Loyola University Chicago

Loyola eCommons
Master's Theses Theses and Dissertations

2013

Social-Environmental Predictors of Health-Related

Quality of Life in Youth with Spina Bifida
Caitlin Beth Murray
Loyola University Chicago

Recommended Citation
Murray, Caitlin Beth, "Social-Environmental Predictors of Health-Related Quality of Life in Youth with Spina Bifida" (2013). Master's
Theses. Paper 2242.
http://ecommons.luc.edu/luc_theses/2242

This Thesis is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in
Master's Theses by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
Copyright © 2013 Caitlin Beth Murray
 LOYOLA UNIVERSITY CHICAGO

SOCIAL-ENVIRONMENTAL PREDICTORS OF HEALTH-RELATED

QUALITY OF LIFE IN YOUTH WITH SPINA BIFIDA

A THESIS SUBMITTED TO

THE FACULTY OF THE GRADUATE SCHOOL

IN CANDIDACY FOR THE DEGREE OF

MASTER OF ARTS

PROGRAM IN CLINICAL PSYCHOLOGY

BY

CAITLIN B. MURRAY

CHICAGO, IL

DECEMBER 2013
Copyright by Caitlin B. Murray, 2013
All rights reserved.
 TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. iv

LIST OF FIGURES .............................................................................................................v

CHAPTER ONE: INTRODUCTION ..................................................................................1

CHAPTER TWO: REVIEW OF THE RELEVANT LITERATURE .................................4

Conceptualizing Quality of Life .......................................................................................4
Definition of Health-Related Quality of Life ...................................................................5
Measurement of Health-Related Quality of Life ..............................................................6
Health-Related Quality of Life in Youth With Chronic Illnesses ....................................8
Health-related Quality of Life in Youth with SB .............................................................9
General Issues With Current Research ...........................................................................16
The Current Study...........................................................................................................19
Hypotheses ......................................................................................................................21

CHAPTER THREE: METHODS ......................................................................................24

Participants......................................................................................................................24
Procedure ........................................................................................................................27
Measures .........................................................................................................................29
Statistical Treatment .......................................................................................................39

CHAPTER FOUR: RESULTS ..........................................................................................43

Preliminary Analyses ......................................................................................................43
Descriptive Information on Health-Related Quality of Life ...........................................44
Longitudinal Analyses of Health-Related Quality of Life in Sample A.........................50
Longitudinal Analyses of Health-Related Quality of Life in Sample B .........................54

CHAPTER FIVE: DISCUSSION ......................................................................................58

APPENDIX A: QUESTIONNAIRE MEASURES ...........................................................69

APPENDIX B: OBSERVATIONAL CODING MANUAL ...........................................101

REFERENCES ................................................................................................................111

VITA ................................................................................................................................120

iii
 LIST OF TABLES

Table 1. Youth Demographic Information of Samples A and B at Time 1 .................27

Table 2. Scale Descriptives, Cross-Informant Correlations, and Differences on

PedsQL Generic Score: Sample A .........................................................................45

Table 3. One-Sample t Tests Comparing Time 1 and 2 HRQOL: Spina Bifida

Versus Chronically Ill and Healthy Populations: Child Report.............................47

Table 4. One-Sample t Tests Comparing Time 1 and 2 HRQOL: Spina Bifida

Versus Chronically Ill and Healthy Populations: Parent Report ...........................48

Table 5. Correlation Matrix for Socio-Environmental Predictors of Time 2

Health-Related Quality of Life: Sample A ............................................................52

Table 6. Hierarchical Regressions: Time 1 Socio-Environmental Predictors of

Time 2 Psychosocial HRQOL ...............................................................................55

Table 7. Correlation Matrix for Socio-Environmental Variables and Time 2

Health-Related Quality of Life: Sample B ............................................................56

iv
 LIST OF FIGURES

Figure 1. Study model: Social-environmental predictors of health-related

quality of life in youth with spina bifida. Model tested in two samples:
Sample A = 140 children with spina bifida (Ages 8-15 at Time 1; Ages 10-
17 at Time 2); Sample B = 61 children with spina bifida (Ages 15/16 at
Time 1; Ages 16/17 at Time 2). ...............................................................................3

v
 CHAPTER ONE

INTRODUCTION

Health-related quality of life (HRQOL) is a multidimensional construct including

an  individual’s  physical  and  mental  health  and  psychosocial  well-being (De Civita et al.,

2005), and the measurement of HRQOL has been recognized as a key marker of health

outcomes in pediatric populations (Eiser & Jenney, 2007). Due to medical and

technological advances, an increasing number of individuals with chronic illnesses are

living longer. As such, research that investigates improvements in HRQOL in youth with

chronic illnesses has become essential. Indeed, the number of studies examining HRQOL

in pediatric populations has increased markedly; spina bifida (SB) is one among several

chronic illnesses that has received increased attention with regard to HRQOL assessment

in the past decade (see Sawin & Bellin, 2010 for a review).

The experience of a chronic illness may have deleterious consequences on several

aspects  of  a  youth’s  life.  In  particular,  SB is a relatively common congenital birth defect

associated with a multitude of physical and cognitive impairments (e.g., orthopedic

abnormalities, urinary and bowel difficulties; Fletcher & Brei, 2010) as well as individual

and contextual social-environmental difficulties (e.g., poor social competence, a stressful

family environment; Alriksson-schmidt, Wallander, & Biasini, 2007). Due to the range of

physical, cognitive, and social impairments associated with this condition, youth with SB

may be at an increased risk of reduced HRQOL. Research on HRQOL for children and

adolescents with SB has begun to identify demographic, illness-related, and social-

1
 2

environmental factors that are associated with HRQOL. Elucidating factors that influence

HRQOL is an important step in informing the development of interventions to improve

HRQOL in youth with SB. Thus far, extant research investigating HRQOL in youth with

SB has tended to focus on non-modifiable demographic and illness-specific correlates,

such as age, gender, and degree of mobility impairment. Studies examining modifiable

social-environmental  factors  on  youth’s  HRQOL  may  be  particularly  important  in  

informing future interventions for youth with SB.

In addition, despite the importance of this work, current research has several

methodological weaknesses, including the utilization of mixed samples, small sample

sizes, single informants, and cross-sectional designs. The current study seeks to address

these weaknesses and bridge critical gaps in the literature by testing a longitudinal, multi-

method and multi-informant model of individual and contextual social-environmental

predictors of HRQOL in youth with SB across two independent studies (see Figure 1).

The following sections provide a general overview of historical and current

conceptualizations, measurement issues related to quality of life as a construct, and an

extensive review of studies that have investigated demographic, illness-specific, and

social-environmental correlates of HRQOL in youth with SB. Methodological

weaknesses and gaps in current literature are identified and a detailed description of the

current study is provided.

 Time 1: Social-environmental Predictors

Community Support
Time 2: HRQOL

Family Environment
(e.g. stress, conflict,
cohesion)

Health-related QOL [Total

Social
score]
Adjustment (e.g.
PedsQL (Sample A)
social problems,
HRQOL-SB (Sample B)
social acceptance,
peer support
Covariates:
Demographic:
Age, SES; Gender; IQ
Illness Parameters:
Illness severity;
Pain frequency

Figure 1. Study model: Social-environmental predictors of health-related quality of life in youth with spina bifida. Model tested
in two samples: Sample A = 140 children with spina bifida (Ages 8-15 at Time 1; Ages 10-17 at Time 2); Sample B = 61
children with spina bifida (Ages 15/16 at Time 1; Ages 16/17 at Time 2).

3
 CHAPTER TWO

REVIEW OF THE RELEVANT LITERATURE

Conceptualizing Quality of Life

In 1948, the World Health Organization (WHO) defined quality of life (QOL) as

the presence of physical, mental, and social well-being, and not just the absence of

disease (as cited in Testa & Simonson 1996). Since this time, the construct of QOL has

received considerable attention and has undergone various conceptualizations. Two

approaches to the assessment of QOL have emerged. Generic QOL assessment is based

primarily on developmental models and takes a broader view of an individual’s  life,  

including relations with family and friends, job or school functioning, and life goals

(Koot & Wallander 2001). Rapheal and colleagues, for example, created a measure of

overall  quality  of  life  to  assess  an  adolescent’s  satisfaction  and  the importance of various

activities and concepts, such as health, sexuality, personal hygiene, and exercise

(Raphael, Rukholm, Brown, Hillbailey, & Donato, 1996).

Conversely, HRQOL is based on developmental as well as health outcomes, and

is a more specific  measure  of  an  individual’s  perceptions  of  well-being that may be

impacted by his or her disease or condition. HRQOL typically addresses symptoms,

functional  status,  psychological  and  social  functioning,  and  an  individual’s  perceived  

ability to participate in and enjoy physical and social activities, given the constraints of

his/her health status (Eiser & Morse, 2001a).

4
 5

Definition of Health-Related Quality of Life

Functional impairment and health-related quality of life (HRQOL) are often used

interchangeably, but meta-analytic research suggests that there is an important distinction

between these terms (Smith et al., 1999). The construct of HRQOL includes assessment

of  the  individual’s  perception  of  the  impact that a disease or condition has on his or her

physical health status, as well as on the psychosocial health dimensions of emotional,

social, and role functioning; HRQOL assessment focuses on the experience of the illness.

Functional impairment, on the other hand, is a concept related to but distinct from

HRQOL. Functional  impairment  has  been  defined  as  limitations  in  a  person’s  ability  to  

perform activities relevant to daily life including physical, social, and personal activities

(Stein & Jessop, 1990). While functional impairment measures activity limitations due to

a chronic disability or illness, health-related quality of life measures the perceived impact

of  an  illness  on  an  individual’s  physical  and  psychosocial  functioning.  In other words, the

meaning of the illness to the individual (i.e., the  individual’s  experience  with  the  illness)

is considered to be separate from functional limitations caused by disease process.

Psychosocial health is a particularly component of HRQOL, and has been called the

“hidden morbidity”  in pediatric clinical practice due to the underidentification of

psychosocial problems in routine pediatric care (Varni et al., 2002). The need to reveal

and identify this hidden morbidity has led to increased support to use a quality of life

construct that is multidimensional, consisting of physical, mental, and social health

dimensions as delineated by the World Health Organization (as cited in Testa &

Simonson 1996).
 6

For patients with chronic health conditions, the goal of health care is to restore

them to the fullest health possible by improving symptom management, treatment

adherence, and their ability to cope with the negative impact of their condition. For this

reason, some researchers have indicated that HRQOL may be more important than

biomedical measures when assessing patients with chronic health conditions (Coons &

Kaplan, 1993). Thus, to ensure that children receive the best medical care possible with

qualified and competent professionals, it is essential that we assess their experiences as it

relates to their chronic illness.

Measurement of Health-Related Quality of Life

In the past decade, there has been a dramatic increase in the number of measures

of HRQOL in pediatric populations (Drotar, 2004). However, the measurement of

HRQOL has presented a number of methodological challenges and debates. For one,

researchers have developed and employed both generic and disease-specific HRQOL

instruments, yet there is a lack of agreement regarding which are preferred. Generic

instruments usually include a global or summary measure of multiple domains of

HRQOL. Global ratings of HRQOL allow for comparisons across different groups (e.g.

cancer vs. SB) and have undergone a significant amount of development and testing.

Examples of these instruments include the Child Health Questionnaire (CHQ-50;

Langraf, Abetz & Ware, 1996) the Pediatric Quality of Life Questionnaire (PedsQL

4.0TM; Varni, Seid, & Kurtin, 2001), the Child Health and Illness Profile (CHIP; Starfield

et al., 1995), and the Youth Quality of Life (YQOL; Edwards, Huebner, Connell, &

Patrick, 2002). However, researchers have noted that these instruments may have limited
 7

utility in detecting clinically significant changes  in  an  individual’s  condition  over  time  

(Quittner, 2003).

Unlike generic measures of HRQOL, disease-specific measures include items that

address specific challenges associated with a given condition or illness. Several disease-

specific instruments have been developed for children and adolescents, including asthma

(Varni, Burwinkle, Rapoff, Kamps, & Olson, 2004), epilepsy (Cramer, Westbrook,

Devinsky, Perrine, Glassman & Camfield, 1999) cancer (Goodwin, Boggs, & Graham-

Pole, 1994; Varni, Katz, Seid, Quiggins, Friedman-Bender, 1998), cystic fibrosis (Modi

& Quittner, 2003), diabetes (Ingersoll &Marrero, 1991), and SB (Parkin et al., 1997).

Advantages include the ability to detect small but clinically meaningful changes, and a

greater clinical relevance to patients, families, and healthcare providers. The Food and

Drug Administration (FDA) has also recognized disease-specific measures of HRQOL as

potential primary or secondary outcome measures in clinical trials (Goss & Quittner,

2007). However, such disease-specific measures do not permit comparisons across

different illness groups.

More recently, unique challenges have emerged with the shift of HRQOL

measurement from adults to children. First, new dimensions of functioning relevant to

children and adolescents have been identified, such as social and academic functioning at

school (Modi & Quittner, 2003). There has also been increased attention on creating

developmentally  appropriate  measures,  suited  to  the  child’s  age,  reading  ability,  and  

emotional maturity (Eiser & Jenney, 2007). Finally, HRQOL has historically been

assessed using parent proxy reports. Researchers have noted special circumstances in
 8

which a child is too young, ill, or fatigued to complete questionnaire measures; otherwise,

youth self-report is considered to be an essential source in the measurement of HRQOL.

Although children with chronic illnesses and their parents may have higher agreement

rates compared to healthy populations (Eiser & Morse, 2001b), research indicates that

there is lack of congruence between child and parent proxy report of HRQOL, especially

in the emotional and physical domains (Modi & Quittner, 2003; Theunissen et al., 1998).

Thus, several researchers have emphasized the importance of assessing HRQOL from

both the child  and  parent  perspectives.  Assessment  of  parents’  reports  of  HRQOL  may  be  

especially  valuable  because  of  the  parents’  role  in  disease  management  and  healthcare  

utilization (Eiser & Jenney, 2007).

Health-Related Quality of Life in Youth With Chronic Illnesses

Morbidity and mortality have historically been used to evaluate the efficacy of

management and treatment of chronic illnesses. However, due to medical and

technological advances, an increasing number of these individuals are living longer. As

such, measuring and improving the quality of life in individuals with chronic health

conditions has become increasingly important. Literature on HRQOL in pediatric

populations has steadily increased in the past decade, and HRQOL has been studied in a

variety of chronic medical conditions Available research in pediatric populations

assessing HRQOL includes oncology (e.g., Varni, Limbers, & Burwinkle, 2007), arthritis

(e.g., Tennant et al., 2001), cystic fibrosis (Modi & Quittner, 2003) chronic pain (e.g.,

Hunfeld et al., 2001), and SB (e.g., Sawin & Bellin, 2010).

 9

The assessment of HRQOL has many research and clinical applications, and may

be critical for several reasons. First, it provides useful information regarding illness

burden  and  a  child’s  progress  and  responsiveness to treatment over time. Second,

HRQOL  is  a  multidimensional  construct  that  may  provide  a  clearer  picture  of  a  child’s  

functioning across several domains compared to the assessment of a single general

domain. Finally, data on HRQOL can be used to compare the efficacy of medical or

psychological interventions, establish the efficacy of new medications, and inform social

policies.

Health-related Quality of Life in Youth with SB

SB is a relatively common congenital birth defect, occurring in 3 out of 10,000

live births in the United States (Centers for Disease Control and Prevention, 2010). This

condition arises during the first month of pregnancy, during which the spinal column fails

to develop fully, resulting in exposure of a portion of the spinal cord (Sarwark, 1996).

