Citation
Sleep & Social-Ecological Influences of Quality of Life in Overweight Rural Youth Using a Risk-Resistance Framework

Material Information

Title:
Sleep & Social-Ecological Influences of Quality of Life in Overweight Rural Youth Using a Risk-Resistance Framework
Creator:
Graef, Danielle M
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (99 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology
Clinical and Health Psychology
Committee Chair:
JANICKE,DAVID
Committee Co-Chair:
MCCRAE,CHRISTINA SMITH
Committee Members:
WIENS,BRENDA A
MATHEWS,ANNE
Graduation Date:
12/19/2014

Subjects

Subjects / Keywords:
Child psychology ( jstor )
Munchausen syndrome by proxy ( jstor )
Obesity ( jstor )
Overweight ( jstor )
Parents ( jstor )
Pediatrics ( jstor )
Preliminary proxy material ( jstor )
Proxy reporting ( jstor )
Proxy statements ( jstor )
Quality of life ( jstor )
Clinical and Health Psychology -- Dissertations, Academic -- UF
duration -- obesity -- pediatric -- sleep -- treatment-seeking
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Psychology thesis, Ph.D.

Notes

Abstract:
Both being obese and having insufficient sleep increases risk for poorer quality of life (QOL). There is limited information regarding how these conditions interact and who is at greatest risk for poorer QOL. The study objectives were to examine the influences of sleep and hypothesized moderating variables of peer support, family functioning, and parental distress on parent-proxy and child self-reported QOL. Participants included 143 obese children (8-12 years) and their parents participating in a weight-management study. Demographics, anthropometrics, objectively measured sleep, and subjective measures of child QOL, child-reported peer support, and parent-reported distress and family functioning were collected prior to treatment. Approximately 88% of children obtained less than 8 hours of sleep and the mean sleep efficiency was 83.29%. Lower family income, higher child BMI z-score, and poorer family functioning were associated with lower parent-proxy QOL. There was a significant interaction between peer support and total sleep time in predicting child-reported QOL, such that there was a positive association between sleep and QOL when support was lower. There was a main effect of family functioning, such that poorer functioning was associated with lower parent-proxy QOL. There were no other main or interaction effects. Child- and parent-reported child sleep problems were associated poorer child QOL compared to children who did not experience sleep troubles.Consistent with previous research, a significant number of children in the current study obtained insufficient sleep. Obese children in our study obtained poorer sleep efficiency despite non-significantly different weekday total sleep than non-obese children in previous research, suggesting that lower sleep in obese populations does not necessarily equate with sleep efficiency and that examining the role of sleep fragmentation is important to consider. The relationship between child sleep, positive peer support, and QOL in our study suggests that improved peer support and sleep may be associated with improvements in well-being. The next step in research should include a multi-method, prospective assessment of sleep to examine subjective child sleep problems, sleep diaries, and well-validated objective sleep measures. This information is needed before drawing definitive conclusions on these relationships. ( en )
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.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: JANICKE,DAVID.
Local:
Co-adviser: MCCRAE,CHRISTINA SMITH.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-12-31
Statement of Responsibility:
by Danielle M Graef.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
12/31/2016
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

SLEEP & SOCIAL ECOLOGICAL INFLUENCES OF QUALITY OF LIFE IN OVERWEIGHT RURAL YOUTH USING A RISKRESISTANCE FRAMEWORK By DANIELLE M. GRAEF A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PART IAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014 1

PAGE 2

2014 Danielle M. Graef 2

PAGE 3

ACKNOWLEDGMENTS I would like to acknowledge my mentor, Dr. David Janicke, for his guidance and support of my professional development throughout my training at the University of Florida Additionally I want to recognize the members of my committee who have also been important in the development of my project : Dr. Anne Mathews, Dr. Christina McCrae, and Dr. Brenda Wiens. I lastly want to thank my family, friends, and colleagues for their support and advice throughout my work on this project and my graduate training 3

PAGE 4

TABLE OF CONTENTS P age ACKNOWLEDGMENTS .................................................................................................. 3 LIST OF TABLES ............................................................................................................ 6 LIST OF FIGURES .......................................................................................................... 7 ABSTRACT ..................................................................................................................... 8 CHAPTER 1 INTRODUCTION ........................................................................................................ 10 Pediatric Obesity & Associated Health and Psychosocial Consequences .............. 10 Pediatric Obesity and Quality of Life (QOL) ............................................................ 11 The Role of Sleep in the Pediatric Biopsychosocial Functioning ............................ 13 Sleep & Obesity Relationship ................................................................................. 16 Mechanisms of the SleepWeight Relationship ....................................................... 18 Quality of Life in Overweight Children with Insufficient Sl eep ................................. 20 Resilience Theoretical Model & Adaptation to Chronic Conditions ......................... 21 Risk and Resistance Variables in Pediatric Chronic Ill ness Literature .................... 21 Risk Resistance Application to Sleep & Psychosocial Functioning in Obese Children ............................................................................................................... 24 Assessment of Pediatric Sleep ............................................................................... 26 Polysomnography ............................................................................................. 26 Sleep Diaries .................................................................................................... 27 Actigraphy ........................................................................................................ 28 Accelerometry .................................................................................................. 29 Conclusions ...................................................................................................... 31 Purpose of the Study .............................................................................................. 31 Aims and Hypotheses ............................................................................................. 32 Describing Sample Characteristics, Sleep, QOL, and Psychosocial Functioning .................................................................................................... 32 Examining HealthRelated and Social Ecological Predictors of Child QOL ...... 34 Exploratory Aims .............................................................................................. 35 2 METHODS ................................................................................................................. 37 Description of the Larger Study .............................................................................. 37 Procedure ............................................................................................................... 37 Participants ............................................................................................................. 38 Measures ................................................................................................................ 39 Demographic Information ................................................................................. 39 Height and Weight ............................................................................................ 39 Pediatric Quality of Life .................................................................................... 40 4

PAGE 5

Objective Pediatric Sleep ................................................................................. 40 Subjec tive Child Sleep Problems ..................................................................... 42 Peer Support .................................................................................................... 42 Family Functioning ........................................................................................... 43 Parent Psychosocial Functioning ..................................................................... 44 Statistical Analysis Plan .......................................................................................... 44 Analyses of Sample Characteristics, Sleep, QOL, and Psychosoc ial Functioning .................................................................................................... 44 Analyses Examining Health and Social Ecological Predictors of Child QOL .... 46 Exploratory Aims .............................................................................................. 48 3 RESULTS ................................................................................................................... 49 Analyses of Participant Characteristics and Sleep Behaviors ................................. 49 Ai m 1: Participant Characteristics & Description of Child Functioning .............. 49 Aim 2: Describing Child QOL & SocialEcological Functioning ......................... 51 Aim 3: Sociodemographic Differences Across Primary Predictor and Outcome Variables ........................................................................................ 52 Moderation and Mediation Analyses ....................................................................... 53 Aim 4: Moderator Analyses .............................................................................. 53 Aim 5. Mediation Analyses ............................................................................... 55 Exploratory Analyses .............................................................................................. 56 Aim 6 ................................................................................................................ 56 Aim 7 ................................................................................................................ 57 4 DISCUSSION ............................................................................................................. 70 Aim 1: Desc ribing Child Sleep Behaviors ................................................................ 70 Aim 2: Describing Child QOL and Psychosocial Functioning .................................. 74 Aim 3: Sociodemographic Differences in Primary Predictor and Outcome Variables .............................................................................................................. 76 Aim 4: Sleep and Social Ecological Predictors of Child QOL .................................. 77 Aim 5: Insulin Resis tance as a Mediator Predicting QOL ........................................ 81 Study Limitations .................................................................................................... 83 Clinical Implications and Future Directions ............................................................. 85 LIST OF REFERENCES ............................................................................................... 87 BIOGRAPHICAL SKETCH ............................................................................................ 99 5

PAGE 6

LIST OF TABLES Table page 3 1 Demographic characteristics of sample. ............................................................. 58 3 2 Mean sleep variables across all days and time of measurement ........................ 59 3 3 Mean sleep variables separately across day of the week and school break ...... 60 3 4 Means and standard deviations of predictor and outcome variables .................. 62 3 5 Intercorrelations of predictor and outcome variables .......................................... 63 3 6 Total sleep time and social ecological predictors of child self reported total QOL .................................................................................................................... 64 3 7 Conditional effect of TST on child self reported QOL at values of the moderator ........................................................................................................... 64 3 8 Total wake time and social ecological predictors of child self reported total quality of life ........................................................................................................ 66 3 9 Total sleep time and social ecological predictors of parent proxy of child quality of life ........................................................................................................ 67 3 10 Total wake time and social ecological predictors of parent proxy of child quality of life ........................................................................................................ 67 3 11 Indirect effect of total sleep time through HbA1c on child and parent proxy QOL .................................................................................................................... 68 3 12 Indirect effect of total wake time through HbA1c on child and parent proxy QOL .................................................................................................................... 68 3 13 Mean QOL of obese children with and without childreported sleep troubles ..... 69 3 14 Mean QOL of obese children with and without parent reported sleep troubles .. 69 6

PAGE 7

LIST OF FIGURES Figure P age 3 1 Main effect of school break on time of sleep onset ............................................. 61 3 2 Main effects of day and school break on time of sleep offset ............................. 61 3 3 Day by time interaction on average wake after sleep onset (WASO) ................. 62 3 4 Interaction of total sl eep time and peer support in predicting self reported quality of life ................................................................................................................... 65 7

PAGE 8

A bstract 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 SLEEP & SOCIAL ECOLOGICAL INFLUENCES OF QUALITY OF LIFE IN OVERWEIGHT RURAL YOUTH USING A RISK RESISTANCE FRAMEWORK By Danielle M. Graef December 2014 Chair: David M. Janicke Major: Psychology Both being obese and having insufficient sleep increases risk for poorer quality of life ( QOL ) There is limited information regarding how these conditions interact and who is at greatest risk for poorer QOL. The study objectives were to examine the influences o f sleep and hypothesized moderating variables of peer support, family functioning, and parental distress on parent proxy and child self reported QOL. Participants included 143 obese children (8 12 years) and their parents participating in a weight managem ent study. Demographics, anthropometrics, objectively measured sleep, and subjective measures of child QOL, child reported peer supp ort, and parent reported distress and family functioning were collected prior to treatment. Approximately 88% of children o btained less than 8 hours of sleep and the mean slee p efficiency was 83.29% Lower family income, higher child BMI z score, and poorer family functioning were associated with lower parent proxy QOL There was a significant interaction between peer support and total sleep time in predicting child reported QOL, such that there was a positive association between sleep and QOL when 8

PAGE 9

support was low er There was a main effect of family functioning, such that poorer functioning was associated with lower parent pro xy QOL. There were no other main or interaction effects. C hild and parent reported child sleep problems were associated poorer child QOL compared to children who did not experience sleep troubles. Consistent with previous research, a significant number o f children in the current study obtained insufficient sleep Obese children in our study obtained poorer sleep efficiency despite nonsignificantly different weekday total sleep than nonobese children in previous research, suggesting that lower sleep in obese populations does not necessari ly equate with sleep efficiency and that examining the role of sleep fragmentation is important to consider. T he relationship between child sleep, positive peer support, and QOL in our study suggests that improved peer su pport and sleep may be associated with improvements in well being. The next step in research should include a multi method prospective assessment of sleep to examine subjective child sleep problems, sleep diaries, and well validated objective sleep measur es. This in formation is needed before drawing definitive conclusions on these relationships 9

PAGE 10

CHAPTER 1 INTRODUCTION Pediatric Obesity & Associated Health and Psychosocial Consequences The prevalence of high adiposity in children in the United States is of great health concern. Nearly 32% of children and adolescents (ages 2 to 19 years) from 20092010 were overweight or obese (i.e., body mass index (BMI) at or above the 85th percentile for age and gender) and 16.9% were considered obese (i.e., BMI at or above the 95th percentile) ( Ogden, Carroll, Kit, & Flegal, 2012 ) Researchers have consistently found demographic differences in child risk for higher weight status. Compared to nonHispanic white youth, the prevalence of overweight and obesity is significantly higher amongst non Hispa nic black (males: OR=1.27; females: OR=1.99) and Hispanic individuals (males: OR=1.81; females: OR=1.47) ( Ogden et al., 2012) Additionally, male s are more likely to be obese than females and 12year trends reveal that the significant increases in the prevalence of obesity is more pronounced in males, with a 1 .03 annual increase in odds of obesity ( Ogden et al., 2012 ) The prevalence and trend in pediatric obesity is significant given the health consequences associated with high weight status. Children who are overweight are at increased risk for high cholesterol, triglycerides, and glucose ( Freedman, Dietz, Srinivasan, & Berenson, 1999 ; Garnett, Baur, Srinivasan, Lee, & Cowell, 2007) as well as hypertension and metabolic syndrome ( Freedman et al., 1999) Moreover, those who are overweight during middle childhood are approximately seven times more likely to exhibit risk factors for cardiovascular disease in adolescence, including higher cholesterol, triglycerides, fasting blood glucose, insulin resistance, and hypertension ( Garnett et a l., 2007 ) Pediatric obesity is associated with poorer health status in 10

PAGE 11

adulthood. Children who are overweight or obese are at increased probability for overweight (5286%) and obesity (1459%) in adulthood compared to nonoverweight peers, placing them at greater risk of mortality in adulthood ( Guo, Wu, Chumlea, & Roche, 2002; Raitakari, Juonala, & Viikari, 2005; Whitaker, Wright, Pepe, Seidel, & Dietz, 1997) In addition to health consequences of high weight status, children who are obese are also at greater risk of psychosocial difficulties compared to nonove rweight peers. Obese individuals are at increased risk for low self esteem ( McClure, Tanski, Kingsbury, Gerrard, & Sargent, 2010; Zeller, Saelens, Roehrig, Kirk, & Daniels, 2004) and negative perceptions of their physical appearance and self worth ( Braet, Mervielde, & Vandereycken, 1997) compared to nonoverweight peers. Children who are obese are also more likely to report experiencing peer victimization ( Janicke et al., 2007; Janssen, Craig, Boyce, & Pickett, 2004) with the type of peer victimization differing across gender. Specifically, obese boys are more likely to report overt victimization and females are more likely to report relational victimization when compared to nonobese peers ( Pearce, Boergers, & Prinstein, 2002) The physical and psychosocial consequences of pediatric obesity can ultimately impact the self perceived well being in children and adolescents. Pediatric Obesity and Quality of Life (QOL) The World Health Organization (WHO) considers health a dynamic state of complete physical, mental, spiritual, and social well being and not merely the absence of disease and infirmity ( Grad, 2002) The assessment of pediatric healthrelated quality of life (i.e., QOL) has become an important measure in patient health outcomes, as it provides a more detailed understanding of the experiences children with chronic 11

PAGE 12

health conditions have across a variety of domains, including total, physical health, and psychosocial functioning (i.e., emotional, social, and school) ( Franciosi et al., 2012; Varni, Burwinkle, Seid, & Skarr, 2003; Varni, Seid, & Rode, 1999) Examination of QOL provides unique inform ation about a patients well being and contributes to treatment outcome research regarding functional and health status progress in children with chronic conditions. Quality of life measurement provides healthcare providers with an understanding of the unique domains of functioning that individual patients and parents find the most important. In line with viewpoints of the WHO, researchers in the previous two decades have examined QOL across pediatric health conditions, including overweight and obesity. A number of studies have consistently shown a negative association between weight sta tus and quality of life in children. Seminal work by Schwimmer and colleagues ( 2003) found that very obese children and adolescents (i.e., on average above 97th percentile) report QOL significantly lower than that of healthy children and similar to that of children with cancer. Following these findings, others have found that cli nical and community based samples of overweight or obese children are more likely to report lower QOL across general, physical health, and psychosocial domains when compared to nonoverweight peers ( Friedlander, Larkin, Rosen, Palermo, & Redline, 2003; Pinhas Hamiel et al., 2006 ; Shoup, Gattshall, Dandamud i, & Estabrooks, 2008 ; Williams, Wake, Hesketh, Maher, & Waters, 2005) Researchers also have examined QOL of overweight children compared to other chronic illness groups. Children who are obese rep ort lower QOL than children with asthma, cardiac diseases, diabetes, gastrointestinal conditions, and long term survivors of cancer, as well as QOL 12

PAGE 13

comparable to children undergoing treatment for cancer (i.e., acute lymphoblastic leukemia) ( Varni, Limbers, & Burwinkle, 2007 ) Although a subset of the population report significantly greater impairments compared to nonoverweight peers, some children who are o verweight or obese report QOL and psychosocial functioning similar to that of healthy and nonoverweight peers ( Zametkin, Zoon, Klein, & Munson, 2004) The substantial variability within obese populations suggests that there are likely important factors that moderate the relationship between weight status and QOL in children Psychosocial predict ors of child and parent proxy reports of QOL in this population include increased weight related teasing, peer victimization, higher parental stress, and lower perceived social support ( Guilfoyle, Zeller, & Modi, 2010; Janicke et al., 2007; Zeller & Modi, 2006) Despite increased research examining psychosocial influences of child QOL, predictors explain a limited degree of the variance of QOL in children who are obese, suggesting that there may be other predictors that have yet to be accounted for. One area that has received increased attention that may accou nt for poorer QOL is poor or insufficient sleep. The Role of Sleep in the Pediatric Biopsychosocial Functioning Sleep plays an important and unique role in the functioning in children and adolescents, as it can impact health and functioning across physic al, cognitive, emotional, and social domains ( Mindell & Owens, 2010) Insufficient sleep in pediatric populations is of increasing concern. It is recommended that individuals in middle childhood (i.e., ages 6 to 12 years) obtain betw een nine and eleven hours of sleep per night ( Mindell & Owens, 2010) ; however, there has been a 1.5 to 2 hour reduction in child sleep duration in the previous 50 years and the most rapid declines occurring on 13

