zrelief device -0.05 -0.17 0.12 1.5 0.24 -0.17 0.41 5.0 0.17 -0.17 0.34 4.1 -0.03 -0.17 0.14 1.7 -0.17 -0.17 0.00 0.0 Ztherapy min -0.08 -0.57 0.49 1.7 0.88 -0.57 1.45 5.0 -0.57 -0.57 0.00 0.0 -0.11 -0.57 0.46 1.6 -0.24 -0.57 0.33 1.1 Zacute skin care 0.07 -0.23 0.30 3.0 0.26 -0.23 0.49 5.0 0.23 -0.23 0.46 4.7 -0.08 -0.23 0.15 1.6 -0.23 -0.23 0.00 0.0 Ztoilet program -0.06 -0.07 0.01 0.1 -0.06 -0.07 0.01 0.1 0.29 -0.07 0.36 5.0 0.04 -0.07 0.11 1.5 -0.07 -0.07 0.00 0.0 Zweight prog. 0.01 -0.10 0.11 2.5 -0.10 -0.10 0.00 0.0 0.12 -0.10 0.22 5.0 0.06 -0.10 0.16 3.7 -0.05 -0.10 0.05 1.0 Zpercent Mcare -0.13 -0.39 0.26 0.7 1.37 -0.39 1.76 5.0 -0.36 -0.39 0.03 0.0 -0.34 -0.39 0.05 0.2 -0.39 -0.39 0.00 0.1 Total scope 12 26 32 11 2.3108 V is the actual Z score value LV is the lowest value for the variable among clusters TV is the total variance is the difference between the highest value among the clusters and the lowest values among clusters R is the final rank from 0 to 5. Determined ((V-LV)/TV*5)

PAGE 109

Table A-9. Descriptives of dependent variables, independent variables and control variables Cluster 1 Differentiator N=990 Cluster 2 Cost Leader N=2752 Cluster 3 Rehab focus N=1816 Cluster 4 Specialty care N=1634 Cluster 5 Lack of strategy N=2534 Avg Wages 12.61 12.52 12.99 12.76 12.45 RN per Resident 0.06 0.04 0.07 0.08 0.05 LPN per Resident 0.14 0.14 0.17 0.14 0.14 CNA per Resident 0.44 0.42 0.47 0.31 0.44 Therapist per Res. 0.02 0.01 0.04 0.02 0.02 Therapist asst per Res. 0.01 0.02 0.03 0.01 0.01 Percent Licensed PT 0.62 0.39 0.54 0.6 0.84 Percent Licensed OT 0.65 0.41 0.6 0.64 0.86 Percent RN 0.09 0.07 0.1 0.17 0.08 LTC Skin Process 0.83 0.72 0.87 0.78 0.69 Acute Skin Process 0.91 0.83 0.91 0.86 0.81 Relief device 1.95 1.8 1.98 1.79 1.72 LTC Restorative Process 4.28 1.8 2.03 1.88 1.67 Acute Restorative Process 3.69 0.47 0.39 0.58 0.44 Therapy minutes 324.7 357.8 439.6 357.3 349.6 Toilet Program 0.68 0.59 0.58 0.57 0.57 Weight Program 0.34 0.32 0.28 0.31 0.29 Percent Medicare 10.7 10.82 22.2 11.88 10.47 Percent Other 23.39 21.7 26.56 24.68 22.65 Percent Medicaid 65.91 67.48 51.24 63.45 66.88 Acuindex 10.39 10.2 10.45 10.25 10.17 ADL index 10.13 10.02 10.23 10.03 9.98 Percent tube feed 0.07 0.06 0.06 0.06 0.06 Percent tracheotomy 0.01 0.01 0.01 0.01 0 Percent requiring suctioning 0.01 0.01 0.01 0.01 0.01 Percent requiring respiratory care 0.11 0.1 0.12 0.11 0.1 Percent requiring IV 0.01 0.01 0.02 0.01 0.01 Percent requiring injection 0.15 0.15 0.17 0.16 0.15 Quality care deficiency 1.782 1.928 1.749 1.777 1.929 Bladder decline 0.186 0.193 0.209 0.192 0.187 Bowel decline 0.169 0.176 0.188 0.176 0.168 ADL 4 point decline 0.165 0.172 0.17 0.169 0.17 Restraint 0.076 0.079 0.075 0.076 0.077 Long term care pressure ulcer 0.08 0.076 0.093 0.081 0.075 Post acute walking improvement 0.232 0.238 0.246 0.24 0.246 Post acute pressure ulcer 0.187 0.173 0.177 0.174 0.169 Ownership 0.71 0.76 0.8 0.77 0.76 Market Competition 0.13 0.14 0.14 0.16 0.14 Chain 0.52 0.6 0.66 0.63 0.58 Total beds 114 130 116 116 111 109

PAGE 110

Table A-10. Negative binomial of quali ty of care deficiencies with di fferentiators as reference group IRR Std. Err. z P>z Cost leaders 1.059 0.036 1.72 0.09 Rehab focus 0.993 0.038 -0.18 0.86 Specialty care focus 1.001 0.037 0.04 0.97 Lack of strategy 1.124 0.038 3.43 0.00 Ownership 1.099 0.029 3.59 0.00 Percent Medicaid 1.005 0.001 7.69 0.00 Percent Medicare 1.005 0.001 3.75 0.00 Market competition 0.773 0.367 -0.54 0.59 Occupancy 0.830 0.065 -2.38 0.02 Total beds 1.003 0.000 14.74 0.00 Chain 1.024 0.021 1.18 0.24 Acuity index 1.040 0.013 3.20 0.00 ADL index 0.952 0.029 -1.63 0.10 Tube feed care 1.009 0.237 0.04 0.97 Respiratory care 0.972 0.157 -0.18 0.86 Suctioning care 0.811 0.454 -0.37 0.71 IV therapy care 3.826 1.304 3.94 0.00 Tracheostomy care 1.142 0.745 0.20 0.84 Need injection 1.189 0.141 1.46 0.14 _Ifips_2 1.206 0.470 0.48 0.63 Table A-11. OLS regressions of root transformed bladder decline with differentiators as ref. group Coef. Std. Err. t P>t Cost leaders 0.006 0.003 2.03 0.04 Rehab focus 0.012 0.004 3.27 0.00 Specialty care focus 0.009 0.004 2.68 0.01 Lack of strategy 0.005 0.003 1.56 0.12 Ownership 0.006 0.002 2.46 0.01 Percent Medicaid 0.000 0.000 -5.35 0.00 Percent Medicare 0.001 0.000 3.84 0.00 Market competition 0.077 0.064 1.21 0.23 Occupancy 0.014 0.008 1.70 0.09 Total beds 0.000 0.000 3.66 0.00 Chain 0.004 0.002 1.97 0.05 Acuity index 0.009 0.001 6.66 0.00 ADL index 0.021 0.003 6.68 0.00 Tube feed care -0.061 0.027 -2.24 0.03 Respiratory care -0.040 0.014 -2.99 0.00 Suctioning care 0.097 0.067 1.44 0.15 IV therapy care 0.004 0.040 0.09 0.93 Tracheostomy care -0.245 0.074 -3.32 0.00 Need injection??? 0.004 0.012 0.33 0.74 _cons 0.241 0.017 13.91 0.00 110