Individuals with SB face a number of challenges, including physical and cognitive issues.

Physical difficulties may include varying degrees of motor paralyses, sensory loss,

orthopedic problems, and urinary and bowel incontinence (Fletcher & Brei, 2010).

Myelomeningocele is the most common and most severe type of SB, and is associated

with brain abnormalities, hydrocephalus, and cognitive impairments, including problems

with planning, orientation, shifting attention, and working memory (Rose & Holmbeck,

2007). Youth with SB typically learn to follow a strict medical regimen, which may

include taking medications, self-catheterization, following a specific bowel program,

regular skin checks, and pressure relief exercises. Further, cognitive and executive
 10

function may have a significant impact on social adjustment difficulties, which are

common in individuals with SB (Rose & Holmbeck, 2007; Devine et al., 2012).

Given the multitude of physical, neurocognitive, and social challenges associated

with this condition, youth with SB have an increased risk for reduced quality of life

(Cate, Kennedy, & Stevenson, 2002). In addition, advances in medical care, such as clean

intermittent catheterization for management of neurogenic balder, have substantially

reduced morbidity due to kidney disease in this population. Individuals with SB are now

expected to live into adulthood, and the emphasis on medical care has shifted to

improvement and enhancement of quality of life and promotion of independence in this

population (Danielsson et al., 2008). In fact, a study by Cate and colleagues found that

quality of life is dramatically impaired in children and adolescents with SB, such that

parents reported lower quality of life (over one SD lower) compared to youth with other

chronic physical conditions and psychiatric disorders (Cate, Kennedy, & Stevenson,

2002). However, thus far no studies have compared quality of life in youth with SB to

healthy or chronically ill youth using a well-validated measure of HRQOL. Further,

investigations of predictors and correlates of quality of life are essential. Research to date

(see Sawin & Bellin, 2010) has found several correlates of HRQOL in youth with SB

including: (a) demographic, (b) illness-related, and (c) social-environmental factors (e.g.,

social and contextual variables).

Demographic covariates of HRQOL. Common demographic correlates of

HRQOL that have been investigated in this population include age, gender, and

socioeconomic status (SES). As a child becomes older, the impairments and strains
 11

associated with their chronic illness may have a cumulative negative impact as the child

becomes increasingly aware of the severity of their condition. Moreover, beginning in

childhood, females with chronic health conditions may be at risk for worse psychosocial

adjustment (e.g., depression; Sawin, Bellin, Roux, Buran, & Brei, 2009). However,

available research examining the relationship between gender, age and HRQOL in youth

with SB is equivocal. Generally, age and gender are not directly related to HRQOL (e.g.,

Cate, Kennedy, & Stevenson, 2002; Leger 2005; MacNeily, Morrell, & Secord, 2005;

MacNeily, Jafari, Scott, Dalgetty & Afshar, 2009) with the exception of a few studies.

For example, a study by Verhoef and colleagues conducted on youth with SB (ages 16-

25) found that those older than 20 reported worse HRQOL in two domains of the SF-36:

body pain and general health (Verhoef, Post, Barf, Van Asbeck, Gooskens, & Prevo,

2007). Similarly, another study on adolescents and emerging adults with SB (ages 12 to

21) found that older age was significantly associated with lower levels of parent-reported

HRQOL (Sawin, Brei, Buran & Fastnenau, 2002). Additionally, Lemelle and colleagues

found an age and gender interaction, such that older females had lower HRQOL scores

than older men across many domains of the SF-36 (Lemelle, Guillemin, Aubert, Guys &

Lottmann 2006).

The effect of socioeconomic status on child outcomes has been well documented

and may be an important factor associated with HRQOL in children with SB.

Socioeconomic status can have an influence on a multitude of domains  in  a  child’s  life,  

including health (e.g., healthcare access), schooling, physical environment (e.g. home and

neighborhood environments), emotional well-being (e.g. stress), and family interactions

 12

(Park & Turnbull, 2002). Moreover, children with disabilities may be more likely to live

in low-income households; Park and Turnball found that twenty-eight percent of children

with disabilities are living in families whose total income is below the poverty line

compared to sixteen percent of children without disabilities (2002). Regardless, data on

the relationship between SES and HRQOL in youth with SB is inconsistent; some studies

have found that low SES is related to poorer HRQOL (Kulkarni, Cochrane, McNeely, &

Shams, 2008), yet others have not supported this association (Sawin et al., 2002; Bier,

Prince, Tremont, & Msall, 2005). Though findings are generally equivocal, the current

study will include age, gender, and SES as covariates in analyses to control for possible

effects of these demographics variables. Cognitive ability (i.e., IQ) has never been

investigated as a correlate of HRQOL in youth with SB, but will also be included as a

covariate in analyses due to evidence suggesting neurocognitive deficits negatively

impact psychosocial outcomes in this population, particularly social adjustment and

family relationships (Rose & Holmbeck, 2006; Holmbeck, Coakley, Hommeyer, Shapera

& Westhoven, 2000).

Illness-related covariates of HRQOL. The majority of research investigating

illness-related variables has shown no association between shunt status, continence, and

HRQOL in youth with SB (Sawin et al., 2002; Bier et al., 2005; Leger, 2005; MacNeily

et al., 2009; Muller-Godeffroy, Michael, Poster, Seidel, Schwarke, & Thyen, 2008;

Lemelle et al., 2006). Kirpalani and colleagues did find relationships between several

clinical variables (i.e., number of operations, bladder continence, and lesion level) and

parent report of child HRQOL. However, examination of adolescent report of HRQOL

 13

revealed that only bowel continence was significantly related to reduced HRQOL. One

unpublished study (using one of the samples included in the current study; Dissertation

Abstract, Abad, 2008) found that higher lesion levels predicted higher levels of quality of

life, which was a contradictory  finding.  However,  this  finding  supports  the  “marginality  

hypothesis,”  which  suggests  that  youth  with  less  severe  levels  of  disability  might  have  

more challenges with adjustment because they are unable to fit in with their able-bodied

peers, but also do not identify with severely disabled children (Holmbeck & Faier-

Routman, 1995). Additionally, research generally does not support an association

between mobility impairment and HRQOL (Kirpalani et al., 2000; Muller-Godeffroy et

al., 2008; Bier et al., 2005) with the exception of one study by Danielsson and colleagues

(Danielsson, Bartonek, Levey, McHale, Sponseller, & Saraste, 2009). This study found

that youth with limited functional mobility had significantly lower physical HRQOL.

Although several studies investigating illness-related correlates of HRQOL have

shown equivocal results, strong and consistent data have indicated that painful somatic

symptoms have a significant influence on HRQOL in youth with SB. Verhoef and

colleagues found that pain was related to several domains of HRQOL in individuals with

SB (age 16-35), including physical role, general vitality, and social functioning (Verhoef,

Post, Barf, van Asbeck, Gooskens, & Prevo, 2007). Similarly, another study found that

youth (ages 5-17) and young adults (ages 18-30) with SB who had chronic pain had

significantly lower physical and psychosocial HRQOL scores (Wood, Watts, Hauser,

Rouhani, & Frias, 2009). Further, a study investigating children ages 8-19 with SB found

that pain intensity was significantly associated with the PedsQL total summary score of
 14

HRQOL (Oddson, Clancy, & McGrath, 2006). In addition to a pain severity variable, an

illness severity composite was created for the current study to take into account several of

the above illness parameters, including lesion level, shunt status, and number of

surgeries.

Social-environmental predictors of HRQOL. Although the majority research

on HRQOL in youth with SB has primarily focused on demographic and illness-related

variables (See Sawin & Bellin 2010), some studies have assessed the role of individual

and contextual social-environmental factors. Sawin and colleagues found that adolescents

with  higher  communication  efficacy,  and  higher  use  of  the  coping  strategy  “developing  

social support”  reported  better  overall  quality  of  life  (Sawin,  Brei,  Burnan,  &  Fastenau,  

2002). General and disease-specific stress has also emerged as an important correlate of

quality of life in adolescents with SB, such that higher levels of stress may be associated

with decreased quality of life (Sawin, Brei, Burnan, & Fastenau, 2002; Alriksson-

schmidt, Wallander, & Biasini, 2007). Life stress has been associated with numerous

other negative outcomes in youth with SB, including increased levels of anxiety and

depression, and lower levels of self-esteem (Murch & Cohen, 1989). From a

developmental perspective, normative biological and psychological stresses faced in

adolescence may exacerbate the stress of living with a chronic illness and disability, thus

greatly impacting youths’  quality  of  life.

Additionally, several family factors have been examined in the literature; Sawin

and colleagues found that increases in parental hope and decreases in parental

overprotection predicted better HRQOL in youth with SB (Sawin et al., 2002). Similarly,
 15

Kirplani and colleagues found parental hope to be a better predictor HRQOL compared to

other disease characteristics in children and adolescents with SB (Kirpalani et al., 2007).

One study also found that maternal psychological distress predicted decreased quality of

life (unpublished dissertation; Abad, 2007). Further, examining the broader social

environment, Cate, Kennedy, and Stevenson found that family resources, as measured

using parental report of self-efficacy as a caregiver and family financial and community

support, was a significant predictor of quality of life in youth with SB (2002).

Using a stress-resiliency model, a recent study by Alriksson-Schmidt and

colleagues (2007) examined more complex relationships between stress, quality of life,

and the moderating effects of personal and social protective factors (e.g., social

competence, family functioning, and peer social engagement) in a mixed sample of

adolescents with a mobility disability. Although results indicated that social competence,

family functioning, and peer social engagement did not function as moderators, these

factors were found to be positive correlates of quality of life, independent of the level of

stress exposure. Moreover, results indicated a cumulative effect of social competence,

family functioning, and peer social engagement, such that having more of these factors

predicted better quality of life.

However, there were several methodological issues with this study. For one,

researchers included a mixed sample of adolescents with mobility disabilities (e.g.,

cerebal palsy, scoliosis and SB) and did not utilize a well-validated tool for measuring

quality of life (Quality of Student Life Questionnaire; QSL.Q, Keith & Schalock, 1995).

The authors proposed that strengths of this study included the use of parent and youth
 16

report of subjective (the QSL.Q; Keith & Schalock, 1994) and objective quality of life.

Yet, objective quality of life was measured using two items from the quality of life

module of the Behavioral Risk Factor Surveillance System (BRFSS; Center for Disease

Control and Prevention): the number of days that physical and mental health was poor for

the adolescent, respectively (Alriksson-Schmidt, Wallander, Biasini, 2007). These items

ask individuals to retrospectively report the number of days perceived as (themselves or

their daughter/son) being in poor mental or physical health. As such, these data seem to

be indicative of subjective, rather than objective, quality of life. In addition, because

results indicated stronger associations between the protective variables, stress, and quality

of life for adolescent report compared to parent report, some of the reported findings

could be attributed to common method variance.

General Issues With Current Research

Despite the variety of pediatric populations that have been assessed, much of the

research to date has provided HRQOL data on only a few conditions. One study by

Andelman and colleagues reviewed 688 articles and chapters on quality of life in

children, and found that the majority of research conducted between 1967 and 1996

(40%) focused on children with cancer, asthma, rheumatic disease, or epilepsy (1999).

Although research has increased in the past decade, there continues to be a paucity of

research examining quality of life in youth with SB.

A recent literature search conducted by Sawin and Bellin (2010) found 39 studies

addressing quality of life in individuals with SB of all ages. I conducted a similar

literature search with a narrower age criterion to estimate the total number of research
 17

articles addressing QOL or HRQOL in youth (age range 5-19) with SB from 1990 to

2011. This review included articles from Medline, PsychInfo, and Cumulative index to

Nursing and Allied Health Literature. The search was conducted using the key words

“quality  of  life,”  “health-related  quality  of  life,”  and  “SB,”  and  review  articles  and  

reference lists were examined to determine any additional potential studies. Using these

methods, a total of fourteen studies were identified that evaluated quality of life in

children and adolescents with SB.

My preliminary review of these studies revealed several key methodological

issues. The majority (77%) used samples with broad age ranges (e.g., ages 1-18; 5-18),

four studies included mixed samples (e.g. combining samples of cerebral palsy and SB),

and four used a small sample size (< 30 participants). Of the total sample of studies,

eleven (79%) used a single reporter, and only three utilized both parent and youth report.

Specifically, three studies were conducted using youth report, and eight using only parent

proxy report. Given that numerous researchers have noted that children have a unique

perspective on illness burden (Eiser & Morse, 2001), additional studies using both child

and parent proxy report of HRQOL is warranted.

Further, only half of these studies (57%) used well-validated measures of

HRQOL.  HRQOL  instruments  were  classified  as  “well  established”  if  they  had  extensive  

psychometric evidence (e.g. factor structure) and had been used in studies addressing SB

published by more than one investigator or investigative teams. These same criteria have

been used in previous evidence-based assessments of HRQOL in pediatric populations

(Palermo et al., 2008; Sawin & Bellin, 2010). Well-established measures included the
 18

Child Health Questionnaire (CHQ; Landgraf, Abetz, & Ware), the Pediatric Quality of

Life Inventory (PedsQL; Varni, Seid, & Rode, 1999) and the Short-Form health survey

(SF-26; Ware, Kosinski, & Keller, 1996). As of yet, there are no well-established illness-

specific measures of HRQOL for youth with SB. Finally, only a few studies investigated

potential predictors of HRQOL in this population, and most did not use statistical

procedures beyond cross-sectional, correlational analyses.

Beyond methodological issues, research investigating the impact of social-

environmental factors on HRQOL in youth with SB is not comprehensive, and this field

may benefit from identifying a broad range of social-environmental factors. Peer

relationships and friendships are an important domain of development in youth, yet little

attention has been devoted to social adjustment in youth with chronic illnesses (LaGreca,

1992). Treatment requirements and disease management may interfere with school

attendance and the development of stable peer relationships (Olsson, Boyce, &

Toumbourou, 2005). Moreover, youth with chronic illnesses and disabilities may have

increased social difficulties and poorly developed social skills, likely due to impairments

in the central nervous system. The majority of children with SB have hydrocephalus,

which has been linked to nonverbal cognitive deficits resulting in difficulties in social

interaction (Fletcher et al., 2005).

Indeed, research has consistently pointed to the salience of social difficulties

experienced in youth with SB. Youth with SB may have higher rates of social isolation

(Blum, Resnick, Nelson, & Germaine, 1991), social immaturity (Holmbeck et al., 2003),

and smaller social networks compared to typically developing youth (Ellerton, Ritchie, &
 19

Hirth, 1996). A recent study found that youth with SB generally have fewer close peer

relationships, and may experience lower levels of companionship and security in their

friendships (Devine, Holmbeck, Gayes, & Purnell, 2011). However, there is a paucity of

research on the implications of social adjustment on HRQOL in youth with SB.

Family  relationships  are  another  critical  component  to  any  child  or  adolescent’s  

healthy development. A focus on the impact of family functioning on HRQOL is

essential, given the considerable influence of the family on psychosocial adjustment in

youth with chronic illnesses (Drotar, 1997). The presence of a chronic illness may be a

source of considerable stress in a family, and parents may become the primary support

for maintaining medical regimens and encouraging a healthy lifestyle. Further, research

has consistently indicated a direct link between family functioning (e.g., conflict,

cohesion) and HRQOL across pediatric illness groups, such as youth with diabetes

(Pereira, Berg-Cross, Almeida, & Machado, 2008), asthma (Sawyer, Spurrier, Whaites,

Kennedy, Martin, & Baghurst, 2001), obesity (Janicke et al., 2007), and organ transplant

recipients (Devine et al., 2011). Yet few studies have comprehensively assessed the

impact of family environment, such as stress, conflict, and cohesion on HRQOL in youth

with SB within a multi-level social-environmental model (see Figure 1).