PAGE 14

weekdays when compared to weekends ( Jenni, Molinari, Caflisch, & Largo, 2007 ; Leproult & Van Cauter, 2010; Snell, Adam, & Duncan, 2007) Although these sleep trends occur across age groups, there are more significant declines for children as they age. The majority of younger children obtain 10 hours of sleep on average; however, by th e age of 7 years children get less than 10 hours of sleep on weekdays and by the age of 14 years youth obtain an average of 7.55 hours of sleep ( Snell et al., 2007 ; Weiss et al., 2010) Other significant demographic differences in child sleep include gender and race, as males sleep fewer hours than females and black children sleep fewer nighttime hours than white children ( Crosby, LeBourgeois, & Harsh, 2005; Lumeng et al., 2007; Spilsbury et al., 2004) Data suggest, however, that there are no significant racial differences in total sleep time that includes both nighttime sleep and daytime napping ( Crosby et al., 2005) In addition to trends towards short sleep duration, approximately 25 to 40% of children experience general sleep difficulties ( Crosby et al., 2005) Problematic sleep can take on many forms, including delayed sleep onset, short sleep duration, poor sleep quality, sleepdisordered breathing, and excessive daytim e sleepiness. Shorter sleep duration in particular can result in physiological consequences that affect the health and well being of children. Health consequences of insufficient sleep include increased sympathetic tone, altered glucose metabolism associ ated with insulin resistance, increased hypertension and inflammatory markers associated with cardiovascular risk, and increased relative risk of mortality in adulthood ( Cappuccio et al., 2008; Gangwisch et al., 2006; Knutson, 2007; Meier Ewert et al., 2004; Mullington, Haack, Toth, Serrador, & Meier Ewert, 2009) Additional consequences of short sleep include increased 14

PAGE 15

likelihood of exhibiting risk factors for coronary artery calcification in adulthood (i.e., precursor to coronary artery disease), such as glucose disregulation, high body mass index, and hypertension ( King et al., 2008) Long term sleep problems also have important health implications. Five days of sleep restriction in adult males is associated with a 40% reduction in glucose tolerance, 30% reduction in acute insulin response to glucose, and 30% lower glucose effectiveness, all of which are associated with increased risk for diabetes ( Knutson, 2007) Similar findings have been found epidemiological research with women, but there were no cited studies that examined these relationships in child populations ( Knutson, 2007) The health consequences of insufficient sleep carry implications fo r functional well being of children. For instance, shorter sleep may result in altered cortisol release patterns that can impact individual response to stress ( Mullington et al., 2009 ; Spiegel et al., 2004) These results suggest that sleep can play a salient role in the psychosocial functioning. General sleep problems are associated with a variety of impairments in child psychosocial functioning. Behavioral difficulties associated with problematic sleep include increased oppositional behaviors, poorer executive functioning, and increased irritability ( Anderson, Stor fer Isser, Taylor, Rosen, & Redline, 2009; Corkum, Tannock, & Moldofsky, 1998; Fallone, Owens, & Deane, 2002; Moore et al., 2009) Childhood sleep problems are also linked to poorer emotional functioning. Children with poorer sleep exhibit greater emotional lability, increased depressive and anxiety symptoms, increased likelihood of a mental health diagnosis, and lower perceived health ( Anderson et al., 2009; Crabtree, Varni, & Gozal, 2004; Ivanenko, Crabtree, Obrien, & Gozal, 2006; Moore et al., 2009; Nixon et al., 2008; Smaldone, Honig, & Byrne, 2007) A particular 15

PAGE 16

strength in pediatric sleep research is the consistent findings regarding the behavioral and emotional consequences of insufficient or problematic sleep. An area of increased need is research examining the role of insufficient sleep in pediatric QOL. Although subjective and objective measures of problematic sleep are associated with poorer QOL ( Carn o et al., 2008; Crabtree et al., 2004 ; Hart, Palermo, & Rosen, 2005; Hiscock, Canterford, Ukoumunne, & Wake, 2007) researchers commonly focus on children with sleeprelated breathing disorders rather than other sleep variables such as total sleep time and total wake time ( Carno et al., 2008; Crabtree et al., 2004) There are also significant limitations in the research examining pediatric sleep. Specifically, many studies rely solely on subjective and singl e question measures of sleep ( Hart, Cairns, & Jelalian, 2011) These studies are important for establishing relationships between sleep, weight, and QOL; however, there is concern regarding the reliability and validity of using those measures in comparison to objective measures of sleep ( Hart et al., 2011) Another limitation of the literature is the use of varying reference values for sleep duration, as inconsistencies fail to take into account child developmental expectations for sleep and makes comparison across studies difficult. Sleep & Obesity Relationship Risk for higher weight status can be due to a variety of genetic vulnerabilities, medical conditions, basal metabolic rate, medication regimens, and neurotransmitter activity regulating hunger and satiety ( Hill, 2006 ) ; however, it is at the basic level an issue of behavioral and environmental factors that influence the balance between energy intake and energy expenditure ( Birch, 2006; Hill, 2006 ) Insufficient sleep is one behavioral factor that can impact child risk for higher weight status, as a number of studies have revealed that obese children are more likely to experience sleep problems 16

PAGE 17

than nonoverweight peers. Overweight and obese children are more likely to experience a variety of sleep problems, including sleep disordered breathing, shorter sleep duration, later sleep onset, excessive daytime sleepiness, and more fragmented sleep in comparison to nonoverweight peers ( Beebe et al., 2007; Calhoun et al., 2011; Cappuccio et al., 2008 ) Sleep duration in children who are overweight or at risk for higher weight status is an area in the sleep literat ure that has received increased interest. Researchers have consistently found a strong association between short sleep and weight status in youth. There are 60 to 80% greater odds of having shorter sleep amongst obese children and adolescents relative to t heir nonobese peers ( Cappuccio et al., 2008) Children who obtain less than the recommended amount of sleep have 1.42 to 3.45 greater odds of concurrent overweight status and 1.5 to 2.9 greater odds of future overweight status when compared to children receiving adequate sleep and controlling for potential confounding variables (e.g., birth weight, age, gender, degree of physical activity and television watching, and parental BMI) ( Hart et al., 2011 ; Knutson, 2007; Lumeng et al., 2007; Nixon et al., 20 08 ; Seegers et al., 2011 ; Snell et al., 2007) Insufficient sleep in childhood is also associated with greater BMI in adulthood, even when controlling for childhood weight status ( Landhuis, Poulton, Welch, & Hancox, 2008) Long term sleep patterns in early childhood provide evidence for the role sleep can play in predicting increased weight status Consistently short sleepers (i.e., those obtaining less than 10 hours between 2.5 and 6 years of age) are at the greatest risk for overweight and obesity compared to those obtaining 11 hours of sleep and short 17

PAGE 18

sleepers that increase in their sleep duration by the age of 3.5 years ( Seegers et al., 2011; Touchette et al., 2008) Decreasing sleep by one hour per night at the age of 10 years is associated with a 1.51 to 2.07 higher odds ratio of being overweight or obese, respectively ( Seegers et al., 2011) Improvement in sleep can also play a significant role in addressing risk for higher weight status. A one hour increase in sleep is associated with a 20 to 80% reduction in odds of overweight or obesity ( Knutson, 2007; Landhuis et al., 2008) with greater reductions in risk being associated with improved sleep earlier in child development ( Lumeng et al., 2007) Mechanisms of the Sleep Weight Relationship Insufficient sleep may impact child weight status through a variety of proposed mechanisms. First, ch ildren obtaining less sleep have an increased opportunity to eat, which is associated with increased total caloric intake ( Hart et al., 2011) A second pathway between sleep and risk for overweight is sleepiness or fatigue that may lead to decreased energy expenditure (i.e., decreased physical activity or increased sedentary activity). Insufficient sleep can also be linked to sleepiness related impairments, such as poor mood and decreased motivation, which are associated with changes in eating behaviors ( Hart et al., 2011) Changes in eating behaviors can include increased energy intake, greater consumption of calories from fat, increased intake from calorical ly dense foods, increased snacking, and changes to eating patterns and timing of meals that are disproportionate to energy expenditure ( Hart et al., 2011 ; Touchette et al., 2008 ; Weiss et al., 2010; Westerlund, Ray, & Roos, 2009) Another commonly discussed possible pathway between sleep and weight status involves neuroendocrine changes that occur in response to sleep deprivation ( Knutson, 2007; Leproult & Van Cauter, 2010; Touchette et al., 2008) Specific areas of impact 18

PAGE 19

include the release of cortisol, leptin, and ghrelin. Leptin is an appetite inhibiting hormone that is released from adipose tissue and ghrelin is a peptide released from the stomach that stimulates hunger and fat storage ( Knutson, 2007) Under normal conditions, the patterns of hormonal release over a 24hour period include a maximum of cortisol occurring in the early morning, a meal dependent release of leptin (i.e., minimum release in morning and an evening maximum), and maximum release of ghrelin in the morning that decreases quickly after food consumption ( Leproult & Van Cauter, 2010) Sleep is hypothesized to impact weight status through disruption of brain signaling of leptin and ghrelin release. Insufficient sleep can result in decreased leptin and inc reased ghrelin levels, negatively impacting accurate brain signaling for caloric need and the correct perception of insufficient food intake (i.e., increasing energy intake) ( Knutson, 2007 ; Leproult & Van Cauter, 2010) Emerging research suggests that insufficient sleep may impact energy expenditure by changes in the release of leptin and ghrelin. Research wi th rats suggests that leptin increases energy expenditure through greater thermogenesis in brown fatty tissue and ghrelin can negatively impact energy expenditure through decreases in nonexercise activity thermogenesis (i.e., NEAT, nonexercise activities that expend energy) ( Knutson, 2007) Insufficient sleep is hypothesized to disrupt these processes. However, there is limited research in humans to firmly establish this rela tionship. It is clear from laboratory and epidemiological research in adults that sleep deprivation can impact risk for higher weight status through neuroendocrine disregulation, suggesting that the same may be true for pediatric populations. 19

PAGE 20

Quality of L ife in Overweight Children with Insufficient Sleep Sleep duration is an important factor to consider given the impact sleep and weight status can have on pediatric QOL. It appears that both being obese and having insufficient sleep increases risk for dysf unction in emotional and quality of life domains. Unfortunately, there is limited information regarding how these health conditions interact and who is at greatest risk for poor outcomes such as QOL. A small number of studies have shown a negative association between sleep disordered breathing and QOL in children who are obese, such that greater sleep problems are associated with lower QOL ( Carno et al., 2008; Crabtree et al., 2004 ; Rosen, Palermo, Larkin, & Redline, 2002) Similar to the general sleep literature, however, there is a dearth of research examining the impact of sleep outside of sleep related breathing disorders and research fails to examine the impact of short sleep within pediatric overweight populations ( Hart et al., 2005; Hiscock et al., 2007 ) Research on pediatric chronic health conditions in the previous several decades have stressed the importance of including multiple informants of child outcomes ( e.g., functional abilities, psychosocial functioning, quality of life) across a variety of domains and models that examine variation in child adaptation through proposed moderators and mediators ( Barakat, 2008) Although researchers within the sleep literature have controlled for family income or parent education status, research oft en fails to examine the contributions of the family and ot her influences on QOL in obese children. The association between sleep duration and QOL in children who are overweight or obese may be a more complex than a straight forward approach to examining pathology and adaptation, as there may be other important environmental and behavioral factors that play a role in this relationship. It is therefore important to delineate the unique conditions 20

PAGE 21

under which sleep impacts pediatric QOL (i.e., moderators) and the processes through which sleep impacts QOL (i.e., mediators). To expand upon existing research, the goal of the proposed study is to examine moderators and mediators using a risk resilience framework. Resilience Theoretical Model & Adaptation to Chroni c Conditions Researchers have proposed a theoretical model that includes risk and resistance factors to explain the variability of child psychosocial functioning, physical health, and well being ( Masten, 2004, 2005; Varni & Wallander, 1988) Important risk factors include the childs health condition (e.g., poor s leep, obesity), functional limitations associated with a health condition, and stressors (e.g., major life events, hassles) ( Wallander, Thompson, & AlrikssonSc hmidt, 2003; Wallander, Varni, Babani, Banis, & Wilcox, 1989) Health related risk factors occur in the larger environmental context, with environmental factors serving as resistance factors. Resis tance variables can be important resources (i.e., universally important irrespective of degree of disease) and serve as protective factors (i.e., important moderators in relationship between disease status and outcomes). Examples of resistance factors incl ude stable interpersonal factors (e.g., stress processing or coping abilities), as well as social ecological factors (e.g., family members coping or adaptation and social support) ( Wallander et al., 2003 ; Wallander et al., 1989) A large principle in the model is examination of modifiable risk and resistance variables that could be addressed in intervent ions for children with chronic conditions ( Wallander & Varni, 1998) Risk and Resistance Variables in Pediatric Chronic Illness Literature Resistance variables have been examined more frequently in the general pediatric chronic illness literature. In children with disabilities, parent, child, and family 21

PAGE 22

variables are more strongly associated with adjustment than disease or disability ( Lavigne & Faier Routman, 1992 ; Witt, Riley, & Coiro, 2003 ) For instance, poorer maternal mental health, increased family burden, and lower child perceived social support are associated with greater psychosocial maladjustment in children, while healthier family functioning (e.g., higher family cohesion, parental warmth) and parental coping is associated with greater child adjustment ( Adam, Snell, & Pendry, 2007; Thompson & Gustafson, 1996 ; Wallander & Varni, 1998 ; Witt et al., 2003 ) Resistance variables such as social competence, family functioning, and peer social engagement or support accounts for a large proportion (i.e., 1341%) of the variance in child and parent proxy reports of child QOL ( Alriksson Schmidt, Wallander, & Biasini, 2007 ; Ingerski, Janicke, & Silverstein, 2007) This suggests that social ecological variables could possibly account for the unique conditions under which health conditions (e.g., obesity, poor sleep) impact child functioning and QOL. The role of peer and family functioning has received increased focus on the chronic illness literature. In the pediatric pain literature, research suggests that peer rejection moderates the relationship between pain and child depressive symptoms, as children with high pain and high peer rejection are at higher risk of reporting a greater number of depressive symptoms ( Sandstro m & Schanberg, 2004) Conversely, higher classmate support is associated with fewer depressive symptoms, irrespective of degree of daily hassles in children with pediatric rheumatic diseases ( von Weiss e t al., 2002) Others have examined the moderating role broader family relationships can have on QOL of youth with chronic conditions. Logan and Scharff ( 2005) fo und that the relationship between pain and functional disability in youth with migraines was stronger 22

PAGE 23

in youth with more dysfunctional family environments than those in more adaptive family environments. Findings in the general pediatric literature suggest that social ecological factors can have both main effects and moderating effects to lessen the impact that chronic and acute stress can have in the presence of health conditions. There are several possible benefits that family and peer support can provide in child adaptation to chronic conditions. Stressors associated with chronic conditions limits the resources available to a child to successfully cope and maintain well being, while social support is a resource that can aid in the coping process and can be involved in stress resistance ( Wallander et al., 1989) Researchers have suggested that social support in the family and peer domains increase self efficacy and improve cognitive appraisals regarding the availability of others, which is associated with higher self perceived health status ( Taal, Rasker, Seydel, & Wiegman, 1993; Zeller & Modi, 2006) Others have suggested that social support can positively impact an individuals motivation, coping, and self care behaviors that could inf luence QOL ( Gallant, 2003) Wysocki and colleagues ( 2006) found that family communication and problem solving in youth with diabetes were associated with greater treatment adherence and greater metabolic control. The psychological and psychosocial well being of parents can also play a role in child adaptation to their own chronic health difficulties. Parental distress can influence c hild adaptation through the impact that it can have on the quality of parenting (e.g., use of effective communication and problem solving) and parental modeling of adaptive coping processes that are essential to a childs adaptive cognitivebehavioral development ( Armstrong, Birni Lefcovitch, & Ungar, 2005) Thus, family and peer support, as well as parent mental health or distress may impact child coping, 23

PAGE 24

problem solving, and self care behaviors that can ultimately influence health outcomes and perceptions of well being across physical and psychosocial d omains. In addition to social ecological resistance factors, healthrelated consequences of chronic conditions can be a risk factor for poorer pediatric QOL. Important variables examined in the diabetes literature include glucose metabolism and glycated h emoglobin. Diabetic children with poor glycemic control (i.e., high hemoglobin A1c [HbA1c]), for example, are more likely to report poorer QOL compared to those with better glycemic control ( Ingerski, Laffel, Drotar, Repaske, & Hood, 2010) Health consequences of short sleep, such as increased glycated hemoglobin (i.e., HbA1c), are associated with increased risk for diabetes and inflammatory and macrovascular complications that impact morbidity, mood, and quality of life ( Biuso, Butterworth, & Linden, 2007; Gerich, 2005; Irwin & Miller, 2007 ; Knutson, 2007; Zee & Turek, 2006) These results suggest that it is important to examine the pathways through which sleep impacts pediatric QOL, as higher HbA1c exhibited in some obese children may also contribute to QOL. Risk Resistance Application to Sleep & Psychosocial Functioning in Obese Children Although some researchers have examined psychosocial consequences of short sleep and social ecologica l resistance factors in the general chronic illness literature, the literature is sparse ( Barakat, 2008) This is particularly true for research investigating these factors within pediatric sleep and obese populations, as there is no known published research to date examining the association between risk and resistance factors and QO L outcomes with these individuals. The research findings within other chronic illnesses suggest that there are several variables that may severe 24

PAGE 25

as resistance factors that limit the potential negative impact of poor sleep on QOL in children who are obese ( e.g., family and peer support, parent distress). Insufficient sleep can limit a childs resources to cope with the stressors associated with being obese, ultimately negatively impacting QOL. Family and peer support can provide children with additional res ources when fatigued to engage in effective coping and problem solving skills to improve upon their QOL. Parental distress can add to the level of stress the child experiences and interferes with coping and problem solving processes necessary for successf ul adaptation and perceived well being. Thus, these domains of social support can act as important moderators in obese children who experience insufficient sleep. It is also important to consider possible mediating variables in the risk resistance fram ework. In the proposed model, Wallander and colleagues ( Wallander et al., 1989) also described illness related risk factors that can have an additive effect on child adjustment. That is, the severity of a condition or illness impacts adjustment indirectly through its affects on other illness parameters, functional independence, and disability stress. As described earlier, shor t sleep is associated with increased HbA1c and higher HbA1c is associated with poorer pediatric QOL. It is plausible to theorize that sleep indirectly impacts child QOL through increased HbA1c. Sleep duration, metabolic consequences of short sleep, and soc ialecological variables are vital yet under investigated factors in research, especially in obese populations. It is important to delineate the role of modifiable healthrelated and social ecological variables in order to inform clinical interventions addressing specific targets for change that can impact outcomes across health, psychosocial, and quality of life domains. 25