PAGE 111

Table A-12. OLS regressions of root transformed bowel decline with differentiators as ref. group Coef. Std. Err. t P>t Cost leaders 0.008 0.004 2.31 0.02 Rehab focus 0.009 0.004 2.22 0.03 Specialty care focus 0.012 0.004 2.92 0.00 Lack of strategy 0.005 0.004 1.41 0.16 Ownership 0.005 0.003 1.83 0.07 Percent Medicaid 0.000 0.000 -3.76 0.00 Percent Medicare 0.001 0.000 3.82 0.00 Market competition 0.040 0.050 0.79 0.43 Occupancy 0.013 0.009 1.49 0.14 Total beds 0.000 0.000 2.94 0.00 Chain 0.004 0.002 1.79 0.07 Acuity index 0.011 0.001 7.26 0.00 ADL index 0.025 0.003 7.43 0.00 Tube feed care -0.039 0.029 -1.34 0.18 Respiratory care -0.040 0.015 -2.74 0.01 Suctioning care 0.014 0.077 0.18 0.86 IV therapy care 0.016 0.044 0.36 0.72 Tracheostomy care -0.101 0.085 -1.19 0.23 Need injection 0.020 0.013 1.55 0.12 _cons 0.176 0.017 10.11 0.00 Table A-13. OLS regressions of r oot transformed ADL decline with differentiators as ref. group Coef. Std. Err. t P>t Cost leaders 0.008 0.003 2.94 0.00 Rehab focus 0.003 0.003 0.86 0.39 Specialty care focus 0.005 0.003 1.67 0.10 Lack of strategy 0.004 0.003 1.50 0.13 Ownership 0.001 0.002 0.50 0.62 Percent Medicaid 0.000 0.000 0.10 0.92 Percent Medicare 0.000 0.000 2.97 0.00 Market competition 0.083 0.053 1.57 0.12 Occupancy 0.006 0.007 0.84 0.40 Total beds 0.000 0.000 -0.71 0.48 Chain -0.001 0.002 -0.72 0.47 Acuity index 0.002 0.001 1.46 0.14 ADL index -0.003 0.003 -1.01 0.31 Tube feed care -0.028 0.026 -1.10 0.27 Respiratory care -0.005 0.013 -0.42 0.67 Suctioning care -0.054 0.057 -0.95 0.34 IV therapy care -0.013 0.033 -0.39 0.70 Tracheostomy care -0.072 0.066 -1.10 0.27 Need injection -0.008 0.011 -0.69 0.49 _cons 0.374 0.016 23.87 0.00 111

PAGE 112

Table A-14. OLS regressions of l og transformed restraints with differentiators as ref. group Coef. Std. Err. t P>t Cost leaders 0.011 0.004 2.48 0.01 Rehab focus 0.004 0.006 0.74 0.46 Specialty care focus 0.006 0.005 1.32 0.19 Lack of strategy 0.006 0.005 1.24 0.22 Ownership 0.020 0.004 5.50 0.00 Percent Medicaid 0.000 0.000 4.06 0.00 Percent Medicare 0.000 0.000 1.16 0.25 Market competition 0.053 0.070 0.76 0.45 Cost leaders -0.010 0.012 -0.84 0.40 Total beds 0.000 0.000 7.90 0.00 Chain -0.007 0.003 -2.24 0.03 Acuity index 0.011 0.002 6.07 0.00 ADL index 0.016 0.004 3.80 0.00 Tube feed care -0.049 0.041 -1.20 0.23 Respiratory care -0.012 0.022 -0.54 0.59 Suctioning care -0.138 0.101 -1.37 0.17 IV therapy care 0.028 0.054 0.51 0.61 Tracheostomy care 0.060 0.110 0.54 0.59 Need injection -0.055 0.018 -3.12 0.00 _cons -0.004 0.023 -0.19 0.85 Table A-15. OLS regressions of r oot transformed LTC ulcers with differentiators as ref. group Coef. Std. Err. t P>t Cost leaders -0.001 0.002 -0.45 0.66 Rehab focus 0.012 0.003 4.19 0.00 Specialty care focus 0.004 0.003 1.47 0.14 Lack of strategy -0.004 0.002 -1.71 0.09 Ownership 0.006 0.002 3.32 0.00 Percent Medicaid 0.000 0.000 -0.71 0.48 Percent Medicare 0.001 0.000 6.94 0.00 Market competition -0.001 0.029 -0.04 0.97 Occupancy -0.035 0.006 -5.55 0.00 Total beds 0.000 0.000 3.31 0.00 Chain -0.002 0.002 -1.57 0.12 Acuity index 0.008 0.001 8.42 0.00 ADL index 0.007 0.002 3.11 0.00 Tube feed care 0.109 0.024 4.6 0.00 Respiratory care 0.031 0.011 2.74 0.01 Suctioning care 0.045 0.060 0.74 0.46 IV therapy care 0.150 0.032 4.7 0.00 Tracheostomy care -0.031 0.065 -0.47 0.64 Need injection 0.046 0.010 4.74 0.00 _cons 0.151 0.012 12.84 0.00 112

PAGE 113

Table A-16. GLM with gamma distribu tion with PA walking improve with differentiators as ref. group Coef. Std. Err. z P>z Cost leaders 0.031 0.019 1.65 0.10 Rehab focus 0.050 0.022 2.32 0.02 Specialty care focus 0.037 0.021 1.76 0.08 Lack of strategy 0.060 0.019 3.17 0.00 Ownership -0.087 0.014 -6.04 0.00 Percent Medicaid -0.004 0.000 -10.18 0.00 Percent Medicare -0.002 0.001 -2.81 0.01 Market competition -0.425 0.248 -1.71 0.09 Occupancy -0.160 0.043 -3.73 0.00 Total beds 0.000 0.000 -0.53 0.60 Chain -0.032 0.012 -2.74 0.01 Acuity index -0.028 0.007 -4.07 0.00 ADL index -0.110 0.017 -6.49 0.00 Tube feed care -0.666 0.143 -4.66 0.00 Respiratory care 0.139 0.081 1.71 0.09 Suctioning care 0.444 0.419 1.06 0.29 IV therapy care 0.240 0.213 1.13 0.26 Tracheostomy care -0.238 0.449 -0.53 0.60 Need injection 0.038 0.068 0.56 0.58 _cons 0.38 0.283 0.77 0.44 Table A-17. OLS regressions of root transformed PA skin ulcers w ith differentiators as ref. group Coef. Std. Err. t P>t Cost leaders -0.045 0.020 -2.30 0.02 Rehab focus -0.015 0.021 -0.74 0.46 Specialty care focus -0.027 0.021 -1.28 0.20 Lack of strategy -0.076 0.020 -3.81 0.00 Ownership 0.034 0.014 2.38 0.02 Percent Medicaid 0.001 0.000 3.39 0.00 Percent Medicare 0.001 0.001 0.70 0.48 Market competition -0.031 0.308 -0.10 0.92 Occupancy 0.014 0.046 0.31 0.75 Total beds 0.000 0.000 5.08 0.00 Chain -0.021 0.012 -1.76 0.08 Acuity index 0.034 0.007 4.88 0.00 ADL index 0.027 0.019 1.42 0.16 Tube feed care 0.568 0.156 3.64 0.00 Respiratory care 0.088 0.082 1.07 0.28 Suctioning care 0.535 0.413 1.30 0.20 IV therapy care 0.270 0.194 1.40 0.16 Tracheostomy care -0.160 0.431 -0.37 0.71 Need injection 0.183 0.069 2.63 0.01 _cons -2.513 0.098 -25.77 0.00 113

