EXAMINATION OF THE MULTIDIMENSIONAL FACTORS RELATED TO LONG TERM ADJUSTMENT IN ADULT SURVIVORS OF CHILDHOOD CANCER By ANNA CEJKA B.A., University of Colorado Boulder, 2006 M.A., University of Colorado Denver, 2008 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Clinical Health Psychology 2015
ii The thesis for Doctor of Philosophy degree by Anna Cejka has been approved for the Clinical Health Psychology Program By Peter Kaplan, Chair Kristin Kilbourn, Advisor Kevin Everhart Brian Greffe April 21 st 2015
iii Anna Cejka (Ph. D., Clinical Health Psycholog y ) Examination of the Multidimensional Factors Related to Long Term Adjustment in Adult Survivors of Childhood Cancer Thesis directed by Professor Kristin Kilbourn. ABSTRACT This dissertation responds to critical gaps in current research on multidimensional factors that affec t long term adjustment in adult survivors of childhood cancer. The study utilizes measures of multiple realms of functioning, including: quality of life, physical health, psychosocial functioning, and individual survivor characteristics It identifies the most important correlates of long term adjustment and posits a working model by which future interventions for improving adjustment in this population could be based The study has important implications for future researchers working with adult survivors of childhood cancer, clinicians and physicians involved in their care ; and policy makers evaluating the most effective use of resources for programs that support adult survivors of childhood cancer. Survivors with higher levels of posttraumatic growth, opt imism, resilience and social support also tended to report better overall long term adjustment. The form and content of this abstract are approved. I recommend its publication. Approved: Kristin Kilbourn
iv ACKNOWLEDGMENTS So many people provided support and encouragement during the development of this dissertation. I am grateful and blessed to have had each one be a part of this process. First, I want to thank my husband, Nathan Cejka, and my sons Romen and Miles Cejka for tolerating the many demands on my time, and numerous life obstacles achieving a doctorate has caused, as well as for their unwavering support and love throughout. I want to thank my parents, Sally and Eric Morgenthaler, for maintaining faith that I would f inish successfully, and always being the helping hands I needed in so many areas over the years. They never doubted that I would be able to realize my dream of earning the first PhD in the history of our family. I want to generally thank my family, friends and colleagues, for without the vibrancy joy, and passion they bring to my life, this achievement would mean very little. I want to thank my advisor, Kristin Kilbourn, for the years of support, encouragement, and understanding, and for being the career role model to which I have aspired. I want to thank the other members of my committee, Peter Kaplan, Kevin Everhart, and Brian Greffe, for their patience, guidance and encouragement throughout this process. Finally, I want to thank the team of physicians a nd nurses at the TACTIC clinic for the work they do for the adult survivor of childhood cancer population, and their dedication to helping these patients thrive and live healthier lives after treatment.
v TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ......................... 1 An Overview of Pediatric Cancer ................................ .......................... 1 Pediatri c Cancer Statistics ................................ ............................... 1 The Physical Impact of Cancer Treatment in Pediatric Populations ................................ .............................. 3 The Psychosocial Impact of Cancer Treatment in Pediatric Populations ................................ .............................. 4 Transition to Adulthood : The Effect of the Childhood Cancer Experience on the Adult Survivor ................................ ...... 5 The Physical Impact of Childhood Cancer Treatment in Adults ................................ ................................ .................... 6 The Psychosocial Impact of Childhood Cancer Treatment in Adults ................................ ................................ .................... 7 Beyond Survival: Personal Growth and Thriving Through the Cancer Experience ................................ .................... 9 Resilience and Adult Survivors of Childhood Cancer ............. 11 Aims and Hypotheses ................................ ................................ ....... 14 II. METHOD ... ................................ ................................ ................................ 17 Procedure ................................ ................................ .......................... 17 Measures ................................ ................................ ........................... 19 The Impact of Cancer Scale ................................ .......................... 19 Resiliency ................................ ................................ ...................... 19 Posttraumatic Growth ................................ ................................ .... 19 Optimism ................................ ................................ ....................... 20
vi Perceived Social Support ................................ .............................. 20 Emotional Health ................................ ................................ .......... 20 Mea sures of Long Term Adjustment ................................ ............ 21 Qual ity of Life and Health Status ................................ ............. 21 Life Sati sfaction ................................ ................................ ....... 21 Self Care Ac tivities and Health Behavio rs .............................. 21 Demo graphic and Medical Variables ................................ ............ 22 Participants ................................ ................................ ....................... 22 III. RESULTS ................................ ................................ ........................... 24 Power Analysis ................................ ................................ ................. 24 Data Analysis ................................ ................................ .................... 24 Means and Distributions for the Psychosocial and Medical Variables ................................ .............................. 25 Assessing Quality of Life ................................ .............................. 29 The Association Between Psychosocial Functioning Resili ence and Posttraumatic Growth ................................ ...... 34 The Association Between Participant Characteristics and Long Term Adjustment ................................ ..................... 35 T he Association Between Psychosocial Functioning, Posttraumatic Gr owth and Long Term Adjustment ................. 40 The Relationship Between Late Effects and Psychosocial Functioning ................................ ......................... 41 Comparison of Psychosocial Variables, Cancer Type and Age at Diagnosis ................................ ................................ 42 IV. DISCUSSION.............................................................. .... ............ .............. .45 Comparing Differences Between Cancer Types and
vii Age at Diagnosis ................................ ................................ ...... 45 Psychological Functioning and Health Rela ted Quali ty of Life ....... 47 Impact of Cancer Scale ................................ ................................ ..... 48 Posttraumatic Growth and Psychosocial Functioning ...................... 49 Long Term Adjustment ................................ ................................ .... 50 Long Term Adjustment an d Participant Characteristics ............... 50 Long Term Adjustment and Psychosocial Functionin g ................ 50 Limitations ................................ ................................ ........................ 53 Conc lusions and Fut ure Directions ................................ ................... 55 REFERENCES.................................... .. .......... .................. .............................. .....57 APPENDIX : Psychosocial Questionnaires . ....... ................. .... ........... ..... ...... .... ...62
1 CHAPTER I INTRODUCTION An Overview of Pediatric Cancer Pediatric Cancer Statistics In the Unite d States on average between 1 and 2 children in 10,000 develop cancer each year. Of the 12 major types of childhood cancers, leukemias and central nervous system cancers account for the majo rity of the new cases (ACS, 2014 ). Leukemias alone make up 34% of new cancer cases in children. The most common types in children are acute lymphoblastic leukemia (ALL) and acute myelo genous leukemia (AML) (ACS, 2014 ). Brain and central nervous system cancers are the second most common cancer in chi ldren and account for 27 % of new cancer cases (ACS, 2014 ). There are numerous types of brain and spinal cord tumors, and treatment and prognosis for each varies according to tumor type, stage and treatment response variables. Other types of common childhoo d cancers include neuroblastomas, Wilms tumors, and lymphomas which collectively account for approximately 16% of childhood cancers (ACS, 2014 ). Treatments for childhood cancer vary based on the type and s everity of the cancer (ACS, 2014 ). Most common tr eatments for childhood cancer include chemotherapy, surgery, radiation therapy and bone marrow transplants. Chemotherapy is a group of medications given either intravenously, orally or intrathecally (into spinal fluid) (ACS, 2014 ). Children may receive dai ly, weekly or monthly chemotherapy for varying amounts of time depending on the established treatment protocol for their type of cancer, and the resulting side effects of chemotherapy can be both short term and long term. Short term side effects can includ e fatigue, nausea, vomiting, hair loss, fatigue, anemia,
2 abnormal bleeding and increased risk for infection due to a decrease in w hite blood cell count (ACS, 2014 ). Long term side effects or "late effects" of chemotherapy include neurocognitive problems, h ormonal dysregulation growth irregularities, loss of limbs or organs, blindness, hearing problems, dental issues, infertility, and increased risk of second ary cancer occurrence (ACS, 2014 ; Elkin, Phipps, Mulhern & Fairclough, 1997). Radiation therapy is also frequently used in combination with chemotherapy in treatment for many types of childhood cancer, including brain tumors, Wilms tumor and head and neck cancers (ACS, 2014 ). Radiation therapy involves highly focused beams of x rays, gamma rays or part icle beams aimed at the tumor in order to kill the cancer cells and ultimately shrink the size of the tumor. Side effects of radiation therapy are dependent on the dosage of radiation and the location (ACS, 2014 ). Fatigue is the most common side effect of radiation and can last for up to six week s following treatment (ACS, 2014 ). Skin damage or radiation burns, including redness and irritation are usually temporary, although there can be permanent changes to the ski n following treatment (ACS, 2014 ). Similar to chemotherapy, radiation therapy can also cause hair loss and result in low levels of platelets and white blood cells (ACS, 2014 ). Long term or late effects associated with radiation therapy include neurocognitive problems, infertility, lymphedema, dent al issues, and secondary malignancies (NCI, 2012) As a result of advances in the treatment of childhood cancer, the survival rate for childhood cancers has increased substantially over the past 30 years. Most recent estimates indicate that 80% of children treated for cancer before age 20 survive 5 years, and 70% s urvive into adulthood (ACS, 2014 ; End Results Group, SEER Program, 200 7 ). Because survival rates have increased, there is an increased emphasis on interventions
3 that will enhance the psychological, social, and behavioral functioning of children and adults living with and surviving pediatric cancer s The Physical Impact of Cancer Treatment in Pediatric Populations It is undeniable that pediatric cancer survivors face many challenges associated with their cancer treatment. This includes, but is not limited to, painful procedures, frequent trips to the hospital and long inpatient stays which can lead to increased distress both in the patient and within the family (Kazak, Cant, Jensen et al., 2003). P ediatric cancer treatment can also have a long term impact on overall health and quality of life. Considerable physiological changes can occur as a result of the cancer and treatment, the results of which can be life long. As previously mentioned, chemothe rapy and radiation may damage organs and tissues and cause permanent physical impairment while increasing the risk for secondary or late effects of treatment such as delayed growth, endocrine issues, reduced fertility, increased fatigue, sleep disturbances and secondary malignancies (Elkin, Phipps, Mulhern & Fairclough, 1997; Clanton et al., 2011). Pediatric cancer survivors are also at risk for neurocognitive impairment which may become more apparent as the child matures. Those who received cranial radi ation and/or high dose or intrathecal chemotherapy, especially at a young age, are most likely to experience neurocognitive late effects (Robinson, Kuttesch & Champion, 2010). It is estimated that nearly 20% of survivors have some impairment (Clanton et al ., 2011). Cognitive impairment can also cause disruption in academic and occupational achievement in children and young adults (Charlton et al., 1991; Eiser, 1990a).