PAGE 26

Assessment of Pediatric Sleep The assessment of pediatric sleep, including sleep duration, can include the use of several measures, including polysomnography, sleep diaries, actigraphy, and armband accelerometers. Each approach to sleep assessment provides unique information about sleep ( Morin, 2003) and carries different strengths and weaknesses that are worth noting when attempting to establish relationships between sleep, weight, and QOL. Polysomnography Polysomnography (PSG) is considered the gold standard for sleep assessment of sleep physiology (Kushida et al., 2005). It provides several measures, including heart rate provided by electrocardiogram (ECG), abdominal and chest movement, air flow, and arterial oxygen saturation (SpO2) provided by pulse oximetry that can be used to calculate a variety sleep related variables ( Spruyt, Gozal, Dayyat, Roman, & Molfese, 2011) Other monitors include electroencephalogram (EEG) and bilateral elctrooculogram (EOG), as well as body position sensors ( Spruyt et al., 2011) Polysomnography is ideal for assessing the physiology of sleep and for diagnosing sleep disorders such as sleeprelated breathing disorders (e.g., OSA, central apnea) and periodic limb movement disorder ( Buysse, Ancoli Israel, Edinger Lichstein, & Morin, 2006) PSG is not ideal for clinically evaluating sleep difficulties such as insomnia, as PSG fails to capture all aspects of insomnia, sleep patterns are increasingly variable as sleep worsens in individuals, and the use of PSG fo r longer periods of time to obtain stable sleep patterns is neither feasible nor cost effective ( Buysse et al., 2006; Morin, 2003) Additionally, there is concern for the first night effect in assessing sleep with PSG in a 26

PAGE 27

laboratory setting and whether patterns seen are representative of sleep in the home setting ( Ancoli Israel et al., 2003) One method of addressing the weaknesses of PSG and to account for sleep variability is obtaining and examining PSG recordings for two to three consecutive nights in the laboratory setting. An alternati ve consideration is the use of homebased, ambulatory PSG monitoring in order to decrease the burden and subject reactivity that can occur in labbased PSG ( Morin, 2003) It is important to note, however, that these strategies fail to address the concerns that PSG cannot capture all aspects of insomnia. Researchers and clinicians therefore often utilize several subjective and objective measures of sleep that complement one another and can address the individual weaknesses ( Morin, 2003). Sleep Diaries A daily sleep diary is a subjective measurement of a variety of aspects of sleep and is considered the ideal for the assessment of sleep difficulties such as insomnia ( Buysse et al., 2006) Despite being unable to establish the necessary timeline criteria for making a diagnosis of insomnia, sleep diaries provide quantitative subjective reports of sleep param eters such as time of sleep onset, time of sleep offset, estimates of sleep onset latency, the frequency and duration of nighttime awakenings, and subjective ratings of sleep quality ( Buysse et al., 2006; Morin, 2003) There are often significant discrepancies between sleep diaries and PSG on aspects of sleep, but the use of sleep diaries allows for a more cost eff ective measure of sleep over longer periods of time in an individuals home environment and provides unique information regarding perceptions of sleep ( Buysse et al ., 2006; Morin, 2003) It is important for sleep diaries to include monitoring for two weeks in order to obtain a more reliable estimate of the 27

PAGE 28

severity of sleep difficulties and sleep patterns due to t he substantial variability that can occur ( Morin, 2003) Actigraphy Actigraphy uses a portable piezoelectric accelerometer device that provides objective behavioral information about sleepwake patterns and circadian rhythms over time through the measurement of movement ( Littner et al., 2003; van Wouwe, Valk, & Veenstra, 2011) These devices l og the number of counts that occur above a prespecified threshold and are averaged over oneminute intervals ( van Wouwe et al., 2011) Actigraphs are typically worn on the wrist and are recommended for the assessment of sleep in healthy populations, as their reliability is weaker for sleep detection in sleepdisordered populations in which sleep tends to be more fragmented ( Ancoli Israel et al., 2003) It is also recommended for a minimum use of three consecutive 24hour periods; however, use for seven consecutive days is ideal for a more reliable estimate of sleep parameters ( Acebo et al., 1999 ; Littner et al., 2003) There is a stro ng utility of actigraphy in the assessment of insomnia, excessive daytime sleepiness, and circadianrhythm disorders in children when used in conjunction with other sleep measures, such as sleep diaries ( Ancoli Israel et al., 2003; Littner et al., 2003) Use of sleep diaries allows for correction of problematic sleep artifacts associated with actigraphy (e.g., breathing movements, external movements in vehicles) and provides more accurate information of parameters such as sleep onset latency ( Ancoli Israel et al., 2003) The use of actigr aphy in the detection of sleeprelated breathing disorders (e.g., sleep apnea) and in patients with movement disorders independent of other sleep measures is not currently recommended ( Buysse et al., 2006; Sadeh & Acebo, 2002) Actigraphy is considered a more cost effective and 28

PAGE 29

reliable estimate of total sleep time and time in bed; however, there are significant di screpancies and inconsistencies worth noting. One notable weakness is the lack of consistency in the device, epoch length, and algorithm utilized across studies ( Meltzer, Montgomery Downs, Insana, & W alsh, 2012 ) Another concern is the devices poor specificity across studies when compared to polysomnography ( Meltzer et al., 2012) Therefore, continued research in further validating the use of actigraphy in the measurement of pediatric sleep is warranted. Accelerometry The Sensewear Armbands ( "What is bodymedia fit.," 2011 ) are devices worn on the back of the upper right arm that provide measures of energy expenditure and distinguish between sleep from wake using multiple microelectronic mechanical sensors ( BaHammam, Alrajeh, Albabtain, Bahammam, & Sharif, 2010; Soric et al., 2012) The device includes several biometric sensors that measure heat flux, skin temperature, galvanic skin response, near b ody temperature, and movement using a tri axial accelerometer using a sampling frequency of 32Hz ( Soric et al., 2012; van Wouwe et al., 2011) The armband classifies data as sleep by using measures of motion, changes to heat flux and skin temperature, galvanic skin response, and instantaneous agitation; however, temperature and motion are the main indicators us ed to infer sleep ( Teller & Crossley, 2004) The tri axial accelerometer provides indicators of motion and whether an individual is lying down, while temperature sensors can offer information regarding whether an individual is asleep ( BaHammam et al., 2010; Teller & Crossley, 2004) Skin temperature sensors reflect core body temperature, whic h can in conjunction with near body temperature indicate sleep due to the fact that those temperatures tend to 29

PAGE 30

follow the circadian rhythm ( BaHammam et al., 2010; Teller & Crossley, 2004) The armbands utilize a pattern recognition (i.e., machine learning) approach to estimate sleep and the sleep detection algorithm for the device was developed using labbased polysomnography data (Sleep and Armband Examples, 2008, p. 123; Soric et al., 2012; Vyas, Farringdon, Andre, & Stivoric, 2012). The demographic specific sleep algorithms (i.e., specific to age, gender, height, weight) were developed using a three step process, including a screening, analysis using PSG, and clarification of short wakeful events (Sleep and Armband Examples, 2008, p. 123). The armband data is first screened for blocks of time when the subject may be asleep, such as lying down while resting but awake (e.g., watching TV, reading) or resting for the purposes of obtaining sleep (Sleep and Armband Examples, 2008, p. 123). PSG lab data is then used to apply the PSG algorithm to the screened data and short wakeful events are clarified using self report (e.g., sleep/activity logs; Sleep and Armband Examples, 2008, p. 123) ("Sleep and Armband Examples," 2008) ("Sleep and Armband Examples," 200 8). The validity of the armband in estimating sleepwake patterns in children and adolescents is promising. Measures of total sleep time, sleep onset latency, and sleep efficiency are not significantly different than the goldstandard polysomnography, but do overestimate wake after sleep onset by an average of 14 minutes ( Soric et al., 2012) Although Soric and colleagues ( 2012) found large variability in the agreement between the two devices that would suggest that analysis at the individual level is not currently recommended, individual variation was less substantial than armband use with adults and armbands can be used for grouplevel comparisons ( Soric et al., 2012) Other strengths of the device include it having similar ag reement with polysomnography 30

PAGE 31

across a wide range of sleep efficiency rates, there being no significant effect of child gender or weight status, and it being a minimally invasive approach to obtaining sleep values over longer periods of time than possible w ith polysomnography ( Soric et al., 2012) Conclusions There is no single sleep assessment measure that can fully capture all aspects of sleep and each measure carries their own set of strengths and weaknesses that are i mportant to consider. A single time point measurement of sleep, regardless of whether it is subjective or objective, has limitations given the high variability in sleep behavior. Additionally, discrepancies exist between physiological, behavioral, and subjective measures of sleep. It is therefore important for sleep assessment to include multiple time points and multiple modalities of measurement that can serve to complement one another and address their individual weaknesses ( Buysse et al., 2006; Morin, 2003) Purpose of the Study The purpose of the current study was to address the limitations within the existing literature on weight status, sleep duration, healthrelated variables (i.e., HbA1c), and social ecological variables (i.e., peer support, family functioning, and parent distress) as predictors of QOL in a treatment seeking sample of overweight and obese children This study addressed gaps in the literature by using an objective measure of sleep (i.e., accelerometers) when examining the impact of sleep duration and wake time on overall QOL of children who are obese, while also investigating proposed moderators to determine under which conditions insufficient sleep negatively impacts QOL in children who are obese. Additionally, this study examined potential healthrelated pathways through which sleep impacts QOL (i.e., HbA1c). Exploratory aims included exam ining a 31

PAGE 32

proposed standard reference value for sleep duration (i.e., meeting recommendations for 9 hours of sleep) and a subjective measure of child sleep problems when examining child QOL. Aims and Hypotheses Describing Sample Characteristics, Sleep, QOL and Psychosocial Functioning Aim 1 To describe the extent of poor sleep patterns in a treatment seeking sample of primarily obese children Additionally, the study sought to determine whether child sleep variables (i.e., average time of sleep onset and offset, sleep duration, sleep efficiency, wake after sleep onset, total wake time, and time spent in daytime napping) differed across the day of the week (i.e., weekdays versus weekends) and by whether the time of sleep measurement occurred while school was in session (i.e., in school versus on school break). Hypothesis 1.1. It was expected that children on average would exhibit sleep duration that was lower than the developmental recommendations for sleep (i.e., less than 9 hours) and their sleep efficienc ies would be less than the optimal 90% or greater efficiency. There were no a priori hypotheses regarding descriptive sleep patterns for the other sleep variables. Hypothesis 1.2. It was hypothesized that there would be a significant main effect of day of the week (i.e., weekday versus weekend) on sleep measures. Specifically, weekday sleep would include longer sleep onset latency, shorter sleep duration, greater wake after sleep onset (WASO) and total wake time (TWT), lower sleep efficiency, and an earlier time of sleep offset when compared to sleep occurring on weekends. 32

PAGE 33

Hypothesis 1.3. There would be a significant main effect of time (i.e., school vacation) on child sleep measures, such that sleep occurring while on school break would include a later tim e of sleep onset, greater sleep duration, lower WASO, greater sleep efficiency, and later time of sleep offset. Hypothesis 1.4. It was expected that there would be a significant interaction between day and time of measurement on child sleep measures, such that the relationship between sleep and day of the week (weekday vs. weekend) would be strongest for those who wore the armbands while school was in session. That is, there would be no significant differences in sleep behaviors across the day of the week for children who are on school vacation. Aim 2 To describe the QOL, positive peer support, and parental distress in a sample of primarily obese children Hypothesis 2.1. Child self report and parent proxy report of child overall QOL would be significantly lower than a nationally representative sample of healthy, non overweight children. Hypothesis 2.2. Children in our sample would report positive peer support that was lower that a sample of children without a chronic health condition ( M =21.4, SD =3.1) ( Storch et al., 2004) Hypothesis 2.3. Although it was e xpected that a number of parents would report clinically significant distress (i.e., T offs. 33

PAGE 34

Aim 3 Examine soci odemograpic differences in child QOL (i.e., child self and parent proxy report) and sleep variables (i.e., total sleep and total wake times), including child age, gender, race, family income, and child BMI z score. Hypothesis 3.1. Variables associated wit h less nighttime sleep would include older age, male gender, racial or ethnic minority status, and higher child weight status. Hypothesis 3.2. Variables associated with lower QOL would include higher child weight status and lower family income. Examining Health Related and Social Ecological Predictors of Child QOL Aim 4 Examine whether parent psychosocial functioning, child peer support, and general family functioning each serve to moderate the relationship between sleep variables (i.e., total sleep tim e, total wake time) and child QOL in our overweight and obese sample. Hypothesis 4.1. It was expected that sleep would have a main effect on child QOL. Children with shorter total sleep time and greater total wake time would exhibit lower parent and chil d reported total QOL compared to those with longer sleep duration and shorter wake time. Hypothesis 4.2. Peer support, general family functioning, and parent psychosocial functioning would have a significant main effect on parent and childreported total QOL, such that higher functioning in these domains would be associated with higher QOL. Hypothesis 4.3. These social ecological variables would moderate the relationship between sleep and child QOL, as the relationship between sleep and QOL 34

PAGE 35

would be stro ngest for those with poorer functioning within these domains. That is, children with greater peer support, family functioning, and parent psychosocial functioning in the presence of shorter sleep or greater time spent awake would exhibit greater QOL than t hose with lower functioning in these domains. Aim 5 Determine whether the degree of insulin resistance, as measured by HbA1c, mediates the relationship between sleep variables and QOL. Hypothesis 5.1. It was expected that higher insulin resistance woul d be associated with poorer child QOL. Additionally, the relationship that each shorter sleep duration and total wake time have with child QOL would decrease after accounting for child insulin resistance. That is, insulin resistance would partially mediate the relationship between sleep variables and QOL. Exploratory Aims Aim 6 Compare the QOL (i.e., total, physical health, emotional, social, and school) of parents and children who report child sleep problems and those who deny sleep problems. Hypothes is 6.1. Subjectively reported sleep problems would be associated with poorer parent and childreported QOL across all domains compared to those who did not report sleep problems. Aim 7 Compare the QOL (i.e., total, physical health, emotional, social, and school) of those meeting and failing to meet developmental recommendations for at least 9 hours of sleep per night. 35

PAGE 36

Hypothesis 7.1. Children meeting developmental recommendations for sleep duration would have significantly higher QOL compared to those failing to meet sleep recommendations. 36

PAGE 37

CHAPTER 2 METHODS Description of the Larger Study This study is part of a larger grant funded randomizedcontrolled trial evaluating the impact of behavioral interventions on child weight status in a rural overweight and obese sample, the Extension Family Lifestyle Intervention Project for Kids (E FLIP for Kids). Child parent dyads in the E FLIP for Kids study were randomized to one of three intervention conditions: a behavioral family intervention, a behavioral parent only condition, or a streamlined education control condition. All data from the current dissertation were taken from initial screening and baseline assessment visits completed with each family prior to beginning the treatment program. Procedure Particip ants for this study were recruited through a variety of methods, including press releases to local radio stations, direct mailings to households and healthcare providers, and brochure distribution to local schools, churches, community events. Families in terested in the study were able to call the research office using a toll free phone number in order to learn more about the program and complete a phone screening for initial eligibility criteria. An in person screening visit was scheduled if they meet in itial criteria, followed by a baseline assessment visit at the Cooperative Extension office in the county in which they resided in order to complete the assessment measures. At the initial screening visit, families completed informed consent and assent and completed measures of height and weight to determine study eligibility. Families were included in the study if the family resided in a rural county, the childs body mass index (BMI) was at or above the 85th percentile for age and gender 37

PAGE 38

norms described by the CDC ( Kuczmarski et al., 2002) and if the parent or legal guardian was under the age of 75 years and living in the same home as the child. Caregivers or children were excluded from the study if they had a medical condition that would make participating in a program requiring changes to physical activity or caloric intake unsafe, high resting blood pressure (i.e., 140/90 mm Hg), or used medications including monoamine oxidase inhibitors, antibiot ics for tuberculosis or HIV, antipsychotic agents, systemic corticosteroids, chemotherapy drugs, and weight loss drugs within six months of beginning the study. Participants were also excluded if they were participating in another weight management program if the parent or child did not provide informed consent or assent, and if the child was identified as having extreme oppositional behaviors or severe cognitive or developmental delays. Those families meeting eligibility at the screening visit attended a baseline assessment visit to complete a series of questionnaires and physical assessments one to two weeks prior to the sta rt of the intervention program. Participants This study included overweight and obese children between the ages of 8 and 12 years and their parent or legal guardian living in one of 10 rural counties in northcentral Florida. There were 305 families that completed inperson eligibility screening, while 269 completed a baseline assessment, 250 of which who initiated treatment in order to receive the accelerometers. There were significant differences in family income [ t (307)= 2.79, p <.01] and parent highest level of education [ t (68.45)= 3.65, p <.01] between those who initiated treatment following baseline versus those who did not initiate treatment, such that those of higher family income and parent education were more likely to initiate treatment. Of the 250 remaining participants, 107 children (42.8%) did 38

PAGE 39

not wear the armband for the required amount of time for useful and valid data analysis in the current study. There were no significant sociodemographic differences in those treatment starters who did and did not have enough armband data to be included in the analyses. There were an additional 19 participants who did not complete quest ionnaires for the proposed moderators (i.e., family functioning, peer support, parent stress) due to their completion of combined screening and baseline assessments. Additionally, seven participants have incomplete data for HbA1c values due to either error s in the measurement device or inability to obtain enough blood for analysis. Thus, descriptive and preliminary analyses involving sleep variables will include the entire sample of 143 participants, while moderator and mediator analyses will include the av ailable participants with complete data in order to not have a significant loss of data. There were no significant sociodemographic differences between those who had complete and missing data across all moderators and the mediator. Measures Demographic Information Parents completed a background information questionnaire during the initial screening visit. Demographic information collected included child age, gender, race or ethnicity, and household income. Height and Weight A trained health technician or nurse collected anthropometric measures. Child height was measured to the nearest 0.1 centimeter using a Harpedon Stadiometer and weight was measured to the nearest 0.1 kilogram using a digital scale. Body mass index was calculated as kilograms per meters squared and BMI z scores were 39