PAGE 114

Table A-18. Negative binomial of qua lity of care deficiencies with rehab focus as reference group IRR Std. Err. z P>z Cost leader 1.067 0.032 2.14 0.03 Differentiator 1.007 0.038 0.18 0.86 Specialty care focus 1.008 0.033 0.25 0.80 Lack of strategy 1.132 0.035 4.03 0.00 Ownership 1.099 0.029 3.59 0.00 Percent Medicaid 1.005 0.001 7.69 0.00 Percent Medicare 1.005 0.001 3.75 0.00 Market competition 0.773 0.367 -0.54 0.59 Occupancy 0.830 0.065 -2.38 0.02 Total beds 1.003 0.000 14.74 0.00 Chain 1.024 0.021 1.18 0.24 Acuity index 1.040 0.013 3.20 0.00 ADL index 0.952 0.029 -1.63 0.10 Tube feed care 1.009 0.237 0.04 0.97 Respiratory care 0.972 0.157 -0.18 0.86 Suctioning care 0.811 0.454 -0.37 0.71 IV therapy care 3.826 1.304 3.94 0.00 Tracheostomy care 1.142 0.745 0.20 0.84 Need injection 1.189 0.141 1.46 0.14 _Ifips_2 1.206 0.470 0.48 0.63 _Ifips_3 1.025 0.198 0.13 0.90 Table A-19. OLS regressions of r oot transformed bladder decline with rehab focus as ref. group Coef. Std. Err. t P>t Cost leader -0.006 0.003 -1.97 0.05 Differentiator -0.012 0.004 -3.27 0.00 Specialty care focus -0.003 0.003 -0.86 0.39 Lack of strategy -0.007 0.003 -2.32 0.02 Ownership 0.006 0.002 2.46 0.01 Percent Medicaid 0.000 0.000 -5.35 0.00 Percent Medicare 0.001 0.000 3.84 0.00 Market competition 0.077 0.064 1.21 0.23 Occupancy 0.014 0.008 1.70 0.09 Total beds 0.000 0.000 3.66 0.00 Chain 0.004 0.002 1.97 0.05 Acuity index 0.009 0.001 6.66 0.00 ADL index 0.021 0.003 6.68 0.00 Tube feed care -0.061 0.027 -2.24 0.03 Respiratory care -0.040 0.014 -2.99 0.00 Suctioning care 0.097 0.067 1.44 0.15 IV therapy care 0.004 0.040 0.09 0.93 Tracheostomy care -0.245 0.074 -3.32 0.00 Need injection 0.004 0.012 0.33 0.74 _cons 0.253 0.017 14.72 0.00 114

PAGE 115

Table A-20. OLS regressions of r oot transformed bowel decline with rehab focus as ref. group Coef. Std. Err. t P>t Cost leader -0.001 0.003 -0.29 0.77 Differentiator -0.009 0.004 -2.22 0.03 Specialty care focus 0.002 0.003 0.64 0.52 Lack of strategy -0.004 0.003 -1.26 0.21 Ownership 0.005 0.003 1.83 0.07 Percent Medicaid 0.000 0.000 -3.76 0.00 Percent Medicare 0.001 0.000 3.82 0.00 Market competition 0.040 0.050 0.79 0.43 Occupancy 0.013 0.009 1.49 0.14 Total beds 0.000 0.000 2.94 0.00 Chain 0.004 0.002 1.79 0.07 Acuity index 0.011 0.001 7.26 0.00 ADL index 0.025 0.003 7.43 0.00 Tube feed care -0.039 0.029 -1.34 0.18 Respiratory care -0.040 0.015 -2.74 0.01 Suctioning care 0.014 0.077 0.18 0.86 IV therapy care 0.016 0.044 0.36 0.72 Tracheostomy care -0.101 0.085 -1.19 0.23 Need injection 0.020 0.013 1.55 0.12 _cons 0.186 0.017 10.67 0.00 Table A-21. OLS regressions of r oot transformed ADL decline with rehab focus as ref. group Coef. Std. Err. t P>t Cost leader 0.006 0.003 2.18 0.03 Differentiator -0.003 0.003 -0.86 0.39 Specialty care focus 0.002 0.003 0.86 0.39 Lack of strategy 0.002 0.003 0.63 0.53 Ownership 0.001 0.002 0.50 0.62 Percent Medicaid 0.000 0.000 0.10 0.92 Percent Medicare 0.000 0.000 2.97 0.00 Market competition 0.083 0.053 1.57 0.12 Occupancy 0.006 0.007 0.84 0.40 Total beds 0.000 0.000 -0.71 0.48 Chain -0.001 0.002 -0.72 0.47 Acuity index 0.002 0.001 1.46 0.14 ADL index -0.003 0.003 -1.01 0.31 Tube feed care -0.028 0.026 -1.10 0.27 Respiratory care -0.005 0.013 -0.42 0.67 Suctioning care -0.054 0.057 -0.95 0.34 IV therapy care -0.013 0.033 -0.39 0.70 Tracheostomy care -0.072 0.066 -1.10 0.27 Need injection -0.008 0.011 -0.69 0.49 _cons 0.377 0.016 23.89 0.00 115

