<%BANNER%>

Quantifying the Diabetes Online Community

Permanent Link: http://ufdc.ufl.edu/UFE0045316/00001

Material Information

Title: Quantifying the Diabetes Online Community An Investigation of Online Forums for Diabetes
Physical Description: 1 online resource (54 p.)
Language: english
Creator: Balkhi, Amanda Marie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: diabetes -- forum -- internet -- parents -- pediatric
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Type 1 Diabetes (T1D) is a chronic autoimmune disease of which 15,000 new cases are diagnosed each year in the United States. Youth with T1D are especially burdened by the rigorous care required by their illness and significantly rely on their parents for many care behaviors. Parents coping with the burden of their child’s T1D often are left with elevated levels of parenting stress and fear. While in-person social support groups for parents of these children have been examined, this study is the first of its kind to directly examine the impact of online social support communities on parents of children with T1D. Participants were 102 parents whom were recruited from online support communities and completed measures of parenting stress, hypoglycemic fear, sleep quality, and internet usage. Participants rated social support and knowledge as the primary benefits of forum participation and a majority of parents indicated that the care they provide for their child is impacted by their experiences on T1D forums. Parent age was the only demographic variable significantly related to forum membership. Forum membership had a significant positive relationship to hypoglycemic fear behaviors and parenting stress frequency with a small effect size. There was a significant indirect effect of sleep quality on the relationship between forum membership and parenting stress frequency, however there was not significant mediation. Recommendations for longitudinal study and recommendations for health care professionals, including patient education and debriefing regarding forum use, are discussed.
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 Amanda Marie Balkhi.
Thesis: Thesis (M.S.)--University of Florida, 2013.
Local: Adviser: Geffken, Gary R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045316:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045316/00001

Material Information

Title: Quantifying the Diabetes Online Community An Investigation of Online Forums for Diabetes
Physical Description: 1 online resource (54 p.)
Language: english
Creator: Balkhi, Amanda Marie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: diabetes -- forum -- internet -- parents -- pediatric
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Type 1 Diabetes (T1D) is a chronic autoimmune disease of which 15,000 new cases are diagnosed each year in the United States. Youth with T1D are especially burdened by the rigorous care required by their illness and significantly rely on their parents for many care behaviors. Parents coping with the burden of their child’s T1D often are left with elevated levels of parenting stress and fear. While in-person social support groups for parents of these children have been examined, this study is the first of its kind to directly examine the impact of online social support communities on parents of children with T1D. Participants were 102 parents whom were recruited from online support communities and completed measures of parenting stress, hypoglycemic fear, sleep quality, and internet usage. Participants rated social support and knowledge as the primary benefits of forum participation and a majority of parents indicated that the care they provide for their child is impacted by their experiences on T1D forums. Parent age was the only demographic variable significantly related to forum membership. Forum membership had a significant positive relationship to hypoglycemic fear behaviors and parenting stress frequency with a small effect size. There was a significant indirect effect of sleep quality on the relationship between forum membership and parenting stress frequency, however there was not significant mediation. Recommendations for longitudinal study and recommendations for health care professionals, including patient education and debriefing regarding forum use, are discussed.
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 Amanda Marie Balkhi.
Thesis: Thesis (M.S.)--University of Florida, 2013.
Local: Adviser: Geffken, Gary R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045316:00001


This item has the following downloads:


Full Text

PAGE 1

1 QUANTIFYING THE DIABETES ONLINE COMMUNITY: AN INVESTIGATION OF ONLINE FORUMS FOR DIABETES By AMANDA MARIE BALKHI A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREME NTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Amanda Marie Balkhi

PAGE 3

3 To m y f amily

PAGE 4

4 ACKNOWLEDGMENTS I first would like to acknowledge my husband for his everlasting support and lo ve. His determination has and will always be an integral part of my success and has inspired me to keep working until I see results. Additionally, his insistence that research can and should have real world implications has encouraged me to continue asking how results found in a lab can translate to meaningful interpretations. I also am thankful for my family. They have stood by me every day to this point and will continue to be here to help me reach my goals. I would especially like to thank my parents an d grandparents who have stood as role models for me and my father and grandfather who encouraged me to pursue an education regardless of the obstacles. I am eternally grateful to my academic family. Drs. Gary Geffken and Joseph McNamara have put their fai th in me since I first joined this team and have encouraged every interest of mine. I also owe thanks to Adam Reid who has continued to remind me to think big in all endeavors. Finally, I would like to thank the Division of Medical Psychology undergraduate s for always being willing to lend a hand.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ...... 4 LIST OF TABLES ................................ ................................ ................................ ................ 7 LIST OF FIGURES ................................ ................................ ................................ .............. 8 LIST OF ABBREVIATIONS ................................ ................................ ................................ 9 ABSTRACT ................................ ................................ ................................ ........................ 10 INTRODUCTION ................................ ................................ ................................ ............... 12 Type 1 Diabetes in Children ................................ ................................ ....................... 12 Treatment Regimen ................................ ................................ ............................. 14 Role of Parents ................................ ................................ ................................ .... 15 Online Health Communities ................................ ................................ ........................ 17 Summary ................................ ................................ ................................ ..................... 18 Study Aims ................................ ................................ ................................ .................. 19 METHODS ................................ ................................ ................................ ......................... 21 Procedure ................................ ................................ ................................ .................... 21 Participants ................................ ................................ ................................ ................. 21 Measures ................................ ................................ ................................ .................... 22 Pediatric Inventory for Parents (PIP) ................................ ................................ ... 22 Hypoglycemic Fear Scale, Parent Version (HFS P) ................................ ........... 23 Pittsburg Sleep Quality Index (PSQI) ................................ ................................ .. 23 Michigan Diabetes Res earch and Training Center's Brief Diabetes Knowledge Test (DKT) ................................ ................................ ..................... 24 Statistical Analyses ................................ ................................ ................................ ..... 24 Qualitative Coding Methods ................................ ................................ ................. 24 Quantitative Analysis ................................ ................................ ........................... 25 RESULTS ................................ ................................ ................................ .......................... 29 Preliminary Analyses ................................ ................................ ................................ .. 29 Results for Aim One ................................ ................................ ................................ .... 30 Why Visit the Diabetes Online Community ................................ ......................... 30 Biggest B enefits of the Diabetes Online Community ................................ .......... 30 Does the Diabetes Online Community Impact Care ................................ ........... 31 Trustworthiness ................................ ................................ ................................ .... 31 Results for Aim Two ................................ ................................ ................................ .... 32 Hypoglycemic Fear ................................ ................................ .............................. 32 Parenting Stress ................................ ................................ ................................ ... 33

PAGE 6

6 Results for Aim Three ................................ ................................ ................................ 33 Hypoglycemic Fear Behaviors ................................ ................................ ............. 33 Parenting Stress Freque ncy ................................ ................................ ................ 34 Summary ................................ ................................ ................................ ..................... 34 DISCUSSION ................................ ................................ ................................ .................... 42 Limitations ................................ ................................ ................................ ................... 45 Future Directions ................................ ................................ ................................ ........ 45 LIST OF REFERENCES ................................ ................................ ................................ ... 48 BIOGRAPHICAL SKETCH ................................ ................................ ................................ 54

PAGE 7

7 LIST OF TABLES Table page 2 1 Parent demographic variables ................................ ................................ ............... 27 2 2 Child with Type 1 Diabetes de mographic variables ................................ .............. 28 3 1 Correlations between variables ................................ ................................ ............. 36 3 2 Internet and forum use per week ................................ ................................ ........... 37 3 3 Hierarchical linear regression predicting forum membership ............................... 38 3 4 Hierarchical linear regression predicting hypoglycemic fear and parenting stres s ................................ ................................ ................................ ...................... 39

PAGE 8

8 LIST OF FIGURES Figure page 3 1 Proposed mediation model of hypoglycemic fear behaviors ................................ 40 3 2 Proposed mediation model of parenting stress frequency ................................ ... 41

PAGE 9

9 LIST OF ABBREVIATIONS H B A1 C Glycosylated hemoglobin is a measure of plasma glucose concentration. It is assessed using a simple blood test an d renders an estimated average blood sugar for the previous three months. T1D Type 1 Diabetes