4 The Psychosocial Impact of Cancer Treatment in Pediatric Populations P ediatric cancer patients also experience a r ange of psychosocial problems. It was once assumed that moderate to severe levels of distress could be expected in all pediatric cancer patients ( Koocher & O'Malley, 1981 ; Cella et al. 1987 ; Mulhern et al. 1989) However, ther e a number of studies that show evidence that moderate levels of distress occurs in a minority of pediatric cancer patients (Fritz, Williams, & Amylon, 1988; Kupst & Schulman, 1988). In fact, Lavigne and Faier Routman (1992) found that among children with chronic illnesses, children with cancer were at significantly lower risk for psychological distress compared to age matched peers without cancer. Other studies have found that pediatric cancer survivors' psychosocial adjustment is not significantly differe nt from controls or no rms. R ates of depression and anxiety may be at or lower than the rates throughout the general pediatric population (Boman & Bodegard, 1995; Elkin et al., 1997; Gray et al., 1992; Kazak, 1994; Kazak et al., 1997; Kupst et al., 1995; M ackie et al., 2000; Madan Swain et al., 1994; Radcliffe, Bennett, Kazak, Foley, & Phillips, 1996; Simms, Kazak, Golomb, Goldwein, & Bunin, 2002) A lthough many studies have found that the average level of psychosocial adjustment in pediatric cancer patients and survivors is close to level s assessed in the general population, there is still a significant percentage (25 30%) of families and children who exhibit high levels of distress A number of research studies suggest that up to 50% of pediatric ca ncer patients and their families may experience maladjustment (Boman & Bodegard, 1995; Friedman & Meadows, 2002; Koocher & O'Malley, 1981; Kupst et al., 1995). Since distress and maladjustment occurs in a significant number of children and families affecte d by pediatric cancer, the question remains whether these children go on to experience
5 psychosocial challenges in adulthood. Identifying factors that may contribute to prolonged maladjustment is important for future research in pediatric cancer survivor po pulations. In addition to psychological effects of cancer treatment, there are also often social ramifications. C a ncer treatment typically increases a child's dependence on his or her parents, and decreases their participation in peer and school based acti vities due to long hospital stays and frequent trips to receive treatment (Pendley et al., 1997; Spirito et al., 1990; Vannatta et al., 1998a, 1998b). Those diagnosed as adolescents experience more disruption in their social development as compared to adul t survivors who were diagnosed at younger ages (Turner Sack, Menna & Setchell, 2012). Disruptions in social and academic development can threaten the accomplishment of developmental tasks, resulting in problems with psychosocial functioning that extend int o adulthood (Stam, Grootenhuis & Last, 2005) The long term effects of these potential developmental delays have just recently become a topic of interest amongst psychosocial oncology researchers. Transition to Adulthood: The Effect of the Childhood Cancer Experience on the Adult Survivor A dvances in pediatric cancer treatment s have led to steady increases in the survival rates of pediatric cancer survivors ( ACS, 2000; End Results Group, SEER Program, 200 7 ). However, there is little information about emotio nal and physical quality of life in adult survivors of childhood cancer. Specific emotional, social and physical outcomes are difficult to study due to methodological issues and large variances within
6 the population itself ( i.e. age at diagnosis, cancer type, and treatment regimen). An overview of some of the research that has been conducted in this area is presented below. The Physical Impact of Childhood Cancer Treatment in Adults As mentioned above, treatment regimens for pe diatric cancers are typically very intense and they often lead to a number of long term medical problems. The current research estimates that adult survivors of childhood cancer experience a variety of long term or late effects with 60 90% developing on e o r more chronic health conditions and 20 40% experiencing severe or life threatening complications into adulthood (Oeffinger, Mertens, & Sklar, 2006; Geenen, Cardous Ubbink, Kremer et al., 2007; Wasilweski Maker, Mertens & Patterson, et al., 2010; Stevens, Mahler & Parkes, 1998). Results from the ongoing Childhood Cancer Survivor Study indicate that 10 23% of adult survivors of childhood cancer report moderate to extreme pain (Hudson, Mertens, Yasui et al., 2003; Tsao, Leisenring, Robinson, Zeltzer, 2007), 16 40% experience significant fatigue (Mulrooney, Mertens, Neglia, et al., 2003; Mulrooney, Ness, Neglia, et al., 2008), and 12 16% report sleep problems (Mulrooney, Ness, Neglia, et al., 2008). Stevens et al. (1998) found that more than half of adult sur vivors of childhood cancer had at least one chronic medical problem. Oeffinger et al. (2000) showed that 69% of patients had at least one late effect, and 36% had two or more. Due to these late effects, survivors often report a lower Health Related Quality of Life (HRQoL) than their peers. HRQoL refers to the impact of health and illness on a person's QoL. Results of previous studies generally suggest that there is an increased risk for HRQoL concerns in those who have experienced one or more of the followi ng; a relapse of their cancer, severe functional impairment, and cranial irradiation. Female survivors
7 and those who identify themselves as minorities also report lower HRQoL (Stam, Grootenhuis, Caron & Last, 2005). The Psychosocial Impact of Childhood Ca ncer Treatment in Adults Findings regarding the long term psychosocial impact of the childhood cancer vary, however many studies have shown that a subset of the population (estimated at 20%) experience clinically sign i ficant levels of depression, anxiety, Posttraumatic Stress Disorder (PTSD) and are at risk for cognitive late effects as adults (Christ, Lane, & Marcove, 1995; Dolgin, Somer, Buchvald, & Zaizov, 1999; Zebra c k et al., 2002). Other psychological complicatio ns may include low self esteem, social isolation, and school or work difficulties (Ostroff & Stenglass, 1996). The majority of results, however, show that long term adult survivors report the same or better QoL (Langeveld et al., 2004; Pemberger et al., 20 05; Zelter et al., 2008) and psychological functioning than controls (Elkin et al., 1997; Langeveld et al., 2002). For instance, a number of studies report that adult survivors of childhood ca ncer do not show elevated levels of anx iety or depression or lo wer self esteem as compared to healthy controls (Barakat, Kazak, Meadows, Casey & Stuber, 1997; Eiser, Hill, & Vance, 2000; Elkin, Phipps, Mulhern and Fairclough, 1997; Fritz & Williams, 1988; Gray et al., 1992; Weigers, Chester, Zebrack & Goldman, 1998). Elkin and colleagues discovered that survivors' mean scores on all SCL 90 subscales were lower than the standardized sample and the distribution of scores on the Anxiety, Psychoticism, GSI, and PST scales were significantly below normative values (Elkin, Phipps, Mulhern & Farclough, 1997) Nevertheless, there was a small subset of survivors who displayed clinical elevations on the SCL 90. This group had experienced more frequent disease relapse and more severe functional impairment (Elkin, Phipps,
8 Mulhern & Farclough, 1997). R esearch examining PTSD symptoms in adult survivors of childhood cancer indicates that most survivors (approximately 83%) do not meet DSM I V criteria for the disorder and function at normal levels (Meeske et al., 2001). However, a lthough most are well adjusted, some of the research has shown that childhood cancer survivors in early adulthood are more likely to report PTSD symptoms and to experience significant psychosocial and physical impairment compared with healthy peers (Schwar tz & Drotar, 2006) Although most survivors of childhood cancer appear to be well adjusted, there remain common emotional, social, and physiological concerns as well as changes to the survivor s life course as a result of their cancer experience. Pediatr ic cancer survivors frequently report a sense of uncertainty, fears of recurrence, and physical impairment issues (Zebrack & Chesler, 2002). When thinking about their future, young adult survivors of childhood cancer often report concerns about the interru ption of life goals, infertility and discrimination in employment and insurance (Binger, 1984; Chesler & Barabin, 1986; Pendley, Dahlquist, & Dreyer, 1997; Richardson, Nelson, & Meeske, 1999; Van Dongen Melman, Pruyn et al., 1995) Additionally, adult surv ivors of childhood cancer often report that their cancer experience impacted their expected course of life. Research in this area falls under a couple of terms such as "hindrance to goals" (Schwartz & Drotar, 2009) or "hampered course of life" (Stam, Groot enhuis & Last, 2005). These terms refer to the developmental delays in milestones or life goals in the adult cancer survivor's life as a result of having experienced pediatric cancer. Changes may occur in a number of capacities including education, emplo yment, insurance, military service, marriage, decisions about having