The Current Study

Given the importance of the social environment to adjustment in youth with SB,

and the relative dearth of knowledge in this area, the current study utilized a social-

ecological model to investigate individual and contextual social-environmental predictors

of HRQOL in youth with SB. Socio-ecological models (e.g., Bronfenbrenner, 1986) have
 20

provided a useful organizational framework for understanding the influence  of  a  youth’s  

social environment on HRQOL (e.g., Alriksson-Schmidt, Wallander, & Biasini, 2007). In

this model, the individual is at the center, surrounded by expanding levels of social-

contextual  influence.  A  more  comprehensive  model  of  a  youth’s  social environment

would  place  the  youth’s  individual  social  adjustment  at  the  center,  family  environment  at  

the next level, and the broader social world (e.g., community support) at the most distal

sphere of social influence. Thus, this study investigated youths’  individual  social  

adjustment (e.g., social problems, social acceptance, and peer support), family

environment (e.g., family stress, conflict, and cohesion), and community support as

predictors of HRQOL in youth with SB. In addition, the impact of social-environmental

factors on HRQOL, above and beyond demographic and illness-related parameters, was

examined.

Finally, the current study sought to address several methodological issues that

exist in studies to date. The use of single methods (e.g., subjective report only), single

reporters (e.g., child or parent report only), newly established measures of HRQOL, and

cross-sectional designs are among the most prominent weaknesses of current literature in

this area. Thus, this study includes (a) child, mother, and father reports (b) a well-

established measure of HRQOL (PedsQL; Varni, Seid, & Rode, 1999) (c) observational

and subjective measures of family environment (i.e., of family conflict and cohesion) and

(d) longitudinal data to predict future HRQOL. Further, the utility of this model was

examined in two independent samples of youth with SB recruited from separate federal-

and agency-funded studies to provide a cross-validation of the study hypotheses.

 21

Hypotheses

The two studies from which these samples were recruited were developed

independently, thus some constructs were assessed using different instrument measures.

Most importantly, in the first sample of 140 youth and their parents (ages 8-15 at initial

assessment; hereafter referred to as Sample A), the Pediatric Quality of Life Inventory

(PedsQL; Varni, Seid, & Rode, 1999) was used to measure HRQOL. In the second

sample of 61 youth and parents ages (ages 15-16 when a HRQOL measure was first used

in this longitudinal study beginning when participants were 8-9 years old; hereafter

referred to as Sample B), the Health-Related Quality of Life-SB (HRQOL-SB; Parkin, et

al., 1997) was used to measure HRQOL. Please see the Methods section for further

details of both Samples.

The first study aim of this study was to provide important descriptive information

regarding HRQOL in youth with SB. Similar to previous research on proxy reports of

HRQOL  (i.e.,  Panepinto,  O’Mahar,  DeBaun,  et  al.,  2005)  moderate agreement (e.g.,

correlations) was expected between child self-report and care-giver proxy report of

HRQOL in Sample A (hypothesis 1). Second, it was expected that youth with SB would

display impaired HRQOL across all domains (i.e., physical, emotional, social, school,

and total HRQOL) compared to mean scores of healthy populations of youth; and would

display lower mean scores compared to other illness groups (hypothesis 2). HRQOL

scores obtained in this sample were compared to a study conducted for the purposes of

demonstrating reliability and validity of the PedsQL scales in healthy and chronically ill

patient populations (Varni et al., 2001). This study by Varni and colleagues (2001)
 22

included children ages 5-18 with and without a chronic illness (N = 683 chronically ill; N

= 730 without a chronic illness). Finally, as noted, impairments and strains associated

with a chronic illness may have a cumulative negative impact as a child and their family

becomes increasingly aware of the severity of their condition and the health of the child

deteriorates. Thus, it was expected that HRQOL would significantly decrease over time

(hypothesis 3). For these first three hypotheses, Sample A only was utilized for several

reasons: 1) The use of the PedsQL as a measure of HRQOL in Sample A allowed for

comparisons with normed HRQOL data on chronically ill and healthy populations 2)

Parent-report of HRQOL was not obtained for Sample B and 3) In Sample B, the

HRQOL questionnaire for Time 1 (i.e., the Quality of Life in Epilepsy scale, or QOLIE)

was different from the HRQOL questionnaire used at Time 2 (i.e., the HRQOL-SB;

Parkin, et al., 1997), thus, an examination of changes in HRQOL over time could not be

conducted in this sample.

The second aim of this study to examine social-envionmental predictors of

HRQOL in both samples. In Sample A, it was expected that all social-environmental

factors measured when youth were ages 8-15 would prospectively predict HRQOL (in

youth ages 10-17) across both informants (e.g., child and parent; hypothesis 4).

Specifically, it was expected that better social acceptance, peer support, family cohesion,

and community support would prospectively predict improved HRQOL. Further, it was

hypothesized that decreased social problems, family conflict and stress would

prospectively predict improved HRQOL across informants. It was expected that social-

environmental predictors of HRQOL would be significant above and beyond all

 23

demographic and illness-related variables, including age, gender, SES, cognitive ability,

illness severity and pain frequency (hypothesis 5). The social-ecological model in this

study conceptualizes intrinsic personal characteristics as the most proximal influence on

child outcomes (with the child at the center). In accordance with this model, as well as

research indicating the salience social deficits in children and adolescents with SB, it was

expected that social adjustment variables (e.g., social problems, social acceptance, and

peer support) would be the strongest prospective predictors of HRQOL in this sample

(hypothesis 6).

Further, it was expected that the hypotheses 4-6 delineated above regarding socio-

environmental predictors of HRQOL would be replicated in Sample B (hypothesis 7).

Specifically, in Sample B, it was expected that individual and contextual social-

environmental predictors measured when youth are ages 15-16 would prospectively

predict HRQOL (in youth ages 16-17), across informants and above and beyond

demographic and illness-related covariates. It was again expected that individual social

adjustment variables would be most predictive of future HRQOL in this sample. Similar

to Sample A, individual and contextual social-environmental predictors included social

adjustment (e.g., social problems, social acceptance, peer support), family environment

(e.g., cohesion, conflict, stress), and community support. Demographic and illness-related

covariates included gender, SES, illness severity, cognitive ability and pain frequency.
 CHAPTER THREE

METHODS

Participants

Recruitment and methodological procedures were similar for both samples.

Participants  were  recruited  from  children’s  hospitals  and  a  statewide  SB association in

the Midwest using recruitment letters. Families were also approached and given

information about the study during regularly scheduled clinic visits. Interested

participants were screened by phone or in person by a member of the research team, and

were invited to participate if they met the following criteria: (1) diagnosis of SB,

including MM, lipomeningocele, and myelocystocele, (2) ages 8-15 (Sample A) or ages

8-9 (Sample B), (3) involvement of at least one parent, (4) cognitive ability to complete

questionnaires, and (5) residence within 300 miles of the laboratory to allow for home

visits. Criteria regarding language proficiency differed between samples; lack of

proficiency in English was an exclusionary criterion for Sample B, whereas participants

proficient in English or Spanish were included in Sample A.

Sample A. Participants in this sample were recruited starting in 2006 as part of a

larger NIH-funded longitudinal study examining neurocognitive, family, and social

domains in youth with SB (e.g., Devine et al., 2012). Of the 246 families approached,

42% (N = 104) could not be contacted or declined to participate and two families did not

meet inclusionary criteria, resulting in an initial sample size of 140 participating families.

Children of families who declined participation did not differ from those who accepted

24
 25

participation with respect to type of SB (myelomeningocele  vs.  other),  χ2 (1) = .000, p >

.05,  shunt  status,  χ2 (1) = .003, p > .05, or occurrence/nonoccurrence of shunt infections,

χ2 = 1.08, p > .05.

The current study included two data collection waves from the larger study: Time

1 (N = 133 youth ages 8-15 years old; M = 11.4; 55.6% female), and Time 2 (N = 101

youth ages 10-17 years old; M =13.3). Of the 133 participants at Time 1, 87.9% had a

diagnosis of myelomeningocele, 8.3% lipomeningocele, and 3.8% other. The majority of

the children had spinal lesions in the lumbosacral or lumbar spinal regions (62.9%),

whereas 19.0% and 18.1% had sacral and thoracic lesions, respectively. Additionally,

80.3% of the children had a shunt. With regard to ambulation methods, 81.1% of the

children used braces and 61.4% used a wheelchair.

Sample B. Participants in this sample were recruited starting in 1993 as part of a

separate, agency-funded (March of Dimes; e.g., Holmbeck et al., 2003) longitudinal

study that included a stronger focus than Sample A on family relationships and

psychosocial outcomes in children and adolescents with and without SB. This study has

included six data collection waves thus far; data were first collected when youth were 8-

and 9-years old, with subsequent data collection occurring every two years until youth

were 18- and 19-years old. Of the 310 families that were approached during Time 1, 39%

(N = 120) could not be reached or declined to participate, and 39% (N = 120) did not

meet inclusionary criteria, resulting in an initial sample size of 70 participating families.

Children of families who declined participation did not differ from those who accepted
 26

participations  with  respect  to  lesion  level,  χ2 (2) = 0.62, p > .05, or type of SB

(myelomeningocele  vs.  lipomeningocele),  χ2 (1) = 1.63, p > .05.

For Sample B, the current study utilized two data collection waves from the larger

study: Time 4 (N = 61 youth ages 14-16, M = 14.5; 41% female) and Time 5 (N = 53

youth ages 16-18, M = 16.6; 46.2% female). These data collection waves were chosen

due to the administration of a more reliable and validated HRQOL instrument (HRQLSB;

Parkin et al., 1997) starting at Time 5; a 6-item quality of life scale (Quality of Life in

Epilepsy scale (QOLIE) will be used to control Time 4 quality of life. Of the 61

participants still participating in the study at Time 4, 85.2% had a diagnosis of

myelomeningocele, 9.8% lipomeningocele, and 4.9% other. Additionally, over half of the

children had spinal lesions in the lumbosacral or lumbar spinal regions (54.1%), 19.7%

were sacral, and 8.1% thoracic. Further, 72.1% of the children had a shunt, 62.3% of the

children used braces to ambulate and 57.4 % used a wheelchair. Demographic

information of both samples is provided in Table 1. Additionally, there were no

differences in Time 1 levels of child-reported HRQOL between participants at Time 2

and non-responders in either sample (N of child non-responders in Sample A = 26; t(126)

= -0.33 and N of child non-responders in Sample B = 9; t(59) = -1.71, p’s  >  .05.

However, there was a significant difference in Time 1 parent-reported HRQOL in Sample

A, such that parent non-responders reported lower levels of HRQOL (N of non-

responders = 27; t(127) = -2.45, p < .05). For conceptual clarity related to describing two

independent samples, Time 4 and Time 5 in Sample B will hereafter be referred to as

Time 1 and Time 2, respectively.

 27

Table 1. Youth Demographic Information of Samples A and B at Time 1

Sample A Sample B
Characteristic n = 140 n = 61
Age M (SD) 11.40 (2.48) 14.50 (.60)
Gender (%)
Male 46.4% 59.0%
Female 53.6% 41.0%
Ethnicity N (%)
Caucasian 53.3% 70.9%
Hispanic 27.9% 7.3%
African American 12.9% 5.5%
Other 5.7% 16.3%
Hollingshead SES, M (SD)* 39.7 (15.9) 42.5 (11.0)
Note. Hollingshead SES are based on initial data from both samples (when children were 8-15 in Sample A
and 8/9 in Sample B).

Procedure

The following is a description of the general procedures used for both study

samples; unless otherwise noted, procedures were similar across the two samples. The

sponsoring  institution  and  hospitals’  Institutional  Review  Boards  approved  these  studies.

Trained graduate and undergraduate research assistants conducted three-hour home visits

at each data collection wave. Parental informed consent, child assent, and medical release

forms were obtained prior to data collection at each visit. Parents and youth completed

questionnaires, counterbalanced to control for order effects. Questionnaires included

measures of social adjustment, family environment, community support, and HRQOL.

Parents also completed a demographics and medical questionnaire.

In addition to completing study questionnaires, families from both samples

participated in counterbalanced, structured videotaped interaction tasks that differed

slightly according to study. Participants in Sample A completed: (1) a warm-up game (2)
 28

discussion of two age-appropriate vignettes, (3) discussion of transferring disease-

specific responsibilities to the child, and (4) a conflict task. These videotaped interactions

were coded using the macro-coding system developed by Holmbeck, Zebracki, Johnson,

Belvedere, and Hommeyer (2007a, 2007b). Interaction tasks that were coded to obtain

observational data on the family environment included: vignettes, transferring of

responsibilities, and conflict tasks. In the vignettes task, families were given two age-

appropriate vignettes of situations adolescents might typically encounter, and were asked

to discuss possible resolutions to these situations. In the transferring of responsibilities

task, families were asked to discuss one to two responsibilities that could be transferred

from the parent to the child (e.g., independent catheterization). Finally, in the conflict

task, families were first asked to complete a brief version of a measure called the Parent-

Adolescent Conflict Scale called the Issues Checklist (IC; Robin & Foster, 1989)

assessing the intensity of 20 common issues discussed between the parent and child

within the past two weeks, some of which were specific to SB. Families were then

presented with the five issues that they rated as most common and of highest intensity,

and they were allowed to discuss and attempt to resolve three or more of these issues

during the videotaped task. Families were given 10 minutes to complete each of these

tasks.

Participants in Sample B completed: (1) a warm-up game (2) an unfamiliar game

task (3) a structured family interaction task (SFIT; Ferreira, 1963), and (4) a conflict task,

based on a procedure developed by Smetana and colleagues (1991). Interaction tasks

coded in this sample included: the unfamiliar board game, family interaction, and conflict
 29

tasks, each of which was completed/discussed for a total of 10 minutes. In the unfamiliar

board game, families were asked to play a game they had not yet encountered. In the

Structured Family Task, families were first asked to complete a questionnaire

individually,  reflecting  five  commonly  discussed  family  issues  (e.g.,  “If  your  family  won  

a  contest,  what  would  you  want  the  prize  to  be?”).  Families  were  then  asked  to  engage  in  

a joint dialogue and make decisions pertaining to each of these five questions.

Administration of the conflict task was identical to that which was administered to

participants in Sample A. Families in both samples received monetary compensation for

completion of study procedures at each time point.

Measures

Unless otherwise noted, the following measures were assessed in both samples.

See Appendix A for all observational and questionnaire measures.

Demographics. Parents completed a questionnaire that assessed several

demographic factors. Demographic information regarding the parent included education,

employment status, and income. Information regarding the child included gender, age,

race, and ethnicity. The Hollingshead Four Factor Index for socioeconomic status was

computed  based  on  parents’  education  and occupation (Hollingshead, 1975), with higher

scores indicating higher SES.

