1 EMOTIONAL AND BEHAVIORAL FUNCTIONING OF SIBLINGS OF PEDIATRIC CANCER PATIENTS: DISCORDANCE BETWEEN PARENT AND SELF REPORT AND ASSOCIATIONS WITH FAMILY FUNCTIONING By RYAN EMILY FUSSELL A DISSERTATION PRESENTED TO THE GRADUA TE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 2
2 2012 Ryan E mily Fussell
3 ACKNOWLEDGMENTS I would first and foremost like to thank my chair, Dr. S helley Heaton and my co chair, Dr. Sheila Eyberg for their guidance and support over the last few years. I appreciate the opportunity you both have given me to pursue a field of study that I am passionate about despite it not falling within your areas of i nter est. I have learned so much under your tutelage and feel immensely grateful for having the opportunity to work with both of you. I would also like to thank my committee members for their advice and support thr oughout this process: Stephen Boggs, Ph.D., Deidre Pereira, Ph.D., Jeff Farr ar, Ph.D., and William Slayton M.D. This project would also not have been possible without the support and funding from the U niversity of F lorida Center for Pediatric Psychology and Family Studies. I also thank Drs. Sanje ev Tuli, Carolyn Carter, and Amy Smith for their support of the project and allowing me into their clinics I am particularly grateful to Stephanie Bryan, PA C, for her unwavering support, never ending optimism, and immense help in recruiting participants for the study. To graduate students, Nadia Bhu iy an and Marissa Gowey and the undergraduate research assistants that helped on many aspects of the study, this dissertation would not have been possible without you. I am very grateful for the participation o f mothers an d children and I appreciate their willingness to contribute to the study. Finally, I must thank my family: my fianc, Jason, for always believing in me and helping me reach my goals and my parents and sisters for making me believe that I coul d do anything I put my mind to.
4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURE S ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Family Environment ................................ ................................ ................................ 13 Patient Factors ................................ ................................ ................................ ........ 15 Sibling Factors ................................ ................................ ................................ ........ 19 Emotional Functioning ................................ ................................ ...................... 20 Behavioral Functioning ................................ ................................ ..................... 22 Differences Between Different Cancer Groups ................................ ................. 23 Informant Factors ................................ ................................ ................................ .... 25 Measurement Factors ................................ ................................ ............................. 28 Limitations of the Existing Literature and Goals of the Current Study ..................... 31 Study Objectives and Hypotheses ................................ ................................ .......... 32 Aim 1: Describe Psychological Functioning of Families with a Child with Cancer ................................ ................................ ................................ ........... 32 Aim 2: Describe Differences Between Parent and Child Report ....................... 33 Aim 3: Differences Between Cancer Diagnoses ................................ ............... 35 Aim 4: Factors A ssociated with Psychological Functioning (Exploratory) ......... 36 2 METHODS ................................ ................................ ................................ .............. 38 Participants ................................ ................................ ................................ ............. 38 Procedure ................................ ................................ ................................ ............... 40 General Procedures ................................ ................................ ......................... 40 Measures ................................ ................................ ................................ .......... 40 Demographic and medical information ................................ ....................... 40 Emotional and behavioral functioning of children ................................ ....... 41 Maternal psychological functioning ................................ ............................ 43 Family functioning ................................ ................................ ...................... 44 Statistical Analyses ................................ ................................ .......................... 45 Aim 1 ................................ ................................ ................................ .......... 45 Aim 2 ................................ ................................ ................................ .......... 46 Aim 3 ................................ ................................ ................................ .......... 46 Aim 4 ................................ ................................ ................................ .......... 46
5 3 RESULTS ................................ ................................ ................................ ............... 48 Psychological Functioning of Siblings of Pediatric Cancer Patients ........................ 49 Mother Report ................................ ................................ ................................ .. 49 report ................................ ................................ ................................ ...... 50 Comparison of oncology siblings to normative sample based on ................................ ................................ ........................ 51 Self Report ................................ ................................ ................................ ....... 51 Comparison of siblings in different cancer groups ................................ ..... 52 Comparison of oncology and primary c are siblings based on self report ... 54 Comparison of oncology siblings to normative sample based on self report ................................ ................................ ................................ ...... 55 Comparison Between Mo ther and Sibling Report ................................ ............. 55 Psychological Functioning of Pediatric Oncology Patients ................................ ...... 56 Mother Report ................................ ................................ ................................ .. 56 Comparison of oncology and primary care patients based on mother report ................................ ................................ ................................ ...... 57 Comparison of oncology patients to the normative sample based on mother report ................................ ................................ .......................... 57 Self Report ................................ ................................ ................................ ....... 58 Comparison of oncology and primary care patients based on self report .. 58 Comparison of oncology patients and normative sample based on self report ................................ ................................ ................................ ...... 59 Comparison b etween Mother and Patient Report ................................ ............. 59 P sychological Functioning of the Mothers ................................ ............................... 60 Family Functioning ................................ ................................ ................................ .. 61 Oncology Group ................................ ................................ ............................... 61 Perceived family functioning of oncology mothers ................................ ..... 61 Perceived family functioning of oncology siblings ................................ ...... 62 Perceived family fu nctioning of oncology patients ................................ ...... 62 Comparison between oncology mothers and siblings perception of family functioning ................................ ................................ .................... 62 Comparison betw een oncology mothers and patients perception of family functioning ................................ ................................ .................... 63 Primary Care Group ................................ ................................ ......................... 64 Perceived family functioning of primary ca re mothers ................................ 64 Perceived family functioning of primary care siblings ................................ 64 Perceived family functioning of primary care patients ................................ 64 Comparison of oncology and primary care mothers report of perceived family functioning ................................ ................................ .................... 65 Exploratory Analyses ................................ ................................ .............................. 65 Relationships between Sibling Psychological Functioning and Family Functioning ................................ ................................ ................................ .... 65 Relationships between Medical Variables and Siblings Psychological Functioning ................................ ................................ ................................ .... 66
6 4 DISCUSSION ................................ ................................ ................................ ......... 88 Psychological Functioning of Siblings of Pediatric Cancer Patients ........................ 89 Utility of Evaluating Findings at Group and Individual Levels ........................... 89 Comparison of oncology and primary care siblings ................................ .... 91 Utility of multi informant assessment methods ................................ ........... 93 Comparison of oncology group to published questionnaire normative sample ................................ ................................ ................................ .... 93 Psycholog ical Functioning of Pediatric Oncology Patients and Comparison to Primary Care Group and Normative Controls ................................ ...................... 95 Psychological Functioning of the Mothers ................................ ............................... 96 Family Functioning of Oncology Families ................................ ............................... 98 Limitations and Future Directions ................................ ................................ ........... 99 Strengths and Conclusions ................................ ................................ ................... 103 LIST OF REFERENCES ................................ ................................ ............................. 114 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 119
7 LIST OF TABLES Table page 2 1 Summary of measures completed by study participants ................................ .... 44 3 1 Demographic characteristics of participants in the o ncology and primary care groups ................................ ................................ ................................ ................ 68 3 2 Medical characteristics of oncology group ................................ .......................... 69 3 3 Psychological functioning of oncology patients and their siblings across informant type ................................ ................................ ................................ ..... 70 3 4 Comparison of primary care siblings and a normative sample across informant type ................................ ................................ ................................ ..... 72 3 5 Psychological functioning of primary care children across informant type .......... 74 3 6 Proportion of oncology and primary care siblings with mothers reporting psychological problems ................................ ................................ ...................... 76 3 7 Proportion of oncology and primary care siblings with self reported psychological problems ................................ ................................ ...................... 77 3 8 Proportion of oncology siblings with mothers reporting psychological problems in cancer and tumor groups ................................ ................................ 78 3 9 Proportion of oncology siblings reporting psychological problems in cancer and tumor groups ................................ ................................ ............................... 79 3 10 Proportion of oncology and primary care patients with mothers reporting psychological problems ................................ ................................ ...................... 80 3 11 Comparisons of oncology patients to primary care and normative samples ....... 81 3 12 Proportion of oncology and primary care patients reporting psychological problems ................................ ................................ ................................ ............. 83 3 13 Psychological functioning of mothers ................................ ................................ 84 3 14 Proportion of mothers rep orting poor functioning ................................ ................ 85 3 15 Family functioning reported by oncology siblings and mothers .......................... 85 3 16 Family functioning reported by oncology patients and mothers .......................... 86 3 17 Family functioning reported by primary care group family members .................. 86
8 3 18 Proportion of family members reporting poor family functioning ......................... 87 3 19 Comparison of mothers perceived family functioning between oncology and primary care groups ................................ ................................ ............................ 87 4 1 Proportion of oncology siblings with mothers reporting psychological problems in cancer and tumor groups ................................ .............................. 106 4 2 Proportion of oncology siblings reporting psychological problems in cancer and tumor groups ................................ ................................ ............................. 107
9 LIST OF FIGURES Figure page 4 1 Proportion of siblings with mothers reporting psy chological problems .............. 108 4 2 Proportion of oncology siblings reporting psychological problems .................... 109 4 3 Proportion of oncology patients with mothers re porting psychological problems ................................ ................................ ................................ ........... 110 4 4 Proportion of oncology patients re porting psychological problems ................... 111 4 5 Proportion of oncology mothers r eporting psychological problems ................... 112 4 6 Proportion of discordant pairs in the oncology group on the Behavior Assessment System for Children 2 nd Edition ................................ .................... 112 4 7 Scores for perceived family fu nctioning in the oncology group ......................... 113
10 Abstract of Dissertation Presented to the Graduate School of the Univ ersity of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EMOTIONAL AND BEHAVI ORAL FUNCTIONING OF SIBLINGS OF PEDIATRI C CANCER PATIENTS: DISCORDANCE BETWEEN PARENT AND SELF REPORT AND ASSOCIATIONS WITH FA MILY FUNC TIONING By Ryan E mily Fussell August 2012 Chair: Shelley C. Heaton Cochair: Sheila M. Eyberg Major: Psychology Pediatric cancer is a life altering disease which can profound ly impact the psychological functioning of the children it affects However, pe diatric cancer not only affects the ill child but the entire family system. R esearch examining the psychological functioning and adjustment of parent s of pediatric cancer patients has been widely disseminated; however studies assessing the emotional and be havioral functioning of the non ill siblings have only recently started to appear R esearch that has been conducted in this area suggests that oncology siblings c an exhibit some emotional and behavioral difficulties in resp onse to the ill child s and resultant changes in family dynamics. Additionally, despite preliminary research suggesting that children with different types of cancer (i.e., tumors versus leukemia) have different emotional and behavioral outcomes, not enough is known to guide trea tment guidelines as programs for these children and their families begin to be developed This project use d standardized clinical measures to evaluate the psychological functioning of 35 healthy siblings of oncology patients recruited from a pediatric hosp ital Additionally, we examine d other factors that may affect sibling functioning including psychological
11 functioning of the parent and ill child and perceptions of broader family functioning This study recruited 20 families from an outpatient clinic to act as a primary care comparison group Results suggest that overall, patients, mothers, and siblings are doing well at the individual level. However, siblings are endorsing more difficulties in the family unit and more psychological problems than reporte d by their mothers. The results of this study further highlight the importance of assessing the psychological functioning of the whole family unit when working with a child diagnosed with cancer Finally, the current study design provide s a framework for f uture studies investigating the complex set of factors that may affect functioning of individual family members
12 CHAPTER 1 INTRODUCTION Childhood cancer is a devastating disease that greatly impacts the lives of the patients affected. However, cancer not only affects the patients, but it can significantly affect the quality of life and psychosocial functioning of the surrounding family members as well. Extensive research has been conducted examining the psychological functioning of ill children and their parents (Alderfer, Long, Lown, et al., 2010; Prchal and Landolt, 2009). However, recent attention has also been given to the emotional and behavioral functioning of siblings of children with cancer (i.e., Alderfer Labay, & Kazak, 2003; Labay & Walco, 2004 ; Houtzager, Oort, Hoekstra Weebers, et al., 2004 a ). These lives are often disrupted by the constant changes in routine, reduced parental involvement or emotional availability, reduced social support and access to friends, and increased responsibilities or assumption of new roles within the family (Prchal et al, 2009; Guite, Lobato, Shalon, et al., 2007). Given all these changes, it is not surprising that these chi ldren may experience some emotional or behavioral distress; however, research has only started to unearth which factors contribute to the expression of psychologi cal symptoms in these children. The present study sought to add to the developing research for siblings of pediatric cancer patients. As children are not immune to the influence of the family members that surround them, it is important to have a framework in mind that includes the influence that other family members and environmental factors may ha ve on the emotional and behavioral functioning of the siblings. One component includes individual factors such as: demographic characteristics, health status, and pre illness functioning
13 and coping, which may vary for each family member. Another component may include family characteristics such as communication, role and routine changes, emotional availability and responsiveness, and available resources such as socioeconomic status and social support. While each of these factors is important to acknowledge individually, the interaction of these factors may greatly impact the functioning of the individual sibling as well. This paper did and behavioral functioning, but rather recognize that there are sever al factors external to the child that may influence their emotional and behavioral functioning. Thus, s ignificant emphasis is and the adjustment of the family unit in our discussion of the heal thy siblings. Furth ermore, while this project sought to address significant gaps in the li terature, the ultimate goal lies in providing further evidence for the implementation of psychological services for this vulnerable population. Family Environment Whe n we are examining the psychological adjustment of siblings in response to a cancer diagnosis and treatment, it is important to note that children function within a family context, and thus there may be factors that contribute to or buffer the development of emotional and behavioral symptoms. The idea that family functioning empirically analyze the effect of family functioning on siblings psychological functioning (Alder fer et al., 2010). A review of studies in this area found 28 articles assessing the family unit; however, it should be noted that while there has been an emphasis in the literature on including family factors in research, 19 of these studies provided solel y qualitative information. Thus, an area of needed development is to provide quantitative
14 accounts of the family context in which siblings exist. The overall findings from this meta analysis (Alderfer et al., 2010) suggest that siblings perceive that they are receiving less attention and spending less time with their parents, which may be an accurate assessment, as the majority of parents across studies expressed that they have difficulty attending to the needs of both of their children (Alderfer et al., 20 10). Other important factors that were noted to change after receiving a cancer diagnosis were disruptions of routines, increasing responsibilities for the sibling, and difficulty depending on their family for the support they are accustomed to. The quanti tative studies that do exist provide us with important information and are functioning, results showed that family harmony was viewed as a problem by children dealing w Usually siblings receive most information regarding the health status of their sibling from the family unit; however, the siblings are usually the family members who are often over looked (Lobato & Kao, 2002). Research by a team at Brown University noted that children often benefit from receiving developmentally appropriate information about their siblings illness. More specifically, children who participated in a clinical program de signed to increase sibling knowledge and connectedne ss, displayed significantly fewer behavioral symptoms on the Child Behavior Checklist, per parent repo rt, post treatment (Lobato et al. 2002). These results suggest that the family environment, including family connectedness and communication, can have a significant and positive universal finding. Labay and Walco (2004) found that siblings who endorsed positive
15 relationsh ips with family members had higher levels of maladjustment and decreased social skills. They hypothesized that these children are actually at higher risk for maladjustment because the stress of the medical outcomes and procedures are increased because of t he close relationship. Another study that points to the importance of functioning of the family unit in predicting maladjustment in siblings is a study of 99 siblings of pediatric cancer patients by Sloper and While (1996). They indicated that 24 of the 99 siblings had levels of behavioral problems in the borderline or clinical range 6 months after diagnosis of cancer. The authors suggested that while the majority of the siblings did not have clinically significant emotional or behavioral symptoms, factors that contributed to the siblings perceived changes in relationships and communication with others (Sloper et al., 1996). These findings suggest that while the majori ty of children do not exhibit clinical levels of symptoms, there may be certain familial factors that place them at risk for maladjustment. A limitation of this research is that the majority of researchers do not use control group s; therefore the reader ca nnot ascertain whether findings are specific to families dealing with cancer or to families in general ( Alderfer, 2010 ) Thus this study seeks to quantitatively assess family functioning in the domains of communication, roles, conflict, emotional avai labi lity and responsiveness, in pediatric oncology and a primary care group s Patient Factors When working with siblings of pediatric cancer patients, it is important to note that individual characteristics of the ill child, such as health status and treatment s, can
16 impact the functioning of the healthy sibling. Fortunately, the world of pediatric cancer treatment and research is vastly changing. Within the last couple decades researchers and clinicians have made substantial strides to improve treatments target ing pediatric patients, and to that end, have had amazing success at increasing 5 year survivorship outcomes (Patenaude & Kupst, 2005). Due to these changes in survivorship, research has emerged addressing psychological functioning in addition to t he child outcomes. Pediatric oncology patients experience significant life stressors associated with painful and scary medical procedures, uncertainty about their diagnoses, separation from family and friends, and in some cases, physical changes assoc iated with their condition or medical treatments. One might expect significant difficulty adjusting and the possibility of maladaptive psychological symptoms due to these changes in the ildren do not experience substantial psychological impairment during (Eiser, Hill, & Vance, 2000; Kazak, Rourke, Alderfer, Pai, Reilly, & Meadows, 2007) or after complet ing treatment (Patenaude et al. 2005). In a meta analysis completed by Eiser et al., ( 2000), the majority of research studies reported that pediatric cancer patients did not exhibit significantly more global symptoms than population norms. In fact, one study reported that these patients exhibited significantly less symptoms than population norms (Elkin, Phipps, Mulhern, & Fairclough, 1997 as cited in Eiser et al, 2000). Given the fact that the majority of children do not express clinical levels of psychological distress, it has led many researchers to examine factors that contribute to the d evelopment of psycholog ical symptoms in some children.