PAGE 40

calculated using a modified LMS procedure from the Center for Disease Control and Prevention ( Kuczmarski et al., 2002) This method gives estimates of lambda (L), mu (M), and sigma (S) for age in moths for each gender, allowing for a smoothed BMI curve in order to calculate child BMI z scores ( Kim et al., 2005; Kuczmarski et al., 2002) Pediatric Quality of Life The PedsQLTM 4.0 ( Varni et al., 2003) is a 23 item QOL instrument to assess subjective reports (i.e., child self report and parent proxy report) of child well being across physical and psychosocial (i.e., emotional, social, and school functioning) domains. Parents and children rated how difficult specific situations were for children over the course of the previous month on a 4point scale ranging from 0 (Never) to 4 (Almost Always). Summary scores from this measure include a total scale score, a psychosocial health score (i.e., social, emotional, and school functioning), and a physical health score, with higher scores indicating higher perceived QOL. This measure has strong internal consistency, clinical validity, reliability, and demonstrates adequate construct vali dity when make a distinction between health and acutely and chronically ill children ( Varni et al., 2003) The current study used the total scale score of parent proxy and child self reported QOL for the primary analyses, whose internal respectively). Objective Pediatric Sleep Children were instructed to wear a Sensewear Armband Accelerometer ( 2013 Bodymedia, Inc., Pittsburgh, PA) for seven consecutive days, 24 hours per day (except when bathing or swimming) as a measure of sleep. 40

PAGE 41

Data inclusion criteria In order to include child sleep data in the study, children needed to wear the accelerometer over the course of a week. Previous studies have suggested at least four days of armband data in order to maintain reliability of at least 0.80 for energy expenditure ( Trost, Pate, Freedson, Sallis, & Taylor, 2000) There are no published research recommendations on the number of nights of sleep data necessary for acceptable reliability in accelerometers; however, similar instruments (i.e., actigraphs) recommend five days of sleep data for more acceptable reliability and validity ( Acebo et al., 1999) Participants were included in the current study if they had five nights of sleep data (i.e., three weeknights and two weekend nights). Sleep data processing and measure descriptions. Sleep data was obtained using the Sensewear Professional Software, ver sion 7.0. The Sensewear programs enables users to export the 24hour data into a spreadsheet that provides minuteby minute epochs of data, with sleep being scored as 1 and awake being scored as 0. Total sleep duration for each night was calculated by summing the number of minutes scored as sleep. For the analyses in the current study, average sleep duration and total wake time were used as the primary independent variables and descriptive analyses used the following measures: average minutes of dayti me napping, time of sleep onset and offset, sleep onset latency, total sleep time, sleep period, wake after sleep onset, total wake time, and nighttime sleep efficiency. The excel files for each participant included dateand timestamped data and were uti lized to calculate these sleep variables. Sleep onset was defined as the first minute in which there were at least 3 consecutive minutes scores as sleep and sleep offset was the last minute of at least 5 consecutive minutes scored as sleep, as identified by the Sensewear algorithm. Sleep 41

PAGE 42

onset latency was considered the number of minutes in which the child was laying down prior to sleep onset. Sleep period was defined as the number of minutes elapsed between the sleep onset and sleep offset times. Wake af ter sleep onset was considered the number of minutes elapsed during the sleep period that was scored as not asleep (i.e., sleep scored as 0). Sleep efficiency was measured as the percentage of minutes during the sleep period that were considered sleep (i.e., [sleep minutes/sleep period] x 100), providing information regarding sleep fragmentation. Finally, total wake time was defined as the total number of minutes that the child was awake throughout the night, including the time spent lying in bed attem pting sleep initiation (i.e., SOL), wake after sleep onset, and the time the child spent snoozing after sleep offset. These criteria were used as outlined in the actigraph literature ( Acebo et al., 1999) as there are no other described c riteria. Subjective Child Sleep Problems The PedsQLTM includes an item used to assess the subjective frequency sleep difficulties in the previous month stated I have trouble sleeping for children. The parent proxy measure asks parents to rate how often their child has trouble sleeping in the previous month. The questions were answered on a 5point likert scale ranging from zero (never) to four (almost always). The question was dichotomized so that responses of sometimes, often, or almost always were grouped as having sleep troubles and responses of never or almost never were grouped as not having sleep troubles in order to examine both child and parent perceptions of child sleep problems. Peer Support Children completed the Social Experiences Questionnaire (SEQ) as a 15item measure of self perceived social support and peer victimization. Items were rated on a 42

PAGE 43

5 point Likert scale, with higher scor es representing greater social support. The SEQ has three subscales (i.e., overt and covert victimization, positive support) that each consist of five items. The positive support subscale (i.e., 5 items) was used in the current study. Family Functioning P arents completed the Family Assessment Device (FAD) as a 60item measure of family functioning ( Miller, Epstein, Bishop, & Keitner, 1985) Items were rated on a 4point Likert scale, with parents responding to items in how well it describes your own family. The FAD provides scores across seven dimensions, including global family functioning, pr oblem solving (i.e., ability to resolve problems), communication (i.e., ability to convey information in a clear and direct manner), affective responsiveness (i.e., responding with the appropriate emotion), roles (i.e., effective division of family respons ibilities), affective involvement (i.e., how much family members are interested in and involved with one another), and behavioral control (i.e., how standards of behavior are maintained and expressed). The current study used the general family functioning domain as a measure of global family functioning, with higher scores indicating poorer family functioning. Total scores were divided by the number of questions (i.e., 12 questions) in order to obtain a mean score that ranges from one to four. Clinical cuto ff scores differentiating healthy from unhealthy general family functioning is 2.00 ( Miller et al., 1985) The tot al score was used in the primary analyses as to decrease the role that a restricted range had in our findings. The FAD is a measure that is well established in its use in chronic illness populations and demonstrates good psychometric properties ( Alderfer et al., 2008) The internal consistency for general 43

PAGE 44

Parent Psychosocial Functioning Parents comple ted the Brief Symptom Inventory18, which is an 18item measure rated on a 5point Likert scale of a variety of anxiety related and depressive symptoms e xperienced by parents. The BSI 18 provides a global index of overall psychological distress (i.e., Global Severity Index) and three symptom scales (i.e., somatization, depression, anxiety). It is a brief measure of psychological problems in medical and community populations and is highly correlated (i.e., >.90) with scores on the SCL90 ( Asner Self, Schreiber, & Marotta, 2006) This study utilized the global index of psychological distress. Insulin Resistance Glycated hemoglobin was collected as a me asure of glucose control over the previous two to three months. Several drops of blood were collected by a trained nurse or health technician and analyzed by using point of care equipment. This measure was completed during the baseline assessment. Statis tical Analysis Plan The Statistical Package for the Social Sciences was utilized for all analyses (SPSS, version 17.0, SPSS Inc., Chicago, Illinois). The data was examined for potential outliers and those cases that fell outside of two standard deviations from the mean were excluded from the analyses. Analyses of Sample Characteristics, Sleep, QOL, and Psychosocial Functioning Aim 1 Means and standard deviations were calculated for each sleep variable for all measured days, weekdays, and weekends. A repeated measures ANOVA was 44

PAGE 45

conducted to examine whether the time of measurement (i.e., in school versus on break) and the day of the week (i.e., weekday versus weekend) predicted child sleep variables. The betweensubjects factor was the time of measurement and the withinsubjects factor was the day of the week. Aim 2 Means and standard deviations were calculated for QOL, social support, and parental distress variables. Paired samples t tests were conducted to examine potential differences between child and parent reports of child total QOL. Concordance of parent and child ratings was conducted through use of intraclass correlations. Onesample t tests were conducted in order to compare the QOL of the current sample to large samples of healthy, nonoverweight children described in previous research. Similarly, onesample t tests were conducted to compare the peer support reported by children in our sample to children with and without other chronic conditions described in previous research. Aim 3 Bi variate correlations were conducted for individual sociodemographic data, QOL, and sleep variables. First, Pearson and Spearmanrho correlations investigated the relationships among sociodemographic variables (i.e., child age, gender, weight status, and race, as wel l as family socioeconomic status and parent educational status), average child sleep (total sleep and wake times), and child QOL (self and parent proxy report). Given that the majority of the sample was identified as white or black, significant correlations including race will be examined further utilizing t tests that compare blacks and whites. Any significant associations were controlled for in later analyses. 45

PAGE 46

Analyses Examining Health and Social Ecological Predictors of Child QOL Aim 4 In order to examine whether social ecological variables served as moderators in the relationship between child sleep and QOL, moderation analyses were conducted using the IBM SPSS Statistics Developer Version 20 PROCESS macro ( Hayes, 2012) Moderation analyses were conducted using 5000 bootstrapped samples. Bootstrapping is a statistical procedure that allows for multiple estimates of the conditional effect by drawing repeated samples with replacement ( Hayes, 2012) The macro also allows for mean centering of predictor variables when forming product terms and increases understanding of the product term within the samples data ( Hayes, 2012; Preacher & Hayes, 2008) The output of these analyses provides unstandardized bweight therefore be used when providing results. Any sociodemographic variables that were associated with the outcome variables (i.e., parent proxy and child self reported total QOL) were included as covariates in the models. Separate analyses were conducted for each sleep measure (i.e., total sleep time, total wak e time) and each informant of child total QOL (i.e., parent proxy and child self report) when examining the potential moderating role of social ecological variables (i.e., positive peer support, parent distress, general family functioning). Significant mo derator effects were further examined using probing of interactions. PROCESS allows for a pick a point approach to probing interactions, which creates conditional estimates of the moderator and estimates the effect of the predictor at those values ( Bauer & Curran, 2005 ; Preacher, Curran, & Bauer, 2006) For the purposes of the current study, the conditional eff ect was estimated using the mean of the moderator, 46

PAGE 47

as well one standard deviation above and below the mean of the moderator ( Hayes, 2012; Hayes & Matthes, 2009) It is important to note that if the distribution of the moderators are skewed, these conditional values may not fit appropriately within the range of the data and it would be more appropriate to generate conditional effects at the 10th, 25th, 50th, 75th, and 90th percentiles of the moderator ( Hayes, 2012; Hayes & M atthes, 2009) The method of probing significant interactions was determined by whether the distributions of the moderators are skewed (i.e., standard deviation for nonskewed data and percentile for skewed data). Aim 5 In order to examine whether insul in resistance (i.e., as measured by glycated hemoglobin [HbA1c]) partially mediated the relationship between each sleep variable (i.e., total sleep time, total wake time) and parent and child reports of child total QOL, mediation analyses were conducted us ing the IBM SPSS Statistics Developer Version 20 macro Process ( Hayes, 2012) To test for mediation effects, the macro included 5000 bootstrapped samples. Bootstrapping allows for multiple estimates of the indirect effect by drawing repeated samples with replacement ( Hayes, 2012 ) This method has advantages over the traditional Baron and Kenny ( 1986) approach, as it minimizes Type II error through the use of fewer required inferential tests and does not rely on the flawed assumption of the shape of the sampling distribution and normality of direct effects ( Preacher & Hayes, 2008) Conditional and unconditional indirect effects were calculated based on bias corrected bootstrapped confidence intervals and the significance of the indirect effect was based on the bootstrapped 95% confidence interval not containing zero ( Hayes, 2012) 47

PAGE 48

Exploratory Aims Aim 6 The PedsQLTM item assessing the subjective frequency child sleep difficulties in the previous month was used to examine the relationship between child sleep problems and QOL. The total QOL summary and the emotional functioning scores were recalculated following removal of the sleep item in order to decrease the likelihood of finding a spurious association between sleep problems and QOL. The physical health score and the remaining domains of psychosoci al functioning (i.e., social and school) remained the same since they did not contain the sleep item. A series of t tests were conducted in order to compare child and parent proxy reports of child sleep problems across all domains of parent proxy and childreported QOL. Cohens d coefficients were also calculated in order to provide measures of effect size. Aim 7 A multivariate analysis of variance (MANOVA) analysis was used to examine whether meeting or failing to meet developmental sleep recommendations has significant effect across all domains of child QOL (i.e., total, physical health, emotional, school). Two separate MANOVAs were conducted for each parent and child reported QOL. Follow up univariate ANOVAs were conducted to examine the direction of t he effects. These follow up analyses were conducted with Bonferonni correction in order to account for alphawise error (i.e., alpha <.0125). 48

PAGE 49

CHAPTER 3 RESULTS Analyses of Participant Characteristics and Sleep Behaviors Aim 1: Participant Characteristics & Description of Child Functioning The sample of 143 children between the ages of 7.7 and 12.9 years (M=10.29, SD=1.4) had complete sleep and QOL data. The sample was primarily obese (i.e., 89.5%; BMI z 2 (SD=5.6) and the mean BMI z score being 2.18 (SD=0.4). The sample consisted of 77 girls (53.8%) and 66 boys (46.2%). The majority of the sample of parents reported their childs race as White (67.8%), while the remaining 32.2% of parents identified their child as Black (12.6%), Bi or multi racial (11.9%), no response (6.3%), Native Hawaiian (0.7%), and Asian (0.7%). Approximately 12% of parents identified their childs ethnicity as Hispanic. Participating parents or legal guardians consisted primarily of females (94.3%) and 69.2% indicated that they were married. Average parent BMI was 36.15 kg/m2 (SD=10.6), which is classified as class II obesity. The median family income ranged from $40,000 to 59,999. Demographic information for the sample is presented in Table 3 1. Hypothesis 1.1: Describing child sleep behaviors. The first aim of the study was to describe the sleep patterns of the sample of primarily obese children and examine the potential impact of time of measurement (i.e., school year versus school break) and day of the week (i.e., weekday versus weekend) on those sleep variables. Means and standard deviations for all sleep variables can be found in Table 32. Average time of sleep onset and offset were 10:59pm and 7:15am, respectively. Children needed an average of 13.73 minutes to initiate sleep (i.e., SOL) and spent an 49

PAGE 50

average of 84.45 minutes awake after sleep onset (i.e., WASO). Children spent an average of 496.92 minutes (SD=54.76) in bed from sleep onset to offset (i.e., sleep period), whic h equates to approximately 8 hours and 17 minutes. Mean child total sleep time ( M =6.92 hours, SD =0.85) was well below the recommended quantity of 9 hours of sleep per night. Moreover, 88.1% of the sample averaged less than 8 hours of sleep per night, and n one of the participants averaged above 9 hours of sleep per night. Mean total wake time ranged from 8.30 to 237.60 minutes ( M =112.42, SD =37.05). Moreover, the mean sleep efficiency was 83.29% ( SD =6.54), which was below the cut off for the healthy range of greater or equal to 90%. Twenty eight percent of the sample had a sleep efficiency that was less than 80% and only 16.8% had a sleep efficiency that was equal or greater to 90%. The mean time spent in daytime napping was 6.35 minutes ( SD =15.26) but vari ed greatly, ranging from 0 to 89 minutes. Hypotheses 1.2 to 1.3: Main effect of day of the week and school break in predicting sleep variables. T able 33 displays the means and standard deviations of the sleep variables separately for day of the week and school break. There was a main effect of school break on the time children initiated sleep (i.e., time of sleep onset), such that children on school break went to sleep much later than those in school (F(1,138)=11.54, 2 p= .08, p =.001). There were also main effects of day of the week (F1,138)=20.99, 2 p=.13, p <.001) and school break (F(1,138)=41.48, 2 p=.23, p <.001) on the time of sleep offset, such that children had a later time of sleep offset on weekends (vs. weekdays) and school break (vs. when children in school). There were no other main effects in predicting sleep onset latency, sleep period, sleep efficiency, total sleep time, total wake time, or wake after sleep onset ( p s >.05). 50

PAGE 51

Hypothesis 1.4: Interaction effect of day of the week x school br eak in predicting sleep variables. There were few significant findings when examining the potential role of the interaction between the day of the week and school break. There were no interaction effects in predicting sleep onset, sleep offset, sleep onset latency, sleep period, sleep efficiency, total sleep time, or total wake time ( p s >.05). There was a significant interaction effect detected between day of the week and school break in predicting wake after sleep onset ( F (1,138)= 4.13, p =.044). Children on school break spent fewer minutes awake after sleep onset on weekends ( M =87.36, SE=8.48) than on weekdays ( M =70.46, SE =7.85), while children in school had a similar WASO for both weekdays ( M =84.33, SE =3.66) and weekends ( M =85.81, SE=37.67). See Figur e 33 for additional information. Aim 2: Describing Child QOL & SocialEcological Functioning Hypothesis 2.1: Describing child QOL. Table 3 4 displays the means and standard deviations for the primary predictor and outcome variables. The mean child self r eported total QOL for the entire sample was 75.44 ( SD =13.85), while the mean parent proxy report of total child quality of life was 73.59 ( SD =14.92). A Wilcoxon signed ranks test was conducted to examine potential differences in parent and child reports of child total QOL given that the distribution of these variables was nonnormal. There were no significant differences between child and parent reports of child QOL (Z= 1.16, p =.25); however, concordance of parent and children ratings were considered poor (i.e., ICC=.35). Children in the current study reported total QOL that was not significantly different from a nationally representative sample of obese children ( M =74.0, t (142)= 1.24, p =.22); however, QOL was significantly lower than samples of children wi th 51

PAGE 52

diabetes ( M =80.35, t (142)= 4.24, p <.001) and healthy/nonoverweight children ( M =83.84, t (142)= 7.25, p <.001) (Varni et al., 2007). Hypotheses 2.2 and 2.3: Describing child social ecological functioning. Child reports of positive peer support from the Social Experiences Questionnaire ranged from 8 to 25 ( M =18.52, SD =4.18). Similar to a sample of children with type 1 diabetes ( Storch et al., 2004) obese children in our study reported significantly lower positive peer support when compared to a sample of control children without a chronic health condition ( t = 7.67, p <.001). The T scores for the degree of parent reported distress from the BSI 18 ranged from 33 to 74 ( M =43.43, SD =9.04) and was in the healthy range overall, as only approximately 3% of parents reported clinically significant distress (i.e. M =1.73, SD =0.47) was in the healthy range. However, 28.2% of parents indicated family functioning to be above the clinical cut off of 2.0 for dysfunctional family functioning. Aim 3: Sociodemographic Differences Across Primary Predictor and Outcome Variables Analyses also included examination of potential demographic differences across either the primary predictor or outcome variables (i.e., total sleep time, total wake time, parent proxy QOL, childreported QOL). Bivariate Pearson correlations between child age and average total sleep time was significant ( r= 0.26, p <.01), such that older children obtained significantly less sleep than younger children on average. Spearmanrho corre lations between race and average total sleep time was also significant ( r= 0.23, p <.01). The follow up t test revealed that children identified as African American (i.e., both Hispanic and non Hispanic Black) obtained significantly less sleep on average (M=379.66 minutes or 6.33 hours) than children identified as Caucasian ( M =416.94 52

