xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002006384
007 cr mnu|||uuuuu
008 090610s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002729
Meyer, Aja M.
Impact of seizure-related variables and psychopathology on health-related quality of life in pediatric epilepsy
h [electronic resource] /
by Aja M. Meyer.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 145 pages.
Dissertation (Ph.D.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: Psychopathology typically is a lasting condition that is persistent from childhood to adulthood. Therefore, it is imperative that children with health conditions and comorbid psychiatric disorders are treated for both conditions as they are likely to have a significant negative impact on children's overall health-related quality of life (HRQL). More specifically, it is important to identify the variables that affect HRQL in children with epilepsy. Research has shown that biomedical variables such as seizure severity and frequency have only moderate relationships with HRQL; therefore, additional factors, such as depression and anxiety, must be identified so that they also may be a focus of treatment. The purpose of this study was to ascertain the relationship among seizure-related variables, health-related quality of life, and psychopathology (i.e., anxiety and depression) in children with epilepsy (n = 51). The seizure-related variables examined in this study include type of seizure, seizure frequency, and seizure treatment with anti-epileptic drugs (AEDs). Canonical correlation analyses indicated that self-report and parent report of anxiety and depression were most strongly correlated with HRQL. Additionally, seizure frequency and number of anti-epileptic drugs also were correlated with HRQL. It is hoped that results from this study will inform both the medical and psychosocial treatment children with epilepsy receive. This comprehensive care needs to go beyond simply attempting to control seizures with minimal adverse drug reactions. Results from this study will contribute to the literature underscoring the importance of identifying, diagnosing, and treating children with epilepsy who have comorbid psychopathology so that they have the best possible psychosocial outcomes.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Kathy L. Bradley-Klug, Ph.D.
x Psychological and Social Foundations
t USF Electronic Theses and Dissertations.
Impact of Seizure-Related Variables and Psychopathology on Health-Related Quality of Life in Pediatric Epilepsy by Aja M. Meyer Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychological and Social Foundations College of Education University of South Florida Major Professor: Kathy L. Bradley-Klug, Ph.D. Kathleen Armstrong, Ph.D. Richard Marshall, Ph.D. Anthony Onwuegbuzie, Ph.D. Date of Approval: November 3, 2008 Keywords: seizure disorders, depression, anxiety, canonical correlation, neurologist Copyright 2008, Aja M. Meyer
i Table of Contents List of Tables v Abstract vi Chapter 1: Introduction 1 Statement of the Problem 1 Pediatric Epilepsy 2 Pediatric Epilepsy and Psychopathology 3 Pediatric Epilepsy and Health-Related Quality of Life 4 Implications for School Psychology 5 Summary 7 Purpose of the Study 8 Research Questions 9 Research Question 1 9 Research Question 2 9 Hypotheses 9 Significance of the Study 10 Definition of Terms 10 Chapter 2: Review of the Related Literature 13 Overview 13 Defining Epilepsy 13 Prevalence of Epilepsy in Children 14 Classification of Seizures 15 Types of Seizures 17 Generalized Tonic-Clonic Seizures 17 Myoclonic Seizures 17 Absence Seizures 17 Atonic Seizures 18 Other Classificati ons of Seizures 18 Classification by Cause of Seizure 18 Common Types of Seizures in Children 19 Neonatal Seizures 19 Febrile Seizures 20 Childhood Absence Epilepsy 20 Juvenile Myoclonic Epilepsy 21 Benign Rolandic Epilepsy 21 Benign Occipital Epilepsy 21
ii Infantile Spasms 22 Lennox-Gastaut Syndrome 22 Potential Causes of Childhood Epilepsy 22 Diagnosis of Epilepsy in Children 24 Treatment of Pediatric Epilepsy 25 Initiating Treatment 25 Treatment with Medication 26 Use of AEDs: Monotherapy versus Polytherapy 27 Psychopathology and Epilepsy 27 Potential Causes of Psychopa thology in Pediatric Epilepsy 28 Relationship between seizure type and psychopathology 28 Relationship between seizur e frequency and psychopathology 28 Relationship between treatment with antiepileptic drugs and psychopathology 29 Neurological dysfunctio n and psychopathology 30 Response to epilepsy and psychopathology 30 Comorbid Disorders: Epilepsy and Internalizing Disorders 31 Treating Children with Epile psy and Comorbid Disorders 36 Summary of Psychopathology and Epilepsy 39 Health-Related Quality of Life and Epilepsy 40 Defining Health-Related Quality of Life (HRQL) 42 Measuring HRQL in Children with Epilepsy 43 Health-related quality of life measures 46 Epilepsy-specific HRQL Measures for Children 47 Summary 52 Chapter 3: Method 55 Overview 55 Selection-Eligibility Criteria 55 Sampling 57 Sampling Scheme 57 Sample Size 57 Instruments 57 Demographics and Seizure Variables Questionnaire (DSV Questionnaire) 57 Behavior Assessment System for Children, Second Edition (BASC-2; Reynolds & Kamphaus, 2004) 59 Health-Related Quality of Life Questionnaire (HRQL; Ronen et al., 2004) 62 Procedures 64 Research Design 69 Data Analyses 69 Descriptive Statistics 69 Variables 70 Independent variables 70
iii Dependent Variables 70 Inferential Statistics 70 Canonical Correlation Analysis 70 Chapter 4: Results 73 Demographics 73 Independent Variables 74 Dependent Variables 76 Defining Characteristics of BASC-2 Anxiety Scale: Child Self-Report 79 Anxiety and Seizure Type 79 Anxiety and Seizure Frequency 79 Anxiety and Use of AEDs 80 Defining Characteristics of BASC-2 Anxiety Scale: Parent Report 80 Anxiety and Seizure Type 80 Anxiety and Seizure Frequency 81 Anxiety and Use of AEDs 81 Defining Characteristics of BASC-2 De pression Scale: Ch ild Self-Report 82 Depression and Seizure Type 82 Depression and Seizure Frequency 82 Depression and Use of AEDs 83 Defining Characteristics of BASC-2 Depression Scale: Parent Report 83 Depression and Seizure Type 83 Depression and Seizure Frequency 84 Depression and Use of AEDs 84 Understanding Aspects of Health-Relat ed Quality of Life in the Current Study 85 Research Findings 86 Question One 86 Statistical Assumptions for Ca nonical Correlation Analysis 89 Canonical Correlation Analys is (Child Report) 89 Question Two 93 Canonical Correlation Analysis (Parent Report) 96 Summary of Canonical Correlation Analysis 98 Chapter 5: Discussion 100 Summary of Study 100 Relationship between Seizure-Related Variables and Health-Related Quality of Life 101 Seizure Frequency and Health-Related Quality of Life 101 Treatment with AEDs and Health-Related Quality of Life 103 Seizure Type and Health-Related Quality of Life 103 Relationship between Psychopathology, Seizure-related Variables, and Health-Related Quality of Life 104 Limitations of the Study 105 Internal and External Validity 105
iv Sample Size 107 Contributions of the Study 108 Future Research 110 Final Thoughts 113 References 115 Appendices 130 Appendix A: Demographics and Se izure Variables Questionnaire 131 Appendix B: Child Cover Letter and Assent Form 132 Appendix C: Parent Cover Letter and Consent Form 134 Appendix D: Health-Related Qual ity of Life Â– Parent Form 136 Appendix E: Health-Related Qu ality of Life Â– Child Form 141 About the Author End Page
v List of Tables Table 1: Descriptive Statistics for Participants 77 Table 2: Means, Standard Deviations, Skewness, and Kurtosis for Participant Ratings 78 Table 3: Skewness and Kurtosis Coefficients for BASC-2 85 Table 4: PearsonÂ’s Correlations (Child Self-Report) 88 Table 5: Canonical Correlations (Child Self-Report) 92 Table 6: PearsonÂ’s Correlations (Parent-Report) 95 Table 7: Canonical Correlations (Parent-Report) 99
vi Impact of Seizure-Related Variables and Ps ychopathology on Health-Related Quality of Life in Pediatric Epilepsy Aja M. Meyer ABSTRACT Psychopathology typically is a lasting condi tion that is persistent from childhood to adulthood. Therefore, it is imperative that children with health conditions and comorbid psychiatric disorders are treated for both conditions as they are likely to have a significant negative impact on childrenÂ’s overa ll health-related quality of life (HRQL). More specifically, it is important to identify the variables that affect HRQL in children with epilepsy. Research has shown that biomed ical variables such as seizure severity and frequency have only moderate relationships with HRQL; therefore, additional factors, such as depression and anxiety, must be identi fied so that they also may be a focus of treatment. The purpose of this study was to ascert ain the relationship among seizure-related variables, health-re lated quality of life, and ps ychopathology (i.e., anxiety and depression) in childr en with epilepsy ( n = 51). The seizure-related variables examined in this study include type of seizure, seizur e frequency, and seizure treatment with antiepileptic drugs (AEDs). Canonical correlation analyses indicated that self-report and parent report of anxiety a nd depression were most str ongly correlated with HRQL. Additionally, seizure frequency and number of anti-epileptic drugs also were correlated with HRQL. It is hoped that results from this study will inform both the medical and
vii psychosocial treatment children with epilepsy receive. This comprehensive care needs to go beyond simply attempting to control seizur es with minimal adverse drug reactions. Results from this study will contribute to the literature underscoring the importance of identifying, diagnosing, and treating child ren with epilepsy who have comorbid psychopathology so that they have the best possible psychosocial outcomes.
1 Chapter 1 Introduction Statement of the Problem A thorough review of the li terature reveals that appr oximately 10% to 15% of school-aged children in the United States are inflicted with a chronic medical condition (Bethell et al., 2002; Davido ff, 2004; Van Dyck et al., 2002). Of these conditions, epilepsy is one of the most physically and psyc hosocially debilitating chronic illnesses in children (Devinsky, 2003). Epilepsy is a neurol ogical condition that affects the central nervous system and has various origins, symptoms, courses, and prognoses (Shafer, 2002). Pediatric epilepsy is one of the mo st common neurological conditions, with a prevalence rate of 3.6 to 5.0 per 1,000 for ch ildren, birth to adolescence, in developed countries (Hiemenz, Hynd, & Jimenez, 1999; Ronen, Streiner, & Rosenbaum, 2003). According to the United States Census Bureau, there were 58 million students, kindergarten through 12th grade, enrolled in public schools in 2003. Thus, the abovecited pediatric epilepsy prevalence rates indicate that approximately 290,000 school-aged children in the United States have epileps y. However, according to Sachs and Barrett (1995), as many as 500,000 children and adolesce nts in the United States may experience recurrent seizures. Physical health and mental health ar e intertwined (Huberty, Austin, & Huster, 2000); therefore, it is important to assess th e mental health of individuals who have chronic medical conditions such as epilepsy. Gi ven that depression a nd anxiety have been
2 found to be the most prevalent co-morbid psyc hiatric disorders in adults with epilepsy (Caplan, Siddarth, & Gurbani, 2005), it is cr ucial for practitioners to ascertain how depression and anxiety manifest in children wi th epilepsy. This is especially important given recent research findings with adu lts that indicate de pression, through the mechanism of sleep deprivation, can have a direct impact on seizure frequency, increasing the rate of seizures and negativ ely impacting health-re lated quality of life (HRQL) (Jackson & Turkington, 2005). Empirical studies demonstrate a rela tionship between psychopathology and poor quality of life in children (Bastiaansen, Koot, & Ferdinand, 2005). Furthermore, research has found that children with physical illn ess have a poorer HRQL (Baker, Spector, McGrath, & Soteriou, 2005). Therefore, the risk for compromised quality of life is confounded in children with physical/medical conditions (e.g., epilepsy) and comorbid psychopathology (Wagner & Smith, 2006). In light of these research findings, there is a need to explore the impact of psychopathol ogy (especially depressi on and anxiety) on the HRQL of children with epilepsy so that appr opriate treatment may be provided early for optimal health outcomes. Pediatric Epilepsy Pediatric epilepsy is a neurological condition wherein abrupt changes in how cells in the brain send electrical signals cause seizures in children. Epilepsy is a condition of chronic, recurri ng seizures. A seizure occurs when there is a sudden discharge or disturbance of th e electrochemical firing of neurons in the brain that may cause a change in an individualÂ’s sensa tion, awareness, and/or behavior (DuLac,
3 MacDonald, & Kelly, 1995). Pediatric epilepsy is a condition that ha s various origins, symptoms, courses, and prognoses for children (Shafer, 2002). Numerous factors contribute to the medi cal and psychosocial difficulties faced by children with epilepsy. For example, the type and frequency of the seizures may be correlated with increased difficulties both medically and psychosocially. Frequency of seizures has been found to be correlated with behavioral problems in children with epilepsy (Austin, Risinger, & Beckett, 1992; Lambert & Robertson, 1999). Furthermore, there are some data to support a relationshi p between seizure fre quency and anxiety, whereby increased levels of stress and anxi ety appear to escalate the frequency of seizures (Vazquez & Devinsky, 2003). Finally, th e use of antiepileptic drugs (AEDs) may produce negative side effects in children with epilepsy. Mendez, Doss, Taylor, and Salguero (1993) found that the use of multiple AEDs was associated with depression in children with epilepsy. Presently, there is a dear th of research examining the relationship between various seizure-related variables a nd psychopathology, and the data that have been collected are inconclusive. Therefore, it is imperative that researchers determine the impact of seizure-related variables on psychopathology in children with epilepsy. Pediatric Epilepsy and Psychopathology Along with the goal of reducing or elimina ting seizures, practitioners also must focus on the individualÂ’s psychosocial adjust ment to promote the most optimal outcomes for children with epilepsy. Therefore, it is pr oblematic that treatment of pediatric epilepsy has focused primarily on seizure control and treatment adherence while overlooking learning and behavior problems as we ll as comorbid psychopathology. Because recognition of the relationshi p between pediatric epilepsy and comorbid psychopathology
4 is newly emerging in the literature, a compre hensive approach to assessing and treating these children has not yet been establishe d (Caplan et al., 2005). Broadband assessment instruments provide information on an indivi dualÂ’s functioning in multiple domains; a number of empirically-based broadband measures of emotional and behavioral symptoms in children may be utilized to assess the presence of psychopathology (e.g., Child Behavior Checklist; Achenbach, 2000; Beha vior Assessment System for Children; Reynolds & Kamphaus, 2004). Pediatric Epilepsy and Health-Related Quality of Life Quality of life refers to an individual's ability to engage in and enjoy normal life activities (Bastiaansen et al., 2005). More sp ecifically, health-rela ted quality of life (HRQL) represents the functional effects of an illness and its subs equent treatment on a patient, as perceived by the individual (B aker, 2001). Several groups of researchers (Bastiaansen, Koot, Ferdinand, & Verhulst, 2004; Sawyer, Whaites, & Rey, 2002) have conducted studies examining quality of life (QoL ) in children with psychiatric disorders. Their findings revealed that children with psychiatric disorders had poor QoL. Moreover, not only were these childrenÂ’s QoL considerab ly lower than children from the population at large, but their QoL also was poorer th an the QoL of physical ly ill children. These results suggest that there is a relationship between poor QoL and psychopathology in children (Bastiaansen et al., 2005). In addition, these findings have revealed that not only do children with physic al illness have a poorer QoL (as compared with healthy children), but children with psyc hiatric disorders also have lowered QoL. In light of these findings, it is reasonable to hypot hesize that children w ith a physical illness
5 such as epilepsy as well as a psychiatric di sorder (e.g., anxiety and depression) are at a substantially greater risk for a poor QoL. Implications for School Psychology Children with chronic health conditions have increased risk factors for poor academic and psychosocial functioning compared with their healthy peers (Thompson & Gustafson, 1996). A number of researcher s (e.g., Aldenkamp, Overweg-Plandsoen, & Arends, 1999; Besag, 1995; Lhatoo & Sander, 2001) have found that children with epilepsy display significant difficulties in l earning and behavior compared to healthy children. Furthermore, research has shown that children with chronic neurological conditions such as epilepsy are more likely to experience academic difficulties compared to those whose illness does not have a neur ological basis (Howe, Feinstein, & Reiss, 1993). Although it has been hypothesized that aca demic difficulties may be due to limited intellectual ability, researchers have f ound that these differences in academic achievement in children with epilepsy are not attributed to limited cognitive functioning (Seidenberg & Berent, 1992; Sturniolo & Galletti, 1994). Un fortunately, the origin of these difficulties still has not been ascerta ined. It is hypothesized that learning and behavior problems may be due to factor s such as seizure-related variables, pharmacological treatment (e.g., AEDs), or environmental factors (e.g., social interactions) that impact children with epilepsy (Bourgeois, 1998). When children experience learning and be havior problems, it is likely that there will be a direct impact on their academic and behavioral performance at school. Because chronic health conditions such as epilepsy impact a multitude of domains, educational laws have been created to ensure appropr iate support for indi viduals with chronic
6 conditions in the educational system. One such law that pertains to the education of children and adolescents with disabilities is the Reauthorization of Individuals with Disabilities Education Act (IDEA; United St ates Department of Education, 2004). Other Health Impaired is a category under IDEA that provides servi ces to students with health conditions such as epilepsy. The reauthorizat ion of IDEA stipulat es that educators, including school psychologists, are obligated to provide suppor ts and services to students whose chronic health condition negatively impacts their academic and behavioral functioning. Furthermore, the Rehabilitati on Act of 1973 states that children with physical or mental impairments` are eligible to receive reasonab le accommodations in school settings under the Section 504 Accomm odation Plan. School psychologists could play a role in the development and imple mentation of these accommodation plans to support the unique needs of st udents with chronic conditions An important difference between the IDEA statute and the Section 504 st atute is the manner in which disabilities are defined, with Section 504 having a much broader definition and encompassing a greater number of student s with disabilities. Bannon and Ross (1998) found that educat ors typically have an inadequate understanding of chronic childhood illnesses such as epilepsy. This is problematic given that the initial iden tification of seizure-related sy mptoms, such as unusual staring, blinking, and head drops, is commonly ma de in the school se tting (Sachs & Barrett, 1995). Therefore, it is impera tive that educators are knowle dgeable about the symptoms and the treatment of epilepsy, as well as the impact of the condition on a childÂ’s academic, behavioral, and psychosocial func tioning. Educators should be aware of
7 studentsÂ’ specific condition, including their pharmacological and psychological treatment, and potential positive and negative effects of the treatment. Furthermore, school psychologists, especi ally those with expe rtise in pediatric health issues, are in a unique position to act as a liaison among families, educators, and medical staff. As such, school psychologists are able to advocate for students with chronic health conditions to help alleviate ps ychosocial problems, as well as to improve their medical compliance and enhance their in tegration into the school setting (Sachs & Barrett, 1995). This can be accomplishe d by the school psychologist providing psychoeducation about epilepsy to the individua l student, classmates and peers, and the entire school staff. Additionally, school ps ychologists can conduct individual or group therapy sessions to address a variety of di fficulties the student with epilepsy may be experiencing. These difficulties may include problem solving deficits, coping deficits, social skills deficits, and/or internalizing (e.g., anxiety and depression) and externalizing (e.g., aggression, rule breaking) behaviors. Summary Psychopathology typically is a condition th at persists from childhood to adulthood (Hofstra, Van der Ende, & Ver hulst, 2000). Therefore, it is imp erative that children with health conditions and comorbid psychiatric disorders are tr eated for both conditions as they are likely to have a si gnificant negative impact on ch ildrenÂ’s overall HRQL. More specifically, it is important to identify the va riables that mediate HRQL in children with epilepsy given that research ers have found that the pres ence of psychopathology also negatively impacts childrenÂ’s HRQL (Basti aansen et al., 2004; Sawyer et al., 2002). Research has shown that biomedical variable s such as seizure severity and frequency
8 have only moderate relationships with HRQL ; therefore, additional factors, such as depression and anxiety, must be identified so that they also may be a focus of treatment (Wagner & Smith, 2006). The awareness of possi ble mediating factors, such as seizurerelated variables and psychopathology, may aid pr actitioners in the ea rly identification of children with epilepsy who are at-risk for poor HRQL. Purpose of the Study The purpose of this study was to ascert ain the relationship among seizure-related variables, health-re lated quality of life, and ps ychopathology (i.e., anxiety and depression) in children with ep ilepsy. The seizure-related variables that were examined in this study include type of seizure, seizur e frequency, and seizure treatment with antiepileptic drugs (AEDs). Specifically, type of se izure was classified as either generalized or partial. Seizure frequency was classified into two groups: 0-5 seiz ures per year or 6-12 seizures per year. Treatment with AEDs wa s classified as either monotherapy (i.e., treatment with one AED) or polytherapy (i.e., treatment with more than one AED). It was hoped that results from this study would in form both the medical and psychosocial treatment children with epilepsy receive. This comprehensive care needs to go beyond simply trying to control seizures with mi nimal adverse drug reactions. Seizure frequency and severity is only one importa nt outcome variable. Other factors, such as psychological disorders, also affect children with epileps y along with their families and close social networks. This study further contributes to the literature underscoring the importance of identifying, diagnosing, and treating child ren with epilepsy who have comorbid psychopathology so that they have the best possible psychosocial outcomes.
9 Research Questions The following research questions were addressed in this study: Research Question 1. What seizure-related variables (i .e., type of seizure, seizure frequency, and treatment with AEDs) and ps ychopathology (Â“at-ris kÂ” or Â“clinically significantÂ” range for anxiety an d/or depression) best predict health-related quality of life as reported by children 8 to 11 years of age diagnosed with epilepsy? Research Question 2. What seizure-related variables (i .e., type of seizure, seizure frequency, and treatment with AEDs) and ps ychopathology (Â“at-ris kÂ” or Â“clinically significantÂ” range for anxiety an d/or depression) best predict health-related quality of life as reported by parents of children 8 to 11 y ears of age diagnosed with epilepsy? Hypotheses Two research hypotheses were tested in the current study. First, it was hypothesized that children with ge neralized seizures (as compar ed with partial seizures), frequent seizures (i.e., 612 per year compared with 0-5 per year), who receive polytherapy (i.e., more than one AED co mpared with monotherapy) and obtain T scores of 60 and higher on the depression scale and/or anxiety scal e of the BASC ( T scores of 60 to 69 are in the Â“at-riskÂ” range and T scores above 69 fall in th e Â“clinically significantÂ” range) would obtain a HRQL score in the Â“h ighly compromisedÂ” (i.e., 5-10) range as reported via parent report. Second, it was hypothesized that children with generalized seizures (as compared with partial seizures ), frequent seizures (i.e., 6-12 per year compared with 0-5 per year), who receive polytherapy (i.e., more than one AED compared with monotherapy) and obtain T scores of 60 and higher on the depression scale and/or anxiety scale of the BASC ( T scores of 60 to 69 are in the Â“at-riskÂ” range and
10 T scores above 69 fall in the Â“clinically signi ficantÂ” range) would obt ain a HRQL score in the Â“highly compromisedÂ” range (i.e ., 5-10) as reporte d via self-report. Significance of the Study This study provides valuable informa tion about the relationship among seizurerelated variables, psychopathology, and health -related quality of life in children with epilepsy. These data will allow practitioners both psychologists and physicians, to assess and treat children with epilepsy by addressing the factors that are correlated with poor health-related quality of life. In addition, this stu dy sheds light on the relationship between health-related quality of life and ps ychopathology (i.e., depr ession and anxiety) in children with epilepsy. Because the re search clearly suppor ts the importance of addressing comorbid psychopathology in children with epilepsy, it is critical that these children are identified, diagnos ed, and treated to enhance th eir overall development. Definition of Terms Antiepileptic Drugs (AEDs). AEDs are a category of drugs that can help control the frequency and severity of seizures; they also are known as anticonvulsants (Kaiser, 2002). Â“At-riskÂ” range for anxiety. For the purposes of this ma nuscript, this phrase refers to children with epilepsy who fall into the Â“at-riskÂ” range (T score between 60-69) for anxiety per scores obtained on the child an d/or parent version of the Behavioral Assessment Scale for Children, Second Ed ition (BASC-2, Reynolds & Kamphaus, 2004). Please note: Scaled scores in the at-risk range on the BASC-2 are between one and two standard deviations from the mean and may signify developing problems that should be monitored.