Illness parameters

Neurocognitive functioning. The Wechsler Abbreviated Scale of Intelligence

(WASI; Wechsler, 1999) was used  as  a  measure  of  youths’  cognitive  ability  at  time  1  in  

Sample A. The WASI is a well-validated measure of child intelligence, with normative

 30

means of 100 and standard deviation of 15. In the current study, the WASI vocabulary

and matrix-reasoning subtests were utilized and an estimated full scale IQ (FSIQ) was

computed. The WASI vocabulary subtest is a 42-item  task  used  to  measure  child’s  

expressive vocabulary and verbal knowledge. The matrix-reasoning subtest was used to

measure nonverbal fluid reasoning and general intellectual ability, requiring subjects to

process and organize 34 visual patterns with shapes.

The Peabody Picture Vocabulary Test-Revised Edition (PPVT; Dunn & Dunn,

1981)  was  used  as  a  measure  of  youths’  cognitive  ability  at  time  1  in  Sample B. The

PPVT-R was administered by presenting each participant with a series of pictures. The

examiner then stated a word describing one of the pictures, and asked the participant to

point to the picture that described the word. The PPVT has shown high levels of validity

and reliability, correlating moderately with other measures of verbal intelligence (Sattler,

2002).

Illness severity. Parents filled out a medical form and data was collected from

medical charts to assess the following information: type of SB (i.e., myelomeningocele,

meningocele, or lipomeningocele), shunt status, lesion level (i.e., sacral, lumbar, or

thoracic) and ambulation method (i.e., ankle-foot orthoses [AFOs], knee-ankle-foot

orthoses [KAFOs] or hip-knee-ankle-foot orthoses [HKAGOs] wheelchair, or no

assistance). Based on previous research (e.g., Hommeyer, Holmbeck, Wills, & Coers,

1999), an overall illness severity composite score was computed according to a

participant’s  inclusion  in  a  specific  group  for  all  of  the  above  variables: shunt status (no =

1, yes = 2), myelomeningocele (no = 1, yes = 2), lesion level (sacral = 1, lumbar = 2,
 31

thoracic = 3), and ambulation status (no insistence/AFOs = 1, KAFOs/HKAFOs = 2,

wheelchair = 3). Thus, scores ranged from four to ten, with higher scores indicating

higher levels of severity.

Pain frequency. In Sample A, pain frequency was measured using a pain

questionnaire that assessed multiple aspects of pain experienced in the past three months

(e.g., location, severity, duration, frequency). In the current study, one item was utilized

to  assess  youths’  perception  of  pain  frequency,  rated  on  a  7-point scale (0 = not at all to 6

= daily). In Sample B, pain frequency was measured using one question from the somatic

subscale of the Child Behavior Checklist (CBCL; Achenbach, 1991). Youth reported on

the extent to which aches and pains without known medical cause had been present in the

past three months, rated on a 3-point scale (0 = not true, 1=somewhat or sometimes true

3= very true or often true).

Social-environmental predictors

Social adjustment. In accordance with research utilizing similar friendship

paradigms in youth with SB (Devine et al., 2011), child social adjustment was assessed

using three constructs: social competence, social skills, and peer social support.

Social competence was evaluated using mother and father report on the Social

Competence subscale from the Child Behavioral Checklist (CBCL; Achenbach, 1991).

The CBCL requires respondents to rate each of 119 behaviors on a 3-point likert scale (0

= not true, 1= somewhat/sometimes true, 2 = very or often true). The CBCL Social

Competence scale raw scores were computed by summing responses across the 9 items

contained in the subscale. This subscale consists of questions regarding a) participation in

 32

organizations, clubs, teams, or groups, b) number of close friends, c) amount of time

spent with friends outside of regular school hours, and d) behavior with others (i.e. how

well the child gets along with their brothers and sisters, other kids, their parents) and

behavior when alone (i.e., how well the child does things by themselves). Mean scores of

all available mother and father were utilized in analyses, with higher scores reflecting

greater social competence. Internal consistency for this measure is well established and

was demonstrated to be adequate in families of youth with SB (Holmbeck et al., 2003).

Social skills was evaluated using mother, father, and teach report of Social Skills

Rating System (SSRS; Gresham & Elliot, 1999). The SSRS is a standardized, norm-

referenced instrument that assesses behaviors that are considered essential to social

adjustment and adaptive functioning. On the parent and teacher forms, each item asks the

respondent to rate how often the child demonstrates a particular social skill (e.g., makes

friends easily, answers the phone appropriately, responds appropriately when hit or

pushed by children)  and  how  important  the  skill  is  to  the  child’s  development.  Teachers

and  parents  were  asked  to  rate  how  often  the  child  engages  in  the  behavior,  from  “0  =  

never”  to  “1  =  sometimes”  to  “2  =  very  often.”    The parent form consists of 38 items, the

teacher form consists of 34 items. The SSRS has shown adequate to good internal

consistency  across  forms  (α  =  .51;;  Gresham  &  Elliot,  1990).  In  the  current  study,  

coefficient alphas for the social skills subscales ranged from .91 to .92 for the teacher

form, and .88 to .94 for the parent form for both Samples A and B.

The Perceived Emotional/Personal Support Scale (PEPSS; Slavin, 1991) was used

to evaluate peer social support. This measure asks youth to nominate three individuals
 33

from each of the three categories: family members, non-family adults, and peers. The

current study utilized data on how respondents rate their peer relationships according to

three dimensions: (a) how much they spoke with their friend about personal concerns, (b)

how close they feel to their friend and (c) how satisfied they are with the support they

receive from their friend. A total score was computed by averaging items across these

three dimensions. The Peer Support subscale of the PEPSS has demonstrated adequate

reliability  and  consistency  (α=.89;;  Salvin  1991) and was adequate for the current study

samples (α=.89  for  both  Sample  A  and  Sample  B).

Family environment. Three domains of family environment were assessed in the

current study: family stress, conflict, and cohesion. All three constructs were assessed

using subjective measures. In addition, family conflict and cohesion were examined using

data derived from the observational coding systems.

The total score of the Family Inventory of Life Events (FILE; McCubbin,

Patterson, & Wilson, 1982) was used to assess family stress. This 71-item measure

examines family stress across several domains: intra-family strains, marital strains,

pregnancy and childbearing strains, finance and business strains, work-family transitions

and strains, illness and family care strains, losses, transition in and out, and family legal

violations. Mothers and fathers report whether or not the family has experienced the

event  and  total  scores  are  calculated  by  summing  all  “yes”  responses,  with  higher  scores  

indicating higher levels of family stress. The FILE has shown adequate internal

consistency (McCubbin, Patterson, & Wilson, 1982), and was adequate for the current

study samples (α’s  =  .84-.86).

 34

The construct family conflict and cohesion was assessed using both subjective

(questionnaire) and objective (observational) measures.

Questionnaire data. Parents and youth completed a shortened version of the

original 90-item Family Environment Scale (FES; Moos & Moos, 1994), which is a

common measure of social-environmental family characteristics. Total scores from the

family conflict and cohesion subscale scores of the Relationship Domain were used in the

analyses. The cohesion subscale of this measure assessed the degree of commitment,

help, and support family members provided for one another. The conflict subscale

evaluated the amount of openly expressed anger, aggression, and conflict between family

members. Studies using the FES have reported low to adequate reliability for parents of

youth with SB ages 8-11  years  (α’s  =  0.60  &  0.70;;  mother  and  father  report,  respectively;;  

Coakley, Holmbeck, Friedman, Greenley, & Thill, 2002). Alpha coefficients were

adequate for mother and father report for the current study samples (.76-.86).

Observational data. In Sample A, a manual-based macro-coding system

developed by Holmbeck, Zebracki, Johnson, Belvedere, and Hommeyer (Family

Interaction Macro-coding System Edition 2 or FIMS-II; Kaugers et al., 2011) was used to

obtain observational data for family conflict and cohesion. This coding system is an

adaptation of a system developed by Holmbeck, Belvedere, Gorey-Ferguson, &

Schneider (1995). In Sample B, the Family Interaction Macro-coding System Edition 1

(FIMS-I; Holmbeck et al., 1995) was used to evaluate these same constructs. Both of

these coding systems require trained undergraduate and graduate research assistants to

view each videotaped task and rate families according to interaction style, conflict, affect,
 35

control and problem-solving using and family systems/general family atmosphere using

5-point Likert scale ratings. Although coding domains overlapped, these two coding

systems differed slightly in item code content included under each domain. Generally, the

FIMS-I has fewer codes per domain. For example, under the interaction system domain,

the FIMS-I codes included (a) clarity of thought/idea expression, (b) listens to others, (c)

confidence in stating opinion, (d) requests input from other family members, (e) comfort

level during interaction, (f) involvement in the task, (g) receptive to statements made by

others, and (e) provides explanations for positions. The FIMS-II included all of these

codes  (with  the  exception  of    “comfort  level  during  interaction”),  as  well  as  six  additional  

codes: (1) off-task behavior (e.g., discussing topics unrelated to the task), (2) attunement

(i.e.,  the  level  which  family  members  are  “in  sync,”  (3)  mutuality  (i.e.,  the  degree  of  “we-

ness”  and  reciprocity  between  family  member  dyads),  (4)  positive  escalation  (i.e.,  

consecutive chains of positive behaviors between a dyad), (5) maturity, and (6) child is

needy.

Both of these coding systems tap into five key aspects of parenting and family

functioning: parental acceptance, behavioral control, psychological control, family

cohesion, and family conflict. For example, a coding item included in the cohesion

subscale  of  both  coding  systems  assessed  whether  “Parents  present  as  a  united  front”  

through verbal and non-verbal cues (5 = Always to 1 = Not at all). Utilizing these coding

methods, two coders rated each of the interaction tasks, and item level means of the raters

were averaged across tasks to yield a single score for each coding item across families in

both samples. Coding item mean scores was computed to create the final composite
 36

cohesion and conflict subscales: the same coding items comprised the cohesion (7 items)

and conflict (3 coding items) scales in both samples. Research has demonstrated adequate

scale-level rater reliability and internal consistency reliability of the FIMS-I conflict and

cohesion subscales in youth ages 14-15 with SB (α’s  =  0.78  – 0.84; Kaugers et al., 2011).

The current study indicated similar internal consistency reliability in youth ages 14-18

with SB in Sample A (α’s  =  0.91  for  cohesion  scale  and  0.61  for  conflict  scale)  and  

Sample  B  (α’s  =  0.83  for  cohesion  scale and 0.73 for conflict scale). Inter-rater reliability

was adequate for both the  conflict  and  cohesion  scales  in  both  Samples  (α’s  = .65-.85).

Community support. Community support was examined using the total score of

the 16-item Social and Community Support Questionnaire, which was modified from the

75-item ACCESS Needs Assessment for Parents Scale (Kennedy et al., 1998). This

measure assesses parental perception of SB-specific community resources available to

their child. More specifically, parents are asked to endorse items that are important to

them (e.g., adequate health insurance, wheelchair accessibility), and then rate the extent

to which this resource is available to the family using a 5-point likert scale (1 = Not taken

care of at all to 5 = Well taken care of), with higher scores indicating better community

support. In the current study, internal consistency was adequate for both samples (α’s  =  

0.81 – 0.87).

Health-related quality of life

Health-related quality of life. In Sample A,  youths’  HRQOL was assessed by

self-report and mother- and father-proxy report using the Pediatric Quality of Life Scale

(PedsQL™ 4.0;Generic Core Scale) which has well-established reliability and validity in
 37

children with both acute and chronic health conditions. The PedsQL yields a 15-item

psychosocial total score as well as four subscale scores to  assess  a  child’s physical,

emotional, social, and school functioning. The physical subscale of the PedsQL contains

questions that implicate physical function, and the majority of children in the sample

have physical impairments that may lead to a measured reduction in HRQOL. Thus, the

psychosocial total score, which includes emotional, social and school functioning

subscales, was utilized in regression analyses to capture quality of life beyond simple

impairment.

Children and parents were asked how much of a problem each item has been over

the past month using a 5-point Likert scale rating (0 = never a problem to 4 = almost

always a problem). Raw scores were then transformed into standard scores ranging from

0 to 100, with higher scores indicating better HRQOL. In the current study, internal

consistency was adequate (α’s  =  0.83 – 0.90) for the PedsQL subscales (physical,

emotional, social, and school) as well as the psychosocial total scores at both time points.

In Sample B, youth were administered a health-related quality of life instrument

specific to SB (HRQOL-SB; Parkin, et al., 1997). The HRQOL-SB is a 47-item measure

is that taps ten domains: (a) social, (b) emotional, (c) intellectual, (d) financial, (e)

medical, (f) independence, (g) environmental, (h) physical, (i) recreational, and (j)

vocational quality of life. Youth were asked how much they feel (e.g., that you have a lot

of pain; happy with yourself; etc.) about all of these topic areas using a 5-point Likert

scale (1 = only little to 5 = a lot). The HRQOL-SB has demonstrated adequate internal

consistency and construct reliability (Parkin, et al., 1997). In addition, the total score on
 38

Quality of Life in Epilepsy scale (QOLIE; Devinsky et al., 1995; Sample B) were used to

control for Time 1 HRQOL. In the current sample, internal consistency was adequate for

the QOLIE and HRQOL-SB (α’s  =  0.89  and  0.95, respectively).

Finally, it is important to note that the social adjustment measures listed above

were chosen in order to reduce item overlap in the Social Domains of the PedsQL and

HRQOL-SB. Item overlap may become an issue because statistical analyses containing

similar or overlapping measured independent and dependent variables can cause inflation

or overestimation of study results. The social domains of these measures ask if the child

has problems with 1) getting along with others, 2) other children not wanting to be their

friend, 3) kids teasing them, 4) doing things other kids can do, and 5) keeping up when

they play with other kids due to the severity of their condition. These items were

determined to be different from items within all of the Social Adjustment independent

variables. In particular, items within the CBCL Social Competence subscale pertained to

participation and involvement in social activities and general social behavior (e.g.,

participation in clubs, behavior towards others and while alone). Items within the Social

Skills Rating System (SSRS) tapped into several social skills, including appropriate

reactions to and interactions with adults and peers. Finally, the Perceived Emotional

Support  Scale  (PESP)  contained  items  that  described  the  child’s  satisfaction with and

quality of their current peer support. Finally, the PedsQL is a multidimensional

questionnaire, and thus any issues regarding item overlap may be not be particularly

salient.
 39

Statistical Treatment

Preliminary analyses. Prior to hypothesis testing, preliminary analyses

determined the psychometric properties of all measures. Analyses also determined

whether variables are skewed or contained outliers. Hierarchical regression analyses were

used to determine the utility of prospective social-environmental predictor variables in

explaining the variance in Time 2 total HRQOL. Assuming a power of .80, and an alpha

of .05, a sample of 50 is required to detect large effect sizes (R2 = .35) and a sample size

of 107 is required to detect medium effect sizes (R2 = .15) for analyses with 8 predictors

and a single dependent variable (Cohen, 1992). Thus, the current study had enough power

to detect medium to large effect sizes in Sample A, and large effect sizes in Sample B.

Given the relatively large number of potential covariates and predictors in the

study model, preliminary analyses examined correlations among all covariates and

predictors of youth- and parent- reported HRQOL to determine which variables would be

most appropriate for inclusion in subsequent regression analyses. Only variables that

were significantly (p < .05) related to HRQOL were included in regression analyses, and

no more than eight predictors were used in each regression model. Only covariates that

had p-values of .10 or more were included in regression analyses. In order to reduce the

number of potential analyses, mother- and father- reports  of  their  youths’  HRQOL,  social  

problems, social acceptance, as well as family environment and community support were

combined if significantly correlated. Youth and parent report of HRQOL were analyzed

separately in an effort to reduce the effects of common method variance.