PAGE 116

Table A-22. OLS regressions of log transformed restraint with rehab focus as ref. group Coef. Std. Err. t P>t Cost leader 0.011 0.004 2.48 0.01 Differentiator 0.004 0.006 0.74 0.46 Specialty care focus 0.006 0.005 1.32 0.19 Lack of strategy 0.006 0.005 1.24 0.22 Ownership 0.020 0.004 5.50 0.00 Percent Medicaid 0.000 0.000 4.06 0.00 Percent Medicare 0.000 0.000 1.16 0.25 Market competition 0.053 0.070 0.76 0.45 Occupancy -0.010 0.012 -0.84 0.40 Total beds 0.000 0.000 7.90 0.00 Chain -0.007 0.003 -2.24 0.03 Acuity index 0.011 0.002 6.07 0.00 ADL index 0.016 0.004 3.80 0.00 Tube feed care -0.049 0.041 -1.20 0.23 Respiratory care -0.012 0.022 -0.54 0.59 Suctioning care -0.138 0.101 -1.37 0.17 IV therapy care 0.028 0.054 0.51 0.61 Tracheostomy care 0.060 0.110 0.54 0.59 Need injection -0.055 0.018 -3.12 0.00 _cons -0.004 0.023 -0.19 0.85 Table A-23. OLS regressions of r oot transformed skin ulcer with rehab focus as ref. group Coef. Std. Err. t P>t Cost leader -0.013 0.002 -5.69 0.00 Differentiator -0.012 0.003 -4.19 0.00 Specialty care focus -0.008 0.002 -3.34 0.00 Lack of strategy -0.016 0.002 -6.89 0.00 Ownership 0.006 0.002 3.32 0.00 Percent Medicaid 0.000 0.000 -0.71 0.48 Percent Medicare 0.001 0.000 6.94 0.00 Market competition -0.001 0.029 -0.04 0.97 Occupancy -0.035 0.006 -5.55 0.00 Total beds 0.000 0.000 3.31 0.00 Chain -0.002 0.002 -1.57 0.12 Acuity index 0.008 0.001 8.42 0.00 ADL index 0.007 0.002 3.11 0.00 Tube feed care 0.109 0.024 4.60 0.00 Respiratory care 0.031 0.011 2.74 0.01 Suctioning care 0.045 0.060 0.74 0.46 IV therapy care 0.150 0.032 4.70 0.00 Tracheostomy care -0.031 0.065 -0.47 0.64 Need injection 0.046 0.010 4.74 0.00 _cons 0.163 0.012 13.90 0.00 116

PAGE 117

Table A-24. GLM of PA walking improve with gamma distribution with rehab focus as ref. group Coef. Std. Err. z P>z Cost leader -0.019 0.018 -1.09 0.28 Differentiator -0.050 0.022 -2.32 0.02 Specialty care focus -0.014 0.019 -0.73 0.47 Lack of strategy 0.010 0.018 0.57 0.57 Ownership -0.087 0.014 -6.04 0.00 Percent Medicaid -0.004 0.000 -10.18 0.00 Percent Medicare -0.002 0.001 -2.81 0.01 Market competition -0.425 0.248 -1.71 0.09 Occupancy -0.160 0.043 -3.73 0.00 Total beds 0.000 0.000 -0.53 0.60 Chain -0.032 0.012 -2.74 0.01 Acuity index -0.028 0.007 -4.07 0.00 ADL index -0.110 0.017 -6.49 0.00 Tube feed care -0.666 0.143 -4.66 0.00 Respiratory care 0.139 0.081 1.71 0.09 Suctioning care 0.444 0.419 1.06 0.29 IV therapy care 0.240 0.213 1.13 0.26 Tracheostomy care -0.238 0.449 -0.53 0.60 Need injection 0.038 0.068 0.56 0.58 _Ifips_2 0.218 0.283 0.77 0.44 _Ifips_3 0.209 0.188 1.11 0.27 Table A-25. OLS regressions of root transformed PA skin ulcers with rehab focus as ref. group Coef. Std. Err. t P>t Cost leader -0.030 0.016 -1.91 0.06 Differentiator 0.015 0.021 0.74 0.46 Specialty care focus -0.012 0.017 -0.69 0.49 Lack of strategy -0.061 0.016 -3.80 0.00 Ownership 0.034 0.014 2.38 0.02 Percent Medicaid 0.001 0.000 3.39 0.00 Percent Medicare 0.001 0.001 0.70 0.48 Market competition -0.031 0.308 -0.10 0.92 Occupancy 0.014 0.046 0.31 0.75 Total beds 0.000 0.000 5.08 0.00 Chain -0.021 0.012 -1.76 0.08 Acuity index 0.034 0.007 4.88 0.00 ADL index 0.027 0.019 1.42 0.16 Tube feed care 0.568 0.156 3.64 0.00 Respiratory care 0.088 0.082 1.07 0.28 Suctioning care 0.535 0.413 1.30 0.20 IV therapy care 0.270 0.194 1.40 0.16 Tracheotomy care -0.160 0.431 -0.37 0.71 Need injection 0.183 0.069 2.63 0.01 _cons -2.528 0.097 -26.06 0.00 117

PAGE 118

Table A-26. Negative binomial of qua lity of care deficiencies with specialty care as reference group IRR Std. Err. z P>z Cost leader 1.058 0.030 2.00 0.05 Rehab focus 0.992 0.033 -0.25 0.80 Differentiator 0.999 0.037 -0.04 0.97 Lack of strategy 1.123 0.032 4.02 0.00 Ownership 1.099 0.029 3.59 0.00 Percent Medicaid 1.005 0.001 7.69 0.00 Percent Medicare 1.005 0.001 3.75 0.00 Market competition 0.773 0.367 -0.54 0.59 Occupancy 0.830 0.065 -2.38 0.02 Total beds 1.003 0.000 14.74 0.00 Chain 1.024 0.021 1.18 0.24 Acuity index 1.040 0.013 3.20 0.00 ADL index 0.952 0.029 -1.63 0.10 Tube feed care 1.009 0.237 0.04 0.97 Respiratory care 0.972 0.157 -0.18 0.86 Suctioning care 0.811 0.454 -0.37 0.71 IV therapy care 3.826 1.304 3.94 0.00 Tracheostomy care 1.142 0.745 0.20 0.84 Need injection 1.189 0.141 1.46 0.14 _Ifips_2 1.206 0.470 0.48 0.63 Table A-27. OLS regressions of root transformed bladder decline with specialty care focus as ref. group Coef. Std. Err. t P>t Cost leader -0.003 0.003 -1.13 0.26 Rehab focus 0.003 0.003 0.86 0.39 Differentiator -0.009 0.004 -2.68 0.01 Lack of strategy -0.004 0.003 -1.55 0.12 Ownership 0.006 0.002 2.46 0.01 Percent Medicaid 0.000 0.000 -5.35 0.00 Percent Medicare 0.001 0.000 3.84 0.00 Market competition 0.077 0.064 1.21 0.23 Occupancy 0.014 0.008 1.70 0.09 Total beds 0.000 0.000 3.66 0.00 Chain 0.004 0.002 1.97 0.05 Acuity index 0.009 0.001 6.66 0.00 ADL index 0.021 0.003 6.68 0.00 Tube feed care -0.061 0.027 -2.24 0.03 Respiratory care -0.040 0.014 -2.99 0.00 Suctioning care 0.097 0.067 1.44 0.15 IV therapy care 0.004 0.040 0.09 0.93 Tracheostomy care -0.245 0.074 -3.32 0.00 Need injection 0.004 0.012 0.33 0.74 _cons 0.251 0.017 14.71 0.00 118