11 Â“ At-riskÂ” range for depression. For the purposes of this manuscript, this phrase refers to children with epilepsy who fall into the Â“at-riskÂ” range ( T score between 60-69) for depression per scores obtained on the ch ild and/or parent ve rsion of the BASC-2 (Reynolds & Kamphaus, 2004). Â“ Clinically significantÂ” range for anxiety. This phrase refers to children with epilepsy who fall into the Â“clin ically significantÂ” range ( T score between 70-100) for anxiety per scores obtained on the child and/or parent version of the BASC-2 (Reynolds & Kamphaus, 2004). Please note: Scaled scores in the clinically significant range on the BASC-2 are two standard deviations or more from the mean and indicate a high level of maladaptive behavior. Â“Clinically significantÂ” range for depression. This phrase refers to children with epilepsy who fall into the Â“clin ically significantÂ” range ( T score between 70-100) for depression per scores obtained on the child and/or parent version of the BASC-2 (Reynolds & Kamphaus, 2004). Epilepsy Epilepsy is a chronic neurological condition with abrupt disturbances in the electrochemical firing of neurons in the brain cause seizures. This condition of unprovoked, recurring seizures a ffects the central nervous sy stem causing change in an individualÂ’s sensation, awaren ess, and/or behavior and has various origins, symptoms, courses, and prognoses (DuLac et al., 1995; Shafer, 2002). Monotherapy. The term monotherapy refers to the use of one antiepileptic drug (Labiner & Ahern, 2002). Polytherapy. The term polytherapy refers to th e use of more than one type of antiepileptic drug (Labiner & Ahern, 2002).
12 Psychopathology. Psychopathology refers to the study of the manifestation of behaviors and experiences that may be i ndicative of mental ill ness or psychological impairment (Caplan et al., 2005). Seizure A seizure is an individual episode in which an abrupt discharge of electrical activity in the br ain may cause changes in se nsation, behavior, and/or consciousness (DuLac et al., 1995; Ho-Turner & Bennett, 1999). Health-Related Quality of Life (HRQL) Health-related quality of life represents the functional effects of an illness and its resu lting therapy on a patient, as perceived by the individual patient (Baker, 2001).
13 Chapter 2 Review of the Related Literature Overview This chapter provides a review of the lit erature relevant to this study. Pediatric epilepsy is discussed, including its relevance to school psychology, a nd the importance of working within an interdisciplinary framework so that the Â“whole childÂ” is treated and the medical, educational, and psychosocial impact of the disease are addressed. Furthermore, prevalence rates, classification, types of seizures, diagnosis, and potential causes of epilepsy in children are presente d. Pharmacological treatment of pediatric epilepsy also is covered, including use of antiepileptic drugs. Next, seizure-related variables that may negatively impact ch ildren with epilepsy are discussed. The relationship between pediatric epilepsy and psychopa thology (specifically depression and anxiety) and the importance of assessing and treating psychopat hology in conjunction with medical/seizure treatment also is delineat ed. Health-related quality of life (HRQL) is reviewed, including various epil epsy-specific HRQL measures. The chapter ends with a discussion of the importance of assessi ng both the presence of psychopathology and HRQL when developing and implementing tr eatment plans for children with epilepsy. Defining Epilepsy Epilepsy is a neurological condition that affects the nervous system; it is a condition of recurrent seizures. It is important to differentiate Â“seizuresÂ” from Â“epilepsyÂ” from the onset of this review. In contrast to epilepsy, a seizure is an individual episode in
14 which an abrupt discharge of electrical ac tivity in the brain may cause changes in sensation, behavior, and/or consciousness (DuLac et al ., 1995; Ho-Turner & Bennett, 1999). Seizure disorders (i.e., epilepsy) are chronic, recurring disturbances in the electrochemical firing of the neurons in th e brain (Ho-Turner & Bennett, 1999). Epilepsy is a cluster of disorders that have various origins, symptoms, courses, and prognoses (Shafer, 2002). Seizure disorder and epileps y are synonymous terms that may be used interchangeably in this document. Prevalence of Epilepsy in Children Epilepsy occurs in approximately 5 out of 1,000 children and is considered to be the most prevalent neurological disorder of childhood (Black & Hynd, 1995; Hiemenz et al., 1999; Ronen et al., 2003). Although some se izure disorders self-limit as children mature, it is reported that more than 80% of adults with epilepsy had their first seizure during childhood (Ho-Turner & Benn ett, 1999). Incidence and pr evalence data are useful in developing hypotheses about the cause of an illness. Furthermore, this information aids in assessing the health care requirements in a population and in determining diagnostic probabilities (Cowan, 2002). When studying epilep sy, however, it is difficult to compare incidence and prevalence rates among vari ous populations because of the lack of homogeneous methods for defining epile psy (Ho-Turner & Bennett, 1999). The incidence of epilepsy in children ha s been estimated from diverse populations using different methods of case definition. For example, when examining recurrent, unprovoked seizures, the annual incidence rates per 100,000 child ren ranged from a low of 41.0 in Nova Scotia (Camfield, C., Camf ield, P., Gordon, Wirrell, & Dooley, 1996) to 82.3 in Northern Sweden (Blom, Heijbel, & Bergfors, 1978). Cowan (2002) reported that
15 the annual incidence rates acros s populations are quite sim ilar (i.e., approximately 50-72 per 100,000), particularly when considering the vast differences in methods of obtaining case studies. Despite the vast differences in bo th methodology and populations used in determining prevalence rates of childhood epilepsy, the majority of estimates are approximately 4 to 5 per 1,000 children (Eri ksson & Koivikko, 1997; Kurtz & Tookey, 1998). Kurtz et al. (1998) found that preval ence rates ranged from 2 to 3 per 1,000 in children up to 7 years of age. However, thes e researchers found that prevalence rates tend to increase with age, with a rate of approximately 4 to 6 per 1,000 in children 11 to 15 years of age (Kurtz & Tookey, 1998). Furthermor e, rates appear to be somewhat higher in males than in females (Eriksson & Koiv ikko, 1997). Differences in rates related to ethnicity also have been reported in the literature. Murphy, Yeargin-Allsopp, and Decoufle (1995) found slight ly higher prevalence rate s among African-American children compared to Caucasian children. Furt hermore, differences in prevalence rates have been found in developed versus devel oping countries. For example, Latin America and Africa have been found to have signifi cantly higher prevalence rates (e.g., 10 to 15 per 1,000) compared to developed countries. Th e higher rate in developing countries may be due to parasitic infections that ar e common in these countries (Cowan, 2002). Classification of Seizures It is difficult to define seizures clinically because there is an in finite variety in the clinical manifestations of seizures. However, the international classification scheme is the most widely accepted and utilized format for clinical classification of epileptic seizure
16 disorders (Chadwick, 1994). This classifica tion has two major divisions: (a) partial seizures, and (b) generalized seizures. Partial seizures occur in one or more restricted regions of the brain. These seizures are a secondary effect of a localized physiologic or structural abnormality in the brain such as a tumor, dysplasia, or tr auma (Prego-Lopez & Devinsky, 2002). Partial seizures originate loca lly in the cortex and are typically preceded by an aura (e.g., visual, auditory, or olfactory hallucin ations) that reflects the functi on of the area of the cortex where the seizure takes place. These seizures al so may be associated with post-ictal focal disturbances. Furthermore, partial seizures may spread to become generalized with a secondary tonic-clonic seizure (Chadwick, 1994). Partial seizures are further classified as either simple, complex, or secondarily generali zed. Simple partial seizures alter behavior but do not impair consciousness, whereas comple x partial seizures alter consciousness by impairing awareness, responsiveness, and memory (Prego-Lopez & Devinsky, 2002). Secondarily generalized seizures begin as pa rtial seizures (occurri ng in one area of the brain) and then quickly sp read throughout both hemispheres, becoming generalized seizures. In contrast to partial seizures, generali zed seizures occur bilaterally (i.e., in both hemispheres of the brain) with consciousness lost suddenly; theref ore, the patient does not experience an aura (Chadwick, 1994). These seizures are subcategorized into several main types, including generalized tonic clon ic, myoclonic, absence, and atonic seizures. These major types of generalized se izures will be defined briefly.
17 Types of Seizures Generalized tonic-clonic seizures These seizures also are called Grand Mal seizures. These are the most common type of ge neralized seizure that typically begin with a stiffening of the limbs (i.e., tonic phase) followed by a jerking of the limbs and/or face (i.e., clonic phase). Breathing may decrease or stop comple tely during the tonic phase, but typically return s (although sometimes irregular) during the clonic phase. The presentation of these seizures varies. For ex ample, some individuals may experience only the clonic or only the tonic phase. Furthermor e, some may experience a tonic-clonic-tonic seizure pattern (Ho-Turner & Bennett, 1999). Myoclonic seizures These seizures present with rapid, brief contractions of muscles that typically occur on both sides of the body; however, they may involve only one limb. Most of the epileptic syndromes that include myoc lonic seizures usually begin in childhood. These seizures occur in a variet y of epilepsy syndromes that have different characteristics such as juvenile myoc lonic epilepsy, Lennox-Gastaut syndrome, and progressive myoclonic epilepsy. Juvenile myoclonic epilepsy and Lennox-Gastaut syndrome will be presented in more detail later in this chapter. Absence seizures These seizures also are called petit mal seizures. Individuals who experience absence seizures may experien ce lapses of awareness that sometimes involve staring episodes that la st only a few seconds. Absence seizures are characterized by a brief impairment of consciousness, which usually lasts no more than a few seconds. The individual simply stares vacantly; neithe r speaking nor appeari ng to hear what is said. Then, as abruptly as it began, the impair ment lifts and the child continues with his or her previous activity. Absence seizures ar e more common in children than in adults;
18 however, they are frequently so short-lived th at they avoid detection, even when the child is experiencing numerous attacks on a daily basis. Because of the briefness of the episodes, it is all too common for children to experience these seizures for several months or even years before they are diagnosed (Williams et al., 2002). These seizures most frequently develop between 4 and 12 years of age, and very rarely begin after 20 years of age. Atonic seizures These seizures also may be referre d to as drop attacks, astatic or akinetic seizures. They produce an abrupt lo ss of muscle tone, which may include the head dropping, a loss of posture, or a sudden collapse. Because they occur abruptly, individuals who experience atonic seizures typically fall with force. Because this type of seizure may result in significant injury to the head and face, protective headgear is sometimes used by both children and adults. Unfortunately, these seizures tend to be resistant to drug therapy. Electroencephalographic (EEG) results may be used to help differentiate partial and generalized seizures. In patients with par tial seizures, inter-ictal (i.e., post seizure) EEG findings typically show localized spikes and occasionally associated focal slow waves. However, in patients with primarily generalized seizures, EEG findings reveal synchronous, high amplitude, generalized spike-wave discharge (Chadwick, 1994). Other Classifications of Seizure Classification by cause of seizure. Causes of epileptic seizures are generally categorized as either genetic, idiopathic (unknown), or cryptogeni c (poorly defined). The cause of idiopathic seizures is co mpletely unknown with no evidence of an underlying abnormality. An individual with epile psy is biologically typical, with the
19 exception of the occurrence of se izures. Idiopathic seizures ar e believed to be inherited and are defined by age-related onsets. The cau se of cryptogenic seizures is undetermined; however, the cause appears to be related to another neurological or cognitive condition (Ho-Turner & Bennett, 1999). In addition, seizures may be either developmental or acquired, as is the case with symptomatic seiz ures. Symptomatic seizures also are called reactive or provoked seizures--when in response to an irritation (e.g., fever or trauma to the brain). They have a known cause that may originate from such ailments as trauma, developmental aberrations, metabolic imbala nces, or fever (Ho-Turner & Bennett, 1999). Therefore, the cause of this type of seizure may be either developm ental or acquired. The various potential causes of epileptic syndromes will be detailed further in a later section of this chapter. Common Types of Seizures in Children There are several different types of ge neralized and partial seizures that are common in childhood. Each of these will be reviewed briefly in this section. The prevalence rates and common symptoms of the various types of seizures will be presented. Neonatal seizures. Neonatal seizures occur from birth up to approximately one month of age. The occurrence of these seizures is higher in infants with a familial history of epilepsy. Of the infants who experience be nign neonatal seizures that are inherited, approximately 14% will develop another childhood epilepsy syndrome (Ho-Turner & Bennett, 1999). However, the majority of in fants experience a spontaneous resolution of seizures after two months (Ho-Turner & Bennett, 1999).
20 Febrile seizures Febrile seizures are nonepile ptic; however, in some cases epilepsy may develop. These seizures occur in approximately 4% of children between 6 months and 5 years of age. These are the mo st common type of seiz ures in infancy and are caused by high fever (typi cally over 102.2 degrees Fahren heit) (Duchowny & Harvey, 1996). One-half of febrile seizures occur betw een 12 and 24 months of age, with a peak in occurrence between 18 and 24 months of age. The most important predictor of recurrence of these seizures is age at o ccurrence of first seizure (Duchowny, 1993). An infant who experiences the first febrile se izure at less than one year of age has approximately a 50% chance of recurrence, with the risk of recurrence decreasing as the child matures (Hulihan, 1997). Epilepsy has been reported to develop in approximately 1% to 10% of children who have a history of febrile seizures (Duchowny, 1993). Epilepsy is most likely to develop in children who have experienced complex febrile seizures. Duchowny (1993) reported that epilepsy develops in 6% of child ren with two or more features of complex febrile seizures; however, only 0.9% of child ren with none of these features develop epilepsy. Childhood absence epilepsy. This syndrome most often develops between 4 and 8 years of age. Children with th is type of epilepsy are typical ly neurologically normal, and therefore, they are less likely to experience generalized tonic-clonic seizures compared to those who experience other types of absence-seizur e disorders (Duchowny, 1993). Currently, there is no consensus regarding th e prognosis for indivi duals with childhood absence epilepsy. However, it is promis ing that these seizures generally respond favorably to anticonvulsant medications in approximately 80% to 95% of children.
21 Furthermore, childhood absence seizures typi cally abate by mid-a dolescence (Duchowny, 1993). It is still unclear whet her this type of seizure ha s a genetic component; however, there typically is a family history of seizures in patients who experience childhood absence epilepsy (Duchowny, 1993). Juvenile myoclonic epilepsy. This type of epilepsy is ch aracterized by brief, rapid jerks of the shoulders and arms. Many patie nts also have genera lized tonic-clonic seizures, and many also suffer absence seizures Age at onset is typically between 10 and 17 years. Myoclonic seizures may be precipit ated by sleep deprivation or alcohol use. Benign rolandic epilepsy (benign childhood epilepsy with centrotemporal spikes). This type of epilepsy is characterized by focal seizures with sensor imotor symptoms most commonly affecting the face and oropha rynx, although at times with secondary generalization to a tonic-clonic seizure. Being the most common epilepsy of childhood, it occurs in children who ar e otherwise neurologically normal. Seizures may be predominantly or exclusively nocturnal. The ch aracteristic EEG pattern of this disorder can assist in diagnosis in a child who presents with a focal seizure. The same EEG pattern has been found in many first-degree relatives of patients with this disorder, indicating a strong hereditary component (Zupanc, 1996). Virtually all cases of benign rolandic epilepsy remit by mid-adolescence. Benign occipital epilepsy. Also known as epilepsy with occipital paroxysms, this is a less common type of idiopathic focal epilepsy. These seizures originate in the occipital lobe of the brain and have the potenti al to generalize to a tonic-clonic seizure. Because approximately 47% of children have a fa mily history of seizure, it is believed that there is a genetic component for these seizures (Duchowny, 1993).
22 Infantile spasms Infantile spasm (IS) is a type of seizure observed in an epilepsy syndrome of infancy and early childhood also known as West Syndrome This syndrome is typically detected early in life (i.e., be tween 3-6 months of ag e). Infants with IS commonly present with a sudden bending forw ard and stiffening of the body, arms, and legs. The spasms tend to occur soon after waking and each spasm typically lasts for 1 to 5 seconds. These spasms occur in clusters, and range from 2 to as many as 100 spasms at a time. Infantile spasms usually remit by 5 years of age, but these spasms are frequently replaced with other types of seizure. Along with the infant ile spasms, individuals with West Syndrome also experience hypsarrhythmia (chaotic brain wave patterns) and mental retardation. A close relationship has been detected between IS and Lennox-Gastaut Syndrome, which is an epileptic disorder of later childhood. Lennox-Gastaut syndrome. This syndrome is a severe condition characterized by mental retardation, multiple seizure types th at respond poorly to medication, and often a characteristic EEG pattern with slow sp ike-and-wave discharges. Many patients are treated with three or more drugs in combin ation, which may result in medication toxicity. The simplification of anticonvulsant regimens with reduction in the number of drugs used and the discontinuation of barbiturat es, may reduce frequency of seizures and ameliorate behavioral disturbances. Potential Causes of Childhood Epilepsy Cowan (2002) reported that approximately 55% to 75% of all epilepsy cases are of unknown cause (i.e., idiopathic), with only 25% to 45% being attributed to specific risk factors. There are a number of factors that appear to co ntribute to seizure threshold (an individualÂ’s propensity to experience a se izure) in childhood epilepsy. Ho-Turner and
23 Bennett (1999) identified three main factors th at affect seizure occurrence: (a) genetic predisposition, (b) seizure threshold, and (c) environmental stressors. It has long been established that cer tain types of childhood epilepsy have a genetic or hereditary com ponent (Ho-Turner & Bennett, 1999). A family history of epilepsy appears to increase the risk of developing epilepsy by two to three times (Hauser, Annegers, & Rocca, 1996). The reason is unclear, but the risk of developing epilepsy is greater in the children of mothers wi th a history of epilepsy than in children of fathers with epilepsy (Annegers et al.). More evidence for a genetic link to the development of epilepsy was found in a study conducted by Kjeldsen, Kyvik, Christensen, & Friis, (2001). The research ers studied 11,900 pairs of Danish twins to ascertain the genetic and environmental factors to the etio logy of epilepsy. A concordance rate of .37 for monozygotic tw ins and .08 for dizygotic twins was found for epilepsy. In summary, the researchersÂ’ analyses of the data suggested that approximately 70% to 88% of the proclivity for developing ep ilepsy may be explained by genetic factors (Cowan, 2002). Age alone has a significant influenc e on an individualÂ’s susceptibility to developing epilepsy. Newborns typically have a high cortical threshold; therefore, production of a seizure response is quite difficult (Ho-Turner & Bennett, 1999). However, as children mature (e.g., around 2 years of age), they develop a greater susceptibility to certain type s of seizures. Therefore, th e majority of childhood epilepsy syndromes are first observed around this chr onological age. Seizur e threshold continues to increase proportionally throughout childhood and adolescence until adult levels are reached (Ho-Turner & Bennett, 1999). Becau se an infantÂ’s brain is continually
24 developing through adulthood, to understand ch ildhood epilepsy it is important to study the developing brain at various stages. More over, there are different implications and manifestations of epilepsy depending on the stag e of cortical development (i.e., the age of the child), as well as differences in pr ognosis and treatment (Ho-Turner & Bennett, 1999). Environmental stressors that may trigger a seizure episode include such factors as high fever, extreme fatigue or excitement, and metabolism rates (Ho-Turner & Bennett, 1999). Furthermore, metabolic disorder s, hypoxia (i.e., insufficient blood oxygen), infectious diseases, and high fevers have been known to produce reactive seizures. Additionally, chemical imbalances, drug or poison ingestion, congenital abnormalities (e.g., aneurysms), or traumatic brain injuries may lead to seizure (Cowan, 2002; DuLac et al., 1995; Heller, Alberto, & Forney, 1996). Over all, in children who survive central nervous system infections, the risk of e xperiencing subsequent unprovoked seizures is increased substantially (approximately th ree-fold). Approximately 50% of childhood epilepsy cases go into remission and do not re quire long-term treatm ent, whereas around 25% are intractable and have a less pr edictable outcome (DuLac et al., 1995). Diagnosis of Epilepsy in Children Because of the variety of factors involve d, it can be difficult to obtain an accurate diagnosis of epilepsy. The childÂ’s medical hist ory must be obtained to help determine the cause of the episode. For example, a histor y of meningitis, encepha litis, head trauma, cancer, or cerebrovascular disease could indicat e the origin of the se izure. A review of the child's medications also should be completed because some medications are commonly implicated in seizure. For instan ce, some antipsychotic drugs (especially
25 clozapine and phenothiazines), radiocontrast dyes, alkyla ting agents, and -lactam antibiotics are commonly implicated (P rego-Lopez & Devinsky, 2002). Furthermore, general anesthetics, tricyc lic antidepressants, newer an tidepressants (e.g., selective serotonin reuptake inhibitors [SSRIs], bupropi on hydrochloride [Wellbutrin]), -blockers, and decongestants also can cause seizure. Conversely, a childÂ’s medical history may indicate a condition other than seizure. A hi story of other neurologi cal disorders in the child or childÂ’s family members may aid in differential diagnosis (Prego-Lopez & Devinsky, 2002). Treatment of Pediatric Epilepsy Initiating Treatment Once an individual is diagnosed with ep ilepsy, a decision must be made regarding treatment. An initial factor in developing a tr eatment plan is the child's estimated risk of seizure recurrence. Hart and Ea ston (1986) reported that patients with epilepsy are apt to experience a second seizure w ithin six months after the first episode. A number of researchers have studied the risk of recu rrence after the initial seizure. A 5-year recurrence risk of 34% was found by Hauser Rich, and Annegers (1990), whereas Hopkins, Garman, and Clark (1988) found a 3-year recurrence risk of 52%. The strongest predictors of seizure recurrenc e are the initial cause of se izure, focal abnormalities, and epileptiform abnormalities found on the EEG (Prego-Lopez & Devinsky, 2002). Seizure recurrence also may be related to type of seiz ure. For example, after experiencing the first tonic-clonic seizure, depending on additional risk factors, the recurrence rates fluctuate from 15% to 60%. After two tonic-clonic seiz ures, the risk of experiencing another seizure increases to approximately 85% (H auser, 1986). Acute symptomatic seizures,
26 which occur immediately after brain insult, ha ve an increased risk of recurrent seizure (Prego-Lopez & Devinsky, 2002). In addition, family history of seizure disorders, abnormal patterns on an EEG, or a history of neurologic insult increases the risk of recurrence after the first seizure (H auser, Anderson, & Loewenson, 1982). Treatment with Medication A study conducted by the First Seizure Trial Group (1993) investigated 397 patients with seizure disorder. The researchers found that the risk of seizure recurrence in untreated patients was 2.8 times higher than in patients treated with antiepileptic drugs (AEDs). The initial selection of medication to treat seizures is based on a number of factors that include seizure type, EEG results concomitant medications, and past medical history (Hulihan, 1997). When prescribing AEDs, it is important to be cognizant of the many side effects that may accompany the AEDs. Some childre n may experience feelings of sedation, psychomotor slowing, and/or gastrointestin al upset (Prego-Lopez & Devinsky, 2002). In rare cases, AEDs can cause severe side effect s such as liver failure, bone marrow failure, and/or pancreatitis (Prego-Lopez & Devi nsky, 2002). Furthermore, some AEDs are known to be more likely to induce seizures in patients than other AEDs; therefore, it is imperative that medications are carefully se lected for each individual depending on the patientÂ’s own unique set of factors. Additio nally, several medications have been known to initiate and/or exacerbate seizure ep isodes. Psychotropic drugs, such as certain antidepressants and major tranquilizers are most commonly indicated, and a majority of tricyclic and nontricyclic antidepressant drugs have been linked with an increase in seizures as well (Hulihan, 1997).