9 children, recreation and value formation (Christ et al., 1995; Dolgin et al., 1999; Griffith & Hart, 2000; Hobbie et al., 2000; Kokkonen & Vainionpaa, 1996 Langeveld et al., 2002; Meeske et al., 2001; Ostroff & Steinglass, 1996; Pui et a;, 2003; Vann, Biddle, Daeschner, Chaffeem & Gold, 1995). Previous research on the course of life of adult survivors of childhood cancer indicates that many survivors achieve fewer milestones, or achieve th em at an older age than peers. After reaching adulthood, many survivors live with their parents longer, and have a lower prevalence of marriage and parenthood compared to age matched peers (Byrne et al., 1989; Langeveld et al., 2002, 2003; Rauk et al., 199 9; Teeter et al., 1987; Zevon et al., 1990). Also, a lower percentage of survivors were employed even though their reported education level was as high as th a t of their peers (Stam, Grootenhuis & Last, 2005) In particular, Schwartz and Drotar (2009) found that survivors who experienced impaired health status as a result of their diagnosis and treatment were significantly more likely to report that having had cancer as a child hindered their ability to achieve life goals. These delays suggest that there may be developmental consequences to experiencing cancer as a child that may extend into adulthood. Beyond Survival: Personal Growth and Thriving Through the Cancer Experience As the numbers of adult survivors of childhood cancer continues to grow, the inte rest in research on their subjective experience as adults has increased. Although researchers initially approached studying these survivors from a traditional biomedical model of disease and negative psychosocial sequelae, many are now looking to a strengt h based perspective that considers cancer as a catalyst for growth ( Carpenter, 1997; Carver, 1998; Chesler, 2000; Garmezy, 1991; O'Leary, 1998; O'Leary, Alday &
10 Ickovics, 1998; Parry & Chesler, 2005; Tedeschi & Calhoun, 1996 ). Even in the face of a traumat ic experience, some people can thrive and even achieve a higher level of functioning following the event (O'Leary, 1998; O'Leary & Ickovics, 1995). This phenomenon, often referred to as "posttraumatic growth", is a positive psychological change that invol ves shifting the perception of the trauma as a threat into a challenge (Calhoun & Tedeschi, 1999). Research on adult cancer survivors has revealed that many experience positive changes and benefits as a result their illness (Thornton, 2002), and that these often include broad positive changes in one's priorities, sense of self, sense of meaning, and outlook on life (Parry & Chesler, 2005; Adrykowski, Brady & Hunt, 1993; Frtiz & Williams, 1988; Kupst et al., 1995). These positive changes may lead to better l ong term adjustment in childhood cancer survivors, as those who display posttraumatic growth may utilize mo re positive coping strategies. Although many cancer survivors report positive outcomes as a result of their cancer experience (Parry & Chesler, 2005; Kazak, 1994 ), it is important to note that this is not the case for all survivors of childhood cancer (Parry & Chesler, 2005; Kazak, 1994 ). Nor does positive outcomes and benefit finding in one area mean that a survivor experiences these same benefits in all areas of physical and psychosocial functioning (Parry & Chesler, 2005). The reality is that many long term survivors of childhood canc er experience ongoing physical and psychosocial challenges (Kazak, 1994; Stuber et al., 1996; Weigers et al., 1998; Zebrack & Chesler, 2001). Just as posttraumatic growth can lead to more positive coping, research has shown that, in general, those with a m ore positive outlook in the face of trauma also tend to see the experience more as a challenge than as a threat. Optimism has also been found to
11 increase the utilization of more adaptive coping strategies and posttraumatic growth (Scheier, Weinraub, & Carv er, 1986). The more optimistic a person is, the better their long term emotional well being and those high in optimism also tend to have increased self efficacy and show increased resilience following a traumatic experience (Snyder, 2000; Carver, 2005). Re silience and Adult Survivors of Childhood Cance r Resilience in psychology refers to a person's ability to cope with and "bounce back" from adversity (physical and psychological). Although there may be personality traits (i.e. optimism, adaptive coping str ategies) that are related to resilience, resiliency itself is viewed as a process (Philippe, Laventure, & Beaulieu Pelletier et al., 2011). Researchers who study resilience have identified four potential responses to adversity, one of them being a "resilie nt" response (Carver, 1998; Tedeschi & Calhoun, 1996). The model below demonstrates that a person's response to adversity over time can vary by their level of functioning (O'Leary & Ickovics, 1995). For example, initially following an adverse event, the in dividual is likely to experience a decrease in functioning (either physical, psychological or both). Following this downturn, four outcomes can occur over time. The first is a continued downward slide where the level of functioning decreases to a point of no return and the individual "succumbs" to the situation. A second outcome could be that the person survives but is impaired and not able to return the homeostasis that existed prior to the adverse event. A third possible outcome is the return to the prior level of functioning that the person regards as "normal". This outcome is termed "resilience" and also indicates an optimum level of recovery. Finally, the most positive outcome is thriving, where following adversity the person functions at a higher level than
12 before the adversity occurred (Carver, 1998). Figure 1.1 Resilience by level of functioning and time Factors that lead to resilience can be consistent and general across adverse situations but can also be situation specific (Werner, 1995). Resili ence generally results from a person's ability to find and utilize the various internal and external resources available to them in adverse situations, including social support, financial options, and emotional and physical strengths that will help them ma intain a sense of well being (Carver, 1998). Resilience in those faced with life challenges has been linked with optimism as well as posttraumatic growth (Snyder, 2000; Carver, 2005). The research in this area suggests that not all survivors of childhood cancer report high levels of resilience following treatment, and ongoing physical and psychosocial worries and difficulties
13 continue for many survivors, even those who display some qualities of thriving (Kazak, 1994; Stuber et al., 1996; Weigers et al., 1 998; Zebrack & Chesler, 2001). Although previous empirical research has examined quality of life, health status and late effects, psychosocial functioning and posttraumatic growth in adult survivors of childhood cancer, to our knowledge there has been no r esearch documenting the association between these factors and the process of resilience in this population. Those who are more resilient may be better equipped to handle the future health challenges and/or late effects of treatment that are often observed in childhood cancer survivors. Higher levels of resilience in adults who have experienced cancer as a child may be related to better long term adjustment to future medical and/or psychosocial challenges associated with past cancer treatments. Since there is limited research simultaneously investigating multiple realms of functioning in adult survivors of childhood cancer, the current study propose d analysis of factors of psychosocial functioning (i.e., social support, optim ism, resilience, emotional functioning), posttraumatic growth, individual survivor characteristics, and long term adjustment (i.e., quality of life, health status and self care activities). As researchers learn more about this growing population, it has be come evident that identifying variables that predict positive adjustment throughout the adult lives of these childhood cancer survivors is important for the development of intervention programs aimed at promoting resilience. In order to provide programs th at accurately address the needs of this population, a broader understanding of the adult survivors of childhood cancer is necessary.