 40

HRQOL descriptive information. It was expected that child and caregiver reports

of HRQOL in Sample A would display low to moderate agreement (hypothesis 1).

Agreement between child and self-report of caregiver proxy measures was addressed

using bivariate correlations and paired-sample t-tests. It was also expected that youth

with SB in Sample A would display impaired HRQOL across all domains (i.e., physical,

emotional, social, school, and total HRQOL) compared to healthy youth, and would

display similar or lower mean scores compared to youth with chronic illnesses (Varni,

Seid, & Kurtin, 2001; hypothesis 2). Mean HRQOL subscale and total scores were

compared to normed references groups of healthy and illness populations that have been

published in previous research (Varni et al., 2001) using independent samples t-tests.

Additionally, standard deviations were compared to provide further information on

clinically significant differences in HRQOL. Finally, it was expected that HRQOL would

decrease over time in Sample A (hypothesis 3). This hypothesis was tested using repeated

measure ANOVAS. Specifically, youth and caregiver report of HRQOL was examined

to determine whether there was a decrease in HRQOL from Time 1 (ages 8-15) to Time 2

(ages 10 to 17).

Social-environmental predictors of HRQOL. For Sample A, it was expected that

better social acceptance, peer support, family cohesion, and community support at Time 1

would predict improved HRQOL at Time 2, and increased social problems, family

conflict, and family stress at Time 1 would predict reduced HRQOL at Time 2, across

both informants (i.e., youth and parent report of HRQOL; hypothesis 4). It was also

expected that social-environmental factors would explain the variance in Time 2 HRQOL
 41

beyond relevant demographic and illness-related variables, including age, gender, SES,

cognitive ability, illness severity and pain frequency (hypothesis 5). To test these

hypotheses, separate analyses were conducted for each reporter of HRQOL (youth or

parent). Variables were entered in the following steps: (1) Time 1 HRQOL total score, (2)

Time 1 demographics and illness-related covariates, (3) Time 1 social adjustment

predictors (i.e., social problems, social acceptance, peer support), (4) Time 1 family

environmental predictors (i.e., family stress, family conflict, and family cohesion), and

(5) Time 1 community support. Within each step, variables were entered simultaneously.

Finally, it was expected that variables within the social adjustment domain (i.e., social

problems, social acceptance, and peer support) would be most predictive of HRQOL at

Time 2 in this sample (hypothesis 6). To test this hypothesis, R2 –change values were

compared at each step in the hierarchical regression models to compare the relative

contribution of social adjustment to all other domains/steps.

Finally, to provide additional support for study hypotheses, it was expected that

hypotheses 4-6 delineated above would be replicated in Sample B. Specifically, in

Sample B, it was expected that individual and contextual social-environmental predictors

measured at Time 1 would predict Time 2 youth-reported HRQOL above and beyond

demographic and illness-related covariates. Individual and contextual social-

environmental predictors included social adjustment (i.e., social problems, social

acceptance, peer support), family environment (i.e., family stress, observed or perceived

family conflict, and family cohesion), and community support. Similar to hypothesis 6, it

was expected that individual social adjustment variables would be most predictive of
 42

future HRQOL in this sample. To test this hypothesis, statistical procedures run on

Sample A were replicated.

 CHAPTER FOUR

RESULTS

Preliminary Analyses

All independent and dependent variables were tested for skewness. Results

indicated that across respondents, reports of community support, family conflict, family

cohesion, social competence, social skills, peer support (at Time 2) and heath-related

quality of life (at Time 1 and Time 2) were not highly skewed for either sample

(Skewness values = 0.85 to -0.84 Sample A and 1.68 to -1.41 in Sample B). Thus, it was

not necessary to perform variable transformations prior to analyses.

Preliminary analyses included an examination of the degree of relationship across

reports for variables in which there were multiple responders. These correlational

analyses were used to determine whether reports could be combined to reduce the

number of longitudinal analyses. Mother and father reports of community support, family

conflict, family cohesion, social competence, and social skills were significantly

correlated at Time 2 for both samples (r’s  =  0.31 - 0.63, M = 0.42, p’s  <  .05). In addition,

mother and father reports of social skills were significantly correlated in Sample A and B

(r’s  =  0.56 and 0.39, respectively, p’s  <  .05). Both mother and father report of social

skills were also significantly correlated with teacher report of social skills in Sample A

(r’s  =  0.21 and 0.23, respectively, p’s < .05), but mother report of social skills was not

significantly correlated with teacher report in Sample B, r = 0.05, p > .05. Thus, mother

and father reports of community support, family conflict, family cohesion, and social

43
 44

competence were combined for both samples. Further, a social skills composite score of

all three reporters was created for Sample A, but teacher report was not combined with

parent report of social skills in Sample B. Finally, mother and father reports of HRQOL

were significantly correlated at both Time 1 and Time 2 in Sample A (r’s  =  0.54 and

0.62, p’s  <  .05 respectively). Thus, mother and father report of HRQOL at Time 1 and

Time 2 were combined for Sample A; this was not necessary for Sample B as only youth

report of HRQOL was collected.

Descriptive Information on Health-Related Quality of Life

Hypothesis 1. Descriptive statistics on the PedsQL in Sample A are summarized

in Table 2. As hypothesized, there were moderate levels of agreement between youth and

caregiver reports on HRQOL in Sample A. Specifically, cross-informant correlations for

the PedsQL psychosocial total score were 0.24 and 0.42 for Time 1 and Time 2,

respectively; physical functioning cross-informant correlations were 0.31 and 0.41,

respectively; emotional functioning cross-informant correlations were .13 and .40,

respectively; and school functioning cross-informant correlations were 0.34 and 0.39,

respectively. The only nonsignificant correlation was parent-proxy and child-reported

emotional functioning at Time 1 (r = .13 p > .05). Further, there were no significant

differences between youth and parent report of Time 1 psychosocial HRQOL [t(120) =

0.98, p > .05], and Time 1 and Time 2 emotional functioning [t(120) = 1.10, p > .05;

t(102) = 0.76, p > .05, respectively], and school functioning [t(120) = 0.53, p > .05;

t(102) = 0.76, p > .05]. However, there were significant differences between youth and

parent report of Time 2 psychosocial HRQOL, [t(102) = 2.15, p < .05], and Time 1 and
Table 2. Scale Descriptives, Cross-Informant Correlations, and Differences on PedsQL Generic Score: Sample A

Cross-
Child Caregiver informant Difference
Variable N (Child/caregiver) M (SD) M (SD) r t
Total psychosocial score
Time 1 129 / 128 62.53 (16.81) 60.89 (12.70) .24* 0.98
Time 2 106 / 105 68.10 (15.70) 64.42 (13.60) .42** 2.15*

Physical functioning
Time 1 129 / 128 58.52 (21.02) 52.83 (20.12) .31** 2.72*
Time 2 106 / 105 61.06 (20.60) 51.96 (20.07) .41** 4.20**

Emotional functioning
Time 1 129 / 128 64.22 (20.72) 66.15 (14.86) .18* 1.10
Time 2 106 / 105 69.53 (19.13) 70.48 (15.49) .30** 0.76

Social functioning
Time 1 129 / 128 66.47 (22.16) 58.89 (15.52) .13 3.48**
Time 2 106 / 105 72.69 (19.20) 62.62 (18.86) .40** 4.89**

School functioning
Time 1 129 / 128 56.90 (22.13) 57.38 (17.56) .34** 0.53
Time 2 106 / 105 62.08 (20.56) 60.31 (16.81) .39** 0.76
*p < .05. **p < .01.

45
 46

Time 2 physical HRQOL [t(120) = 4.20, p < .01; t(102) = 2.15, p < .05, respectively] and

social HRQOL [t(120) = 3.48, p < .01; t(102) = 4.89, p < .01, respectively]. More

specifically, parent proxy scores of HRQOL were significantly lower than youth report of

psychosocial HRQOL at Time 1 and physical and social HRQOL for both time points

(see Table 2), such that parents reported youth to have significantly worse HRQOL

compared to youth report. Thus, hypothesis 1 was supported such that there was

moderate agreement between child-self report and caregiver proxy report of HRQOL In

general, it appears that parents rated similar or significantly lower HRQOL compared to

youth self-report.

Hypothesis 2. It was also expected that youth with SB in Sample A would

display impaired HRQOL across all domains (i.e., physical, emotional, social, school,

and total HRQOL) compared to mean scores of healthy populations of youth, and would

display similar to lower mean scores compared to other illness groups. According to

youth-report at Time 1 and Time 2 of the PedsQL, youth with SB reported significantly

lower total psychosocial HRQOL, as well as physical, emotional, social, and school

HRQOL compared to healthy as well as chronically ill samples as reported in the study

by Varni and colleagues (2001; p’s  <  .001, see Table 3). Similar findings were found for

parent-report at Time 1 and Time 2 of the PedsQL, such that parent-reported quality of

life was significantly lower than total psychosocial HRQOL, as well as physical,

emotional, social and school HRQOL compared to healthy and chronically ill samples as

reported in Varni and colleagues (2001, p’s  < .001, see Table 4). Thus, hypothesis 2 was
Table 3. One-Sample t Tests Comparing Time 1 and 2 HRQOL: Spina Bifida Versus Chronically Ill and Healthy
Populations: Child Report

Scale N (T1/T2) M (T1/T2) SD (T1/T2) T (T1/T2)

Psychosocial total
Spina bifida 129 / 106 62.53 / 68.10 16.81 / 15.70
Chronically ill 367 77.10 15.84 −9.84**  /  −5.90**
Healthy 399 82.38 15.51 −13.40**  /  −9.37**

Physical
Spina bifida 129 / 106 58.52 / 61.06 21.02 / 20.60
vs. Chronically ill 366 77.36 20.36 −10.17**  /  −8.16**
vs. healthy 400 84.41 17.26 −13.98**  /  −11.69**

Emotional
Spina bifida 129 / 106 64.22 / 69.53 20.72 / 19.13
vs. Chronically ill 366 76.40 21.48 −6.67**  /  −3.70**
vs. healthy 400 80.86 19.64 −9.12**  /  −6.10**

Social
Spina bifida 129 / 106 66.47 / 72.69 22.16 / 19.20
vs. Chronically ill 367 81.60 20.24 −7.75**  /  −4.78**
vs. healthy 399 87.42 17.18 −10.74**  /  −7.90**

School
Spina bifida 129 / 106 56.90 / 62.08 22.13 / 20.56
vs. Chronically ill 362 73.43 19.57 −7.30**  /  −5.69**
vs. healthy 386 78.63 20.53 −9.97**  /  −8.29**
**p < .00.

47
Table 4. One-Sample t Tests Comparing Time 1 and 2 HRQOL: Spina Bifida Versus Chronically Ill and Healthy
Populations: Parent Report

Scale N (T1/T2) M (T1/T2) SD (T1/T2) T (T1/T2)

Psychosocial total
Spina bifida 128 / 105 60.89 / 64.42 12.70 / 13.60
Chronically ill 367 77.10 15.84 −14.48**/  −9.55**
Healthy 399 82.38 15.51 −19.20**/  −13.50**

Physical
Spina bifida 128 / 105 52.83 / 51.96 20.12 / 20.07
vs. Chronically ill 366 77.36 20.36 −13.79**  /  −8.16**
vs. Healthy 400 84.41 17.26 −17.75**  /  −11.69**

Emotional
Spina bifida 128 / 105 66.15 / 70.48 14.86 / 15.49
vs. Chronically ill 366 76.40 21.48 −7.80**/  −3.92*
vs. Healthy 400 80.86 19.64 −11.19**/  −6.87**

Social
Spina bifida 128 / 105 58.89 / 62.62 15.52 / 18.86
vs. Chronically ill 367 81.60 20.24 −16.55**/  −10.31**
vs. Healthy 399 87.42 17.18 −20.79**/  −13.47**

School
Spina bifida 128 / 105 57.38 / 60.31 17.56 / 16.81
vs. Chronically ill 362 73.43 19.57 −10.33**  /  −8.00**
vs. Healthy 386 78.63 20.53 −13.68**  /  −11.17**
**p < .001.

48
 49

supported, such that youth with SB have significantly impaired HRQOL compared to

both healthy and chronic illness groups.

Further, according to youth-report of Time 1 HRQOL, mean scores were at least

one SD lower than those found in healthy youth in Varni et al. (2001) on HRQOL scores,

with the exception of emotional HRQOL. Of particular note, children were 1.31 SDs

below the healthy population mean on total psychosocial HRQOL, 1.50 SDs below the

population mean on physical HRQOL, 1.22 SDs below the population mean on social

functioning, an 1.06 SDs below the population mean on school functioning. Youth report

of emotional HRQOL was only 0.85 SDs below the sample mean as reported by Varni

and colleagues (2001).

Caregiver report of HRQOL at Time 1 yielded slightly more robust findings for

total, physical and social functioning; results indicated that youth with SB were 2.01 SDs

below the healthy population mean on total psychosocial HRQOL, 2.23 SDs below the

population mean on physical HRQOL, 2.30 SDs below the population mean on social

functioning, and 1.60 SDs below the population mean on school functioning. Similar to

findings for youth-report at Time 1, parent-proxy on emotional HRQOL was 0.94 SD

below the mean healthy population score.

Time 2 comparison analyses yielded different findings for youth at Time 2. Youth

and parent-proxy report reported the following results: child and parent reports were 0.91

and 1.73 SDs below the population mean on total psychosocial HRQOL, respectively;

1.35 and 1.73 SDs below the population mean on physical HRQOL, respectively; 0.58

and 2.29 SDs below the population mean on emotional HRQOL, respectively; 0.86 and
 50

0.63 SDs below the population mean on social HRQOL, respectively; and 0.81 and 1.43

SDs below population mean on school HRQOL, respectively. Thus, data at Time 2

indicated that findings were less robust. One exception was parent-report of emotional

HRQOL, which was 1.35 SDs higher compared to parent report at Time 1. Further, this

data provide evidence that physical HRQOL was a consistently poor area of functioning

in this population across informants and at both time points.

Hypothesis 3. Repeated-measures ANOVAS tested the hypothesis that total

psychosocial HRQOL would decrease over time in Sample A (hypothesis 3). This

Hypothesis was not supported. In fact, results indicated that there was a significant

increase in youth-reported psychosocial HRQOL from Time 1 to Time 2 for Sample A,

F(1, 99) = 9.46, p < .01. Results also indicated that there was a marginally significant

increase in parent-reported psychosocial HRQOL from Time 1 to Time 2 in this same

sample, F(1, 101) = 3.11, p = .08.

Longitudinal Analyses of Health-Related Quality of Life in Sample A

Hypotheses 4-6. In Sample A, it was proposed that all social-environmental

factors measured when youth were ages 8-15 would prospectively predict HRQOL (in

youth ages 10-17) across both informants (e.g., child and parent; hypothesis 4). It was

also expected that social-environmental predictors of HRQOL would be significant above

and beyond all demographic and illness-related variables (hypothesis 5). Finally, it was

expected that social adjustment variables (e.g., social problems, social acceptance, and

peer support) would be the strongest prospective predictors of HRQOL in this sample

(hypothesis 6).
 51

First, two-tailed Pearson correlations were calculated between all covariates,

independent variables, and measures of HRQOL across informants (see Table 5).