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Scie nce QUANTIFYING THE DIABETES ONLINE COMMUNITY: AN INVESTIGATION OF ONLINE FORUMS FOR DIABETES By Amanda Marie Balkhi May 2013 ABSTRACT Chair: Gary Roy Geffken Major: Psychology Type 1 Diabetes (T1D) is a chronic autoimmune disease of which 15,000 new cases are diagnosed each year in the United States. Youth with T1D are especially burdened by the rigorous care required by their illness and significantly rely on their parents for many care behaviors. Parents coping with the burden of often are left with elevated levels of parenting stress and fear. While in person social support groups for parents of these children have been examined, this study is the first of its kind to directly examine the impact of online social support communities on parents of children with T1D. Participants were 102 parents whom were recruited from online support communities and completed measures of parenting stress, hypoglycemic fear, sleep quality, and internet usage. Participants rated soc ial support and knowledge as the primary benefits of forum participation and a majority of parents indicated that the care they provide for their child is impacted by their experiences on T1D forums. Parent age was the only demographic variable significant ly related to forum membership. Forum membership had a significant positive relationship to hypoglycemic fear behaviors and parenting stress frequency with a small effect size. There was a

PAGE 11

11 significant indirect effect of sleep quality on the relationship be tween forum membership and parenting stress frequency, however there was not significant mediation. Recommendations for longitudinal study and recommendations for health care professionals, including patient education and debriefing regarding forum use, ar e discussed.

PAGE 12

12 CHAPTER 1 INTRODUCTION Diabetes is a class of conditions hallmarked by high levels of blood glucose. Diabetes impacts an estimated 25.8 million people in the United States and has serious complications such as heart disease, blindness, kidne y disease, amputations, and nervous system damage and is the seventh leading cause of death in the United States (Center for Disease Control and Prevention, 2011) Direct and indirect costs of diabetes are believed to total more than $132 billion in the Un ited States in 2002 alone (American Diabetes Association, 2003) In children, it is estimated that 1 out of every 523 youth under the age of 20 are affected by diabetes (SEARCH for Diabetes in Youth Study Group, 2006) Diabetes is characterized by either b y the destruction of pancreatic cells that produce insulin (Type 1) or gradually decreasing insulin sensitivity (Type 2) (Center for Disease Control and Prevention, 2011) While Type 2 Diabetes is more commonly diagnosed in adults, Type 1 Diabetes is a co ndition that is primarily diagnosed in childhood and that is responsible for more than half of all diabetes diagnoses in children and adolescents across all racial groups with the exception of American Indian youth (SEARCH for Diabetes in Youth Study Group 2006) Type 1 Diabetes in Children Type 1 Diabetes (T1D) is a chronic autoimmune disease that impacts the beta cells of the pancreas and renders the human body unable to produce insulin, a hormone necessary to process the glucose needed for the body to function. Recent estimates of the prevalence of T1D indicate that 15,000 new cases emerge in youth each year in the United States with cases disproportionately affecting non Hispanic white youth (SEARCH for Diabetes in Youth Study Group, 2012) Current tre nds show that these

PAGE 13

13 numbers are not lessening, despite continued research interest in the field of T1D prevention and treatment (SEARCH for Diabetes in Youth Study Group, 2012) The cause of T1D is curre ntly unknown, but epidemiological research has id entified factors that influence the likelihood of developing T1D. In a sample of Finnish chil dren, infant feeding patterns specifically early intro have been associated with later development of T1D (Akerblom, et al., 1993; Childhood Diabetes in Finland Study Group, 1992) These findings were replicated in a study of children in Qatar, and Vitamin D and sun light were identified as additional predictors of the development of T1D (Bener, et al., 2009) There is also emerging evidence th at appears to point to a seasonality of T1D onset, in which children are more likely to be diagnosed during periods of low sun exposure (Moltchanova, Schreier, Lammi, & Karvonen, 2009) While the recent interest in studying the etiology of T1D is promising researchers have yet to pinpoint interventions to prevent or cure T1D. T1D has been associated with a number of negative medical outcomes, including reduced cardiorespiratory fitness and shorter height as compared to healthy same aged peers (Brown, Ahme d, Clayton, & Dunger, 1994; Williams, Guelfi, Jones, & Davis, 2011) Perhaps the most common complication of T1D is hypoglycemia. Hypoglycemia Hypoglycemia is a potentially fatal com plication and is a well documented concern for individuals with T1D (Taborsky, Ahren, & Havel, 1998) Additionally, hypoglycemia can result in coma, seizures, nausea and other physiological symptoms of insufficient sugar. Hyperglycemia, which is characteri zed by an excess of glucose in the blood stream, is another serious complication. More specifically, hyperglycemia can produce feelings of

PAGE 14

14 excessive thirst, blurring vision, and long term complications such as kidney disease (Miller, 1999) These complicat ions tend to be especially salient for children and adolescents because they tend to exhibit poorer awareness and poorer disease management behaviors which lead to hypoglycemic and hyperglycemic events (Jones & Davis, 2003; Ly, Gallego, Davis, & Jones, 200 9; Ryan & Becker, 1999). Treatment Regimen Effective treatment of T1D and prevention of medical complications (e.g., hyperglycemia or hypoglycemia) requires continuous management of blood glucose by monitoring blood glucose and administering insulin. Comm on methods of insulin administration include direct injection or through a subcutaneous infusion point controlled by a pump worn by the patient. Both injection and pump based insulin administration have been shown to be efficacious for use with children an d adolescents (Plotnick, Clark, Brancati, & Erlinger, 2003 ; Weinzimer, et al., 2004) Longitudinal studies have indicated that intensive treatment of T1D (i.e. more frequent insulin administration) is effective at reducing long term complications and at i mproving glycemic control in the short term when compared to less frequent insulin administration (Abid, et al., 2011; Reichard, Nilsson, & Rosenqvist, 1993) Despite literature showing the success of these interventions for managing T1D, effective manage ment of diabetes with children requires frequent adjustment and flexibility. This is because glucose levels are influenced by carbohydrate intake and se levels is particularly challenging as they tend to be more active and consume more snacks. In addition, younger children are traditionally more biologically sensitive to changing insulin doses (Kiess, et al., 1998) To reduce the difficulty of child

PAGE 15

15 tre atment, many technological advances have been developed to aid the process of calculating and administering insulin. One such advancement is the continuous blood glucose monitor, a compact sensor that quickly measures the concentration of blood glucose and transmits it to a handheld receiver. These monitors have been shown to be effective for use in children and adolescents and appear to allow for more intensive monitoring of blood glucose with little interference in day to day activities (Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, 2008; STAR 3 Study Group, 2010). Additional attempts have been made to construct detailed decision making algorithims that would allow for seemless administration of insulin by a insulin pump based on feedback from a glucose monitor (e.g. Hovorka, et al., 2010), however these technologies have yet to come into wide use. Role of Parents Research examining parents a showed 22% of parents endorsing clinical levels of Posttraumatic Stress symptomology diagnosis (Landolt, Vollrat h, Laimbacher, Gnehm, & Sennhauser, 2005). Parents also widely report difficulties with sleep, due both to worry and nighttime care activities (Monaghan, Herbert, Cogen, & Streisand, 2012) Additionally, with the immense responsibility and knowledge that i s required to adjust insulin dosages throughout the day and the lack of technological interventions to do so automatically, parents are often dual role for parents w ho are required to act as both parent and health care professional, a position which can lead to increased levels of anxiety, depression and

PAGE 16

16 parenting stress (Streisand, et al., 2008; Streisand, Swift, Wickmark, Chen, & Holmes, 2005; Sullivan Bolyai, Deat rick, Gruppuso, Tamborlane, & Grey, 2003) It is this dual role and associated difficulties that establishes parents as an ideal population for interventions targeting glycemic control and psychosocial factors (Winkley, Landau, Eisler, & Ismail, 2006) How ever, despite parents of children with T1D being at a greater risk for developing these psychological problems, more information is needed to guide appropriate support for these parents. Sleep is one factor that has received little attention in research o n parents of children with T1D. Recent research by Monaghan and colleagues (2012) has demonstrated that parents of children with T1D experience frequent nighttime tota l sleep time such as these have been associated with numerous negative cognitive and physical outcomes including increased perceived stress, difficulties with attention, and decreased immune function (Bryant, Trinder, & Curtis, 2004; Tworoger, Davis, Viti ello, Lentz, & McTiernan, 2005;Van Dongen, Maislin, Mullington, & Dinges, 2003) Given these potential negative side effects of sleep interruption, it has become increasingly clear that the relationship between sleep and the psychological effects of parent ing a child with T1D need to be directly assessed. The only study to the author's knowledge to have done so only examined nighttime care giving behaviors, not quantitative measures of sleep quality ( Monaghan, Herbert, Cogen, & Streisand, 2012 ). Despite th e difficulties these parents face from sleep and the psychological effects of parenting, most direct interventions to reduce parental burden are time consuming and cost prohibitive. Brief interventions have been designed to address