14 Aims and Hypotheses Aim 1 : Assess quality of life in adult survivor s of childhood cancer using the Impact of Cancer Scale Cancer Survivor version A cluster analysis technique was utilized to examine any potential underlying classification structure that exists within this area A better understanding of these clusters will help solidify future hypotheses regarding t he relationship b etween these variables and improved long term adjustment Aim 2: Examine the relationship between psychosocial functioning (e.g. distress, social support, optimism and resilience), participant characteristics and posttraumatic growth Hyp othesis 1: Psychosocial functioning will be significantly associated with parti cipant characteristics and post traumatic growth such that: (a) Distress in participants will be negatively associated with post traumatic growth. (b) Social support optimism and resilience will be positively associated with post traumatic growth (c) Time since diagnosis will be positively associated with posttraumatic growth
15 Aim 3 : Examine the association between participant characteristics (e.g. demographic and medical variables) and long term adjustment We conceptualize long term adjustment as a broad representation of multiple doma ins of current quality of life, current health status, self care activities and life satisfaction. Therefore, long term adjustment was investigated using measures specific to these areas. Hypothesis 2: Survivor characteristics such as time since diagnosis cancer type, and treatment regimen will be significantly associated with long term adjustment such that: (a) Time since diagnosis will b e positively associated with long term adjustment (b) Multiple treatment regimens ( i.e. chemotherapy and radiation) will be positively associated with a greater number of late effects and negatively associated with long term adjustmen t Aim 4 : Examine the association between the predicting variables of psychosocial functioning (e.g. distress, social support, optimism and resilience ) and posttraumatic growth with long term adjustment Hypothesis 3: Psychosocial functioning will be significantly associated w ith long term adjustment such that: (a) Emotional distress will be negatively correlated with long term adjustment (b) The level of perceived social support will be positively associated with long term adjustment.
16 (c) O ptimism will be positively associated with long term adjustment. (d) R esilience will be positively associated with l ong term adjustment. Aim 5: Examine the relationship between the number of medical conditions and/or late effects and psychosocial functioning. Hypothesis 4 (Exploratory): T he number of reported medical conditions will be associated with psychosocial functioning in adult survivors of childhood cancer. (a) The number of medical conditions will be negatively associated with psychosocial functioning (b) P osttraumatic growth mediates the relationship between the number of medical conditions and psychosocial functioning Figure 1.2 Proposed model of factors leading to long term adjustment "#$% & '()*! +,-./0*($0 1.23405!#6!"46( 7(2308!9(320(,!1#" :(36!;2)(!+<04=404(/ "46(!:204/62<04#$ >/5<8#/#<423! ?.$<04#$4$% @*#04#$23 ?.$<04#$4$% :#<423!:.AA#)0 BA04*4/* 9(/434($<( :.)=4=#)!;82)2<0()4/04 C(,4<23! <#$,404#$/D"20(! (66(<0/ +%(!20!,42%$#/4/ ;2$<()!05A( ')(20*($0!9(%4*($ ! >#/00)2.*204
17 CHAPTER II METHOD Procedure Participants in this study were recruited from the roster of patients who attended the TACTIC clinic over the last four years. The TACTIC clinic is designed to aid adult survivors of childhood cancer in the transition to survivorship and adult primary care. Each patien t meets with a pediatric oncologist, internal medicine physician, nurse practitioner and psychologist during their visit. Patients were asked to complete a questionnaire about their treatment history, their health behaviors and their current health and psy chological functioning prior to attending the clinic and a follow up survey inquiring about health related changes the patients made as a result of their visit. The goals of this clinic is to identify the psychosocial, educational, and health care needs of this unique population, as well as to aid the patient in transitioning from using their pediatric oncologist as their primary health care provider to a adult primary care Participants were recruited for this study as a part of a larger ongoing research p rogram associated with the TACTIC clinic. The potential participants were contacted via electronic correspondence a maximum of three times and invited to participate in an online survey through REDCap (Research Electronic Data Capture) a secure server Pa rticipants were informed prior to completing the survey that their participation in the study was optional and no compensation would be offered. Participants were eligible for the study if they have previously been seen in the TACTIC clinic. Eligibility cr iteria for enrollment into the TACTIC study include the following: 1) Participant must have history of one or more cancer diagnoses that occurred prior to age 18 2) age 18 or above; 3) must be able to read
18 and understand English as the questionnaires were only provided in English. This research was reviewed and approved by the Colorado Multiple Institution Review Board ( 05 0597 ). Access to the questionnaire was provided to the participants with a hyperlink to their email address. The questionnaires were posted on a HIPPA and COMIRB consistent data c ollection website called REDCap. Consent was conducted at the start of the online questionnaire with the consent form provided to patients electronically and submission of the questionnaire was equivalent to co nsent. Study participants were informed that their choice to participate in this study in no way affects their treatment in the TACTIC clinic, and that they were free to terminate their participation at any point. Contact information for the primary invest igator was provided at the end of the consent form if the participant had questions about the study. Once the participant completed the questionnaire, their data was encrypted and stored in a secure database that had restricted access to the primary invest igator. Data from participants was stored according to a unique identification number in the REDCap database. The questionnaire was sent to 60 potential participants between the dates of 10/20/2014 to 12/30/14 Those who did not compl ete the first survey received two other reminder e mails. Nineteen completed the survey following the first e mail while 2 and 3 completed it following the reminder e mails. All participants who accessed the survey through their personalized hyperlink in the e mail completed the survey.
19 Measures Impact of Cancer Scale Quality of life was measured using The Impact of Cancer Scale Cancer Survivor version (IOC CS) The IOC CS is an instrument developed specifically for adult survivors of childhood cancer and is a comprehensive, multidimensional scale that investigates general domains of QoL, HRQoL, health behaviors, body image, interaction with the medical team and concerns about insurance, fertility concerns, identity, the meaning of cancer in their life (in cluding posttraumatic growth), memory, financial concerns, interpersonal and family relationships, and life goals (Zebrack et al., 2010). The IOC CS is an ideal measure for evaluating quality of life in this population, since it's development was based on qualitative interviews of adult survivors of childhood cancer and therefore contains questions about concerns that are specific to these survivors. Resiliency Resilience in participants was evaluated using the Brief Resilience Scale (Smith et al., 2008) The Brief Resilience Scale is a short 6 item self report measure that inquires about the participant's perceptions of their ability to "bounce back" following a stressful event. Items such as "It does not take me long to recover from a stressful event" ar e reported from 1 (strongly disagree) to 5 (strongly agree). Posttraumatic Growth P osttraumatic growth was evaluated using the Posttraumatic Growt h Inventory (PTGI). The PTGI is composed of 21 items with standardized response choices from 0 (I did not exp erience this change as a result of my crisis) to 5 (I experienced this change to a very great degree as a result of my crisis). Created by Tedeschi and Calhoun (1996), the
20 PTGI contains five subscales, including: Relating to Others, New Possibilities, Pers onal Strength, Spiritual Change, and Appreciation for Life. Optimism Optimism was measured using the Life Orientation Test Revised (LOT R). The LOT R was developed to assess individual differences in optimism (Scheier, Carver & Bridges 1994). The LOT R i s a brief 6 item inventory with standardized response choices from 1 (I agree a lot) to 4 (I disagree a lot) to questions such as, "In uncertain times, I usually expect the best." Perceived Social Support Current perceived social support was evaluated us ing the Interpersonal Support Evaluation List 12 (ISEL 12) (Cohen & Hoberman, 1983). The original ISEL was composed of 40 items, however the authors found good reliability and validity scores for a shorter 12 item version with three subscales which include : (1) Appraisal, or the availability of others to offer support, (2) Tangible, which is material or instrumental support, and (3) Belonging, or the perceived availability of others for interpersonal relationships. The answer options range from "definitely false" to "definitely true". Evaluating the quality of the participants' social support is an important part of identifying factors related to resilience. Emotional Health The Impact of Events Scale (IES R) was used to assess subjective distress brought on by the participant's cancer experience (Weiss & Marmar, 1997). The IES R asks respondents to identify a specific stressful life event and then rate how much distress they experienced over the past week from a list of difficulties on a scale from 0 ("n ot at all")
21 to 4 ("extremely"). Current emotional functioning was assessed using one added question that inquires about the participant's emotional health and four additional questions asking the survivor to rate their level of fatigue, pain, anxiety and depression in the past two weeks Fatigue, anxiety and depression questions were taken from the Profile of Mood States Linear Analogue Self Assessment (POMS LASA; Sutherland, Lockwood, & Cunningham, 1989). The four th dimension, pain, was added due to the high frequency of pain in oncology patients. These four questions are rated on a 0 to 10 scale, with zero representing no experience of the construct and ten being the "most possible". Measures of Long term Adjustm ent Quality of Life and Health Status General quality of life, and current health status were measured u sing the IOC CS (Zebrack et al., 2010) and one additional question that inquir ed about the participant's current physical health. Survivors were asked about new cancer diagnoses since their TACTIC visit and late effects of cancer treatment using questions from the original TACTIC questionnaire. Life Satisfaction Participants' level of life satisfaction was investigated using the Satisfaction With Life S cale ( Diener, Emmons, Larsen & Griffin, 1985 ). This measure is composed of 5 items designed to measure global judgments of satisfaction with one's life. Self Care Activities and Health Behaviors Self care activities measured include d diet (e.g. servings o f fruits and vegetables and diet quality) and exercise (e.g. type, times per week and for how long). Health behaviors such as alcohol consumption, cigarette use, and sunscreen routine were also