PAGE 119

Table A-28. OLS regressions of root transformed bowel decline with specialty care as ref. group Coef. Std. Err. t P>t Cost leader -0.003 0.003 -1.10 0.27 Rehab focus -0.002 0.003 -0.64 0.52 Differentiator -0.012 0.004 -2.92 0.00 Lack of strategy -0.006 0.003 -2.18 0.03 Ownership 0.005 0.003 1.83 0.07 Percent Medicaid 0.000 0.000 -3.76 0.00 Percent Medicare 0.001 0.000 3.82 0.00 Market competition 0.040 0.050 0.79 0.43 Occupancy 0.013 0.009 1.49 0.14 Total beds 0.000 0.000 2.94 0.00 Chain 0.004 0.002 1.79 0.07 Acuity index 0.011 0.001 7.26 0.00 ADL index 0.025 0.003 7.43 0.00 Tube feed care -0.039 0.029 -1.34 0.18 Respiratory care -0.040 0.015 -2.74 0.01 Suctioning care 0.014 0.077 0.18 0.86 IV therapy care 0.016 0.044 0.36 0.72 Tracheostomy care -0.101 0.085 -1.19 0.23 Need injection 0.020 0.013 1.55 0.12 _cons 0.188 0.017 10.99 0.00 Table A-29. OLS regressions of r oot transformed ADL decline with specialty care as ref. group Coef. Std. Err. t P>t Cost leader 0.003 0.002 1.39 0.16 Rehab focus -0.002 0.003 -0.86 0.39 Differentiator -0.005 0.003 -1.67 0.10 Lack of strategy -0.001 0.002 -0.28 0.78 Ownership 0.001 0.002 0.50 0.62 Percent Medicaid 0.000 0.000 0.10 0.92 Percent Medicare 0.000 0.000 2.97 0.00 Market competition 0.083 0.053 1.57 0.12 Occupancy 0.006 0.007 0.84 0.40 Total beds 0.000 0.000 -0.71 0.48 Chain -0.001 0.002 -0.72 0.47 Acuity index 0.002 0.001 1.46 0.14 ADL index -0.003 0.003 -1.01 0.31 Tube feed care -0.028 0.026 -1.10 0.27 Respiratory care -0.005 0.013 -0.42 0.67 Suctioning care -0.054 0.057 -0.95 0.34 IV therapy care -0.013 0.033 -0.39 0.70 Tracheostomy care -0.072 0.066 -1.10 0.27 Need injection -0.008 0.011 -0.69 0.49 _cons 0.379 0.016 24.46 0.00 119

PAGE 120

Table A-30. OLS regressions of r oot transformed restraint with specialty care as ref. group Coef. Std. Err. t P>t Cost leader 0.005 0.004 1.22 0.22 Rehab focus -0.006 0.005 -1.32 0.19 Differentiator -0.002 0.005 -0.41 0.68 Lack of strategy -0.001 0.004 -0.15 0.88 Ownership 0.020 0.004 5.50 0.00 Percent Medicaid 0.000 0.000 4.06 0.00 Percent Medicare 0.000 0.000 1.16 0.25 Market competition 0.053 0.070 0.76 0.45 Occupancy -0.010 0.012 -0.84 0.40 Total beds 0.000 0.000 7.90 0.00 Chain -0.007 0.003 -2.24 0.03 Acuity index 0.011 0.002 6.07 0.00 ADL index 0.016 0.004 3.80 0.00 Tube feed care -0.049 0.041 -1.20 0.23 Respiratory care -0.012 0.022 -0.54 0.59 Suctioning care -0.138 0.101 -1.37 0.17 IV therapy care 0.028 0.054 0.51 0.61 Tracheostomy care 0.060 0.110 0.54 0.59 Need injection -0.055 0.018 -3.12 0.00 _cons 0.002 0.023 0.08 0.93 Table A-31. OLS regressions of root transformed LTC skin ulcer with specialty care as ref. group Coef. Std. Err. t P>t Cost leader -0.005 0.002 -2.50 0.01 Rehab focus 0.008 0.002 3.34 0.00 Differentiator -0.004 0.003 -1.47 0.14 Lack of strategy -0.008 0.002 -3.97 0.00 Ownership 0.006 0.002 3.32 0.00 Percent Medicaid 0.000 0.000 -0.71 0.48 Percent Medicare 0.001 0.000 6.94 0.00 Market competition -0.001 0.029 -0.04 0.97 Occupancy -0.035 0.006 -5.55 0.00 Total beds 0.000 0.000 3.31 0.00 Chain -0.002 0.002 -1.57 0.12 Acuity index 0.008 0.001 8.42 0.00 ADL index 0.007 0.002 3.11 0.00 Tube feed care 0.109 0.024 4.60 0.00 Respiratory care 0.031 0.011 2.74 0.01 Suctioning care 0.045 0.060 0.74 0.46 IV therapy care 0.150 0.032 4.70 0.00 Tracheostomy care -0.031 0.065 -0.47 0.64 Need injection 0.046 0.010 4.74 0.00 _cons 0.155 0.012 13.38 0.00 120

PAGE 121

Table A-32. GLM with gamma distribution PA walking improve with specialty care as ref. group Coef. Std. Err. Z P>z Cost leader -0.005 0.016 -0.33 0.74 Rehab focus 0.014 0.019 0.73 0.47 Differentiator -0.037 0.021 -1.76 0.08 Lack of strategy 0.024 0.016 1.45 0.15 Ownership -0.087 0.014 -6.04 0.00 Percent Medicaid -0.004 0.000 -10.18 0.00 Percent Medicare -0.002 0.001 -2.81 0.01 Market competition -0.425 0.248 -1.71 0.09 Occupancy -0.160 0.043 -3.73 0.00 Total beds 0.000 0.000 -0.53 0.60 Chain -0.032 0.012 -2.74 0.01 Acuity index -0.028 0.007 -4.07 0.00 ADL index -0.110 0.017 -6.49 0.00 Tube feed care -0.666 0.143 -4.66 0.00 Respiratory care 0.139 0.081 1.71 0.09 Suctioning care 0.444 0.419 1.06 0.29 IV therapy care 0.240 0.213 1.13 0.26 Tracheostomy care -0.238 0.449 -0.53 0.60 Need injection 0.038 0.068 0.56 0.58 _Ifips_2 0.218 0.283 0.77 0.44 _cons -0.35 0.188 1.11 0.27 Table A-33. OLS regressions of root transformed PA skin ulcers w ith specialty care as ref. group Coef. Std. Err. T P>t Cost leader -0.018 0.016 -1.15 0.25 Rehab focus 0.012 0.017 0.69 0.49 Differentiator 0.027 0.021 1.28 0.20 Lack of strategy -0.049 0.016 -3.00 0.00 Ownership 0.034 0.014 2.38 0.02 Percent Medicaid 0.001 0.000 3.39 0.00 Percent Medicare 0.001 0.001 0.70 0.48 Market competition -0.031 0.308 -0.10 0.92 Occupancy 0.014 0.046 0.31 0.75 Total beds 0.000 0.000 5.08 0.00 Chain -0.021 0.012 -1.76 0.08 Acuity index 0.034 0.007 4.88 0.00 ADL index 0.027 0.019 1.42 0.16 Tube feed care 0.568 0.156 3.64 0.00 Respiratory care 0.088 0.082 1.07 0.28 Suctioning care 0.535 0.413 1.30 0.20 IV therapy care 0.270 0.194 1.40 0.16 Tracheostomy care -0.160 0.431 -0.37 0.71 Need injection 0.183 0.069 2.63 0.01 _cons -2.540 0.096 -26.36 0.00 121