17 One of the most influential individual factors affecting functioning of pediatric patients is type of diagnosis. (C OG) there are three main types of pediatri c cancer diagnoses: leukemias (cancers of the blood), lymphomas (cancers of the immune system), and solid tumors, which can be further separated into brain and Central Nervous System ( CNS ) tumors and other solid tumors Ea ch type of cancer has its own prognosis, treatment course, and associated symptomatology Due to these differences, it could be expected that children with different oncological diagnoses would exhibit various physical and psychological outcomes. In a stud y utilizing data from the national research project, the Childhood Cancer Survivor Study, data showed that sur vivors who were diagnosed with leukemia, CNS tumors, and n euroblastomas were more likely to exhibit problems with depression/anxiety and antisocia l behaviors than children with other types of cancer (Schultz, Ness, Whitton, et al., 2007). Furthermore, children with CNS tumors were more likely to exhibit difficulties with social competence. Given the research above, it may be particular ly important f or health care workers treating these children t o closely monitor any changes in or deterioration of mental health status. Considering that each pediatric cancer diagnosis comes with its own unique treatment regimen, another important individual factor to consider is type of oncological treatment. Most researchers in the field agree that children with cancers with central nervous system (CNS) involvement have an increased risk for short (Moore Challinor, Pasvogel, et al., 2003) and long term psychological sequelae (Hill, Kornblith, Jones, et al., 2003). In a study of patients who were currently undergoing or had re cently completed treatment for acute lymphoblastic l eukemia with CNS involvement, results
18 showed elevated scores in the domains of somatization, depression, anxiety, and withdrawal (Moore et al., 2003). In addition to these internalizing symptoms, these authors also reported high er levels of attention and learning problems. The authors also postulate d that children who experience academic or cognit ive difficulties in response to their treatments might also be at higher risk for behavioral symptoms. These results suggest that children diagnosed with ALL with CNS involvement, who have subsequently received intrathecal chemotherapy and/or intracranial radiation, may have additional emotional and cognitive symptoms than chi ldren without CNS involvement. Finally treatment status may act as an individual factor that may play a role in the patient psychological functioning. Pediatric cancer treatment, unlik e many other chronic disorders, goes through several stages with various side effects and stressors on the individual patient and the family. Some research suggests that children who have been recently diagnosed with cancer exhibit more psychological sympt oms than a comparison group ( Sawyer, Antoniou, Toogood, et al. 2000 ). However, in that same study, parents rated their children comparably to the community sample one year into treatment and then at subsequent yearly follow ups. In a di fferent study of ch ildren with acute lymphoblastic l eukemia, researchers found differences in physical and psychological functioning 6 weeks and 1 year after diagnosis (Moore et al., 2003). The authors found that individual treatment factors, such as treatment intensity and medical complications, were predictive of decreased motor and emotional functioning. Results from this study show that treatment intensity is not significantly correlated with health related quality of life one year post diagnosis compared to 6 weeks; howe ver, complications associated with oncological treatment continue to be correlated with
19 emotional functioning. Their results a lso suggest that children with acute lymphoblastic l eukemia may have more physical and emotional complaints due to the intense tre atments they received during their first phase of chemotherapy (Moore et al., 2003). Putting these results together, it suggest s that health care workers not only need to acknowledge the type of cancer, but also the treatment phase in working with these fa milies. Sibling Factors Individual patient factors as mentioned above can in influence not only the psychological fu nctioning of the patient, but the surrounding family members as well. If we continue to emphasize the involvement of various factors contri buting to the emotional and behavioral functioning of siblings, it is also important to acknowledge individual sibling factors that may contribute to their emotional and behavioral functioning such as: demographics, health status treatments for their sibl ing, and pre and post diagnosis coping. Earlier research in the area of pediatric psycho oncology suggest ed that healthy siblings of pediatric cancer patients are at higher risk for psychological maladjustment ( Cohen, Friedrich, Jaworski, Copeland, & P endergrass, 1994 provide references ) However, recent studies suggest that while siblings may exhibit some emotional and behavioral changes in response to the ill child and resultant treatment, their symptoms are usually not clinically significant at the group level (Alderfer et al., 2010). However, some researchers report that the percentage of siblings whose scores fall within the sub clinical or clinical ranges is higher than normative groups (Houtzager, Grootenhuis, Caron, et al., 2004 b ). A simi lar question to the one proposed in the oncology group arises: if the majority of siblings do not
20 experience serious psychological maladjustment, what factors are implicated for the minority of siblings that do experience signific ant psychological difficul ties? Emotional Functioning Often when a child is diagnosed with cancer, they are thrust into a world of changes: frequent medical appointments and painful procedures, significant time away from home and school, and large disruptions to their daily routine However, what is often overlooked is that siblings have significant changes to their lives as well (Alderfer et al., 2010). Not only do these children have to worry about the uncertainty of their en have limited contact with one or both of their parents and the daily structure that they are accustomed to (family factors). Possibly due to these changes, siblings often have significant difficulty adjusting and these difficulties may present as emotio nal symptoms. Common emotional symptoms found within this population include depression, anxiety, and post traumatic stress symptoms (Alderfer et al., 2010). Recent research in the area of pediatric psychology suggests that anxiety is a common symptom afte Grootenhuis & Last, 2001). In fact, a clinical psychology group in the Netherlands received frequent consults about siblings of pediatric cancer patients to the point where they created a support g roup to enhance sibling knowledge about the disease and to foster peer support in order to decrease anxiety (Houtzager et al., 2001). A study examining the effectiveness of their program found that 18 out of 24 siblings initially had anxiety scores above t he 80 th percentile on the State Trait Anxiety Inventory before age based norms. The exception to this finding was in the sub group of adolescent
21 females where anxi ety was comparable to the normative sample before and after treatment. At the end of the treatment, the young males and the adolescent boys exhibited normative levels of symptoms; however, the younger females continued to endorse higher levels of anxiety t han age based norms. The same research team reported that anxiety appeared to be associated with various indiv idu al factors such as age, number of days in the hospital, types of control strategies used, and other family factors (Houtzager et al., 2004a ) O verall, anxiety decreased as days in the hospital and the chil F amilies who rated themselves as more cohesive reported higher levels of anxiety, which supports the idea that individual and family factors may interact. Findin gs by Houtzager et al. (2004 a ) may suggest that being closer to family could actually serve as a risk factor of anxiety possibly due to more time spent with the ill child or more exposure to ill Going beyond generalized anxiety, one of t he most recent areas of research for siblings of pediatric cancer patients is post traumatic stress. Alderfer, Labay, and Kazak (2003) reported that approximately 49% of siblings in their sample experienced mild post traumatic stress reactions years after diagnosis. More concerning results showed that 32% of siblings experienced moderate to severe levels of post traumatic stress reactions. Results also showed that girls and siblings older than six were more likely to endorse post traumatic stress symptoms. Alderfer and colleagues (2003) postulated that these children are at particular risk because they may take on caregiving roles and are exposed to more of the medical procedures and familial distress. This also provides an example of how an individual facto r (age of sibling) can interact with family factors
22 (caregiving changes), highlighting that different age groups may experience different psycholo gical symptoms possibly in part to their roles played within the family unit. Given the stressful nature of h aving a sibling with cancer, it would not be surprising to see significant depressive symptoms, in addition to anxiety, in these children. However, in a recent meta analysis of psychological functioning in siblings of children with cancer, the majority of studies showed that siblings show comparable or lower levels of depressive symptoms than community controls (Alderfer et al., 2010). Interestingly, compared to the findings for anxiety symptoms, there was not a trend for a higher percentage of children fal ling within the sub clinical or clinical ranges, at the group level, than the normative controls. In line with the question asked above, what factors then are contributing to the expression of anxiety in these children, but not depression? This study sough t to add to the research examining which individual and family factors are associated with the presence of psychologi cal symptoms in these children. Behavioral Functioning While emotional symptoms are frequently discussed within the sibling literature, the re is less research examining behavioral symptoms. It is not unfounded that these children would exhibit acting out behaviors given the dramatic changes in their lives and the significant decrease in daily routines, structure, and at times, lack of parenta l involve ment or supervision. In a large scale study of siblings of cancer patients, researchers showed that the sibling group had significantly higher scores on the Externalizing scale, including aggression and conduct problems, of the Child Behavior Chec klist comp ared to age based norms (Cohen et al., 1994). While this was an important finding, what was more concerning was that a significantly larger proportion of
23 siblings had scores one or two standard deviations above the mean compared to the normative group. Results indicated that the extent of externalizing behaviors was predicted by the individual factors of sibling age at time of diagnosi s and age at the initiation of the study, as parents were more likely to endorse externalizing symptoms in their y ounger children than older children. However, these results are not universal, as other studies have shown decreased rates of aggression, substance use, and anti social behavior in pediatric cancer survivors in comparison to their peers (Verrill, Schafer, Vannatta, and Noll, 2000). The strength of the Cohen et al. study lies in their examination of factors that would make children more prone to developing externalizing symptoms, but there are likely othe r factors, which this study aimed to examine, that pl ay a role in the development of increased emo tional and behavioral problems. Differences B etween Different Cancer Groups While it is important to examine the emotional and behavioral symptoms in siblings in general, it is also important to acknowledge that psychological functioning may also be associated with the type of cancer the ill child has. There are many types of pediatric cancer diagnoses that are associated with a vast array of prognoses, treatments, and physical outcomes that may subsequently affe ct families in different ways (Sidhu, Passmore, & Baker, 2006). In a large study examining outcomes for siblings of pediatric cancer patients, siblings were found to have significantly higher levels of psychological distress than siblings of non ill child r en (Cadman, Boyle, & Offord, 1988). Siblings of cancer patients may have particular difficulties with psychological adjustment given the physical (Sidhu et al., 2006) and cognitive impacts of the disease (Freeman et al., 2003), intensive treatments, and va riable prognoses (Sidhu et al., 2006). The next section will discuss the differences that have been found between cancer groups.
24 While there has been an increase in studies examining the emotional and behavioral functioning of pediatric cancer patients and their siblings, little research has examined possible differences across different oncological diagnoses. Research that has been done in this area suggests that children diagnosed with leukemia and lymphomas exhibit different psychological symptoms than c hildren with solid or brain tumors, and respond differently to psychological treatment (Houtzager et al., 2001). As described in a previous section, researchers examined the effect of a group therapy treatment on levels of anxiety in siblings of pediatric cancer patients. Results showed that siblings showed greater levels of anxiety before starting the treatment; however, their ill and treatment, the sibli decreased. An interesting finding from this research suggests that siblings of children with brain tumors did not benefit from the treatment at the same levels as siblings of leukemia and lymphoma patients. The Surveilla nce Epidemiology and End Results suggest that the 5 year survival rates (ages 0 19) for brain and other central nervous system cancers is 75.2% and 68.8% for neuroblastomas, compared to 84.0% for children with acute lymphoblastic l eukemia (Ries, Melbert, K rapcho, et al., 2007). Given that the survival rates for brain and solid tumors are lower than for leukemia, siblings of treatment and subsequent anxiety related to their pr ognosis (Houtzager et al., 2001). Future research should aim to examine differences between cancer groups in order to ascertain whether some groups are at particularly high risk for maladjustment, thus providing evidence for appropriate assessment.
25 Informa nt Factors Most studies examin ing the psychological functioning of children with cancer and their siblings have heavily relied upon parent behavior re port (Guite, Lobato, Kao, et al., 2004) In some instances parent report of emotional and behavioral symptoms may be necessary, such as when the child is too young or unable, due to medical factors, to complete their own ratings (Eiser and Morse, 2001). However, if the child is old enough to pr ovide their own account of their functioning, they may provide clinically useful information that the parent may not have provided. Furthermore, it is conceivable that may differ depending on who the informant was as well as the nature of psychological functioning of the informants themselves thus producing a R eview s of parent child discrepancy literature indicate that parents of ill children usually endors e more negative outcomes than the patients rate themselves (Eiser et al., 2001). However, in a study of siblings of children with cancer, siblings endorsed more internalizing, externalizing, and overall problems than their parents on the Child Behavi or Che cklist, post diagnosis (Houtzager et al., 2005). In the same study, siblings also reported lower quality of life than their parents. The authors hypothesized that the parents may have b een less likely to see emotional problems in the siblings because they may be spending significant t ime attending to the ill child Additionally, they noted that parents of children with cancer are frequently out of the home attending to the ill child and thus may not be spending significant amounts of time with the siblings, which could reduce their ability to see emotional disturbances should they exist.
26 The psychological state of the parent is also an important individual factor to consider when utilizin g parental proxy report of a child functioning. Research suggests that parents who endorse problems with their own psychological functioning are more likely to endorse psychological symptoms in their child ( Cohen et al., 1994 ). The nature of this relation ship is unclear, but could be refl ective of a reporting bias (e.g. depressed e ffect where children of depressed parents may actually have more mood or behavior problems becaus An example of this is illustrated in a study examining predictors of sibling maladjustment (Cohen et al., 1994). This study examined the various factors that predict psychological symptoms in siblings of children with cancer at various points in their treatment protocols. Results suggest that mothers who endorsed significant depressive symptoms were more likely to endorse higher scores on the Child Behavior Checklist and lower scores on a measure of social competence. This has large implications when working with these families clinically as children may be less likely to engage in treatment if they do not feel as distressed as their parents think they are (Davidson, 2005). This also suggests that individua l factors of different family members may interact to affect the emotional and behavioral functioning of the sibling. is also an important individual factor when comparing parent and child similarities between parent and needs and emotional state (Eiser et al., 2001). As a discussed previously in this review, Houtzager et al. (2004 b ) found that sibling s endorse significantly more emotional
27 symptoms two years post diagnosis than their parents. In the discussion of these findings, the authors suggest that the younger children may exhibit maladjustment in terms of their social relationships instead of in e motional symptoms because they are more dependent on their family members for coordinating social visits than teenagers. expression of stress, anxiety, or a method of gainin g their parents attention since they do not yet have the verbal skills to communicate their feelings. This idea would also support the finding that parents and children agree more on behavioral symptoms, those that they can externally ob serve, than emotion al symptoms. P arent child relationship s can be greatly affected by a cancer diagnosis ( Zeltzer, Dolgin, Sahler, et. al., 1996). Parents can become less communicative and can spend significantly less time with the non ill child Conversely, family factors s uch as parent child communication may improve between the ill child and the parent given their increased time with one another and the need for clear communication in order to properly treat the child (Eiser et al, 2001). An area of needed research is to d etermine whether there are differences between parent child report with the ill child and parent, sibling and parent, and controls in which there is no illness within the family unit. In summary, the research discussed above shows that when working with fa milies living with cancer, it is important to assess not only the patient, but also the surrounding family unit. More importantly, while most studies use parental proxies to assess patient and sibling functioning, it seems imperative that mental health pro fessionals receive information from the patient and the parent However with the addition of multiple informants, you also have the increase d time of scoring and interpretation. This
28 provides justification for the use of broadband screening measures in ord er to screen for more serious psycho patho logy than giving each family member multiple measures to assess different domains of functioning. This last factor will be discussed in the section below. Measurement Factors P ediatric patients and their families h ave often received services through referrals from their medical teams through a traditional consultation approach. However, several problems arise with this model of psychological intervention; 1) many families are not referred until they have significant problems and 2) many families who are not experiencing significant psychopathology or significant distress, but have sub clinical symptoms or are having difficulty adjusting, may be missed. The present trend in psycho oncology research is veering away fro m focusing on major psychopathology to a more preventative and universal model of psychological care (Kazak et al., 2007). As Kazak et al. (2007) note d the majority of patients and families reasonably have some minor adjustment difficulties, but overall m ost families do not exhibit major psychopathology. Using their knowledge from 20 years of research in pediatric psycho oncology, they have implemented a model of care that emphasizes screening for all, but intervention for only those who need it, recognizi ng that that majorit y of families do not require in depth clinical services ( Pediatric Psychosocial Preventative Health Model [PPPHM] for more information, Kazak et al., 2007). A strength of their program lies in their ability to identify familie s that need the most aid and provide needed services. However, as the reader can gather from the research summarized above, the majority of psychological research in this field has focused on specific areas of functioning, such as anxiety, depression, and post traumatic stress disorder or disease specific
29 measures, as opposed to more broadband measures of functioning. While individually each measure may not take much time to complete, if these families are asked to complete several measures in order to tap into multiple areas of emotional and behavioral functioning, they could be spending hours completing these measures. Since most families will not experience clinically significant symptoms, and it is time intensive to complete several measures, it seems a ppropriate to have families complete broadband questionnaires that evaluate various domains of psychological functioning at one time (Wolfe Christensen, Mullins, Stinnet t, Carpentier, & Fedele, 2009). The Child Behavior Checklist ( CBCL; Achenbach, 1991a, b) is often used in clinical settings as a broadband measure of emotional and behavioral functioni ng in children with cancer ( Wolfe Christensen et al., 2009 provides listing of studies that have utilized CBCL). The CBCL is a useful tool in examining several domains of psychological functioning; however, Perrin, Stein, & Drotar (1991) note that there are some problems with using this measure within an oncology population (Perrin, Stein, & Drotar (1991) as cited in Wolfe Christensen et al., 2009). First, there are concerns that pediatric oncology patients may have higher scores on the internalizing composite scale due to endorsement of somatic items associated with their illness and treatment. Second, the CBCL is less sensitive to sub clinical levels of symptom s. Third, the CBCL may not provide accurate social competence scores given the scales that make up the composite. In response to these concerns, Wolfe Christensen et al. (2009) proposed the use of the newer broadband measure, the Behavior Assessment System for Children, 2 nd Edition ( BASC 2; Reynolds & Kamphaus)
30 The BASC 2 is a widely used measure of emotional, behavioral, and adaptive functioning (Reynolds et al., 2004 ), which includes both parent and self report forms. In contrast to the popular CBCL, Wo lfe Christensen et al. (2009) suggested that the BASC 2 has several advantages over the CBCL including individual subscales for emotio nal symptoms, a wide range of T scores that allows participants to score below 50, scales to assess validity of participan t responses, and normative scores for various diagnostic groups (Wolfe Christensen et al., 2009). While the BASC 2 has been extensively used in children with clinical disorders such as A ttention D eficit/ H yperactivity D isorder (ADHD) and learning disorders, there is limited research examining the use of the BASC 2 in children with medical diagnoses. In order to address this gap in the literature, Wolfe Christensen and colleagues (2009) initiated a study to compare scores of pediatric oncology patients to a h ealthy control sample. Using a discriminate function analysis, the authors found that the BASC 2 was able to distinguish between the two groups in terms of emotional and cognitive complaints. While their results provided evidence for the use of the BASC 2 as a good screening tool in the pediatric cancer population, they had 2 limitations that should be addressed: (1) their sample failed to find differences between differ ent subtypes of cancer (i.e., leukemia/l ymphoma, solid tumors, and brain tumors) and (2) they only included parent report versions of the questionnaire. Future research should seek to assess whether there are differences between oncological groups (leukemia vs. brain tumor) in a different sample of partic ipants and should include child self r eport versions of the forms.
31 Limitations of the Existing Literature and Goals of the Current Study Despite the vast research in the area of psychosocial functioning of families with a child with pediatric cancer, there are several limitations that should b e addressed. One of the consistent limitations throughout the literature is the lack of appropriate control group ncer diagnosis. This study sought to expand on the literature in this area an d provide a primary care group of children with non ill siblings in order to assess whether many of the relationships cited in the literature (i.e., discordance between parent and child report, increased maladjustment with non cohesive family units) are co nsistent for families in general or whether they are specific to the cancer population. Additionally, the present study adds to the current literature base by providing quantitative reports of family functioning as rated by both the parent and the children While there is still some disagreement about whether siblings exhibit negative psychological outcomes in response to living with a brother or sister diagnosed with cancer, there appears to be consistent evidence that parents and children endorse differe nt psychological symptoms, and that in order to provide the most comprehensive siblings themselves in addition to parent report. This study sought to advance the litera broadband emotional and behavioral functioning and the functioning of the family based studi es have examined the psychological functioning of siblings, this study went one step further by assess ing what aspects of family functioning affect the psychologi cal adjustment of the siblings.
32 Study Objectives and Hypotheses This study add s to the literat ure by examining the emotional and behavioral functioning of pediatric patients, thei r siblings, and their parents. Additionally, this study attempts to add to the literature in the field differentiating the psychological symptoms that present between diff erent family members. A recent addition to the literature, and one in which t his study aims to incorporate, is the inclusion of individual ( patient sibling, functioning. A fi nal contribution of the present study collects similar information from a primary care group of siblings to assess whether the proposed findings are specific to siblings in the oncology population or whether siblings from healthy populations exhibit simila r patterns. Study aims and hypotheses are as follows: Aim 1 : Describe Psychological Functioning of Families with a Child with Cancer The preliminary goal of this project was to describe an often overlooked aspect of the psychological functioning of familie s living with a child with cancer: the psychological functioning of siblings. As noted in the literature, the patient is not the only family member at risk for experiencing emoti onal and behavioral disturbance siblings and parents are also at risk for psy chological maladjustment. In order to study sibling and parent wellbeing in a pediatric cancer population, research should include measures to mothers However, as noted in the review of the literature, parents and their children may have differing opinions about their functioning. Thus, the current study incorporates both the report of their psychological functioning. Sub Aim 1 Describe psych ological functioning of children in families living with a child with cancer, including the child with cancer and siblings.
33 Hypothesis 1.1.1 : It is not anticipated that s iblings of oncology patients (from here will ex hibit clinically significant symptoms at the group level, but a larger percentage of them will fall in the at risk or clinical ranges compared to a primary care comparison group primary care It is not expected o nco logy s iblings will endorse significantly higher scores than the normative sample. This was tested for parent report and self report versions of the measure. Hypothesis 1.1.2 : We do not expect p ediatric oncology patients (from here on will as a group to have clinically elevated scores on a broadband psychological measure. It is hypothesized however, that the percentage of children falling into the at risk or clinical range will be significant higher than a primary care comparis on group primary care It is anticipated that o ncology patients will not endorse significantly higher scores than the normative sample. This hypothesis was tested for parent report and self report versions of the me asure. Sub Aim 2 Explore psychological functioning of mother s of children with cancer. Hypothesis 1.2.1: Given the small expected sample sizes, exploratory analyses will be conducted to compare groups. Mothers ncology mother as a group will have significantly higher levels of self reported psychological symptoms than mothers of children without cancer primary care Aim 2 : Describe Differences Between Parent and Child Report The seco nd goal of this project was to assess how families living with a cancer
34 behavioral function ing and family adjustment differently and one goal of this project was to provide further support to the literature citing these differences. Thus, the current study incorporates both the report of their psycholo gical functioning. For example, clinicians could incorporate questions about the researchers may choose to avoid using parent proxy methods alone if it is determined t hat this lone perspective does not provide the most accurate assessment of the chi psychological functioning. Sub Aim 1 Describe differences in ratings of the oncology functioning based on parent and sibling self report. Given the small expected sample sizes, traditional analyses comparing group means may not find statistical differences that may exist. Thus, other methods of compar isons will also be employed ( methods section). Hypothesis 2.1.1: Oncology m others and siblings will not have a high level of agreement in their endorsement of externalizing symptoms. Mothers will endorse significantly more externalizing symptoms (i.e., hyperactive aggressive behavior) than their non ill child (sibling). Hypothesis 2.1.2: Oncology m other s and siblings will not have a high level of agreement in their endorsement of emotional symptoms. Mothers will endorse significantly more emotional sympto ms (i.e., depressive and anxious symptoms) than their non ill child (sibling).