22 measured. Questions that address self care and health behavior s from the original TACTIC questionnaire were re administered to participants Demographic and Medical Variables This study also assess ed a variety of demographic an d medical variables. These included information about the survivor's treatment history (i.e age at diagnosis, type of cancer, and treatment regimen) that was gathered at the time of the initial TACTIC Clinic visit. Other demographic and medical information were gathered including age, gender, level of completed educatio n and marital/relationshi p status. This is presented in Table 2.1 Participants Participants included 24 adult participants ( 15 females, 9 males) who attended the TACTIC (Thriving After Cancer Treatment is Complete) interdisciplinary clinic at the University of Colorado Hospital during the dates of September 2010 and May 2013. The most common type of childhood cancer diagnosis reported was leukemia (34%), followed by lymphoma (31%), and sarcoma (19%). One participant reported a childhood cancer diagnosis of a Wilms tumor and one reported being diagnosed with a brain tumor. Interestingly, the percentage of participants with childhood cancer diagnoses of leukemia is exactly the national base rate that American Cancer Society reports (ACS, 2014). However, the percentage of participants with childhood cancer diagnoses of central nervous system tumors are far below the national base rate of 27% (ACS, 2014) Also, the percentage of participants with a childhood cancer diagnosis of lymphoma is half again higher than the nati onal base rate of 16% (ACS, 2014 ). The participant sample was not representative of national base rates of race and ethnicity, as 100% (24) of the
23 participants identified as White. Participants ranged in age from 24 to 41, with an average age of 30. All of the participants had achieved a high school degree, and most had attended at least some college. The number of years of education the participants achieved ranged from 14 to 20 with an average of 17 years. Demographics for the sample are included in Table 2.1 Table 2.1 Sample Population Demographic Statistics Demographic Descriptive Statistics Variable N Range Minimum Maximum Mean Std. Deviation Age 24 17.00 24.00 41.00 30.1364 4.89213 Years of Education 24 6 14 20 16.96 1.781 Age at Diagnosis 24 19.5 .5 20.0 9.967 6.5967 Years Since Treatment 24 26 8 34 17.96 6.836 Number of Treatments 24 3 1 4 2.17 .816 Variable Frequency Percent Gender Female Male 15 9 62.5 37.5 Relationship Single In a relationship 15 9 62.5 37.5 Race Caucasian 24 100 Cancer Type Brain Leukemia Lymphoma Sarcoma Wilms 1 9 8 5 1 4.2 37.5 33.3 20.8 4.2
24 CHAPTER III RESULTS Power Analysis Statistical power is the likelihood that a statistical test will reject the null hypothesis when it is false (Miles & Shelvin 2001). There are three factors that affect power, including: 1) statistical significance (i.e. = 0.05 or 0. 01), 2) the size of the effect of a given construct in the study population, and 3) the sample size itself. Using GPOWER (Erdfelder, Faul, & Buc hner, 1996), the bivariate correlations with = 0.05, power= 0.80, would require a sample size of 21 to detect a large effect ( r = 0.50 ) and a sample size of 64 to detect a small effect ( r = 0.30 ) Therefore a p value of 0.05 would allow us to detect a large effect size in our sample of 24 participants. Data Analysis Quantitative data analyses procedures were utilized The data was examined for skewness kurtosis, normality, and outli ers prior to analyses. Of all the variables, 7 were identified as being non normal : Anxiety, Dep r ession, Distress, Emotional Health, Physical Health, Pain and Medical Conditions In all of these instances, the data was skewed as a result of floor and ceiling effects and all skewed variables were physica l and emotional health ratings. Because the data was not normally distributed, in the following analyses the Spearman's rho coefficient was used rather than the Pearson coefficient, as Spearman's rho is non parametric and does not depend on assumptions of normality (Hauke & Kossowski, 2011).
25 Means and Distributions for the Psychosocial and Medical Variables Overall, participants reported experiencing low levels of anxiety, depression, and distress as measured by the POMS LASA and the Impact of Events Scale (IES) (Su therland, Lockwood, & Cunningham, 1989) The Impact of Events scale includes two subscales: Avoidance and Intrusion. Both were highly skewed due to floor effects where participants reported low levels of distress. Similarly, participants reported experienc ing low levels of pain and very good to average physical health with a low number of additional medical conditions Please refer to table s 3.1 and 3.2 for a list of the means and standard deviations and published norms for medical and psychosocial va riables Table 3.1 Means for Medical Variables Means for Medical Variables Physical Health ( Added Question ) N=24 Pain ( Added Question ) N=24 Medical Conditions ( Added Question ) N=24 Current Sample Current Sample Current Sample Mean 2.54 2.71 1.38 Std. Deviation .932 2.896 1.416
26 Table 3.2 Mea ns and Comparison Means for Psychosocial Variables Means and Comparison Means for Psychosocial Variables Emotional Health ( Added Question ) ( N=24 ) Anxiet y ( POMS LASA) ( N=24 ) Depression ( POMS LASA) ( N=24 ) Distress (IES) ( N=24 ) Current Sample n/a Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Mean 2.58 3.46 7.1 1.58 7.5 25.95 33.26 Std. Deviation 1.100 3.189 5.8 2.165 9.2 10.04 18.98 The histograms below provide a visual representation of the frequencies of the ratings for all the variables that were found to have significant floor and ceiling effects. These medical and psychosocial variables were included in the survey to evaluate the perceived emotional and physical health ratings of the participants. Our findings are consistent with previous research on cancer survivors in that they tend to report the same or better levels of emotional functioning as age matched peers (Boman & Bodega rd, 1995; Elkin et al., 1997; Gray et al., 1992; Kazak, 1994; Kazak et al., 1997; Kupst et al., 1995; Mackie et al., 2000; Madan Swain et al., 1994; Radcliffe, Bennett, Kazak, Foley, & Phillips, 1996; Simms, Kazak, Golomb, Goldwein, & Bunin, 2002) Also, 25% of participants reported moderate to severe pain, which are similar frequencies of pain ratings compared to the ongoing Childhood Cancer Survivor Study (Hudson, Mertens, Yasui et al., 2003; Tsao, Leisenring, Robinson, Zeltzer, 2007). Please ref er to histograms 3.1 to 3.7 to see the current sample's frequencies for each psychosocial and medical variable
27 Figure 3.1 Histogram of frequencies of anxiety ratings Figure 3.2 Histogram of the frequencies of depression ratings Figure 3.3 Histogr am s of the frequencies of the total Impact of Events Scale (distress) ratings
28 Figure 3.3 cont. Histograms of the frequencies of Intrusion and Avoidance subscale for the Impact of Events Scale (distress) ratings Figure 3.4 Histogram of the frequencies of overall emotional health ratings Figure 3.5 Histogram of the frequencies of physical health ratings
29 Figure 3.6 Histogram of the frequencies of number of reported medical conditions Figure 3.7 Histogram of the frequencies of pain ratings Assessing Quality of Life The first aim of this study was to a ssess quality of life in adult survivors of childhood cancer using the Impact of Cancer Scale Cancer Survivor version Initially a hierarchical cluster analysis technique was proposed in order to identify groups of participants with similar response patterns on the IOC CS. However, the smaller sample size prevent ed the use of cluster analysis with a high number of variables, such as each variable in the IOC CS (82), as the dimensionality of the high number of variables is too high for the number of case s to be grouped. Formann (1984) suggests the minimal sample size to include no less than 2 k cases (k = number of variables). Indeed, SPSS would not perform a cluster analysis with 24 cases.
30 In 2010, after the initial development of the IOC CS, Zebrack and his colle a gues used a cluster analysis to identify subscales of the quality of life measure. In total, they identified 10 subscales, 5 reflecting positive realms of quality of life, and 5 "negative" subscales reflecting aspects of the cancer experience that led to poorer quality of life. The 5 positive subscales are: Body & Health, Talking with Parents, Personal Growth, Health Literacy and Socializing The 5 negative subscales are: Life Challenges, Thinking & Memory Problems, Financial Problems, Sibling Concerns and Relationship Concerns As an alternative to performing a new cluster analysis, we used the previously identifi ed subscales and analyzed interc orrelations among the subscales as well as relationships between the IOC CS subscales and the other measures used in the study. The means and standard deviations for our population on the various subscales, as compared to the published norms are presented in Table 3.3
31 Table 3.3 IOC CS Subscale Means and Published Norms (Zebrack et. al, 2010) IOC CS Subscale Means Life Challenges ( N =24) Body & Health (N=24) Talking with Parents (N=24) Personal Growth (N=24) Thinking & Memory (N=24) Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Mean Std. Deviation 1.97 .627 2.28 0.81 3.59 0.76 3.44 0.54 4.54 0.86 3.89 1.11 4.08 0.69 3.45 0.92 2.30 0.79 2.30 0.89 Health Literacy (N=24) Socializing (N=24) Financial Problems (N=24) Sibling Concerns (N=24) Relationship Concerns (N=24) Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Current Sample Pub. Norms Mean 3.87 3.58 3.86 3.97 1.52 1.64 2.39 1.90 2.69 2.26 Std. Deviation .615 0.89 .827 0.85 .589 0.93 1.334 1.00 1.440 1.02 Previous research indicates that psychological, social, physical and spiritual realms of quality of life are interrelated (Zebrack, 2000; Wyatt & Friedman, 1996). If this were true for our population, we would expect to see associations between related subscales of the IOC CS. As hypothesized, the Life Challenges subscale was significantly correlated with the Thinking & Memory Problems ( rs  = .4 55, p < .05) and Socializing ( rs  = .392, p < .05) subscales. Higher levels of life challenges was significantly associated with thinking and memory challenges as well lower levels of socializing Talking with Parents was significantly negatively as sociated with Financial Problems s uch that as participants reported more financial problems, they reported poorer quality of communication with their parents ( rs  = .447, p < .05).