PAGE 17

17 parenting stress alone ( e.g. Monaghan, Hilliard, Cogen, & Streisand, 2011) However, effective, and widely accessible that targets parenting stress and sleep. Interventions using the internet as a primary method of communication have been proposed to address this difficulty as internet delivery allows programs to be both easily assessable and cost effective (Griffiths, Lindenmeyer, Powell, Lowe, & Thorogood, 2006; Ritterband, Gonder Frederick, Cox, Clifton, West, & Borowitz, 2003) Online Health Communities Online health communities are relatively new phenomena that gained popularity shortly after the 1997 launch of national social networking websites (Boyd & Ellison, 2007) and are targeted towards individuals suffering from a shared health condition. These websites are often set up with the same purpose as an in person support group, providing a place to share information and knowledge of the health condition; however, given that most communities do not assure the accuracy of forum postings, online forums possess an inherent risk that in person groups do not present (White & Dorman, 2001). Previous research has examined these naturally occurring forums for a diverse array of conditions, with a prima ry focus on adult sufferers of a specific health condition. However, the examination of many of these naturally occurring forums has been primarily qualitative (e.g. Coulson, 2005; Reeves, 2001 ). Additionally, the quantitative studies that have been cond ucted focus heavily on participant expression alone, not forum interactions (e.g. Davison, Pennebaker, & Dickerson, 2000; Lieberman, et al., 2003; Lieberman & Goldstein, 2006; Weinert, 2000) These online interactions can have similar effects to their in person counterparts. Researcher moderated forums have

PAGE 18

18 established the efficacy of online forums for establishing social support and conveying health related information for a variety of conditions (Ayers & Kronenfeld, 2007; Barrera Jr, Glasgow, McKay, B oles, & Feil, 2002; Wangberg, 2007) knowledge, no research has directly assessed the impact of online forums on parents of children with T1D. The Diabetes Online Community : The diabetes online community is one health forum tha t has developed rapidly over the past decade. While no organization has yet to gather conclusive data on the number of users within the diabetes online community, it is known that the number of registered users on three of the largest forums totals nearly 500,000 users (Warshaw, Hernandez, Tenderich, & Edelman, 2011). Additionally, some information suggests that the diabetes online community is seen as a helpful resource. Greene, Choudhry, Kilabuk and Shrank (2010) undertook the only existing analysis of th e diabetes online community. These researchers examined the content of posts made to Facebook groups with "diabetes" in the title on August 8, 2009; the results of this cursory examination suggested that patients with diabetes seek help and openly share pe rsonal information on the internet. However, current studies have yet to examine or report demographic variables and reported usage habits of group participants. Summary T1D is a disease with a large burden of care that often falls on the parents of child ren and adolescents. Parenting these children requires knowledge and fortitude and often has adverse effects on the functioning of these dedicated parents. Disruptions in sleep and elevated levels of hypoglycemic fear and parenting stress are common in the se parents. Interventions to target these negative effects of parenting a child with T1D have been developed, but there has yet to be an intervention that has the reach

PAGE 19

19 and cost effectiveness of pre existing online health communities. However, despite the existence of a well established diabetes online community with a dedicated subset of participating parents, there has yet to be a study that assesses the demographics of this community and that evaluates the effects of joining and participating in this co mmunity. The need for a study of the diabetes online community is underscored by the increasing rates of T1D diagnosis and the recent growth of the diabetes online community (SEARCH; Warshaw, 2011). To this end, the current study will address these limitat ions outlined by White and Dorman (2001) by directly assessing the characteristics of those who use these communities. Study Aims The current study aimed to be the first to assess the parents participating in the diabetes online community through online f orums. The first aim sought to provide demographic information on the participants of the diabetes online community and their perceptions of the community. It was hypothesized that participants would mirror the demographics of previous clinic based samples in that they will be mainly non Hispanic white mothers. It was also hypothesized that participants would rate social support and medical advice as the main reasons for participation. Finally, it was hypothesized that a majority of parents will indicate th at the care they provide to their child is impacted by their experiences within the diabetes online community. The perceived trustworthiness of the diabetes online community compared to the internet in general will also be assessed. The second aim of this study is to examine forum membership. Forum membership is quantified as the number of forums the participant has joined and utilized. Membership was chosen over forum use time to distinguish forum viewing

PAGE 20

20 (which can be done without membership) and forum i nvolvement/cohesion. It is hypothesized that younger parent age and poorer child glycemic control (i.e. higher H B A1 C ) will predict an increased number of forum memberships. Additionally, increased forum memberships are expected to predict higher reported parenting stress and hypoglycemic fear. The third and final aim of the present study is to examine the impact of sleep quality on the significant relationships demonstrated in the second aim. Based on the work of Monaghan and colleagues (2012), it is expe cted that parents will demonstrate sleep disruptions that will impact their experience of hypoglycemic fear and parenting stress. It is hypothesized that sleep quality will partially mediate the relationship between forum memberships, parenting stress, and hypoglycemic fear.

PAGE 21

21 CHAPTER 2 METHODS Procedure Following approval from the local institutional review board, owners or designated representatives of T1D forums based in the United States were contacted. With the permission of the forum owner or rep resentative, a recruitment message was posted to the forum requesting parents of children with T1D to participate in a survey of the diabetes online community. Study inclusion criteria mandated that parents be a member of at least one T1D forum for one mon th or more and have a child with T1D that lived with the respondent at least half of the time and for whom the parent was primarily responsible for diabetes related care. Participants were informed that as compensation, an anonymous donation of $5 would be made to the Juvenile Diabetes Research Foundation (JDRF) for the first 100 completed surveys ($500 total). Parents followed a link to an electronic consent form and clicked to indicate their consent. They then completed a brief online battery of demograph ic, sleep quality, parenting stress, and hypoglycemic fear questionnaires followed by a qualitative assessment of their interactions in online T1D forums. Recruitment was completed in four months. Participants One hundred and two participants completed the survey. Of these, the majority were non Hispanic White female caregivers (Table 2 1). Parents were on average 40 years and 10 months (Range = 28 54; SD = 6.07) and averaged two children (Range = 1 5; SD = 0.91). Only five families reported having more tha n one child with T1D; parents were instructed to answer study questions based on their first child with T1D. The average age of the first child with T1D was nine years, eight months (Range = 2 23;

PAGE 22

22 SD = 4.27) and the average H B A1 C chosen from a categorical list of ranges, was between 7.5 and 7.9 (Range = 3.5 to 13.4). The average number of years since diagnosis for the first child with T1D was 3.79 (Range = 0 14; SD = 2.72). Of the five second children with T1D, the average age was eight years and 0 month s (Range = 3 13; SD = 4.36) and the average H B A1 C was between 7.0 and 7.4 (Range = 7.0 to 8.4). Average number of years since diagnosis for the second child with T1D was 3.00 years (Range = 1 6; SD = 1.87). All but seven of the children with T1D were repor tedly enrolled in school (Table 2 2). Measures Demographic information was collected regarding both participating parents and their children. In addition to completing questions assessing their use of the internet forum, parents completed the following mea sures to assess common stressors of parents with children with T1D and their diabetes knowledge. Pediatric Inventory for Parents (PIP) The PIP is a 42 item questionnaire, rated on a 5 point Likert Scale that assesses illness and care in four domains: communication, (Streisand, Braniecki, Tercyack, & Kazak, 2001) event frequency and one of event difficulty, are calculated with higher scores indicating more frequency and difficulty. This measure has been used in a population of parents of children with T1D with good internal consistency and validity (frequency = .94; difficulty = .95) (Streisand, S wift, Wickmark, Chen, & Holmes, 2005) In our sample, internal consistency was good (frequency = .93; difficulty = .93).