Predictors that were not significantly correlated with the outcome were not entered into

the regression model for that outcome. Only two variables were significantly correlated

with Time 2 youth-reported psychosocial HRQOL in Sample A: child pain intensity and

parent-reported social competence. In other words, child pain intensity when youth were

ages 8-15 (Time 1) was negatively associated with child report of HRQOL when youth

were ages 10-17 (Time 2), such that higher pain intensity at Time 1 was associated with

lower HRQOL at Time 2, r = -.29, p < .05. In addition, parent-reported social

competence at Time 1 was positively associated with child-reported HRQOL at Time 2,

such that lower social competence was associated with reduced HRQOL, r = .31, p < .05.

The following variables were not significantly correlated with Time 2 youth-reported

psychosocial HRQOL in Sample A: (a) covariates: child age, socioeconomic status,

gender, child IQ; (b) social-environmental predictors: parent-reported community

support, family stress, family conflict, family cohesion, social skills, child-reported peer

social support, as well as observed family conflict and cohesion (p’s  >  .05;;  see  Table  5).

Only two social-environmental variables were significantly correlated with Time

2 parent-reported psychosocial HRQOL in Sample A: parent-reported community

support and the composite score (i.e., mother, father, and teacher) of social skills. In other

words, higher levels of community support (r = .23, p < .05) and better social skills (r =

.21, p < .05) were associated with parent-proxy report of HRQOL. The following

variables were not significantly correlated with Time 2 parent-reported psychosocial

Table 5. Correlation Matrix for Socio-Environmental Predictors of Time 2 Health-Related Quality of Life: Sample A
Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1.Time 2 QOL- Child — .41* −.02 −.01 −.02 .01 −.16 −.29* .08 .09 −.01 −.11 −.07 .03 .31* .17 −.16
2. Time 2 QOL- Parent — −.01 .08 −.12 −.02 −.18 −.04 .23* −.05 −.14 −.09 .07 .04 .11 .21* −.05
3. Child age — −.08 −.08 −.23* .22 .05 −.01 −.09 −.01 −.06 .02 −.14 −.08 .08 .12
4. SES — −.10 .47* −.14 −.11 .12 .18* .05 −.18* .08 .34* .33* .09 −.06
5. Gender — .05 −.06 −.10 −.11 .02 .02 .03 −.02 .00 −.05 .08 .20*
6. Child IQ — −.31* −.07 −.04 .15 .10 −.15 .05 .33* .44 .23* .07
7. Illness severity — .30* .05 −.03 −.03 −.12 −.03 .00 −.19* −.06 .06
8. Child pain intensity — .05 −.13 −.15                                                                                                                                                                                                                                                  
.02 .43 −.08 −.22* −.04 .10
9. Community support — −.18* −.16 −.11 .24 .05 .14 .17 −.01
10. Family Stress — .32* .13 −.30* .04 .10 .01 −.08
11. Family Conflict-Qx — .10 −.60* −.08 −.06 −.32* −.09
12. Family Conflict-Mac — .09 −.47* −.17 −.21* −.09
13. Family Cohesion-Qx — .15 .03 .27 .01
14.Family Cohesion-Mac — .31* .36* .01
15. Social Competence — .41* .12
16. Social Skills — .11
17. Peer Social Support —

*p < .05.

52
 53

HRQOL in Sample A: (a) covariates: child age, socioeconomic status, gender, child IQ,

illness severity, child pain intensity; (b) social-environmental predictors: parent-reported

family stress, conflict, cohesion, social skills, child-reported peer social support, as well

as observed family conflict and cohesion. Thus, results were similar to correlational

analyses of Time 2 youth-reported HRQOL (see Table 5).

Subsequently, longitudinal hierarchical regression analyses were conducted to

determine whether socio-environmental variables were related to subsequent change in

HRQOL over time. Separate regression analyses were conducted for child-reported

HRQOL and parent proxy-report of HRQOL. To control for previous quality of life,

HRQOL at Time 1 was entered in the first step. Time 1 covariates obtained that were

moderately correlated with Time 2 HRQOL (i.e., p = .10 or lower) were entered in the

second step. Finally, socio-environmental variables were entered in the remaining steps.

For the first hierarchical regression analysis predicting child-report of HRQOL at Time 2,

Time 1 HRQOL was entered in the first step, illness severity (r = -.16, p = .10) and child

pain intensity were entered in the second step, and parent-report of social competence

was entered in the last step.

Hypotheses 4-6 were partially supported according to this analysis. Although

several of the proposed covariates and social-environmental predictors were not

associated with Time 2 HRQOL, results indicated that social competence predicted child

report of HRQOL over time in Sample A, above and beyond illness related variables, β =

0.23, p < .05. However, illness severity and child pain intensity did not significantly

predict HRQOL, β’s  = -0.04 and -0.08, respectively, p’s > .05. Social competence
 54

accounted for 4.0% of the variance in child-reported HRQOL at Time 2 above baseline

levels. The final model accounted for 28% of variance in youth-reported HRQOL at Time

2, R2 = .28, adjusted R2 = .24, F(4, 80) = 7.77, p < .05. However, longitudinal analyses of

parent-reported quality of life yielded different results, such that none of the covariates or

socio-environmental factors predicted HRQOL in youth with SB (see Table 6).

Longitudinal Analyses of Health-Related Quality of Life in Sample B

Hypothesis 7. Finally, it was expected that hypotheses 4-6 delineated for Sample

A regarding socio-environmental predictors of HRQOL would be replicated in Sample B.

Again, two-tailed Pearson correlations were calculated between all covariates,

independent variables, and measures of HRQOL across informants in this sample. Only

two variables were significantly correlated with Time 2 youth-reported psychosocial

HRQOL in Sample A: socioeconomic status and teacher-report  of  youths’  social  skills.  

Socio-economic status was positively associated with child report of HRQOL at Time 2

(youth ages 16-17), such that higher SES was associated with better HRQOL at Time 2, r

= .40, p < .05. In addition, teacher-report of social skills was positively correlated with

HRQOL at Time 2, such that better social skills was associated with better HRQOL, r =

.31, p < .05. The following covariates and socio-environmental variables were not

significantly associated with Time 2 youth-reported HRQOL: (a) covariates: child age,

gender, IQ, illness severity, pain frequency (b) social-environmental variables: parent-

reported community support, family stress, family conflict, family cohesion, social

competence and social skills, child-reported peer social support, and observed family

conflict and cohesion (see Table 7).

Table 6. Hierarchical Regressions: Time 1 Socio-Environmental Predictors of Time 2 Psychosocial HRQOL

Quality of life outcome Covariates and predictors b SE b β p

Sample A
Time 2 HRQOL- Child Time 1 PedsQL- Child 0.44 0.10 0.46** 0.00
Illness severity −0.35 1.03 −0.04 0.73
Child pain intensity −0.86 0.50 −0.18 0.08
T1 Social Competence (M) 0.40 0.19 0.23* 0.03

Time 2 HRQOL- Parent Time 1 PedsQL- Parent 0.74 0.09 0.64** 0.00
Gender −4.17 2.20 −0.15 0.06
Illness Severity −0.42 0.77 −0.05 0.59
Community Support (MF) 1.47 1.44 0.08 0.31
Family Stress (MF) −0.20 0.15 −0.10 0.20
Social Skills (MF) 7.01 4.76 0.12 0.14
Sample B
Time 2 HRQOL- Child Time 1 HRQOL-SB – Child 0.05 0.06 0.13 0.42
Socioeconomic Status 0.02 0.01 0.42* 0.01
Social Skills (TR) 0.51 0.22 0.33* 0.02
Note. MF = mother and father report; TR = teacher report.

*p < .05; **p < .01.

55
Table 7. Correlation Matrix for Socio-Environmental Variables and Time 2 Health-Related Quality of Life: Sample B

Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

1. Time 2 QOL- Child — −.02 .40* .02 −.12 −.05 .02 .07 −.02 .20 .10 −.24 .02 .20 .21 .31* .01
2. Child age — −.10 −.09 −.17 −.05 −.24 .11 .10 .03 .19 −.07 −.20 −.06 .01 −.18 −.07
3. SES — .04 .29* .02 −.15 −.10 .07 −.17 −.16 .22 .33* .28* .46* .06 .24
4. Gender — −.09 .02 .06 .12 .13 −.09 −.19 .01 .09 .07 .06 −.19 .29*
5. Child IQ — −.19 −.04 −.22 .26* −.45* .12 .23 .26 .37* .23 .16 .37*
6. Illness severity — −.05 .45* −.03 −.12 −.10 .10 −.10 −.13 −.06 −.18 −.20
7. Child pain frequency — −.05 −.15 .03 −.08 .13 .05 −.03 −.20 .03 .01
8. Community support — −.01 .10 −.10 .20 −.20 −.05 −.17 −.08 .07
9. Family Stress — .15 .12 −.32* .00 .08 −.08 −.15 .12
10. Family Conflict -Qx — .42* −.42* −.48 −.25 −.24 .12 −.36*
11. Family Conflict-Mac — .04 −.49* −.13 −.28* .04 −.10
12. Family Cohesion-Qx — −.16 −2.6 −.07 −.03 .16
13. Family Cohesion-Mac — −.13 .27* .30* −.11
14. Social Competence — .63* .26 .18
15. Social Skills-MF — .18 .16
16. Social Skills-T — .02
17. Peer Social Support —

Note. Child report. Combined mother and father report.

*p < .05.

56
 57

Again, a longitudinal hierarchical regression analysis was conducted to determine

whether teacher-report of social skills predicted subsequent change in HRQOL over time.

Analysis for Sample B was similar to that which was conducted for Sample A: Time 1

HRQOL was entered in the first step, SES was entered in the second step, and teacher-

report of social skills was entered in the last step. Results indicated that socioeconomic

predicted HRQOL at Time 2 in this sample, β = 0.33, p < .05. Also, similar to results

found in Sample A, analyses in Sample B indicated that social skills predicted child-

report of HRQOL over time, above and beyond socioeconomic status, β = 0.33, p < .05.

Socioeconomic status explained 15.1% of the variance in Time 2 HRQOL above baseline

levels; teacher-reported social skills accounted for 10.7% of the variance in Time 2

HRQOL above baseline levels. The final model accounted for 27% of variance in youth-

reported HRQOL at Time 2, R2 = .27, adjusted R2 = .22, F(3, 38) = 4.78, p < .05. Thus,

hypothesis 7 was partially supported, as variables within the social domain (see Figure 1)

were found to be the most important prospective predictors of child-reported HRQOL

across samples (i.e., social competence in Sample A; social skills in Sample B; see Table

6).
 CHAPTER FIVE

DISCUSSION

The present study examined descriptive data as well as prospective predictors of

overall psychosocial HRQOL in youth with SB. Consistent with previous research on

several pediatric populations, such as cancer, sickle cell disease, obese and chronic pain

patients (Varni, Limbers, & Burwinkle, 2007; Palermo, Schwartz, Drotar, & McGowan,

2002; Schwimmer, Burwinkle, Varni, 2003; Modi & Quittner, 2003; Hunfeld et al.,

2001), results indicated that children and adolescents with SB are at-risk for poor

HRQOL. At both time points, scores on HRQOL across all domains (i.e., physical,

emotional, social, and school HRQOL) were significantly lower than those reported by

Varni and colleages (2001) in populations of children with other chronic health

conditions, and scores were also well below the normative scores for healthy children.

Several scholars have proposed that patient self-report should be considered the

standard for measuring perceptions of HRQOL (Riley, 2004; Sawin & Bellin, 2010).

These researchers assert that the adolescent should serve as the primary informant,

because HRQOL includes subjective appraisal of life contexts, events, and experiences.

When the youth is unable to report on HRQOL due to physical or cognitive limitations,

seeking parent-proxy report has been supported (Haas, 1999). Although moderate cross-

informant correlations of child and parent HRQOL were found in the current study, the

mild to moderate cognitive limitations associated with this condition (Fletcher & Brei,

2010) may highlight the need to gather data from both child and parent perspectives.

58
 59

Thus, due to potential differences in parental observations of functioning versus the

child’s  reliance  on  internal  cues,  as  well  as  the  cognitive  impairments  associated with SB,

future studies should follow the guidance provided by several researchers (e.g., Modi &

Quittner, 2003; Palermo et al., 2008) in including both youth and caregiver proxy report

of HRQOL.

Moreover, significant discrepancies were found between youth and caregiver-

proxy report of HRQOL. Specifically, child and parents reported significantly different

social and physical HRQOL at Time 1 and 2, such that caregivers indicated their child

had lower HRQOL compared to youth self-report. There may be several explanations for

this finding. First, caregivers may have relied solely on behavioral and visual cues to

assess  their  child’s  functioning  (Varni,  Seid,  &  Kurtin,  2001),  while  the  child  may  have  

relied on internal emotional cues to assess of their level of functioning relative to other

youth (Gold, Mahrer, Yee & Palermo, 2009). Second, the cognitive limitations associated

with  spina  bifida  may  result  in  youths’  lowered  awareness  of  the  physical  and  social  

consequences of their illness, and again caregivers may have a more complete

understanding of the functioning of their child as it is understood by behavioral

observation. Finally, youth with SB may have adapted to and accepted their chronic

condition while their caregivers may have not (Berrin, Malcarne, Varni, et al). Overall,

research on cross-informant discrepancies is inconsistent. Some research has indicated

that children report more physical complaints and problems with motor functioning

(Theunissen et al., 1998) as well as greater emotional distress (Modi & Quittner, 2003;

Verrips, Vogels, den Ouden, Paneth, & Verloove-Vanhorick, 2000). However, in line
 60

with the current study, other studies have indicated that youth report better HRQOL

compared to parent-proxy report (e.g., cerebal palsy, cystic fibrosis & chronic pain;

Berrin, Malcarne, Varni, et al., Britto, Kotagal, & Chenier et al., Gold, Mahrer, Yee &

Palermo, 2009).

Although contrary to previous research and study hypotheses, the  current  study’s  

results indicated that youth-report of HRQOL significantly increased over time, while

caregiver-report of HRQOL remained stable. Such findings may be explained by several

factors. First, HRQOL data were based on group means, yet individual trajectories may

vary; some patients may improve while others decline over time. Individual HRQOL can

yield patient-specific and clinically-relevant information for healthcare providers.

Second, this finding may point to a resiliency factor in families of youth with SB as

indicated in previous research (Holmbeck et al., 2002), such that youth with SB may be

better able to adapt to and accept their condition compared to other illness populations.

Despite this resiliency, it should be noted that HRQOL was significantly lower than

population means for chronically ill and healthy youth at both time points. Thus, although

HRQOL may increase over time, youth with SB still had relatively low HRQOL scores

across time. The stability in parent report of HRQOL can be supported by some

longitudinal studies that have also found this trend in pediatric illness groups. For

example, a study of parents of 124 children with newly diagnosed epilepsy found that

HRQOL remained relatively stable across most scales, and only detected a statistical

trend for improvements in emotional functioning over time (Modi, Ingerski, Rausch, &

Glauser). Finally, as some data suggest that young adults with SB may be at a heightened
 61

risk for psychological distress (Bellin et al., 2009), it will be important for studies to

investigate trajectories of HRQOL in this population from childhood to adulthood. In

particular, future research should determine whether trajectories of HRQOL continue to

increase, or have a bell-shaped curve due to the difficulties with an individual securing

employment and gaining independence as well as the continued deterioration of their

condition.