35 Hypothesis 2.1.3 : Oncology m others and siblings will not have a high level of agreement in their endorsement of adaptive symptoms. Mothers will report that the siblings ha ve significantly lower adaptive skills (i.e., activities of daily living, social skills, and function al communication) than the siblings will report themselves. Sub Aim 2 Describe differences in ratings of family functioning based on mother and sibling self report in the oncology group Similar to psychological functioning, we will employ various methods of comparison as traditional analyses comparing group means may not find statistical differences. Hypothesis 2.2.1 : Oncology m others and siblings will not have a high level of agreement in their endorsement of family functioning. Mothers will describe si gnificantly better family functioning than will the sibling in the family. Hypothesis 2.2.2 : Oncology m others and patients will not have a high level of agreement in their endorsement of family functioning Oncology m others will describe significantly bett er family functioning than will patient s in the family. Hypothesis 2.2.3 : Oncology m others will report worse family functioning than mothers in the primary care group. Aim 3 : Differences Between Cancer Diagnoses The third goal of this project address ed an other under researched area within the pediatric cancer literature: the differences in psychological functioning between different types of cancer groups. Research cited earlier in the review has shown that siblings of cancer patients often exhibit differe nt emotional and behavioral symptoms based on the diagnosis of their sibling. For example, some research has shown that siblings of brain and solid tumor patients exhibit significantly more anxiety than siblings of lymphoma and leukemia patients (Houtzager et al., 2001). Over the past decade, significant
36 improvements have been made in treatments for pediatric cancer patients, subsequently increasing 5 year survivorship rates. However, there are still significant differences in prognoses and survivorship rat es between different cancer groups. Despite substantial improvements in medical outcomes, little research has been conducted examining how siblings function in light of various diagnoses. Thus the third aim of this study addressed that deficit. For the pur poses of this study, pediatric oncology patients will be divided into four subgroups: leukemias, lymphomas, sarcomas/solid tumors, and brain/central nervous system tumors (CNS) (based on standar ds utilized in the literature). Hypothesis 3.1: Siblings of sa rcomas/solid tumors and brain/CNS tumor patients will have significantly higher levels of emotional and externalizing behavior symptoms and worse adaptive skills than siblings of leukemia and lymphoma patients. These differences will be described qualitati vely as we do not expect a large enough sample to conduct statistical comparisons. Aim 4 : Factors Associated with Psychological Functioning (Exploratory) The final aim of this project was to examine various factors that are associated with psychological s ymptoms in siblings of children with cancer. While information regarding the psychological functioning of siblings is interesting and informative, research shows that the majority of siblings, as a group, do not exhibit clinical levels of symptoms (Alderfe r et al., 2010). However, at the individual level, a larger percentage of siblings fall in the sub clinical or clinical ranges on behavioral measures. Therefore, future research in this area will specifically need to account for factors that are associated with the development of these symptoms in siblings with clinically significant levels.
37 Hypothesis 4.1: I ndividual and family factors such as demographic characteristics, health status, and family functioning will be correlated with emotional and behavior functioning in oncology siblings
38 CHAPTER 2 METHODS Participants This study was reviewed and approved by the University of Florida Institutional Review Board. A total of 55 families provided consent and completed study measures. Participants were recruit ed from two academic medical center clinics in rural Florida : a pediatric o ncology c linic ( n = 35 ) and a primary care clinic ( n = 20 ). Families were seen at the clinics for either a routine medical appointment or an inpatient hospital stay. All participant s were identified by the respective medical teams and subsequently approached by a member of the research team to review and obtain informed consent from the mother. Assent was obtained from children who were 8 or older (the minimum age required to complet e the measures). All families were advised that refusal to participate in the study would not impact their medical care nor would their medical teams see their individual responses Pediatric participants included 2 groups: a pediatric oncology group and a primary care group. Children in both groups met the following inclusion criteria: (1) between the ages of 2 and 17 and (2) able to independently complete study measures if old enough to complete study measures (over the age of 8 for the B ehavior A ssessme nt S ystem for C hildren 2 nd Edition and 12 for the F amily Assessment Device ) Only mothers were asked to participate in the study given their previous precedence in the literature and greater likelihood of being at the routine appointments. Mothers were req uired to: (1) be over the age of 18, (2) live in the home with the oncology patient and at least one sibling, and (3) able to independently complete study measures. Families were required to have at least one other sibling living in the home who was: (1) b etween the ages of 2
39 and 17, (2) able independently complete study measures, and (3) living in the home with the patient and the mother at the time of the assessment. If there was more than one sibling living in the home, the sibling closest in age to the patient was included in the study. Additionally, oncology patients met these additional criteria: (1) diagnosed with of any form of cancer (e.g., leukemia, brain tumor) (2) completed at least 8 weeks of active treatment or (3) if in remission, were no mo re than one year beyond completion of active treatment. Children and their mothers were excluded if the mother reported any of the following: (1) being legally blind in both eyes, (2) being legally deaf in both ears or (3) diagnosed with mental retardatio n. Additionally, the study excluded patients who: (1) were diagnosed with another severe medical illness (as determined by the medical staff) and/or (2) had a history of bone marrow transplantation. In order to reduce the risk of confounding factors, sibli ngs were excluded from the study if mothers reported that they were diagnosed with a serious chronic medical or psychiatric illness (e.g., sickle cell anemia, HIV, etc.). Children in both groups were not excluded for mild medical diagnoses (e.g., allergies ) unless they had severe/serious complications (e.g., frequent hospitalizations, emergency room visits). Emotional and behavioral symptoms can occur in both healthy children (C enters for D isease C ontrol (CDC) Simpson, Bloom, Cohen, Blumberg, 2005) and tho se diagnosed with cancer (Moore et al., 2003) thus in order to maintain generalizability to both groups, children with mild to moderate psychiatric disorders (e.g., depression, ADHD; as determined by the medical and research team) were not excluded in eith er the oncology or primary care groups. Potential limitations of including these participants
40 are discussed in the discussion. Patients and siblings with severe psychiatric disorders (e.g., schizo phrenia, autism) were excluded. Procedure General Procedures Parents and children completed appropriate measures during their routine appointment or inpatient stay Special emphasis was placed on the families not discussing the measures with one another. Mothers completed the same measure for multiple children, thu s they were encouraged to think clearly about which child they their ratings. Siblings who were not present during the appointment or inpatient stay completed appropr iate measures over the phone after orally assenting to study procedures. Participants were given $5.00 compensation for their participation in the study. Measures A review of each measure, the utility of the measure for this particular line of research, an d a comment on the psychometric strength of the measure is presented below. A summary of measures completed by each part icipant is presented in Table 2 1. Demographic and medical i nformation Basic medical and demographic information was collected from bot h the oncology and primary care group. Medical history. Medical information was collected from a review of the patient medical records. The chart review extracted information regarding the patient
41 diagnosis, types and lengths of treatments received, di sease progression, history of hospitalizations, and dates and types of applicable surgeries or procedures. Demographic q uestionnaire Mothers completed a demographic questionnaire l history, behavioral concerns. Emotional and behavioral functioning of c hildren The Behavior Assessment System for Children, Second Edition (BASC 2: Reynolds & Kamphaus, 2004) was used as the primary measure of emotional and behavioral functioning of the oncology and primary care children. The BASC 2 is a multi dimensional psychological questionnaire that evaluates various aspects of and behavioral functioning. There are distinct versions of the BASC 2 that allow for parent report, teacher report, or child self report ratings of functioning in the identified child For the current study, the parent report and child self report versions of the BASC 2 were utilized. Ratings on t he parent report versions of the BASC 2 produce composite scores for measured Externalizing Problems (e.g., aggression, hyperactivity and conduct problems), Internalizing Problems (e.g., anxiety, depression, and somatization), Behavioral Symptoms (e.g., atypicality, withdrawal), and Adaptive Skills (e.g., adaptability, leadership, activities of daily living, functional communication, and social skills ). Standardized scores are also produced for the i ndividual scales within each composite index, as indicated in parentheses in the preceding sentence. There are different parent report versions of the BASC 2 for 3 distinct age groups : 2 5 years old, 6 11 years old, and 12 21 years old. Mothers completed the appropriate
42 question naires based on the age of their children, one for the patient and another one for the sibling. Age norm referenced standardized scores for each of the primary scales and their composite score domains were generated using norms from the published software for each group (i.e., parent report, patient report, sibling report) (Reynolds & Kamphaus, 2004) BASC 2 scales have a mean of 50 and a standard deviation of 10. T scores are considered at risk between 60 69 (30 39 for the Adaptive Skills) while scores abo ve 70 are considered clinically significant Within a normative sample, the Parent Rating Scales (BASC 2: PRS) have test retest reliability scores that range from 78 to .92 across age groups for the composite scales. Internal con sistency scores ranged from .85 to .95. Children (within the oncology and primary care group s ) of appropriate age completed the self report version of the BASC 2. Youth Self report Profiles (BASC 2: SR P ) are available for ages 8 11, 12 21 and 18 25 ( the 1 8 2 5 form was not utilized for the current study ) with composite scales measuring School Problems, Internalizing Problems, Inattention/Hyperactivity, Emotional Symptoms and Personal Adjustment. Test retest reliability scores for the self report profiles composite scores range from .74 to .89 in a normative sample. Internal consistency scores for the composite scales range from .74 to 84 I n our sample of oncology mothers, patients and .882 for the siblings on the PRS P; .883 for patients and .873 for siblings on the PRS C, and .512 for the patients and .791 on the PRS A. In our sample of primary care mothers the PRS P; .812 for patients and .946 for siblings on the PRS C, and .871 for the patients and .934 on the PRS
43 for the self report forms was .459 for patients and .529 for siblings on the SRP C, and .672 for the patients and .824 for the siblings on the SRP A. In our sample of primary care C, and .908 for patients and .651 for siblings on the SRP A. respo nses on items assessing suicidal and/or homicidal thoughts or actions to ensure that appropriate measures were taken to ensure safety and provide any necessary treatment referral s Maternal p sychological f unctioning The Brief Symptom Inventory (BSI: Dero gatis, 1993) was used as a brief assessment of psychological symptoms in oncology and primary care mothers. The BSI is a 53 item scale designed to quickly assess symptoms of psychological distress. The items comprise 9 symptom subscales of Somatization, Ob sessive Compulsive behavior, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism. There are also three global scales: Global Severity Index (GSI), Positive Symptom Distress Index (PSDI), and Positi ve Symptom Total (PST). Gender norm referenced standardized scores for each of the scales were generated using norms from the published manual for each group (i.e., oncology and primary care group s) Internal consistency ratings have been reported for the various scales from .71 to .85 (Derogatis & Melisaratos, 1983). Test retest reliability scores have been shown for the global indices to range from .80 .90. was .960 and .957 for the primary care group A safety ass urance plan was put in place to ensure that appropriate measures were taken to ensure safety.
44 Family f unctioning Mothers and children (12 years and older) completed th e McMaster s Family Assessment Device (FAD: Epstein, Baldwi n, & Bishop, 1983) to assess perceived current functioning of the family unit. The FAD is a 60 item measure assess ing the global functioning of family units. Furthermore, the FAD assesses specific components of family functioning such as Problem Solving, Communication, Roles, Affective Responsiveness, Affective Involvement, and Behavioral Control. The FAD has been found to have internal reliability scores ranging from .57 to .86 and all scales had i nternal reliabilities within the adequate range, with the subscale of Roles only producing marginal reliability ( Miller, Epstein, Bishop, et al., 1985 ) oncology group were as follows : .928 for mothers, .865 for siblings, and .863 for the patients for the primary care group were as follows: 894 for the mothers and .810 for the patients. There were only 4 primary care siblings, making the group too small for assessment of measurement reliability. Table 2 1 Summa ry of measures completed by study participants Oncology Group Primary C are Group Oncology Patient Oncology Sibling Oncology Mother Primary Care Patient Primary Care Sibling Primary Care Mother Self Report BASC 2 Self Report BASC 2 Parent Report BASC 2 ( for patient & sibling ) Self Report BASC 2 Self Report BASC 2 Parent Report BASC 2 ( for patient & sibling ) FAD FAD FAD FAD FAD FAD BSI BSI Demographic Questionnaire Demographic Questionnaire Note: BASC 2 = Behavior Assessment System for Child ren, FAD = Family Assessment Device, BSI = Brief Symptom Inventory
45 Statistical Analyses The Statistical Pa ckage for the Social Sciences 20 .0 (SPSS) was used to conduct statistical analyses. Descriptive statistics, including means and standard deviations were calculated for each group. Statistics are provided for the entire sample and by group. Separate scores are provided for the parent, sibling, and patient report on measures completed by both groups. Total and sub scale scores are provided for the BSI. For the BASC 2, the mean scores are provided for each of the subscales Percentages of individuals above the a t r isk (T scores of 60 69 ) and clinically s ignificant (T cutoff score s were calculated for each group. For the FAD, the General Func tioning score as well as the scores for the 6 behavioral subscales (Problem Solving, Communication, Roles, Affective Responsiveness, Affective Involvement, and Behavior Control) are provided. indicative of poorer functioning. Clinical cutoff scor es are provided in the Tables 3 15 through 3 17. Aim 1 To examine sub aim 1, independent samples t tests were used to compar e the oncology patients and siblings to primary care children in behavioral symptomatology a s rated on the BASC 2. Separate t tests were conducted for the self report and parent report meas ures. To examine sub aim 2, independent samples t tests were used to compare means of the oncology mothers and primary care mother s, assuming equal variances. Welsh t tests were also condu cted to compare the oncology sample to the normative population.
46 Aim 2 To examine hypotheses 2.1.1 through 2.1.3 and 2.2.1 and 2.2.2, various methods of comparison were conducted to compare mother and child scores. First, inde pendent samples t tests were used to compare the oncology mothers and children across scales on the BASC 2. Second, discrepancy scores were calculated to measure the level of agreement between mothers and their children using a model proposed by Garber et al. (1998) and utilized by Guite et al. (2004). Using this model s scores of perceived psychological and perceived family functioning were subtracted from the mothers with higher scores indicating more d iscrepancy Standard deviations of the group as a whole were then calculated for each scale of the BASC 2 and FAD. T he dyad discrepancy scores were then compared to the standard de viation for each s cale According to model used by Guite et al. (2004) d yad s were considered discordant if the calculated discrepancy scores were more than half of the standard deviation above zero Third, Pearson product moment correlations were calculated to measure the relationship between mother and child report. Aim 3 As we did no t hav e a large enough sample to examine differences between diagnostic groups (i.e., leukemia versus brain tumor), differences between groups were described qualitatively. Percentages of children falling in the at risk or clinically significant rang es were calculated by groups for both mother and sibling report. Aim 4 Our sample did not allow for testing of a model that predicts psychological functioning. However, correlations between oncology report BASC 2 and the FAD scores were con ducted to in order to examine factors that may impact the
47 as a whole. Pearson product moment correlations were run for normal variables while Spearman correlation s were run non normal variables. For the purposes of the exploratory analyses, large correlations were considered to be scores greater or equal to .5. Significant correlations are described below. Given that the focus of this project is on individual and f amily functioning, analyses examining relationships with teachers or school functioning were not conducted.