PAGE 23

23 Hypoglycemic Fear Scale, Parent Version (HFS P) The HFS P is a 26 item questionnaire used to assess the fear of hypoglycemic episodes a mong parents of children with T1D (Clarke, Gonder Frederick, Snyder, & Cox, 1998) The HFS P is comprised of two subscales, the Behavior subscale and Worry subscale. The Behavior subscale consists of 11 items describing activities that parents may do to av oid hypoglycemia. The Worry subscale includes 15 items assessing aspects of hypoglycemia that may cause anxiety in parents. Each question is rated on a five point Likert scale (1 = Never, 5 = Almost Always), and both subscale and total scores can be derive d. This scale demonstrated good internal consistency in our sample (.89 for Total score, .78 for the Behavior subscale, and .89 for the Worry subscale). Pittsburg Sleep Quality Index (PSQI) The PSQI is a brief measure of sleep quality that asks participan ts to report their sleep quality over the past month (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). The PSQI evaluates seven aspects of sleep to compute a global sleep quality score from 0 to 21: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficency, sleep disturbances, use of sleeping medication, and daytime dysfunction. While a clinical cut sleepers, the PSQI has also been used as a continuous measure of sleep ( Lopez Wagner, Hoffman, Sweeney, Hodge, & Gilliam, 2008; Meltzer, 2008). The PSQI has demonstrated good reliability and validity in other studies; internal consistency within our sample was acceptable ( Carpenter & Andrykowski, 1998).

PAGE 24

24 Michigan Diabetes Research and Training Center's Brief Diabetes Knowledge Test (DKT) The DKT is a 23 item questionnaire used to assess general diabetes knowledge (Fitzgerald, Funnell, Hess, Barr, Anderson, Hiss, & D avis, 1998) Scores can be generated for knowledge of diabetes with and without insulin use. The insulin use score was used for this study, a score of 23 indicates that all questions were answered correctly. Each question is has three to four alternative a nswers, from which the participant chooses one. This test has demonstrated good reliability and validity in previous samples (Fitzgerald, Anderson, Funnell, Hiss, Davis, & Barr, 1998) The DKT demons Statistical Analyses Qualitative Coding Methods Participants answered two qualitative questions regarding their activity within the diabetes online community: 1) Why do you visit Type 1 Diabetes re lated websites (forums, blogs, etc.)? and 2) What is the biggest benefit to reading or engaging in the Type 1 Diabetes Support Forums? Participant entries were reviewed and using a grounded theory based methodology four categories were established for each question: 1) Social Support, 2) Knowledge Seeking, 3) both Social Support and Knowledge Seeking, and 4) none of the above. Two raters who were nave to the research question independently coded each response into one of the four categories. Inter rater re liability for these raters was found to be good, Kappa = 0.85 p <.001 (Question 1) and Kappa = .836 p <.001 (Question 2). For items in which raters did not agree, a third rater chose the final response.

PAGE 25

25 Quantitative Analysis All data analysis was conducted using the Statistical Package for the Social Sciences, version 20 (IBM, 2011) Data were collected using SurveyMonkey and exported directly to SPSS. Data were checked by the author and participants with less than 20% of data missing per measure were consi dered for multiple imputation. MCAR test was conducted to ensure that all missing data were missing completely at random. Ten multiple imputation data sets were created to address missing data present on the PIP and HFS P. Missing data from the PS QI could not be imputed due to measure restrictions. For preparatory analyses, all variables were checked to ensure that they met normality assumptions, as well as assumptions required to conduct a paired samples t test, linear regression, and indirect ef fect analysis. Variables that did not meet normality assumptions (sleep quality) were transformed using a Blom transformation (Blom, 1958) In all analyses, total time spent on the internet was controlled for to allow for examination of the effect of forum membership. A partial correlation was conducted to examine the intercorrelation between the variables of interest and relevant demographic variables. To control for multicollinearity, variables demonstrating significant multicollinearity (HFS P subscales and PIP subscales) were residualized to allow for examination of unique variance. Independent samples t tests were then conducted to assess for differences by gender and number of children with diabetes. All variables used in future regressions were then s tandardized to allow for comparison of effects. Due to missing data, the sample size for the second aim was 94. For the final aim, sleep was tested as a mediator between forum membership and those variables demonstrating a significant linear relationship i n the second aim.

PAGE 26

26 Due to missing data on the PSQI and the independent variable, the sample size for the third aim was 89. This analysis was done using the SPSS macro PROCESS designed for testing mediation models (Hayes, 2012). Using procedures outlined by Hayes (2012), this analysis tests the ability of the independent variable (Forum Membership) to predict the dependent variable (Hypoglycemic Fear or Parenting Stress) directly and indirectly through a mediating variable (Sleep Quality). The additional impa ct of the indirect path on the direct path is calculated and the resulting significance is used to distinguish if partial mediation (where the resulting path is less significant than the initial path) or full mediation (where the resulting path is no longe r significant) has been achieved. Following recommendations by Preacher and Hayes (2008), the bootstrapped resampling technique was chosen to test the significance of the mediation model; this technique allows for statistical resampling of the collected da ta to estimate direct and indirect effects. Parameter estimates and 95% confidence intervals were calculated based on 5,000 random resamples generated by the SPSS macro; 5,000 resamples were chosen to reduce variability in results, as outlined by Preacher and Hayes (2004). Significance of the indirect effect is determined by the calculation of a 95% confidence interval that does not contain zero.

PAGE 27

27 Table 2 1. Parent demographic v ariables Demographic % Gender Male 5.9 Female 92.2 Race White 93.1 Asian 1.0 Biracial 3.9 Ethnicity Not Hispanic or Latino 89.2 Hispanic or Latino 5.0 Marital Status Married 92.2 Divorced/Separated 4.0 Single 3.9 Highest Level of Education High Sc hool Diploma or Equivalent 3.9 Some College, No Degree 25.5 12.7 38.2 Graduate Degree 19.6 Employment Status Employed 1 39 Hours per Week 29.4 Employed 40 or More Hours per Week 35.3 Not Employed, Looking for Work 4.9 Not Employed, Not Looking for Work 29.4 Disabled/Unable to Work 1.0 Annual Family Income Less than $20,000 4.9 $20,000 $34,999 7.8 $35,000 $49,999 8.8 $50,000 $74,999 22.5 $75,000 $99,999 21.6 $100,000 $149,999 20.6 $150,000 or More 11.8 Chronic Illness Responding Parent Type 1 Diabetes 4.9 Responding Parent Other Chronic Illness 14.7 Spouse Type 1 Diabetes 3.9 Spouse Other Chronic Illness 12.7 Note: Some categories do not add up to 100%. All remaining percentages are from the responses "N/A" or were left blank.

PAGE 28

28 Table 2 2. Child with Type 1 Diabetes demographic v ariables Demographic % First Child with Type 1 Diabetes (n=102) Gender Male 44.1 Female 55.9 Developmental Disability Yes 6.9 No 91.2 Other Chronic Illness Yes 23.5 No 74.5 Insulin Replacement Method Pump 75.5 Injection 23.5 Use Continuous Glucose Monitor Yes 27.5 No 71.6 School Accommodations (504 or IEP) Yes 46.1 No 37.3 Second Child with Type 1 Diabetes (n=5) Gender Male 60.0 Female 40.0 Developmental Disability Y es 0.0 No 100.0 Other Chronic Illness Yes 20.0 No 80.0 Insulin Replacement Method Injection 0.0 Pump 100.0 Use Continuous Glucose Monitor Yes 20.0 No 80.0 School Accommodations (504 or IEP) Y es 40.0 60.0 Note: Some categories do not add up to 100%. All remaining percentages are from the responses "N/A" or were left blank.