27 Use of AEDs: Monotherapy versus Polytherapy For each individual child, it is important to weigh the benefits and risks of pharmacological treatment. The Collaborative Group for Epidemiology of Epilepsy (1986) found that approximately 33% of patients who receive long-term antiepileptic treatment experience adverse reactions. Monothe rapy (i.e., use of one antiepileptic drug) is thought to be tolerated better than polyt herapy (i.e., use of multiple antiepileptic drugs). However, monotherapy has been report ed to cause problems in 20% of patients (Labiner & Ahern, 2002). Very young children app ear to have increased susceptibility to adverse medication effects (Ronen et al., 2003), as well as children with other medical problems and those taking concomitant medications. These adverse symptoms often lead to treatment failure; therefore, children s hould be monitored for compliance and adverse reactions to pharmacotherapy. Evidence suggests that early identificati on and diagnosis (soon after the onset of the initial seizure) improves outcomes for these at-risk children (Ronen et al., 2003). Early diagnosis and pharmacotherapy may inde ed reduce seizure recurrence as well as decrease the number of an tiepileptic drugs needed (i .e., monotherapy rather than polytherapy). Additionally, early diagnosis and treatment is likely to minimize the impact of epilepsy on a child's overall health-related quality of life. Psychopathology and Epilepsy It has been well documented in the literature that children with epilepsy are at-risk for psychopathology (McDermott, Mani, & Krishnaswami, 1995; Rodenburg, Stams, Meijer, Aldenkamp, & Dekovic, 2005). A preval ence rate ranging from 21% to 60% for psychopathology in children with epilepsy has been reported--a three to six times higher
28 risk than that of the general population (Caplan et al., 2004; Devinsky, 2003). An epidemiological study conducted by McDerm ott, Mani, & Krishnaswami (1995) found that children with epilepsy are at a 4.7 tim es greater risk for psychopathology compared to children from the general population. This pr evalence rate is substa ntially higher than rates in children with other chro nic illnesses (Wagner & Smith, 2006). Potential Causes of Psychopathology in Pediatric Epilepsy Relationship between seizure type and psychopathology. There are many factors that contribute to psychosocial difficulties in children with epilepsy, such as features of the actual seizures. The type, location, and frequency of the seizures may have a strong correlation with psychosocial pr oblems. For example, if the seizures are located in the temporal lobes or limbic structures, they may directly affect emotions and coping, because these are crucial areas in the brain th at control these emotions. Furthermore, if seizures are located in cortical areas, an individualÂ’s cognitive and/or physical functioning may be adversely affected. Therefor e, it is critical to di agnose and locate the focus and type of seizure because seiz ures significantly impact the child, both neurologically and psychosocially (Duchowny, 1993; Ho-Turner & Bennett, 1999). Relationship between seizure frequency and psychopathology. Seizure frequency is another variable that has been examined by researchers to determine its impact on psychopathology in patients with epilepsy. Seve ral researchers (e.g., Austin et al., 1992; Lambert & Robertson, 1999) found that seizur e frequency contributed to behavioral difficulties in children with epilepsy. Although to date there is not a sufficient amount of data to ascertain the relati onship between seizure frequenc y and the presence of anxiety,
29 there are some data that indicate significan t stress and anxiety can escalate the frequency of seizures (Vazquez & Devinsky, 2003). Relationship between treatment with antiepileptic drugs and psychopathology. Fortunately, antiepileptic drugs often can aid in the management of seizures; unfortunately, seizure control alone does not ameliorate the negative affects of epilepsy on an individualÂ’s quality of life. Moreover, th ere is a plethora of research findings that support the negative side effect s of antiepileptic drugs. Re searchers who have studied AED treatment in children with epilepsy have found that behavioral si de effects are most frequently reported. Several researchers (Fiordelli, Beghi, Bogliun, & Crespi, 1993; Mendez et al., 1993) found that the use of multip le AEDs (i.e., polytherapy) is associated with depression in children with epilepsy. O guz, Kurul, and Dirik (2002) examined the relationship of epilepsy-related factors to a nxiety and depression sc ores in children and adolescents, 9-18 years of age, diagnosed w ith epilepsy. Results indicated that children and adolescents who were treated with mo re than one AED obtained depression and anxiety scores that were signi ficantly higher than the childr en and adolescents who were treated with one AED. Currently, the research examining th e relationship between seizure-related variables and psychopathology is both limited an d inconclusive. Mixed results have been found on the relationship among a number of seizure related va riables and their relationship with psychopathology. For exam ple, some researchers have found that epilepsy duration, seizure frequency, and polytherapy are related to psychiatric disturbances (e.g., Cramer, Blum, & Reed, 2003; Oguz et al., 2002), wh ereas others have not generated significant findings (e.g., Etti nger, Weisbrot, Nolan, Gadow, & Vitale et
30 al., 1998). Because there is little conclusive evidence in this area, it is important for researchers to determine the impact of specific seizure-related variables on psychopathology so that preventative plans may be put in place for children at-risk of developing psychiatric diffi culties and/or poor HRQL. Neurological Dysfunction and Psychopathology. It has long been hypothesized that neurological dysfunction is the primar y cause of psychopathology in children with epilepsy. Researchers have found that children with neurological disorders have elevated levels of psychopathology compared to childr en with disorders that do not involve the nervous system (Lavigne & Faier-Routman, 1992). This may be due, in part, to the reality that children with epilepsy have the risk a ssociated with chronic illness and the risk associated with a central nervous system disorder, combined. Austin et al. (2002) reported that children with epilepsy have a 2.5 times higher risk for psychopathology than do children whose chronic illnesses do not involve the central nervous system. Response to epilepsy and psychopathology. An individualÂ’s re sponse to seizures has been compared to the concept of lear ned helplessness because individuals with epilepsy continually experience aversive events (i.e., seizures) over which they have no control (Lambert & Robertson, 1999). The se izure episodes may produce extreme fear and embarrassment in the individual with epilepsy. Moreover, these negative emotions may lead to the development of an external lo cus of control, which in turn may lead to anxiety and/or depression (H ermann, Seidenberg, & Bell, 200 0). It is important to note that underlying psychiatric illness may contribut e to a lower quality of life in children with epilepsy.
31 There are a number of negative emo tions that children with epilepsy may experience in relation to thei r neurological condition. For exam ple, fear and anxiety have been identified frequently in children ne wly diagnosed with epilepsy (Austin, Smith, Risinger, & McNelis, 1994). Fear and anxiety can present si gnificant emotional burdens on children with epilepsy, and these internal emotions often go unnoticed (Shafer, 2002). Children diagnosed with seizure disorders ma y have a number of fears related to their epilepsy, including the fear that they will di e during a seizure episode, that they will suffer brain damage, and/or they will experi ence a seizure episode in public. Children and adolescents may fear an overall loss of control. That is, th ey may fear a loss of control related directly to th eir medical condition (e.g., over bodily functions during a seizure) as well as a loss of control over their social development and relationships (e.g., losing friends). Comorbid Disorders: Epilepsy and Internalizing Disorders Given that epilepsy, anxiet y, and depression are all rath er common disorders, it is not surprising that for a number of ch ildren, these conditions coexist (Jackson & Turkington, 2005). Therefore, it is problematic that few researchers have examined the mechanism of depression and anxiety in ep ilepsy and even less attention has been focused on the treatment of these comorb id disorders (Jacks on & Turkington, 2005). There is consensus among several re searchers who have found mood disorders (e.g., depression) in 12% to 26% of children w ith epilepsy (Caplan et al., 2005; Oguz et al., 2002; Rodenburg et al., 2005). The reporte d prevalence of mood disorders varies depending on a multitude of factors, including instruments used for assessment (e.g., self-
32 report measures, psychiatric interviews), t ype of informant (e.g., parent, teacher, and child), age range of child, and samp le sizes of research studies. The prevalence of comorbid epileps y and psychopathol ogy is significant, especially compared to othe r chronic illnesses. Davies Heyman, and Goodman (2003) examined the rate of emotional disorders in children with epilepsy, diabetes, and healthy children. Records from the Child Benefit Re gister (CBR) were used to obtain a representative sample of 10,316 children between 5 and 15 years of age throughout England, Wales, and Scotland. Through data collection from a main caregiver and teacher, 67 children with epilepsy (mean age 10 years 2 months), 47 children with diabetes (mean age 10 years 4 months), and 10,202 controls (mean age 9 years 11 months) were identified. The researchers derived DSM-IV psychi atric diagnoses from the Development and Well-Being Assessment along with data ob tained through clinical interviews. Rates of psychiatric disorder were 37% for epile psy, 11% for diabetes, and 9% for medically healthy children (i.e., without chronic medi cal conditions). Furthermore, parents of children with epilepsy reported more emotiona l and behavioral problems compared to parents of children with diabetes and parent s of controls. Regression analyses revealed that epilepsy was independently associated with all behavioral variables, but this association was not found with children with diabetes. Therefore, the researchers concluded that emotional and behavioral problems are freque nt in children with epilepsy and that there is a need for e ffective mental health services for this population (Davies et al., 2003).
33 Caplan et al. (2005) examined affec tive disorders, anxiety disorders, and suicidality in children with complex partial seizure disorder and absence epilepsy. Seizure-related, cognitive, linguistic, family history, social competence, and demographic variables and their association with ps ychopathology also were examined. The study involved 171 epilepsy patients; 100 children with complex partial seizures, 71 with childhood absence seizures, and 93 medically hea lthy children, all 5 to 16 years of age. Affective and anxiety disorders were present in 33% of children with epilepsy compared to only 6% in the healthy group. Furthermore, 20% of children with epilepsy had suicidal ideation compared to 9% of children in the healthy group. When examining the prevalence broken down by specific disorder, 63% of children had anxiety disorders, and 26% had comorbid affective/anxiety and disrup tive disorders. Caplan et al. (2005) noted that only one third of the children who were diagnosed with a psyc hiatric disorder had previously received psychiatric assessment or treatment. This finding highlights the lack of attention to psychiatric disorders in ch ildren with epilepsy. Even more remarkable, results from this study revealed that only one third of children w ith suicidal ideation received any type of psychiatric care. Caplan et al. (2005) found that children with epilepsy who also had affective and anxiety disorders obtained significantly hi gher mean scores on the Child Behavior Checklist (CBCL) internalizing and anxiety/ depression factor scores. In addition, children with epilepsy had significantly high er scores on the Child Depression Index (CDI) and the Multidimensional Anxiety Scale for Children (MASC). Furthermore, children with complex partial seizures had si gnificantly higher ra tes of depression and comorbid depression and anxiety compared to children with absence epilepsy.
34 Conversely, children with absence epilepsy had significantly higher rates of anxiety disorders as compared to children with complex partial seizures. In summary, this study was the first to identify a high rate of both affective and anxiety disorders (i.e., 33%) and a high rate of suicidal ideation (i.e., 20%) in children with epilepsy. Caplan and colleagues reported that only 33% of the children with epilepsy identified with affective disorders, anxiety di sorders, and/or suicid e ideation had received mental health services. Because of this fi nding, along with the high rates of depression, anxiety, and suicide in adults with epilepsy, and the age-related increase of suicide in adolescence, it is critical that psychopathology is identified an d treated early to lessen the negative impact of these conditions on child ren with epilepsy (Caplan et al., 2005). Several meta-analyses were conducted by Rodenburg et al. (2005) to examine the types and severity of psychopathology in children diagnosed with epilepsy. These researchers examined which types of ps ychopathology were most prevalent when children with epilepsy were compared to children with a different chronic illness to ascertain whether psychopathology is generic to chronic illness or specific to epilepsy. Children with epilepsy were compared to four control groups (i .e., normative groups, healthy study controls, children with a chronic illness, and siblings). A multi-informant perspective was utilized, including parent re ports, teacher reports, and self-reports. Fortysix studies met inclusion criteria for the meta-analyses. Data were analyzed from a total of 2,434 children with epileps y, with the average sample size of 65 for children with epilepsy. The mean age of children was 10 year s, with a range of 4 to 21 years of age (Rodenburg et al., 2005).
35 To examine the severity of psychopa thology in children with epilepsy, metaanalyses of studies compari ng children with epilepsy with children from the general population were conducted. Results from these an alyses revealed medium to large effect sizes (i.e., d = .57 to .61) for parent report, teacher report, and self-report. Moreover, large effect sizes were consistently found fo r differences between children with epilepsy and normative controls for the whole range of psychopathology. Meta-analyses of total behavior problems revealed large effect sizes for parent report and teacher report. Larger effect sizes (i.e., d = .45 ve rsus .23) were found on compar isons between children with epilepsy and normative controls on externaliz ing problems compared to internalizing problems. Furthermore, meta-analyses of tota l behavior problems revealed that effect sizes for differences between children with epilepsy and healthy study controls were medium (i.e., .15) for parent report and teacher report (Rodenburg et al., 2005). These findings confirm that children with epilepsy are at high risk for developing psychopathology. These children appear to be vulnerable to the whole range of psychopathology, including attention problems, thought problems, and social problems, as well as internalizing and externalizing beha vior. For parent report, teacher report, and self-report, when comparing ch ildren with epilepsy to children from normative groups, effect sizes were larger fo r internalizing than for exte rnalizing behavior problems. However, effect sizes for externalizing probl ems were still quite large for parent report and teacher report, indicating that externaliz ing problems also are frequently present in children with epilepsy. Researchers conclude d that their findings indicated that psychopathology in children with epilepsy may be associated partially with generic
36 factors related to chronic il lnesses. Therefore, psychopath ology may be partly diseasespecific--that is, partly attributed to epilepsy (Rodenburg et al., 2005). It is well documented that there is ofte n a delay in the diagno sis of internalizing disorders in children (e.g., anxiety and depres sion) because they lack overt behavioral symptoms that are easy to identify. This probl em is the same for children with epilepsy experiencing symptoms of anxiety and/or depression. It is comm on for internalizing disorders such as depression and anxiety to ma nifest differently in children compared to adults. That is, children may present with exte rnalizing behaviors, such as irritableness or aggressiveness. Therefore, it is important to administer broadband measures to assess a wide range of behavioral and emotional difficulties in children with epilepsy. Treating Children with Epilepsy and Comorbid Disorders Due to the complexities that may arise when treating epilepsy, the best approach for managing pediatric epilepsy is to utili ze a disease-based model that distinguishes epilepsy syndromes. Accurate diagnosis is essential to enhance medication selection, determine length of treatment, and to counsel children and family members regarding the overall impact of epilepsy and the likely prognosis. Children who ar e diagnosed with comorbid disorders are likely to be prescribed multiple medications for treatment. This is especially problematic for children with epilepsy because there are a number of medications prescribed to treat co-morbi d disorders that may exacerbate seizures (Hulihan, 1997). For instance, a number of over-the-counter cold medications may contain epileptogenic compounds that could induce seizure. Furthermore, nonprescription anorectics, as well as several antibiotics, have been linked with seizure episodes in nonepileptic children (Scheuer, 1992). This emphasizes the importance of close
37 monitoring of the patient and the measuremen t of drug levels. In most cases, children may safely use a combination of medications to treat multiple disorders (Hulihan, 1997). The treatment of children with comorbid epilepsy and depressi on is a challenging undertaking. Research findings indicate th at depressive episodes in children and adolescents tend to be quite lengthy, with an average length of 7 to 9 months. In addition, approximately 40% of individuals relapse (i.e ., have another depressive episode) within a two-year period (Plioplys, 2003). Therefore, the treatment of comorbid affective disorders requires a long-term commitment by both patient and caregivers. Although medication must be carefully monitored by a physician, antidepressa nts may be used to treat comorbid depression or anxiety in children and adolescents with epilepsy. According to Plioplys (2003), somatic tr eatment and psychotherapy are the main psychiatric treatment options for comorbid depression and epilepsy. It is critical to address comorbid psychi atric disorders in patients with epilepsy because these comorbid disorders present a profound impact on this population (Davies et al., 2003; Devinsky, 2003). The identification, diagnosis, and treatment of comorbid psychopathology can be quite challenging. Research findings have supported a relationship among comorbid psychopathol ogy, poor functional outcomes, and lower quality of life (Devinsky, 2003). However, rese archers have only recently begun to focus on this association in adults with epilepsy; this relationship in children with epilepsy has received even less attention (Hermann et al., 2000). Another diffi culty in addressing comorbidity in epilepsy is the possibility that the psychopathology may be treatmentrelated. For example, the psychopathology ma y be a consequence of treatment with AEDs (Devinsky, 2003).
38 Another factor that further complicates the treatment of comorbid disorders in epilepsy is that very little is known regard ing the pathogenic mechanisms underlying the comorbidity (Devinsky, 2003). Recent resear ch findings have pointed to a possible bidirectional relationship between psychiatri c disorders (e.g., depression) and epilepsy; however, further research is sorely need ed (Kanner & Palac, 2000). Although suitable treatment strategies are being developed with use of medications that have mood stabilizing effects, the most optimal treat ment plans remain unknown (Devinsky, 2003). Due to primary care providersÂ’ lack of clinical expertise in recognizing, diagnosing, and treating both epilepsy a nd psychopathology (Jackson & Turkington, 2005), collaboration with psychologists a nd psychiatrists likely would improve pharmacological treatment for these patients. One of the most si gnificant barriers to receiving appropriate and effective treatment for depression may be that primary care physicians and neurologists are reluctant to prescribe antidepressant s to their patients. This is because all classes of antidepressants are contraindicated in persons with epilepsy, and when prescribed, must be used with great caution in this population (Jackson & Turkington, 2005). AEDs that eff ectively control seizures wi thout the adverse effects on behavior and cognition, in combination with psychopharmacological treatment to address comorbid psychopathology, would be an ideal treatment option for children with epilepsy and comorbid psychopathology. Psychiatric side effects can be linked to all types of AE Ds used to treat individual patients with epilepsy. That is one group of AEDs has sedati on effects such as fatigue and cognitive slowing (Ketter, Post, & Th eodore, 1999). Another group of AEDs is an activating-type that tend to cause activati on, weight loss, and possibly antidepressant
39 effects. Because of the correlation betw een use of AEDs and psychopathology, it is critical that knowledge regarding the tolera bility of AEDs is increased, possibility through clinical trials and/or patient self-report m easures. To diagnosis and treat comorbidity in children with epilepsy accura tely and effectively, the administration of AEDs and the potential relations hip to psychiatric adverse eff ects must be considered. It is important to differentiate psychopathology that is independent of treatment of an underlying illness (e.g., epilepsy) and psyc hopathology that is a consequence of pharmacological treatment of an underlying disorder (Devinsky, 2003). In conclusion, deciding how to treat seizures can be a very complex task. There are a multitude of factors that can affect patients and their familiesÂ’ decisions regarding treatment. It is imperative that the patient and family understand both the benefits and risks of various treatment options. Depending on past experiences a nd cultural beliefs, people will have different attitudes and beliefs regarding treatment. It is important to involve the child in his or he r own treatment plan as much as possible. When the team (i.e., patient, family members, and practitione rs) is able to establish common goals and objectives, it is most likely that the treat ment will be successful (Shafer, 2002). Summary of Psychopathology and Epilepsy Although often unrecognized and untreate d, psychopathology (especially anxiety and depression) is common in children with epilepsy (Plioplys, 2003) It is unfortunate that, to date, the majority of research has focused on psychopathology in adults with epilepsy, and studies with children have been sparse. Children with epilepsy may have different types of risk factors, but these ri sk factors are equally significant to the risk
40 factors that adults endure. Just as adults with epilepsy need to adjust to their conditions, children also must adapt to the di agnosis and treatment of epilepsy. Epilepsy, compounded with a psychiatric comorbidity, further complicates pharmacological and psychosocial treatment planning. The benefits and risks to pharmacological treatment in ch ildren should be weighed carefully, and this is especially true when dealing with comorbidities. Furthe r research is needed to develop effective prevention and intervention plans for ch ildren with epilepsy and associated psychopathology. It is imperative that indivi duals with seizures are carefully screened by clinicians to detect any mood disturbances or ot her psychopathology (Shafer, 2002). Co-morbid psychopathology in children with epilepsy must be identified and treated with psychotropic medication and psychotherapy as needed. The high prevalence rates for comorbidity in children with epilepsy highlig hts the need for mental health services in pediatric epilepsy clinics. In 2003, the Epilepsy Foundation of America recommended that the development of eviden ce-based mental health treatm ent specifically for children with epilepsy take precedence in th e overall management of epilepsy. Health-Related Quality of Life and Epilepsy Attention typically has focused on seizure control in patients with epilepsy (Berg et al., 2001). However, research findings are accumulating that shed light on the tremendous psychosocial impact of epilepsy on patients (Devinsky, 2003; Ronen et al., 2003). As the primary goal in health care is the attainment of optimal physical and psychosocial well-being for patients, the im pact of chronic illness on psychosocial
41 functioning should not be overlooked. The cycl ical relationship betw een physical health and mental health must be addressed to promote overall health and well-being. Epilepsy can have a significant imp act on childrenÂ’s overall development, including learning, behavioral, social, and em otional domains. Historically, epilepsy and seizures have been stigmatized (Baker, 2001). This may be due to the fact that selfcontrol and predictability of behavior is expected in our society, and for individuals who experience seizures, this expected behavior is not met. Unfortunate ly, individuals with epilepsy cannot predict when their next seizur e will occur, and this uncertainty can make others around them uneasy. Given that the av erage adult does not have an adequate understanding of seizure disorders, epilepsy can be a scary condition with which to deal, especially for children. This presents a chal lenging situation for children diagnosed with epilepsy, especially in the school setting. Because all children, those with a nd without epilepsy, have difficulty understanding the disorder, it is crucial that educators inform st udents about epilepsy, including causes, symptoms, and treatment options. Fear and uncertainty surrounding the condition may lead peers to distance th emselves from children with epilepsy. Unfortunately, it is quite common for adults and same-age peers who are uninformed about epilepsy to segregate ch ildren with epilepsy from thei r peers. When children with epilepsy are not permitted to engage in age-a ppropriate activities with peers, their sense of control and overall quality of life may be significantly restrict ed (Shafer, 2002). The development of social relations hips is of the utmost importa nce for children; therefore, their exclusion from social activities can se verely jeopardize their emotional well-being and sense of self. Quality of life is an especi ally important health outcome to address in
42 children with epilepsy because they appear to be at special risk (Austin et al., 1994). Research has shown that epilepsy has a signi ficant negative affect on QOL, with a number of factors identified that may cause poor QOL, such as seizure severity, stigma, fear, and psychiatric problems (Shafer, 2002). Defining Health-Related Quality of Life (HRQL) Health-Related Quality of Life (HRQL) represents the functional effects of an illness and its resulting therapy on a patient, as perceived by the individual patient. HRQL may be defined as an individualsÂ’ em otional reaction to the illness and their life situation, or by their ability to meet pers onal needs (Baker, 2001) One of the most comprehensive definitions is provided by Schipper (1990). In the definition, the following five broad domains are covered: phys ical, occupational, psychological, social, and somatic. When treating children with epil epsy, it is imperative that all areas of functioning are addressed because all of th ese areas can significantly impact overall functioning. Therefore, an ecological approach to treatment is necessary. One of the chief aims of treating children with epilepsy should be to enable them to feel better and to improve their function in daily activities. Most clinicians acknowledge that it is cri tical to incorporate HRQL measures into routine clinical practice to en sure that patients with epile psy receive the best possible treatment (Ronen et al., 2003). It has been well-e stablished in the lite rature that chronic medical illnesses, to a certain degree, imp act patientsÂ’ quality of life (Baker, 2001). Additionally, researchers have f ound that individuals with epil epsy typically experience a more severe decline in their quality of life compared to individuals with other types of chronic illness. This may be due, in part, to the nature of epilepsy; the uncertainty of
43 when another seizure will occur, how to cont rol the seizures, and when--if ever--they will abate (Baker). Quality of life in children with chro nic conditions is a relatively new area of research (Calaminus, Weinspach, Teske, & Gobel, 2000). However, quality of life data has provided important insight for understanding the impact of a chronic illness on childrenÂ’s psychosocial functioning and deve lopment (Noll et al., 1999). To assess HRQL in children with epilepsy, several groups of researchers have investigated various factors that may contribute to a lowered hea lth-related quality of life. Although minimal research has been conducted measuring the relationship between epilepsy and HRQL, the two areas that researchers have studied include seizure-specific variables, and to a lesser extent, psychiatric disorders. The most common ly assessed seizure variables include type of seizure, seizure frequency, seizure severit y, and age at onset of seizure (Ronen et al., 2003). However, the relationships between thes e variables and level of HRQL are only moderate. The following sections will revi ew the literature related to both seizure variables and psychiatric disorders, and thei r particular influence on HRQL in children with epilepsy. Measuring HRQL in Children with Epilepsy Because the chief goal of management of children with epilepsy should be to reduce medical and psychosocial complications as much as po ssible, it is important that practitioners are aware of the childÂ’s percep tion of her/his quality of life. An efficacious instrument that measures HRQL is necessary to identify areas of functioning that are being negatively impacted so that comprehe nsive and effective treatment plans can be developed. Research has shown that factor s specific to seizures (e.g., frequency and
44 severity) are merely one pertinent outc ome variable (Ronen et al., 2003). Other dimensions, such as social, psychological and behavioral dime nsions also have significant impacts on children with epilepsy and their families. There have been several HRQL scales specific to epilepsy developed to measure the impact and burden of epilepsy on child ren (Ronen et al., 2003). However, few researchers have studied the score reliabili ty and validity of these measures. It is important that HRQL scales measure a multitude of factors, including the childÂ’s resilience, co-morbid conditions, and societal or cultural variables. The presence of these variables will impact each indivi dual patient to a different degree; however, all have the potential to impact significantl y patientsÂ’ HRQL on a daily basis. Ronen et al. (2003) recommend utilizing childrenÂ’s own perspec tives of their HRQL as well as parent reports. Once HRQL measures, both generic a nd epilepsy-specific, have been determined to be efficacious, they should be used routin ely by clinicians in their daily practices (Ronen et al., 2003). Austin et al. (1994) conduc ted a study examining quality of life in children with epilepsy and children with asthma. All childr en who participated in the study were diagnosed with either epilepsy ( n = 136) or asthma ( n = 134) and were between 8 and 12 years of age. Additional inclus ion criteria also required that the children were currently receiving medication for their conditions, ha d no other chronic physical conditions, and had at least average intellect ual functioning. Data were collected from the children, their guardians, and their teachers th rough a variety of assessment strategies (e.g., interviews, school records, and questionnair es). Participants were comp ared on four dimensions of