48 CHAPTER 3 RESULTS A total of 55 families ( oncology group n = 35; primary care group n = 20 ) were included in the present study. Demographic chara cteristics of study partic ipants are presented in Table 3 1. In the oncology group, 48 eligible families were approached and informed about the study. Thirty eight provided informed consent ( 92.1 %) and subsequently completed study measures. Of the 35 parti cipating families in the oncology group 22 oncology patients and 22 siblings completed at least one of the self report measures. Fewer oncology patients and siblings completed measures than mothers as children had to be at least 8 years old to complete at least one of the study measures. The oncology patients ranged in age from 2 to 17 years (M ean age = 8.9) and 57.1% were male. The diagnoses of the oncology patients included leukemias (62.8%), lymphomas (8.6%), brain/CNS tumors (11.4), and solid tumors/sar comas (17.1%). The mean length of time since 1 st diagnosis was 12.38 months ( SD = 12.10) and the majority of oncology patients were receiving treatment at the time of this study (85.7%). Medical characteristics of the oncology group are presented in Table 3 2. In the primary care group 30 eligible families were approached and informed about the study. Twenty three provided informed consent and 20 ( 87 %) completed study measures. P atients in the primary care group ranged in age from 2 to 17 with a mean age of 8.9 and were 45 % male. Of the 20 families in the primary care group, 12 patients and 6 siblings completed at least one of the self report measures. Socio economic status (SES) was calculated for each group using the Hollingshead Four Factor Index of Socia l Status (Hollingshead, 1975), which generates
49 a score using information regarding parental educational and occupational status. Approximately 38% of our oncology group and 35% of the primary care group reported household incomes of over $50,000. The Holli ngshead score for the primary care group was higher ( M = 41.75, SD = 17.95) than the oncology group ( M = 36.03, SD = 12.58), but this difference was not statistically different, t (53) = 1.39, p = .172, suggesting that the two groups are comparable in terms of socioeconomic status. Chi square tests and i ndependent s ample t tests were used to compare the oncology and primary care groups across the other demographic variables. The two groups were similar on the variables of sibling gender, 2 (1) = 2.31, p = .128, patient gender, 2 (1) = .753, p = .386, patient age, t (53) = .072, p = .942, and t (53) = 1.01, p = .318. Groups were statistically different in terms of sibling age, t (53) = 2.95, p = .005, with the mean of the oncolo gy group approximately three and a half years older than the primary care group. Psychological Functioning of Sibling s of Pediatric Cancer Patients Aim 1 sought to examine the psychological functioning of families with a child with cancer (i.e., siblings, patients, and mothers). The following section will describe the psychological functioning of the siblings of pediatric cancer patients in comparison to the primary care group and a normative sample In order to address Aim 2, differences in perception betw een mothers and children, findings are discussed in terms of mother and sibling self report. Mother Report At the group level, mothers did not report at risk or clinically significant emotional or behavioral problems for the siblings in the oncology or pri mary care group s ( Table s 3 3 and 3 4 ). Examination of ratings at the individual level revealed that a subset of
50 mothers endorse d at risk or clinically significant levels of emotional or behavioral problems across the two sibling groups ( Table 3 5 ). The oncology mothers endorsed at risk or clinically significant concerns in Externalizing Problems ( 20.0%), Hyperactivity (20%), Aggression (22.9 %), Conduct Pro blems (8.6%), Internalizing Problems (17.1%), Anxiety (8.6%), Depression (22.9%), Somatization ( 11.4 %), Behavioral Symptoms (17.1%), Atypicality (5.7%), Withdrawal (11.4%), Attention Problems (25.7 %), Adaptive Problems (11.4% ), Adaptability (20 %), Social Skills (28.6 %), Leadership (14.8 %), Activities of Daily Living (14.3 %), and Functional Communicat ion ( 17.1 %). The primary care mothers endorsed at risk or clinically significant concerns in Externalizing Problems (10 %), Hyperactivity (15 %), Conduct Problems (14.3%), Internalizing Problems (10%), Anxiety (20%), Somatization (15%), Atypicality (15%), At tention Problems (15%), Adaptability (15 %), Social Skills (10%), Leadership (14.3%), Activities of Daily Living (25%), and Functional Communication (20%). Comparison of o ncology and p rimary care siblings based on m r eport Independent samples t tes ts and Mann Whitney t ests were used to compare the oncology and primary care groups across the BASC 2 scales ( Table 3 5) Overall, ratings made by mothers in the oncology group produced slightly higher scores (indicative of worse psychological function ing albeit within normal limits in terms of group means) than mothers of primary care siblings across most scales of the BASC 2 with the exception of Anxiety, Atypicality, Adaptive Problems, Adapta bility, and Social Skills. However, the only statistically significant difference between the two groups was found on the Depression scale where oncology mothers reported more depressive symptoms in oncology siblings than primary care mothers ( t (52) = 2.18, p = .034 )
51 Comparison of oncology s iblings to normative s ample b ased on m r eport Welsh t tests were also conducted to compare the oncology to the B ehavior A ssessment S ystem for Children 2 nd Edition (BASC 2) normative sample ( Table 3 5) Overall, mothers are endorsing slightly lower scores across most scales on the BASC 2 (indicative of better psychological functioning) for oncology siblings. Nevertheless, there were no significan t differences on any scale of the BASC 2 between the study sample and the BASC 2 normative samples Self R eport oncology siblings themselves did not produce clinically elevated scores on the BASC 2 at the group level ( Table 3 3) However, with in the group, a subset of o ncology s iblings did endorse at risk or clin ically significant emotio nal or behavioral con cerns ( Table 3 6 ) on the BASC 2 in the following domains: School Problems (5.3%), Attitude to School (21.1%), Attitude to Teachers (15.8%), Sensation Seeking (10.5%), Internalizing Problems (5.3%), Atypical ity (5.3%), Locus of Control (10.5%), Anxiety (10.5%), Depression (5.3%), Somatization (10.5%), Inattention (15.8%), Attention Problems (21.1%), Hyperactivity (21.1%), Personal Adjustment (5.3%), Interpersonal Relations (5.3%), and Self Reliance (10.5%). P rimary care siblings endorsed at risk concerns at the group level in Attention Problems ( M = 60.0, SD = 9.92) a nd Inattention/Hyperactivity ( M = 60.33, SD = 11.98). At the individual level, a subset of primary care siblings endorsed at risk or clinically s ignificant concern s in School Problems (33.3%), Attitude to Teachers (16.7%), Sensation Seeking (25%), Internalizing Problems (16.7%), Social Stress (16.7%), Anxiety (16.7%), Depression (16.7%), Somatization (25%), Inattention (21.1%), Attention Problems ( 50%), Hyperactivity (33.3%), Self Esteem (16.7%), and Emotional
52 Symptoms (16.7%) ( Table 3 7) These proportions should be interpreted with caution in light of the very small sample size ( N = 6) which may not be an accurate representation of the broader population between the oncology and primary care group s; however, there were no statistically significant differences between groups. Comparison of s iblings in d ifferent c ancer g roups Aim 3 sought to examine differences between different cancer groups (i.e., leukemia versus brain tumor). The following section discussed differences in sibling se lf report between cancer groups Leukemias Table s 3 8 and 3 9 show the proportion of c hildren in the oncology group falling into the at risk and clinically significant ranges (as rated by mothers and siblings), broken down by diagnosis. In the Leukemia group, a subset of mothers endorsed at risk or clinically significant concerns in Hyperac tivity (9.1%), Internalizing Problems (9.1%), Depression (13.6%), Somatization (13.6%), Behavioral Symptoms (9.1%), Atypicality (22.7%), Withdrawal (18.2%), At tention Problems (22.7%), Adapta bility (9.1%), Social Skills (13.6%), and Functional Communicatio n (18.2%). Similar to their mothers, a subset of siblings in this group endorse d at risk or clinically significant levels of concerns across scales of the BASC 2. Elevations were found in the following scales: Attitude to School (30%), Attitude to Teachers (30%), Internalizing Problems (10%), Locus of Control (20%), Anxiety (10%), Depression (10%), Inattention/Hyperactivity (30%), Attention Problems (40%), Hyperactivity (40%), Personal Adjustment (30%), and Self Reliance (10%). Overall, these results sugges t that mothers and siblings of l eukemia patients perceive that the siblings are doing
53 relatively well in terms of emotional, behavioral, and adaptive functioning but there are a subset of participants perceiving s ome psychological difficulties. Lymphomas More mothers in the l ymphoma group endorsed at risk and clinically significant levels of problems. A t risk and clinically significant concerns were found for Externalizing Problems (66.7%), Hyperactivity (33.3%), Aggression (100 %), Internalizing Problems (66.7%), Anxiety (66.7%), Depression (100%), Somatization (33.3%), Behavioral Symptoms (66.7%), Attention Problems (33.3%), Adaptive Problems (33.3%), Adaptability (66.7%), Social Skills (66.7%), Leadership (33.3%), Activities of Daily Living (33.3%) and F unctional Communication (33.3%). There were only 2 siblings that were old enough to complete the BASC 2 measure, so the following results should be interpreted with caution. Half of the siblings in this group endorsed at risk concerns in Atypicality, Soma tization, Personal Adjustment, and Interpersonal Relations. Interesting, siblings in this group did not endorse any clinically significant problems on any scale of the BASC 2. Solid t umors Similar to the Lymphoma group, mothers in the solid tumor group ar e endorsing higher percentages of at risk and clinically significant symptoms compared to the leukemia group. A subset of mother endorsed at risk or clinically significant concerns in the following BASC 2 domains: Externalizing Problems (50%), Hyperactivit y (50%), Aggression (50%), Conduct Problems (33.3%), Internalizing Problems (16.7%), Depression (16.7%), Behavioral Symptoms (33.3%), Attention Problems (16.7%), Adaptive Problems (16.7%), Adaptability (33.3%), Social Skills (50%), Leadership.(16.7%), Acti vities of Daily Living (16.7%) and Functional Communication (16.7%).
54 Overall, siblings of solid tumor patients endorsed good functioning across BASC 2 scales. Twenty percent of siblings endorsed at risk or clinically significant levels in School Problems, Attitude to School, Attitude to Teachers, Sensation Seeking and Self Reliance Brain t umors M others in the brain tumor group also endorsed higher percentages of at risk and clinically significant concerns compared to the leukemia group; however, there w ere only 4 children in this group, so these results should be interpreted with caution. A subset of mothers endorsed at risk or clinically significant concerns in the following BASC 2 domains: Externalizing Problems (25%), Hyperactivity (25%), Aggression ( 25%), Internalizing Problems (25%), Depression (25%), Behavioral Symptoms (25%), Attention Problems (50%), Adaptive Problems (25%), Adaptability (50%), Social Skills (50%), Leadership.(66.7%), and Activities of Daily Living (50%). These results suggest tha t a small portion of mothers in our sample perceive d significant difficulties in emotional, behavioral, and adaptive functioning in their non ill child ren. Finally, there were only two siblings old enough to complete the BASC 2 measure; however, these chi ldren endorsed good functioning across areas of the BASC 2. One child endorsed clinically significant concerns in Attitude to Teachers and Anxiety. There were only two children in this group; however, these children suggest that children in our sample are endorsing good emotional, behavioral, and adaptive functioning. Comparison of o ncology and p rimary care siblings based on self r eport Independent samples t tests and Mann Whitney tests were used to compare oncology and primary care siblings. Overall, prima ry care siblings endorsed lower
55 scores (indicative of better functioning, albeit within normal limits) on most scales of the BASC 2 except for Self Esteem Significant differences were found in School Problems, t (32) = 2.56, p = .025, Attitude to School, U = 74.0, p = .036, Attitude to Teachers, t (32) = 2.68, p = .020, and Somatization, t (32) = 3.65, p = .002. Comparison of oncology s iblings to normative s ample based on self r eport Oncology s BASC 2 normative sample and they endorsed better emotional, behavioral, and adaptive functioning across all scales o f the BASC 2 except Depression ( Table 3 4) Significant differences were found for the following scales: Internalizing Problems, t (19) = 3.05, p = 0.01, Emotional Symptoms, t (19) = 3.42, p = 0.00, Personal Adjustment, t (19) = 3.45, p = 0.00, Atypicality, t (19) = 3.16, p = 0.01, Social Stress t (19) = 4.88, p = 0.00, Sense of Inadequacy, t (18) = 2.33, p = 0.03, Interpersonal Relations, t (19) = 3.75, p = 0.00, and Sel f Esteem, t (19) = 8.26, p = 0.00. Comparison Between Mother and Sibling Report Analyses were conducted to compare the perceptions of oncology mothers and the siblin gs using independent samples t tests The following scales are represented across parent r eport and sibling report versions of the BASC 2 f orms and thus could be compared in this study : Internalizing Problems, Anxiety, Depression, Somatization, Hyperactivity, Atypicality, and Attention Problems. Oncology mothers rated the oncology siblings as ratings indicated, with the exception of A nxiety but none of the comparisons were statistically significant. However, 50% or more of the mother sibling dyads were classified as discordant across a ll areas of the BASC 2 ( Table 3 3). The majority of Pearson p roduct m oment c orrelations between mother and sibling report for the BASC 2 scales fell within
56 the small range with the exception of Anxiety ( r = .328, moderate) and Attention Problems ( r = .406, moderate). None of the Pearson correlations were statistical ly significant. The i ntraclass c orrelation results followed a similar pattern with all correlations classified as small with the exception of Anxiety ( r = .323, moderate) and Attention Probl ems ( r = .405, moderate). None of these correlations were statistically significant. P sychological Functioning of Pediatric Oncology Patient s The following section continue s to address Aim 1 by describing the psychological functioning of pediatric cancer p atients from the perspective of the mothers and the patients themselves. The current study sample is also compared to the primary care group and the normative sample. Additionally, this following section also addresses Aim 2 as it describes differences bet ween mother and patient report. Mother Report Overall, oncology mothers did not endorse any at risk or clinically significant behaviors at the group level for the oncology patients ( Table 3 3 ) as rated by the BASC 2. In terms of the percentage of mothe rs endorsing at risk or clinically significant levels of symptoms ( Table 3 10) t he following scales had a small pro portion of mothers endorsing at risk or clinically significant concerns: Externalizing Problems (5.7%), Hyperactivity (8.6%), Aggression (5.7%), Internalizing Problems (17.1%), Anxiety (14.3%), Depression (20%), Somatization (40%), Behavioral Symptoms (5.7%), Atypicality (5.7%), Withdrawal (8.6%), Attention Problems (14.3%), Adaptive Problems (11.4%), Adaptability (8.6%), Activities of Dai ly Living (14.3%), and Functional Communication (14.3%).
57 Similarly, primary care mothers did not endorse any clinically significant emotional, behavioral, or a daptive difficulties for the primary care patients at the group level ( Table 3 5) Comparable to above, a subset of mothers endorsed at risk or clinically significant concerns in the following BASC 2 domains: Externalizing Problems (15%), Hyperactivity (15%), Aggression (25%), Conduct Problems (23.1%), Anxiety (10%), Depression (10%), Somatization (20%), Behavioral Symptoms (10%), Atypicality (10%), Withdrawal (20%), Attention Problems (25%), Adaptability (15%), Social Skills (15%), Leadership.(15.4%), Activities of Daily Living (20%) and Functional Communication (10%) ( Table 3 10) Comparison of o ncology and p rimary care patients based on mother r eport Independent s ample t tests and Mann Whitney t ests were used to compare emotional, behavioral, adaptive functioning in the two groups. The only significant difference between groups was in Somatiz ation, t (53) = 2.44, p = .018, where oncology mothers endorsed significant more symptoms (e.g., ) than primary care mothers. Comparison of oncology p atients to the normative s ample based on m other r eport Welsh t tests we re also conducted to compare scores produced by oncology ratings to the published BASC 2 normative sample ( Table 3 11) Overall, oncology m others endorsed better emotional, behavioral, and adaptive functioning in the oncology patients o n every scale on the BASC 2 except for Internalizing problems, Adaptive Skills, Somatization, and Activities of Daily Living. However, significant differences were only found in Externalizing Problems, t (34) = 3.03, p = 0.005 Behavioral Symptoms, t (34) = 2.77, p = 0.0 09 Aggression, t (34) = 2.94, p = 0.006
58 Conduct Problems, t (28) = 2.78, p = 0.01, Somatization, t (34) = 3.16, p = 0.00, Atypicality, t (34) = 2.33, p = 0.03, and Social Skills, t (34) = 3.16, p = 0.00. Self Report In general, neither o ncology patients nor primary care patients endorse d any emotion al or behavioral concerns at the group level ( T able s 3 3 and 3 5 ). Similar to their mothers, a small proportion of patients endorsed at risk or clinically significant concerns in the following BASC 2 domain s ( Table 3 12) : Sensation Seeking (13.6%), Anxiety (13.6%), Somatization (36.4%), Sense of Inadequacy (9.1%), Inattention/Hyperactivity (9.1%), Attention Problems (9.1%), Hyperactivity (27.2%), Relations with Parents (9.1%), Interpersonal Relations (9. 1%), and Self Relian ce (9.1%). Primary care patients had larger proportions of participants endorsing at risk or clinically significant concerns on the BASC 2 ( Table 3 12) such as School Problems (33.3%), Attitude to Teachers (33.3%), Attitude to Tea chers (25%), Sensation Seeking (28.6%), Internalizing Problems (16.7%), Locus of Control (16.7%), Social Stress (25%), Anxiety (16.7%), Depression (25%), Sense of Inadequacy (16.7%), Inattention/Hyperactivity (16.7%), Attention Problems (50%), Hyperactivit y (8.3%), Personal Adjustment (16.7%), Self Esteem (8. 3%), and Self Reliance (25%). Comparison of o ncology and p rimary care patients based on s elf r eport Independent samples t tests and Mann Whitney t ests were used to compare the groups across the BASC 2 s cales ( Table 3 11) Overall, there were no significant differences between the oncology and primary care patients with the exce ption of the school scales and S omatization Oncology patients endorsed significantly more symptoms of S omatization than prim ary care patients t (1 7 ) = 2.85 p = 011 Interesting, primary care patients endorsed significa ntly more difficulty with the school
59 scales: School Problems, t (32) = 2.32 p = .02 6 ; Attitude to School, t (32) = 2.42 p = .021 ; and Attitude to Teachers, t (32) = 3.46 p = 001 Comparison of oncology p atients a nd normative s ample based on s elf r eport scores were compared to the BASC 2 normative sample ( Table 3 11) and results showed that they endorsed better emotional, behavior, and a daptive functioning with the exception of Inattention/Hyperactivity, Sensation Seeking, Anxiety, Somatization, and Hyperactivity. Significant differences were found for the following scales of the BASC 2: School Problems, t (21) = 4.06, p = 0.00, Personal A djustment, t (21) = 2.85, p = 0.01, Attitude to School, t (21) = 2.66, p = 0.02, Attitude to Teachers, t (21) = 6.05, p = 0.00, Atypicality, t (21) = 2.28, p = 0.03, Depression, t (21) = 2.91, p = 0.01, Somatization, t (12) = 2.66, p = 0.02, Relations with Paren ts, t (21) = 2.13, p = 0.05, Interpersonal Relations, t (21) = 2.31, p = 0.03, and Self Reliance, t (21) = 2.24, p = 0.04. Com parison b etween Mother and Patient Report In order to test for differences in perception between oncology mothers and patients Inde pendent samples t tests were used to compare group means on the following scales: Internalizing Problems, Anxiety, Depression, Somatization, Hyperactivity, Atypicality, and Attention Problems. Mothers endorsed worse functioning in Internalizing Problems, D epression, Atypicality, and Attention Problems; yet there were no significant differences between the groups. Even though there were no significant differences between groups, 50% or more of all the mother patient dyads were qualified as discordant across all scales of the BASC 2 ( Table 3 3) Pearson p roduct m oment c orrelations between mother and patient report for the BASC 2 scales ranged from small (.185, Hyperactivity) to large (.707, Somatization). Three correlations
60 were statistically significant: Internalizing Problems, r = .484, p = .011, Depression, r = .602, p = .002, and Somatization, r = .707, p = .003. S imilar results were found with intraclass correlation c oefficients with the correlations ranging from small (.185, Hyperactivity) to large ( .690, Somatization). Three of these correlations were also statistically significant: Internalizing Problems, r = .406 p = 027 Depression, r = 530 p = .00 5 and Somatization, r = 690 p = 003 Psychological Functioning of the Mothers Aim 1 also had the goal of describing the psychological functioning of mothers. Additionally, we proposed that the oncology mothers in our sample would have worse psychological functioning than the primary care group and a normative sample. The following section address e s these aims. On the whole, oncology mothers did not endorse significant psychological distress at the group level as m easured by the B rief S ymptom I nventory (BSI) ( Table 3 13 ) In order to have a significant BSI score the participant either had to hav e an elevated Global Severity score or two elevated subscale scores. Only 8 (22.9%) oncology mothers had elevated Global Severity scores and 13 (37.1%) had two or more elevated scales ( Table 3 14) S imilar to the oncology group primary care mothers di d not endors e significant psychological distress at the group level Only 1 (5%) mother had an elevated Global Severity score and 2 (10%) h ad two or more elevated scales. See Table 3 14 for the proportions of mothers reporting poor functioning across other scales of the BSI. Overall, oncology mothers endorsed more psychological distress than primary care mothers across all subscales; however, none of the scale s were elevated at the group level. Mann Whitney tests were used to compare symptom subscales betwe en
61 groups ( Table 3 13) There was a significant difference in Global Severity between the patient ( M = 53.26, SD = 11.34) and primary care ( M = 45.30, SD = 9.93) groups, ( U = 209.5 p = .04 2 ) Oncology m others endorsed significantly more distress than primary care mothers in the following subscales: Obsessive Compulsive ( U = 206.5, p = .011), Interpersonal Sensitivity ( U = 235.5, p = .037), Anxiety ( U = 219.0, p = .020), Hostility ( U = 193.5, p = .005), and Paranoid Ideation ( U = 238.0, p = .041), and P ositive Symptom Total ( U = 216.5, p = .019). No significant differences were found for Somatization, Depression, Phobic Anxiety, Psychoticism, or Positive Symptom Distress. tests were also computed to compare the proportion of mother s endors ing clinical symptoms in the oncology and primary care group s. None of these comparisons were statistically signif icant. Family Functioning was to describe the functionin g of the family unit as a whole. The following section will describe perceived family functioning from the sibling, patient, and mother perspective. Additionally, this section will compare this group to the primary care group. Oncology Group Perceived f ami ly f unctioning of o ncology m others Both oncology and primary care mothers endors e d good family functioning and none of the F amily A ssessment D evice (FAD) scales were elevated at the group level ( Tables 3 15 and 3 16) A little less than a quarter of oncolo gy mothers endorsed difficulties in General Functioning (22.9%). Mothers in this group endorsed the most difficulties in the area of Roles (40%). Only 5% of mothers in the primary care group endorsed clinical impairments in General Functioning ( T able 3 1 8).