PAGE 122

Table A-34. GLM regressions of nursing home cost with gamma distribution Coef. Std. Err. z P>z Differentiator 0.002 0.016 0.12 0.90 Rehab focus 0.080 0.015 5.42 0.00 Specialty care focus 0.039 0.014 2.91 0.00 Lack of strategy -0.006 0.012 -0.54 0.59 Ownership -0.108 0.012 -8.87 0.00 Percent Medicaid -0.009 0.000 -27.82 0.00 Percent Medicare 0.001 0.001 1.49 0.13 Market competition -0.188 0.222 -0.85 0.39 Occupancy -0.904 0.035 -26.03 0.00 Total beds -0.001 0.000 -11.94 0.00 Chain -0.043 0.010 -4.38 0.00 Acuity index -0.012 0.006 -2.09 0.03 ADL index 0.001 0.014 0.07 0.94 Tube feed care 0.012 0.120 0.10 0.92 Respiratory care 0.159 0.069 2.29 0.02 Suctioning care 0.330 0.334 0.99 0.32 IV therapy care 0.056 0.175 0.32 0.75 Tracheotomy care 0.636 0.360 1.77 0.07 Need injection 0.093 0.056 1.66 0.09 _Ifips_2 -0.407 0.235 -1.73 0.08 _cons 7.526 0.150 49.92 0.00 Table A-35. Odds of nursing home of being in the highest operating margin tier Odds Ratio Std. Err. Z P>z Cost leader 1.25 0.10 2.79 0.01 Rehab focus 1.46 0.13 4.12 0.00 Specialty care focus 1.04 0.09 0.44 0.66 Differentiator 1.20 0.13 1.71 0.09 Ownership 4.65 0.37 19.15 0.00 Percent Medicaid 1.00 0.00 2.76 0.01 Acuindex 1.04 0.03 1.27 0.21 Total beds 1.00 0.00 1.94 0.05 Market competition 0.92 1.23 -0.06 0.95 Chain Affiliation 1.39 0.09 4.92 0.00 _Ifips_2 131.45 180.57 3.55 0.00 _Ifips_3 153.83 145.15 5.34 0.00 Note: 53 observations completely determ ined. Standard errors questionable 122

PAGE 123

123 Table A-36. Odds of nursing home of being in the highest total margin tier Odds Ratio Std. Err. z P>z Cost leader 1.22 0.09 2.58 0.01 Rehab focus 1.27 0.11 2.74 0.01 Specialty care focus 0.98 0.09 -0.25 0.80 Differentiator 1.31 0.14 2.6 0.01 Ownership 1.64 0.12 6.62 0.00 Percent Medicaid 1.06 0.03 1.98 0.05 Acuindex 1.00 0.00 1.29 0.20 Total beds 0.92 1.25 -0.06 0.95 Market competition 1.04 0.07 0.58 0.56 Chain Affiliation 10.53 15.18 1.63 0.10 Ifips_2 18.39 18.84 2.84 0.00 Note: 57 observations completely determ ined. Standard errors questionable .

PAGE 124

Table A-37. Correlations of independent variables 124 Cost Leader RN Focus Rehab Focus Differentiate Lack of Strategy Ownership Case mix Size Occupancy Pct Medicaid Pct Medicare Competition System Cost Leader 1 RN Focus -0.282 1.000 Rehab Focus -0.301 -0.215 1.000 Differentiator -0.212 -0.151 -0.161 1.000 Lack of Strategy -0.373 -0.267 -0.284 -0.200 1.000 Ownership -0.010 0.006 0.038 -0.041 -0.001 1.000 Case mix -0.032 -0.005 0.069 0.033 -0.047 0.062 1.000 Size 0.126 -0.017 -0.020 -0.026 -0.079 -0.029 0.082 1.000 Occupancy 0.033 -0.046 -0.041 0.034 0.019 -0.084 0.106 -0.027 1.000 Pct Medicaid 0.124 0.000 -0.288 0.041 0.100 0.231 0.024 0.160 0.048 1.000 Pct Medicare -0.136 -0.051 0.434 -0.077 -0.149 0.101 0.098 -0.001 -0.039 -0.477 1.000 Competition -0.010 0.071 -0.025 -0.034 -0.004 0.023 -0.095 -0.178 -0.105 0.088 -0.089 1.000 System -0.004 0.024 0.059 -0.061 -0.026 0.171 0.015 -0.066 -0.046 0.038 0.101 0.093 1

PAGE 125

LIST OF REFERENCES Agresti A, Finlay B, Statistical methods for social sciences, third edition, New Jersey 1997 Allen, M.J., & Yen, W. M. (2002). Introduction to Measurement Theory. Long Grove, IL: Waveland Press. American Physical Therapy Association History of Therapy caps Available at: http://www.apta.org/AM/Template.cfm?Section=Media&TEMPLATE=/CM/ContentDisplay.cfm&CONTENT ID=47095 Accessed March 30th, 2008 Amit, R., & Schoemaker, P. f. H. 1993. Strategic assets and organizational rent. Strategic Management Journal, 14: 33-4B Ansoff, H. Igor, Corporate Strategy, McGraw-Hill, New York, 1965 Abt Associates (2004), National Nursing Home Quality Measures Users Manual, Available at: http://www.cms.hhs.gov/NursingHomeQuality Inits/downloads/NHQIQMUsersManual.pdf Accessed March 30th, 2008 Barney J B. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17: 99-129. Britton L. C.; Clark, T. A. R.; Ball, D. F.. Modify or Extend? The Application of the Structure Conduct Performance Approach to Service Industries. By: Service Industries Journal, Jan92, Vol. 12 Issue 1, p34-43, 10p; Center for Medicare and Medicaid 2007. National Health Expenditures by type of service and source of funds, CY 1960-2006. Available at: http://www.cms.hhs.gov/NationalHealthExpendData/02_N ationalHealthAccountsHistorical.asp#TopOfPage Accessed March 30th, 2008 Center for Disease Control. 2007. NCHS faststats Nursing home care. Available at: http://www.cdc.gov/ Cohen nchs/fastats/nursingh.htm Accessed March 30th, 2007 Center for Medicare and Medicaid 2007 Me dicare Long Term Care Available at: http://www.medicare.gov/longtermcare/static/home.asp Accessed March 30th, 2008 Cool K Schendel, D STRATEGIC GROUP FORMATION AND PERFORMANCE: THE CASE OF THE U.S. PHARMACEUTICAL INDUSTRY, 1963-1982. Management Science ; Sep87, Vol. 33 Issue 9, p11021124, 23p Castle NG. Strategic groups and outcomes in nursi ng facilities. Health Care Manage Rev. 2003 JulSep;28(3):217-27. Cool, K., Schendel, D., (1988) Performance differences among strategic group members. Strategic Management Journal, 9 (3), pp 207-223 Caves, R., Porter, M., (1977), From entry barriers to mobility barriers: conjectural decisions and contrived deterrence to new competition. Quar terly Journal of Economics p.241-261 Centers for Medicare and Medicaid: RAI MDS Users Manual Available at: 125