45 quality of life (physical, psychol ogical, social, and school) to determine the specific areas that are problematic for children with epilepsy. Data were analyzed via a 2 x 2 between-subjects multivariate analysis of covariance with type of illness as the independe nt variable and length of time since onset of illness as a covariate. The most signifi cant finding from the study was that children with epilepsy had a more compromised quality of life in the psychological, social, and school domains as compared to the group with asthma, despite a later age of disease onset and fewer illness episodes for the children w ith epilepsy compared to the children with asthma. Contrarily, children w ith asthma had a more compro mised quality of life in the physical domain (Austin et al., 1994). Findings revealed that th e physical domain had the lowest correlations with the other three domains. Medication side effect s also were significantly related to with internalizing and externalizi ng problems at home, and exte rnalizing problems at school. The difference in quality of life between groups suggests that difficulties in children with epilepsy are not exclusively the result of the chronic conditio n; poorer HRQL of children with epilepsy is at least somewhat specific to seizure disorders. It is hypothesized that another aspect of the i llness is related to poorer quality of life in the psychological, social, and school domains. The researchers hypothesize d that whatever caused the epilepsy may also directly affect coping behaviors a nd school achievement (Austin et al., 1994). A limitation to this study was that th e physical domain was not measured as comprehensively as the psychological, social and school domains; therefore, future research should investigate additional variables such as illness severity and use of antiepileptic medications. It also is a chal lenge for future researchers to ascertain a
46 stronger understanding of the relationship betw een learning problems and quality of life, such as the sequence of events to determin e whether these learning difficulties precede or follow the emergence of psychological and social problems. These findings suggest that a focus on seizure control will not address the full range of problems that children with epile psy experience. Furthermore, these results emphasize the need to ascertain the risk a nd protective factors fo r developing problems that compromise quality of life. This underscores the need for prevention and intervention programs that ameliorate psyc hological, social, a nd school performance problems in children with epilepsy. Research has revealed that seizure contro l plays only a minor role in the social adjustment of children; therefore, additio nal factors are likely contributors. Other researchers also have concl uded that although patients with frequent seizures had poorer psychosocial profiles than did those with infr equent or no seizures, important predictors of psychopathology and social dysfunction seemed to exist in the patients with refractory epilepsy that could not be explained by physic al or demographic da ta (Austin et al., 1994). Health-related quality of life measures. Both generic and condition-specific HRQL scales have been developed to meas ure quality of life in children. Generic measures such as the Child Health Questi onnaire (CHQ; Landgraf, Abetz, & Ware, 1996) and Pediatric Quality of Life Questionnaire (PedsQL; Varni, Seid, & Kurtin, 2001) were developed to assess a broad range of HRQL domains in individuals with chronic conditions, irrespective of the specific condi tion. However, because research has shown that children with epilepsy are more suscepti ble to a poorer QOL than children with other
47 chronic illnesses, use of epilepsy-specific scales may be warranted. Furthermore, a generic instrument may lack the sensitivity to detect subtle characteristics of specific conditions in a manner that provides meaningful information to patients and professionals (Ronen et al., 2003). Disease-specific measures for children have been developed for a number of beneficial reasons. For example, disease-spec ific measures are specifically designed to assess areas of functioning that are most likel y to be affected by a particular illness. Furthermore, they utilize items that are clin ically relevant and meaningful for children (Quittner, 1998; Spieth & Harris, 1996). Epilepsy-specific HRQL instruments are developed to assess characteristics of a par ticular condition. Therefor e, these scales are typically more relevant and se nsitive to the nuances of epile psy. However, a drawback to the condition-specific scales is that they addr ess a more restricted range of issues than generic instruments. Another limitation to the utilization of condition-specific instruments is that they often do not ha ve well-documented psychometric properties because they are less widely used than generic measures (Ronen et al., 2003). If researchers determine that the advantages to utilizing disease-speci fic HRQL instruments outweigh the potential disadvantages, it is important that the psychometric properties (validity, reliability) of these measures ar e assessed so that psychometrically sound instruments are used to assess HRQL. Epilepsy-specific HRQL measures for children. Accurately measuring HRQL in children with epilepsy can be challenging. Only recently have researchers focused on studying epilepsy and its impact on HRQL (R onen et al., 2003). To garner the most useful data when utilizing a HRQL rating scale, it is important to use a measure
48 developed for children because adult HRQL measures are typically inappropriate for use in children and adolescents (Carpay & Arts 1996). Different types of information are pertinent depending on the patientÂ’s age. For example, scales developed specifically for children focus on social issues and their physical appearance, whereas adult scales focus more on financial issues, caree r issues, and being self-suffi cient. Furthermore, HRQL measures for children must accommodate the changes that occur throughout childrenÂ’s development, whereas these domains are unnecessary for adult patients. Eiser and Morse (2001) recommend that to assess most accurately childrenÂ’s HRQL, children should rate their own HRQL. Research findings have revealed that children identify more items impacting th eir HRQL than do their own parents. Furthermore, young children (i.e., seven years of age) appear to be consistent and accurate in their understanding of the items and response options on rating scales, and they have demonstrated very good test-retes t score reliability (Eiser & Morse, 2001). These findings support the effectiveness of util izing self-report measures to assess HRQL in children with epilepsy. It is essential that epilepsy-specifi c HRQL scales are developed to focus on problems relevant to children. For example, it would be advantageous if scales could detect subtle changes in the child includi ng evaluation of different therapies. More research must be undertaken to identify accura tely attributes of HRQL in children with epilepsy. One of only two research groups that ha ve addressed this ar ea is Ronen et al. (2003). These researchers used separate focus groups for children with epilepsy, ages 6 to 10 years, and their parents, so that each c ould discuss their own perceptions of life with
49 epilepsy. Five dimensions were identified: (a) the experience of epilepsy; (b) life fulfillment and time use; (c) social issues; (d) impact of epilepsy; and (e) attribution. In their follow-up study, 381 child ren with epilepsy and thei r parent(s) independently completed a 67 item HRQL questionnaire. Fa ctor analyses revealed five HRQL dimensions that the children considered most important: (a) interpersonal/social impact; (b) areas of worries and concerns; (c) intrapersonal/emotional consequences; (d) issues of keeping epilepsy a secret; and (e) quest for normality and resilience. Factor analysis of the parents' reports of their children's HRQL identified only the firs t four factors. In addition, the parents thought their children we re worried as much about the future as about present issues whereas in fact the child ren worried almost exclusively about present matters. Based on the results of the factor an alyses, the questionnaire was reduced to 25 items on both the child form and the parent fo rm without sacrificing the integrity of the questionnaires. Statistics were computed to determine the psychometric properties of the questionnaires. Results showed that the que stionnaires were psychometrically robust for parents and children 8 years of age and older. Internal c onsistency, construct-related validity, and test-retest reliabi lity were found to be adequa te. Scores obtained from the matched parent and child were poorly to moderately correlated. In another study, Arunkumar, Wyllie, Kotagal, Ong, and Gilliam (2000) examined parent and patient-validated conten t for quality of life assessment in children with epilepsy. The researchers obtained info rmation from both parents of children with epilepsy and the children themselves. A total of 80 parents of children with epilepsy (3 months to 20 years of age) and 48 of the children were asked to share concerns about
50 living with epilepsy in order of importance. These data were helpful in establishing questionnaires for use with children and parents (Arunku mar et al., 2000). Gilliam et al. (1997) utilized the pare nt-proxy generic Child Health Questionnaire (CHQ; Langraf et al., 1996) to assess functi oning and quality of life in 33 children before undergoing epilepsy surgery. Significantly poore r scores were found in the domains of emotional impact on parents, time impact on parents, and the ge neral health index. Decreased scores also were found for domains of self-esteem, gene ral behavior, and the physical function index. Results indicated that the interventi on did not allow the children who had epilepsy surgery to reach the same le vels as healthy controls even though their HRQL scores did increase following surgical tr eatment. However, a significant limitation to this study was that it failed to assess which HRQL domains improved after surgery, and which, if any, subgroups of patients impr oved following surgical treatment (Gilliam et al., 1997). A different group of researchers (Mille r, Palermo, & Grewe, 2003) utilized the same parent-proxy CHQ scale to compare 41 ch ildren with epilepsy, 4 to 19 years of age, to healthy controls. The majo rity of the childrenÂ’s seizur e disorders were severe, and more than one half of the patients had co-m orbid neurological impairments. Results from this study revealed that the pr esence of co-morbid disorders and the use of multiple AEDs were the strongest predictors for poor HRQL in children with epilepsy. A limitation to this study was that HRQL markers unrelated to epilepsy were not used to compare these scores for children with epilepsy a nd scores for the healthy controls. Devinsky, Westbrook, Cramer, Glassm an, Perrine, and Camfield (1999) conducted a study to assess the risk factors for lowered HRQL in 197 adolescents, 11 to
51 17 years of age, with epilepsy. These resear chers correlated antiep ileptic drug toxicity, socio-demographics, academic and social variables, and heal th-related variables (including epilepsy), with self-reported HR QL via the Quality of Life in Epilepsy Inventory for Adolescents (QOLIE-AD48; Cramer, Westbrook, Devinsky, Perrine, Glassman, & Camfield, 1999). The variables th at were related to lower HRQL included older age, lower socioeconomic status, increa sed seizure severity, and antiepileptic drug neurotoxicity. Unfortunately, re mediable factors that may ha ve been responsible for the lower HRQL in older adolescen ts and those with a lower so cioeconomic status were not identified in this study (Devinsky et al., 1999). Despite recent achievements in developi ng HRQL measures, there is a need to improve our understanding of the functional and experiential dimensions associated with complex neurodevelopmental disorders (Rone n, Rosenbaum, & Streiner, 2000). It is difficult to attribute better or poorer quality of life to the nature of epilepsy alone, when so many disparate factors play key roles in people's lives. These factors include, among others, a child's resilience, co-morbid condi tions, parental well-b eing, family factors, attitudes, and societal/cultural variables. Recent studies have shown that clinical symptoms such as seizure frequency and sever ity, or other biomedical markers, have only moderate correlations with HRQL (Ronen et al.). Furthermore, HRQL may change over time with the development of the child and the family's accommodation to the situation. In addition, we need to learn what truly en compasses comprehensive patient care, define the goals of management, and attempt to eval uate the impact of interventions wherever possible.
52 Future assessments should include meas ures where the items originated from children with epilepsy, and allow them to ra te their own HRQL. However, parent-proxy report measures may be useful in addition to the child's self-assessment. Even though the ratings of the child and parent may differ, it may be beneficial to obtain ratings from both informants to gain multiple perspectives of the impact of epilepsy (Ronen et al., 2003). The combination of self-report and parent responses may help to understand family dynamics of coping with epilepsy. In addition, these data may help to identify multiple issues that could be addressed through fam ily therapy because a childÂ’s chronic illness impacts not only the child, but the entire family. Therefore, future research is needed to assess the advantage of using a multi-info rmant method for meas uring health-related quality of life in chil dren with epilepsy. Summary Pediatric epilepsy is a complex neurol ogical condition with a number of possible co-morbid features. Although there is a d earth of research examining comorbid psychopathology in children with epilepsy, it is apparent from the existing literature that there is a significant relati onship between psychopathology (i .e., depression, anxiety) and epilepsy in both children and adults. Seiz ures alone may be difficult to manage as treatment with medication presents numerous challenges. There exists the typical challenge of medical adherence as well as th e potential negative side effects that are encountered when using antiepileptic me dications. Furthermore, when comorbid disorders are present, treatment difficulties are exacerbated. The coping process in epileps y is not linear; it may vary with different life stages and in relation to each patient's unique experiences. When a child is first diagnosed with
53 epilepsy, the child and family are likely to have a variety of emotions and reactions to the diagnosis. These initial emotions may include fear, anger, and disbelief, and subsequent emotions may involve anxiety and depression (Shafer, 2002). The types and intensity of emotions are likely to change over the course of the illness, depending on factors such as the severity and frequency of seizures and side effects of medicat ions. Because of the multitude of factors that impact the presen ce of psychopathology and poor health-related quality of life, it is imperative that epile psy management include early assessment and treatment of seizures as well as psychosocial problems to a ssist children in successfully coping with their chronic medical condition. It is imperative that researchers ascertai n the risk factors that lead to poor healthrelated quality of life in children with epile psy. Knowledge of these factors may help to determine effective treatments to enhance the HRQL of children with epilepsy and their families. Through the attainment of information related to HRQL of children with epilepsy there are many ways that children and their families can be supported. Empirically-validated instruments can aid in the detection, assessment, and follow-up of issues impacting children with epilepsy and their families. It also is important to recognize the need to address opportunities for dissemination, translation of information, and implementation of the obtained research fi ndings into clinical pediatric practice. To assess the effect of epilepsy on child ren, the childrenÂ’s own perceptions should be addressed in addition to their families a nd teachers, and any other significant persons in the childÂ’s environment. When a child is coping with a chronic condition, it impacts the childÂ’s entire family. Likewise, the childÂ’s condition may impact his or her performance at school, both behaviorally a nd academically. It is imperative that school
54 faculty and staff (e.g., teachers, psychologists, and administrators) as well as students are aware of the childÂ’s condition, including the sp ecific diagnoses, symptoms, and treatment plans. Because children spend a significant amount of their day in school, educators are in the position to recognize symptoms that may indicate intern alizing problems (e.g., depression and anxiety) and ex ternalizing problems (e.g., aggr ession) that may negatively impact their levels of success at school. When children with epilepsy and comorbid psychopathology are identified early, appropria te interventions may be developed and implemented that will help to ameliorate in ternalizing and externalizing difficulties, as well as academic problems. It is critical that health care and educational profes sionals recognize the spectrum of issues facing children with epilepsy and become familiar with available resources for these children. The ideal model of care for epilepsy stre sses collaboration among patients, families, primary care physicians, epilepsy sp ecialists, mental health professionals, educators, rehabilitation experts, and co mmunity resources. Finally, the children diagnosed with epilepsy and their caregive rs should be counseled regarding the associated risks and benefit of various treatment options.
55 Chapter 3 Method Overview This chapter presents the methods for th e current study in detail. Characteristics of the participants are reviewed, such as inclusion and exclusion criteria, and sampling considerations such as sampling scheme and sample size. Next, all procedures for the study, in a step-by-step fashion, are thoroughl y presented. The research design for the study is discussed, as well as the statistical analyses that were c onducted to address each research question. Selection-Eligibility Criteria The purpose of this study was to de termine whether seizure type, seizure frequency, anti-epileptic drug (AED) trea tment, anxious symptomatology, and/or depressive symptomatology were correlated with a lowered health-related quality of life in children with epilepsy. A central aim of the study was to identify children with epilepsy who are most at-risk for poor HRQL based on seizure variables and psychopathology. Therefore, children with a vari ety of types of epileps y and variability in seizure characteristics were invited to participate in this study. The inclusion criteria for participation in th e current study were (a) a diagnosis of epilepsy confirmed by a pediatri c neurologist, (b) chronologi cal age of child between 8 years, 0 months and 11 years, 11 months at the time of the study, (c) pharmacological treatment with AEDs, (d) English proficiency of child and guardian, (e) consent from the
56 childrenÂ’s parents/guardians, a nd (f) verbal assent from th e children. The age range of 8 to 11 years was chosen because the researcher was interested specifically in elementaryschool aged children (i.e., pre-adolescen ce) diagnosed with epilepsy. Additionally, adolescent participants were not included b ecause this population would introduce unique age-related variables that potentially would confound the fi ndings of the current study. Pharmacological treatment with AEDs was an inclusion criterion be cause the literature has shown that AEDs, especially use of multiple AEDs, are correlated with lowered health-related quality of lif e (Devinsky, 2003; First Seizure Trial Group, 1993; Miller et al., 2003). This researcher was interested in examining the impact of AEDs, specifically the number of AEDs prescribed (i.e., monot herapy or polytherapy) on the presence of anxiety and/or depression and health-related quality of life. Exclusion criteria for par ticipation in this study incl uded children diagnosed with acquired epilepsy (e.g., due to head injury) and those with epilepsy who were not taking any antiepileptic drugs. Furthermore, for purposes of this study, the pr incipal investigator was interested in discovering the distinct impact of pediatric epilepsy on depressive symptoms, anxious symptoms, and HRQL. B ecause additional medi cal or psychiatric conditions would likel y contribute to depressive sy mptoms, anxious symptoms, and lowered HRQL, children with additional dia gnoses were not included in the current study. Additionally, children who were prescrib ed psychotropic medications other than anti-depressants and anti-a nxiety medication did not m eet criteria for the study.
57 Sampling Sampling Scheme A non-random, convenience sampling method was used in this study. That is, the principal investigator attempted to recruit pa rticipants for the current study based on their accessibility (e.g., geographic location). This sampling scheme was used because it was the most practical sampling scheme given th at epilepsy is a relatively low-incidence condition. Sample Size According to Tabachnick and Fidell (2001) the number of vari ables in a research study, in part, determines the sample size need ed to obtain statistica lly significant results; therefore, when utilizing canonical correlation analyses, a rule of thumb in the social sciences is to have approximately 10 cases fo r each variable in the study. However, if score reliability is high, then a much lower ra tio of cases to variables is adequate. The current study examined the relationship betw een five independent variables and five dependent variables utilizing a 5% level of statistical si gnificance and a .80 level of statistical power. In view of that, the goal of this st udy was to obtain data from a minimum of 50 parent-child pairs to have sufficient power to obtain statistically significant findings. The final sample size obta ined in this study was 51 child-guardian pairs. Instruments Demographics and Seizure Variables Questionnaire (DSV Questionnaire) A Demographics and Seizure Variable s Questionnaire was developed by the principal investigator for parents/guardians to complete to obtain information on their
58 child diagnosed with epilepsy. This questionnai re was utilized in the study to obtain data on specific variables that were statistically an alyzed to determine the impact of seizurerelated variables on HRQL, depressive symp toms, and anxious symptoms in children with epilepsy. The specific seizure-related va riables (i.e., seizure type, seizure frequency, and use of AEDs) were included because prio r research suggested that these variables may negatively impact HRQL in children with epilepsy. Furthermore, demographic data were collected so that descriptive informa tion about the population participating in the study could be ascertai ned and presented. The following information was gleaned: (a ) age of child, (b) gender of child, (c) race of child, (d) additional diagnoses of chil d, (e) type of seizure (i.e., generalized or partial), (f) seizure frequency (i.e., 0-5 or 6-12), and (g) treatment with AEDs (i.e., monotherapy or polytherapy). The seizure fr equency category ranges of 0-5 and 6-12 were developed based on discussion with a pe diatric neurologist. The rationale was that seizures that occurred more often than on ce per month would be considered uncontrolled (i.e., intractable) seizures and would introduce another subset of pa tients with epilepsy. The format of this questionnaire consiste d of multiple-choice and fill-in-the-blank options. There was a total of 10 items; 4 fill-in-the-blank items and 6 multiple-choice items. After the questionnaire was develope d, a pediatric neurol ogist reviewed the questionnaire to ensure accuracy in the te rms used and the wording of the items. The questionnaire took guardians a pproximately five minutes to complete. The Demographics and Seizure Variables Questionnaire is presented in Appendix A.
59 Behavior Assessment System for Childre n, Second Edition (BASC-2; Reynolds & Kamphaus, 2004) The Behavior Assessment System for Children, Second Edition (BASC-2), is a broadband measure that consists of a variety of items and scales that assess internalizing and externalizing domains of behavioral/e motional problems, as well as social and adaptive competencies. The BASC-2 has both cl inical and adaptive scales. The clinical scales include: Externalizing Proble ms (hyperactivity, aggression, and conduct problems), Internalizing Problems (anxiet y, depression, and somatization), School Problems, Attention Problems, and Learni ng Problems. The adaptive scales include: Adaptive Skills (adaptability), Social Skills (leadership), and Study Skills. The BASC-2 includes a total of six forms: self, parent, and teacher measures for young children (ages 6 through 11), and self parent, and teacher measures for adolescents (ages 12 through 21). Research ha s shown that caregivers and teachers are not always aware of the negativ e effects of an illness on their child, which is evidenced by discrepancies found between child and parent reports (Merrell, 1999). Therefore, both the child self-report form (BASC-SRP-C) a nd the parent form (BASC-PRS-C) were completed independently. Moreover, t here are significant differences among children with anxiety alone, depression alone, and co-morbid anxiety and depression; ther efore, it is imperative to differentiate between anxiety and depre ssion (Brady & Kendall, 1992). Given the comorbidity of both anxiety and depression in ep ilepsy, it is necessary to assess anxiety and depression in children with epilepsy to deve lop effective treatment plans. The BASC-2 provides validity indicators (F, Response Pa ttern, and Consistency) that make the
60 interpretation of clinical scale elevations more meaningful. Furthermore, the BASC-2 separates anxiety and depressi on constructs into two separa te scales. There is no item overlap between these two scales, and this a llows for a greater opportunity for differential diagnosis (Kamphaus & Frick, 1996). For purpos es of the current study, the anxiety and depression scales from the BASC -2 were of most interest. The BASC-2 produces T-scores, which describe distance from the mean, and percentiles, which describe ra rity. The BASC-2 has been normed for a general population as well as clinical populations. Clinical norms are useful when the childÂ’s problems are extreme compared to the gene ral population. Therefore, clin ical norms help to avoid ceiling effects for children who have significan t adaptive and/or behavioral problems. In addition to general norms and clinical norms the BASC-2 also has gender-based norms. Gender-based norms help identify children whos e self-report scores are rare for their age and gender (Reynolds & Kamphaus, 2004). The BASC-2 generates two interpretive ranges (i.e., at-r isk and clinically significant) for composite s cales and subscales based on T scores obtained using norms. Typically, T scores ranging from 60 to 69 on the clinical scales are considered to be in the Â“at-riskÂ” classification range indicating th e potential of developing a problem that requires careful monitoring. T scores of 70 and above on the clinical scales are considered to be in the Â“clinically signifi cantÂ” range. Scale scores in the clinically significant range indicate a high level of maladjustment. The BASC-2 is reported to be a comprehensive and psychometrically s ound assessment instrument (Flanagan & Esquivel, 2006).
61 The Behavior Assessment Scale for Child ren Self-Report of Personality (BASCSRP-C) consists of 152 items and is designed for use with children 8 to 11 years of age to provide insight into childrenÂ’s thoughts a nd feelings. The BASC-SRP-C takes children approximately 30 minutes to complete. The BASC-SRP-C is written at a first-grade reading level; therefore, children as young as si x years of age should be able to complete the measure independently or with minimal assistance. Furthermore, the SRP form contains validity scales to aid in determ ining the quality of the data collected. The BASC-SRP-C is a well-developed br oadband self-report m easure with strong psychometric properties (Merrell, 1999). An alyses conducted with data from the standardization sample revealed that the median internal consistency reliability coefficients of the scale scores were in the low .80 range. Furthermore, the median coefficients for the composite scores were in the mid .90 range, and the test-retest reliability coefficients were in the mid .80 ra nge at one month intervals (Merrell, 1999). In the current sample, the internal consistency reliability coefficient of the scale scores on the BASC-SRP-C Anxiety subscale was 0.72 and the Depression subscale was 0.87. The Behavior Assessment System for Children Â– Parent Rating Scale (BASCPRS-C), is completed by parents or guardi ans and takes approximately 10-20 minutes to complete. The BASC-PRS consists of 134 to 160 items that measure adaptive and problem behaviors in the ho me and community setting. Items on the PRS are rated by using a 4-point Likert-type scale ranging from 0 (never) to 3 (almost always). To complete the PRS, a fourth-gra de reading level is required. It is reported in the BASC-2 manual that the individual PRS scale a nd composite scores yield good reliability coefficients (.80), except for the Adapta bility, Conduct Problems, Hyperactivity, and
62 Somatization scales. Fortunately, for purposes of this study, the former scales were not of specific interest. A number of criterion-related and construc t-related validity studies indicate that the PRS scores yield adequate validity (e.g., Doyle, Ostrander, Skare, 1997; Kamphaus & Frick, 1996). In the current sample, the internal consistency reliability coefficient of the scale scores on the BA SC-PRS-C Anxiety subscale was 0.72 and the Depression subscale was 0.87. Health-Related Quality of Life Qu estionnaire (HRQL; Ronen et al., 2003) The Health-Related Quality of Life qu estionnaire (HRQL; Ronen et al., 2003) was used to assess the quality of life for ch ildren diagnosed with epilepsy. There were two versions of this questionnair e utilized in the current stud y: a child questionnaire and a parent questionnaire. The child questionnaire wa s developed for children 6 to 15 years of age. This measure was designed to assess a childÂ’s quality of lif e in multiple domains. The child questionnaire contains five subs cales: (a) interpersonal/social, (b) present concerns, (c) intrapersonal/emotional, (d ) secrecy, and (e) normality. The parent questionnaire also contains five subscales: (a ) interpersonal/social, (b) present concerns, (c) future concerns, (d) intrapers onal/ emotional, and (e) secrecy. Both the parent form and the child form have 25 items. All items on the questionnaires utilize a 4-point Likert-type sc ale, ranging from Â“really trueÂ” to Â“sort of trueÂ” for two contrastin g statements. An example of an item is, Â“Some kids with epilepsy feel that other kids treat them differentlyÂ” but Â“Other kids with epilepsy feel they are treated the same as everyone else.Â” The indivi dual is instructed to circle the statement that is most like them, and then choose whethe r the one statement is Â“really trueÂ” or Â“sort of trueÂ” for them. The scores for each subscale range from 5 to 20, with lower scores
63 indicating more compromised HRQL. Both pa rent and child HRQL questionnaires are reported by the developer to take approximately five minutes to complete. The HRQL questionnaire is one of only tw o disease-specific health-related quality of life scales that uses a self-response questionnaire that has been found to have sound psychometric measures. Furthermore, the HRQL is the only measure that has parallel questionnaires for both the parent/guardian and the child to complete independently. Ronen et al. (2003) assessed the psychometr ic properties of the HRQL questionnaire. A total of 381 children and 424 parents/guardia ns participated in the study. All child participants were between 8 a nd 15 years of age. Results revealed that scores from the HRQL questionnaire yielded good internal c onsistency, test-retes t reliability, and construct-related validity (Ronen et al.). CronbachÂ’s alpha demonstrated adequate internal consistencies (> 0.70) for scores yielded for all scales except the normality subscale in the self-report scale (i.e., 0.63) and the present worries subscale in the parentproxy scale (i.e., 0.64). In the current sample the internal consistency reliability coefficient of the five subscale scores on bot h the HRQL Â– Child Form and Parent Form were adequate. CronbachÂ’s alpha for scores on the HRQLChild Form were as follows: Social (0.87), Present (0.90) Emotion (0.91), Secrecy (0.90), and Normalcy (0.91). On the HRQL Â– Parent Form, the five subscale scor es revealed the following score reliability coefficients: Social (0.87), Present (0.88), Future, (0.79) Emotion (0.76), and Secrecy (0.80).