62 Perceived family functioning of oncology s iblings O ncology and primary care siblings endorsed clinical impairments in fa mily functioning across domains. Oncology siblings as a group reported impairment in Communication, Affective Responsiveness, and Affective Involvement. Table 3 18 provides percentages of participants who endorsed at risk or clinically significant concerns across all scales Fifty percent of the oncology siblings endorsed clinically significant levels in General Functioning ( n = 6 ). Additionally, at least a quarter or more of the siblings endorsed clinically significant levels of symptoms across all other domains: Problem Solving (33.3%), Communication (58.3%), Roles (41.7%), Affective Responsiveness (41.7%), Affective Involvement (50 %), and Behavior Control (25%). Perceived family f unctioning o f oncology p atients When examining the report of family functioning, they reported clinically signi ficant concerns in Affective Involvement at the group level ( Tab le 3 16) L arge percentages of participants endorsed clinically significant levels across domains. Fifty percent of the oncology patients endorsed cl inically significant levels in G eneral F unctioning (50%). Additionally, there were several other subscales in which over a quarter of the participants endorsed clinical levels: Problem Solving (25%), Communication (25%), Roles (41.7%), Affective Responsiveness (41.7%), and Behavior Control (25%). Over half of the oncology patients endorsed clinically significant levels in Affective Involvement (58.3%). Comparison b etween o ncology m others and s iblings p erception of f amily f unctioning Table s 3 1 5 an d 3 1 6 present the means, standard deviations, and percentages of discordant pairs between oncology mothers and their ch ildren across scales on the
63 FAD. Independent samples t tests were conducted to compare differences in their perceptions of family functioning between the oncology mothers and their children In general, oncology siblings perceived their family as having mo re impairment than their mothers perceived ( Table 3 15) There were significant differences in Communication, t (45) = 2.47, p = .017, Affective Responsiveness, t (45) = 3.08, p = .004, Affective Involvement, t (45) = 2.56, p = .014, and General Fun ctioning, t (45) = 2.1, p = .041. Pearson product moment correlations range d between small for Behavior Control ( r = .167) and moderate for Affective Responsiveness ( r = .438). None of the Pearson correlations were statistically significant. The intraclas s c orrelation results showed the same pattern with correlation s ranging from small for Behavior Control ( r = .167) and moderate for Affective Responsiveness ( r = .402). None of these correlations were statistically significant. Using the discrepancy model results show that 50% or more of all mother patient dyads were classified as discordant on all scales of the FAD. Interestingly, 100% of mother sibling dyads were discordant on the Behavior Control scale. Comparison between oncology m others and p atients p erception of f amily f unctioning T here were no significant differences between onc ology mothers and patients The majority of Pearson p roduct m oment c orrelations between mother and patient report for the FAD scales fell within the small range with the exce ption of Problem Solving ( r = .491, moderate) and Roles ( r = .380, moderate). None of the Pearson correlations were statistical ly significant. The intraclass c orrelation results showed the same pattern with all correlations classified as small with the exc eption of Problem Solving ( r = .408, moderate) and Roles ( r = .353, moderate). None of these correlations
64 were statistically significant. Using the discrepancy model, results show that 50% or more of all mother patient dyads were classified as disco rdant o n all scales of the FAD. Primary C are G roup Perceived family f unctioning of primary c are m others Overall, primary care mothers endorsed positive family functioning. At the group level, there were no clinic ally elevated scores ( Table 3 1 7 ). Percentages of mothers who endorsed at risk or clinically significant concerns were calculated and t here wa s only one mother in this group who endorsed clinically significant levels in the General Functioning domain (5%). Across all other subscales of the FAD, few mo thers endorsed cl inical levels of symptoms ( Table 3 18) P erceived family f unctioning of p rimary c are s iblings Primary care siblings als o endorsed clinical impairments in several areas of family functioning ( T able 3 1 8 ). The following scales were e levated at the group level for the primary care siblings : Problem Solving, Communication, Roles, Affective Involvement, Behavior Control, and General Functioning. The majority of children in this group endorsed clinically significant levels in General Func tioning (75%). Primary care siblings also endorsed significant impairments in Problem Solving (25%), Communication (50%), Roles (50%), Affective Responsiveness (50%), and Affective Involvement (75%). All of the children in this group endorsed clinically si gnificant symptoms in the domain of Behavior Control (100%). P erceived family f unctioning of primary c are p a tients Primary care patients endorsed clinically elevated scores at the group level in Affective Responsiveness and Affective Involvement Over fift y percent of the children endorsed clinically significant levels in General Functioning (57.1%) ( Table 3 18)
65 Children in this group endorsed significant impairments in Communication (71.4%), Roles (42.9%), Affective Responsiveness (57.1%), and Affecti ve Involvement (71.4%). Comparison of o ncology and p ri mary c ar e m others report of perceived family f unctioning Independent samples t tests w ere used to compare mother reports of f amily functioning in the oncology and primary care group s ( Table 3 19) There was a significant difference in General Functioning for the oncology ( M = 1.62, SD = 0.45) and primary care ( M = 1.34, SD = 0.32) groups; t(53) = 2.1, p = .041. Oncology mothers endorsed worse family functioning across all scales; however, significan t differences were only found in the scales of Problem Solving t (53) =2.10, p = .041, Communication t (53) = 2.34, p = .023 and Roles t (53) = 2.11, p = .040 These results suggest tha t oncology mothers are seeing more deficits in family functioning than primary care mothers Exploratory Analyses The final aim of the current study (Aim 4) was to examine various factors that are associated with psychological symptoms in siblings of children with cancer. The following section will examine the relationships between individual and family functioning in addition to relationships between medical var iables and sibling functioning. Relationships b etween Sibling Psychological Functioning and Family Functioning In order to examine factors that may impact the oncolog y their individual functioning and the functioning of their family unit as a whole, correlations between their self report BAS C 2 and the FAD scores were conducted. Pearson product moment correlations were run for normal variables w hile Spearman
66 correlations were run non normal variables. For the purposes of the exploratory analyses, large correlations were considered to be scores greater or equal to 5. Significant correlations are described below. Given that the focus of this proje ct is on individual and family functioning, analyses examining relationships with teachers or school functioning were not conducted. Analyses revealed a few significant correlations between the oncology self report of psychologic al functioning an d their perception of family functioning. Oncology s iblings reported a significant positive relationship between Personal Adjustment and Communication, r = .613, p = .023. I mprovement s in family Behavior Control were associated with better Personal Adjustm ent scores, r = .524, p = .049. Siblings who endorsed higher scores on Affective Involvement also endorsed more symptoms of Hyperactivity, r = .568, p = .034. Additionally, siblings who endorsed more Self Reliance also tended to endorsed poor family commu nication, r = .767, p = .003. Worse Relations with Parents scores were associated with worse family Roles, r = .532, p = .046. Siblings who endorsed worse Interpersonal Relations also endorsed worse family Behavior Control, r = .627, p = .019. Finally, t here was a significant positive relationship between the sibling variable of Locus of Control and Affective Responsiveness, r = .639, p = .01; Affective Involvement, r = .755, p = .004, and General Functioning, r = .596, p = .026. Relationships b etween Med ical Variables and Siblings Psychological Functioning In order to examine the relationship between key medical variables and the oncology siblings psychological functioning, Spearman correlations were conducted using medical variables and both the BASC 2 composite scores. Both the sibling self report s and the mother reports were used. The medical variables used were: type of
67 diagnosis (e.g leukemia or brain tumor), high or low risk diagnosis (as determined by the medical team), months from diagnosis to s tudy enrollment, total days admitted to the inpatient unit (over whole course of treatment), and finally miles to hospital from No significant correlations were found between the mothers BASC 2 scores and type of diagnosis, high or low ris k diagnosis number of inpatient days, or miles from the months since diagnosis and Behavior Symptoms, r = .382, p = .024 where fewer months from diagnosis was associated wi th higher BASC 2 scores. Similarly, there were no significant correlations between type of diagnosis, high or low risk diagnosis, 2 scores. Large positive correlations were found betwee r = .639, p = report of School Problems, r = .583, p = .009.
68 Table 3 1 Demogra phic characteristics of p articipants in the o ncology and pr imary c are g roups Characteristic Oncology Families ( N = 35) Primary Care Families ( N = 20) Comparison Statistic p value Sibling Sibling Age 1 9.46 (4.17) 6.00 (4.23) 2.95 ( t test ) .005 Sibling Sex (male) 45.7 (16) 25 (5) 2.31 (chi square) .128 S ibling Race (Caucasian) 54.3 (19) 50.0 (10) 5.59 (chi square) .348 Hispanic 5.7 (2) 0.0 (0) African American 31.4 (11) 30.0 (6) Asian/PI 0.0 (0) 10.0 (2) Biracial 5.7 (2) 10.0 (2) Other 2.9 (1) 0.0 (0) Patient Patient Age 1 8.9 (4.54) 8.9 (4.70) 0.072 ( t test ) .943 Patient Sex 57.1 (20) 45.0 (9) 7.53 (chi square) .386 Patient Race (Caucasian) 60.0 (21) 50.0 (10) 5.09 (chi square) .405 Hispanic 2.9 (1) 0.0 (0) African American 28.5 (10) 30.0 (6) Asian/PI 0.0 (0) 10.0 (2) Biracial 5.7 (2) 10.0 (2) Other 2.9 (1) 0.0 (0) Mother Age 36.0 (7.4) 34.0 (6.7) 1.01 ( t test ) .318 Note: Data in this table are presented as % ( n ) unless otherwise noted 1 Data pres e nted as group mean (standard deviation) Stati stically significant at p < .0 1
69 Table 3 2 Medical c haracteristics of o ncology g roup Characteristic n % Diagnosis Leukemia 22 62.9 Lymphoma 3 8.6 Solid Tumor 6 17.1 Brain Tumor 4 11.4 Treatment Risk High 20 57.1 Low 15 42.9 Treatment Status In treatment 30 85.7 Relapse 2 5.7 Recently off treatment 3 8.6 Intervention Status Chemotherapy 35 100 Radiation 8 22.9 Surgery 1 35 100 1 Note: All patients had surgery to place either a port and/or central line
70 Table 3 3. Psyc hological functioning of o ncology p atients and their siblings across informant t yp e Patients Siblings BASC 2 Scales Mother Report ( n =35) Self Report ( n = 22) % of Discordant Mother Patient Pairs 1 Mother Report ( n = 35) Self Report ( n = 20) % of Disc ordant Mother Sibling Pairs 1 Behavioral Symptoms 46.97 (6.41) 49.14 (9.6) Withdrawal 47.31 (9.00) 48.17 (8.82) Atypicality 47.46 (6.38) 47.09 (5.94) 90% 46.06 (7.26) 45.05 (6.93) 50% Attention Problems 48.69 (8.39) 48.14 (7.67) 77.3% 52 .26 (10.80) 48.85 (11.56) 80% Adaptive Problems 52.8 (8.61) 50.97 (11.38) Adaptability 52.91 (9.13) 49.91 (11.44) Social Skills 54.49 (8.36) 49.46 (12.0) Leadership 2 51.31 (16.0) 53.46 (11.55) Activities of Daily Living 49.88 (9.95) 50.77 (11.55) Functional Communication 51.49 (9.06) 51.86 (11.12) Externalizing Problems 46.71 (6.37) 50.34 (11.84) Aggression 46.57 (6.84) 49.43 (11.97) Conduct Problems 2 43.1 (13.33) 51.11 (9.5) Hyperactivity 48.26 ( 8.23) 55.5 (8.89) 63.6% 50.80 (11.28) 49.70 (11.29) 80% Internalizing Problems 51.83 (10.52) 48.59 (6.64) 59.1% 47.60 (10.53) 45.4 (6.7) 85% Anxiety 48.31 (11.08) 50.45 (8.60) 77.3% 46.77 (10.37) 47.3 (9.19) 70% Depression 49.77 (9 .51) 46.05 (6.32) 50% 49.83 (10.40) 47.2 (6.17) 75% Somatization 55.69 (10.63) 58.62 (11.67) 76.9% 47.54 (9.57) 47.5 (8.42) 75% Locus of Control 47.86 (8.10) 47.55 (7.81) Social Stress 48.5 (6.10) 44.0 (5.44) Sense of Inadequacy 47.71 (7.92) 45.84 (7.76) School Problems 44.73 (6.04) 44.80 (11.30) Attitude to School 45.36 (8.14) 47.35 (11.74) Attitude to Teachers 43.32 (5.12) 45.45 (12.12) Sensation Seeking 51.0 (11.22) 47.0 (11.23) Inattention /Hyperactivity 51.14 (8.16) 49.35 (11.29) Emotional Symptoms 47.27 (6.91) 45.0 (6.5) (continued)
71 Table 3 3 (c ont.). Psychological f unctioning of o ncology p atients and their siblings across informant t ype Patients Siblings BASC 2 Scales Mother Report ( n =35) Self Report ( N =22) % of Discordant Mother Patient Pairs 1 Mother Report ( n =35) Self Report ( n = 20) % of Discordant Mother Sibling Pairs 1 Personal Adjustment 54.18 (6.84) 55.25 (6.77) Relations with Parents 53.36 (7. 37) 53.2 (7.37) Interpersonal Relations 53.41 (6.89) 54.9 (5.80) Self Esteem 52.18 (9.36) 57.5 (3.99) Self Reliance 53.68 (7.65) 50.2 (9.4) Note: Data are presented as group means (standard deviation) unless noted otherwis e At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 See page 40 for description of discrepancy calculation 2 n = 26 (patients) and n = 27 (siblings) in Oncology group as some scales were not applicable due to participants age
72 Table 3 4 Comparison of primary care s iblings and a normative s ample across informant type Mother Report of Siblings Self Report of Siblings Primary Care ( n = 35) Published Normative Sam ple Primary Care ( n = 35) Published Normative Sample BASC 2 Scales t test / U p value d t test p value d t test p value d t test p value d Behavioral Symptoms 1.31 .198 0 .36 .528 .602 0 .18 Withdrawal 0 .826 .412 0 .23 1.22 .23 0 .42 Atypicality 294.5 .408 0 .45 3.19 .031* 0 1. 1 122.5 .731 0 .29 3.18 .005 1.5 Attention Problems 0 .415 .680 0 .13 1.22 .232 0 .42 1.43 .172 0 .51 0 .444 .662 0 .20 Adaptive Problems 0 .236 .814 0 .07 .503 .618 0 .17 Adaptability 0 .474 .503 0 .13 .046 .963 0 .02 Social Skills 1.75 .86 0 .49 .279 .782 0 .10 Leadership 1 0 .03 .976 0 .01 1.58 .126 0 .68 Activities of Daily Living 0 .322 .749 0 .09 .393 .697 0 .13 Functional C ommunication 0 .121 .232 0 .03 .987 .331 0 .34 Externalizing Problems 265.0 .078 0 .03 .169 .866 0 .06 Aggression 1.08 .287 0 .30 .281 .780 0 .10 Conduct Problems 1 0 .956 .346 0 .34 .616 .543 0 .23 Hyperactivity 1 .06 .295 0 .29 .418 .678 0 .14 97.0 .206 0 .03 0 .120 .906 0 .06 Internalizing Problems 0 .749 .457 0 .21 1.34 .188 0 .46 127.0 .857 0 .54 3.05 .007 1.4 Anxiety 0 .904 .370 0 .25 1.84 .075 0 .63 0 .037 .971 0 .01 1.31 .206 0 .60 Depression 2.18 .034* 0 .60 .096 .92 4 0 .03 1.52 .138 0 .54 2.01 .058 0 .92 Somatization 279.0 .274 0 .25 1.52 .139 0 .52 3.65 .002 1 .72 1.03 .327 0 .48 Locus of Control 1.29 .205 0 .46 1.4 .179 0 .64 Social Stress 0 .754 .464 0 .27 4.88 .000 2.2 Sense of Inadequ acy 1.03 .313 0 .36 2.33 .032* 1.1 School Problems 2.53 .025* 0 .89 2.00 .060 0 .91 Attitude to School 74.0 .036* 0 .24 1.01 .326 0 .46 Attitude to Teachers 2.68 .020* 0 .95 1.68 .110 0 .77 Sensation Seeki ng* 0 .698 .495 0 .34 0 .924 .375 0 .43 Inattention/Hyperactivity 0 .277 .783 0 .10 0 .257 .800 0 .11 Emotional Symptoms 1.1 .276 0 .39 3.42 .003 1.6 (c ontinued)
73 Table 3 4 (cont.) Comparison of primary c are s iblings an d a normative s ample across informant type Mother Report of Siblings Self Report of Siblings Primary Care ( n = 35) Published Normative Sample Primary Care ( n = 35) Published Normative Sample BASC 2 Scales t test / U p value d t test p value Cohen d t test p value d t test p value d Personal Adjustment 1.13 .268 0 .40 3.45 .003 1.6 Relations with Parents 0 .325 .747 0 .11 1.93 0 .069 0 .88 Interpersonal Relations 101.0 .259 0 .69 3.75 0 .001 1. 7 Self Esteem 101.5 .269 0 .99 8.26 0 .000 3.8 Self Reliance 0 .941 .354 0 .33 .095 0 .925 0 .04 Note: Statistically significant at p < .05. **Statistically significant at p < .01 1 n = 27in o ncology group and n = 7 in primary c ar e group as some scales were not applicable due to participants age.