32 Significant correlations were also observed between this study's measure of life satisfaction (as measured by the Satisfaction With Life Scale ) and 6 of the 11 IOC subscales. Life satisfaction was lower in relation to reporting greater Life Challenges ( rs  = .523, p < .01). Life satisfaction was also h igher in childhood cancer survivors who reported more positive outcomes in Body & Health ( rs  = .517, p < .01) and Socializing ( rs  = .468, p < .05). The correlational findings are reported in Table 3.4
33 Table 3.4 IOC CS Subscale Correlations and Correlations with Satisfaction With Life Scale IOC CS Subscale Correlations and Correlations with Satisfaction With Life Scale Life Challenges Thinking & Memory Problems Socializing Talking with Parents Financial Problems Satisfaction with Life Scale Life Challenges Correlation Coefficient 1.000 .455 .392 .181 .180 .523 ** Sig. (2 tailed) .013 .029 .199 .200 .004 N 24 24 24 24 24 24 Thinking & Memory Correlation Coefficient .455 1.000 .191 .184 .062 .326 Sig. (2 tailed) .013 .185 .195 .387 .060 N 24 24 24 24 24 24 Socializing Correlation Coefficient .392 .191 1.000 .018 .133 .468 Sig. (2 tailed) .029 .185 .466 .268 .011 N 24 24 24 24 24 24 Talking with Parents Correlation Coefficient .181 .184 .018 1.000 .447 .321 Sig. (2 tailed) .199 .195 .466 .014 .063 N 24 24 24 24 24 24 Financial Problems Correlation Coefficient .180 .062 .133 .447 1.000 .046 Sig. (2 tailed) .200 .387 .268 .014 .416 N 24 24 24 24 24 24 Satisfaction with Life Scale Correlation Coefficient .523 ** .326 .468 .321 .046 1.000 Sig. (2 tailed) .004 .060 .011 .063 .416 N 24 24 24 24 24 24 *. Correlation is significant at the 0.01 level (2 tailed). **. Correlation is significant at the 0.01 level (2 tailed).
34 The Association between Psychosocial Functioning, Resilience and Post traumatic Growth The second aim of the study was to e xamine the relationship between psychosocial functioning (e.g. distress (IES), social support (ISEL), optimism (LOT) and resilience (BRS), participant characteristics and posttraumatic growth (PTGI). We hypothesized that p sychosocial functioning w ould be significantly associated with participant characteristics and posttraumatic growth such that d istress would be negatively associated with posttraumatic growth and s ocial support, optimism and resilience would be positive ly associated with post traumatic growth We found no significant relationship found between posttraumatic growth and social support or distress indic ating that even as distress and levels of social support change, posttraumatic growth does not significantly change as a result. Posttrauma tic growth was moderately positively correlated with optimism ( rs  = .434, p < .05), and positively associated with resilience ( rs  = .384, p < .05 ). Squaring the correlation coefficients provides an estimate of variance in the relationship between t he variables. Squaring the correlation coefficient indicated that 18.8 % of the variance in posttraumatic growth was explained by the presence of optimism in the participants Similarly, 14.7% of the variance in posttraumatic growth was explained by partici pants' level of resilience. Optimism and resilience were strongly positively correlated ( rs  = .758, p < .001) and squaring the correlation coefficients indicated that 57.4% of the variance in resilience was explained by the level of optimism participants reported (Hypothesis 1a) Both optimism and posttraumatic growth were moderately negatively correlated with the
35 number of years since the participant had received treatment ( rs  = 414, p < .05) ( rs  = .400, p < .05) (Hypothesis 1b). W e also hypothesized that t ime since diagnosis will be positively associated with posttraumatic growth. When looking at how the number of late effects changes as time since treatment increases, there was s a significant positive correlation between the pass age of time and the number of medical conditions the participants reported ( rs  = .471, p < .05). However, there was no significant correlation between the number of cancer related late effects and posttraumatic growth or optimism. The Association Betw een Participant Characteristics and Long Term Adjustment The third aim of the study was to examine the association between participant characteristics such as demographic and medical variables, and long term adjustment. We hypothesized that survivor characteristics such as time since diagnosis cancer type, and treatment regimen would be significantly associated with long term adjustment such that t ime since diagnosis would be positively assoc iated with long term adjustment, and m ultiple treatment regimens (i.e. chemotherapy surgery and radiation) w ould be positively associated with a greater number of late effects and negatively associated with long term adjustment In order to examine the association between the individual predicting variables of psychosocial functioning, participant characteristics, posttraumatic growth and long term adjustment, independent variables were separately correlated to each other using Spearman's rho correlations. Because we conceptualize long term adjustment as a comb ination of multiple factors, we create d a composite scale made up of the standardized scores of each variable included in long term adjustment. Prior to this, a
36 correlation matrix was computed to examine the strength of relationships between the variables, and then internal consistency was measured amongst the variables using Chronbach's alpha (see Table s 3.5, 3.6 and 3.7 ). Significant correlations were found between QOL, physical health and satisfaction with life, with physical health having the strongest correlation with QOL. This also made sense from a theoretical standpoint and was consistent with findings in the literature ( Zebrack, 2000; Wyatt & Friedman, 1996 ). Diet and exercise were moderately correlated with physical health, but no other variable. S unscreen was not significantly correlated with any other variable. Chronbach's alpha for the long term adjustment scale was calculated using all 6 variables (! = .559). Due to diet and exercise being only moderately correlated with one other variable, and the use of sunscreen being unrelated to other variables, Chronbach's alpha analyses indicated that the value would increase without the inclusion of these three variables, thereby creating a more internally consistent scale. This also made sense form a the oretical standpoint since the questions about diet, exercise and sunscreen are health behaviors that are conceptually different from quality of life, satisfaction with life and physical health. A second test of internal consiste ncy was performed without t hese which resulted in an acceptable Chronbach's alpha (! = .764) for the composite scale which included the variables of quality of life, satisfaction with life and physical health.
37 Table 3.5 Long Term Adjustment Variable Correlations Long Term Adjustment Variable Correlations Quality of Life Satisfaction w/ Life Exercise Sunscreen Physical Health Diet Spearman's rho Quality of Life Correlation Coefficient 1.000 .390 .134 .220 .592 ** .050 Sig. (2 tailed) .030 .270 .151 .001 .408 N 24 24 23 24 24 24 Satisfaction w/ Life Correlation Coefficient .390 1.000 .002 .025 .495 ** .188 Sig. (2 tailed) .030 .496 .455 .007 .190 N 24 24 23 24 24 24 Exercise Correlation Coefficient .134 .002 1.000 .068 .445 .265 Sig. (2 tailed) .270 .496 .378 .017 .110 N 23 23 23 23 23 23 Sunscreen Correlation Coefficient .220 .025 .068 1.000 .092 .177 Sig. (2 tailed) .151 .455 .378 .335 .205 N 24 24 23 24 24 24 Physical Health Correlation Coefficient .592 ** .495 ** .445 .092 1.000 .448 Sig. (2 tailed) .001 .007 .017 .335 .014 N 24 24 23 24 24 24 Diet Correlation Coefficient .050 .188 .265 .177 .448 1.00 0 Sig. (2 tailed) .408 .190 .110 .205 .014 N 24 24 23 24 24 24 *. Correlation is significant at the 0.05 level (2 tailed).