64 Procedures Prior to the start of the research study, approval for the proposed study was obtained from the University of South Florid a Institutional Review Board to ensure the ethical treatment of all participants. The principal investigator obtained the contact information (e.g., telephone number) of a pediatri c neurologist in the west central Florida region. This contact information was obtaine d by the principal investigator through a psychologist in the same geographical regi on who was participating on the principal investigatorÂ’s dissertation committee and ha d a working relationshi p with a pediatric neurologist in the region. Th e pediatric neurologist was contacted via telephone to provide information regarding the goals a nd objectives of the study. The principal investigator requested permission from the participating pediatric neurologist to handdeliver packets to his practice to be distribut ed by the office manager to the parent-child pairs who met all inclusion criteria. In January 2008, the principal investig ator sought and obtained approval from Wayne State UniversityÂ’s Institutional Re view Board to add a ChildrenÂ’s Hospital located in the Midwestern United Sates as a second data collection site. The principal investigator contacted a pedi atric neurologist at the ChildrenÂ’s Hospital via electronic mail and written letter to provide information regarding the goals and objectives of the study. The principal investigator requested permission from th e pediatric neurologist to hand-deliver packets to his clinic within th e ChildrenÂ’s Hospital to be distributed by the office manager to the parent-child pair s who met all inclusion criteria. After verbal and/or written consent was obtained from the above-referenced pediatric neurologists, packets (i.e., forms enclosed within a sealable manila envelope)
65 were generated for parent-child pairs. The p ackets for the two sites consisted of the same type of information; however, the consent pr ocess/requirements of the two sites differed and therefore there was a slight variation in the packets for the two sites. Packets for the site in the Southwestern United States c onsisted of the following documents: a cover letter and assent form for the child partic ipant (see Appendix B) and a cover letter and consent form for the guardian participant (see Appendix C). Additionally, the packets included each of the following measures: De mographics and Seizure Variables (DSV) questionnaire to be completed by the guardia n, Behavior Assessment System for Children Â– Parent Rating Scale (BASC-PRS; Reynolds & Kamphaus, 2004), Behavior Assessment System for Children Â– Self Report of Pers onality, child form (ASC-SRP-C; Reynolds & Kamphaus, 2004), Health-Related Quality of Life (HRQL; Ronen et al., 2004) Â– parent form (see Appendix D), and Health-Related Quality of Life (HRQL; Ronen et al., 2004) Â– child form (see Appendix E). The documents included in the packets for the ChildrenÂ’s Hospital in the Midwestern United States were the same as stated for the Southeastern United States site with the following exceptions : instead of an assent form (Appendix B) an assent oral script was utilized (see A ppendix F), and in lieu of the parental consent form (Appendix C), an information sheet was provided to guardian participants (see Appendix G). The instruments were presented in counter-balanced order; that is, one half of the packets had the BASC-2 rating scale first, followed by the HRQL scale and the other one half were in reverse order. Two cover letters were developed; one for the child participant and one for the parent/guardian participant. In age-appropriate language, each cover letter reviewed the purpose and goals of the study, and provided detailed instructions regarding how to
66 complete the questionnaires a nd rating scales. The instructio ns emphasized that all forms should be completed independently, and sp ecific guidelines were provided in the parent/guardian cover letter (and information sheet) regardin g assisting their child with completing the forms. For example, the pare nt/guardian could clarify, as needed, any item for which the child requested clarificatio n; however, the paren t/guardian should not influence the childÂ’s responses. That is, pe rmission was given to read items from the questionnaires out loud to the child if need ed; however, the items were to be read verbatim. Both cover letters explained that all data collected woul d be confidential and would only be used for research purposes. In the event that particip ants had any questions or concerns regarding the research study, the cover letters provided the principal investigatorÂ’s contact information (i.e., tele phone number and electronic mail address) so that additional information could be pr ovided to participants as requested. Finally, the cover letters included a descri ption of an incentive for completing and returning all forms in the packet. All parent-c hild pairs that completed and returned the assent and consent forms, Demographics a nd Seizure Variables Questionnaire, and rating scales were eligible to recei ve a $100 Visa gift card through a random drawing that was completed at the end of the data collection period. The cover lette r requested that the parent-child pair review the materials in the packet, and if they c hose to partake in the study, to sign the consent and assent forms and include them in the packet to be returned to the office manager in the pe diatric neurologistÂ’s office. Information was collected from participan ts in a pediatric neurologistÂ’s office in the Southeastern United Stat es and a pediatric neurology c linic housed in a ChildrenÂ’s Hospital in the Midwestern United States. Wh en parent-child pairs arrived for their
67 regularly scheduled appointments with the pedi atric neurologists, they were asked by the office manager if they were willing to partic ipate in a research study that required them (parent and child) to take approximately 40 minut es to fill out several forms related to the childÂ’s illness. If they consented, they then were given a packet a nd were asked to read through the materials. The mate rials instructed the particip ants to sign the consent and assent forms and to complete the questionnair e and rating scales in the packet prior to their appointment with the pediatric neurol ogist. The forms were filled out in the reception area while they were waiting to m eet with the pediatric neurologist. After materials were completed independently by the parent and child, they were sealed in a manila envelope by the parent, and the offi ce manager collected a nd held the completed packets at the pediatric neurologistÂ’s office until the principal investigator collected the completed packets. After approximately two months of data collection, an adequate number of participants had not be acce ssed (i.e., less than 30 completed packets); therefore, a pediatric neurol ogist practicing within a nonprofit ChildrenÂ’s Hospital was contacted to obtain access to additional par ticipants. The principal investigator was granted access to recruit patie nts from this neurologistÂ’s practice which generated 47 additional participants for the current study. All data obtained from participants were kept confidential and were accessible only to the principal investigator. All forms in each packet had identification numbers to match parent and child forms/responses in the event that the pair of parent-child forms became separated. The identification numbers were arbitrarily created by the principal investigator. The identification number for each parent-child pair was printed on the individual forms to ensure anonymity; therefor e, no identifying information was collected
68 from participants. That is, in each packet, both guardian and child instruments had the same code number (e.g., 001). Approximately 50 packets were deliver ed by the principal investigator to the participating pediatric ne urologistsÂ’ clinic. The office managers at each clinic collected the complete d packets and made a notation (e.g., asterisk) in the childÂ’s private medical file indicating that they retu rned the completed packet. At the conclusion of the study, the principal investigator pr ovided the neurologists Â’ office managers each with a $100 Visa gift card, and the office ma nagers randomly selected an individual who had participated in the study to receive the gi ft card. The office manager selected a name from those marked with a notation in thei r medical file indicating that they had participated in the study. This procedure allowed a ll participants to remain anonymous, as the neurologists and office managers were the only persons who were privy to the participantsÂ’ identities and c ontact information. All completed packets were stored in a locked file cabinet that was accessible only to the principal investigator. Once all data were collected, the principal i nvestigator conducted statistical analys es, interpreted the results, and presented the findings. For participants at the Southeastern United States site, any child participants obtaining depression and/ or anxiety scores in the Â“at-riskÂ” or Â“clinically significantÂ” range on the BASC-2 were notified of thei r increased risk for psychopathology (i.e., depression and/or anxiety). The participants Â’ code numbers and BASC scores in these elevated ranges were disclosed to the ne urologistÂ’s office mana ger within 24 hours of scoring the BASC data. In turn, the office ma nager contacted the guardian participant to relay this information so the guardian coul d make an informed decision regarding the attainment of appropriate mental health services for their child. Similarly, all participants
69 recruited from the Midwestern United States site were provided with an outpatient psychotherapy referral form from the Depart ment of Psychiatry and Psychology at the recruitment site (i.e., childrenÂ’s hospital) in the event that they requested outpatient services. Research Design The current study utilized an exploratory, non-experimental research design, also known as a correlational research design. This design was used because the principal investigator was interested in determining th e statistical associati on between two or more variables. This study utilized a survey method to collect quan titative data from participants. Survey methods are typically us ed to gather large am ounts of data about a construct when there is little empirical da ta available (Bordens & Abbott, 1996). The survey method encompassed a multi-informant approach whereby both self-report and parent-report were utilized to obtain data from multiple viewpoints. Specifically, the child with epilepsy and his or her guardian completed the assessment instruments (i.e., rating scales). This multi-informant approach was utilized because research shows that it is advantageous to collect data from multiple sources. Furthermore, in applied research, the multi-informant approach is characteristic of quality research designs (Holmbeck et al., 1998). Data Analyses Descriptive Statistics Once all questionnaires and rating scales were completed by the participants and returned to the researcher, the data were analyzed. Descriptive statistics included the following: mean, median, standard deviati on, and skewness and kurtosis of individual
70 participantsÂ’ scores across rating scales and questionnaires. In addition, descriptive statistics were computed for the particip antsÂ’ demographic information. No extreme outliers were discovered in the data. The descriptive statistics are illustrated quantitatively in Table 1. Variables Independent variables. The independent variables in this study were symptoms of anxiety and depression (as measured by the BA SC-2) and seizure-related variables. The seizure-related variables were type of epileps y, seizure frequency, and use of antiepileptic drugs (i.e., monothera py or polytherapy). Dependent variables. The dependent variables in this study comprised healthrelated quality of life subscale scores, as measured by the Health-Related Quality of Life (HRQL) questionnaire. The five subscale scores for the self-report measures were social, present, emotional, secrecy, and normalcy factor s. For the parent report measure, the five subscales consisted of social present, future, emotional, and secrecy factors. Each individual subscale ranged fr om 5 to 20, with higher scor es indicating better healthrelated quality of life. Inferential Statistics Canonical correlation analysis. Canonical correlation is a procedure that allows researchers to examine the relationship betw een two sets of variables (Thompson, 1984). The linear correlation between the two sets of la tent (i.e., not directly observed; inferred) variables is maximized with a canonical co rrelation analysis. Se ts of variables are combined to produce a predicted value that ha s the highest correlation with the predicted value of the other (second) set of vari ables (Thompson, 1984). That is, canonical
71 correlation analyses allowed this researcher to determine which combination of seizurerelated variables and psychopathology variables best correlate with health-related quality of life variables in children with epilepsy. More than one linear correlation relating two sets of variables may exist, with each correlation representing a different dimension by which the independent set of variables is related to the dependent set. Specifically, the linear correlation relating the two sets of variables is termed a canonical function and the number of canonical functions that can be generated for a given data set is equal to the number of variables in the smaller of the two variable sets. Because there were five independent variables and five dependent variables in the child data se t and the parent data set, the number of possible canonical functions in each set was five. There are several assumptions for canonical correlation anal ysis. First, there is the assumption of linearity that posits that th e correlation coeffici ent between any two variables is based on a linear relationship, and the canonical correlation is the linear relationship between the variates. It is unlik ely that this assumpti on can ever be truly confirmed; however, canonical corr elation analysis is robust to small deviations from this assumption (Thompson, 1984). Nevertheless, a bivariate scatterplot was used to determine if there is curvature in the rela tionship between each inde pendent variable and dependent variable. If significant curvatur e was found, then the corresponding variable would have been transformed. Second, in canonical correlation analysis, it is assumed that there is multivariate normality, which can be assessed, in part, by en suring that each variable has univariate normality. That is, if univariate distributions are not normal, the multivariate distribution
72 is not normal; however, if all uni variate distributions are normal, it is still possible that the multivariate distribution is not normal Nevertheless, canonical correlation can accommodate metric variables without the st rict assumption of normality (Osborne & Waters, 2002). Third, the assumption of the ab sence of multicollinear ity indicates that there is no redundancy in the independent variab les such that a relationship does not exist between them (Osborne & Waters, 2002). Finally, it is important to not e that a significant limitation to canonical correlation analyses is that relationshi ps can be determined, but the underlying causal mechanism cannot be ascertained.
73 Chapter 4 Results Treatment of the Data The data were entered into an Excel sp readsheet by the researcher and verified by a colleague (i.e., a clinical psychology gra duate student) following the completion of all forms and questionnaires. Each score was ente red for every participant on each individual item. Missing data were coded as a blank sp ace in the Excel document while single items with multiple responses were averaged and th e mean score was inputted. The researcher and a clinical psychology graduate student checked the data by randomly selecting participantsÂ’ identification numbers and matching the data in the database to the entrees completed by hand. Additionally, extreme values were checked across each participant for each item to ensure that data were ente red correctly. The final sample size of this study was 51 child-guardian pairs; no cases were removed from the analysis due to extreme outliers or incomplete da ta. Inter-rater agreement was 100%. Demographics After all data were transferred from th e Excel spreadsheet into an SPSS data editor file, the data were analyzed. Descriptive statistics were computed for participantsÂ’ demographic information (see Table 1). All demographic information collected in the current study was obtained through parent self-report measures.
74 Independent Variables Among the dichotomous variables, gender wa s coded as Â“1Â” for males and Â“2Â” for females. Medical insurance was coded as Â“1Â” for private insurance and Â“2Â” for Medicaid. Regarding seizure type, generali zed seizures were coded as Â“1 Â” and partial seizures were coded as Â“2.Â” Child participants with 0 to 5 seizures within the last 12 months were coded as Â“1Â” and those with 6 to 12 seizures with in the last 12 months were coded as Â“2.Â” Finally, child participants trea ted with one AED were coded as Â“1Â” and those treated with more than one AED were coded as Â“2.Â” The continuous variable (i.e., age of child) was measured in years and months. The mean age of the children participants was 9 years 5 months ( SD = 1 year 8 months). The youngest child participant was 8 ye ars, 0 months of age whereas the oldest child participant was 11 years, 11 months of age. Regarding gender of the child participants, 29 (56.9 %) were identified as male and 22 (43.1 %) identified as female. Twenty-eight (54.9 %) families identified themselves as White, 11 (21.6 %) identified as African American, 4 (7.8 %) identified as Hi spanic, and 8 (15.7 %) identified themselves as mixed race. Information regarding participantsÂ’ medical insurance was obtained in an effort to infer the socioeconomic status of families pa rticipating in the curr ent study. These data were dummy coded; families with private medical insurance were coded as Â“1Â” and families with medical insurance through Medicaid were coded as Â“2.Â” Based on participants self-report, 18 (35.3 %) familie s had private medical insurance. The remaining 33 (64.7 %) families reported that they had insurance through Medicaid.
75 Twenty-one (41.2 %) parent participants repo rted that their chil dÂ’s seizures were generalized seizures, with 30 (58.8 %) parent s reporting that their child experienced partial seizures. Regarding seizure frequenc y, 28 (54.9 %) parent par ticipants indicated that their child experienced between zero and five seizures within the last 12 calendar months. Twenty-three (45.1 %) parent partic ipants reported that their child had experienced between 6 and 12 seizures with in the last 12 calendar months. Finally, 34 (66.7 %) parent participants indi cated that their ch ild currently was pr escribed a single anti-epileptic medication (i.e., monothera py) compared with 17 (33.3 %) parent participants who indicated that their child was prescribed multiple anti-epileptic medications (i.e., polytherapy). Table 2 presents descriptive data for child self-report and parent report on BASC2 and HRQL measures. Means, standard devi ations, skewness, and kurtosis are reported for each individual variable. The BASC-2 genera tes two interpretive ranges (i.e., at-risk and clinically significant) for com posite scales and subscales based on T scores obtained using norms. Typically, T scores ranging from 60 to 69 on the clinical scales are considered to be in the Â“at-riskÂ” classificat ion range indicating the potential of developing a problem that requires careful monitoring. T scores of 70 and above on the clinical scales are considered to be in the Â“clinically signi ficantÂ” range. Scale scores in the clinically significant range indicate a high level of maladjustment. Scor es below 60 are considered to be in the Â“normalÂ” range a nd do not indicate maladjustment. In the current study, child pa rticipantsÂ’ ratings of anxi ous symptoms (as measured on the BASC-2) had a mean score of 54.9 ( SD = 8.8), with scores ranging from 37 to 71. Self-report scores on depressive symptoms (as measured by the BASC-2) had a mean
76 score of 54.7 ( SD = 9.1) and a range of 38 to 73. Pare nt-report of anxious symptoms in the child participant (as measured on the BASC-2) had a mean score of 54.9 ( SD = 8.8), with scores ranging from 37 to 71. Parent-r eport scores on depressive symptoms (as measured by the BASC-2) had a mean score of 54.7 ( SD = 9.1) and a range of 38 to 73. Dependent Variables The outcome measures for this study were health-related qual ity of life (HRQL) rated by child self-report and pa rent-report. Specifically, ther e were subscale scores for the following five HRQL measures as rate d by self report: (a) social concerns, (b), present concerns, (c) emotional concerns, (d) secrecy issues, and (e) normalcy issues. The scores for each subscale range from 5 to 20, with lower scores indicating more compromised HRQL. Descriptive statistics fo r the HRQL self-report measure were as follows: self-report of social concerns (as measured on the HRQL) had a mean score of 15.2 ( SD = 3.38) and a range from 8 to 20. Pres ent concerns had a mean score of 16.1 ( SD = 3.49) and a range from 9 to 20. Emoti onal concerns had a mean score of 15.4 ( SD = 3.60) and a range from 8 to 20. Secrecy issues had a mean score of 15.4 ( SD = 3.34) and a range from 9 to 20. Finally, normalcy issues had a mean score of 15.7 ( SD = 3.48) and a range from 9 to 20. The HRQL parent form had five subscales as well, consisting of (a) social, (b) present, (c) future, (d) emotional, and (e) se crecy. Descriptive statistics for the HRQL parent-report measure were as follows: parent -report of social concerns (as measured on the HRQL) had a mean score of 15.3 ( SD = 3.99) and a range from 6 to 20. Present concerns had a mean score of 15.5 ( SD = 3.88) and a range from 7 to 20. Future concerns had a mean score of 15.7 ( SD = 3.66) and a range from 8 to 20. Emotional concerns had a
77 mean score of 15.3 ( SD = 3.89) and a range from 7 to 20. Finally, secrecy had a mean score of 15.3 ( SD = 4.25) and a range from 6 to 20. Table 1 Descriptive Statistics fo r Participants (n = 51) Variable (%) Age 8.0 Â– 8.11 23.5 9.0 Â– 9.11 29.4 10.0 Â– 10.11 25.5 11.0 Â– 11.11 21.6 Gender Male 56.9 Female 43.1 Race White (NonHispanic) 54.9 African American 21.6 Hispanic/Latino 7.8 Asian American 0.0 Native American 0.0 Mixed-Race/Other 15.7 Medical Insurance Private Medicaid None Seizure Type Generalized Partial Number of Seizures 0 Â– 5 per year 6 Â– 12 per year Anti-epileptic Drugs Monotherapy Polytherapy 35.3 64.7 0.0 41.2 58.8 54.9 45.1 66.7 33.3
78 Table 2 Means, Standard Deviations, Skewness, and Kurt osis for Participant Ratings (n = 51) Variable Self-Report M SD Skewness Kurtosis AnxietyCH 54.9 8.78 -0.33 -0.66 DepressionCH 54.7 9.15 0.24 -0.81 SocialCH 15.2 3.38 -0.51 -0.67 PresentCH 16.1 3.49 -0.52 -1.05 EmotionCH 15.4 3.60 -0.21 -1.33 SecrecyCH 15.4 3.34 -0.27 -1.11 NormalcyCH 15.7 3.48 -0.33 -1.05 Parent-Report AnxietyPR 54.9 8.33 0.38 -0.87 DepressionPR 57.5 7.16 0.36 0.35 SocialPR 15.3 3.99 -0.56 -0.70 PresentPR 15.5 3.88 -0.51 -0.94 FuturePR 15.7 3.66 -0.53 -0.79 EmotionPR 15.3 3.89 -0.52 -0.98 SecrecyPR 15.3 4.25 -0.47 -1.12
79 Defining Characteristics of BASC-2 Anxiety Scale: Child Self-Report Anxiety and Seizure Type Anxious symptoms, as measured by the BASC-2 Child Form, had an overall mean of 56.00 ( SD = 10.58) for children with generalized seizures and 54.13 ( SD = 7.37) for children with partial seizures. These scores did not indicate a sta tistically significant difference in anxiety scores based on type of seizure F (1, 49) = 0.55, p = .461). The effect size for the difference in mean scores between children with ge neralized and partial seizures was small for self-report ( d = 0.011). The skewness and kur tosis values indicated a fairly normal distribution of scores for ch ildren with generalized and partial seizures ( skewness = -0.42, -0.49; kurtosis = -0.88, -0.73, respectively). Anxiety and Seizure Frequency Anxious symptoms, as measured by the BASC-2 Child Form, had an overall mean of 51.68 ( SD = 8.38) for children with seizures 0 to 5 times in the last 12 months and 58.83 ( SD = 7.72) for children with 6 to 12 seizur es within the last year. The effect size for the difference in mean scores between children with more and less seizures was small for self-report ( d = 0.17). These scores indicated a statistically significant difference between self-report of anxiety and frequency of seizures, with children with more seizures (5-12 within the last y ear) indicating increased anxiety symptoms, F (1, 49) = 9.85, p < .01. The skewness and kurtosis values indicated a fairly normal distribution of scores for children with less frequent seizur es (i.e., zero to five se izures in last year) and children with more frequent seizures (i.e., 6 to 12 seizures in last year) ( skewness = -0.27, -0.45; kurtosis = -1.03, -0.40, respectively).
80 Anxiety and Use of AED Anxious symptoms, as measured by the BASC-2 Child Form, had an overall mean of 52.09 ( SD = 7.85) for children treated with monotherapy and 60.53 ( SD = 7.95) for children treated with polyt herapy. These scores indicate d a statistically significant difference between self-report of anxiety a nd treatment with AEDs, with polytherapy indicating increased anxiety symptoms, F (1, 49) = 12.97, p < .001. The effect size for the difference in mean scores between children treated with monotherapy versus polytherapy was small for self-report ( d = 0.210). The skewness and kurtosi s values indicated a fairly normal distribution of scores for child ren with monotherapy and polytherapy ( skewness = -.32, -1.13; kurtosis = -.61, 1.00, respectively). Defining Characteristics of BASC2 Anxiety Scale: Parent Report Anxiety and Seizure Type Anxious symptoms, as measured by the BASC-2 Parent Form, had an overall mean of 58.10 ( SD = 9.34) for children with generalized seizures and 52.67 ( SD = 6.84) for children with partial seizures. These sc ores indicated a sta tistically significant difference between parent-report of anxiety and type of seizur e, with parentsÂ’ indicating increased anxiety symptoms for children with generalized seizures, F (1, 49) = 5.75, p < .020. The effect size for the difference in mean scores between children with generalized and partial seizures was sm all for parent report ( d = 0.11). The skewness and kurtosis values indicated a fairly norma l distribution of scores for children with generalized and partial seizures ( skewness = -0.05, 0.34; kurtosis = -1.43, -0.50, respectively).