PAGE 29

29 CHAPTER 3 RESULTS Preliminary Analyses ssing data resulted in a chi square of 1357.82 (df= 1424; p =.894), indicating that our data was missing completely at random. A bivariate Pearson correlation was conducted to assess the relationship between all dependent and independent variables and demog raphics. Given the high covariance between the worry and behavior subscales on the HFS P, the the hypoglycemic fear behavior subscale was residualized from the hypoglycemic fear worry subscale so to preserve independence of variance and thus reduce multico llinearity. This residualization procedure was also conducted to residualize the parenting stress frequency subscale from the parenting stress difficulty subscale of the PIP to reduce multicollinearity. Bivariate correlations using these residualized varia bles are in Table 3 1. Given the significant correlation with the variables of interest, child age and number of years post diagnosis were controlled for in each successive analysis. An independent samples t test was conducted to assess the impact of paren t gender, child gender, and the number of children with diabetes (one or two) on the variables of interest. No significant differences were found for parent gender or number of children with diabetes. Similarly, no significant differences were found on the variables of interest for child gender, with the exception of scores on the worry subscale of the HFS P in which parents whose child with T1D was a girl ( M =45.5, SD =10.85) endorsed more worry than parents of boys ( M =40.91, SD =10.27); t (100)= 2.18, p = .05

PAGE 30

30 Results for Aim One The goal of aim one was to investigate the demographics and qualitative ratings of the diabetes online community. Inspection of de mographics presented earlier ( Table 2 1) indicate that our sample was composed mainly of non Hispanic wh ite mothers. Detailed information about internet usage was also collected (Table 3 2 ). Participants reported the number of people from the diabetes online community they had met in T1D forums compared to the friendships they had begun in person. Of the total say somewhat worse, 1% say much worse. Regarding diabetes knowledge, the average sc ore was 20.59 ( SD= 1.58; Range = 16 23; n =100). Why Visit the Diabetes Online Community Regarding reasons for visiting the diabetes online community, 18.6% of parents indicated Social Support. Knowledge was indicated as the primary reason for 35.3% of par ents. Finally, 39.2% of parents indicated that knowledge and social support were jointly the biggest benefit of the diabetes online community. The remaining 6.9% of parents had an answer that did not fall into any of the other categories (e.g. "I have a s on with Type 1 Diabetes"). Biggest Benefits of the Diabetes Online Community Of the total participants, 38.2% indicated that the primary benefit of using the diabetes online community was Social Support exclusively. Knowledge was identified as

PAGE 31

31 the primary benefit for 31.4% of the responding parents. Finally, 21.6% of the parents indicated that both social support and knowledge were the primary benefits of the diabetes online community. The remaining 8.8% of parents had an answer that did not fall into any of the other categories (e.g. "Some people argue on the forums"). Does the Diabetes Online Community Impact Care Parents were asked to indicate whether or not their experiences within the diabetes online community impacted the care they provide for their child ; experiences on these forums. An additional 11.8% of parents i ndicated that their activity did not impact care. The remaining participants did not indicate whether or not their activity impacted their care behaviors. Trustworthiness An additional exploratory analysis was conducted based on participant reports of tru st on the internet. Participants were asked to rate the trustworthiness of individuals on the internet and then rate the trustworthiness of individuals within the diabetes online community t Higher scores indicate less trust. A paired samples t test was used to compare parent ratings of trust on the internet to parent ratings of trust within the diabetes online community There was a significant difference in perceived trustworthiness on the internet ( M =2.97, SD =0.94) and on T1D Forums ( M =2.49, SD =0.72); t (100)= 5.621, p < .001. This d =0.57).

PAGE 32

32 Results for Aim Two The second ai m of the current investigation was to examine forum membership. A hierarchical linear regression was conducted to test the hypothesis that parent age and child glycemic control, as measured by H B A1 C would predict number of forum memberships. Total interne t time, child age, and years post diagnosis were entered into block one to serve as control variables. Parent age and child H B A1 C were entered into block two. There was a significant R squared change from 0.02 to 0.08 with the addition of the variables of interest. Examination of the effect demonstrated that parent age was the only significant predictor ( = 0.30) and was positively related to forum membership, H B A1 C was negatively r elated to forum membership ( = 0.01), but this relationship was not significant. Given the significant relationship between parent age and forum membership, it was incorporated as a control variable in the following analyses. Hypoglycemic Fear Two hiera rchical linear regressions were conducted to analyze the impact of forum membership on hypoglycemic fear. For both regressions, total internet time, child age, years post diagnosis, and parent age were entered into block one to serve as control variables; forum membership was entered into block two as the predictor variable for both regressions. In the first regression, there was a significant R squared change from 0.14 to 0.19 when forum membership was added as a predictor. Forum membership significantly p redicted hypoglycemic fear behaviors ( = 0.23; p <.05) with a small effect size, such that as forum membership increased by one standard deviation, fear behaviors also increased by 0.23 of a standard deviation. In the second

PAGE 33

33 regression, forum membership d id not significantly predi ct hypoglycemic fear worry ( Table 3 3). Parenting Stress Two hierarchical linear regressions were conducted to analyze the impact of forum membership on parenting stress. For both regressions, total internet time, child age, yea rs post diagnosis, and parent age were entered into block one to serve as control variables; forum membership was entered into block two as the predictor variable. In the first regression, there was a significant R squared change from 0.15 to 0.22 when for um membership was added as a predictor. Forum membership significantly predicted parenting stress frequency ( = 0.28; p <.01) with a small effect size, such that as forum membership increased by a standard deviation, parenting stress frequency also increas ed by 0.28 of a standard deviation. In the second regression, forum membership did not significantly predict p arenting stress difficulty ( Table 3 4). Results for Aim Three The third and final aim of this study was to examine the indirect effect of sleep q uality on the significant relationships described above. Using the PROCESS macro for SPSS (Hayes, 2012), the indirect effects through sleep quality were assessed using 5000 bootstrapped samples. For both analyses, total internet time, child age, number of years post diagnosis, and parent age were entered as covariates to serve as control variables; effects reported are pooled across the ten imputations. Hypoglycemic Fear Behaviors For a description of the model of the indirect effect, see Figure 3 1 The d irect effect of forum membership on hypoglycemic fear ( c path) was significant ( p <.05).

PAGE 34

34 Additionally, the relationship between forum membership and sleep quality ( a path) was significant ( p <.05) and the relationship between sleep quality and hypoglycemic f ear behaviors ( b path) was not significant ( p >.05). The total indirect effect ( a*b ) was also not significant (Effect = 0.12, Bootstrapped SE = 0.12, 95% confidence interval = [ 0.05, 0.42]). Given the nonsignificant indirect effect, there is no significant mediation. Parenting Stress Frequency For a description of the model of the indirect effect, see Figure 3 2 The direct effect of forum membership on parenting stress frequency ( c path) was significant (p<.05). Additionally, the relationship between foru m membership and sleep quality ( a path) was significant (p<.05) and the relationship between sleep quality and parenting stress frequency ( b path) was significant (p<.01). The total indirect effect ( a*b ) was also significant (Effect = 1.07, Bootstrapped SE = 0.62, 95% confidence interval =[0.25, 2.68]). Given the significant indirect effect, the remaining effect was tested and the relationship between forum membership and parenting stress frequency reached non significance. However, normal theory tests sugg est that this is not a significant mediation ( p =.10). Summary Parents who responded to this survey were majority non Hispanic white females who rated T1D forums as more trustworthy than the internet. Parents also indicated that they gained social suppor t and knowledge from using the forums and that for a majority of parents these forums impacted the care they provided for their child. Parent age was identified as a significant positive predictor of forum membership, while H B A1 C was not significantly rela ted to forum membership. Hypoglycemic fear behaviors and parenting stress frequency were both significantly predicted by forum membership, such that

PAGE 35

35 increased forum membership predicted increased fear behaviors and stress frequency. A significant indirect effect was found for forum membership to parenting stress frequency through sleep quality; however sleep quality did not mediate this relationship. There was no significant indirect effect of mediation for hypoglycemic fear behaviors through sleep quality.

PAGE 36

36 Table 3 1 Correlati ons b etween v ariables Variable 1 2 3 4 5 6 7 8 1. HFS Worry 2. HFS Behavior .05 3. PIP Frequency .37** .48*** 4. PIP Difficulty .22* .16 .04 5. Sleep Quality .33** .23* .44*** .24* 6. Forum Membership .06 .21* .18 .09 .10 7. Years post Diagnosis .05 .06 .22* .13 .12 .11 8. Child Age .09* .21 .12 .28* .10 .06 .47** 9. Annual Income .01 .14 .19 .14 .14 .09 .04 .03 Note: All correlations are pool ed across imputations and controlled for total time on internet. p < .05. ** p < .01. *** p < .001.