74 Table 3 5 Psychological f unctioning of primary c are c hildren across informant t ype Patients Siblings BASC 2 Scales Mother Report ( n = 20) Self Report ( n = 12) Mother Report ( n = 20) Self Report ( n = 6) Behavioral Symptoms 48.65 (8.56) 47.95 (9.01) Withdrawal 50.25 (9.89) 45.7 (7.89) Atypicality 48.0 (6.83) 48.75 (10.81) 47.95 (9.01) 47.0 (6.0) Attention Problems 50.35 (12.63) 54.00 (13.0) 50.3 (10.12) 60.0 (9.92) Adapti ve Problems 52.4 (10.21) 51.35 (8.95) Adaptability 49.35 (9.45) 50.85 (9.55) Social Skills 52.55 (12.23) 54.75 (9.57) Leadership 1 54.62 (10.94) 53.14 (12.8) Activities of Daily Living 52.95 (10.73) 49.5 (8.99) Functional Communicat ion 50.85 (8.66) 48.0 (8.85) Externalizing Problems 49.2 (8.4) 46.2 (9.06) Aggression 48.35 (10.75) 45.85 (9.33) Conduct Problems 1 49.62 (8.31) 47.14 (10.25) Hyperactivity 51.2 (7.82) 49.9 (9.99) 47.9 (9.48) 58.33 (14.0) Internalizing Problems 46.8 (7.88) 50.25 (9.55) 45.05 (11.12) 46.83 (9.13) Anxiety 46.8 (10.0) 50.33 (10.33) 48.75 (10.86) 49.17 (11.02) Depression 47.0 (8.12) 50.08 (9.1) 43.6 (8.42) 46.83 (3.6) Somatization 48.35 (10.87) 45.43 (4.24) 45.9 (11.7) 47.75 (9.5) Locus of Control 51.75 (8.88) 47.83 (5.95) Social Stress 51.58 (13.44) 46.83 (11.29) Sense of Inadequacy 50.75 (8.61) 46.83 (3.6) School Problems 56.17 (15.04) 50.83 (10.17) Attitude to School 54.33 (13.53) 49.0 (6.26) Attitude to Teache rs 56.42 (16.53) 50.33 (9.14) Sensation Seeking 54.43 (8.46) 55.00 (12.33) Inattention/Hyperactivity 52.08 (11.68) 60.33 (11.98) Emotional Symptoms 50.58 (10.53) 46.67 (10.29) (continued)
75 Table 3 5 (cont.). Psychological f unctioning of primary c are children across informant t ype Patients Siblings BASC 2 Scales Mother Report ( n = 20) Self Report ( n = 12) Mother Report ( n = 20) Self Report ( n = 12) Personal Adjustment 51.00 (9.51) 55.67 (6.5) Relations with Parents 52.5 (7.48) 53.5 (6.32) Interpersonal Relations 50.00 (12.18) 58.17 (3.71) Self Esteem 50.17 (10.61) 53.5 (10.56) Self Reliance 50.5 (12.17) 52.33 (5.32) Note: Data are presented as group means (standard deviation) unless noted otherwise. At risk scor es = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 n = 13 (patients) and n = 7 (siblings) in primary c are group as some scales were not applicable due to participants age
76 Table 3 6 Proportion of o ncology and primary c are s iblings with mothers reporting psychological p roblems BASC 2 Scale Oncology ( n = 35) Comparison ( n = 20) p value Externalizing Problems 20.0 (7) 10.0 (2) 0.462 .130 Hyperactivity 20.0 (7) 15.0 (3) 0.731 .062 Aggression 22.9 (8) 5.0 (1) 0.133 .232 Conduct Problems 1 11.1 (3) 14.3 (1) 1.0 .066 Internalizing Problems 17.1 (6) 10.0 (2) 0.696 .097 Anxiety 8.6 (3) 20.0 (4) 0.242 .165 Depression 22.9 (8) 5.0 (1) 0.133 .2 32 Somatization 11.4 (4) 15.0 (3) 0.696 .052 Behavioral Symptoms 17.1 (6) 5.0 (1) 0.402 .175 Atypicality 5.7 (2) 15.0 (3) 0.342 .155 Withdrawal 11.4 (4) 5.0 (1) 0.643 .108 Attention Problems 25.7 (9) 15.0 (3) 0.503 .125 Adaptive Problems 11.4 (4) 0.0 (0) 0.285 .216 Adaptability 20.0 (7) 15.0 (3) 0.731 .062 Social Skills 28.6 (10) 10.0 (2) 0.176 .062 Leadership 1 14.8 (4) 14.3 (1) 1.0 .108 Activities of Daily Living 14.3(5) 25.0 (5) 0.469 .134 Functional Communication 17.1 (6) 20.0 (4) 1.0 .036 No te: Values are presented as percentage of children falling in the at risk/clinically significant range (n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 20 for Adaptive Scales) 1 n = 27 for the Oncology group and n = 7 for Primary Care group as some scales were not applicable due to participants age
77 Table 3 7 Proportion of o ncology and primary care s iblings with self r eported p sycholog ical p roblems BASC 2 Scale Oncology ( n = 19) Primary Care ( n = 6) School Problems 5.3 (1) 33.3 (2) Attitude to School 21.1 (4) 0.0 (0) Attitude to Teachers 15.8 (3) 16.7 (1) Sensation Seeking 1 10.5 (2) 25.0 (1) Internalizing 5.3 (1) 16.7 (1) Atypicality 5.3 (1) 0.0 (0) Locus of Control 10.5 (2 ) 0.0 (0) Social Stress 0.0 (0) 16.7 (1) Anxiety 10.5 (2) 16.7 (1) Depression 5.3 (1 16.7 (1) Somatization 1 10.5 (2) 25.0 (1) Sense of Inadequacy 0.0 (0) 0.0 (0) Inattention 15.8 (3) 21.1(4) Attention Problems 21.1 (4) 50.0 (3) Hyperactivity 21.1 ( 4) 33.3 (2) Personal Adjustment 5.3 (1 0.0 (0) Relations with Parents 0.0 (0) 0.0 (0) Interpersonal Relations 5.3 (1) 0.0 (0) Self Esteem 0.0 (0) 16.7 (1) Self Reliance 10.5 (2) 0.0 (0) Emotional Symptoms 0.0 (0) 16.7 (1) Note: Values are presented as percentage of children falling in the at risk/clinically significant range (n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 n = 12 for the o ncology gr oup and n = 4 for p rimary c are group as some scales were not applicable due to participants age
78 Table 3 8 Proportion o f o ncology s iblings with m other s r eporting p sychological p roblems in c ancer and t umor g roups BASC 2 Scale Leukemias ( n = 22) Lymphoma s ( n = 3) Solid Tumors ( n = 6) Brain T umors ( n = 4) Externalizing Problems 4.5 (1) 66.7 (2) 50.0 (3) 25.0 (1) Hyperactivity 9.1 (2) 33.3 (1) 50.0 (3) 25.0 (1) Aggression 4.5 (1) 100.0 (3) 50.0 (3) 25.0 (1) Conduct Problems 1 0.0 (0) 0.0 (0) 33.3 ( 2) 0.0 (0) Internalizing Problems 9.1 (2) 66.7 (2) 16.7 (1) 25.0 (1) Anxiety 4.5 (1) 66.7 (2) 0.0 (0) 0.0 (0) Depression 13.6 (3) 100.0 (3) 16.7 (1) 25.0 (1) Somatization 13.6 (3) 33.3 (1) 0.0 (0) 0.0 (0) Behavioral Symptoms 9.1 (2) 66.7 (2) 33.3 ( 2) 25.0 (1) Atypicality 22.7 (5) 0.0 (0) 0.0 (0) 0.0 (0) Withdrawal 18.2 (4) 0.0 (0) 0.0 (0) 0.0 (0) Attention Problems 22.7 (5) 33.3 (1) 16.7 (1) 50.0 (2) Adaptive Problems 4.5 (1) 33.3 (1) 16.7 (1) 25.0 (1) Adaptability 9.1 (2) 66.7 (2) 33.3 (2) 50. 0 (2) Social Skills 13.6 (3) 66.7 (2) 50.0 (3) 50.0 (2) Leadership 1 0.0 (0) 33.3 (1) 16.7 (1) 66.7 (2) Activities of Daily Living 4.5 (1) 33.3 (1) 16.7 (1) 50.0 (2) Functional Communication 18.2 (4) 33.3 (1) 16.7 (1) 0.0 (0) Note: Values are presented as % (n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 These scales were not applicable due to age
79 Table 3 9 Proportion of o ncology s iblings repor ting psychological p roblems in cancer and tumor g roups BASC 2 Scale Leukemias ( n = 22) Lymphomas ( n = 3) Solid Tumors ( n = 6) Brain tumors ( n = 4) School Problems 0.0 (0) 0.0 (0) 20.0 (1) 0.0 (0) Attitude to School 30.0(3) 0.0 (0) 20.0 (1) 0.0 (0) Atti tude to Teachers 10.0 (1) 0.0 (0) 20.0 (1) 50.0 (1) Sensation Seeking 1 0.0 (0) 0.0 (0) 20.0 (1) 0.0 (0) Internalizing 10.0 (1) 0.0 (0) 0.0 (0) 0.0 (0) Atypicality 0.0 (0) 50.0 (1) 0.0 (0) 0.0 (0) Locus of Control 20.0 (2) 0.0 (0) 0.0 (0) 0.0 (0) Soc ial Stress 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Anxiety 10.0 (1) 0.0 (0) 0.0 (0) 50.0 (1) Depression 10.0 (1) 0.0 (0) 0.0 (0) 0.0 (0) Somatization 1 0.0 (0) 50.0 (1) 0.0 (0) 0.0 (0) Sense of Inadequacy 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Inattention/Hyperact ivity 30.0 (3) 0.0 (0) 0.0 (0) 0.0 (0) Attention Problems 40.0 (4) 0.0 (0) 0.0 (0) 0.0 (0) Hyperactivity 40.0 (4) 0.0 (0) 0.0 (0) 0.0 (0) Personal Adjustment 30.0 (3) 50.0 (1) 0.0 (0) 0.0 (0) Relations with Parents 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Inte rpersonal Relations 0.0 (0) 50.0 (1) 0.0 (0) 0.0 (0) Self Esteem 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Self Reliance 10.0 (1) 0.0 (0) 20.0 (1) 0.0 (0) Emotional Symptoms 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Note: Values are presented as % (n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 Some scales were not applicable due to participants age
80 Table 3 1 0 Proportion of o ncology and primary care p atients with mo thers reporting psychological p roblems BASC 2 Scale Oncology ( n = 35) Primary Care ( n = 20) p value Externalizing Problems 5.7 (2) 15.0 (3) .342 .155 Hyperactivity 8.6 (3) 15.0 (3) .657 .099 Aggression 5.7 (2) 25.0 (5) .086 .278 Conduct Problems 1 0.0 (0) 23.1 (3) .031* 408 Internalizing Problems 17.1 (6) 5.0 (1) .402 .175 Anxiety 14.3 (5) 10.0 (2) 1.0 .062 Depression 20.0 (7) 10.0 (2) 1.0 .130 Somatization 40.0 (14) 20.0 (4) .149 .318 Behavioral Symptoms 5.7 ( 2) 10.0 (2) .616 .079 Atypicality 5.7 (2) 10.0 (2) .616 .079 Withdrawal 8.6 (3) 20.0 (4) .242 .165 Attention Problems 14.3 (5) 25.0 (5) .469 .134 Adaptive Problems 11.4 (4) 5.0 (1) .643 .108 Adaptability 8.6 (3) 15.0 (3) .657 .099 Social Skills 2.9 (1) 15.0 (3) .131 .287 Leadership 1 0.0 (0) 15.4 (2) .105 .329 Activities of Daily Living 14.3 (5) 20.0 (4) .709 .074 Functional Communication 14.3 (5) 10.0 (2) 1.0 .062 Note: Values are presented as %(n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 n = 26 for the Oncology group and n = 13 for Primary C are group as some scales were not applicable due to participants age Statistically significant at p < .05.
81 Table 3 11 Comparison s of oncology p atients to primary c are and n ormative s ample s Mother Report of Patients Self Report of Patients Primary Care ( n = 35) Published Normative Sample Primary Care ( n = 35) Published Normative Sample BASC 2 Scale s t test / U p value d t test p value d t test p value d t test p value d Behavioral Symptoms 1.24 .221 0 .34 2.77 .009 0 .95 Withdrawal 1.12 .267 0 .31 1.76 .087 0 .60 Atypicality 331.0 .745 0 .08 2.33 .025* 0 .80 0.581 .565 0 .21 2.28 .033* 0 .78 Attention Problems 0 .587 .560 0 .16 0 .919 .365 0 .31 1.66 .107 0 .59 1.13 .271 0 .39 Adaptive Problems 0 .155 .877 0 .04 1.91 .064 0 .66 Adaptability 1.38 .175 0 .38 1.88 .069 0 .64 Soci al Skills 0 .629 .534 0 .17 3.16 .003 1.1 Leadership 1 179.0 .796 0 .24 0 .448 .663 0 .44 Activities of Daily Living 1.07 .290 0 .29 0 .652 .948 0 .22 Functional Communication 0 .254 .800 0 .07 0 .969 .339 0 .33 E xternalizing Problems 0 .124 .221 0 .03 3.03 .005 1.0 Aggression 0 .667 .510 0 .18 2.94 .006 1.0 Conduct Problems 1 145.0 0.23 0 .59 3.78 .009 1.3 Hyperactivity 1.30 .200 0 .36 1.24 .222 0 .42 1.40 .170 0 .49 1.31 .271 0 .45 Internalizing Problems 1.86 .126 0 .51 1.03 .312 0 .35 0 .596 .555 0 .21 0 .998 .334 0 .34 Anxiety 0 .505 .616 0 .14 0 .899 .375 0 .31 0 .036 .971 0 .01 0 .244 .809 0 .08 Depression 285.0 .254 0 .31 0 .143 .888 0 .05 1.52 .139 0 .54 2.91 .008 1.0 Somatizati on 1 2.44 .018 0 .67 3.16 .003 1.08 2.85 .011* 1.0 2.66 .021* 1.4 Locus of Control 1.29 .205 0 .46 1.23 .231 0 .41 Social Stress 0 .923 .363 0 .33 1.14 .266 0 .39 Sense of Inadequacy 1.04 .307 0 .37 1.32 .202 0 .46 School P roblems 2.33 .026* 0 .82 4.06 .000 1.4 Attitude to School 2.42 .021* 0 .86 2.66 .015* 0 .91 Attitude to Teachers 3.46 .001 1.2 6.05 .000 2.1 Sensation Seeking 1 0 .698 .495 0 .25 0 .031 .764 0 .01 (continue d)
82 Table 3 11 (cont.) Comparisons of oncology p atients to primary c are and normative s amples Mother Report of Patients Self Report of Patients Primary Care ( n = 35) Published Normative Sample Primary Care ( n = 35) Published Normative Sample BASC 2 Scal es t test / U p value d t test p value d t test p value d t test p value d Inattention/Hyperactivity 0 .275 .784 0 .10 0 .652 .521 0 .35 Emotional Symptoms 1.15 .277 0 .41 1.84 .08 0. 18 Personal Adjustmen t 1.13 .268 0 .40 2.85 .01 0 .98 Relations with Parents 0 .324 .748 0 .11 2.13 .046* 0 .73 Interpersonal Relations 1.05 .302 0 .37 2.31 .031* 0 .79 Self Esteem 0 .571 .571 0 .20 1.09 .289 0 .37 Self Reliance 0 .094 .356 0 .03 2.24 .036* 0 .77 Note: Statistically significant at p < .05. *Statistically significant at p< .01 1 These scales were not applicable due to participants age
83 Table 3 12 Proportion of o ncology and primary care p atients reporting psychological p roblems BASC 2 Scale Oncology ( n = 22) Primary Care ( n = 12) School Problems 0.0 (0) 33.3 (4) Attitude to School 4.5 (1) 33.3 (4) Attitude to Teachers 0.0 (0) 25.0 (3) Sensation Seeking 1 13.6 (3) 28.6 (2) Internalizing 4.5 (1) 16.7 ( 2) Atypicality 0.0 (0) 16.7 (2) Locus of Control 4.5 (1) 16.7 (2) Social Stress 0.0 (0) 25.0 (3) Anxiety 13.6 (3) 16.7 (2) Depression 4.5 (1) 25.0 (3) Somatization 1 36.4 (8) 0.0 (0) Sense of Inadequacy 9.1 (2) 16.7 (2) I nattention /Hyperactivity 9.1 (2) 16.7 (2) Attention Problems 9.1(2) 50.0 (6) Hyperactivity 27.2 (6) 8.3 (1) Personal Adjustment 0.0 (0) 16.7 (2) Relations with Parents 9.1 (2) 0.0 (0) Interpersonal Relations 9.1 (2) 0.0 (0) Self Esteem 4.5 (1) 8.3 (1) Self Reliance 9.1 (2) 2 5.0 (3) Emotional Symptoms 4.1 (1) 25.0 (3) Note: Values are presented as percentage of children in at risk/clinically significant range (n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T scores 1 n = 12 for the Oncology group and n = 7 for Primary C are group as some scales were not applicable due to participants ag e
84 Table 3 1 3 Psychol ogical functioning of m others BSI Scale Oncology ( n = 35) Primary Care ( n = 20) Mann Whitney Statistic p value Somatization 51.34 (10.09) 50.25 (8.91) 338.0 .830 Obsessive Compulsive 55.17 (11.00) 47.25 (8.76) 206.5 .011* Interpersonal Sensitivity 51.97 (10.24) 46.5 (8.27) 235.5 .037* Depression 51.6 (10.57) 46.1 (7.18) 246.5 .0 52 Anxiety 52.71 (11.40) 45.4 (9.37) 219.0 .020* Hostility 53.57 (9.50) 46.2 (7.80) 193.5 .005* Phobic Anxiety 51.66 (8.44) 49.9 (7.43) 321.0 .57 Paranoid Ideation 52.71 (8.05) 48.55 (9.10) 238.0 .041* Psychoticism 53.23 (10.47) 51.55 (7.72) 339.0 .8 28 Global Severity Index 53.26 (11.34) 45.3 (9.93) 209.5 .014* Positive Symptom Distress Index 52.46 (12.55) 49.40 (7.88) 286.0 .258 Positive Symptom Total 52.0 (10.81) 44.85 (11.74) 216.5 .019 Note : BSI = Brief Symptom Inventory. Data presented as mea n ( SD ) Clinical cutoff score for classification as reporting poor functioning = Global Severity Index T score > 63 or two subscales with T score s > 63 Cutoff refers to a value indicative of clinical sig nificance. Values higher than the cutoff indicate cl inically significant problems *Statistically significant at p < .05. *Statistically significant at p< .0 1
85 Table 3 14 Propor tion of mothers reporting poor f unctioning BSI Scale Oncology ( n = 35) Primary Care ( n = 20) Test p value Cr Somatization 17.1 (6) 10.0 (2) .696 .097 Obsessive Compulsive 22.9 (8) 5.0 (1) .133 .232 Interpersonal Sensitivity 17.1 (6) 10.0 (2) .696 .097 Depression 20.0 (7) 5.0 (1) .234 .205 Anxiety 22.9 (8) 5.0 (1) .133 .232 Hostility 11.4 (4) 0.0 (0) .285 .212 Phobic Anxiety 17.1 (6) 10.0 (2) .696 .097 Paranoid Ideation 14.3 (5) 10.0 (2) 1.0 .062 Psychoticism 20.0 (7) 5.0 (1) .234 .205 Global Severity Index 22.9 (8) 5.0 (1) .133 .232 Positive Symptom Distress Index 28.6 (10) 10.0 (2) .176 .216 Positive Symptom Total 17.1 (6) 10.0 (2) .696 .097 Note : BSI = Brief Symptom Inventory. Values are presented as percentage of mothers with scores in the clinical range ( n ) Clinical cutoff score for classification as reporting poor functioning = Global Severity Index T score > 63 or two subscales with T score s > 63 Cutoff refers to a value indicative of clinical sig nificance. Values higher than the cutoff indicate clinically significant problems Table 3 15 Family functioning reported by o nc ology sib lings and m others FAD Scale Cutoff Score Mothers ( n = 35) Siblings ( n = 11) t test p value % of Discordant Mother Sibling Dyads 1 Problem Solving 2.2 1.91 (0.46) 2.11 (0.34) 1.46 .151 63.6 Comm unication 2.2 1.93 (0.38) 2.23 (0.35) 2 2.47 .017 54.5 Roles 2.3 2.17 (0.40) 2.21 (0.27) 0 .485 .630 54.5 Affective Responsiveness 2.2 1.75 (0.45) 2.20 (0.31) 2 3.08 .004** 54.5 Affective Involvement 2.1 1.85 (0.50) 2.20 (0.52) 2 2.56 .014 90.9 Behavioral Control 1.9 1.56 (0.33) 1.71 (0.35) 1.8 .079 100. 0 General Functioning 2.0 1.62 (0.45) 1.89 (0.33) 2.1 .041 72.7 Note: FAD = Family Assessment Device D ata presented as mean ( SD ) 1 See page 40 for description of discrepancy calculation Cutoff refers to a value indicative of clinical sig nificance. Va lues higher than the cutoff indicate clinically significant problems 2 Above clinical cutoff (each scale has a different cutoff score) *Statistically significant at p < .05. *Statistically significant at p< .01.
86 Table 3 16 Family functioning r eported b y oncology patients and m others FAD Scale Cutoff Score Mothers ( n = 35) Patient s ( n = 11) t test p value % of Discordant Mother Patient Dyads 1 Problem Solving 2.2 1.91 (0.46) 1.91 (0.30) 0 .017 .987 63.6 Comm unication 2.2 1.93 (0.38) 2.04 (0.27) 0 .894 .376 90.0 Roles 2.3 2.17 (0.40) 2.13 (0.29) 0 .350 .738 72.7 Affective Responsiveness 2.2 1.75 (0.45) 1.97 (0.45) 1.45 .156 63.6 Affective Involvement 2.1 1.85 (0.50) 2.19 (0.58) 2 1.89 .066 72.7 Behavioral Control 1.9 1.56 (0.33) 1.64 (0.46) 0 .594 556 90.0 General Functioning 2.0 1.62 (0.45) 1.75 (0.32) 0 .922 .361 45.5 Note: FAD = Family Assessment Device Data presented as mean ( SD ) 1 See page 40 for description of discrepancy calculation Cutoff refers to a value indicative of clinical sig nific ance. Values higher than the cutoff indicate clinically significant problems 2 Above clinical cutoff (each scale has a different cutoff score) and indicative of poor family functioning Table 3 17 Family f unctioning r eported by primary care g roup f amil y m embers FAD Scale Cutoff Score Mothers ( n = 20) Siblings ( n = 4) Patient ( n = 7) Problem Solving 2.2 1.65 (0.43) 2.21 (0.32) 1 1.91 (0.44) Comm unication 2.2 1.68 (0.36) 2.3 (0.31) 1 2.17 (0.43) Roles 2.3 1.93 (0.43) 2.3 (0.27 ) 1 2.12 (0.51) Affective Responsiveness 2.2 1.57 (0.44) 2.13 (0.44) 2 .26 (0.38) 1 Affective Involvement 2.1 1.79 (0.49) 2.36 (0.56) 1 2.33 (0.43) 1 Behavioral Control 1.9 1.47 (0.36) 2.25 (0.29) 1 1.6 (0.33) General Functioning 2.0 1.34 (0.32) 2.02 (0.32) 1 1.89 (0.38) N ote: FAD = Family Assessment Device Data presented as mean ( SD ) Cutoff refers to a value indicative of clinical sig nificance. Values higher than the cutoff indicate clinically significant problems Cutoff refers to a value indicative of clinical sig nificance. Value s higher than the cutoff indicate clinically significant problems 1 Above clinical cutoff and indicative of poor functioning
87 Table 3 1 8 Proportion of family members r eporting poor family f unctioning Oncology Group Primary Care G roup FAD Scale Parent ( n = 35) Patient ( n = 12) Sibling ( n = 12 ) Parent ( n = 20) Patient ( n = 7) Sibling ( n = 4) Problem Solving 20.0 (7) 25.0 (3) 33.3 (4) 10.0 (2) 0.0 (0) 25.0 (1) Communication 22.9 (8) 25.0 (3) 58.3 (7) 10.0 (2) 71.4 (5) 50.0 (2) Roles 40.0 (14) 41.7 (5) 41.7 (5) 15.0 (3) 42.9 (3) 50.0 (2) Affective Responsiveness 14.3 (5) 41.7 (5) 41.7 (5) 5.0 (1) 57.1 (4) 50.0 (2) Affective Involvement 22.9 (8) 58.3 (7) 50 (6) 20.0 (4) 71.4 (5) 75.0 (3) Behavioral Control 14.3 (5) 25.0 (3) 25 (3) 5.0 (1) 0.0 (0) 100.0 (4) General Functioning 22.9 (8) 50.0 (6) 50 (6) 5.0 (1) 57.1 (4) 75.0 (3) Note: FAD = Family Assessment Device Values are presented as percentage of participants falling in cl inically significant range ( n ) Table 3 19 Comparison of mothers p ercei ved family functioning between o ncology and primary care g roup s FAD Scale Cutoff Score Oncology Control t test p value Problem Solving 2.2 1.91 (0.46) 1.65 (0.43) 2.10 .041* Comm un ication 2.2 1.93 (0.38) 1.68 (0.36) 2.34 .023* Roles 2.3 2.17 (0.40) 1.93 (0.43) 2.11 .04* Affective Responsiveness 2.2 1.75 (0.45) 1.57 (0.44) 1.45 .154 Affective Involvement 2.1 1.85 (0.50) 1.79 (0.49) 0 .488 .627 Behavioral Control 1.9 1.56 (0.33) 1. 47 (0.36) 1.01 .318 General Functioning 2.0 1.62 (0.45) 1.34 (0.32) 2.21 .041* Note: Data presented as group means (standard deviation) Statistically significant at p < .05 Cutoff refers to a value indicative of clinical sig nificance. Values higher th an the cutoff indicate clinically significant problems
88 CHAPTER 4 D ISCUSSION Pediatric cancer is a life threatening disease that can have a significant psychological impact on patient s and the ir family members Extensive research has been conducted exam ining the psychological functioning of the patient and their parents (Alderfer, Long, Lown, et al., 2010; Prchal and Landolt, 2009). Recent research has begun to explore the emotional and behavioral functioning of the siblings and factors that may impact t Labay, & Kazak, 2003; Labay & Walco, 2004; Houtzager, Oort, Hoekstra Weebers, et al., 2004a); however, there have been some methodological problems with past research. The present s tudy sought to examine the functioning of oncology siblings within an environmental framework that incorporates the functioning of patients mothers, and framework we hop ed to begin to unearth factors that may contribute to their emotional, behavioral, and adaptive functioning A better understanding of contributing factors might allow for identification of siblings at risk for developing psychological problems, as well as help inform future intervention research The current study adds to the existing literature by using a broadband measure approach to psychological assessment in this population examining findings at both the group and individual level, utilizing multi in formant methods which include child report, and incorporating assessment of broader family functioning.