81 Anxiety and Seizure Frequency Anxious symptoms, as measured by the BASC-2 Parent Form, had an overall mean of 51.07 ( SD = 6.57) for children with seizures 0 to 5 times in the last 12 months and 59.57 ( SD = 7.95) for children with 6 to 12 seiz ures within the previous year. These scores indicated a statistically significant di fference between parent report of anxiety and frequency of seizures, with children with more seizures (5-12 within the last year) indicating increased anxiety symptoms, F (1, 49) = 17.47, p < .001. The effect size for the difference in mean scores between children w ith more and less seizures was small for parent report ( d = 0.26). The skewness and kurtosis va lues indicated a fairly normal distribution of scores for childre n with less frequent seizures (i .e., zero to five seizures in last year) and children with more frequent seizures (i.e., 6 to 12 seizures in last year) ( skewness = 0.90, -0.28; kurtosis = 0.37, -0.58, respectively). Anxiety and Use of AEDs Anxious symptoms, as measured by the BASC-2 Parent Form, had an overall mean of 51.85 ( SD = 7.21) for children treated with monotherapy and 61.00 ( SD = 7.08) for children treated with polyt herapy. These scores indicate d a statistically significant difference between parent report of anxiety and treatment with AEDs, with polytherapy indicating increased anxiety symptoms, F (1, 49) = 18.45, p < .001. The effect size for the difference in mean scores between childre n with monotherapy versus polytherapy was small for parent report ( d = 0.27).The skewness and kurtosi s values indicated a fairly normal distribution of scores for child ren with monotherapy and polytherapy ( skewness = 0.80, -0.16; kurtosis = 0.05, -0.85, respectively). See Table 3 for additional data.
82 Defining Characteristics of BASC-2 De pression Scale: Child Self-Report Depression and Seizure Type Depressive symptoms, as measured by the BASC-2 Child Form, had an overall mean of 56.24 ( SD = 9.71) for children with generalized seizures and 53.63 ( SD = 8.74) for children with partial seizures. These scores did not indicate a sta tistically significant difference in depressive symp toms for children with gene ralized seizures and partial seizures, F (1, 49) = 1.00, p = .322. The effect size for the difference in mean scores between children with generalized and part ial seizures was small for self-report ( d = 0.02). The skewness and kurtosis values indicated a fairly normal distribution of scores for children with generalized and partial seizures ( skewness = 0.06, 0.35; kurtosis = -0.92, -0.59, respectively). Depression and Seizure Frequency Depressive symptoms, as measured by the BASC-2 Child Form, had an overall mean of 50.43 ( SD = 7.71) for children with seizures 0 to 5 times in the last 12 months and 59.91 ( SD = 8.10) for children with 6 to 12 seiz ures within the previous year. These scores indicated a statistically significant difference between self-report of depressive symptoms and frequency of seizures, with chil dren with more seizures (5-12 within the last year) indicating increas ed depressive symptoms, F (1, 49) = 18.27, p < .001. The effect size for the difference in mean scores between children with mo re and less seizures was small for self-report ( d = 0.27). The skewness and kurtosi s values indicated a fairly normal distribution of scores for children with less frequent seizures (i.e., zero to five seizures in last year) and children with more frequent seizures (i.e., 6 to 12 seizures in last year) ( skewness = 0.60, -0.09; kurtosis = -0.12, -0.77, respectively).
83 Depression and Use of AEDs Depressive symptoms, as measured by th e BASC-2 Child Form, had an overall mean of 50.82 ( SD = 7.49) for children treated with monotherapy and 62.47 ( SD = 7.05) for children treated with polyt herapy. These scores indicate d a statistically significant difference between self-report of depressive symptomatology and treatment with AEDs, with polytherapy indicating incr eased depressive symptoms, F (1, 49) = 28.48, p < .001. The effect size for the difference in mean scores between children with monotherapy versus polytherapy was me dium for self-report ( d = 0.37). The skewness and kurtosis values indicated a fairly nor mal distribution of scores for children with monotherapy and polytherapy ( skewness = 0.49, -0.08; kurtosis = -0.37, -0.90, respectively). Defining Characteristics of BASC-2 Depression Scale: Parent Report Depression and Seizure Type Depressive symptoms, as measured by the BASC-2 Parent Form, had an overall mean of 58.14 ( SD = 9.26) for children with generalized seizures and 56.97 ( SD = 5.35) for children with partial seizures. These scores did not indicate a statistically significant difference between parent-report of depressive symptomatology and type of seizure, with parentsÂ’ indicating increased de pressive symptoms for childre n with generalized seizures, F (1, 49) = 0.33, p < .569. The effect size for the difference in mean scores between children with generalized and partial se izures was small for parent report ( d = 0.01). The skewness and kurtosis values indi cated a fairly normal distri bution of scores for children with generalized and partial seizures ( skewness = 0.06, 0.75; kurtosis = -0.68, 1.91, respectively).
84 Depression and Seizure Frequency Depressive symptoms, as measured by the BASC-2 Parent Form, had an overall mean of 54.89 ( SD = 4.78) for children with seizures 05 times in the last 12 months and 60.57 ( SD = 8.36) for children with 6 to 12 seizures within the previous year. These scores indicated a statistically significant di fference between parent report of depressive symptomatology and frequency of seizures, with children with more seizures (5-12 within the last year) having in creased depressive symptoms, F (1, 49) = 9.24, p < .004. The effect size for the difference in mean sc ores between children with more and less seizures was small for parent report ( d = 0.16). The skewness and kurtosis values indicated a fairly normal distribution of scor es for children with less frequent seizures (i.e., zero to five seizures in last year) and children with more frequent seizures (i.e., 6 to 12 seizures in last year) ( skewness = -0.39, -0.21; kurtosis = -0.59, 0.01, respectively). Depression and Use of AEDs Depressive symptoms, as measured by the BASC-2 Parent Form, had an overall mean of 56.03 ( SD = 6.27) for children treated with monotherapy and 60.29 ( SD = 8.13) for children treated with polyt herapy. These scores indicate a statistically significant difference between parent report of depressive symptoms and treatment with AEDs, with polytherapy indicating increased depressive symptomatology, F (1, 49) = 4.29, p < .044. The effect size for the difference in mean scores between children treated with monotherapy versus polytherapy was small for parent report ( d = 0.08). The skewness and kurtosis values indicated a fairly normal distribution of scores for children with monotherapy and polytherapy ( skewness = 0.58, -0.25; kurtosis = 0.59, 1.04, respectively). See Table 3 below for additional data.
85 Table 3 Skewness and Kurtosis Coefficients for BASC-2 ( n = 51) Child Self-Report Parent Report Anxiety Scale Depression Scale Anxiety Scale Depression Scale Skewness -0.33 0.24 0.38 0.36 Std. Error of Skewness 0.33 0.33 0.33 0.33 Kurtosis -0.66 -0.81 -0.87 0.35 Std. Error of Kurtosis 0.66 0.66 0.66 0.66 Understanding Aspects of Health-Related Quality of Life in the Current Study Of the 51 child-parent participant pairs, sc ores derived from the Child Self-Report Form of the BASC-2 Anxiety Scale indicated that 68.6% ( n = 35) of the child participants fell in the norm al range for anxiety (i.e., T < 60), 29.4% ( n = 15) of the child participants fell in the border line range for anxiety (i.e., T = 60-69) and 2.0% (n = 1) fell in the clinically significant range (i.e., T = 70 or higher). On the BASC-2 Depression Scale, 68.6% ( n = 35) of child participants had scor es in the normal range of functioning, 23.6% ( n = 12) of child participants had scores in the borderline range (at-risk) for depression, and 7.8% ( n = 4) of child part icipants had self-reported scores in the clinically significant range for depression. Based on parent report, the BASC-2 Parent Forms revealed that 68.6% ( n = 35) of children obtained scores in the no rmal range for anxiety, 25.5% ( n = 13) of children obtained scores in the bord erline range, and 5.9% ( n = 3) of children obtained scores in
86 the clinically significant range for anxi ety. Regarding depressive symptomatology, parentsÂ’ ratings indicated that 70.6% ( n = 36) of children were in the normal range of functioning, while 23.5% ( n = 12) of children were functioning in the at-risk range for depression, and 5.9% ( n = 3) of children were functioni ng in the clinically significant range for depression. This res earcher hoped to discover whic h seizure-related variables and internalizing symptoms were most predicti ve of health-related quality of life. The following section will address the research questions for this study as presented in Chapter 1. Research Findings Question 1. What seizure-related variables (i .e., type of seizure, seizure frequency, and treatment with AEDs) and ps ychopathology (Â“at-ris kÂ” or Â“clinically significantÂ” range for anxiety an d/or depression) best predict health-related quality of life as reported by children 8 to 11 years of age diagnosed with epilepsy? To address this research question, a canoni cal correlation anal ysis was computed. However, prior to conducting this analysis the researcher want ed to ascertain the relationship between each independent and de pendent variable in the study. Bivariate correlations (i.e., PearsonÂ’s correlations) for all of the variables in the analysis are presented in Table 4. Because 45 correlations were of interest in Table 4, the Bonferroni adjustment had to be applied to ensure that the experiment-wise erro r rate did not exceed its nominal value of 5%. Thus, the adjusted Bonferroni level of .0011 (i.e., .05/45) was used. According to the PearsonÂ’s product mome nt correlation analysis, scores indicated a moderately strong, positive re lationship between ratings of seizure frequency and childrenÂ’s self-repor t of anxiety ( r = .41) and depression ( r = .52). All five subscales on
87 the HRQL measure had negativ e and moderately strong re lationships with seizure frequency (see Table 4). Treatment with AEDs had a moderate, positive relationship with childrenÂ’s ratings of anxiety ( r = .46) and a strong, positive relationship with childrenÂ’s ratings of depression ( r = .61). Furthermore, strong, ne gative relationships were found among all five HRQL subscales and treatment with AEDs. Ch ildrenÂ’s ratings of anxiety were moderately correlated with all five subs cales of the HRQL meas ure, with ratings of depression being very strongly correlated with the five HRQL subs cales (see Table 4 for list of all correlation coefficients). It is not eworthy that all five subscales from the HRQL measure were strongly correlated with one and another, with correlation coefficients ranging from r = .72 to .83. Given that a correlati onal relationship was found between ratings of anxiety and depression, it seem ed reasonable to proceed with canonical correlation analyses to determine whether a statistically significant relationship existed between the two certain sets of variables.
88 Table 4 PearsonÂ’s Correlations (Child Self-Report) Seiz Type Seiz Freq AED Anx CH Dep CH Social CH Pres CH Emot CH Secr CH Norm CH Seiz Type 1 Seiz Freq -.36* 1 AED -.42* .53* 1 Anx CH -.11 .41* .46* 1 Dep CH -.14 .52* .61* .66* 1 Social CH .20 -.58* -.69* -.58* -.78* 1 Present CH .12 -.53* -.59* -.63* -.77* .83* 1 Emot CH .18 -.42* -.59* -.63* -.74* .76* .80* 1 Secr CH .12 -.46* -.44* -.51* -.76* .72* .68* .82* 1 Norm CH .07 -.58* -.52* -.67* -.79* .83* .82* .83* .79* 1 Statistically significant at the Bonfe rroni-adjusted level of .0011 (i.e., .05/45).
89 Statistical Assumptions for Canonical Correlation Analysis The statistical assumptions presented a nd discussed in Chapter 3 were assessed (when feasible) to determine if the sp ecific assumptions pe rtaining to canonical correlation analysis were met. Univariate normality (i.e., skewness and kurtosis) was obtained for all variables ex amined in this study. However, one cannot assume that multivariate normality (i.e., multivariate sk ewness and kurtosis) also was obtained. Canonical Correlation Analysis (Child Report) Using canonical correlation analysis, a co mposite of the quality of life subscales that correlate with a composite of seizurerelated variable, anxi ety, and depression was derived. Two independent canonical correlation analyses were computed; one utilized data derived via self-report, and the other utilized data obtained via pare nt-report. The first analysis consisted of five independent variables (i.e., seizure type, seizure frequency, AED treatment, anxious symptoms [via self-report], and depr essive symptoms [via selfreport]) and five HRQL depende nt variablesÂ—all of which were derived from child selfreport (i.e., social, present, emotion, s ecrecy, and normalcy). Table 5 presents information about the first canonical functi on (the only function that was statistically significant). This first func tion had a canonical correlati on coefficient of .94, which indicates that the most variance is attributed to this first pair of linear combinations (statistically significant at p < .001). The second function, which had a canonical correlation coefficient of .55, had far less vari ance associated with it and therefore was not statistically significant using BartlettÂ’s approximate chi-square test ( 2 = 14.51, df = 16, p = .561).
90 The standardized function coefficients for the first canonical function also are presented in Table 5. These standardized coeffi cients indicate the relative contribution of each independent variable to the variance of its respective within-set canonical variable. An examination of the standardized function coefficients revealed that of the five independent variables (i.e., seizure type, seizure frequency, AED treatment, anxious symptoms, and depressive symptoms), only depression (as rated by child participants) made a significant contribution to the dependent variables composite (i .e., social, present, emotion, secrecy, and normalcy). Additionally, the structure coefficients ar e presented in Table 5 to provide more information about the relationships acro ss pairs of canonical variates. Structure coefficients are the correlations between a gi ven variable and the scores on the canonical composite (i.e., latent variable) in the set to which the variable be longs. Thus, structure coefficients indicate the degree of relationshi p of a given variable in the set with the canonical composite for the variable set (T hompson, 1984). Structure coefficients greater than + 0.30 were considered to be practically significant (Hair, Anderson, Tatham, & Black, 1998). The structure coefficients reve aled that four of the five independent variables (i.e., seizure frequency, AED treat ment, anxious symptoms, and depressive symptoms) made important contributions to th e first canonical variate. The size of the structure coefficient indicated that all four of these independent variables made very large contributions, with depression making the la rgest contribution. W ith regard to the dependent set, all five variables made la rge contributions, social making the largest contribution, followed closely by present.
91 Finally, the cross loadings for the first canonical function are presented in Table 5. The cross loadings are the most stable coe fficients for reliable interpretation because they are the product of each variableÂ’s canonical loading a nd the canonical correlation coefficient. Therefore, the cross loadings allow the most accurate/confident interpretation of relationships between independent and de pendent variables in each canonical function (Hair et al., 1998). Data gene rated from the cross loadings indicate that depression contributes the most to the relationship with the dependent variable composite. Additionally, seizure frequency, AED treat ment, and anxiety, respectively, make significant contributions on the first canonical function for the set of dependent variables. With respect to the dependent variables se t, all five variables made an important contribution to the composite set, with so cial and present making the most significant contribution, followed closely by normalc y, secrecy, and emotion, respectively.
92 Table 5 Canonical Correlations (Child Self-Report) for First Canonical Function Variable Can r Wilks 2 Df P Std. Coeff Struct Coeff Cross Ldgs 1 .937 .077 80.84 25 .0001 Independent Variables: Seize Type .11 -.25 -.24 Seize Freq .24 .84 .79 AED .16 .76 .72 Anx CH .02 .75 .71 Dep CH .70 .98 .92 Dependent Variables: Soc CH -.42 -.97 -.91 Pres CH -.36 -.95 -.89 Emot CH .01 -.85 -.80 Secr CH -.16 -.88 -.83 Norm CH -.13 -.92 -.86 Can r = canonical r Std. Coeff = standardized function coefficient Struct Coeff = structure coefficient Cross Ldgs = cross-loadings
93 Question 2. What seizure-related variables (i .e., type of seizure, seizure frequency, and treatment with AEDs) and ps ychopathology (Â“at-ris kÂ” or Â“clinically significantÂ” range for anxiety an d/or depression) best predict health-related quality of life as reported by parents of children 8 to 11 years of age diagnosed with epilepsy? This research question was addressed by computing a second canonical correlation analysis with data obtained via pare nt report. Prior to conducting this analysis, bivariate correlations (i.e., P earsonÂ’s correlations) were comput ed for all of the variables in the analysis. These results are presented in Table 6. As previously indicated, because 45 correlations were of interest in Table 4, the Bonferroni adjustment had to be applied to ensure that the experiment-wise error rate did not exceed its nominal value of 5%. Thus, the adjusted Bonferroni level of .0011 (i.e., .05/45) was used. According to the PearsonÂ’s product moment correlation analys is, scores indicated a moderately strong, negative relationship between ra tings of seizure type and parent report of anxiety ( r =.32). Seizure frequency had a moderately strong, positive relationship with anxiety ( r = .51) and depression ( r = .40). Additionally, treatment with AEDs had a moderately strong, positive relations hip with anxiety ( r = .52) and a small corr elation with depression ( r = .28; significant at the .05 alpha level). ParentsÂ’ ratings of anxiety were highly correlated with all five subscales of the HR QL measure, with corre lations ranging from 0.74 to -0.78. Parent ratings of depression al so were highly correlated with the five subscales of the HRQL measure, ranging fr om -0.58 for social concerns to -0.74 for secrecy issues. Seizure type had a small, positive correlation with four of the five subscales on the HRQL measure (i.e., social, future, em otion, and secrecy). Seizure frequency had
94 moderately strong, negative corre lations with all five subscal es on the HRQL measure. Finally, treatment with AEDs had strong, ne gative correlations with all five HRQL subscales (see Table 6). It is noteworthy that all five subscales fr om the HRQL measure were strongly correlated with one and another, with correlati on coefficients ranging from r = .83 to .93. Given that a correlational re lationship was found among ratings of seizurerelated variables, psychopathology, and HRQL it seemed reasonable to proceed with canonical correlation analyses to determine whet her a statistically significant relationship existed among certain sets of variables.
95 Table 6 PearsonÂ’s Correlations (Parent-Report) Seiz Type Seiz Freq AED Anx PR Dep PR Social PR Pres PR Fut PR Emot PR Secr PR Seiz Type 1 Seiz Freq -.36* 1 AED -.42* .53* 1 Anx PR -.32* .53* .52* 1 Dep PR -.08 .40* .28* .51* 1 Social PR .30* -.53* -.39* -.78* -.58* 1 Pres PR .26* -.60* -.46* -.76* -.73* .87* 1 Fut PR .33* -.60* -.53* -.78* -.67* .90* .93* 1 Emot PR .31* -.53* -.39* -.77* -.69* .83* .90* .88* 1 Secr PR .32* -.52* -.42* -.74* -.74* .87* .91* .91* .92* 1 Statistically significant at the Bonfe rroni-adjusted level of .0011 (i.e., .05/45).
96 Canonical Correlation Analysis (Parent Report) Using canonical correlation analysis, a co mposite of the quality of life subscales that correlate with a composite of seizurerelated variables, anxi ety, and depression was derived. The second (parent) analysis consisted of five independent va riables (i.e., seizure type, seizure frequency, AE D treatment, anxious sympto ms [via parent report], depressive symptoms [via pare nt report]) and five HRQL dependent variables; all of which were derived from parent report (i.e., social, present, future, emotion, and secrecy). Table 7 presents information about the first ca nonical function (the only function that was statistically significant). This first function had a canonical correlation coefficient of .927, which indicated that the most variance was attributed to this first pair of linear combinations (statistically significant at p < .001). The second func tion had a correlation coefficient of .533 had far less variance a ssociated with it and therefore was not statistically significant using Bartle ttÂ’s approximate chi-square test ( 2 = 20.26, df = 16, p = .209). The standardized function coefficients also are presented in Table 7. The standardized coefficients indicate the rela tive contribution of each independent variable to the variance of its respec tive within-set canonical variab le. The data suggest that depression ( r = .43) as rated by parent pa rticipants contribu tes the most to the relationship with the first canonical variable followed by seizure frequency ( r = .36) and anxiety ( r = .31). With respect to the dependent variable s set, present and emotion made moderate contributions to the composite set. Additionally, the structure coefficients ar e presented in Table 7 to provide more information about the relationships acro ss pairs of canonical variates. Structure
97 coefficients are the correlations between a gi ven variable and the scores on the canonical composite (i.e., latent variable) in the set to which the variable be longs. Thus, structure coefficients indicate the degree of relationshi p of a given variable in the set with the canonical composite for the variable set (T hompson, 1984). Structure coefficients greater than + 0.30 were considered to be practically significant (Hair et al., 1998). An examination of the structure coefficients re vealed that all five independent variables made significant contributions to the de pendent variables compositeÂ–with seizure frequency and anxiety making the largest c ontributions. The structure coefficients revealed that all five independent variable s made significant contributions to the first canonical variateÂ—with seizur e frequency and anxious symp toms making the largest contributions, and seizure type making the sm allest contribution. With regard to the dependent set, all five variables made very large contributions, with present making the largest contribution, followed closely by emotion, future, secrecy, and social, respectively. Finally, the cross loadings are presented in Table 7. An examination of the cross loadings revealed that seizure frequency ( r = .79), anxiety ( r = .78), and depression ( r = .74) made large contributions to the depende nt variables composite, with AED treatment making a moderate contribution to the composite ( r = .60). The fifth independent variable, seizure type, made the smallest (although still statis tically significant) contribution to the composite ( r = -.44). With respect to th e dependent variables set, all five variables made significant contributions to the composite set, with present concerns making the largest contribution, followed clos ely by future, emotion, secrecy, and social, respectively.
98 Summary of Canonical Correlation Analyses Results of the canonical correlation anal ysis examining data derived via child report revealed that four out of five independent vari ables assessed (i.e., seizure frequency, AED treatment, anxi ety, and depression) were significantly correlated with the five HRQL subscales. Specifically, depres sive symptomatology provided the largest contribution on the first canonical function fo r the set of dependent variables. Data derived via parent report indi cated that all five independe nt variables made significant contributions to the dependent variables com posite, with seizure frequency, anxiety, and depression making the largest contributions, fo llowed by a moderate contribution of the independent variable, AED treatment, and a relatively smaller contribution of the independent variable, seizure type. The impli cations of these findings are presented and discussed in the following chapter.
99 Table 7 Canonical Correlations (Parent-Repo rt) for First Canonical Function Variable Can r Wilks 2 Df P Std. Coeff Struct Coeff Cr Ld 1 .927 .074 81.90 25 .0001 Independent Variables: Seize Type -.10 -.47 -.44 Seize Freq .36 .85 .79 AED .07 .65 .60 Anx PR .31 .84 .78 Dep PR .43 .80 .74 Dependent Variables: Soc PR .04 -.87 -.81 Pres PR -.35 -.98 -.91 Future PR -.26 -.96 -.89 Emot PR -.33 -.96 -.89 Secrecy PR -.13 -.96 -.89
100 Chapter 5 Discussion Summary of Study The present study was conducted to expl ore which seizure-related variables and/or psychopathology variables were most significantly correlate d with health-related quality of life (HRQL) in children with ep ilepsy. This study was nove l in nature because it was one of only several studies to explore the relationship of anxi ety and depression on HRQL in children with epilepsy. Findings s uggested that anxiety and depression were statistically significantly correlated with HRQL for both self-repor t and parent report measures. Moreover, several seizure-related variables were found to correlate with the social and present concerns subscales on HRQL. That is, on self-report measures, both seizure frequency and number of antiepileptic drugs (AEDs) were correlated with HRQL. For parent report measures, seizure type frequency, and number of AEDs were correlated with HRQL. However, it is im portant to note that although statistical significance was found, it is questionable whet her these findings were clinically significant. That is, the majority of scores generated from childand parent-ratings fell in the sub-clinical range of f unctioning for anxiety and depression as measured on the BASC-2. This chapter summarizes the results from Chapter 4, discusses implications of the results, examines limitations, and s uggests directions for future research.