To date, this study is the first to provide multi-informant, multi-method

longitudinal data on the impact of social-environmental factors on HRQOL in two

independent samples of youth with SB. Results of this study indicated that few social-

environmental factors predicted decreased future HRQOL. Specifically, only one illness

variable and one social-environmental variable were significantly associated with youth-

reported HRQOL at Time 2: Pain intensity and parent-reported social competence. In

other words, two-tailed Pearson correlations indicated that higher child pain intensity

when youth were ages 8-15 (Time 1) was associated with lower HRQOL when youth

were ages 10-17 (Time 2). However, the following social-environmental factors were

associated with decreased HRQOL: lower parent-reported social competence, lower

parent-reported community support, and a composite score (i.e., teacher, father, and

teacher) of social skills at Time 1. No other demographic, illness-related, or social-

environmental factors were related to Time 2 youth-report of HRQOL in Sample A.

Similar correlational results were found in Sample B, such that only two variables were

significantly associated with Time 2 youth-reported HRQOL. In particular, decreased

SES and youth social skills (as reported by teachers) were related to lower HRQOL.
 62

Finally, hierarchical linear regression analyses determined that decreased social

competence (in Sample A) and social skills (in Sample B) predicted reduced HRQOL.

Although study findings did not support the expectation that variables from the

community and family environment domains would significantly predict HRQOL in both

samples, the hypothesis that variables within the social adjustment domain of the study

model (see Figure 1) would have the greatest impact on future youth HRQOL was

supported.

Few studies have investigated the impact of poor social adjustment in youth with

SB, thus poor social adjustment and acceptance by peers in this population should receive

increased attention. Historically, research on SB has focused on physical and

neuropsychological domains, with less attention paid to psychological and social

variables (Devine & Holmbeck, 2010). Researchers have found that youth with SB are at

risk for having fewer friendships, social immaturity, and may have poor quality

friendships (Blum, Resnick, Nelson, & St. Germaine, 1991; Ellerton, Stewart, Ritchie &

Hirth, 1996; Devine, Holmbeck, Gayes, & Purnell, 2012). Showcasing the importance of

the social domain in this population, a camp-based intervention originally designed to

increase independence in this population incorporated additional modules aimed to

increase social-related goals, such as greeting others appropriately, contributing to

conversation, speaking clearly and audibly, and asserting self appropriately (Holbein et

al., in press). Results from this intervention of 119 individuals aged 7 to 41 with SB

indicated improvement in campers’  independence,  social  goal  attainment, and social

functioning (Holbein et al., in press). To my knowledge, no interventions in youth and

 63

young adults with SB have incorporated measures of HRQOL as an outcome measure,

which may provide critical information related  to  youths’  overall  improvement. Indeed,

researchers have recently begun to recognize the important role of HRQOL in evaluating

the effectiveness of medical and psychosocial interventions (Sawyer et al., 2006; Seid,

Varni, Segall, & Kurtin, 2004; Varni, Limbers, & Burwinkle, 2007). Further, future

studies may benefit from investigating the impact of several aspects of social competence

in dyadic and general friendships (e.g. social adjustment, social performance, social

skills; Devine, Holmbeck, Gayes, Purnell et al., 2012) on HRQOL, using multiple

methods (e.g. observational coding of peer intervention). Such research may provide

important information for interventions in this population that aim to improve youths’  

social competence and HRQOL.

Given the considerable influence of the family on psychosocial adjustment in

youth with chronic illnesses, the finding that none of the family environment variables

predicted HRQOL was surprising. Despite methodological limitations, some studies have

found associations between family and parent variables in youth with SB (e.g., parental

hope, parental overprotection, maternal psychological distress; Sawin et al., 2002; Abad,

2007 unpublished manuscript) as well as other pediatric illness groups (e.g. diabetes,

asthma, obesity, organ transplant recipients; Pereira, Berg-Cross, Almeida, & Machado,

2008; Sawyer, Spurrier, Whaites, Kennedy, Martin, & Baghurst, 2001; Janicke et al.,

2007; Devine et al., 2011). Regardless, it is possible that individual psychological and

behavioral variables may have more predictive utility than family and contextual

domains. For one, future studies could utilize measures of SB-related stress and family
 64

conflict, which may have more important implications  on  a  youths’  HRQOL  compared  to  

the broad measures of family environment used in the current study. For example, a

longitudinal study of 124 children ages 2-18 with newly diagnosed Type 1 diabetes

measured both general and disease-specific parent-child behaviors and HRQOL, and

found that diabetes-specific family conflict and negative diabetes-specific family

communication were associated with lower HRQOL (Weissberg-Benchell et al., 2009).

In addition, behavioral factors such as poor sleep quality and insomnia have been

implicated in impaired HRQOL in patients with chronic illnesses (Katz & McHorney,

2002; Palermo & Kiska, 2005). In children and adolescents with SB, studies have

revealed the presence of insomnia symptoms (e.g., difficulty initiating and maintaining

sleep; Quine, 1991) and central and obstructive sleep disordered breathing (SDB; 16-

20%; Waters, Forbes, Morielli, et al., 1997). High rates of SDB and other sleep

disturbances in this population may be due to central nervous system malformations and

pulmonary function abnormalities (Waters, Forbes, Morielli, et al., 1997). Thus,

examination of associations between sleep disturbances and HRQOL may be an

important consideration for future research.

Although HRQOL represents an important area of study, it is a complex

construct. The measurement of HRQOL involves several challenges. For one, HRQOL is

a multidimensional construct. A HRQOL psychosocial total score was utilized in

hierarchical analyses, which may have obscured domain differences. For example, it is

possible that increased family stress and conflict may predict decreased emotional and

social HRQOL in youth with SB, but may not predict role-related (i.e., school) HRQOL.
 65

In addition, given the complex medical profiles of youth with SB, HRQOL in this

population may have different predictors than those found for other pediatric populations.

Every pediatric condition has a complex array of symptomatology, as well as a prognosis

and course that may differentially impact a  youth’s HRQOL (Kazak, Rourke, & Crump,

2003). While many chronic illnesses share common features, such as family conflict,

fatigue, pain and/or discomfort in the child, stigmatization by peers, and financial burden,

there are also striking differences in the nature and course of every chronic illness. Some

conditions are highly visible, such as SB, whereas other illnesses have no external

physical manifestations except when the child becomes severely ill (e.g., epilepsy,

diabetes). In addition, SB is a congenital disorder with a chronic and stable course, unlike

conditions such as cancer or children with organ transplants. Thus, youth with SB may

not face the increased and unpredictable stress of conditions such as cancer. Instead,

youth with SB may experience a chronic type of stress due to daily struggles that a

complex medical regimen, ambulation needs, and urologic difficulties require. In

summary, while the social-environmental predictors used in this study were based on

previous research of HRQOL in pediatric populations, the broad differences across

numerous  diagnoses  may  account  for  this  study’s  unique  results.  

Beyond conceptual and theoretical issues influencing the modest findings of the

current study, statistical factors may also account for study findings. The analyses

conducted in this study were fairly conservative. First, HRQOL was controlled at earlier

time points, thus eliminating some of the variance in the dependent variable. Thus, the

change in HRQOL may not have been large enough to yield significant variability in the
 66

residuals that remained after controlling for previous levels of HRQOL. Analyses were

also conservative given the utilization of multiple reporters in the dependent and

independent measures, which addressed common method variance in findings.

Specifically, mother and father reports were combined for several predictor variables as

well as the main outcome variable of HRQOL.

This study represents an important step in identifying that youth with SB are at-

risk for poor HRQOL, and detecting modifiable individual social characteristics that

predict future HRQOL; however, several limitations should be considered. First, because

there was a significant difference in Time 1 parent-reported HRQOL in Sample A, such

that parent non-responders reported lower levels of HRQOL (N of non-responders = 27;

t(127) = -2.45, p < .05), our results may not be representative of youth with particularly

poor quality of life, as these families may have dropped out of the study. Second, Sample

B was relatively small, and statistical power would be enhanced in future studies that use

a larger sample size. Third, Sample A and B were not matched according to number of

participants, age, ethnicity, and several other important demographic variables, which

may have limited our ability to compare results between these two samples. In fact,

Sample B had a relatively small sample size compared to Sample A, and was primarily

composed of Caucasian participants. On the other hand, because similar findings were

found across both independent samples, one may argue that the external validity of the

study was expanded and results may be applicable to heterogeneous populations of youth

with this condition. Fourth, the HRQOL measure used in Sample B has not been well-

established in literature, compared to the psychometrically sound and frequently used

 67

measure of HRQOL used in Sample A (the PedsQL). In sum, although similar results

were found in both samples, more definitive conclusions could be drawn if researchers

were to compare two samples with similar demographics using the same well-established

outcome measure. Although the use of the PedsQL in Sample A allowed for normed

comparison data analysis on chronically ill and healthy youth populations, future studies

should use a matched comparison sample to provide methodologically sound and

sensitive HRQOL comparison analyses. Future research should also investigate a broader

range of individual behavioral and psychosocial predictors of HRQOL, such as SB-

related family conflict, SB-related stress, anxiety, coping, and sleep disturbances.

Mediation models could be tested to identify temporal associations among the factors.

Finally, as noted above, continuing to follow youth and measure HRQOL into adulthood

may elucidate important linear and curvilinear trajectories of individual functioning.

In conclusion, the results of this study suggested that youth with SB are at-risk for

poor HRQOL, and poor social adjustment at Time 1 predicted decreased HRQOL two

years later across two distinct samples. Clinics should routinely examine risk factors of

poor HRQOL in this population. In particular, youth with lower social competence and

poor social skills may represent a subgroup that is particularly vulnerable to poor

HRQOL outcomes. Interventions aimed to improve social competence may help to

improve long-term HRQOL in this population. Currently, clinical and hospital settings

often use ambulation status and bladder/bowel  function  to  determine  a  child’s  HRQOL,  

despite the lack of evidence of the predictive utility of these variables (Sawin & Bellin,

2010; Kirpalani, et al., 2002; Sawin, Brei, Buran, & Fastenau, 2002). Clinicians could
 68

benefit from the use of standard self-administered questionnaires to assess social

adjustment routinely in clinics. Further, it may be useful for interdisciplinary teams to

include a psychologist or social worker to assist in identifying and treating youth with

social adjustment risk factors in order to promote optimal HRQOL. Further research is

required to better understand the role of social adjustment in youth with SB in order to

identify strategies to reduce its impact on broader domains of functioning.

 APPENDIX A

QUESTIONNAIRE MEASURES

69
70
71
72
73
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
 APPENDIX B

OBSERVATIONAL CODING MANUAL

101
[102]
[103]
[104]
[105]
[106]
[107]
[108]
[109]
[110]
 REFERENCES

Abad, M. C. (2007). Predictors of quality of life in youth with spina bifida (Unpublished
doctoral dissertation). Loyola University Chicago, Chicago, IL.

Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/4-18 and 1991
profile. Burlington: Department of Psychiatry, University of Vermont.

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(3), 370–379.

Andelman, R. B., Zima, B. T., & Rosenblatt, A. B. (1999). Quality of life of children:
Toward conceptual clarity. In Maruish M. (Ed.), The use of psychological testing
for treatment planning and outcomes assessment (1383–1413). Mahwah, NJ:
Erlbaum.

Bier, J. B., Prince, A., Tremont, M., & Msall, M. (2005). Medical, functional, and social
determinants of health-related quality of life in individuals with
myelomeningocele. Developmental Medicine & Child Neurology, 47(9), 609–
612.

Blum, R. W., Resnick, M. D., Nelson, R., & St. Germaine, A. (1991). Family and peer
issues among adolescents with spina bifida and cerebral palsy. Pediatrics, 88(2),
280–285.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development:

Research perspectives. Developmental Psychology Developmental Psychology,
22(6), 723–742.

Cate, I. P., Kennedy, C., & Stevenson, J. (2002). Disability and quality of life in spina
bifida and hydrocephalus. Developmental Medicine & Child Neurology, 44(5),
317–322.

Centers for Disease Control and Prevention. (2011). Neural tube defect ascertainment
project. Retrieved June 26, 2011, from http://www.nbdpn.org

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155–159.

Cramer, J. A., Westbrook, L. E., Devinsky, O., Perrine, K., Glassman, M. B., &
Camfield, C. (1999). Development of the quality of life in epilepsy inventory for
adolescents: The QOLIE-AD-48. Epilepsia, 40(8), 1114–1121.

111
 112

Danielsson, A., Bartonek, Ê., Levey, E., McHale, K., Sponseller, P., & Saraste, H.
(2008). Associations between orthopaedic findings, ambulation and health-related
quality of life in children with myelomeningocele. Journal of Children's
Orthopaedics, 2(1), 45–54.

de Civita, M., Regier, D., Alamgir, A. H., Anis, A. H., FitzGerald, M. J., & Marra, C. A.
(2005). Evaluating health-related quality-of-life studies in pediatric populations:
Some conceptual, methodological and developmental considerations and recent
applications. PharmacoEconomics, 23(7), 659–685.

Devine, K. A., Holmbeck, G. N., Gayes, L., & Purnell, J. Q. (2012). Friendships of
children and adolescents with spina bifida: Social adjustment, social performance,
and social skills. Journal of Pediatric Psychology, 37(2), 220–231.

Devine, K. A., Reed-Knight, B., Loiselle, K. A., Simons, L. E., Mee, L. L., & Blount, R.
L. (2011). Predictors of long-term health-related quality of life in adolescent solid
organ transplant recipients. Journal of Pediatric Psychology, 36(8), 891–901.

Devinsky, O., Vickrey B. G., Cramer J., Perrine K., Hermann, B., Meador, K., & Hays,
R. D. (1995). Development of the quality of life in epilepsy inventory. Epilepsia,
36(11), 1089–104.

Drotar D. (2004). Validating measures of pediatric health status, functional status, and
health-related quality of life: Key methodological challenges and strategies.
Ambulatory Pediatrics, 4(4), 358–364.

Drotar, D. (1997). Relating parent and family functioning to the psychological adjustment
of children with chronic health conditions: What have we learned? What do we
need to know? Journal of Pediatric Psychology, 22(2), 149–165.

Dunn, L. M., & Dunn, L. M. (1981). PPVT Peabody Picture Vocabulary Test, Revised
Forms L and M. Circle Pines, MN: American Guidance Service.

Edwards, T., Huebner, C., Connell, F., & Patrick, D. (2002). Adolescent quality of life,
part I: Conceptual and measurement model. Journal of Adolescence, 25(3), 275–
286.

Eiser, C., & Jenney, M. (2007). Measuring quality of life. Archives of Disease in
Childhood, 92(4), 348–350.

Eiser, C., & Morse, R. (2001). Can parents rate their child's health-related quality of life?
results of a systematic review. Quality of Life Research, 10(4), 347–357.

Eiser, C., & Morse, R. (2001). The measurement of quality of life in children: Past and
future perspectives. Journal of Developmental and Behavioral Pediatrics: JDBP,
22(4), 248–256.
 113

Ellerton, M. L., Stewart, M. J., Ritchie, J. A., & Hirth, A. M. (1996). Social support in
children with a chronic condition. Canadian Journal of Nursing Research, 28(4),
15–36.

Ferreira, A. (1963). Decision-making in normal and pathologic families: A study.

Archives of General Psychiatry, 8, 68–73.

Fletcher, J. M., & Brei, T. J. (2010). Introduction: Spina bifida—A multidisciplinary

perspective. Developmental Disabilities Research Reviews, 16(1), 1–5.

Fletcher, J. M., Copeland, K., Frederick, J. A., Blaser, S. E., Kramer, L. A., Northrup, H.,
& Dennis, M. (2005). Spinal lesion level in spina bifida: A source of neural and
cognitive heterogeneity. Journal of Neurosurgery, 102(3), 268–279.