38 Table 3.6 Inter Item Reliability for the Long Term Adjustment Scale Inter Item Reliability for Long Term Adjustment Scale Scale Mean if Item Deleted Scale Variance if Item Deleted Corrected Item Total Correlation Squared Multiple Correlation Cronbach's Alpha if Item Deleted Diet 49.8442 49.193 .133 .263 .446 Quality of Life 32.2609 33.747 .604 .490 .198 Satisfaction w/Life 33.6703 13.477 .364 .398 .524 Physical Health 48.8007 42.001 .649 .564 .322 Sunscreen 48.6703 50.676 .008 .118 .465 Exercise 48.1486 46.466 .084 .343 .456 Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .544 .764 3 Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .443 .559 6 Table 3.7 Inter Item Reliability for the Long Term Adjustment Scale with Higher Alpha Inter Item Reliability For Long Term Adjustment Scale with Higher Alpha Scale Mean if Item Deleted Scale Variance if Item Deleted Corrected Item Total Correlation Squared Multiple Correlation Cronbach's Alpha if Item Deleted Quality of Life 22.0417 27.955 .593 .444 .277 Satisfaction w/ Life 23.4340 6.492 .556 .309 .636 Physical Health 38.5590 36.533 .528 .365 .535
39 Following the determination of the variables that would lead to the most internally consistent scale of long term adjustment, those variables were standardized in order to prevent one variable from carrying more weight in analyses. This was done by subtracting the sample mean and dividing b y the sample standard deviation. The variables were then summed to create the long term adjustment scale (Bibinger, 2013). In order to determine if long term adjustment in participants was related to the amount of time that had passed since they received t reatment (Hypothesis 2a ), a Spearman's rho correlation was used. No significant relationship was found between the amount of time that had passed since participants' diagnosis, and their overall adjustment. It was hypothesized (Hypothesis 2b) that multi ple treatment regimens would be positively associated with a higher number of late effects, and negatively associated with long term adjustment. As expected, a significant positive relationship was found between multiple treatment regimens (i.e. having com binations of chemotherapy, radiation and surgery), and the number of self reported medical conditions ( rs  = .471, p < .05). There was, however, no significant relationship discovered between a participant having multiple treatments the number of late effects, and long term adjustment, indicating that multiple treatment regimens and the number of late effects may not be related to long term adjustment. These findings are reported in Table 3.8
40 Table 3.8 Long Term Adjustment Intercorrelations Long Term Adjustment Interc orrelations Years Since Treatment Long Term Adjustment Number of Medical Conditions Multiple Treatments Spearman's rho Years Since Treatment Correlation Coefficient 1.000 .241 .471 .355 Sig. (2 tailed) .257 .020 .089 N 24 24 24 24 Long Term Adjustment Correlation Coefficient .241 1.000 .275 .056 Sig. (2 tailed) .257 .194 .795 N 24 24 24 24 Number of Medical Conditions Correlation Coefficient .471 .275 1.000 .462 Sig. (2 tailed) .020 .194 .023 N 24 24 26 24 Multiple Treatments Correlation Coefficient .355 .056 .462 1.000 Sig. (2 tailed) .089 .795 .023 N 24 24 24 24 *. Correlation is significant at the 0.05 level (2 tailed). The Association Between Psychosocial Functioning, Posttraumatic Growth and Long Term Adjustment The fourth aim of the study was to examine the association between psychosocial functioning (e.g. distress, social support, optimism and resilience), posttraumatic growth and long term adjustment We hypothesiz ed that psychosocial functioning would be significantly associated with long term adjustment such that emotional distress would be negatively correlated with long term adjustment, and the level of perceived social support would be positively associated wit h long term adjustment. We also hypothesized
41 that optimism, resilience, and posttraumatic growth would be positively associated with long term adjustment. Spearman's rho correlations were used to examine the relationship between variables of psychosocial functioning, posttraumatic growth and long term adjustment. As expected (Hypothesis 3a), levels of emotional distress were negatively correlated with long term adjustment ( rs  = .471, p < .05), indicating that lower distress levels are related to highe r levels of long term adjustment. Additionally, social support, optimism, resilience and posttraumatic growth were all significantly positively associated with long term adjustment ( rs  = .548, p < .01) ( rs  = .649, p < .001) ( rs  = .642, p < .001 ) ( rs  = .504, p < .05) ( Hypothesis 3b e). The strongest of these correlations existed between long term adjustment and optimism. Squaring the correlation coefficient indicated that 42% of the variance in long term adjustment was explained by the level of optimism participants reported Squaring the correlation coefficient indicated that resilience accounts for 41% of the variance in long term adjustment. In summary, lower levels of distress and higher levels of social support, optimism, resilience and p osttraumatic growth are all strongly associated with adult survivors of childhood cancer's long term adjustment. The Relationship Between Late Effects and Psychosocial Functioning The fifth aim of the study was to examine relationship between the number of late effects and psychosocial functioning (e.g. distress, social support, optimism and resilience ). We hypothesized that the number of reported late effects would be negatively associated with psychosocial functioning and that posttraumatic growth would mediate the relationship between the number of late effects and psychosocial functioning
42 Spearman's rho correlations were used to examine the relationship between variables of psychosocial functioning and the number of late effects. No significant associations were found between the number of late effects and distress, optimism or resilience (Hypothesis 4a). However, a negative relationship was found between the number of late effects and social support so that as the number of late effects of cance r increased, their social support decreased ( rs  = .368, p < .05). The second part of the hypothesis examined the relationship between the number of late effects and posttraumatic growth. There was also no significant relationship found between the number of late effects and posttraumatic growth. A mediation model was not done due to the lack of correlation among these variables. Compari son of Psychosocial Variables, Cancer Type and A ge at D iagnosis In order to determine if differences existed betwe en psychosocial variables among participants diagnosed with certain types of cancer, participants were grouped into two groups, either hematological (blood cancers), or tumor based cancers Independent t tests were used to compare participants who had been diagnosed with tumors or hematological cancers, and differences were found between the groups on their reported levels of distress, depression rate, and whether or not they were currently smoking. Those with hematological cancers reported significantly le ss distress compared to those with tumor based cancers (M = 23.36, SD = 5.62; M = 30.44, SD = 14.09) ( t (24) = 1.77, p < .01). Similarly, those with hematological cancers reported significantly lower levels of depression compared to those with tumor based cancers (M = 1.20, SD = 1.57; M = 2.22, SD = 2.9) ( t (24) = 1.13, p < .05). Finally, those who had been diagnosed with hematological cancers also tended to report that they had smoked in the past 7 days,
43 compared with those who had been diagnosed with tumo r based cancers (M = .13, SD = .352; M = .00, SD = .00) ( t (24) = 1.13, p < .05). No differences were found between males ve rsus females on any variable. These findings are reported in Table 3.8 Table 3 .8 T Test Comparisons Between Cancer Types Hem a tological Cancers Mean (SD) Tumor based cancers Mean (SD) t Significance Mean Difference Distress 23.26 (5.6) 30.44 (14) 1.771 .008 7.17 Depression 1.20 (1.5) 2.22 (2.9) 1.126 .045 1.02 Smoking .13 (.35) 0 (0) 1.127 .014 .133 We were also interested in the impact of age on diagnosis on psychosocial adjustment. Age at diagnosis was found to be significantly positively associated with resilience, so that as the age at which a participant was diagnosed increased, so did their reported level of resilience ( rs  = .400, p < .05). Age at diagnosis was also found to be significantly positively associated with increased body and health quality of life on the IOC CS, so that as the age at which a participant was diagnosed increased, so did their reported body image and perceptions of current health ( rs  = .435, p < 05). These findings are reported in table 3.9. T Test Comparisons Between Cancer Types
44 Table 3.9 Age at Diagnosis Intercorrelations Age at Diagnosis Intercorrelations Age at Diagnosis Body & Health Resilience Spearman's rho Age at Diagnosis Correlation Coefficient 1.000 .435 .400 Sig. (2 tailed) .017 .026 N 24 24 24 Body & Health Correlation Coefficient .435 1.000 .340 Sig. (2 tailed) .017 .052 N 24 24 24 Resilience Correlation Coefficient .400 .340 1.000 Sig. (2 tailed) .026 .052 N 24 24 24 *. Correlation is significant at the 0.05 level (2 tailed).