101 Relationship between Seizure-Related Variabl es and Health-Related Quality of Life Seizure Frequency and Health-R elated Quality of Life Seizure frequency, as rated by both child and parent-report, was found to be correlated with all five subs cales on the HRQL measure. Th ese findings indicate that children who have more frequent seizures (b etween 6 and 12 per year) are more likely to experience a lowered HRQL, especially associ ated with present concerns (i.e., current worries) and social concerns. The social s ubscale of the HRQL assesses interpersonal issues such as childrenÂ’s interactions with peers, their perceptions of friendships, and experiences of bullying behavior. The reduced HRQL scores could be relate d, in part, to the social stigma of epilepsy that has been well documented in the literature (Baker, 2001). The social stigma associated with epilepsy and seizures may st em from the unpredictability of behavior evident in individuals who e xperience seizures. Although the effect of race on healthrelated quality of life (HRQL) was not a fo cus of the current research, a significant percentage of minority children and parents participated in this study, and the potential impact of race on HRQL should not be overl ooked. In fact, Nei Wu et al. (2008) found that ethnic minority groups may face a double stigma (i.e., minority status and neurological condition) or a triple stigma (w ith the addition of mental health concerns). Minority groups may be more impacted by so cial concerns and/ or negative social interactions; therefore, these groups may be less likely to seek mental health services. Given that epilepsy has histor ically been stigmatized, it is not surprising that adding psychopathology/mental health difficulties exacerbate the perc eived stigma for
102 individuals who experience seiz ures. This may be an important variable to examine in future research studies th at assess HRQL in the pedi atric epilepsy population. Health-related locus of control (HLC) is defined as the degree to which individualsÂ’ believe that thei r health is controlled by intern al or external factors (Lau, 1982). In other words, HLC measures the degr ee to which an indi vidual believes that his/her health is or is not determined by his/her own behavior. HLC may play a significant factor in childrenÂ’s present conc erns regarding their epilepsy condition. For example, an individual with epilepsy typically has a poor sens e of an internal locus of control and a strong awareness of an external locus of contro l (Asadi-Pooya, Schilling, & Glosser et al., 2007). This is especially true for individuals with epilepsy because of the unpredictability of when their seizures w ill occur. This unpredictability may lead individuals with epilepsy to experience an increase in feelings of helplessness and hopelessness, both with their cogniti ve and behavioral functioning. Given the lack of control over their seizure occurrence and the subsequent externally-based locus of control, children with epilepsy would benefit from increasing their self-efficacy as well as self-reliance in an attempt to decrease their dependence on others (e.g., parents, clinicians) as experts. One method to increas e the self-efficacy of children with epilepsy would be to educat e children about their condition. Children should be knowledgeable about their me dical condition, including the symptoms, treatment options, and likely pr ognosis related to th eir seizure disorder. Education should be used as a first line of defense to incr ease childrenÂ’s understandi ng of their condition. Additionally, as is developmen tally-appropriate, children should play an active role in
103 their medical care because their involvement w ill likely lead to increased feelings of empowerment. Treatment with AEDs and Health -Related Quality of Life Results from the current study also suppor ted previous findings that number of AEDs and HRQL are correlated (Cushner-Wei nstein et al., 2008; Herranz, Armijo, & Artega, 1988). To analyze the relationship between child de pression and anti-epileptic drug treatment, Cushner-Weinstein and colleagues found a signi ficant correlation between rates of depression (specificall y, interpersonal prob lems) and polytherapy. Furthermore, these researchers found that children on polytherapy ha d scores indicating significant difficulties with interpersonal relationships, feelings of ineffectiveness, and negative self-esteem compared to scores of children on monotherapy. The current findings substantiate the relationship betw een number of AEDs and HRQL, indicating that increased numbers of AEDs are correlated with poorer quality of life. These findings highlight the importance of prescribing th e least number of AEDs possible that successfully control seizure activity. Seizure Type and Health-Related Quality of Life Type of seizure was found to be corre lated with HRQL based on the parent ratings of children with seizure disorders, whereas childrenÂ’s self-report on the HRQL subscales did not reveal a signi ficant relationship with seizur e type. It is important to note, however, that the correlation between pare nt ratings of HRQL and seizure type was small. Differing perceptions of parents and child ren is the most likely explanation for this finding. That is, the perceptions, concerns, and worries of parents are likely quite disparate from those of children (Eiser & Mo rse, 2001). Previous research has revealed
104 mixed findings regarding the eff ect of seizure type on the qu ality of life of individuals with epilepsy, with some researchers findi ng a significant correlation between type of seizure and HRQL (e.g., Baker, Gagnon, & Mc Nulty, 1998) and other researchers failing to find a significant relationship (e.g., J acoby, Baker, Steen, Potts, & Chadwick, 1996). Given the inconsistent findings regarding th e relationship between these two variables, future researchers should attempt to ascertain the effect of seizure type on the healthrelated quality of life of children with epilepsy. Relationship between Psychopathology, Se izure-related Variables, and HRQL Findings from the current study indicate that a statistically significant relationship exists among anxious symptomatology, seizur e frequency, and HRQL (both parent and self-reported). It has been pres umed that individualsÂ’ level of anxiety would be directly related to the number of seiz ures they experience, with a positive correlation between seizure frequency and anxiety. Therefore, it was expected that individuals with more frequent seizures would re port increased anxious sympto matology, and those with both frequent seizures and increas ed anxiety would, in turn, be correlated with a lowered health-related quality of life. However, researchers are now hypothesizing that it may not be the actual number of seizures that pred icts level of anxiety; instead, it may be the individualÂ’s perceived level of control that is correlated with anxiety (Pedroso de Souza & Salgado, 2006). That is, research ers have found that it is the perception with which individuals regard their seizure disorder that dictates their qu ality of life; this perception has a larger impact on HRQL than seizure-re lated variables such as type of seizure, seizure frequency, age of onset, and disease duration (Meador, 1993).
105 Depression, as rated by selfand pare nt-report, was found to be significantly correlated with lowered HRQL in the current study. This result was not surprising given the previous research that revealed increased rates of depression in the adult literature (Robertson & Trimble, 1983). It is important for clinicians to recognize that children may manifest symptoms of depression differently than adults. For example, children tend to present with disruptive behavi ors and/or irritability (Car lson & Cantwell, 1980; Ettinger et al., 1998). These symptoms ma y not be readily identified as symptoms of depression, making the identification of depression in children rather challenging. This reality highlights the need for practitioners to inquire directly about and assess specific depressive symptomatology. Previous studies have found large pe rcentages of adult patients with both unrecognized and untreated psychopathology; na mely depression (Boylan et al., 2004; OÂ’Donoghue et al., 1999). The few studies examining rates of de pression in children with epilepsy reveal similar results (Ettinger et al., 1998). Given that the current study, along with previous studies, indicate that depression has a significant negati ve effect on healthrelated quality of life in indi viduals with epilepsy, it is im portant to consider patientsÂ’ psychosocial functioning in addition to the stan dard seizure-related va riables in order to provide comprehensive care to this population. Limitations of the Study Internal and External Validity It is important to establish internal and external validity in a well-designed and effectively implemented study. However, the results of the current study have several possible threats to both intern al and external validity. One possible threat to internal
106 validity of the findings is instrumentati on (Onwuegbuzie, 2003); that is, because the instruments for this study are designed to obt ain information from participants via selfreport, the researcher will not be able to as certain with complete confidence the accuracy of the data collected. In addition, given that the data being request ed are sensitive in nature (and have potential ne gative connotations), there is an increased likelihood that participants would minimize psychopatholog ical symptomatology. Furthermore, there exists the possibility of a threat to construct-rela ted validity (Onwuegbuzie, 2003) because items on the questionnaire and/or rati ng scales may be unclear or confusing to some participants. However, the score reli ability of each scale and subscale on each measure has been previously validated. Differential selection of pa rticipants, also known as sele ction bias, may be a threat to the internal validity of th e findings. That is, there may be a notable distinction between participants who completed and returned th e questionnaires and ra ting scales and those who did not complete and return these form s (Best & Kahn, 2003). In addition, there may be a threat to historical valid ity, such that external events may impact the participantsÂ’ responses (Best & Kahn, 2003). Several potential thr eats to external validity of the results of this study exist as well. One possible threat to external valid ity is population valid ity (Best & Kahn, 2003), whereby results are generalized to populations that are not included in the study, such as children with medical conditions other than ep ilepsy or children with co-morbid medical conditions. Another threat to ex ternal validity is ecological validity. This refers to the extent to which findings can be generalized across settings, conditions, and contexts (Onwuegbuzie, 2003). To control for the threat to ecological validity, the results of this
107 study will not be generalized to children dia gnosed with epilepsy who do not also meet all inclusion criteria from this study. Temporal validity also may be a threat to external validity in this study. This refe rs to the extent that result s from a study are applicable across time. Therefore, it is important to c onsider the function of time during the period of data collection for th e study (Onwuegbuzie, 2003). Sample Size The sample size was minimally acceptable, and the principal investigator would have preferred to obtain a larger number of participants. However, the researcher had a rather small pool from which to draw th e sample given the specific inclusion and exclusion criterion. Thus, with th e small sample size, the ideal statistical procedures were not carried out because statistical significan ce may not have been ascertained. Given the smaller sample size obtained in this study, results should be interpreted with some caution as a larger sample size may have generated disparate results. It also is important to note the potent ial positive effects of medication on ratings of anxiety and depression. Re search findings have revealed that anti-epileptic drugs (AEDs) used to manage seizures may have an effect on mood, with AEDs improving mood in children and adults with epilep sy (Messenheimer, 2002; Uvebrant & Bauzien. 1994). Therefore, given that all participants in this research study used AEDs to manage seizures and AEDs may improve mood, thes e results may partially explain why the majority of selfand parent-re porters rated their le vels of anxiety and depression in the sub-clinical range of functioning (i.e., 68.6% to 70.6% of the sample received scores in the normal range).
108 Contributions of the Study Although there is a dearth of research in the area of pediatric epilepsy and psychopathology, the research that has been conducted has documented a statistically significant relationship between epilepsy a nd psychopathology, specifically depression and anxiety (Caplan, Siddarth, & Gurbani, 2005). The current study contributes to the sparse literature in this area of pediatri c research, by providing valuable information about the relationship among seizure-related va riables, anxiety, depression, and healthrelated quality of lif e in children diagnosed with epilepsy. It has been well established in the literature that health-related problems negatively impact childrenÂ’s academic perf ormance; therefore, the importance of addressing chronic health needs of children in schools is apparent (e.g., Power, Shapiro, & DuPaul, 2003). School psychologists are in th e position to make large contributions to the health promotion of children in schools a nd ensure that both educational and health needs are appropriately addr essed and successfully met. School psychologists should facilitate the coordination of educational, health, and mental health services for children with medical conditions (Power & Bl om-Hoffman, 2004). Additionally, school psychologists should play a role in the pr ovision of additional supports and services provided through IDEA and/or a Section 504 Accommodation Plan. Results from this study will hopefully highlight the importance of identifying, diagnosing, and treating children with epile psy who have co-morbid psychopathology and poor health-related qua lity of life so that they may ha ve the best possible educational and psychosocial outcomes. The Pediatric Psychosocial Preventative Health Model (PPPHM; Kazak, 2006) views pedi atric health care through th e view of a biopsychosocial
109 lens. That is, children and their families function in a multitude of complex systems, and when a child is affected with a chroni c health condition the normal course of development is interrupted. The PPHM has thr ee levels of prevention and intervention: universal, targeted, and clini cal/treatment. The universal le vel represents the largest group of children and families entering the health care system in which children and families are distressed yet remain resilient. At the universal level it is assumed that the individuals are functioning normally and have ad equate coping strategies; therefore, at this level general informati on and support is provided. A sm aller number of children and families require support and services at the second (Targeted) level. At this level, children and families are experiencing acute distress and risk factors are identified; therefore, support and services tailored to the familyÂ’s sp ecific needs are provided and distress is closely monitored. The third level (Clinical/Treatment) is reserved for children and families who are at-risk for persistent distress and for whom significant risk factors are identified. At this level, anxiety, depression, and other psychosocial difficulties may become apparent; therefore, these children and families receive the most intensive support and treatment. Given that other three-tiered models have been effectively introduced and implemented in the public school system (e.g., Response to Intervention), the PPPHM also could be successfully implemented in the school setting with ps ychologists and other mental health specialists playing important role s in the provision of services to students and their families. An effectual start may be to incorporate training for both psychologists and physicians regarding epilepsy and co -morbid psychopathology and its impact on HRQL. Physicians must work collaboratively with psychologists to address the medical
110 aspects and the psychosocial aspects of seiz ure disorders in children. Additionally, the importance of early, routine sc reenings to assess anxious a nd depressive symptomatology in children with seizure disorders should be emphasized. This effort could involve the combination of didactic training as well as supervisors and chief re sidents modeling best practices with the use of routine, early sc reenings for anxiety, depression, and healthrelated quality of life for all children diagnosed with epilepsy. Additionally, it is imperative that information regarding the medi cal and psychosocial aspects of epilepsy is communicated to parents so that they ar e well-informed, empowered, and able to advocate most effectively for their children. While interventions historically have focused on the medical management of seizure disorders, findings from this study unde rscore the importance of treating anxiety and depression in children with epilepsy. A dditionally, it is the hope of this researcher that children with epilepsy and their familie s will be informed of available preventative and early intervention services to address both the medical and psychosocial aspects of epilepsy. To provide a high quality level of care, practitioners need to incorporate prevention and intervention services that address childrenÂ’s ps ychosocial functioning rather than focusing solely on the medical ma nagement of their seiz ures. These services may help enable children to lead a life w ith minimal psychosocial difficulties stemming from epilepsy and improve thei r overall quality of life. Future Research Previous researchers have discovered that behavioral problems in children with epilepsy is a better predictor of parental depres sion than is the severity of the childrenÂ’s seizure condition (Shore, Austin, Huster, & Dunn, 2001). In light of these findings,
111 important data would be glean ed by investigating rates of depression in the parents of children with epilepsy as well as parental c oping skills. It would be advantageous for future researchers to glean information re garding the relationship between parental psychosocial functioning and the over all functioning of their children. Another important area that has not b een explored by researchers is what strategies children with epilepsy use to define their personal and social reality and how to implement strategies that would increase their sense of personal cont rol (Attarian, Vahle, & Carter, Hykes, & Gilliam 2003). That is, ch ildren with epilepsy would benefit from psychosocial interventions that would empower them and provide them with a stronger sense of support from their communities. Futu re studies must address the mechanisms that lead children to feel unsupported by their peers and community as well as what strategies could be implemented to build so cial support. Psychologi sts in education and medical settings are in the ideal posit ion to provide both psycho-education and psychosocial support to children with epilepsy within the various systems in which they function. Additionally, invaluable in formation could be gleaned regarding resilient factors that may protect children with epilepsy from psychopathology such as anxiety and depression. Future studies would benefit fr om examining resiliency in children and families inflicted with epilepsy to determine what factors help children become resilient. Moreover, information about resiliency in chil dren with epilepsy could be incorporated in early intervention services to decrease the pr obability that children with epilepsy will develop co-morbid psychopathology and lo wered health-related quality of life.
112 It is hoped that this study has laid the groundwork for subs equent studies to explore further the role of psychopathol ogy (especially anxiety and depression) on health-related quality of life in children with epilepsy. A future research path that would greatly augment the current research in this area would be to utilize a mixed methods research design, incorporating both quan titative and qualitative components. Furthermore, it may be beneficial to comple te a qualitative research study prior to a quantitative study design with children dia gnosed with seizure disorders. The abovementioned research components would enab le researchers to understand better how children with epilepsy conceptualize intern alizing disorders such as anxiety and depression. Although researchers have develope d measures to assess HRQL for specific pediatric populations, children may conceptua lize HRQL differently. Therefore, studies that include qualitative com ponents such as focus groups may be better aligned with childrenÂ’s own perceptions. Additionally, it is important that future researchers explore the relationship between parentsÂ’ and child renÂ’s ratings of psychopathology and HRQL. Adding a qualitative component, such as focus groups, would provide invaluable information about the similarities and differenc es between parent and child perceptions of epilepsy and its impact on psychosocial functioning and overall quality of life. Furthermore, future researchers may want to investigate additional variables that were not examined in this study. For exampl e, it may be advantageous to determine which factors contribute to in creased perceived control in children with epilepsy. It is critical to examine outcome variables in addi tion to health-related quality of life because it is likely that seizure disorders (especially with co-morbid anxiety and depression) have a broad impact on childrenÂ’s functioning. Fo r example, it would be beneficial to
113 understand the relationship between psyc hopathology and academic performance in children with seizures. This study could be re plicated with additiona l variables to assess the relationship among psychopathology, academic functioning, and HRQL. These data would allow for more comprehensive interv entions to alleviate negative emotions, cognitions, and behaviors as well as address va riables that impact the educational success of children with epilepsy. Future researchers also may benefit from including a larger age range of participants, perhaps including a dolescents to examine risk and resilience factors specific to an adolescent population. The inclusion of older children/adolescents also may provide greater insight regarding the effect of seizure-related variables and psychopathology on HRQL as younger children are likely less in trospective and may have more difficulty identifying anxious and depressive symp tomatology and the effect on their HRQL. Finally, another relationship that should be teased out is the correlation between the socio-economic status of families and the presence of psychopathology and/or lowered health-related quality of life in children with epilepsy. Determining the relationship between SES and childrenÂ’s psychosocial f unctioning may provide valuable information for the development of interventions and s upport services for child ren with epilepsy. Final Thoughts Psychopathology and poor health-related quality of life is a significant concern for children with epilepsy and their families. Given that children who manifest symptoms of anxiety and depression have signifi cantly lowered HRQL, it is apparent that psychopathology has implications for the mental health of children with epilepsy. This study was one of only several to examine inte rnalizing disorders a nd HRQL in children
114 with epilepsy. It was hypothesized that sign ificant correlations would be found between specific seizure-related variables, anxiety, depression, and the outco me variable (HRQL). It is imperative that clinicians and research ers alike utilize existi ng research findings to develop prevention and intervention programs for children at gr eatest risk for poor HRQL. The overall quality of life of children w ith epilepsy should be a primary focus of their treatment. It is hoped that findings genera ted from this study serve to be an impetus to developing interventions that will improve their quality of life. Given the importance of early identification and intervention for children with co-morbid epilepsy and psychopathology, it is critical that professiona ls who are in the position to identify these children obtain the knowledge and resources necessary to provide children with the supports and services needed as early as po ssible after seizure diagnosis. Practitioners working with children and families impacted by epilepsy have the privileged opportunity to identify and ameliorate the numerous challenges this popu lation endures. Although research findings reveal that as many as 33% of children with epilepsy have co-morbid psychopathology (Caplan et al., 2005), it is also eviden t in the literature that the majority of children with psychopa thology do not receive appropriate mental health services. If the relationship among seizure-related variables, internalizing disorders, and health-related quality of life is ascertaine d, then children with epilepsy who are at-risk for poorest HRQL may be iden tified early and receive mental health services as appropriate. It is quite clear that changes in treatment must occur so that children with epilepsy may receive adequa te medical and psychological services.
115 References Achenback, T. M. (2000). Child Behavior Checklist. In: Encyclopedia of psychology Vol. 2. Kazdin, Alan E.; Washington, DC, US: American Psychological Association, 69-70. Aldenkamp, A. P., Overweg-Plandsoen, W. C., & Arends, J. (1999). An open, nonrandomized clinical comparative study evaluating the effect of epilepsy on learning. Journal of Child Neurology, 14, 795-800. Annegers, J. F., Hauser, W. A., & Elvebac k, L. R. (1979). Remission of seizures and relapse in patients with epilepsy. Epilepsia, 20, 729-737. Arunkumar, G., Wyllie, E., Kotagal, P., Ong, H. T., & Gilliam, F. (2000) Parent and patient-validated content for pediatri c epilepsy quality-of-life assessment. Epilepsia, 41, 1474-1484. Asadi-Pooya, A. A., Schilling, C. A., Glosser, D., Tracy, J. I., & Sperling, M. R. (2007). Health locus of control in patients with epilepsy and its relationship to anxiety, depression, and seizure control. Epilepsy & Behavior, 11, 347-350. Attarian, H., Vahle, V., Carter, J., Hykes, E., & Gilliam, F. (2003). Relationship of depression & intractability of seizures. Epilepsy & Behavior, 4, 298-301. Austin, J. K., Dunn, D. W., Caffrey, H. M., Pe rkins, S. M., Harezlak, J., & Rose, D. F. (2002). Recurrent seizures and behavi or problems in children with first recognized seizures: A prospective study. Epilepsia, 43, 1564-1573.
116 Austin, J. K., Risinger, M. W., Beckett, L. A. (1992). Correlates of behaviour problems in children with epilepsy. Epilepsia, 33, 1115-1122. Austin, J. K., Smith, M. S., Risinger, M. W., & McNelis, A. M. (1994). Childhood epilepsy and asthma: Comparison of qualitFy of life. Epilepsia, 35, 608-615. Baker, G. A. (2001). Assessment of qualit y of life in people with epilepsy: Some practical implications. Epilepsia, 42 (3), 66-69. Baker, G. A., Gagnon, D., & McNulty, P. (1998). The relationship between seizure frequency, seizure type and quality of life: Findings from three European countries. Epilepsy Research, 30, 231-240. Baker, G. A., Spector, S., McGrath, Y., & So teriou, H. (2005). Impact of epilepsy in adolescence: A UK controlled study. Epilepsy & Behavior, 6 556-562. Bannon, M. J., & Ross, E. M. (1998). Administ ration of medicines in school: Who is responsible? British Medical Journal, 316, 1591-1593. Bastiaansen, D., Koot, H. M ., & Ferdinand, R. F. (2005). Psychopathology in children: Improvement of quality of life wit hout psychiatric symptom reduction? European Child & Adolescent Psychiatry, 14, 364-370. Bastiaansen, D., Koot, H. M., Ferdinand, R. F ., & Verhulst, F. C. (2004). Quality of life in children with psychiat ric disorders: Self-, pare nt, and clinician report. Journal of the American Academy of Ch ild & Adolescent Psychiatry, 43, 221-230. Berg, A. T., Shinnar, S., Levy, S. R., Testa, F. M., Smith-Rapaport, S., Beckerman, B., & Ebrahimi, N. (2001). Defining early seiz ure outcomes in pediatric epilepsy: The good, the bad and the in-between. Epilepsy Research, 43 75Â–84.
117 Besag, F. M. (1995). Myoclonus and infantile spasms. In M. M. Robertson & V. Eapen (Ed.), Movement and allied disorders in childhood (pp. 149-175). Oxford, England: John Wiley & Sons. Best, J. W., & Kahn, J. V. (2003). Research in education Boston: Pearson Education. Bethell, C. D., Read, D., St ein, R., Blumberg, S. J., Wells, N., & Newacheck, P. W. (2002). Identifying children with spec ial health needs: Development and evaluation of a short screening instrument. Ambulatory Pediatrics, 2, 38Â–48. Black, K. C., & Hynd, G. W. (1995). Epilepsy in the school aged child: Cognitivebehavioral characteristics and effects on academic performance. School Psychology Quarterly 10, 345-358. Blom, S., Heijbel, J., & Bergfors, P. G. ( 1978). Incidence of epilepsy in children: A followup study three years after the first seizure. Epilepsia 19, 343-350. Bordens, K. S., & Abbott, B. B. (1996). Research design and methods: A process approach (3rd ed.). Mountain View, CA: Mayfield. Bourgeois, B. F. (1998). Temporal l obe epilepsy in infants and children. Brain & Development, 20, 135-141. Boylan, L. S., Flint, L. A., Labovitz, D. L., J ackson, S. C., Starnter, K., & Devinksy, O. (2004). Depression but not seizure frequency predicts quality of life in treatmentresistant epilepsy. Neurology, 62, 258-261. Brady, E. U., & Kendall, P. C. (1992). Comorb idity of anxiety and depression in children and adolescents. Psychological Bulletin, 111, 244-255.
118 Calaminus, G., Weinspach, S., Teske, C., & Gobe l, U. (2000). Quality of life in children and adolescents with cancer. First results of an evaluation of 49 patients with the PEDQOL questionnaire. Klinische Padiatrie, 212 211Â–215. Camfield, C. S., Camfield, P. R., Gordon, K., Wirrell, E., & Dooley, J. M. (1996). Incidence of epilepsy in childhood a nd adolescence: A population-based study in Nova Scotia from 1977 to 1985. Epilepsia, 37, 19Â–23. Caplan, R., Siddarth, P., & Gurbani, S. ( 2005). Depression and Anxiety Disorders in Pediatric Epilepsy. Epilepsia, 46, 720-730. Caplan, R., Siddarth, P., Gurbani, S., Ott, D., Sankar, R., & Shields, W. D. (2004). Psychopathology and Pediatric Complex Pa rtial Seizures: Se izure-related, Cognitive, and Linguistic Variables. Epilepsia, 45, 1273-1281. Carlson, G., & Cantwell, D. (1980). Unma sking masked depression in children and adolescents. American Journal of Psychiatry, 137, 445-449. Carpay, H. A., & Arts, W. F. (1996). Outc ome assessment in epilepsy: Available rating scales for adults and methodological i ssues pertaining to the development of scales for childhood epilepsy. Epilepsy Research, 24, 127-136. Chadwick, D. (1994). Epilepsy. Journal of Neurology, Ne urosurgery & Psychiatry 57, 264-277. Collaborative Group for Epidemiology of Epilepsy. (1986). Adverse reactions to antiepileptic drugs: A multi-center survey of clinical practice. Epilepsia, 27, 323330. Cowan, L. D. (2002). The epidemiol ogy of the epilepsies in children. Mental Retardation and Developmental Disabilities Research Reviews, 8, 171Â–181.