PAGE 126

http://www.cms.hhs.gov/NursingHomeQualityInits/downloads/MDS20rai1202ch3.pdf Accessed March 30th, 2008 Centers for Medicare and Medicaid : State Operations Manual Appendix PP Guidance to Surveyors for Long Term Care Facilities Available at: http://www.cms.hhs.gov/manuals/Dow nloads/som107ap_pp_guidelines_ltcf.pdf Accessed March 30th, 2008 Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297-334 Day, G., Strategic Market Planning, West Publishing Co., St. Paul, 1984 Dess, G., Davis, P., (1984), Porter's (1980) Generic Strategies as Determinants of Strategic Group Membership and Organizational Performance. Academy of Management Journal ; Sep84, Vol. 27 Issue 3, p467-488, 22p, 5 charts Fabrigar L. R. Wegener D. T. MacCallum R. C., & Strahan E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods 4 2729. Feng Z Grabowski DC Intrator O Mor V The effect of state medicaid case-mix payment on nursing home resident acuity.Health Serv Res. 2006 Aug;41 (4 Pt 1):1337. Feng Z, Grabowski DC, Intrator O, Zinn J, Mor V. Medicaid payment rates, case-mix reimbursement, and nursing home staffing--1996-2004. Med Care. 2008 Jan;46(1):33-40. Fiegenbaum, A; Thomas, H., INDUSTRY AND STRATEGIC GROUP DYNAMICS: COMPETITIVE STRATEGY IN THE INSURANCE INDUSTRY, 1970-84. Journal of Management Studies, Jan93, Vol. 30 Issue 1, p69-105, 37p Fleishman JA, Cohen JW, Manning WG, Kosinski M. Us ing the SF-12 health status measure to improve predictions of medical expenditures. Med Care. 2006 May;44(5 Suppl):I54-63. Ford EW, Duncan JD, Ginter PM., The Structure of State Health Agencies: A Strategic Analysis. MCR&R 60:1 (March 2003) Fries BE, Schneider DP, Foley WJ, et al. Refining a case-mix measure for nursing homes: Resource Utilization Groups (RUG-III). Med Care.1994;32:668685. Gapenski, L., Understanding Healthcar e Financial Management, Third Ed ition, 2000 Health Administration Press GAO Report to Congressional Requesters (July 2001), Nursing Home Quality: Prevalence of Serious Problems, While Declining Reinforces Importance of Enhanced Oversight, http://www.gao.gov/new.items/d03561.pdf Grabowski DC, Castle NG. Nursing homes with pers istent high and low quality. Med Care Res Rev. 2004 Mar;61(1):89-115. Grabowski DC. A longitudinal study of Medicaid paymen t, private-pay price and nursing home quality. Int J Health Care Finance Econ. 2004;4:526. 126

PAGE 127

Harrington, C., Zimmerman, D., Karon, S.L. Robinson, J. & Beutel, P.(2000). Nursing home staffing and its relationship to deficiencies. Journal of Gerontology: Social Sciences, 55B, S278-S287 Harrington, C., Kovner, C., Mezey, M., Kayser-Jones, J., Burger, S., Mohler, M., Burke, S. & Zimmerman, D. (2000) Experts recommend minimum nurse staffing standards for nursing homes in the United States. The Gerontologist: 40, 5-16 Harrington C, Woolhandler S, Mullan J, Carrillo H, Hi mmelstein DU. Does investor ownership of nursing homes compromise the quality of care? Am J Public Health. 2001 Sep;91(9):1452-5. Hill, C.W. Differentiation Versus Low Cost or Di fferentiation and Low Cost: A Contingency Framework. Academy of Management Review 13 (1988): 401-12 Hitt M., Bierman L.,Shimizu K., Kochhar L. Direct and Moderating Effects of Human Capital on Strategy and Performance in Professional Service Firms: A Resource-Based Perspective Michael A. Hitt, Leonard Bierman, Katsuhiko Shimizu and Rahul Kochhar The Academy of Management Journal Vol. 44, No. 1 (Feb., 2001), pp. 13-28 Hofer, C. Schendel, D., Strategy Formulation: Anal ytical Concepts, West Publishin, St. Paul, 1978. Harrigan, C An Application of Clustering for Strategic Group Analysis. Strategic Management Journal ; JanMar85, Vol. 6 Issue 1, p55-73, 19p Hulin, C., Roznowski, M., (1985) Organizational Technol ogies: Effect on organizational characteristics and individuals responses. Research in Or ganizational behavior, vol7, p.39-85 Hunt, M.S. Competition in the Major Home App liance Industry, 1960-1970. Unpublished Doctoral dissertation, Harvard, 1972. Instititute of Medicine, 1986. Improving the qualit y of care of nursing homes. Washington DC: National Academy Press Johnson CE, Dobalian A, Burkhard J, Hedgecock DK, Harman J. Predicting lawsuits against nursing homes in the United States, 1997-2001. Health Serv Res. 2004 Dec;39(6 Pt 1):1713-31. Kachigan S.K., Multivariate Statistical analysis: A conceptual in troduction, Radius, New York, 1991 Kaiser, H. F. (1960). The application of electronic computers to factor analysis. Educational and Psychological Measurements, 20, 141-51 Kaplan, and Norton. 2001. Transforming the balanced scorecard from performance measurement to strategic management: Part I. Accounting Horizons (March): 87-104 Katz, R., Cases and Concepts in Corporate St rategy, Prentice-Hall, Englewood Cliffs, N.J.1970 Ketchen DJ, Shook CJ. The Application of Cluster Analys is In Strategic Management Research: An Analysis and Critique. Strategic Manageme nt Journal 17 (1996): 441-58 Kling, J A.; Smith, K A. Identifying Strategic Groups in the U.S. Airline Industry: An Application of the Porter Model. Transportation Journal, Winter95, Vol. 35 Issue 2, p26-34, 9p 127