89 Psychological Functioning of Siblings of Pediatric Cancer Patients Utility of Evaluating Findings at Group and Individual Levels Overall (at the group l evel) siblings and mothers in the oncology group endorsed good emotional, behavioral, and adaptive functioning However, examination of the data revealed a subset of individuals experiencing psychological problems. Traditionally prior research has assess ed psychological functioning in these families solely by examining group means. However, as our results show, restricting examination of the data in this manner excludes important information. The current study adds to the literature by highlighting the im portance of calculating the proportion of children who fall in the at risk or clinically significant range on psychological measures We are not the first to use this method to identify subsets of individuals experiencing problems. Martin et al. (2012) use d this approach when examining the social emotional functioning of children diagnosed with Neurofibromatosis Typ e 1 and Plexiform Neurofibromas. These researchers found similar results to the current study : W hile the children did not have elevated scores a t the group level, a substantial proportion of children had scores suggestive of at risk or clinically significant social emotional problems The authors suggest that a subset of children experiencing significant social emotional problems is not accounted for by solely examining group means. While the methods employed in the current study allowed us to identify a subset of individuals experiencing psychological problems, normative data is not available to determine if the rates of problems are higher in fam ilies of children with cancer than a general population. For example, the B ehavior A ssessment S ystem for C hildren 2 nd Edition (BASC 2) manual does not report the percentage of children falling in the at risk or clinically significant ranges within the norm ative reference sample (Martin et al, 2012) Thus, it is difficult to
90 compare the proportion of children falling into at risk or clinical ranges in our sample to the expected proportions in a normative population. Future research should explore the percent ages of children falling into the at risk or clinical ranges for both normative and pediatric populations. When considering the methods challenge presented above, we attempted to find applicable way s to compare our sample to children in the United States (U.S.) as a whole. According to the most recent (2009) national report from the Federal Interagency Forum on Child and Family Statistics, approximately 5.3% of children (ages 4 17) in the United States have serious emotional or behavioral difficulties as r ated by their parents (Federal Interagency Forum on Child and Family Statistics, 2011). Thus, in order to compare our sample to the U.S. population as a whole, we compared the proportion of children (siblings and patients) falling in the at risk or clinica lly si gnificant range on the BASC 2, to the comparison proport ion in the U.S. population ( Figures 4.1 through 4.4) In other words, observed rates that were higher than 5.3% were considered to represent higher rates of problems tha n would be occurring in the general national population of children. Using this reference point, we found that a large proportion of siblings and patients fell in the at risk or clinically significant ranges across scales on the BASC 2 in both self report and parent report con ditions. Thus, while our groups may be reporting good psychological function ing at the group level, the proportion of children experiencing psychological problems appears higher in families of children with cancer than in the broader US population. It is t his subset of children who may benefit from further assessment or intervention. As suggested by Martin et al. (2012), and as
91 supported by our results, future studies should aim to examine factors associated with poor psychosocial outcomes for this subgroup of children. Comparison of o ncology and p rimary c are s iblings When comparing the psychological functioning of siblings in the oncology and primary care group s we found that oncology mothers reported observing more psychological symptoms in these children across nearly all of the measures although the group means indicated generally good reported functioning However, oncology mothers did report observing significantly higher symptoms of depressed mood in the siblings than were reported in the primary car e group. This finding is consistent with prior studies that have noted that depression is a common symptom of poor adjustment for siblings of cancer patients (Alderfer, 2010). It will be important for future researchers to examine potential factors that le ad to the development or maintenance of depressive symptom ato logy in the siblings of children with cancer including longitudinal studies depressive symptoms over time. It should also be mentioned that a large proportion (50%) siblings in our primary care group endorsed at risk or clinically significant levels of behaviors indicative of attention problems The fairly high rate of reported attention problems in our primary care group was unexpected and raises questions regarding the representativeness of our primary care group sample. Though we purposefully did not exclude children in either group who had pre existing psychological conditions (in an effort to have study samples representative of their broader population cohort), we did not expect to find attention problems reported in half of the children in our primary care group Though reported attentional problems alone do not indicate presence of an attentio nal disorder (such as Attention Deficit / Hyperactivity Disorder ; ADHD ) clinically significant levels of
92 attentional problems reported on the BASC 2 are not expected to occur in children who do not have psychiatric or neurological diagnoses. Thus, it is important to consider whether the rate of observed attention problems in our primary care group is similar to what we might exp ect in the general population. According to C enters for D isease C ontrol data from 2008 2010, approximately 9.4% of children ages 5 17 in the United States have been diagnosed with A ttention D eficit D is order or A ttention D eficit/ H yperactivity D isorder (based off of parent report) (National Center for Health Statistics, 2011). Our sample has a much higher proportion (50%) of children endorsing attention problems and thus our co mparison sample may not be r epresentative of American children as a whole. One hypothesis for the increased rate of attention problems in our control population is that mothers with inattentive children may be more likely to agree to participate in a project examining psychological f unctioning However, the number of families who declined to participate or failed to complete all study measures was quite small (n = 10 ) and assuming their inclusion and lack of reported attention problems would only bring the proportion of reported probl ems down to 30 % (still higher than the estimated 9.4% of children in the general population with clinically significant attention problems). It is also possible that this particular sample was not representative of the primary care clinic as a whole, and t hat a large r sample would have included more children with normative levels of a ttention p roblems. Regardless, it is unlikely that this issue with our primary care group accounts for the psychological problems found in the cancer group. Rather, it is possi ble that additional group differences in psychological functioning may be identified with a more representative primary care group
93 Util ity of multi informant assessment m ethods Our results suggest that oncology mothers described worse sibling functioning than the siblings themselves; however, none of the comparisons across informant type ( mother versus child) were statistically different. However, an interesting finding showed that over 50% of mother child dyads (both siblings and patients) were found to b e discordant across the psychological scales on the BASC 2. This supports research that functioning the same way (Achenbach, Mc Conaughy & Howell, 1987) Howeve r, the perce ntage of discordant pairs is very high across scales (up to 90%), which may suggest that the method used to define discordant pairs may be particularly sensitive to minor (and perhaps less meaningful) differences between dyads. Future researchers should de termine whether a higher cutoff point (i.e., making discrepancy score less than one standard deviation instead of half a deviation) may be more appropriate in defining discordant pairs. Nevertheless, r esults from the present study provide further evidence for the importance of gathering information on sibling functioning from both the parents and siblings themselves when assessing emotional, behavi oral, and adaptive functioning. Comparison of o ncology g roup to p ub lished questionnaire normative s ample The pr esent study also examined the proportion of mothers who reported observing psychological problems in their children (siblings) and our results support our initial hypothesis that a larger proportion of oncology mothers endorsed at risk or clinical ly signi ficant levels of psychological symptoms than the normative sample acro ss many of the domains measured However, when we statistically compared the observed proportions for the two groups there were no significant differences in the rates of
94 mother reporte d sibling psychological problems While at first glance it appears as if oncology mothers are not endorsing statistically different functioning than normative controls, further examination of the effect sizes show a different picture. For example, oncology mothers rated siblings as having more symptoms of anxiety, physical complaints, and unusual behaviors than are present in the published parent report normative sample data, but these differences were not statistically significant. However, the magnitude o f the group differences indicated medium to large effect sizes Thus, it is likely that with a larger study sample, these observed group differences would be statistically significant and more clearly reflect higher rates of mother reported psychological p roblems in s iblings of children with cancer Future studies should aim to include larger samples (at least 88 study participants for a medium effect) to increase the likelihood of finding statistically significant differences should they exist. Similar att ention should also report. Independent samples t tests and Mann Whitney tests found some statistically significant differences between the oncology siblings and their primary care and normative co mparison groups Examination of the associated effect sizes of these comparisons reveals that the majority of the comparisons produce medium or large effects sizes This suggests that a larger study sample may result in even greater levels of statistical significance of differences b etween groups. This replicates prior research suggesting overall good psychological functioning in siblings of pediatric cancer patients However, current results also highlight the need for additional research and clinical focus on a subgroup of siblings who do appear to be experiencing psychological problems.
95 Psychological Functioning of Pediatric Oncology Patients and Comparison to Primary Care Group and Normative Controls Supporting our initial hypothesis, mothers and patients did not endorse clinically significant levels of emotional, behavioral, or adaptive symptoms on the BASC 2. In fact, oncology mothers endorsed better functioning across most scales of the BASC 2, with effect sizes ranging from .22 to 1.3. Despite our small sample, we were able to d etect better psychological functioning across many scales of the BASC 2. One area in which oncology mothers and patients endorsed significantly more symptoms than the BASC 2 normative sample was in Somatization. The Somatization scale on the BASC 2 is int ended to be a measure of psychological distress expressed in a physical manner (Reynolds & Kamphaus, 2004). For example, a healthy child experiencing anxiety may complain of frequent headaches or stomach aches without a medical etiology for their pain. Whi le this may be an effective means of measuring psychological distress in a healthy population, the same may not hold true for a medical population. Pediatric oncology patients may experience physical pain and side effects from their diagnosis and treatment (i.e., pain, nausea). The higher levels of physical complaints in our population may result more from medical side effects from their treatment as opposed to true psychological distress. Results from the current study suggest that this scale may not appro priate for use in a pediatric oncology population or should be interpreted with caution. Interestingly, primary care patients endorsed significantly m ore difficulty adjusting to the school setting. The school scales on the BASC 2 are intended to measure a According to the BASC
96 a pervasive pattern of dissatisfaction with schooling, school personnel, and the structure It should be noted that approximately 60% of children in the oncology group were not enrolled in a traditional school setting at the time of the study, while only 25% of the primary care group were not enrolled in school If o ncology patients were enrolled in a hospital homebound or home school program, they were encouraged to answer school questions in relation to that program. It is possible that children who were not enrolled in school had lower scores on these scales due to the fact that these items are not applicable to home school or hospital homebound settings children, their parents are also filling two roles, parent and teacher, and it may be difficult for these children to be objective about their parent in the teacher role. For pediatric patients not enrolled in non traditional settings, it may be more appropriate for clinicians to ask patient centered questions that directly relate to their individual when doing school ) These limitations should be considered when choosing psychometric measures for future research in pediatric oncology population s Psychological Functioning of the Mothers Overall, oncology mothers as a group endorsed good personal psychological functioning. Previous research has s uggested that mothers of pediatric cancer patients may experience increased psychological distress (Vrijmoet Wiersma, VanKlink, Kolk, et al., 2008). Contrary to our hypothesis our sample of mothers did not endorse clinical levels of symptoms at the group level. Our measure of psychological distress, the B rief S ymptom I nventory (BSI) has been shown to be a valid measure of psychological
97 distress in cancer populations (Derogatis, 1993). One hypothesis for why mothers in our sample failed to show any psychol ogical distress is that the BSI is not a reliable calculated from the current study data set suggests good internal consistency of this measure in our oncology group. Thus, we have to cons ider the possibility that there is something unique about our sample population that leads to better outcomes than have been previously reported Given that our sample is reporting good psychological functioning, it is possible that they may not be represe ntative of the broader population of oncology mothers. Additionally, research has suggested that mothers who endorse their own psychological problems are more likely to endorse problems in their children (Cohen et al., 1994). We must take this into conside perception of their children. Oncology mothers at the group level did not endorse clinically elevated scores for the patients or siblings. Mothers in our sample may have been less likely to report psychological problems i n their children because they have a positive perception of their own functioning. Supporting our initial hypothesis, oncology mothers endorsed worse functioning than primary care mother s across several areas of the BSI We propose that oncology mothers in our sample endorsed more symptoms (albeit not clinically significant levels) of obsessive compulsive behavior, paranoia, and anxiety because treatment and functioning is out of their control, so they attempt to manage other factors i n their environment to regain some control. Nevertheless, the majority of mothers in our sample are doing well, with only a few moth ers endorsing poor functioning.
98 Family Functioning of Oncology Families Even though oncology siblings are endorsing emotiona l and behavioral functioning within normal limits, one area in which they perceived significant difficulties is family functioning. Perhaps the most interesting findings from the present study stem from the family functioning data Oncology siblings percei ved particular difficulties in family communication, involvement and interaction, and emotional investment. Oncology al., 1985). Additionally, siblings reported probl ems on t he emotional scales which are intended to measure e degree to which family members are involved and interested in the activities o (Miller, et al., 1985) Mothers and patients often spend a considerable amount of time together due to and forth from medical visits, and the impo rtant caregiving tasks at home. While this may lead to better communication, emotional expression, and understanding between the mothers and patients, siblings do not always get the opportunity to spend as much time with their mothers, which may in fact im pede effective communication and impact relationships. As seen in the discrepancy between mothers and siblings in our sample, mothers may be in a difficult position to see this impact on their child because they may not see the day to day influence of thei r changing roles or p resence within the family unit. Additionally, our results support previous research that suggests that individual functioning is correlated with family functioning. Siblings perceive that their own personal adjustment is closely relate
99 behavioral control (e.g., limit setting and boundaries) within the home. Additionally, siblings perceive themselves as being more self decreases and more in control as the fam emotional expression and recognition, investment in others, and overall functioning decreases Children also perceive worse relations with their parents as family roles deteriorate and perceive worse interpersonal ior control decreases. While these results cannot imply a causal relationship, they do suggest that the functioning of the family as a whole may impact become more independent as they perceive changes in their routines, family relationships, and structure. perception of family functioning in addition to their individual emotional and behavioral functioning, as they may be e xperiencing difficulties that are not perceived by their parents There are some limitations to our assessment of family functioning. Our measure of family functioning could only be completed by children 12 or older, thus our results did not include the re port of younger children who may perceive differences in family functioning. Future research shoul d seek to develop measures of perceive family functioning that can be completed by younger children. Finally, as many correlations were conducted on this data we cannot rule out the possibility of Type 1 error Future researchers could consider utilizing a smaller p value to reduce the likelihood of Type 1 error and recruiting a l arger sample of comparison children. Limitations and Future Directions There are several limitations to the present study that should be addressed. lymphoblastic l eukemia (ALL) is the most commonly diagnosed cancer in childhood
100 (C O ncology G roup 2012). Given the higher prevalence of ALL in the oncology population, it only sta nds to reason that more patients with ALL would be eligible for the study. In fact, 62% of the participants recruited at the clinic (62%) were diagnosed with ALL. We were able to recruit smaller groups of patients diagnosed with lymphomas, solid tumors, or brain tumors. The small sample sizes recruited in each of these groups prevented us from conducting meaningful comparisons between groups (i.e., brain tumors, lymphomas). However, in reviewing our results, we felt it would be informative to combine the le ukemia and lymphoma groups and the brain and solid tumor groups (given that they have similar treatments and risk profiles), and then examine whether there are differ ences between those groups ( Table s 4 1 an d 4 2). Results show that there were no statistically significant differences between the leukemia/lymphoma group and the brain/solid tumor group. However, o ur results are likely more representative of children diagnosed with leukemia and their families as opposed to children and families with o ther diagnoses Our results suggest that the percentage of patients and siblings falling into the at risk or clinically significant ranges differs between diagnoses; however, future studies should seek to recruit larger sample sizes that include children w ith various diagnoses in order to detect possible differences between diagnoses at the group level. Additionally, we did not include patients who had undergone bone marrow transplants. Bone marrow transplantation requires that patients stay in the hospital for several weeks to months at a time which present may present with unique challenges. Thus future studies should examine similar constructs with this population to patients and families to assess whether they have similar outcomes.