PAGE 53

minutes or 6.95 hours) ( t (116)= 2.95, p =.004). There were no sociodemographic variables that were significantly correlated with child total wake time. Parent proxy report of child total QOL was significantly correlated with child BMI z score ( r= 0.21, p <.05), such that parents of children with a higher BMI z score rated their childs QOL lower than children with a lower BMI z score. There was also a significant association betw een family income and parent proxy report of child total QOL ( r= 0.22, p <.01), such that parents who rated their child as having higher overall QOL were more likely to have higher family income than those with lower family income. Child weight status and fa mily income were included in as covariates in analyses that included parent proxy report of child QOL as the outcome variable. Information regarding variable intercorrelations is displayed in Table 35. Moderation and Mediation Analyses Aim 4: Moderator a n alyses The fourth aim of the study was to examine whether social ecological variables served as moderators in the relationship between child sleep (i.e., total sleep duration, total wake time) and parent proxy or child self reported total QOL. All six moderator models and follow up analyses are presented in Table 3 6 through Table 3 10. Main and interaction effects for total sleep time and social ecological variables in predicting self reported QOL. There were no main effects of TST (B weight= .0157, SE=. 02, p =.47), positive peer support (B=.2967, SE=.30, p =.32), parent distress (B= .0248, SE=.30, p =.93), or general family functioning (B= .2170, SE=.23, p =.35) in predicting child self reported QOL (see Table 36). There was no significant interaction bet 2 =.007, p= .37) 2=.03, p =.068) in predicting 53

PAGE 54

self reported QOL. However, analyses did reveal a significant interaction between positive peer suppor t and TST in predicting self reported QOL (B= 2 =.047, p =.02). Probing of the interaction using a median split of positive peer support (see Table 37) revealed that when the degree of positive peer support was low, there was a high effec t of TST on child self reported QOL (B=.0603, SE=.027, p =.03). That is, higher TST was associated with higher QOL for children who reported lower positive peer support (see Figure 34). In contrast, TST was not a significant predictor of child QOL at eit her the mean of the sample (B=.0157, SE=.022, p =.47) or when positive peer support was high (B= .0288, SE=.03, p =.34). Main and interaction effects for total wake time and social ecological variables in predicting self reported QOL. There were no main e ffects of TWT (B= .0442, SE=.04, p =.25), positive peer support (B=.2721, SE=.31, p =.38), parent distress (B=.0315, SE=.34, p =.93), or general family functioning (B= .1362, SE=.24, p =.58) in predicting child self reported QOL (see Table 38). Positive peer support (B=.0042, S 2=.002, p= .68), parent distress (B= 2=.023, p= .13), and family functioning (B= 2=.022, p= .13) did not serve as moderators of the relationship between TWT and child self reported QOL. Main and interaction effec ts for total sleep time and social ecological variables in predicting parent proxy QOL. There were no main effects of TST (B=.0447, SE=.04, p >.05) or positive peer support (B=.1665, SE=.35, p =.64) in predicting parent proxy report of child total QOL. Parent distress approached significance as a predictor of parent proxy reports of child QOL (B= .6473, SE=.32, p =0.55). There was a main effec t of family functioning (B= .9116, SE=.26, p <.001), 54

PAGE 55

such that poorer general family functioning was associated with poorer parent reported child QOL. Analyses did not reveal significant interactions between positive peer support and TST (B= 2 =.0002, p =.63), parent distress and TST 2=.002, p =.63), or family functioning and TST (B=. 0022, SE=.005, 2=.001, p= .68) in predicting parent reported child QOL. See Table 39. Main and interaction effects for total wake time and social ecological variables in predicting parent proxy QOL. There were no main effects of positive peer support (B= .1665, SE=.35, p =.64), parent distress (B= .4801, SE=.37, p =.19), or TWT (B= .0447, SE=.04, p >.05) in predicting parent proxy report of child QOL (see Table 310). There was a main effect of general family functioning (B= .8653, SE=.28, p =.002), such that poorer general family functioning was associated with poorer parent reported child QOL. Positive peer support (B= 2=.0008, p= .36), parent distress (B= 2=.01, p= .30), and family functioning (B= .0046, 2=.002, p= .61) did not serve as significant moderators of the relationship b etween child TWT and parent proxy report of child QOL. Aim 5: Mediation a nalyses The fifth aim of the study was to determine whether glycated hemoglobin (HbA1c) mediated the relationship between each sleep variable (i.e., total sleep duration, total wake time) and parent and child reports of child total QOL. Bi variate correlations between the independent and dependent vari ables are presented in Table 3 5 All four mediator models are presented in Figure 3 5 through Figure 3 8. Predicting child self repor ted QOL. Total sleep time (TST) was not a significant predictor of child HbA1c levels (B= .0005, SE=.0004, p >.05). Similarly, TWT was not a significant predictor of HbA1c (B= .0004, SE=.001, p >.05). The direct effect 55

PAGE 56

of TST (B=.0073, SE=.02, p >.05), HbA1 c (B=3.08, SE=4.97, p >.05), and TWT (B= .0328, SE=.04, p >.05) were not significant in predicting child QOL. Both of the confidence intervals for TST and TWT predicting childreported total QOL contained zero, indicating that hemoglobin A1c was a not signi ficant mediator of the relationship between either predictor and child QOL (see Table 3 11 and Table 3 12). Predicting parent proxy report of child QOL. Total sleep time (TST) was not a significant predictor of child HbA1c levels (B= .0002, SE=.0004, p >.0 5). Similarly, TWT was not a significant predictor of HbA1c (B= .0004, SE=.001, p >.05). The direct effect of TST (B=.0112, SE=.02, p >.05), HbA1c (B= .3176, SE=5.17, p >.05), and TWT (B= .0433, SE=.04, p >.05) were not significant in predicting parent prox y QOL. For TST and TWT, the confidence intervals predicting parent proxy QOL contained zero, indicating that HbA1c was not a significant mediator of the relationship between bot h predictors and QOL (see Table 3 11 and Table 3 12). Exploratory Analyses Aim 6 T tests revealed that children who endorsed sleep problems on the PedsQLTM questionnaire were more likely to report lower QOL across total ( t (248)=5.32, d =.68, p <.001), physical health ( t (149.81)=4.54, d =.74, p <.001), social functioning ( t (134.42)=3.47, d =.60, p <.001), and emotional functioning ( t (249)=5.38, d =.68, p <.001) domains. There were no significant differences for childreported school functioning ( t (248)=1.88, d =.24, p =.061). T tests also revealed that parents who endorsed child sleep problems were more likely to report poorer child QOL across total ( t (249)=4.39, d =.56, p <.001), physical health ( t (249)=3.16, d =.40, p =.002), school functioning ( t (249)=3.11, d =.39, p =.002), and emotional functioning ( t (249)=6.50, d =.82, p <.001) 56

PAGE 57

domains. There were no significant differences for parent proxy reports of child social functioning ( t (249)=1.88, d =.24, p =.061). See tables 313 and 314. Aim 7 MANOVA analyses could not be conducted to examine differences in QOL between children who do and do not meet de velopmental recommendations for sleep (i.e., 9 hours), as none of the participants met this cutoff. 57

PAGE 58

Table 31 Demographic characteristics of sample. Mean SD Child Age Child BMI BMI z score 10.29 28.83 2.18 1.4 5.6 0.4 % Gender Boys Girls 46. 2 53.8 Child Race White Black 67.8 12.6 Bi or Multi racial Native Hawaiian 11.9 0.7 Asian No response 0.7 6.3 Child Ethnicity Hispanic Non Hispanic 12.6 87.4 Median Family Income Below $19,999 $20,000 $39,999 15.6 32.6 $40,000 $5 9,999 $60,000 79,999 22.0 14.2 $80,000 $99,999 Above $100,000 7.1 8.5 SD = Standard deviation 58

PAGE 59

Table 32 Mean sleep variables across all days and time of measurement Sleep Variable Mean Standard Deviation (SD) Sample Range Time of Sleep Onset 22:59pm 0.06 20:58 to 4:58 Time of Sleep Offset 7:15am 0.50 4:45 to 13:35 Sleep Onset Latency (minutes) 13.73 9.04 1.20 to 48.20 Sleep Period (minutes) 496.92 54.76 182.60 to 586.00 Total Sleep Time (minutes) 415.80 52.79 274.80 to 594.0 0 Total Sleep Time (hours) 6.92 0.85 4.58 to 8.67 Wake After Sleep Onset (minutes) 84.45 33.42 16.80 to 195.00 Sleep Efficiency (%) 83.29 6.54 65.59 to 96.30 Total Wake Time (minutes) 112.14 37.05 8.30 to 237.60 Daytime Nappi ng (minutes) 6.35 15.26 0.00 to 89.00 59

PAGE 60

Table 33 Mean sleep variables separately across day of the week and school break Sleep Variable All Days (SD) Weekdays (SD) Weekend (SD) Time of Sleep Onset On Break 0:14 (0.80) ^ 0:02 (0.09) 0:30 (0.08) In Scho ol 22:46 (0.05) 22:20 (0.05) 23:17 (0.06) Time of Sleep Offset On Break 8:40 (0.08) ^ 8:28 (0.08) 9:00 (0.09) In School 6:59 (0.03) 6:33 (0.03) 7:41 (0.05) Sleep Onset Latency (SOL) On Break 13.78 (8.17) 12.86 (9.11) 15.16 (11.35) In School 13.80 (9.28) 13.60 (10.54) 14.10 (14.12) Sleep Period On Break 507.87 (51.15) 506.76 (57.84) 509.55 (76.50) In School 494.16 (55.84) 489.09 (69.63) 502.24 (66.80) Sleep Efficiency On Break 85.31 (7.87) 84.44 (9.94) 86.61 (6.05) In School 82.94 (6.25) 82.93 (7.10) 82.96 (7.32) Wake After Sleep Onset (WASO) On Break 80.60 (41.63) 87.36 (53.16) 70.45 (31.77) In School 84.92 (31.92) 84.33 (36.87) 85.81 (37.67) Total Sleep Time (TST) On Break 437.16 (66.44) 424.39 (62.71) 440.14 (70.32) In School 411 .42 (49.58) 404.34 (66.83) 417.72 (69.70) Total Wake Time (TWT) On Break 106.13 (43.23) 109.47 (52.58) 101.11 (37.36) In School 113.45 (36.04) 113.28 (41.04) 125.12 (129.69) Daytime Napping On Break 7.06 (19.62) 9.92 (28.44) 2.77 (8.31) In School 6.2 3 (14.55) 6.57 (18.21) 5.72 (15.60) p <.05, weekday vs. weekend main effect ^ p <.05, school vs. on break main effect p <.05, day of the week and school break interaction effect 60

PAGE 61

Figure 31 Main effect of school break on time of sleep onset Fig ure 3 2 Main effects of day and school break on time of sleep offset 61

PAGE 62

Figure 33 Day by time interaction on average wake after sleep onset (WASO) Table 34 Means and standard deviations of predictor and outcome variables Mean SD Sample Range Scale Ra nge Total Quality of Life (Parent report) Total Quality of Life (Child report) 73.59 75.44 14.92 13.85 32.6 to 100.0 30.4 to 98.9 0.0 to 100.0 0.0 to 100.0 Average Total Sleep Time (Minutes) 415.80 52.79 274.8 to 594.0 Avera ge Total Wake Time (Minutes) Positive Peer Support General Family Functioning (Total score) Family Functioning (Av erage Score) Parental Distress (T score) Hemoglobin A1c 112.42 18.52 20.78 1.73 43.43 5.49 37.05 4.18 5.67 0.47 9.04 0.27 8.3 to 237.6 8.0 to 25.0 12.0 to 39.0 1.0 to 3.3 33.0 to 74.0 4.5 to 6.4 5 .0 to 25 .0 12.0 to 48.0 1.0 to 4.0 62

PAGE 63

Table 35 Intercorrelations of predictor and outcome variables Measure 1 2 3 4 5 6 7 8 9 10 11 12 13 1. BMI z score 2. Child Age 0.02 3. Family income 0.07 0.03 4. Gender 0.12 0.02 0.02 5. Race 0.10 0.04 0.04 0.08 6. Total Sleep Time 0.11 0.26 0.08 0.13 0.23 7. Total Wake Time 0.10 0.12 0.11 0.12 .001 0.38 8. Peer Support 0.01 0.10 0.14 0.16 0.11 0.10 0.06 9. Family Function 0.08 0.04 0.25 0.09 0.02 0.03 0.08 0.07 10. Parent Distress 0.01 0.17 0.12 0.17 0.01 0.08 0.10 0.13 0.44 11. HbA1c 0.20* 0 .20* .004 0.03 0.07 0.11 0.10 0.06 0.06 0.16 12. PedsQL TM Total (Parent) 0.21 0.03 0.22 0.15 0.02 0.05 0.08 0.06 0.37 0.20* 0.10 13. PedsQL TM Total (Child) 0.13 0.09 0.08 0.06 0.08 0.15 .004 0.15 0.07 0.05 0.09 0.35 PedsQL, Pediatric Quality of Life Inventory p <.05 p <.01 63

PAGE 64

Table 36 Total sleep time and social ecological predictors of child self reported t otal QOL Total Sleep Time 1 B SE R 2 Model 1 (N=115) .065 Positive Peer Support .2967 0.3 0 Total Sleep Time (TST) .0157 0.02 Peer Support TST .0120 0.005 .047 Model 2 (N=111) .046 Parent Stress .0248 0.30 Total Sleep Time (TST) .0219 0.02 Parent Stress TST .0088 .005 .03 Model 3 (N=114) .029 Family Functioning 2170 0.23 Total Sleep Time (TST) .0231 0.02 Family Functioning TST .0041 .005 .007 1Excluding IV/DV outliers p <.05 Table 37. Conditional effect of TST on child self reported QOL at values of the m oderator Moderator a Point Estimate Lower 95% BCa CI b Upper 95% BCa CI b Low Peer Support .0603 c .0064 .1141 Mean Peer Support .0157 .0268 .0583 High Peer Support .0288 .0881 .0306 aValues for the moderator are the mean and plus/minus one SD from the mean bBCa are bias corrected cConfidence intervals that do not contain zero are deemed to be significant 64

PAGE 65

Figure 34 Interaction of total sleep time and peer support predicting self reported QOL 6 5

PAGE 66

Table 38 Total wake time and social ecological predictors of c hild self rep orted t otal quality of life Total Wake Time 1 B SE R 2 Model 1 (N=109) 0.022 Positive Peer Support .2721 0.31 Total Wake Time (TWT) .0442 0.04 Peer Support TWT .0042 0.01 .002 Model 2 (N=105) 0.03 Parent Stress .0315 0.34 Total Wake Tim e (TWT) .0266 0.04 Parent Stress TW T .0169 0.01 .023 Model 3 (N=108) 0.04 Family Functioning .1362 0.24 Total Wake Time (TWT) .0412 0.04 Family Functioning TWT .0126 .008 .022 1Excluding IV/DV outliers p .05 66

PAGE 67

Table 39 Total s le ep time and social ecological predictors of parent proxy of c hild quality of life Total Sleep Time 1,2 B SE R 2 Model 1 (N=112) .054 Positive Peer Support .1638 0.35 Total Sleep Time (TST) .0006 0.03 Peer Support TST .0030 0.006 .002 Model 2 (N=109) .10 Parent Stress .6473 0.33 Total Sleep Time (TST) .0201 0.02 Parent Stress TST .0026 .005 .002 Model 3 (N=112) .162 Family Functioning .9116 0.26 Total Sleep Time (TST) .0136 0.02 Family Functioning TST .0022 .005 .00 1 1Excluding IV/DV outliers; 2Covarying for child BMI z score and family income p <.05; p <.01 Table 310. Total wake t ime and social ecological predictors of parent proxy of c hild quality of life Total Wake Time 1,2 B SE R 2 Model 1 (N=112) .06 Pos itive Peer Support .1665 0.35 Total Wake Time (TWT) .0447 0.04 Peer Support TWT .0110 0.01 .008 Model 2 (N=103) .09 Parent Stress .4801 0.37 Total Wake Time (TWT) .0543 0.04 Parent Stress TWT .0123 0.01 .01 Model 3 (N=106) .15 Family Functioning .8653 0.28 Total Wake Time (TWT) .0461 0.04 Family Functioning TWT .0046 0.01 .002 1Excluding IV/DV outliers; 2Covarying for child BMI z score and family income p <.05; p <.01 67

PAGE 68

Table 311. Indirect e ffect of total sleep time t hrough HbA1c on child and parent p roxy QOL Mediator a Point Estimate Lower 95% BCa CI b Upper 95% BCa CI b HbA1c (Child report QOL) .0017 .0155 .0024 HbA1c (Parent proxy QOL) .0001 .0044 .0063 a BCa are bias corrected and accelerated b Confidence inter vals that do not contain zero are deemed to be significant Table 312. Indirect effect of total w ake time t hrough HbA1c on child and parent p roxy QOL Mediator a Point Estimate Lower 95% BCa CI b Upper 95% BCa CI b HbA1c (Child report QOL) .0009 .0158 .0 037 HbA1c (Parent proxy QOL) .0004 .0057 .0164 a BCa are bias corrected and accelerated b Confidence intervals that do not contain zero are deemed to be significant 68