Goodwin, D. J., Boggs, S. R., & Graham-Pole, J. (1994). Development and validation of
the pediatric oncology quality of life scale. Psychological Assessment
Psychological Assessment, 6(4), 321–328.

Goss, C. H., & Quittner, A. L. (2007). Patient-reported outcomes in cystic fibrosis.

Proceedings of the American Thoracic Society, 4(4), 378–386.

Gresham, F. M., & Elliot, S. N. (1990). Social skills rating system. Circle Pines, MN:
American Guidance Service.

Harter, S. (1985). Manual for the self-perception profile for children. Denver, CO:
University of Denver.

Holbein, C. E., Murray, C. B., Psihogios, A. M., Wasserman, R. M., Essner, B. S.,
O’Hara,  L.  K.,  Holmbeck,  G.  N.  (in  press).  A  camp-based psychosocial
intervention to promote independence and social function in individuals with
spina bifida: Moderators of treatment effectiveness. Journal of Pediatric
Psychology.

Hollingshead, A. B. (1975). Four factor index of social status. New Haven, CT:
Department of Sociology, Yale University.

Holmbeck, G. N., & Faier-Routman, J. (1995). Spinal lesion level, shunt status, family
relationships, and psychosocial adjustment in children and adolescents with spina
bifida myelomeningocele. Journal of Pediatric Psychology, 20(6), 817–832.

Holmbeck, G. N., Belvedere, M. C., Gorey-Ferguson, L., & Schneider, J. (1995). Family
macro-coding manual (Manual, March of Dimes Triadic Version). Unpublished
coding system, Loyola University Chicago, Chicago, IL.
 114

Holmbeck, G. N., Westhoven, V. C., Phillips, W. S., Bowers, R., Gruse, C.,
Nikolopoulos, T., & Davison, K. (2003). A multimethod, multi-informant, and
multidimensional perspective on psychosocial adjustment in preadolescents with
spina bifida. Journal of Consulting and Clinical Psychology, 71(4), 782–796.

Holmbeck, G. N., Zebracki, K., Johnson, S. Z., Belvedere, M., & Schneider, J. (2007a).
Child-peer interaction macro-coding manual. Unpublished manual, Loyola
University of Chicago, Chicago, IL.

Holmbeck, G. N., Zebracki, K., Johnson, S. Z., Belvedere, M., & Schneider, J. (2007b).
Parent-child interaction macro-coding manual. Unpublished manual, Loyola
University of Chicago, Chicago, IL.

Holmbeck, G., Hommeyer, J., Wills, K., & Coers, S. (1999). Condition severity and
psychosocial functioning in pre-adolescents with spina bifida: Disentangling
proximal functional status and distal adjustment outcomes. Journal of Pediatric
Psychology, 24(6), 499–509.

Hunfeld, J. A., Perquin, C. W., Duivenvoorden, H. J., Hazebroek-Kampschreur, A. A.,

Passchier, J., van Suijlekom-Smit, L. W., & van der Wouden, J. C. (2001).
Chronic pain and its impact on quality of life in adolescents and their families.
Journal of Pediatric Psychology, 26(3).

Ingersoll, G. M., & Marrero, D. G. (1991). A modified quality-of-life measure for youths:
Psychometric properties. The Diabetes Educator, 17(2), 114–118.

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
overweight youth. Obesity, 15(7), 1799–1807.

Keith, K. D., & Schalock, R. L. (1994). The measurement of quality of life in

adolescence: The quality of student life questionnaire. American Journal of
Family Therapy, 22(1), 83–87.

Kennedy, S. E., Martin, S. D. G., Kelley, J. M., Walton, B., Vlcek, C. K., Hassanein, R.
S., & Holmes, G. E. (1998). Identification of medical and nonmedical needs of
adolescents and young adults with spina bifida and their families: A preliminary
study. Children's Health Care, 27(1), 47–61.

Kirpalani, H. M., Rosenbaum, P. L., Van Nie, A. J., Willan, A. R., King, D., Parkin, P.
C., & Fehlings, D. L. (2000). Quality of life in spina bifida: Importance of
parental hope. Archives of Disease in Childhood, 83(10), 293–297.

Koot, H. M., & Wallander, J. L. (2001). Quality of life in child and adolescent illness:
Concepts, methods, and findings. New York, NY: Brunner-Routledge.
 115

Kulkarni, A. V., Cochrane, D. D., McNeely, P. D., & Shams, I. (2008). Medical, social,
and economic factors associated with health-related quality of life in Canadian
children with hydrocephalus. The Journal of Pediatrics, 153(5), 689–695.

La Greca A. M. (1992). Peer influences in pediatric chronic illness: An update. Journal of

Pediatric Psychology, 17(6), 775–784.

Landgraf, J. M., Abetz, L., & Ware, J. E. (1996). Child health questionnaire (CHQ): A
user's manual. Boston, MA: The Health Institute, New England Medical Center.

Leger, R. R. (2005). Severity of illness, functional status, and HRQOL in youth with
spina bifida. Rehabilitation Nursing, 30(5), 180–187.

Lemelle, J. L., Guillemin, F., Aubert, D., Guys, J. M., Lottmann, H., Lortat-Jacob, S., &
Schmitt, M. (2006). Quality of life and continence in patients with spina bifida.
Quality of Life Research, 15(9), 1481–1492.

MacNeily, A. E., Jafari, S., Scott, H., Dalgetty, A., Afshar, K. (2009). Health related
quality of life in patients with spina bifida: A prospective assessment before and
after lower urinary tract reconstruction. The Journal of Urology, 182(4), 1984–
1992.

McCubbin, H. I., Patterson, J., & Wilson, L. (1982). Family Inventory of Life Events
Scale (FILE). St. Paul: University of Minnesota, Department of Family Social
Services.

Modi, A. C., & Quittner, A. L. (2003). Validation of a disease-specific measure of health-

related quality of life for children with cystic fibrosis. Journal of Pediatric
Psychology, 28(8), 535–546.

Moos, R. H., & Moos, B. S. (2002). Family environment scale manual: Development,
applications, research. Palo Alto, CA: Center for Health Care Evaluation,
Department of Veterans Affairs.

Müller-Godeffroy, E., Michael, T., Poster, M., Seidel, U., Schwarke, D., & Thyen, U.
(2008). Self-reported health-related quality of life in children and adolescents
with myelomeningocele. Developmental Medicine & Child Neurology, 50(6),
456–461.

Murch, R. L., & Cohen, L. H. (1989). Relationships among life stress, perceived family
environment, and the psychological distress of spina bifida adolescents. Journal
of Pediatric Psychology, 14(2), 193–214.

Nova Southeastern University Center for Psychological Studies. (2008). Family

environment scale. Retrieved March 3, 2008, from http://www.cps.nova.edu/
~cpphelp/FES.html
 116

Oddson, B. E., Clancy, C. A., & McGrath, P. J. (2006). The role of pain in reduced
quality of life and depressive symptomology in children with spina bifida. The
Clinical Journal of Pain, 22(9), 784–789.

Olsson, C., Boyce, M., & Toumbourou, J. (2005). The role of peer support in facilitating
psychosocial adjustment to chronic illness in adolescence. SAGE Family Studies
Abstracts, 27(4), 78–87.

Palermo, T. M., Long, A. C., Lewandowski, A. S., Drotar, D., Quittner, A. L., & Walker,
L. S. (2008). Evidence-based assessment of health-related quality of life and
functional impairment in pediatric psychology. Journal of Pediatric Psychology,
33(9), 983–996.

Park, J., Turnbull, A. P., & Turnbull, H. R., III. (2002). Impacts of poverty on quality of
life in families of children with disabilities. Exceptional Children, 68(2), 151–
170.

Parkin, P. C., Kirpalani, H. M., Rosenbaum, P. L., Fehlings, D. L., Nie, A. V., Willan, A.
R., & King, D. (1997). Development of a health-related quality of life instrument
for use in children with spina bifida. Quality of Life Research, 6(2), 123–132.

Pereira, M. G., Berg-Cross, L., Almeida, P., & Machado, J. C. (2008). Impact of family
environment and support on adherence, metabolic control, and quality of life in
adolescents with diabetes. International Journal of Behavioral Medicine, 15(3),
187–193.

Quittner, A. L., Davis, M. A., & Modi, A. C. (2003). Health-related quality of life in
pediatric populations. In M. Roberts (Ed.), Handbook of pediatric psychology (pp.
696–709). New York, NY: Guilford Press.

Raphael, D., Rukholm, E., Brown, I., Hillbaily, P., & Donato, E. (1996). The quality of
life profile—Adolescent version: Background, description, and initial validation.
Journal of Adolescent Health Journal of Adolescent Health, 19(5), 366–375.

Robin, A. L., & Foster, S. L. (1989). Negotiating parent-adolescent conflict: A

behavioral-family systems approach. New York, NY: Guilford Press.

Rose, B. M., & Holmbeck, G. N. (2007). Attention and executive functions in

adolescents with spina bifida. Journal of Pediatric Psychology, 32(8), 983–994.

Sarwark, J. F. (1996). Spina bifida. Pediatric Clinics of North America, 43(5), 1151–
1158.

Sattler, J. M. (2002). Assessment of children: Behavioral and clinical applications. San

Diego, CA: Sattler.
 117

Sawin, K. J., & Bellin, M. H. (2010). Quality of life in individuals with spina bifida: A
research update. Developmental Disabilities Research Reviews, 16(1), 47–59.

Sawin, K. J., Brei, T. J., Buran, C. F., & Fastenau, P. S. (2002). Factors associated with
quality of life in adolescents with spina bifida. Journal of Holistic Nursing, 20(3),
279–304.

Sawin, K. J., Bellin, M. H., Roux, G., Buran, C. F., & Brei, T. J. (2009). The experience
of self-management in adolescent women with spina bifida. Rehabilitation
Nursing, 34(1), 26–38.

Sawyer, M. G., Spurrier, N., Whaites, L., Kennedy, D., Martin, A. J., & Baghurst, P.
(2000). The relationship between asthma severity, family functioning and the
health-related quality of life of children with asthma. Quality of Life Research: An
International Journal of Quality of Life Aspects of Treatment, Care and
Rehabilitation, 9(10), 1105–1115.

Seja Kaugars, A., Zebracki, K., Kichler, J. C., Fitzgerald, C. J., Neff Greenley, R.,
Alemzadeh, R., & Holmbeck, G. N. (2011). Use of the family interaction macro-
coding system with families of adolescents: Psychometric properties among
pediatric and healthy populations. Journal of Pediatric Psychology, 36(5), 539–
551.

Slavin, L. A. (1991). Validation studies of the PEPSS, a measure of perceived emotional

support for use with adolescents. Journal of Adolescent Research Journal of
Adolescent Research, 6(3), 316–335.

Smith, K. W., Avis, N. E., & Assmann, S. F. (1999). Distinguishing between quality of
life and health status in quality of life research: A meta-analysis. Quality of Life
Research, 8(5), 447–459.

Starfield, B., Riley, A. W., Green, B. F., Ensminger, M. E., Ryan, S. A., Kelleher, K., . . .
Vogel, K. (1995). The adolescent child health and illness profile: A population-
based measure of health. Medical Care, 33(5), 553–566.

Stein, R. E., & Jessop, D. J. (1990). Functional status II(R). A measure of child health status.
Medical Care, 28(11), 1041–1055.

Tennant, A., Kearns, S., Turner, F., Wyatt, S., Haigh, R., & Chamberlain, M.A. (2001).
Measuring the function of children with juvenile arthritis. Rheumatology, 40(11),
1274–1278.

Testa, M. A., & Simonson, D. C. (1996). Assessment of quality-of-life outcomes. New

England Journal of Medicine, 334(13), 835-840.
 118

Theunissen, N. M., Vogels, T. C., Koopman, H. M., Verrips, G. W., Zwinderman, K. H.,
Verloove-Vanhorick, S. P., & Wit, J. M. (1998). The proxy problem: Child report
versus parent report in health-related quality of life research. Quality of Life
Research, 7(5), 387–397.

Varni, J. W., Burwinkle, T. M., Rapoff, M. A., Kamps, J. L., & Olson, N. (2004). The
PedsQL & trade; in pediatric asthma: Reliability and validity of the pediatric
quality of life Inventory & trade; generic core scales and asthma module. Journal
of Behavioral Medicine, 27(3), 297–318.

Varni, J. W., Katz, E. R., Seid, M., Quiggins, D. J., Friedman-Bender, A., & Castro, C.
M. (1998). The pediatric cancer quality of life inventory (PCQL). I. instrument
development, descriptive statistics, and cross-informant variance. Journal of
Behavioral Medicine, 21(2), 179–204.

Varni, J. W., Limbers, C., & Burwinkle, T. M. (2007). Literature review: Health-related
quality of life measurement in pediatric oncology: Hearing the voices of the
children. Journal of Pediatric Psychology, 32(9), 1151–1163.

Varni,  J.  W.,  Seid,  M.,  &  Kurtin,  P.  S.  (2001).  PedsQL™  4.0:  Reliability  and  validity  of  
the  pediatric  quality  of  life  inventory™  version  4.0  generic  core  scales  in  healthy  
and patient populations. Medical Care, 39(8), 800–812.

Varni, J. W., Seid, M., & Rode, C. A. (1999). The  PedsQL™:  Measurement  model  for  
the pediatric quality of life inventory. Medical Care, 37(2), 126–139.

Varni, J. W., Seid, M., Knight, T. S., Uzark, K., & Szer, I. S. (2002). The PedsQLTM 4.0
Generic Core Scales: Sensitivity, responsiveness, and impact on clinical decision-
making. Journal of Behavioral Medicine, 25(2), 175–193.

Verhoef, M., Post, M. M., Barf, H. A., van Asbeck, F., Gooskens, R. M., & Prevo, A. H.
(2007). Perceived health in young adults with spina bifida. Developmental
Medicine & Child Neurology, 49(3), 192–197.

Ware, J. E., Kosinski, M., & Keller, S. D. (1996). A 12-item short-form health survey:
Construction of scales and preliminary tests of reliability and validity. Medical
Care, 34(3), 220–233.

Wechsler, D. (1999). Wechsler abbreviated scale of intelligence WASI: Manual. San

Antonio, TX: Pearson.

Weissberg-Benchell, J., Nansel, T., Holmbeck, G., Chen, R., Anderson, B., Wysocki, T.,
& Laffel, L. (2009). Generic and diabetes-specific parent–child behaviors and
quality of life among youth with type 1 diabetes. Journal of Pediatric Psychology,
34(9), 977–988.
 119

Wood, D., Watts, G., Hauser, K., Rouhani, P., & Frias, J. (2009). Impact of chronic pain
and other health problems on the quality of life in children and young adults with
spina bifida. International Journal of Child and Adolescent Health, 2(3), 395–
404.
 VITA

The author, Caitlin Beth Murray, was born on October 15th, 1986 in Ithaca, New

York to John and Kathy Murray. She graduated Cum Laude in Psychology from the

University of North Carolina, Chapel Hill in May 2008. In December 2013, she

graduated with a Master of Arts in Clinical Psychology from Loyola University Chicago.

Ms. Murray is currently pursuing her doctoral degree in Clinical Psychology at Loyola

University Chicago. Her research interests are in the area of pediatric psychology.

Currently, her research addresses the impact of painful somatic symptoms and sleep

disturbances on mental health and health-related quality of life in youth with spina bifida.

She currently resides in Chicago, Illinois.

120