PAGE 37

37 Table 3 2. Internet and forum u se per w eek Demographic % Total Time Spent on the Internet 1 to 3 Hours 2.9 4 to 6 Hours 11.8 6 t o 9 Hours 16.7 10 to 14 Hours 21.6 15 to 19 Hours 11.8 20 to 24 Hours 10.8 25 to 29 Hours 11.8 30 to 34 Hours 3.9 35 to 39 Hours 2.9 40 to 44 Hours 3.9 45 to 49 Hours 1.0 75 to 79 Hours 1.0 Total Ti me Spent on Type 1 Diabetes Forums Less than 30 Minutes 7.8 30 Minute to 1 Hour 14.7 1 to 3 Hours 39.2 4 to 6 Hours 10.8 6 to 9 Hours 11.8 10 to 14 Hours 7.8 15 to 19 Hours 3.9 20 to 24 Hours 2.0 25 to 29 Hours 2.0

PAGE 38

38 Table 3 3. Hierarchical linear r egression predicting f orum m embership Variable R 2 R 2 f Step 1 Control Variables .02 .02 0.56 Total Internet Time .04 Child Age .05 Years Post Diagnosis .09 Step 2 Predic tors .08 .06 2.98* Parent Age .30** H B A1 C .01

PAGE 39

39 Table 3 4. Hierarchical l inear r egression p redicting h ypoglycemic f ear and parenting s tress Hypoglycemic Fear Behaviors Hypoglycemic Fear Worry Variable R 2 R 2 f R 2 R 2 f Step 1 Control Variables .14 .14 3.58 .08 .08 2.08 Total Internet Time .29** .10 Child Age .10 .29* Years Post Diagnosis .01 .07 Parent Age .17 .29* Step 2 Predictors .19 .05 5.36* .08 .00 .00 Fo rum Membership .23* .01 Parenting Stress Frequency Parenting Stress Difficulty Step 1 Control Variables .15 .15 3.97 .07 .07 1.73 Total Internet Time .21* .04 Child Age .17 .23* Years Post Diagnosis .22* .02 Parent Age .28* .07 Step 2 Predictors .221 .07 8.42** .07 .00 0.26 Forum Membership .28** .05 Note: p

PAGE 40

40 Figure 3 1. Proposed m ediation m odel of h ypoglycemic fear b ehaviors Forum Membership Hypoglycemic Fear Behaviors Sleep Quality a = 0.14 se=0.07* b = 0.89 se=0.69 c = 0.95 se=0.44* c' = 0.12 se=0.12

PAGE 41

41 Figure 3 2. Proposed m ediation m odel of p arenting s tress f requency Sleep Quality Forum Membership Parenting Stress Frequency a = 0.14 se=0.07* b = 7.81 se=2.37** c = 3.28 se=1.52* c' = 1.07 se=0.65

PAGE 42

42 CHAPTER 4 DISCUSSION The current study is the first of the author's knowledge to directly assess parents of children with T1D who use online T1D support forums. This expands on the only previous study of the diabetes online community (Green, Choudhry, Ki labuk, & Shrank, 2010) by assessing both participant demographics and participant perceived benefits. Non Hispanic white mothers with only one child with T1D were identified as the primary participants; this mirrors T1D clinical and epidemiological researc h participant demographics (Monaghan, Herbert, Cogen, & Streisand, 2012; Plotnick, Clark, Brancati, & Erlinger, 2003; SEARCH for Diabetes in Youth Study Group, 2006) Additionally, results suggest that parents who use the T1D forums place a significant amo unt of trust in the information they receive and the relationships created online. This is exemplified by 50% of all participants indicating they had met an individual from the T1D forums in person, more than half of all participants rating friendships on the T1D forums as better or equal to friendships in person, and 84.3% of participants indicating that their child's health care was influenced by experiences on the T1D forums. Finally, within the current sample more than two thirds of participants spent m ore than one hour on T1D forums a week. This underscores the importance of assessing online interactions, as a majority of participants spend more time on the T1D forums in one week than they might with their child's endocrinologist or other health care pr ofessional. A significant gender difference between parents of females and parents of males on the worry subscale of the HFS P was also detected. While the author is unaware of this difference appearing in other studies using the HFS P, previous research h as noted significant gender differences in the same pattern on the HFS when administered to

PAGE 43

43 adolescents (Gonder Frederick, et al., 2006) While it is possible that some questions on the form maybe more worrisome for girls than boys, and thus invoke higher scores for females, it is also possible that there is a subtle differential treatment of girls and boys by either parents or health care providers. For example, it may be that girls are warned more to avoid certain complications of hypoglycemia and this is responsible for the elevated scores. Additionally, no significant association between the child's H B A1 C and forum membership was found, suggesting that parents who are involved in these forums may not be exclusively searching for short term strategies to improve glycemic control and instead are establishing relationships regardless of thier child's health status (i.e. glycemic control). Important information regarding parental perceptions of T1D forums has not been discussed in previous studies. The resu lts of this examination suggest that parents place a significant amount of trust in the people and information that they find on these T1D forums, both when compared to the internet at large and to the people they meet in real life. While this may be benef icial to parents, who are rating social support as the biggest benefit of their T1D forum membership, it should also be treated with caution. More than half of parents indicated that knowledge was a reason that they participated in the T1D forums and a lar ge majority indicated that these forums impacted the care provided to their children. This identifies the T1D forums as a potentially strong source for information dissemination but also as a potential venue for the unintended spread of misinformation. For um membership was examined in relation to parenting stress, hypoglycemic fear, and sleep quality. This study examined the impact of forum membership on

PAGE 44

44 parenting variables including parenting stress and hypoglycemic fear. It is important to note that while forum membership is a relatively simplistic measure of forum involvement and cohesion, it provides an easily quantifiable measure that is tailored to this medium. The results of these analyses suggest that forum membership significantly impacts the amount of parenting stress and hypoglycemic fear that parents experience. Specifically, hypoglycemic fear behaviors and parenting stress frequency appear to be the variables most susceptible to increasing based on increased forum membership. This is important, a s the hypoglycemic fear behaviors subscale captures behaviors use of these str ategies can lead to poorer disease outcomes (e.g. stress frequency may be indicative of poorer psychosocial functioning of the parent (e.g. Finally, the fact that sleep quality did not mediate the relationships under study suggests that the impact of sleep is more complicated than previously thought. As outlined in the intr oduction, sleep has been linked to cognitive and physical wellbeing and sleep disruptions have been demonstrated in parents of children with T1D. Given this, it was expected that sleep quality would significantly mediate the established relationships of fo rum membership to parenting stress frequency and hypoglycemic fear behaviors. There was a significant indirect effect of forum membership on parenting stress frequency through sleep, suggesting that sleep is an integral part of this relationship, however s ignificant mediation was not obtained, possibly due to measure

PAGE 45

45 restrictions. This relationship can hopefully become clearer with a larger sample size, a sleep measure with improved internal consistency, and more tailored measures. Limitations While this s tudy provides necessary pilot data for future analyses, it is not without limitations. The primary limitation of the study was the cross sectional and self report nature of the analysis. By recruiting and assessing participants online, it is possible that the demographics collected are more representative of the users willing to participate in research and not the demographics of the website as a whole. However, research suggests that internet based samples are comparable to paper and pencil samples in term s of psychosocial wellbeing (Gosling, Vazire, Srivastava, & John, 2004). Additionally, our sample's demographics were similar to those seen in larger epidemiological studies. The use of self to objectively as sess the impact of the measured variables on the participant and the cross sectional nature of the study does not allow for exploration of causation. Finally, it and par enting stress before participating in forums, nor did it include a group that were not members of internet forums, and thus causation cannot be implied or directly tested. Future Directions The current study serves as a first step towards the qualitative description of and the quantitative assessment of online forums for parents of children with T1D. Future research should more directly assess forum involvement and perceptions of group cohesion. The current study used forum membership for this purpose, but a detailed assessment regarding participant's perceptions of their involvement in the group may allow for more insight into the variables discussed in this paper. Additionally,