PAGE 69

Table 313. Mean QOL of obese children with and without childreported sleep t roubles Children with Sleep Troubles Children without Sleep Troubles Cohens d P value Child Reported QOL Mean (SD) Mean (SD) Total Score 70.04 (14.3) 79.30 (12.4) .68 <.001** Physical Summary Score 73.15 (15.6) 82.12 (1 3.3) .74 <.001** Emotional Functioning 61.92 (23.9) 77.42 (20.5) .68 <.001** Social Functioning 68.95 (23.6) 78.94 (17.4) .60 .001** School Functioning 72.67 (18.4) 77.01 (16.7) .24 .061 significant difference between thos Table 314. Mean QOL of obese children with and without parent reported sleep t roubles Children with Sleep Troubles Childr en without Sleep Troubles Cohens d P value Parent Reported QOL Mean (SD) Mean (SD) Total Score 66.86 (14.9) 75.87 (15.1) .56 <.001** Physical Summary Score 69.95 (20.5) 78.47 (19.4) .40 .002** Emotional Functioning 55.81 (19.3) 7 5.07 (17.9) .82 <.001** Social Functioning 66.47 (19.4) 71.45 (19.3) .24 .061 School Functioning 68.72 (21.2) 76.76 (17.9) .39 .002** 69

PAGE 70

CHAPTER 4 DISCUSSION The current study is the first to examine the associations between objectively measured sleep and child QOL, as well as potential social ecological risk resistance factors within a strictly overweight and obese sample of children Although there are a large number of children who are overweight or that are at greater risk for poorer QOL, there is often variability in reported QOL within this population (Hughes, Farewell, Harris, & Reilly, 2007; Schwimmer et al., 2003). Sleep is an important factor to consider given the impact sleep and weight status can have on pediatric Q OL (Beebe et al., 2007; Carno et al., 2008). Additionally, research often fails to examine the contributions of environmental variables that play a role in the relationship between sleep and well being The purpose of the current study therefore was to address the limitations within the literature by examining the unique conditions under which sleep may impact QOL (i.e., peer support, family functioning, and parent distress) and the processes through which sleep may impact QOL (i.e., HbA1c) using a risk r esilience framework. Aim 1: Describing Child Sleep Behaviors Children in the current study exhibited profound sleep insufficiencies, as approximately 88% of children averaged 8 or fewer hours of sleep per night, spent a large amount of time awake during th e night, and only about 17% of the sample had optimal sleep efficiency. These findings are consistent with research examining the trend towards shorter sleep in a large, nationally representative sample of children (Snell et al., 2007). However, it is not eworthy that 11% of the national sample obtained less than 8 hours of sleep and is in contrast to the approximately 88% of our sample who obtained less than 8 hours. This percentage of children experiencing an insufficient total sleep time in our 70

PAGE 71

sample is even greater than that reported by Beebe and colleagues (2007), who compared a sample of obese adolescents attending a weight management clinic ( Mageth percentile) to a control group of healthy weight and overweight adolescents ( Mage=12.6 years; Range BMI=24th to 92nd percentile). Beebe et al. found that 26% of healthy weight and overweight adolescents ( MBMIz= 0.26, SD =.61) and 56% obese adolescents ( MBMIz= 2.48, SD =.30) averaged less than 8 hours of sleep. The proportion of obese children in the current study obtaining inadequate sleep compared to nonobese peers in the Beebe et al. study suggests that trends towards decreased sleep may be more substantial in obese treatment seeking samples Although obese children in both samples exhibited low percentages of children achieving eight hours of sleep per night, children in the current study sample averaged significantly fewer number of minutes of sleep per night compared to nonobese control group ( t (142)= 5.50, p <.001) and significantly greater minutes of sleep per night compared to obese adolescents ( t (142)=8.98, p <.001) in the Beebe et al. study. This difference in mean numbers, but not in percentage achieving eight hours of sleep, may be due to the greater variability in the data in the current study sample. Interestingly, children in our sample had significantly lower sleep efficiency compared to control youth from the Beebe study ( t (142)= 6.41, p <.001), but a greater sleep efficiency than the obese youth in the Beebe study t (142)=5.66, p <.001). One possible explanation for the poorer sleep efficiency in the Beebes obese group relative to the current study is that the youth in Beebes obese group were significantly heavier than the youth in the current sample. 71

PAGE 72

It was hypothesized that sleep patterns would differ across weekdays when compared to weekends and for children who were on school break versus attending school the week thei r sleep was measured. Expected differences included significantly greater SOL, shorter sleep duration, greater WASO, lower sleep efficiency, and earlier time of sleep offset on weekdays (versus weekends) and during the week of school (versus school break). These hypotheses were partially supported. Children in our study went to sleep significantly later when on school break and woke up significantly later on weekends and when on school break. These findings are generally consistent with the literature exami ning sleep patterns in childhood and adolescence (Beebe et al., 2007; Hansen, Janssen, Schiff, Zee, & Dubocovich, 2005; Wing, Li, Li, Zhang, & Kong, 2009; Nixon et al., 2008; Wolfson & Carskadon, 1998). As expected, children on school break spent fewer minutes awake after sleep onset (WASO) on weekends when compared to weekdays, while there were no significant differences in WASO across the day of the week for children who were in school the week of sleep measurement. Differences in sleep patterns across days of the week and school attendance are important to consider given that irregularities in the sleep schedule and timing is frequently associated with shorter sleep duration and poorer daytime psychosocial functioning (Nixon et al., 2008; Pesonen et al., 2010; Wolfson & Carskadon, 1998). Contrary to Hypothesis 1.4, there were no significant main or interaction effects of day of the week (weekday vs. weekend) and school break across the remaining sleep variables (i.e., SOL, TST, TWT, and sleep efficiency). An additional unexpected finding was the delayed sleep preference (i.e., time of sleep onset and offset increased by one hour on weekends) in children who were attending school during measurement, such 72

PAGE 73

that the sleepwake shift did not result in weekend s leep extension (i.e., greater TST) or greater sleep efficiency. This is contrary to research with adolescents that indicate that a shift to later and more preferred bedtimes on weekends is associated with improved SOL, TST, and sleep efficiency (Giannotti, Cortesi, Sebastiani, & Ottaviano, 2002; Wolfson & Carskadon, 1998). One potential explanation for the limited differences across sleep variables is the degree of variability of sleep in our sample. Although a smaller range of data can limit detection of s ignificant relationships, highly variable data (i.e., higher standard deviation) may indicate that the sample mean does not accurately represent the variable being measured due to higher random error (Field, 2009). However, the majority of participants fel l within two standard deviations of the mean, suggesting that this was not the case in this study. Comparison of weekday versus weekend total sleep time and sleep efficiency between the current study sample relative to the obese and control groups in the Beebe et al study are also noteworthy. Beebe and colleagues (2007) found that differences in sleep duration and sleep efficiency between the control group (i.e., nonoverweight and overweight) and obese group was more pronounced on weekends than on weekday s due to a later time of sleep onset in the obese group. There were similar findings when comparing our primarily obese sample to the Beebe et al. (2007) control group. Specifically, children in the current sample averaged significantly fewer minutes of we ekend sleep ( t (142)= 8.77, p <.001) and weekend sleep efficiency ( t (142)= 7.93, p <.001) compared to the control group. These findings are in contrast to the group comparisons for weekday sleep. The average weekday sleep duration was not significantly dif ferent between the children in our study and the group of nonobese 73

PAGE 74

youth in the Beebe et al. study ( t (142)= 1.12, p =.27). However, children in our sample had a significantly lower sleep efficiency when compared to the Beebe et al. control group ( t (142)= 6.26, p <.001). These findings are noteworthy given that there is a dearth of research examining sleep duration and efficiency in children of varying weight status. The finding that obese children obtain poorer sleep efficiency despite nonsignificantly di fferent TST than non obese children suggests that focusing on the negative effects of lower weekday TST in obese populations does not necessarily equate with sleep efficiency, and that this variable is also important to consider. Aim 2: Describing Child QOL and Psychosocial Functioning It was hypothesized that child and parent proxy reported QOL in the current sample would be lower than a nationally representative sample of healthy weight children. The findings of the study were consistent with previous r esearch (Friedlander et al., 2003; Pinhas Hamiel et al., 2006; Williams et al., 2005). Specifically, children in our sample reported QOL that was similar to a nationally representative sample of obese children and significantly lower than both nonoverweig ht peers and children with diabetes (Varni et al., 2007). Our findings are also similar to a study by Schwimmer and colleagues (2003) in which they reported obese children have greater odds of impaired QOL compared to healthy, nonoverweight children (i.e. OR=5.5) and that obese children report proportionally similar QOL to that of children with cancer (i.e., OR=1.3). However, compared to the group of children with cancer from the Schimmer study, the primarily obese children in our sample reported signific antly greater total QOL for both child ( t (249)=4.23, p <.001) and parent proxy reports ( t (250)=3.53, p =.001). These findings are in contrast to what Schwimmer et al. found when comparing their obese sample to a group of children with cancer described in previous research (i.e., Varni, 74

PAGE 75

Burwinkle, Katz, Meeske, & Dickinson, 2002). This may be due to the differences in weight status between the two samples, as children in our sample had an average BMI z score of 2.19 and those in Schwimmer et al. were more severely obese (i.e., MBMI z =2.6). Consistent with H ypothesis 2.3, the majority of parents reported general family functioning ( M =1.73, SD =0.47) and parental distress ( M =43.43, SD =9.04) that was in the healthy range. Approximately 28% of parents in the cur rent sample reported family functioning in the unhealthy range, which is consistent with research examining proportion of unhealthy family functioning in groups of parents of obese children (29%), those with sickle cell disease (32%), inflammatory bowel disease (26%), and healthy comparisons (24%) (Herzer et al., 2010). Although were no significant differences in family functioning in between families of children with and without a chronic condition (Herzer et al., 2010), the findings of the current study suggest that the large proportion of families that meet the clinical cut off for unhealthy functioning may be at greater risk of poorer child QOL and could benefit from intervention. Nearly 3% of parents reported clinically significant distress (i.e. well below the 2541% of mothers of obese treatment seeking children in previous studies reporting significant distress (Epstein, Myers, & Anderson, 1996; Zeller et al., 2004). The small number of parents who reported significant distres s likely limited our ability to detect relationships and may reflect that parents and children that initiate treatment are more likely to be well adjusted compared to those who do not initiate treatment. It is also possible that parents report less distres s due to living in a rural environment that may be characterized as having fewer or different stressors, including 75

PAGE 76

having a lower population density and crime rates compared to nonrural environments. Alternatively, it can also be argued that rural environments have fewer financial resources compared to nonrural environments that can be associated with greater distress (Tickamyer & Duncan, 1990); however, the median family income in our sample (i.e., $40,000 to 59,999) is higher than the general rural population and may have buffered the impact of environmental resources on parental distress. It was also expected that children would report lower positive peer support when compared to healthy peers. Consistent with the findings of Storch and colleagues (2004) who found that children with type 1 diabetes report lower positive peer support compared to a group of children without a chronic health condition, obese children in our study reported significantly lower peer support than the healthy peers group cited by Storch and colleagues. The findings are also consistent with research comparing obese children to their nonoverweight peers when examining peer ratings of social acceptance (Zeller, Reiter Purtill, & Ramey, 2008), which provides further support that the peer environment (i.e., both the presence of negative attitudes or behaviors and positive peer support or acceptance) is important to examine for the purposes of potential areas for intervention for children who are obese. Aim 3: Sociodemographic Differe nces in Primary Predictor and Outcome Variables It was expected that variables associated with less nighttime sleep would include older child age, male gender, racial or ethnic minority status, and higher child weight status. These hypotheses were partial ly supported, as there was a negative association between child sleep time and child age. Moreover, African American children obtained less sleep than Caucasian children. These findings are consistent with previous 76

PAGE 77

research examining child age and ethnic/m inority status (Crosby et al., 2005; Lumeng et al., 2007; Spilsbury et al., 2004). Contrary to the findings of Crosby and colleagues (2005), there continued to be significant differences in TST between Caucasian ( M =423.47 minutes) and African American chil dren ( M =385.34 minutes) when sleep also included time spent in daytime napping. There were no sociodemographic variables that were significantly associated with child total wake time. It was also hypothesized that variables associated with poorer child QO L would include higher weight status and lower family income. Despite the restricted range of weight status in our sample, higher child BMI z score was associated with lower parent proxy reports of child QOL. There also was a positive association between f amily income and parent proxy QOL. Neither child weig ht status nor family income had significant relationships with childreported QOL. These findings suggest that the degree to which weight status is associated with QOL is dependent upon whether parent or child reports are examined. Although parents may be less aware of child internalizing problems, parents may be more open to reporting obesity related difficulties that children may deny having and therefore perceive their child as having poorer overall QO L when compared to parents of children with lower weight status. However, these findings may have also been influenced by the domain of QOL we examined (i.e., overall), as higher weight status was also associated with lower childreported QOL in the domains of physical and social functioning. Aim 4: Sleep and Social Ecological Predictors of Child QOL It was hypothesized that TST and TWT would be associated with poorer child QOL and that family functioning, peer support, and parental distress would moderate the relationship between child sleep and each parent proxy and child self reported total 77

PAGE 78

QOL. Most of our hypotheses were not supported. There were no main effects of TST or TWT on either child self reported or parent proxy report of child QOL. This lack o f significant findings may be due to the use of the Sensewear Armband, which is less extensively examined than other measures such as actigraphy. Despite the promising initial findings of Soric and colleagues (2012), it is the only study to date to examine the use of armbands as a measure of sleep behaviors in a pediatric population. More extensive research is needed in order to establish how well validated this measure is when examining child sleep, particularly with varying child weight status. It is als o possible that there were no main effects of TST or TWT because objective measures of sleep may not reflect how child perceptions of their sleep influence their perceived QOL. For instance, obtaining less than 8 hours of sleep may be perceived as adequate for some children and poor for other children. The exploratory aim 6 of our study focused on parent and child perceptions of child sleep problems and their association with child QOL. It was expected that childand parent reported child sleep problems would be associated with poorer child and parent proxy report of child QOL across all domains (i.e., total, physical health, emotional, and school) compared to parents and children who denied child sleep problems. These hypotheses were supported for most dom ains of child self report (i.e., total, physical, emotional, social) and parent proxy report (i.e., total, physical, emotional, school). Our findings are consistent with published literature demonstrating that subjective sleep problems (e.g., OSA symptoms) or sleep insufficiency is associated with poorer psychosocial functioning and QOL (Carno et al., 2008; Graef, Janicke, McCrae, & Silverstein, in press; Wolfson & Carskadon, 1998), even when PSG 78

PAGE 79

measured OSA and TST are not associated with QOL (Carno et al ., 2008). Although PSG is a wellvalidated biophysiological measure of sleep used to diagnosis sleep disorders and other objective sleep measures (e.g., actigraphy and armbands) provide useful information regarding sleep behaviors, subjective measures of sleep problems may provide more important information regarding how a parent or child feels sleep is impacting the child (Acebo et al., 1999; Lewandowski, Toliver Sokol, & Palermo, 2011). While the oneitem measure of sleep is a limitation in these analyses, the large to medium effect sizes for parent proxy QOL (i.e., emotional functioning and total QOL, respectively) and the medium effect sizes for childreported QOL (i.e., total, physical, emotional, social functioning) suggest that subjective sleep probl ems could play a more salient, or at least noteworthy, role in how children perceive their well being. There was a main effect of parent reported family functioning on their reports of their childs overall QOL, such that poorer family functioning was ass ociated with parents rating their childs QOL lower than parents reporting greater family functioning. This may be due to family functioning being associated with greater parent and child motivation, coping, self care behaviors, and positive cognitive appr aisals that can assist children in managing their well being (Gallant, 2003; Taal et al., 1993). Parental distress, on the other hand, may be associated with parental modeling of less than optimal coping and distressed parents having less time, energy, and resources to effectively help their child manage their well being (Epstein, Klein, & Wisniewski, 1994; Janicke et al., 2007). The findings of our study did not show a significant association between parental distress and child reports of their QOL. Other research has postulated that parental distress is more closely associated with parent proxy reports of QOL and 79

PAGE 80

psychological adjustment rather than child self reports (Guilfoyle et al., 2010; Zeller et al., 2004). This is somewhat consistent with our findings, as the association between parental distress and parent proxy QOL approached significance (i.e., p =.055). The lack of consistent findings in the literature may be a reflection of child age. When compared to Janicke et al. (i.e., 12.85 years), child age was younger in the current (i.e., 10.29 years) and Guilfoyle et al. (i.e., 11 years) studies. It is possible that the younger children are less aware of their parents distress and degree of family functioning in a way that would impact their self perceived well being compared to adolescents. Degree of parental distress and parent perceptions of overall family functioning should be considered when using parent proxy measures of QOL, as parent reports are typically used for measures of pediatric psychosoc ial functioning and may impact treatment initiation, treatment outcomes, and parent opinions of treatment success (Guilfoyle et al., 2010). Contrary to H ypothesis 4.3 family functioning and parent distress did not serve as moderators in the relationshi p between child sleep and QOL. One possibility for this lack of significant findings are limitations to the measurement of these variables. T hese measures of social support may not be accurate indicators of how children perceive their social support in the family environment (Dubow & Tisak, 1989). Another possibility is that family functioning and parent stress are best characterized by a main effect model, which suggests that the degree of social support in the family environment is beneficial to adjustmen t and QOL, regardless of the degree of health related stress. Although family variables did not serve as moderators, peer support did moderate the relationship between total sleep time and child self reported QOL. 80

PAGE 81

Specifically, children who obtained less sleep and experienced low levels of peer support reported lower QOL than those who reported moderate or high peer support. This finding is consistent with previous research that found that greater peer support is associated with higher QOL in children who are obese and that social support is associated with higher self perceived health status in children with chronic health conditions (Gallant 2003; Taal et al., 1993). Social support therefore may limit the impact of insufficient sleep on QOL in children w ho are obese through providing children with the resources to cope with the stressors associated with insufficient sleep and being obese in a way that is adaptive to their self perceived QOL. Aim 5: Insulin Resistance as a Mediator Predicting QOL The final main hypotheses of the study were that insulin resistance (i.e., as measured by HbA1c) would serve as the healthrelated pathway through which each total sleep time and total wake time impact parent and child reports of child total QOL. None of these hy potheses were supported in the findings. One potential explanation for the lack of findings may be the measure used for insulin resistance in our study (i.e., HbA1c). Although the American Diabetes Association has suggested that HbA1c has utility in predic ting the microvascular complications associated with chronically high glucose (American Diabetes Association, 2010), research with adults have noted that A1c may be inadequate as the only criterion used for diagnosing diabetes or prediabetes (Lorenzo et al., 2010). More recent research has examined the use of A1c in children and adolescents who are obese, who have also found that the traditional cut off of 6.5% for prediabetes is inadequate due to low sensitivity and specificity and poor agreement with the oral glucose tolerance test (Nowicka et al., 2011). Nowicka and colleagues (2011) described more optimal thresholds of 5.8% for detection of type 2 81