PAGE 128

Laberge A, Weech-Maldonado R, Johnson CE, Jia H, Dewald L.Outsourcing vetera ns for long-term care: comparison of community and state veterans' nursing homes. J Health Hum Serv Adm. 2008 Spring;30(4):44167. Leask, G.; Parker, D. Strategic groups, competitive groups and perfo rmance within the U.K. pharmaceutical industry: Improving our understanding of the competitive process. Strategic Management Journal, Jul2007, Vol. 28 Issue 7, p723-745 Marlin D, Sun M, Huonker JW. Strategic groups and performance in the nursing home industry: a reexamination.Med Care Res Rev. 1999 Jun;56(2):156-76 Marlin D, Huonker JW, Sun M. An examination of th e relationship between strategic group membership and hospital performance. Health Care Manage Rev. 2002 Fall;27(4):18-29 Mukamel DB, Glance LG, Li Y, Weimer DL, Spector WD, Zinn JS, Mosqueda L. Does risk adjustment of the CMS quality measures for nursing homes ma tter? Med Care. 2008 May;46(5):532-41. Mehra, Ajay. RESOURCE AND MARKET BASED DETERMINANTS OF PERFORMANCE IN THE U.S. BANKING INDUSTRY. Strategic Management Journal, Apr96, Vol. 17 Issue 4, p307-322 Murray, A.L. A Contingency view of porters generic strategies. Academy of Management Review 13 (1988): 390-400 National Institute of Health, http://www.nlm.nih.gov/medlineplus/nursinghomes.html accessed March 20th, 2008 O'Neill C, Harrington C, Kitchener M, Saliba D. Quality of care in nursing homes: an analysis of relationships among profit, quality, and ownership. Med Care. 2003 Dec;41(12):1318-30. Pagano E, Bo S, Petrinco M, Rosato R, Merletti F, Gre gori D. Factors affecting hospitalization costs in Type 2 diabetic patients. J Diabetes Complications. 2009 Jan-Feb;23(1):1-6. Epub 2008 Apr 16. Pfeffer, J. 1994. Competitive advantage through people Boston: Harvard Business School Press Porter, M.E., Competitive Strategy, New York: Free press, 1980 Rhoades J, Sommers J. Expenses and Sources of Payment for Nursing Home Residents 1996. Rockville, MD: Agency for Healthcare Research and Quality; 2000. MEPS Research Findings No 13. AHRQ Pub. No 01-0010 SAS manual 1990, SAS Technical Report A-108 Scanlon, W. 1980. A theory of the nursing home market. Inquiry 17 (1):25-41. Schnelle JF. Determining the relati onship between staffing and quality. Gerontologist. Feb 2004;44(1):10-12 Shen YC, Eggleston K, Lau J, Schmid CH.Hospital owne rship and financial performance: what explains the different findings in the empirical litera ture? Inquiry. 2007 Spring;44(1):41-68 Short JC, Palmer TB, Ketchen DJ Jr. Resource-based and strategic group influences on hospital performance. Health Care Manage Re v. 2002 Fall;27(4):7-17. 128

PAGE 129

STATA. STATA reference manual release 7. College station: STATA Press; 2001 Swan JH, Harrington C Clemea W Pickard RB Studer L deWit SK Med Care Res Rev. 2000 Sep;57(3):361-78 Medicaid nursing facility reimbursement methods: 1979-1997 The Medicare Payment Advisory Commission 2007 Report to the Congress: Medicare Payment Policy (March 2007) Available at: http://www.MedPAC.gov/documents/Mar07_EntireReport.pdf Accessed March 30th, 2007 The Medicare Payment Advisory Commission 2003 Report to the Congress: Medicare Payment Policy (March 2007) Available at: http://www.MedPAC.gov/documents/Mar03_EntireReport.pdf Accessed March 30th, 2007 Tremblay, V. J., Strategic groups and the demand for beer. Journal of Industrial Ec onomics, Dec85, Vol. 34 Issue 2, p183, 16p United States General Accounting Office 2000 NUR SING HOMES: Aggregate Medicare Payments Are Adequate Despite Bankruptcies, GAO/T-HEHS-00-192 Venkatraman N., Vasudevan R. Measurement of Busin ess Performance in Strategy Research: A Comparison of Approaches The Academy of Management Review Vol. 11, No. 4. (Oct., 1986), pp. 801-814. Walshe K, Harrington C. Regulation of nursing facilities in the United States: an analysis of resources and performance of state survey agencies. Gerontologist. 2002 Aug;42(4):475-87. Weech-Maldonado R, Neff G, Mor V. Does quality of care lead to better financial performance?: the case of the nursing home industry. Health Care Manage Rev. 2003 Jul-Sep;28(3):201-16. Weech-Maldonado, R., Meret-Hanke, L., Neff, M., and Mor, V. (2004). Nurse staffing patt erns and quality of care in nursing homes. Health Care Management Review, 29 (2), 1-10. Weech-Maldonado, R., Laberge, A., Pradham, R., Johnson, C., Yang, Z., Hyers, K., Nursing home quality and financial performance. Working Paper White H. A. Heteroskedasticity-Consistent Covari ance Matrix Estimator and a Direct Test for Heteroskedasticity Econometrica 1980; 48 (4) :8178 Wodchis WP, Fries BE, Pollack H. Payer incentives and physical rehabilitation therapy for nonelderly institutional long-term care residents: evidence from Mi chigan and Ontario. Arch Phys Med Rehabil. 2004 Feb;85(2):210-7. Wright P., A Refinement of Porters Strategies Strategic Management Journal 8 (1987) 93-101 Zinn JS, Aaronson WE, Rosko MD. Strategic groups, performance, and strategic response in the nursing home industry. Health Serv Res. 1994 Jun;29(2):187-205. Zinn J, Feng Z, Mor V, Intrator O, Grabowski D. Re structuring in response to case mix reimbursement in nursing homes: a contingency approach. Health Care Manage Rev. 2008 Apr-Jun;33(2):113-23. 129

PAGE 130

130 BIOGRAPHICAL SKETCH Alexandre Laberge completed his honors bachelor s of physical therapy at the University of Western Ontario in London, Ontario in 1995. He worked as a physical therapist in a variety of settings including hospitals, skilled nursing facilities, assisted living facili ties and in the home health settings. He completed his Masters of Business Administration in informa tion technology at Goldey Beacom College in 2002. He completed his Ph.D. in health se rvices research, management and policy in May in 2009. He worked in skilled nursing facilities for over 6 years as a therapist as well as a rehab manager. In the 4.5 years that he worked as a research assi stant, he worked closely with Dr Weech-Maldonado on a variety of projects that examin ed the relationship between cost, quality and financial performance in nursing homes.