PAGE 46

46 longitudinal studies are also needed to expand the current research. While th e current study could not assess the impact that forum membership has on H B A1 C it is notable that the H B A1 C in this study was lower than in many clinic based populations. Whether this is an effect of the increased knowledge and self efficacy gained from f orum participation or if this is representative of the characteristics of participants before they use the forums should be assessed. Finally, child report data regarding perceptions of parental involvement in care should also be assessed. Previous resear ch has suggested that child behavior is important to assess when looking at T1D health outcomes; child internalizing behavior problems, externalizing behavior problems, and family adaptability have all been linked to glycemic control outcomes (Cohen, Lumle y, Naar King, Partridge, & Cakan, 2004) Additionally, family cohesion and critical parenting have been linked to child disease outcomes; specifically, behaviors that are perceived by youth as "nagging" or "criticizing" have been linked to poorer disease o utcomes (Cohen, Lumley, Naar King, Partridge, & Cakan, 2004; Duke, Geffken, Lewin, Williams, Storch, & Silverstein, 2008) Since many of the behaviors measured by the HFS Behavior scale and the PIP Frequency scale could be perceived this way, it is possib le that there may be a negative impact on child disease outcomes that is increased by forum involvement. The current study demonstrates that parents who use T1D forums perceive benefit, especially in the domains of social support and knowledge. Clinicians should be aware of this and talk openly with the parents of their patients to explain the potential risks and benefits of participation in online forums. Parent education regarding the potential impact of forum use on variables such as hypoglycemic fear a nd parenting

PAGE 47

47 stress should be provided and parents should be encouraged to verify information gained from online forums with their health care team. Additionally, forum sponsors and administrators should be aware of the potential effects of information sha red on their treatment; providing timely and factual information along with necessary disclaimers regarding information gained from these forums should be instituted as stand ard of practice.

PAGE 48

48 LIST OF REFERENCES Abid, N., Porter, L., Day, E., Krone, N., Hogler, W., Kirk, J., et al. (2011). Differences in metabolic effects of twice daily versus multiple daily insulin injections in children with Type 1 Diabetes. Practical Diabete s, 28 384 387. Akerblom, H. K., Saukkonen, T. T., Vaarala, O., Savilahti, E., Paganus, A., Virtanen, S. M., et al. (1993). The case for elimination of cow's milk in early infancy in the prevention of Type 1 Diabetes: The Finnish experience. Diabetes/Metab olism Reviews, 9 269 278. American Diabetes Association. (2003). Economic costs of diabetes in the U.S. in 2002. Diabetes Care 26 917 932. Ayers, S. L., & Kronenfeld, J. J. (2007). Chronic illness and health seeking information on the internet. Health: An interdisciplinary Journal for the Social Study of Health, Illness and Medicine, 11 327 347. Barrera Jr, M. B., Glasgow, R. E., McKay, H. G., Boles, S. M., & Feil, E. G. (2002). Do internet based support interventions change perceptions of social suppo rt?: An experimental trial of approaches for supporting diabetes self management. American Journal of Community Psychology, 30 637 654. Bener, A., Alsaied, A., Al Ali, M., Al Kubaisi, A., Basha, B., Abraham, A., et al. (2009). High prevalence of vitamin D deficiency in Type 1 Diabetes Mellitus and healthy children. Acta Diabetologica, 46 183 189. Blom, G. (1958). Statistical Estimates and Transformed Beta Variables. New York, NY: Wiley. Boyd, D.M., & Ellison, N.B. (2007). Social network sites: Definition history, and scholarship. Journal of Computer Mediated Communication, 13, 210 230. Brown, M., Ahmed, M. L., Clayton, K. L., & Dunger, D. B. (1994). Growth during childhood and the final height in Type 1 Diabetes. Diabetic Medicine, 11 182 187. Bryant, P .A., Trinder, J., & Curtis, N. (2004). Sick and tired: Does sleep have a vital role in the immune system? Nature Reviews Immunology, 4 457 467. Buyssee, D.J., Reynolds, C.F., Monk, T.H., Berman, S.R., & Kupfer, D.J. (1989). The Pittsburg sleep quality i ndex: A new instrument for psychiatric practice and research. Psychiatry Research, 28, 193 213. Carpenter, J.S., & Andrykowski, M.A. (1998). Psychometic evaluation of the Pittsburgh sleep quality index. Journal of Psychocomatic Research, 45 5 13.

PAGE 49

49 Center f or Disease Control and Prevention. (2011). National diabetes face sheet: National estimates and general information on diabetes and prediabetes in the United States, 2011 U.S. Department of Heatlh and Human Services, Centers for Disease Control and Preven tion. Childhood Diabetes in Finland Study Group. (1992). Feeding in infancy and the risk of Type 1 Diabetes Mellitus in Finnish children. Diabetic Medicine, 9 815 819. Clark, W.L., Gonder Frederick, L.A., Snyder, A.L., & Cox, D.J. (1998). Maternal fear o f hypoglycemia in their children with insulin dependent Diabetes Mellitus. Journal of Pediatric Endocrinology and Metabolism, 11, 189 194. Cohen, D.M., Lumley, M.A., Naar King, S., Partridge, T., & Cakan,N. (2004). Child behavior problems and family funct ioning as predictors of adherence and glycemic control in economically disadvantaged children with Type 1 Diabetes: A prospective study. Journal of Pediatric Psychology, 29, 171 184. Coulson, N.S. (2005). Receiving social support online: An analysis of a c omputer mediated support group for individuals living with Irritable Bowel Syndrome. CyberPsychology & Behavior, 8, 580 584. Davison, K.P., Pennebaker, J.W., & Dickerson, S.S. (2000). Who talks? The social psychology of illness support groups. American Psy chologist, 55, 205 217. Duke, D.C., Geffken, G.R., Lewin, A.B., Williams, L.B., Storch, E.A., & Silverstien, J.H. (2008). Glycemic control in youth with Type 1 Diabetes: Family predictors and mediators. Journal of Pediatric Psychology, 33, 719 727. Fitzger ald, J.T., Funnell, M.M., Hess, G.E., Barr, P.A., Anderson, R.M, Hiss, R.G., Davis, W.K. (1998). The reliability and validity of a brief diabetes knowledge test. Diabetes Care, 21, 706 710. Gonder Frederick,L.A., Fisher, C.D., Ritterband, L.M., Cox, D.J., Hou, L., DasGupta, A.A., Clarke, W.L. (2006). Predictors of fear of hypglycemia in adolescents with Type 1 Diabetes and their parents. Pediatric Diabetes, 7, 215 222. Gosling, S.D., Vazire, S., Srivastava, S., & John, O.P. (2004). Should we trust web base d studies? A comparatice analysis of siz preconceptions about internet questionnaires. American Psychologist, 59 (2), 93 104. Greene, J.A., Choudhry, N.K., Kilabukm, E., & Shrank, W.H. (2010). Online social networking by patients with diabetes: A qualitat ive evaluation of communication with Facebook. Journal of General Internal Medicine, 26 287 292. Griffiths, F., Lindenmeyer, A., Powell, J., Lowe, P., & Thorogood, M. (2006). Why are health care interventions delivered over the internet? A systematic revi ew of the published literature. Journal of Medical Internet Research, 8 e10.

PAGE 50

50 Hayes, A.F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from: http://www.afhayes.com/public/process2012.pdf Hovorka, R., Allen, J. M., Elleri, D., Chassin, L. J., HArrie, J., Xing, D., et al. (2010). Manual closed loop insulin delivery in children and adolescents with Type 1 Diabetes: A phase 2 randomised crossover trial. The Lancet, 375 743 751. IBM Corportation. (2011) IBM SPSS Statistics for Windows, Version 20. Chicago, IL: SPSS Inc. Jones, T.W., & Davis, E.A. (2003). Hypoglycemia in children with Type 1 Diabe tes: Current issues and controversies. Pediatric Diabetes, 4 143 150. Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group. (2008). Continuous glucose monitoring and intensive treatment of Type 1 Diabetes. New England Journal of Medicine, 359 1464 1476. Kiess, W., Kapellen, T., Siebler, T., Deutscher, J., Raile, K., Dost, A., et al. (1998). Practical aspects of managing preschool children with Type 1 Diabetes. Acta Paediatrica (425), 67 71. Landolt, M.A., Vollrath, M., Laimbacher J., Gnehm, H.E., Sennhauser, F.H. (2005). Prospective study of Posttraumatic Stress Disorder in parents of children with newly diagnosed Type 1 Diabetes. Journal of the American Academy of Child & Adolescent Psychiatry, 44, 682 689. Lieberman, M.A., Gol ant, M., Giese Davis, J., Winzlenberg, A., Benjamin, H., Humphreys, K.,... Spiegel, D. (2003). Electronic support groups for breast carcinoma: A clinical trial of effectiveness. Cancer, 97, 920 925. Lieberman, M.A., & Goldstein, B.A. (2006). Not all negati ve emoations are equal: The role of emotional expression in online support groups for women with breast cancer. Psycho Oncology, 15, 160 168. Lopez Wagner, M.C., Hoffman, C.D., Sweeney, D.P., Hodge, D., & Gilliam, J.E. (2008). Sleep problems of parents of typically developing children and parents of children with autism. Journal of Genetic Psychology, 169, 245 259. Ly, T. T., Gallego, P. H., Davis, E. A., & Jones, T. W. (2009). Impaired awareness of hypoglycemia in a population based sample of children and adolescents with Type 1 Diabetes. Diabetes Care, 32 1802 1806. Meltzer, L.J. (2008). Brief report: Sleep in parents of children with autism spectrum disorders. Journal of Pediatric Psychology, 33, 380 386.