PAGE 82

diabetes and 5.5% for detection of impaired glucose tolerance. In using these suggested criteria, 45.6% of the sample was considered to be in the healthy range, whereas 38.4% met criteria for impaired glucose tolerance and 15.9% met criteria for type 2 diabetes. It is therefore unexpected that our hypotheses were not supported. It is possible that our lack of findings is a reflection of our generic QOL measure, which may not fully capture the role that diseaserelated symptoms can have on child well being as a diseasespecific measure of QOL could (Guyatt, Feeny, & Patrick, 1993; Varni, Burwinkle, & Lane, 2005). Alternatively, it is possible that our lack of findings reflect the restricted range of HbA1c values. It is also possible that the findings reflect the age of the children in our sample. S ignificant health problems may not be apparent for children in middle childhood and children may not perceive problems with their blood sugar in a manner that would impact their QOL ratings if they have not been specifically diagnosed or tested for diabetes Parents who endorsed yes when asked has a doctor ever said that your child has diabetes or high blood sugar (i.e., N=6) reported lower child QOL than parents who endorsed no, according to a nonparametric t test (i.e., Mann Whitney U; Z= 2.08, p =.04). There were no differences in childreported QOL (Z= 1.86, p =.06). These differences in parent proxy QOL may be due to the changes in parent perceptions of child functioning that occur following being informed by a doctor that their child has greater insulin resistance. An alternative explanation is that parents who perceive their child as functioning poorly are more likely to initiate medical appointments and examine health variables such as insulin resistance. However, only three of the six parents 82

PAGE 83

endorsed the presence of child sleep problems on a oneitem sub jective measure of child sleep. Study Limitations The current study had several strengths, including examination of child and parent perspectives of child QOL, use of an objective measure of sleep duration and subjective measure of child sleep problems, e xamination of social ecological risk resistance variables, and examination of healthrelated pathways through which sleep can impact QOL in a strictly overweight and obese treatment seeking sample of children and their parents. However, the findings of the study were for the large part nonsignificant or produced very small effect sizes, which may be a reflection of several methodological limitations. One limitation of the study was the cross sectional design that limited our ability to make causal statem ents in our results. An additional limitation was that the sample comprised of treatment seeking parents and children that included a highly restricted range of weight status (i.e., 89.5% obese) that may have precluded our ability to detect differences and make them distinguishable from the general population of children who are overweight or obese. However, the main goal of the study was to elucidate the variability of QOL within obese samples by examining the role of objectively measured sleep and potenti al moderators that could explain under which circumstances some children cope with the stressors associated with having a higher weight status in order to maintain adequate QOL. As suggested by the social ecological model of health and adjustment, pediatri c obesity research needs to consider the multitude of factors that could explain the variability of child well being. The limited significance and effect size of 83

PAGE 84

the current study findings indicates that the association that obesity and sleep each have wit h child QOL may interact with a variety of other variables (e.g., cortisol release) that may impact the relationship. Moreover, the generic measure of QOL may be limited in its ability to accurately assess problems most salient to children who are obese despite its ability to compare multidimensional well being across disease groups and that an obesity specific measure such as Sizing Them Up may address the limited scope of a generic measure (Modi & Zeller, 2008). A notable methodological limitation was the use of SenseWear Armbands accelerometers (SWA) as a measure of sleep. Although the SWA has been validated for measurement of energy expenditure and has the advantage of provided extended monitoring in an unobtrusive manner (Soric et al., 2013), there is scant research examining use of SWA as a measure of sleep in children. The only known research to date examining agreement between SWA and PSG in children found that there was no systematic bias in estimating sleep variables and that there were no sig nificant differences between the two measures for total sleep time, sleep onset latency, and sleep efficiency (Soric et al., 2013). However, the armbands overestimated wake after sleep onset by an average of 14 minutes and the large variability of the agreement between SWA and PSG resulted in substantial random error (Soric et al., 2013). These findings suggest that despite promising initial psychometric properties in a labbased setting, SWA is recommended only for group level analyses and is currently li mited in its ability to reliably examine intraindividual sleep patterns. An additional limitation in the use of the armbands in our study was the number of participants (N=107, 42.8%) that were excluded due to the lack of adequate sleep data for useful and valid analysis 84

PAGE 85

of sleep (i.e., three weekdays and two weekend days), resulting in analyses being conducted with less power. While this suggests that there were possible biases that negatively impacted the findings of the study, there were no significant differences between those who did and did not have enough sleep data across sociodemographic variables (i.e., child gender, child age, child BMI z score, family income, treatment condition, parent BMI) and the study predictor or outcome variables (i.e., QO L, peer support, family functioning, parent distress, insulin resistance). Clinical Implications and Future Directions Given the current prevalence of high child adiposity in the U.S., the strong association between pediatric short sleep and weight status and the potential for long term negative impacts that each can have on health care, obesity related sleep difficulties are likely to affect a substantial number of children and adolescents in the future. There is therefore a need for research to develop a better understanding of how sleep may impact not only health functioning and weight management, but also quality of life and psychosocial functioning of children who are overweight or obese. This study is an important step in that direction, as it demons trates that in the context of lower peer support, obese children who obtain less sleep report lower QOL. The use of a oneitem subjective measure of childand parent reported child sleep problems also revealed that greater child sleep problems are associated poorer childand parent reported child QOL compared to those who do not experience sleep troubles. Although objective sleep measures of sleep duration reportedly provide relatively valid estimates of sleepwake patterns compared to the gold standard PSG (Acebo et al., 1999), objective measures prevent determination of the types of sleep problems a child might experience beyond insomnia (e.g., sleeprelated breathing disorders, periodic limb movements). This is 85

PAGE 86

important because there are different mec hanisms underlying each sleep problem (e.g., behavioral versus physiological) and they may therefore have different relationships with child QOL. The next step in research should include a multi method assessment of sleep that uses a longitudinal approach to examine childand parent reported child sleep problems, well validated objective sleep measures (e.g., actigraphy or accelerometry and PSG), and sleep diaries to account for artifacts of the objective measures. This information is needed to obtain mor e conclusive information about the direction of the relationship sleep and QOL, as well as understand potential healthrelated mediators and social ecological moderators between these variables in an obese population. If confirmed in future studies, QOL and sleep may become important variables for clinicians to evaluate when working with children who are overweight and obese. Given the degree of insufficient sleep in our study and the impact it can have on the QOL of those with low peer support, it is not u nreasonable to speculate that improved support and sleep may be associated with improvements in child well being. Prospective research using a larger sample size and use of well validated measures of sleep duration, as well as perceived sleep problems or s leep quality are needed before drawing more definitive conclusions on these relationships. 86

PAGE 87

LIST OF REFERENCES Acebo, C., Sadeh, A., Seifer, R., Tzischinsky, O., Wolfson, A. R., Hafer, A., & Carskadon, M. A. (1999). Estimating sleep patterns with activ ity monitoring in children and adolescents: How many nights are necessary for reliable measures? Sleep, 22, 95 103. Adam, E. K., Snell, E. K., & Pendry, P. (2007). Sleep timing and quantity in ecological and family context: a nationally representative timediary study. Journal of Family Psychology, 21 4 19. Alderfer, M. A., Fiese, B. H., Gold, J. I., Cutuli, J. J., Holmbeck, G. N., Goldbeck, L., . Patterson, J. (2008). Evidencebased assessment in pediatric psychology: family measures. Journal of Pediatric Psychology, 33 10461061. Alriksson Schmidt, A. I., Wallander, J., & Biasini, F. (2007). Quality of life and resilience in adolescents with a mobility disability. Journal of Pediatric Psychology, 32, 370379. American Diabetes Association (2010). Diagnosis and classification of diabetes mellitus. Diabetes Care, 33, S62 69. Ancoli Israel, S., Cole, R., Alessi, C., Chambers, M., Moorcroft, W., & Pollak, C. P. (2003). The role of actigraphy in the study of sleep and circadian rhythms. Sleep, 26, 342392. Anderson, B., Storfer Isser, A., Taylor, H. G., Rosen, C. L., & Redline, S. (2009). Associations of executive function with sleepiness and sleep duration in adolescents. Pediatrics, 123, e701 707. Armstrong, M.I., Birni Lefcovitch, S., & Ungar, M.T. (2005). Pathways between social support, family well being, quality of parenting, and child resilience: What we know. Journal of Child and Family Studies, 14 269 281. Asner Self, K. K., Schreiber, J. B., & Marotta, S. A. (2006). A cross cultural analysis of th e Brief Symptom Inventory 18. Cultural Diversity & Ethnic Minority Psychology, 12, 367375. BaHammam, A., Alrajeh, M., Albabtain, M., Bahammam, S., & Sharif, M. (2010). Circadian pattern of sleep, energy expenditure, and body temperature of young healthy m en during the intermittent fasting of Ramadan. Appetite, 54 426 429. Barakat, L. P. (2008). Editorial: Journal of Pediatric Psychology statement of purpose-section on family influences and adaptation. Journal of Pediatric Psychology, 33 2630. 87

PAGE 88

Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 11731182. Bauer, D. J., & Curran, P. J. (200 5). Probing interactions in fixed and multilevel regression: Inferential and graphical techniques. Multivariate Behavioral Research, 40, 373 400. Beebe, D. W., Lewin, D., Zeller, M., McCabe, M., MacLeod, K., Daniels, S. R., & Amin, R. (2007). Sleep in over weight adolescents: shorter sleep, poorer sleep quality, sleepiness, and sleepdisordered breathing. Journal of Pediatric Psychology, 32 6979. Birch, L. L. (2006). Child feeding practices and the etiology of obesity. Obesity, 14 343 344. Biuso, T. J., B utterworth, S., & Linden, A. (2007). A conceptual framework for targeting prediabetes with lifestyle, clinical, and behavioral management interventions. Disease Management, 10, 6 15. Braet, C., Mervielde, I., & Vandereycken, W. (1997). Psychological aspect s of childhood obesity: a controlled study in a clinical and nonclinical sample. Journal of Pediatric Psychology, 22, 59 71. Buysse, D. J., Ancoli Israel, S., Edinger, J. D., Lichstein, K. L., & Morin, C. M. (2006). Recommendations for a standard research assessment of insomnia. Sleep, 29 11551173. Calhoun, S. L., Vgontzas, A. N., Fernandez Mendoza, J., Mayes, S. D., Tsaoussoglou, M., Basta, M., & Bixler, E. O. (2011). Prevalence and risk factors of excessive daytime sleepiness in a community sample of young children: the role of obesity, asthma, anxiety/depression, and sleep. Sleep, 34 503 507. Cappuccio, F. P., Taggart, F. M., Kandala, N. B., Currie, A., Peile, E., Stranges, S., & Miller, M. A. (2008). Meta analysis of short sleep duration and obesity i n children and adults. Sleep, 31 619626. Carno, M. A., Ellis, E., Anson, E., Kraus, R., Black, J., Short, R., & Connolly, H. V. (2008). Symptoms of sleep apnea and polysomnography as predictors of poor quality of life in overweight children and adolescents. Journal of Pediatric Psychology, 33 269278. Corkum, P., Tannock, R., & Moldofsky, H. (1998). Sleep disturbances in children with attention deficit/hyperactivity disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 37 637 646 88

PAGE 89

Crabtree, V. M., Varni, J. W., & Gozal, D. (2004). Healthrelated quality of life and depressive symptoms in children with suspected sleepdisordered breathing. Sleep, 27 11311138. Crosby, B., LeBourgeois, M. K., & Harsh, J. (2005). Racial differences in reported napping and nocturnal sleep in 2to 8 year old children. Pediatrics, 115, 225 232. Dietz, W. H. (1998). Health consequences of obesity in youth: childhood predictors of adult disease. Pediatrics, 101, 518 525. Epstein, L.H., Klein, K.R., & Wis niewski, L. (1994). Child and parent factors that influence psychological problems in obese children. International Journal of Eating Disorders, 15, 151 158. Epstein, L.H., Myers, M.D., & Anderson, K. (1996). The association of maternal psychopathology and family socioeconomic status with psychological problems in obese children. Obesity Research, 4, 65 74. Fallone, G., Owens, J. A., & Deane, J. (2002). Sleepiness in children and adolescents: clinical implications. Sleep Medicine Reviews, 6, 287306. Field, A. (2009). Everything you ever wanted to know about statistics. In Andy Field (Ed.), Discovering Statistics Using SPSS (3rd ed., pp. 3738). London: Sage. Franciosi, J. P., Hommel, K. A., DeBrosse, C. W., Greenberg, A. B., Greenler, A. J., Abonia, J. P., . Varni, J. W. (2012). Quality of life in paediatric eosinophilic oesophagitis: What is important to patients? Child: Care, Health and Development, 38, 477 483. Freedman, D. S., Dietz, W. H., Srinivasan, S. R., & Berenson, G. S. (1999). The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics, 103, 1175 1182. Friedlander, S. L., Larkin, E. K., Rosen, C. L., Palermo, T. M., & Redline, S. (2003). Decreased quality of life associated wit h obesity in school aged children. Archives of Pediatrics & Adolescent Medicine, 157, 1206 1211. Gallant, M. P. (2003). The influence of social support on chronic illness self management: a review and directions for research. Health Education & Behavior 30, 170195. Gangwisch, J. E., Heymsfield, S. B., Boden Albala, B., Buijs, R. M., Kreier, F., Pickering, T. G., . Malaspina, D. (2006). Short sleep duration as a risk factor for hypertension: Analyses of the first National Health and Nutrition Examination Survey. Hypertension, 47, 833839. 89

PAGE 90

Garnett, S. P., Baur, L. A., Srinivasan, S., Lee, J. W., & Cowell, C. T. (2007). Body mass index and waist circumference in midchildhood and adverse cardiovascular disease risk clustering in adolescence. The American Journal of Clinical Nutrition, 86 549555. Gerich, J. E. (2005). The importance of tight glycemic control. The American Journal of Medicine, 118 7S 11S. Giannotti, F., Cortesi, F., Sebastiani, T., & Ottaviano, S. (2002). Circadian preference, sleep and day time behaviour in adolescence. Journal of Sleep Research, 11, 191199. Grad, F. P. (2002). The Preamble of the Constitution of the World Health Organization. Bulletin of the World Health Organization, 80, 981984. Graef, D.M., Janicke, D.M., McCrae, C., Si lverstein, J.H. (in press). Quality of life in obese youth with and without sleep problems. Childrens Health Care, 42. Guilfoyle, S. M., Zeller, M. H., & Modi, A. C. (2010). Parenting stress impacts obesity specific healthrelated quality of life in a pediatric obesity treatment seeking sample. Journal of Developmental and Behavioral Pediatrics, 31, 17 25. Guo, S. S., Wu, W., Chumlea, W. C., & Roche, A. F. (2002). Predicting overweight and obesity in adulthood from body mass index values in childhood and adolescence. The American Journal of Clinical Nutrition 76, 653 658. Guyatt, G.H., Feeny, D.H., & Patrick, D.L. (1993). Measuring health related quality of life. Annals of Internal Medicine, 118 622 629. Hansen, M., Janssen, I., Schiff, A., Zee, P.C., & D ubocovich, M.L. (2005). The impact of school daily schedule on adolescent sleep. Pediatrics, 115, 15551561. Hart, C. N., Cairns, A., & Jelalian, E. (2011). Sleep and obesity in children and adolescents. Pediatric Clinics of North America, 58 715 733. Har t, C. N., Palermo, T.M., & Rosen, C.L. (2005). Health related quality of life among children presenting to a pediatric sleep disorders clinic. Behavioral Sleep Medicine, 3, 4 17. Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed var iable mediation, moderation, and conditional process modeling [White paper]. Retreived from http://www.afhayes.com/public/process2012.pdf Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41 924 936. 90

PAGE 91

Herzer, M., Godiwala, N., Hommel, K.A., Driscoll, K., Mitchell, M., Crosby, L.E., Piazza Waggoner, C., Zeller, M.H., & Modi, A.C. (2010). Family functioning in the context of pediatric c hronic conditions. Journal of Developmental and Behavioral Pediatrics, 31, 26 34. Hill, J. O. (2006). Understanding and addressing the epidemic of obesity: an energy balance perspective. Endocrine Reviews, 27, 750761. Hiscock, H., Canterford, L., Ukoumunn e, O. C., & Wake, M. (2007). Adverse associations of sleep problems in Australian preschoolers: National population study. Pediatrics, 119, 86 93. Hughes, A.R., Farewell, K., Harris, D., & Reilly, J.J. (2007). Quality of life in a clinical sample of obese children. International Journal of Obesity, 31 39 44. Ingerski, L. M., Janicke, D. M., & Silverstein, J. H. (2007). Brief report: quality of life in overweight youththe role of multiple informants and perceived social support. Journal of Pediatric Psychology, 32 869 874. Ingerski, L. M., Laffel, L., Drotar, D., Repaske, D., & Hood, K. K. (2010). Correlates of glycemic control and quality of life outcomes in adolescents with type 1 diabetes. Pediatric Diabetes, 11, 563571. Irwin, M. R., & Miller, A. H. ( 2007). Depressive disorders and immunity: 20 years of progress and discovery. Brain, Behavior, and Immunity, 21, 374383. Ivanenko, A., Crabtree, V. M., Obrien, L. M., & Gozal, D. (2006). Sleep complaints and psychiatric symptoms in children evaluated at a pediatric mental health clinic. Journal of Clinical Sleep Medicine, 2, 42 48. Janicke, D. M., Marciel, K. K., Ingerski, L. M., Novoa, W., Lowry, K. W., Sallinen, B. J., & Silverstein, J. H. (2007). Impact of psychosocial factors on quality of life in over weight youth. Obesity, 15 17991807. Janssen, I., Craig, W. M., Boyce, W. F., & Pickett, W. (2004). Associations between overweight and obesity with bullying behaviors in school aged children. Pediatrics, 113 11871194. Jenni, O. G., Molinari, L., Caflis ch, J. A., & Largo, R. H. (2007). Sleep duration from ages 1 to 10 years: variability and stability in comparison with growth. Pediatrics, 120, e769776. Kim, E., Hwang, J. Y., Woo, E. K., Kim, S. S., Jo, S. A., & Jo, I. (2005). Body mass index cutoffs for underweight, overweight, and obesity in South Korean schoolgirls. Obesity Research, 13, 15101514. 91