101 There were several fa ctors that restricted study recruitment that are important to consider when designing future studies, as small sample size clearly affected statistical power of the current study analyses Potential participants were primarily approached in two ways: at an outpatient appointment (for both groups) or during an inpatient stay (for the oncology group) W hile the majority of families were open and willing to participate in the study, many families recruited in an outpatient setting were not willing to stay past their appointment time to complete study measures. This was particular true for families in the primary care group. Also, m any families were interested in taking study measures home to complete. We chose to not allow families to complete study measures at home as we were concerned about the potential effect of families members discussing their answers with each other and possibility that the forms would not be returned. However, future studies should consider this possibility to potentially increase study participation. Furthermore some children came accompanied to their appointments by fathers, grandparents, or other guardians, so while they met all other inclusion criteria, they were not eligible to participate without a mother present Finally, we only were able to recruit from one u niversity affiliated hospital pediatric oncology clinic which clearly limited our available sample for recruitment. F uture research teams should consider the possibility of collaboration with other medical centers to approac h more potential participants, which may also increase the likelihood of enrolling partici pants to more varied diagnoses. The present study used a cross sectional design which does not allow us examine how these children function over time. The structure o f the current study did not allow us to make comparisons between families who had recently started cancer treatment
102 versus families who had been undergoing treatment for a number of years. Our sample included a range of patients who started treatment two m onths prior to study participation to oncology patients who had been in treatment for 39 months. The literature is inconsistent regarding whether families experiencing more distress long term or immediately after diagnosis. Future research should continue to address this discrepancy and which factors lead to positive or improved functioning over time. The current study screened for serious mental health disorders (i.e., a utism, schizophrenia), and while they were not specifically targeted, participants diag nosed with mild psychiatric conditions (i.e., d epression) were eligible and recruited for the study. One must take into consideration that parents and children diagnosed with these disorders may have been more likely to endorse maladaptive symptoms on the BASC 2 and/or BSI. Thus we cannot definitely know whether the endorsed symptoms are a function of their adjustment to the cancer diagnosis or their underlying pre existing psychiatric condition. However, this effect is likely minimized in our group compari son since more serious psychiatric conditions were excluded and participants with mild psychiatric conditions were included in both the cancer and primary care group s. The BASC 2 parent report form has three different questionnaire versions, from which a n age appropriate version is selected for use Since children in our sample ranged in age from 2 to 17 years all three BASC 2 versions had to be used in this study. As suggested by Wolfe Christensen et al. (2009), the BASC 2 scales are supposed to measure the same constructs across the three age group versions ; however, given the inherent differences and developmental expectations for children at different ages, there may be some variation between forms. Additionally, while there are both parent and
103 self re port versions of the BASC 2, there are a limited number of psychological symptom scales that are found on both the parent and child report versions. Thus, we were not able to make informant comparison across all scales. Additionally, as noted on the parent report forms, it is possible that constructs being measured (e.g., a nxiety ) are somewhat different between parent and self report versions (i.e. questions assessing symptoms of anxiety are not worded exactly the same on parent and self report forms) The significant age difference between our patient and primary care groups is a potential limitation that warrants discussion The mean age for our community group was approximately 3 years younger than our oncology group We used age referenced scores, so th is method should control for differences based solely on age. However, because we use age references scores, we may not have observed any age effects unique to a cancer population. Nevertheless, some of our findings may have been due to differences between the age groups and not cancer specifically Strengths and Conclusions The current study adds to the literature in several important ways. First, our results provide further evidence that both pediatric oncology patients and their siblings as a group are d oing well. However, the current study also suggests that a subgroup of both patients and their siblings are experiencing at risk or clinically significant emotional and behavioral problems. This is an important methodological finding that suggests that res earch reliant on examination of group means may fail to detect a subset of children who are experiencing significant difficulties. This speaks to the importance of a universal screening approach in clinical settings which would allow mental health professi onals in oncology settings to quickly and efficiently identify the small number of children experiencing emotional and behavioral difficulties without having to do a full
104 psychological evaluation with each family when the majority of families are likely do ing well. Second, the current study highlights the importance of including both parent and self reports measures when working with these families. Our results show that both patients and siblings endorsed better emotional and behavioral functioning than wh at was reported by their mothers. Although these differences were not statistically significant at the group level, u sing a model utilized by Guite, 2004 we found that the majority of mother child dyads were significantly discordant in their report of emot ional, behavioral, adaptive, and family functioning. Although it is difficult to know whether one to consider in the assessment process and in targeting interventions. Furthermore, children who perceiv e fewer problems than their mother may be less likely to participate in or benefit from psychological intervention (Yeh and Weisz, 2001 as cited in Davidson, 2005). This also speaks to the importance of communication betwee n family members, which leads to the final important aspect of th e current study design T his study adds to the literature by providing additional quantitative measures of the broader family functioning. Much of the previous work in thi s area has included qualitative reports and the current study sought to include a quantitative measure collected from both the sibling and the mother, such that informant differences could be explored. One of the most interesting findings from the current study is the discrepancy between cancer sibling and mother report of family functioning, with siblings perceiving significant ly more impairments in family functioning. This may be an area of potential intervention for mental health professionals working wi th these families.
105 While siblings may not need individual psychological intervention, they may benefit from interventions targeting problems within the family system and allowing for a setting in which siblings can discuss the impact of their dia gnosis and treatment on their own life and family interactions. In conclusion, findings support existing literature describing generally good psychological functioning of pediatric oncology patients and their siblings. However, the study design allowed for experiencing psychological problems and highlights the utility of multiple informants and broader assessment of family functioning. It also speaks to the need for additional research examining other fa ctors that may impact individual functioning, differences between different cancer groups, and the importance of primary care group s, and larger samples pediatric oncology populations to have adequately powered studies to examine these questions within a f amily system framework
106 Table 4 1. Propor tion of oncology siblings with mothers reporting psychological p roblems in cancer and tumor g roups BASC 2 Scale Non Solid ( n = 25) Solid Tumors ( n = 10) Test p value Externalizing Probl ems 12.0 (3) 40.0 (4) .155 .316 Hyperactivity 12.0 (3) 40.0 (4) .155 .316 Aggression 16.0 (4) 40.0 (4) .186 .258 Conduct Problems 1 0.0 (0) 20.0 (2) .103 .400 Internalizing Problems 16.0 (4) 20.0 (2) 1.0 .048 Anxiety 12. 0 (3) 0.0 (0) .542 .194 D epression 24.0 (6) 20.0 (2) 1.0 .043 Somatization 16.0 (4) 0.0 (0) .186 .258 Behavioral Symptoms 16.0 (4) 30.0 (3) 1.0 .077 Atypicality 20.0 (5) 0.0 (0) .292 .258 Withdrawal 16.0 (4) 0.0 (0) .186 .258 Attention Problems 24.0 (6) 30.0 (3) .694 .062 Adaptive Problems 8.0 (2) 20.0 (2) .561 .170 Adaptability 16.0 (4) 40.0 (4) .186 .258 Social Skills 20.0 (5) 50.0 (5) .107 .300 Leadership 1 4.0 (1) 30.0 (3) .093 .369 Activities of Daily Living 8.0 (2) 30.0 (3) .128 .284 Functional Communication 20.0 (5) 10.0 (1) .649 .120 Note: Values are presented as % (n). At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 These scales were not applicable due to age.
107 Table 4 2. Proporti on of oncology siblings reporting p sycho logical problems in cancer and t umor g roups BASC 2 Scale Non Solid ( n = 13 ) Brain tumors ( n = 7 ) School Problems 0.0 (0) 14.3 (1) Attitude to School 23.1 (3 ) 14.3 (1) Attitude to Teachers 7.7 ( 1 ) 28.6 (2) Sensati on Seeking 1 0.0 (0) 20.0 (1) Internalizing 7.7 ( 1 ) 0.0 (0) Atypicality 7.7 (1) 0.0 (0) Locus of Control 15.4 (2) 0.0 (0) Social Stress 0.0 (0) 0.0 (0) Anxiety 7.7 (1 ) 14.3 (1) Depression 7.7 (1 ) 0.0 (0) Somatization 1 14.3 (1) 0.0 (0) Sense of Inade quacy 0.0 (0) 0.0 (0) Inattention/Hyperactivity 23.1 (3) 0.0 (0) Attention Problems 30.8 (4) 0.0 (0) Hyperactivity 30.8 (4) 0.0 (0) Personal Adjustment 30.8 (4) 0.0 (0) Relations with Parents 0.0 (0) 0.0 (0) Interpersonal Relations 7.7 (1) 0.0 (0) S elf Esteem 0.0 (0) 0.0 (0) Self Reliance 7.7 (1) 14.3 (1) Emotional Symptoms 0.0 (0) 0.0 (0) Note: Values are presented as % (n) At risk scores = T scores between 60 69 (30 39 for Adaptive Scales) and Clinically Significant scores = T 1 These scales were not applicable due to participants age
108 Figure 4 1 Proportion of s iblings with mothers reporting psychological p roblems. 1 n = 27 for siblings.
109 Figure 4 2 Proportion of oncology siblin gs reporting psychological p roblems 1 n = 7 for siblings.
110 F igure 4 3. Proportion of oncology p atients with mothers reporting psycholo gical p roblems 1 n = 26 *Chi Square test is statistically significant.
111 Figure 4 4. Proportion of oncology p atients r eporting p sychological p roblems 1 n = 13.
112 Figure 4 5 Proportion of oncology mothers reporting psychological problems. Figure 4 6 Proportio n of discordant pairs in the on cology group on the B ehavior A ssessment S ystem for C hildren 2 nd Edition
113 Figu re 4 7 Scores for perceived family functioning in the oncology group
114 LIST OF REFERENCES Achenbach, T.M., McConaughy, S.H., & Howell, C.T. (1987). Child and adolescent behavioral and emotional problems: Implications of cross informant correlations fo r situational specificity. Psychological Bulletin 101 (2 ), 213 232. doi: 10.1037/0033 2909.101.2.213 Achenbach, T. M. (1991a). Manual for the Child Behavior Checklist and 1991 profile. Burlington, VT: University of Vermont, Department of Psychiatry. Achen bach, T. M. (1991b). Integrative guide for the 1991 CBCL/4 18, YST, and TRF profiles. Burlington, VT: University of Vermont, Department of Psychiatry. Alderfer, M.A., Labay, L.E., Kazak, A.E. (2003). Brief report: Does posttraumatic stress apply to sibling s of childhood cancer survivors? Journal of Pediatric Psychology, 28 (4), 281 286. doi :10.1093/jpepsy/jsg016 Alderfer, M.A., Long, K.A., Lown, E.A, Marsland, A.L, Ostrowski, N.L, Hock, J.M., & Ewing, L.J. (20 10 ). Psychosocial adjustment of siblings of child ren with cancer: a systematic review. Psycho Oncology, 19 (8), 789 805. doi : 10.1002/pon.1638 Cadman, D., Boyle, M., & Offord, D.R. (1988). The Ontario child health study: Social adjustment and mental health of siblings of children with chronic health probl ems. Journal of Developmental Behavioral Pediatrics 9 117 121. doi : 10.1097/00004703 198806000 00001 (Ruccione, K. & Reaman, G., Chairs ). (200 5). Family Handbook for Children with Cancer. Retrieved from http://www.curesearch.org/pdf/Family_Handbook_for_CureSearch.pdf Cohen, D.S., Friedrich, W.N., Jaworski, T.M., Copeland, D., & Pendergrass, T. (1994). Pediatric cancer: Predicting sibling adjustm ent Journal of Clinical Psychology 50 303 319. doi : 10.1002/1097 4679(199405)50:3<303::AID JCLP2270500302>3.0.CO;2 # Davidson, K. H. (2005). Understanding parent and child report in a sample of pre pubertal children with mood disorders: Does family psyc hoeducation lead to greater agreement between parents and children? (Doctoral D issertation ) Retrieved from ProQuest Dissertations & Theses. (Publication No. AAT 3177168) Derogatis, L.R. (19 93 ). Brief Symptom Inventory. Baltimore: Clinical Psychometric Res earch. Derogatis, L.R., & Melisaratos, N. (1983). The Brief Symptom Inventory: an introductory report. Psychological Medicine 13 : 595 605. doi : 10.1017/S0033291700048017
115 Eiser, C., Hill, J., & Vance, Y. (2000). Examining the Psychological Consequences of Surviving Childhood Cancer: Systematic Review as a Research Method in Pediatric Psychology. Journal of Pediatric Psychology, 25 (6), 449 460. doi : 10.1093/jpepsy/25.6.449 Eiser, C., & Morse, R. (2001). Can parents rate their child's health related quality of life? Results of a systematic review. Quality of Life Research, 10 (4): 347 357. doi : 10.1023/A:1012253723272 Elkin, T.D., Phipps, S., Mulhern, R.K., Fairclough D. (1997). Psychological functioning of adolescent and young adult survivors of pediatric malig nancy. Medical and Pediatric Oncology; 29 582 588. doi : 10.1002/(SICI)1096 911X(199712)29:6<582::AID MPO13>3.0.CO;2 8 Epstein, N., Baldwin, L., & B ishop, D. (1983). The McMaster Family A ssessment D evice. Journal of Marital and F amily T herapy, 9 171 180. doi : 10.1111/j.1752 0606.1983.tb01497.x Epstein, N., Baldwin, L, & Bishop, D. (2000). Family Assessment Device (FAD). Handbook of psychiatric measures. Washington, D.C. : American Psychiatric Association. Federal Interagency Forum on Child and Family Stati National Indicators of Well Being, 2011. Washington, DC: U.S. Government Printing Office. siblings concerns regarding the diagnosis and phase of illness. Journal of Pediatric Oncology Nursing 20 (3): 133 140. doi : 10.1053/jpon.2003.74 t abdominal pain or emotional disorders. Journal of Abnormal Child Psychology, 26, 381 391. doi: 10.1023/A:1021955907190 Guite, J., Lobato, D., Kao, B., & Plante, W. (2004). Discordance Between Sibling and Parent Reports of the Impact of Chronic Illness and Disability on Siblings. Children's Health Care, 33 (1), 77 92. doi : 10.1207/s15326888chc3301 Guite, J.W., Lobato, D., Shalon, L., Plante, W., & Kao, B. (2007). Pain, disability, and symptoms among siblings of children with functional abdominal pain. Journal of Developmental & Behavioral Pediatrics, 28 2 8. doi : 10.1097/DBP.0b013e3180307c26
116 Hill, J.M., Kornblith, A.B. Jones, D., Freeman, A., Holland, J.F., Glicksman, A.S., et al. (1998). A comparative study of the long term psychosocial functioning of childhood acute lymphoblastic leukemia survivors treated by intrathecal methotrexate with or without cranial radiation Cancer, 82 (1), 208 218. doi : 10.1002/(SICI)1097 0142(19980101)82:1<208::AID CNCR26>3.0.CO;2 5 Hollingshead, A.B. (1975). Four Factor Index of Social Status. New Haven, CT. Yale University Press. Houtzager, B.A., Grootenhuis, M.A., & Last, B.F. (2001). S upportive groups for siblings of pediatric oncology patients: impact on anxiety. Psychooncology, 10 (4), 315 324. doi : 10.1002/pon.528 Houtzager, B.A., Oort, F.J., Hoekstra Weebers, J.E., Caron, H.N., Grootenhuis, M.A., & Last, B.F. (2004 a ). Coping and fami ly functioning predict longitudinal psychological adaptation of siblings of childhood cancer patients. Journal of Pediatric Psychology, 29 (8) 591 605. doi : 10.1093/jpepsy/jsho61 Houtzager, B.A., Grootenhuis, M.A., Caron, H.N., & Last, B.F. (2004b). Qualit y of life and psychological adaptation in siblings of paediatric cancer patients, 2 years after diagnosis. Psycho Oncology, 13 499 511. doi : 10.1002/pon.759 Houtzager, B.A., Grootenhuis, M.A., Caron, H.N., & Last, B.F. (2005). Sibling Self Report, Parenta l Proxies, and Quality of Life: The importance of multiple informants for siblings of a critically ill child Pediatric Hematology and Oncology 22 25 40. doi : 10.1080/08880010590896233. Kazak, A.E., Rourke, M.T., Alderfer, M.A., Pai, A. Reilly, A.F., Mea dows, A.T. (2007). Journal of Pediatric Psychology 32 (9) 1099 1110. doi :10.1093/jpepsy/jsm031 Labay, L.E., Walco, G.A. (2004). Brief report: Empathy and psychological adjustment in siblings of children with cancer Journal of Pediatric Psychology, 29 (4), 309 314. doi: 10.1093/jpepsy/jsho32 Lobato, D. J., & Kao, B. T. (2002) Integrated sibling parent group intervention to improve sibling knowledge and adjustment to chronic illness and disability Journal of Pediatric Psychology, 27 (8): 711 716. doi : 10.1093 /jpepsy/27.8.711 Martin, S., Wolters, P., Baldwin, A., Gillespie, A., Dombi, E., Walker, K., & Widemann, B. (2012). Social emotional Functioning of Children and Adolescents with Neurofibromatosis Type 1 and Plexiform Neurofibromas: Relationships With Cogni tive, Disease, and Environmental Variables. Journal of Pediatric Psychology, 37 (6), 1 12 doi:10.1093/jpepsy/jsr124 Miller, I., Epstein, N., Bishop, D., & Keitner, G. (1985). The McMaster Family Assessment Device: Reliability and Validity. Journal of Mar ital and Family Therapy, 11 (4), 345 356. doi: 10.1111/j.1752 0606.1985.tb00028.x
117 Moore, I.M., Challinor, J, Pasvog el, A., Matthay, K., Hutter, J., & Kaemingk, K. (2003). Behavior adjustment of children and adolescents with cancer: teacher, parent, and sel f report. Oncol ogy Nurs ing Forum ; 30 (5), 84 91. doi : 10.1188/03.ONF.E84 E91 National Center for Health Statistics. Health, United States, 2011: With Special Feature on Socioeconomic Status and Health. Hyattsville, MD. 2012. Patenaude, A. F., & Kupst, M. J. (2005). Psychosocial functioning in pediatric cancer. Journal of Pediatric Psychology, 30 9 27. doi : 10.1093/jpepsy/jsi012 Perrin, E. C., Stein, R. E., & Drotar, D. (1991). Cautions in using the Child Behavior Checklist: O bservations based on research about children with a chronic illness. Journal of Pediatric Psychology, 16 411 421. doi : 10.1093/jpepsy/16.4.411 Prchal, A. & Landolt, M.A. (2009). Psychological Interventions with siblings of pediatric cancer patients: a sys tematic review. Psycho Oncology, 18 : 1241 1251. doi : 10.1002/pon.1565 Assessment System for Children (BASC). New York: Guilford. Ries L Melbert D Krapcho M et al. SEE R Cancer Statistics Review, 1975 2004. Bethesda, MD: National Cancer Institute. Retrieved December 26, 2007, from http://seer.cancer.gov/csr/1975_2004 Sawyer, M. Antoniou, G., Toogood, I., Rice, M., & Ba ghurst, P. ( 2000 ). Childhood cancer: A 4 year prospective study of the psychological adjustment of children and parents. Journal of Pediatric Hematology/Oncology 22 (3), 214 220. doi : 10.1097/00043426 200005000 00006 Schultz, K.A.P., Ness, K.K., Whitton, J ., Recklitis, C., Zebrack, B., Robison, L.L., Zeltzer, L., & Mertens, A.C. (2007). Behavioral and Social Outcomes in Adolescent Survivors of Childhood Cancer: A Report From the Childhood Cancer Survivor Study. Journal of Clinical Oncology; 25 ( 24), 3649 365 6. doi : 10.1200/JCO.2006.09.2486 Simpson, G.A., Bloom, B., Cohen, R.A., Blumberg, S., & Bourdon, (2005). US children with emotional and behavioral difficulties. Data from the 2001, 2002, and 2003 National Health Interview Surveys. Adv Data. 23 1 13. Sidhu R., Passmore, A., & Baker, D. (2006). The effectiveness of a peer support camp for siblings of children with cancer. Pediatric Blood Cancer, 47 : 5 80 588. doi : 10.1002/pbc.20653. Sloper, P. & While, D. (1996). Risk factors in the adjustment of siblings of children with cancer. Journal of Child Psychology and Psychiatry, 37 (5): 597 607. doi : 10.1111/j.1469 7610.1996.tb01446.x.
118 Wolfe Christensen, C., Mullins, L. L., Stinnett, T.A., Carpentier, M.Y., Fedele, D. A. (2009). Use of the behavioral assessment syst em for children 2nd edition: Parent report scale in pediatric cancer populations. Journal of Clinical Psychology in Medical Settings, 16 (4), 322 330. doi 10.1007/s10880 009 9174 7 Verill, J. R., Schafer, J., Vannatta, K., & Noll, R. B., (2000). Aggression antisocial behavior, and substance abuse in survivors of pediatric cancer: Possible protective effects of cancer and its treatment. Journal of Pediatric Psychology, 25 493 502. doi : 10.1093/jpepsy/25.7.493 Vrijmoet Wiersma, J., V anKlink, J. M. M., Kolk, A. M., Koopman, H. M., Ball, L.M., & Egeler, L. M. (2008). Assessment of parental psychological stress in pediatric cancer: A review. Journal of Pediatric Psychology, 33 694 706. doi:10.1093/ jpepsy/jsn007 Yeh, M., & Weisz, J. R. (2001). Why are we here at the clinic? Parent child(dis)agreement on referral problems at outpatient treatment entry. Journal of Consulting & Clinical Psychology, 69 (6), 1018 1025. doi : 10.1037/0022 00 6X.69.6.1018 Zeltzer, K., Dolgin, M., Sahler, O., Roghmann, K., Barbarin, O., Carpenter, P., Copeland, D., Mulhern, R., & Sargent, J. (1996). Sibling adaptation to childhood cancer collaborative study: Health outcomes of siblings of children with cancer. Medical & Pediatric Oncology, 27 98 107. doi : 10.1002/(SICI)1096 911X(199608)27:2<98::AID MPO6>3.0.CO;2 O
119 BIOGRAPHICAL SKETCH Ryan Emily Fusse ll earned a Bachelor of Arts in p sychology and a Bachelor of Science in human d evelopment from the University of California, Davis. Ryan enter ed the University of Florida in 2007. interest in behavior management and pediatric oncology. Ryan completed a predoctoral received her Ph.D. from the University of Florida in August 2012. She is currently pursuing a postdoctoral fellowship in pediatric psyc Los Angeles.