PAGE 51

51 Miller, J. A. (1999). Impact of hyperglycemia on the Renin angiotensin system in early human Type 1 Diabetes Mellitus. Journal of the American Society of Nephrology, 10 1778 1785. Moltchanova, E. V., Schreier, N., Lammi, N., & Karvonen, M. (2009). Seasonal variation of diagnosis of Type 1 Diabetes Melli tus in children worldwide. Diabetic Medicine, 26 673 678. Monaghan, M., Herbert, L. J., Cogen, F. R., & Streisand, R. (2012). Sleep behaviors and parent functioniong in young children with Type 1 Diabetes. Children's Health Care, 41 246 259. Monaghan, M. Hilliard, M. E., Cogen, F. R., & Streisand, R. (2011). Supporting parents of very young children with Type 1 Diabetes: Results from a pilot study. Patient Education and Counseling (82), 271 274. Plotnick, L.P., Clark, L.M., Brancati, F.L., & Erlinger, T. (2003). Safety and effectiveness of insulin pump therapy in children and adolescents with Type 1 Diabetes. Diabetes Care 26 1142 1146. Preacher, K.J., & Hayes, A.F. (2008). Asymptomatic and resampling strategies for assessing and comparing indirect effec ts in multiple mediator models. Behavior Research Methods, 40, 879 891. Preacher, K.J., & Hayes, A.F. (2004).SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, and Computers, 36 717 731. Reeves, P.M. (2001). How individuals coping with HIV/AIDS use the internet. Health Education Research, 16, 709 719. Reichard, P., Nilsson, B., & Rosenqvist, U. (1993). The effect of long term intensified insulin treatment on the development of microva scular complications of diabetes mellitus. The New England Journal of Medicine, 329 304 309. Ritterband, L. M., Gonder Frederick, L. A., Cox, D. J., Clifton, A. D., West, R. W., & Borowitz, S. M. (2003). Internet interventions: In review, in use, and into the future. Professional Psychology: Research and Practice, 34 527 534. Ryan, C.M., & Becker, D.J. (1999). Hypoglycemia in children with Type 1 Diabetes Mellitus: Risk factors, cognitive function, and management. Endocrinology and Metabolism Clinics of N orth America, 28, 883 900. SEARCH for Diabetes in Youth Study Group. (2006). The burden of Diabetes Mellitus among U.S. youth: Prevalence estimates from the SEARCH for diabetes in youth study. Pediatrics, 118, 1510 1518.

PAGE 52

52 SEARCH for Diabetes in Youth Study Group. (2012). Projections of Type 1 and Type 2 Diabetes burden in the US population aged <20 years through 2050: Dynamic modeling of incidence, mortality, and population growth. Diabetes Care, 35 2515 2520. STAR 3 Study Group. (2010). Effectiveness of s ensor augmented insulin pump therapy in Type 1 Diabetes. The NEw England Journal of Medicine, 363 311 320. Streisand, R., Braniecki, S., Tercyack, K.P., & Kazak, A.E. (2001). Childhood illness related parenting stress: The pediatric inventory for parents. Journal of Pediatric Psychology, 26 155 162. Streisand, R., Mackey, E. R., Elliot, B. M., Mednick, L., Slaughter, I. M., Turek, J., et al. (2008). Parental anxiety and depression associated with caring for a child newly diagnosed with Type 1 Diabetes: Op portunities for education and counseling. Patient Education and Counseling, 73 333 338. Streisand, R., Swift, E., Wickmark, T., Chen, R., & Holmes, C. S. (2005). Pediatric parenting stress among parents of children with Type 1 Diabetes: The role of self e fficacy, responsibility, and fear. Journal of Pediatric Psychology, 30 513 521. Sullivan Bolyai, S., Deatrick, J., Gruppuso, P., Tamborlane, W., & Grey, M. (2003). Constant vigilance: Mothers' work parenting young children with Type 1 Diabetes. Journal of Pediatric Nursing, 18 21 29. Taborsky, G.J., Ahren, B., & Havel, P.J. (1998). Autonomic mediation of glucagon secretion during hypoglycemia: Implications for impaired alpha cell responses in Type 1 Diabetes. Diabetes, 47, 995 1005. Tworoger, S.S., Davis, S., Vitiello, M.V., Lentz, M.J., & McTiernan, A. (2005). Factors associated with objective (actigraphic) and subjective sleep quality in young adult women. Journal of Psychosomatic Research, 59, 11 19. Van Dongen, H.P.A., Maislin, G., Mullington, J.M., & Dinges, D.F. (2003). The cumulative cost of additional wakefulness: Dose response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. SLEEP, 26, 117 126. Wangberg, S. C. (2007). An internet based diabetes self care intervention tailored to self efficacy. Health Education Research, 23 170 179. Warshaw, H., Hernandez, M., Tenderich, A., & Edelman, D. (2011). The DOC (Diabetes Online Community): What the heck is going on? [PowerPoint Slides]. T alk presented at the 2011 American Association of Diabetes Educators, Las Vegas, Nevada.

PAGE 53

53 Weinert, C. (2000). Social support in cyberspace for women with chronic illness. Rehabilitation Nursing, 25, 129 135. Weinzimer, S.A., Ahern, J.H., Doyle, E.A., Vince nt, M.R., Dziura, J., Steffen, A.T., Tamborlane, M.W. (2004). Persistence of benefits of continuous subcutaneous insulin infusion in very young children with Type 1 Diabetes: A follow up report. Pediatrics, 114, 1601 1605. White, M., & Dorman, S.M. (2001). Receiving social support online: Implications for health education. Health Education Research, 16, 693 707. Williams, B. K., Guelfi, K. J., Jones, T. W., & Davis, E. A. (2011). Lower cardiorespiratory fitness in children with Type 1 Diabetes. Diabetic Med icine, 28 1005 1007. Winkley, K., Landau, S., Eisler, I., & Ismail, K. (2006). Psychological interventions to improve glycaemic control in patients with Type 1 Diabetes: Systematic review and meta analysis of randomised controlled trails. BMJ Group.

PAGE 54

54 BIO G RAPHICAL SKETCH Amanda Marie Balkhi graduated as class salutatorian from Palm Beach Lakes High School in West Palm Beach, Florida in 2008. She then attended the University of Florida, where she majored in psychology with a minor in l eadership. While at the University of Florida, she founded and edited an undergraduate psychology research journal while also serving as a group co leader for an adult weight management intervention in Gainesville, Florida. Additionally, she completed an independent research pro ject in her third year which examined perceived weight discrimination and patient satisfaction in a sample of South Florida adults. She graduated fro m the University of Florida in the s pring of 2011. Following graduation, Amanda entered the Department of Clinical and Health Psychology at the University of Florida. She is currently in the second year of her training with Dr. Gary R. Geffken. Her first two years have been spent as a student clinician in a treatment outcome study for anxiety disorders, under the supervision of Drs. Gary Geffken and Joseph McNamara. Additionally, Amanda has had the honor of co leading a group of dedicated undergraduates in a variety of projects with the aim of interdisciplinary collaboration between psychiatry and psychology. A manda recently finished a public policy research project in collaboration with the Bob Graham Center for Public Service, which she used to extend her internet use research to an undergraduate population. Her research interests include the intersection of t echnology and treatment outcome with a specific focus on pediatric populations. Her long term goals include the development of risk reduction interventions for parents of children with chronic illnesses who use online forums for social support.