PAGE 92

King, C. R., Knutson, K. L., Rathouz, P. J., Sidney, S., Liu, K., & Lauderdale, D. S. (2008). Short sleep duration and incident coronary artery calcification. Journal of the American Medical Association, 300, 2859 2866. Knutson, K. L. (2007). Impact of sleep and sleep loss on glucose homeostasis and appetite regulation. Sleep Medicine Clinics, 2 187197. Kuczmarski, R. J., Ogden, C. L., Guo, S. S., Grummer S trawn, L. M., Flegal, K., Wei, R., & Johnson, C. L. (2002). 2000 CDC growth charts for the United States: methods and development. Vital Health Statistics, 246, 1 190. Kushida, C.A., Littner, M.R., Morgenthaler, T., Alessi, C.A., Bailey, D., Coleman, J.& Wise, M. (2005). Practice parameters for the indications for Polysomnography and related procedures: An update for 2005. Sleep, 28 499 521. Landhuis, C. E., Poulton, R., Welch, D., & Hancox, R. J. (2008). Childhood sleep time and long term risk for obesit y: a 32year prospective birth cohort study. Pediatrics, 122 955 960. Lavigne, J. V., & Faier Routman, J. (1992). Psychological adjustment to pediatric physical disorders: a metaanalytic review. Journal of Pediatric Psychology, 17 133157. Leproult, R., & Van Cauter, E. (2010). Role of sleep and sleep loss in hormonal release and metabolism. Endocrine Development, 17 11 21. Lewandowski, A.S., Toliver Sokol, M., & Palermo, T.M. (2011). Evidencedbased review of subjective pediatric sleep measures. Journal of Pediatric Psychology, 36, 780 793. Littner, M., Kushida, C. A., Anderson, W. M., Bailey, D., Berry, R. B., Davila, D. G., . Standards of Practice Committee of the American Academy of Sleep, Medicine. (2003). Practice parameters for the role of act igraphy in the study of sleep and circadian rhythms: an update for 2002. Sleep, 26 337 341. Logan, D. E., & Scharff, L. (2005). Relationships between family and parent characteristics and functional abilities in children with recurrent pain syndromes: an investigation of moderating effects on the pathway from pain to disability. Journal of Pediatric Psychology, 30 698 707. Lorenzo, C., Wagenknecht, L.E., Hanley, A.J.G., Reweres, M.J., Karter, A.J., & Haffner, S.M. (2010). A1c between 5.7 and 6.4% as a mar ker for identifying prediabetes, insulin sensitivity and secretion, and cardiovascular risk factors. Diabetes Care, 33, 21042109. Lumeng, J. C., Somashekar, D., Appugliese, D., Kaciroti, N., Corwyn, R. F., & Bradley, R. H. (2007). Shorter sleep duration is associated with increased risk for being overweight at ages 9 to 12 years. Pediatrics, 120, 10201029. 92

PAGE 93

Masten, A. S. (2004). Regulatory processes, risk, and resilience in adolescent development. Annals of the New York Academy of Sciences, 1021, 310319. Masten, A. S. (2005). Peer relationships and psychopathology in developmental perspective: Reflections on progress and promise. Journal of Clinical Child and Adolescent Psychology, 34, 87 92. McClure, A. C., Tanski, S. E., Kingsbury, J., Gerrard, M., & Sa rgent, J. D. (2010). Characteristics associated with low self esteem among US adolescents. Academic Pediatrics, 10, 238 244 e232. Meier Ewert, H. K., Ridker, P. M., Rifai, N., Regan, M. M., Price, N. J., Dinges, D. F., & Mullington, J. M. (2004). Effect of sleep loss on C reactive protein, an inflammatory marker of cardiovascular risk. Journal of the American College of Cardiology, 43, 678 683. Meltzer, L. J., Montgomery Downs, H. E., Insana, S. P., & Walsh, C. M. (2012). Use of actigraphy for assessment in pediatric sleep research. Sleep Medicine Reviews, 16, 463475. Miller, I.W., Epstein, N.B., Bishop, D.S., & Keitner, G.I. (1985). The McMaster family assessment device: Reliability and validity. Journal of Marital and Family Therapy, 11 345 356. Mindell, J. A., & Owens, J.A. (2010). A clinical guide to pediatric sleep: Diagnosis and management of sleep problems (2nd ed.). Philadelphia, PA: Lippincott Williams & Wilkins Publishing. Modi, A.C., & Zeller, M.H. (2008). Validation of a parent proxy, obesity sp ecific quality of life measure: Sizing them up. Obesity, 16 2624 2633. Moore, M., Kirchner, H. L., Drotar, D., Johnson, N., Rosen, C., Ancoli Israel, S., & Redline, S. (2009). Relationships among sleepiness, sleep time, and psychological functioning in adolescents. Journal of Pediatric Psychology, 34, 11751183. Morin, C. M. (2003). Measuring outcomes in randomized clinical trials of insomnia treatments. Sleep Medicine Reviews, 7 263 279. Mullington, J. M., Haack, M., Toth, M., Serrador, J. M., & Meier Ew ert, H. K. (2009). Cardiovascular, inflammatory, and metabolic consequences of sleep deprivation. Progress in Cardiovascular Diseases, 51 294 302. Nixon, G. M., Thompson, J. M., Han, D. Y., Becroft, D. M., Clark, P. M., Robinson, E., . Mitchell, E. A. (2008). Short sleep duration in middle childhood: risk factors and consequences. Sleep, 31, 71 78. 93

PAGE 94

Nowicka, P., Santoro, N., Liu, H., Lartaud, D., Shawn, M.M., Goldberg, R., Guandalini, C., Savoye, M., Rose, P., & Caprio, S. (2011). Utility of hemoglobin A1c for diagnosing prediabetes and diabetes in obese children and adolescents. Diabetes Care, 34, 13061311. Ogden, C. L., Carroll, M. D., Kit, B. K., & Flegal, K. M. (2012). Prevalence of obesity and trends in body mass index among US children and adolesc ents, 19992010. Journal of the American Medical Association, 307, 483490. Pearce, M. J., Boergers, J., & Prinstein, M. J. (2002). Adolescent obesity, overt and relational peer victimization, and romantic relationships. Obesity Research, 10, 386393. Peso nen, A.K., Rikknen, K., Paavonen, E.J., Heinonen, K., Komsi, N., Lahti, J., Kajantie, E., Jrvenp, A.L., & Strandberg, T. (2010). Sleep duration and regularity are associated with behavioral problems in 8year old children. International Journal of Beh avioral Medicine, 17, 298305. Pinhas Hamiel, O., Singer, S., Pilpel, N., Fradkin, A., Modan, D., & Reichman, B. (2006). Healthrelated quality of life among children and adolescents: associations with obesity. International Journal of Obesity, 30, 267 272 Preacher, K. J., Curran, P.J., & Bauer, D.J. (2006). Computational tools for probing interaction effects in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational & Behavioral Statistics, 31 437 448. Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40 879 891. Raitakari, O. T., Juonala, M., & Viikari, J. S. (2005). Obesity in childho od and vascular changes in adulthood: insights into the Cardiovascular Risk in Young Finns Study. International Journal of Obesity, 29, S101104. Rosen, C. L., Palermo, T. M., Larkin, E. K., & Redline, S. (2002). Healthrelated quality of life and sleepdi sordered breathing in children. Sleep, 25 657 666. Sadeh, A., & Acebo, C. (2002). The role of actigraphy in sleep medicine. Sleep Medicine Reviews, 6, 113124. Sandstrom, M. J., & Schanberg, L. E. (2004). Peer rejection, social behavior, and psychological adjustment in children with juvenile rheumatic disease. Journal of Pediatric Psychology, 29, 29 34. Schwimmer, J. B., Burwinkle, T. M., & Varni, J. W. (2003). Healthrelated quality of life of severely obese children and adolescents. Journal of the Americ an Medical Association 289 1813 1819. 94

PAGE 95

Seegers, V., Petit, D., Falissard, B., Vitaro, F., Tremblay, R. E., Montplaisir, J., & Touchette, E. (2011). Short sleep duration and body mass index: a prospective longitudinal study in preadolescence. American Jour nal of Epidemiology, 173, 621629. Shoup, J. A., Gattshall, M., Dandamudi, P., & Estabrooks, P. (2008). Physical activity, quality of life, and weight status in overweight children. Quality of Life Research, 17, 407412. Sensewear Bodymedia. (2008). Sleep and Armband Examples [Brochure]. Pittsburgh, PA: 2013 Bodymedia Inc. Smaldone, A., Honig, J. C., & Byrne, M. W. (2007). Sleepless in America: inadequate sleep and relationships to health and well being of our nation's children. Pediatrics, 119 S29 37. S nell, E. K., Adam, E. K., & Duncan, G. J. (2007). Sleep and the body mass index and overweight status of children and adolescents. Child Development, 78, 309 323. Soric, M., Turkalj, M., Kucic, D., Marusic, I., Plavec, D., & Misigoj Durakovic, M. (2013). V alidation of a multi sensor activity monitor for assessing sleep in children and adolescents. Sleep Medicine, 14, 201205. Spiegel, K., Leproult, R., L'HermiteBaleriaux, M., Copinschi, G., Penev, P. D., & Van Cauter, E. (2004). Leptin levels are dependent on sleep duration: relationships with sympathovagal balance, carbohydrate regulation, cortisol, and thyrotropin. Journal of Clinical Endocrinology & Metabolism, 89, 57625771. Spilsbury, J. C., Storfer Isser, A., Drotar, D., Rosen, C. L., Kirchner, L. H., Benham, H., & Redline, S. (2004). Sleep behavior in an urban US sample of school aged children. Archives of Pediatrics & Adolescent Medicine, 158 988 994. Spruyt, K., Gozal, D., Dayyat, E., Roman, A., & Molfese, D. L. (2011). Sleep assessments in healthy school aged children using actigraphy: concordance with polysomnography. Journal of Sleep Research, 20, 223 232. Storch, E. A., Lewin, A., Silverstein, J. H., Heidgerken, A. D., Strawser, M. S., Baumeister, A., & Geffken, G. R. (2004). Peer victimization and psychosocial adjustment in children with type 1 diabetes. Clinical Pediatrics, 43 467 471. Taal, E., Rasker, J. J., Seydel, E. R., & Wiegman, O. (1993). Health status, adherence with health recommendations, self efficacy and social support in patients with rheumatoid arthritis. Patient Education and Counseling 20, 63 76. Teller, A., & Crossley, M. (2004). Ambient and wearable computing. White Pa pers Body Media www.bodymedia.com 95

PAGE 96

Thompson, R.J., & Gustafson, K.E (1996). Models of adaptation. In R.J. Thompson & K.E. Gustafson (Eds.), Adaptation to chronic childhood illness (pp. 131156). Washington, DC: American Psychological Association. Tickamyer, A.R., & Duncan, C.M. (1990). Poverty and opportunity structure i n rural America. Annual Review of Sociology, 16 67 86. Touchette, E., Petit, D., Tremblay, R. E., Boivin, M., Falissard, B., Genolini, C., & Montplaisir, J. Y. (2008). Associations between sleep duration patterns and overweight/obesity at age 6. Sleep, 31, 15071514. Trost, S. G., Pate, R. R., Freedson, P. S., Sallis, J. F., & Taylor, W. C. (2000). Using objective physical activity measures with youth: how many days of monitoring are needed? Medicine & Science in Sports & Exercise, 32, 426 431. van Wouwe, N. C., Valk, P. J., & Veenstra, B. J. (2011). Sleep monitoring: A comparison between three wearable instruments. Military Medicine 176, 811 816. Varni, J.W., Burwinkle, T.M., Katz, E.R., Meeske, K. & Dickinson, P. (2002). The PedsQL in pediatric cancer: Reliability and validity of the Pediatric Quality of Life Inventory Generic Core Scales, Multidimensional Fatigue Scale, and Cancer Module. Cancer, 94, 20902106. Varni, J.W., Burwinkle, T.M., & Lane, M.M. (2005). Healthrelated quality of life me asurement in pediatric clinical practice: An appraisal and precept for future research and application. Health and Quality of Life Outcomes, 3, 34. Varni, J. W., Burwinkle, T. M., Seid, M., & Skarr, D. (2003). The PedsQLTM 4.0 as a pediatric population health measure: feasibility, reliability, and validity. Ambulatory Pediatrics, 3, 329 341. Varni, J. W., Limbers, C. A., & Burwinkle, T. M. (2007). Impaired healthrelated quality of life in children and adolescents with chronic conditions: a comparative anal ysis of 10 disease clusters and 33 disease categories/severities utilizing the PedsQL 4.0 Generic Core Scales. Health and Quality of Life Outcomes, 5, 43. Varni, J. W., Seid, M., & Rode, C. A. (1999). The PedsQL: Measurement model for the pediatric quality of life inventory. Medical Care, 37, 126 139. Varni, J. W., & Wallander, J.L. (1988). Pediatric chronic disabilities. In D.K. Routh (Ed.), Handbook of Pediatric Psychology. New York: Guilford Press. von Weiss, R. T., Rapoff, M. A., Varni, J. W., Lindsley, C. B., Olson, N. Y., Madson, K. L., & Bernstein, B. H. (2002). Daily hassles and social support as predictors of adjustment in children with pediatric rheumatic disease. Journal of Pediatric Psychology, 27 155165. 96

PAGE 97

Vyas, N., Farringdon, J., Andre, D., & Stivoric, J.I. (2012). Machine learning and sensor fusion for estimating continuous energy expenditure. AI Magazine, 33, 55. Wallander, J., Thompson, R.J., & AlrikssonSchmidt, A. (2003). Psychosocial adjustment of children with chronic physical conditions In Michael C. Roberts (Ed.), Handbook of Pediatric Psychology (3rd ed., pp. 141158). New York, NY: Guilford Press. Wallander, J., & Varni, J. W. (1998). Effects of pediatric chronic physical disorders on child and family adjustment. Journal of Child Psy chology and Psychiatry, 39, 29 46. Wallander, J., Varni, J. W., Babani, L., Banis, H. T., & Wilcox, K. T. (1989). Family resources as resistance factors for psychological maladjustment in chronically ill and handicapped children. Journal of Pediatric Psychology, 14, 157 173. Weiss, A., Xu, F., Storfer Isser, A., Thomas, A., Ievers Landis, C. E., & Redline, S. (2010). The association of sleep duration with adolescents' fat and carbohydrate consumption. Sleep, 33 12011209. Westerlund, L., Ray, C., & Roos, E (2009). Associations between sleeping habits and food consumption patterns among 1011year old children in Finland. British Journal of Nutrition, 102, 15311537. Bodymedia, Inc. (2011). What is bodymedia fit. Retrieved March 18, 2013, from http://www.bodymedia.com Whitaker, R. C., Wright, J. A., Pepe, M. S., Seidel, K. D., & Dietz, W. H. (1997). Predicting obesity in young adulthood from childhood and parental obesity. The New England Journal of Medicine, 337, 8 69 873. Williams, J., Wake, M., Hesketh, K., Maher, E., & Waters, E. (2005). Health related quality of life of overweight and obese children. The Journal of the American Medical Association, 293, 7076. Wing, Y.K., Li, S.X., Li, A.M., Zhang, J., & Kong, A.P.S. (2009). The effect of weekend and holiday sleep compensation on childhood overweight and obesity. Pediatrics, 124, e994e1000. Witt, W.P., Riley, A.W., & Coiro, M.J. (2003). Child functional status, family stressors, and psychosocial adjustment among school aged children with disabilities in the United States. Archives of Pediatrics & Adolescent Medicine, 157, 687 695. Wolfson, A.R., & Carskadon, M.A. (1998). Sleep schedules and daytime functioning in adolescents. Child Development, 69, 875887. 97

PAGE 98

Wysoc ki, T., Harris, M. A., Buckloh, L. M., Mertlich, D., Lochrie, A. S., Taylor, A., . White, N. H. (2006). Effects of behavioral family systems therapy for diabetes on adolescents' family relationships, treatment adherence, and metabolic control. Journal of Pediatric Psychology, 31 928 938. Zametkin, A. J., Zoon, C. K., Klein, H. W., & Munson, S. (2004). Psychiatric aspects of child and adolescent obesity: a review of the past 10 years. Journal of the American Academy of Child and Adolescent Psychiatry, 43 134150. Zee, P. C., & Turek, F. W. (2006). Sleep and health: Everywhere and in both directions. Archives of Internal Medicine, 166, 16861688. Zeller, M. H., & Modi, A. C. (2006). Predictors of healthrelated quality of life in obese youth. Obesity, 14 122 130. Zeller, M.H., Reiter Purtill, J., & Ramey, C. (2008). Negative peer perceptions of obese children in the classroom environment. Obesity, 16, 755 762. Zeller, M. H., Saelens, B. E., Roehrig, H., Kirk, S., & Daniels, S. R. (2004). Psychological adjustment of obese youth presenting for weight management treatment. Obesity Research, 12, 15761586. 98

PAGE 99

BIOGRAPHICAL SKETCH Danielle was a graduate student in the Clinical and Health Psychology Program at the University of Florida, with a conc entration in pediatric psychology. She completed her bachelors degree in psychology at the University of Missouri in 2009. Danielle completed her pre doctoral internship at the Nemours/A.I. duPont Hospital for Children and received her Ph.D. in 2014. Sh e will complete her Post Doctoral Fellowship at St. Jude Childrens Research Hospital. Her clinical and research interests include health promotion the role of sleep disturbance in health behaviors and daytime functioning, and the relationship between beh avior health variables and adjustment in pediatric populations. 99