119 Cramer, J. A., Blum, D., & Reed, M. (2003) The influence of comorbid depression on seizure severity. The Epilepsy Impact Project Group; Epilepsia, 44, 1578-1584. Cramer, J. A., Westbrook, L. E., Devins ky, 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 1114-1121. Cushner-Weinstein, S., Dassoulas, K., Salpekar J. A., Henderson, S. E., Pearl, P. L., Gaillard, W. D., & Weinstein, S. L. (2008). Parenting stress and childhood epilepsy: The impact of depression, l earning, and seizure-related factors. Epilepsy & Behavior, 13, 109-114. Davies, S., Heyman, I., & Goodman, R. (2003) A population survey of mental health problems in children with epilepsy. Developmental Medicine & Child Neurology, 45, 292Â–295. Davidoff, A. J. (2004). Identifying children with special health care needs in the National Health Interview Survey: A new resource for policy analysis. Health Services Research, 39, 53-72. Devinsky, O. (2003). Psychiatric comorbidity in patients with epilepsy: Implications for diagnosis and treatment. Epilepsy & Behavior, 4 S2-S10. Devinsky, O., Westbrook, L., Cramer, J., Gla ssman, M., Perrine, K., & Camfield, C. (1999). Risk factors for poor health-relat ed quality of life in adolescents with epilepsy. Epilepsia 40, 1715-1720. Doyle, A., Ostrander, R., & Skare, S. (1997). Convergent and criterionrelated validity of the Behavior Assessment System for Children-Parent Rating Scale. Journal of Clinical Child Psychology, 26, 276-284.
120 Duchowny, M. (1993). Febrile seizures in chil dhood. In E. Wyllie (Ed.) The treatment of epilepsy: Principles and practice. (p p. 647-653). Philadelphia: Lea & Febiger. Duchowny, M., & Harvey, A. S. (1996). Pedi atric epilepsy syndromes: An update and critical review. Epilepsia, 37(S1), S26-40. DuLac, O., MacDonald, R. L., & Kelly, K. M. (1995). Age-specific antiepileptic drug treatment and development of age-specific antiepileptic drugs. Brain development and epilepsy. New York: Oxford University. Eiser, C., & Morse, R. (2001). Can parents rate their child's health-related qua lity of life? Results of a systematic review. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care & Rehabilitation, 10, 347357. Eriksson, K. J., & Koivikko, M. J. (1997). St atus epilepticus in children: Aetiology, treatment, and outcome. Developmental Medicine & Child Neurology, 39, 652658. Ettinger, A. B., Weisbrot, D. M., Nolan, E. E ., Gadow, K. D., Vitale, S. A., Andriola, M. R., & Lenn, N. J. (1998). Symptoms of depression and anxiety in pediatric epilepsy patients. Epilepsia, 39, 595-599. Fiordelli, E., Beghi, E., Bogliun, G., & Cr espi, V. (1993). Epilepsy and psychiatric disturbance. A cross-sectional study. The British Journal of Psychiatry, 163, 44650. First Seizure Trial Group. (1993). Randomized clinical trial on the efficacy of antiepileptic drugs in reducing the risk of relapse after a first unprovoked tonicclonic seizure. Neurology, 43, 478-483.
121 Flanagan, R., & Esquivel, G. B. (2006). Empiri cal and clinical methods in the assessment of personality and psychopathology: An integrative approach for training. Psychology in the Schools, 43, 513-526. Gilliam, F., Wyllie, E., Kashden, J., Faught, E., Kotagal, P. et al. (1997). Epilepsy surgery outcome: Comprehensive assessment in children. Neurology, 48 13681374. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate Data Analysis, 5th edition. Prentice Hall, Inc. Hart, R. G., & Easton, J. D. (1986). Seizure recurrence after a firs t, unprovoked seizure. Archives of Neurology 43, 1289-1290. Hauser, W. A. (1986). Should people be treated after a first seizure? Archives of Neurology 43, 1287-1288. Hauser, W. A., Annegers, J. F., Rocca, W. A. (1996). Descriptive epidemiology of epilepsy: contributions of population-base d studies from Rochester, Minnesota. Mayo Clinic Proceedings, 71, 576-586. Hauser, W. A., Anderson, V. E., Loewenson, R. B., et al. (1982). Seizure recurrence after a first unprovoked seizure. New England Journal of Medicine 307, 522-528. Hauser, W. A., Rich, S. S., Annegers, J. F ., et al. (1990). Seizure recurrence after a 1st unprovoked seizure: An extended follow-up. Neurology 40, 1163-1170. Heller, K. W., Alberto, P. A., & Forney, P. E. (1996). Understanding physical, sensory, and health impairments: Characteri stics and educationa l implications. Belmont, CA, US: Thomson Brooks/Cole Publishing Co.
122 Hermann, B. P., Seidenberg, M., & Bell, B. (2000). Psychiatric comorbidity in chronic epilepsy: Identification, consequences and treatment of major depression. Epilepsia 41(Sl), S31Â–41. Herranz, J. L., Armijo, J. A., & Artega, R. ( 1988). Clinical side eff ects of Phenobarbital, pirimidone phenytoin, carbamazepine and valproate during montherapy in children. Epilepsia, 29, 794-804. Hiemenz, J. R., Hynd, G. W., & Jimenez, M. (1999). Seizure disorders. In R. T. Brown (Ed.), Cognitive aspects of chr onic illness in children (pp. 238-261). New York: Guilford. Hofstra, M. B., Van der Ende, J., & Verhulst F. C. (2000). Continuity and change of psychopathology from childhood into adulthood: A 14-year follow-up study. Journal of the American Academy of Child & Adolescent Psychiatry, 39, 850-858. Holmbeck, G. N., Belvedere, M. C., Christ iansen, M., Czerwinski, A. M., Hommeyer, J. S., et al. (1998). Assessment of adhe rence with multiple informants in preadolescents with spina bifida: Init ial development of a multidimensional, multitask parent-report questionnaire. Journal of Personality Assessment 70 427Â–440. Hopkins, A., Garman, A., & Clark, C. (1988). Th e first seizure in adult life: Value of clinical features, electroencephalo graphy, and computerized topographic scanning in prediction of seizure recurrence. The Lancet 1(8588), 721-726.
123 Ho-Turner, M., & Bennett, T. L. (1999). Seiz ure disorders. In S. Goldstein & C. R. Reynolds (Eds.), Handbook of neurodevelopmental and genetic disorders (pp. 499-524) New York: Guilford Press. Howe, G. W., Feinstein, C., & Reiss, D. (1993). Adolescent adjustment to chronic physical disorders: I. Comparing ne urological and non-neurological conditions. Journal of Child Psychology and Psychiatry, 34, 1153-1171. Huberty, T. J., Austin, J. K., & Huster, G. A. (2000). Relations of change in condition severity and school self-c oncept to change in achievement-related behavior in children with asthma or epilepsy. Journal of School Psychology, 38, 259-276. Hulihan, J. F. (1997). Seizures in special populations: Children, the elderly, and patients with coexistent medical illness. Postgraduate Medicine, 102, 165-168, 171172, 175-176. Jackson, M. J., & Turkington, D. (2005) Depression and anxiety in epilepsy. Journal of Neurology, Neurosurgery and Psychiatry, 76, 45-47. Jacoby, A., Baker, G., Steen, N., Potts, P., & Chadwick, D. (1996). The clinical course of epilepsy and its psychos ocial correlates: Findings from a UK community study. Epilepsia, 37, 148-161. Kaiser, S., Selai, C. E., & Trimble, M. R. (2002). Long-term follow-up of topiramate and lamotrigine: A perspective on quality of life. Seizure, 11, 356-360. Kamphaus, R. W., & Frick, P. J. (1996). Clinical assessment of child and adolescent personality and behavior. Needham Heights, MA, US: Allyn & Bacon. Kanner, A. M., & Palac, S. (2000). Depr ession in epilepsy: A common but often unrecognized cormorbid malady. Epilepsy & Behavior, 1, 37-51.
124 Kazak, A. (2006). Pediatric Psychosocial Preventative Health Model (PPPHM): Research, practice and collaboration in pediatric family systems medicine. Families, Systems and Health, 24, 381-395. Ketter, T. A., Post, R. M., & Theodore, W. H. (1999). Positive and negative psychiatric effects of antiepileptic drugs in patients with seizure disorders. Neurology, 53, S53-S67. Kjeldsen, M. J., Kyvik, K. O., Christense n, K., & Friis, M. L. (2001). Genetic and environmental influence in the etiology of epilepsy: A population-based study of 11,900 Danish twin pairs. Epilepsy Research, 44, 167-178. Kurtz, Z., & Tookey, R. (1998). Epilepsy in young: 23 year follow up of the British National Child Development Study. British Medical Journal, 316, 339-342. Labiner, D. M., & Ahern, G. L. (2002). Neur ologic consultation for seizures: When, why, and how to pursue. Postgraduate Medicine, 111, 53-64. Lambert, M. V., & Robertson, M. M. (1999). Depression in epilepsy: Etiology, phenomenology and treatment. Epilepsia, 40, S21Â–S47. Landgraf, N.J., Abetz, L., & Ware, J.E. (1996) The child health que stionnaire (CHQ): A user's manual. Boston, The Health Institute, New England Medical Center Lavigne, J. V., & Faier-Routman, J. (1992) Psychological adjust ment to pediatric physical disorders: A meta-analytic review. Journal of Pediatric Psychology, 17, 133Â–157. Lhatoo, S. D., & Sander, J. W. (2001). Th e epidemiology of epilepsy and learning disability. Epilepsia, 42, 6-9.
125 McDermott, S., Mani, S., & Krishnaswami, S. (1995). A population-based analysis of specific behavior problems asso ciated with childhood seizures. Journal of Epilepsy, 8, 110-118. Meador, K. J. (1993). Research use of the new quality of life of epilepsy inventory. Epilepsia, 34, S34-38. Mendez, M. F., Doss, R. C., Taylor, J. L., & Salguero, P. (1993). Depression in epilepsy. Relationship to seizures and anticonvulsant therapy. Journal of Nervous and Mental Disease, 181, 444-447. Merrell, K. W. (1999). Behavioral, social, and emoti onal assessment of children and adolescents. Mahwah, NJ: Erlbaum. Messenheimer, J. (2002). Efficacy and safety of lamotrigine in pediatric patients. Journal of Child Neurology, 17, 2S34-2S42. Miller, V., Palermo, T. M., & Grewe, S. D. (2003). Quality of life in pediatric epilepsy: Demographic and disease-related pred ictors and comparison with healthy controls. Epilepsy & Behavior, 4, 36-42. Murphy, C. C., Yeargin-Allsopp, M., & D ecoufle, P. (1995). The administrative prevalence of mental retard ation in 10-year-old childre n in metropolitan Atlanta, 1985 through 1987. American Journal of Public Health, 85, 319-323. Nei Wu, K., Lieber, E., Siddarth, P., Smith, K ., Sankar, R., & Caplan, R. (2008). Dealing with epilepsy: Parents speak up. Epilepsy & Behavior, 13, 131-138. Noll, R. B., Gartstein, M. A., Vannatta, K., Correll, J., Bukowski, W. M., & Davies, W. H. (1999). Social, emotional, and behavi oral functioning of children with cancer. Pediatrics, 103 71Â–78.
126 Oguz, A., Kurul, S., & Dirik, E. (2002). Re lationship of epilepsy-related factors to anxiety and depression scor es in epileptic children. Journal of Child Neurology, 17, 37-40. Onwuegbuzie, A. J. (2003). Expanding the framew ork of internal and external validity in quantitative research. Research in the Schools, 10 (1), 71-90. Osborne, J. W., & Waters, E. (2002). Four assumptions of multiple regression that researchers should always test. Practical Assessment, Research & Evaluation 8(2). Retrieved March 20, 2007, from h ttp://PAREonline.net/getvn.asp?v=8&n=2. Plioplys, S. (2003). Depression in child ren and adolescents with epilepsy. Epilepsy & Behavior, 4 S39-45. Power, T. J., & Blom-Hoffman, J. (2004). The school as a venue for managing and preventing health problems: O pportunities and challenges. Handbook of pediatric psychology in school settings. R. T. Brown (eds.). Mahwah, NJ: Lawrence Erlbaum Publishers. Prego-Lopez, M., & Devinsky, O. (2002). Evalua tion of a first seizur e: Is it epilepsy? Postgraduate Medicine 111 (1), 34-36, 43-48. Quittner, A. L. (1998). Measurement of quality of life in cystic fibrosis. Current Opinion in Pulmonary Medicine, 4, 326-331. Reynolds, C. R. & Kamphaus, R. W. (2004). Behavior Assessment System for Children, (2nd ed.) Circle Pines, MN: AGS Publishing, Robertson, M. M., & Trimble, M. R. (1983). Depressive illness in patients with epilepsy: A review. Epilepsia, 22, 515-524.
127 Rodenburg, H. R., Stams, G. J., Meijer, A. M., Aldenkamp, A. P., & Dekovic, M. (2005). Psychopathology in children with epilepsy: A meta-analysis. Journal of Pediatric Psychology, 30, 453-468.Ronen, G. M., Rosenbaum, P., & Streiner, D. L. (2000). Outcome measures in pediatric neurology: Why do we need them? Journal of Child Neurology, 15, 775-780. Ronen, G. M., Streiner, D. L., & Rosenbaum, P. (2003). Health-relate d quality of life in childhood epilepsy: Moving beyond seizure c ontrol with minimal adverse effects. Health Quality of Life Outcomes, 1 36-46. Sachs, H., & Barrett, R. P. (1995). Seizure di sorders: A review for school psychologists. School Psychology Review, 24, 131-145. Sawyer, M. G., Whaites, L., & Rey, J. M. (2002). Health-related quality of life of children and adolescents with mental disorders. Journal of the American Academy of Child & Adolescent Psychiatry, 41, 530-537. Scheuer, M. L. (1992). Medical patients with epilepsy. In S. R. Resor& H. Kutt (Eds.), The medical treatment of epilepsy (pp. 557-572). New York: Dekker. Schipper, H. (1990). Quality of life: Pr inciples of the clinical paradigm. Journal of Psychosocial Oncology, 8, 171-185. Seidenberg, M., & Berent, S. (1992). Ch ildhood epilepsy and the role of psychology. American Psychologist, 47, 1130-1133. Shafer, P. O. (2002). Improving the quality of life in epilepsy: Nonmedical issues too often overlooked. Postgraduate Medicine 111 (1). Retrieved March 20, 2007, from http://www.postgradmed.com/issues/2002/01_02/shafer.htm.
128 Shore, C. P., Austin, J. K., Huster, G. A., & Dunn, D. W. (2001). Identifying risk factors for maternal depression in families of adolescents with epilepsy. Journal for Specialists in Pediatric Nursing, 7, 71-80. Spieth, L. E., & Harris, C. V. (1996). Assessm ent of health-related quality of life in children and adolescents: An integrative review. Journal of Pediatric Psychology, 21, 175-193. Sturniolo, M. G., & Galletti, F. (1994). Couns eling children and parent s about epilepsy. Patient Education and Counseling, 55, 422-425. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.) Needham Heights, MA: Allyn & Bacon. Thompson, B. (1984). Canonical correlation analysis: Uses and interpretations. Newbury Park, CA, Sage. (ERIC Do cument Reproduction Service No. ED199269). Thompson, R. J., & Gustafson, K. E. (1996). Psychological adjustment. In adaptation to chronic childhood illness Washington, DC, US: American Psychological Association, pp. 57-86. Uvebrant, P. & Bauzien, P. R. (1994). Intrac table epilepsy in children. The efficacy of lamotrigine treatment, including non-seizure-related benefits. Neuropediatrics 25, 284Â–289. Van Dyck, P. C., McPherson, M., Strickland, B. B., Nesseler, K., Blumberg, S. J., Cynamon, M. L., & Newacheck, P. W. (2002). The national survey of children with special health care needs. Ambulatory Pediatrics, 2, 29Â–37.
129 Varni, J. W., Seid, M., & Kurtin, P. S. (2001). PedsQL-super(TM ) 4.0: Reliability and validity of the Pediatric Quality of Life Inventory-super(TM ) Version 4.0 Generic Core Scales in h ealthy and patient populations. Medical Care, 39, 800812. Vazquez, B., & Devinsky, O. (2003). Epilepsy and anxiety. Epilepsy & Behavior, 4, S20S25. Wagner, J. L., & Smith, G. (2006). Psychosoc ial intervention in pe diatric epilepsy: A critique of the literature. Epilepsy & Behavior, 8, 39Â–49. Williams, J., Sharp, G.B., DelosReyes, E., Bate s, S., Phillips, T., et al. (2002). Symptom differences in children with abse nce seizures versus inattention. Epilepsy & Behavior, 3, 245-248. Zupanc, M. L. (1996). Update on ep ilepsy in pediatric patients. Mayo Clinic Proceedings 71, 899-916.
131 Appendix A: Demographics and Seizure Variables Questionnaire Demographics and Seizure-Re lated Variables Questionnaire Please read the items below and circle or write-in one response for each item. If you have any questions about this form, please ask for assistance. (1) My childs age is: ________________ (2) My childs gender is: Male Female (3) My childs race/ethnic identity is: African American/Black Hispanic/Latino Asian American Native American Caucasian/White Mixed-Race or Other (4) Type of medical insurance: Private Medicaid None (5a) Does your child have any medical/physical diagnoses in addition to epilepsy? (for example: cystic fibrosis, diabetes, asthma) Yes No (5b) If you answered yes to (5a), please provide the diagnosis: ________________________ (6a) Does your child have any em otional/behavioral diagnoses? (for example: attention deficit-hyperactivity disorder, autism, depression) Yes No (6b) If you answered yes to (6a), please provide the diagnosis: ________________________ (7) Please circle the type of seizure your child experiences: Generalized seizures Partial seizures (8) Number of seizures in the past 12 months : 0-5 per year 6-12 per year (9) Please list all medications your child ha s been taking for the past 3 months: ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________
132 Appendix B: Child Cover Letter and Assent Form Hello! You and your parent are being asked to take part in a research study by filling out several surveys. I am trying to find out what types of things impact the quality of life of children with epilepsy. Why You Should Take Part in the Study : I want to learn more about what things impact the lives of kids with epilepsy. Any information yo u give will be confidential: that means it is private. Also, you and your parent will be en tered into a drawing to win a $100 gift card! Filling Out the Surveys : These surveys will ask about your thoughts, behaviors, and attitudes about different things. I expect it will take you about 30 minutes to fill out the surveys. Confidentiality (Privacy) of Your Responses : I do not expect that there will be more than minimal risk to you for taking part in this research study. The information you provide will be kept confidential (private, secret). Your surv eys will have a code number to protect the privacy of your responses. Please Note : Your involvement in this study is voluntary: you get to decide if you want to fill out the surveys. By signing this form, you ar e agreeing to take part in this research. If you choose not to participate, or if you want to stop participating at any time, you will not be punished in any way. You are free to stop participating in this study at any time. What WeÂ’ll Do With Your Responses : I will use the information from this study to let others know what sort of things impact the lives of kids with epilepsy. The results of this study may be published; however, your responses will be combined with responses from other people. The published results will not include your name or any other information that would in any way identify you. Questions? If you have any questions about this research study, you may contact me (Ms. Aja Meyer) at (813) 810-2526. If you have qu estions about your rights as a person who is taking part in a research study, you may contact a member of the Division of Research Compliance of the University of South Florida at 813-974-5638 or the Department of Health and Human Services. Thank you for taking the time to take part in this study. Sincerely, Aja Meyer, M.A. Doctoral Candidate Psychological and Social Foundations University of South Florida
133 Appendix B: (Continued) Assent to Take Part in this Research Study I freely give my permission to take part in this study. I understand that this is research. I have received a copy of this letter and assent form for my records. ________________________ ________________________ ____________ Signature of child Printed name of child Date taking part in the study Statement of Person Obtaining Informed Assent I certify that participants have been provided w ith an informed assent form that has been approved by the University of South FloridaÂ’s In stitutional Review Board and that explains the nature, demands, risks, and benefits involved in pa rticipating in this study. I further certify that a phone number has been provided in the event of additional questions. ________________________ ________________________ ___________ Signature of person Printe d name of person Date obtaining assent obtaining assent Investigator Statement: I certify that participants have been provided w ith an informed consen t form that has been approved by the University of South FloridaÂ’s In stitutional Review Board and that explains the nature, demands, risks, and benefits involved in pa rticipating in this study. I further certify that a phone number has been provided in the event of additional questions. _________________________ ________________________ _____________ Signature of Investigator Printed Name of Investigator Date
134 Appendix C: Parent Cover Letter and Consent Form Dear Parent or Guardian: This letter provides information about a research stud y that will be conducted at Dr. FernandezÂ’s office by Aja M. Meyer, M.A., a doctoral candidate from the Un iversity of South Florida. Please carefully read through this information and sign the bottom of the form. Purpose of study: The purpose of the study is to determine whic h factors are correlated with quality of life in children with epilepsy so they may receive approp riate supports and services to improve their physical and psychosocial well-being. This study is part of my dissertation research entitled, Â“Impact of SeizureRelated Variables and Psychopathology on Health-R elated Quality of Life in Pediatric Epilepsy.Â” Information from this study will inform educators and psychologists about the relationship between seizure variables and internalizing disorders (i.e., anxiety and depression) and the health-related quality of life of children with epilepsy. What participation requires: If you and your child agree to take part in this study, you both will be requested to complete several forms. You will be aske d to complete a short questionnaire requesting basic information about your childÂ’s medical condition, a rating scale that assesses your childÂ’s behavioral and emotional symptoms, and a rating scale that measures your childÂ’s health-related quality of life. These forms should take approximately 40 minutes to complete. Your child will be asked to complete the child version of the two rating scales. It should take your ch ild about 30 minutes to complete these scales. Please note that all forms should be completed independently. However, if your child as ks for help with some items from the forms, you may assist them, but you should not provide the responses for them. Risks and benefit s : There are no direct risks or benefits to you or your child for participating in this study; however, by participating you will provide valuable information about the factors that impact the quality of life of children with epilepsy. If the researcher should find that your child is at-risk for depression or anxiety (based on scores obtained on the BASC-2 ques tionnaire), Dr. Fernandez will be informed that code number participant Â“XXXÂ” is in the at-risk range for depression and/or anxiety. Dr. Fernandez will match the participant with the code number on the consent form and may refer the patient for further evaluation. Compensation : Each child/guardian pair who participates in this study will be included in a drawing to win a $100 gift card. Confidentiality : There are federal laws that say I must keep your study records private. Your records will be kept private by not asking for information that could identify you or your child. However, certain people may need to see your study records, and they must keep your records completely confidential. The only people who are allowed to view your records are the research team members, the University of South Florida Institutional Review Board staff, and governme nt agencies who make sure that your rights and safety are being protected. I may publish what I learn from this study; however, nothing will be published that would let people know who you are. Voluntary Participation: You and your childÂ’s participation in th is study is completely voluntary. Your decision to take part in this study will not affect your current or future relationship with Dr. Fernandez. If at any time you have questions or concerns regarding this study, please contact the principal investigator (Aja M. Meyer) at 813-810-2526 or email@example.com. If you have questions or concerns about you or your child's rights as research subjects, you may contact the Institutional Review Board at the University of South Florida at 813-974-5638 or the Department of H ealth and Human Services. You may retain this letter for your personal records, if desired. Aja M. Meyer, M.A. Psychological and Social Foundations University of South Florida
135 Appendix C: (Continued) Consent for Child to Take Part in this Research Study I give consent for myself and my child to participate in this study. ___________________ ______________________ __________ Signature of Participant Printed Name of Participant Date I do NOT give consent for myself and/or my ch ild to participate in this study. ___________________ ______________________ __________ Signature of Participant Printed Name of Participant Date Investigator Statement: I certify that participants have been provided w ith an informed consen t form that has been approved by the University of South FloridaÂ’s In stitutional Review Board and that explains the nature, demands, risks, and benefits involved in par ticipating in this study. I further certify that a phone number has been provided in the event of additional questions. _________________________ _________________________ _____________ Signature of Investigator Printed Name of Investigator Date
136 Appendix D: Health-Related Quality of Life: Parent Questionnaire
137 Appendix D: (Continued)
138 Appendix D: (Continued)
139 Appendix D: (Continued)
140 Appendix D: (Continued)
141 Appendix E: Health-Related Qualit y of Life: Child Questionnaire
142 Appendix E: (Continued)
143 Appendix E: (Continued)
144 Appendix E: (Continued)
145 Appendix E: (Continued)
About the Author Aja M. Meyer graduated with honors from the University of Fl orida in 2000 with a BachelorÂ’s of Arts Degree in Psychology. She earned her Ph.D. in School Psychology from the University of South Florida in 2008, specializing in pediatri c health issues. Aja completed her pre-doctoral internship at the ChildrenÂ’s Hospital of Michigan and currently is completing a post-doctoral fell owship at Akron ChildrenÂ’s Hospital. Aja specializes in working with children who have neurological impairments and plans to pursue a career within a pe diatric medical setting. During her graduate studies, Aja gain ed experience working in school and medical settings with children and adolescen ts. She received advanced training at the Early Steps Clinic through the University of South FloridaÂ’ s School of Medicine and at the Silver Child Development Center. Aja also conducted research in the areas of autism spectrum disorders and epilepsy and has pres ented her findings at state and national conferences.