45 CHAPTER IV DISCUSSION This study examined the relationship between psychosocial functioning, posttraumatic gro wth and long term adjustment in a group of adult survivors of childhood cancer. It also explored intercorrelations between scales of the widely used quali ty of life measure, the IOC CS, as well as overall psychosocial functioning of adult survivors of childhood cancer. Comparing Differences Between Cancer Types and Age at Diagnosis In addition to evaluating long term adjustment, psychosocial functioning and posttraumatic growth, we investigated potential differences between participants diagnosed with different types of cancer and the variables des cribed above Our findings suggest that participants diagnosed with hematological cancers experience less distress and depression on average compared to those diagnosed with tumor based cancers. Zabora (2001) found similar results when using the Brief Symp tom Inventory to measure distress between patients with a history of different types of cancer, and demonstrated that those with hematological cancers had less distress than those with tumor based cancers. No significant differences in the number of late e ffects of cancer as measured by self report of late effects, were observed between the groups The average number of late effects for both types of cancer in this population was less than 2. This was somewhat surprising given that the treatments are typic ally quite different for these two populations Given that the mean time since diagnosis was 18 years most of the participants have reached the point in time when they are likely to start experiencing secondary effects of treatment and previous research on late effects indicate that the number of late effects in
46 childhood cancer survivors increases with longer time since treatment and with older age ( Robinson et al., 2010 ). Interestingly more participants diagnosed with hematologica l cancers reported having smoked in the past 7 days compared to those who had been diagnosed with tumor based cancers. However, when looking at the number of participants who reported smoking, only 2 of 24 reported they were currently smoking and both had a h istory of hematological cancers, which explains these mean differences. These results point to potential differences in the longer term experiences of survivors diagnosed with different types of cancer Differences in the age at which the participants were diagnosed were also associated with a number of factors. The older a participant was when they were diagnosed was related to increased resilience as measured by the Brief Resilience Scale (Smith et al., 2008) Those who were diagnosed at an older age may have been better able to understand the seriousness of their diagnosis and subsequent treatment, and therefore were better prepared to conceptualize the impact that cancer had o n their li ves It is possible that survivors diagnosed at an older age may have had increased access to support services and various programs aimed at older children. A positive correlation existed between age at diagnosis and body image, so that participants diagnosed at an older age also tended to report a more positive body image as measured by the IOC CS Body & Health scale. This may be because treatment was less damaging to those diagnosed at an older age because their bodies were more developed Kopel (1998) and her colleagues researc hed body image changes over time in survivors of childhood cancer. They found that there was no significant correlation between age at diagnosis or time since diagnosis and body image. Scarring, disfigurement and permanent hair loss
47 experienced as a result of treatments such as radiation and surgery, can often affect body image. These treatments are more frequently used for patients with tumor based cancer and previous research indicates that scarring, disfigurement and permanent hair loss have been associa ted with increased depression in survivors (Kinehan et al., 2012). This may be a contributing factor to the higher levels of distress and depression levels in survivors with a history of tumor based cancers in our population. Psycho logical Functioning and Health Related Quality of Life Previous research in the area of psychological functioning on adult survivors of childhood cancer indicates that they often display better psychological functioning than general population base rates (Elkin et al., 1997 ; Langeveld et al., 2002) Our sample was no exception. Eighty percent of our participants reported low levels of depression, and 60% reported low levels of anxiety. Overall ratings of emotional distress specific to their cancer diagnosis and treatment wer e similarly low, as 80% of participants reported average to low level of distress compared to published norms on the Impact of Events Scale Results from the ongoing Childhood Cancer Survivor Study indicate that 10 23% of adult survivors of childhood cancer report moderate to extreme levels of chronic pain (Hudson, Mertens, Yasui et al., 2003; Tsao, Leisenring, Robinson, Zeltzer, 2007). This was also true for our sample. Twenty five percent of participants reported experiencing moderate to severe pain. Previous research also indicates that 69% of adult survivors of childhood cancer experienced at least one late effect, and 39% reported t wo or more. Our sample reported comparatively lower rates of late effects as 37% of our participants rep orted having no late effects, 30 % reported having only one, and 33% reported having
48 two or more late effects. Those participants report ing more late eff ects, also tended to report lower levels of social support. This could be due to increased physiological impairment as a result of multiple medical conditions, which could in turn prevent them from having the ability to actively participate in social activ ities or seek social support. Research with other medical populations has shown that a low level of social support is associated with poor medical outcomes (Lett et al., 2006). Impact of Cancer Scale Evaluation of the relationships between the subscales of the IOC CS revealed some interesting findings. First, those participants who reported more life challenges, such as worrying about mortality, their future, and negative thoughts about cancer, also had more problems with thinking and memory, su ch as challenges with decision making, concentration, and short term and long term memory problems. Long term neurocognitive deficits are a common side effect of chemotherapy and radiation (ACS, 2014 ; Elkin, Phipps, Mulhern & Fairclough, 1997), but the psy chological sequelae of having these neurocognitive deficits as adult survivors of childhood cancer have not been widely researched. Clearly, neurocognitive deficits could create unique challenges in adulthood, particularly when faced with attaining employm ent, managing finances, and navigating relationships. Life satisfaction was found to be lower in adult survivors of childhood cancer who reported high er life challenges, poorer HRQol poorer body image as a result of their cancer, and inadequate social sup port. One unique finding in our sample was the relationship between survivors' quality of communication with their parents and having financial problems as a result of cancer treatment. P articipants who reported that they and their parents had experienced financial
49 challenges as a result of cancer treatment, also tended to report talking less about their cancer experience with their parents, as measured by the IOC CS. Posttraumatic Growth and Psychosocial Functioning Posttraumatic growth is a positive psychological change that involves shifting the perception of the trauma from a threat into a challenge (Calhoun & Tedeschi, 1999). Many adult cancer survivors experience positive changes, including re prioritizing their lives, integrating positive meanings of cancer, and increased perceived benefits as a result of their illness (Thornton, 2002 Parry & Chesler, 2005; Thornton, 2002; Adrykowski, Brady & Hunt, 1993; Frtiz & Williams, 1988; Kupst et al., 1995). Optimism has b een found to increase posttraumatic growth, and is also highly related to the development of resilience following a traumatic experience (Carver, 2005; Snyder, 2000; Scheier, Weinraub, & Carver, 1986). This was also found to be true in our sample. Posttraumatic growth was strongly associated with levels of optimism and resilience, so that those who had higher levels of optimism and resilience also reported experiencing more posttraumatic growth. Not surprisingly, o ptimism and resilience were also strongly correlated in our sample. Perceptions of posttraumatic growth did not appear to be related to the participants' current levels of distress or social support, which indicates that adult su rvivors of childhood cancer may experience the benefits of posttraumatic growth even when experiencing significant distress or inadequate social support. We predicted that self reported posttraumatic growth would be higher in the cancer survivors who wer e further out from their initial cancer diagnosis and treatment. However, we found that the opposite was true. P osttraumatic growth and optimism were negatively associated with time since diagnosis Although the cause behind this
50 interesting finding is bey ond the scope of this study it could be that as time passes, survivors experience more late effects of their cancer, which impacts their ability to find positive aspects of their cancer experience. Research on posttraumatic growth in other populations who experience trauma, such a war veterans and victims of natural disasters, indicates that posttraumatic growth increases over time (Cordova, Cunningham, Carlson & Andrykowski, 2001). However, in these instances, the traumatic event does not necessarily caus e HRQoL issues, such as late effects, as time since the trauma increases, as it does with cancer survivors. Indeed, the number of late effects in our sample increased as the years since diagnosis increased. Long Term Adjustment Long Term Adjustment and Pa rticipant Characteristics In this study, we conceptualized long term adjustment as a construct made up of quality of life, health related quality of life, and life satisfaction. Although posttraumatic growth was negatively assoc iated with time since diagn osis long term adjustment was not associated with time since diagnosis So although some adult survivors of childhood cancer may struggle with finding positive meaning in their illness and optimism as time goes on, their QoL, HRQoL and overall satisfactio n does not appear to be negatively impacted by passage of time. Participants' report of long term adjustment was also not related to having had multiple treatment regimens or more late effects, and was not significantly different based on the type of cance r. Long Term Adjustment and Psychosocial Functioning Not surprisingly, p articipants' long term adjustment was negatively associated with distress. L ong term adjustment was also positively correlated with perceived social
51 support, optimism, resilience and posttraumatic growth. All of these appear to be important correlates to the maintenance of long term adjustment, more than any factor specifically related to the participants' cancer diagnosis and treatme nt. It is likely that these strong correlations may be associated with the fact that quality of life includes a number of psychosocial subscales leading to strong correlations. Nevertheless, it is important to note that distress appears to be strongly rela ted to long term adjustment while the number of late effects and the type of treatment received were not strongly correl ated with long term adjustment. In addition to providing hope for the future to those survivors who were severely affected by their trea tment and late effects, t his is particularly important when determining the most effective interventions for adult survivors of childhood cancer. It appears that decreasing levels of distress, and increasing social support, positive thinking, and resilienc e will help increase their posttraumatic growth and lead to better overall adjustment in adulthood. Carver (1998) demonstrated that increasing resilience involves helping patients utilize various internal and external resources available to them, including social support, financial options, and emotional and physical strengths that will help them maintain a sense of well being
52 Figure 4.1 Hypothesized model of factors leading to long term adjustment Figure 4.2 Revised model of factors leading to long term adjustment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
53 O ur initial hypotheses were based on the model presented in Figure 4.1 The first model suggests that survivor characteristics are associated with long term adjustment, as well as posttraumatic growth and psych osocial functioning. S urvivor characteristics such as age at diagnosis, cancer type and the number of late effects were found to be associated with psychosocial functioning and posttraumatic growth, but not long term adjustment. Factors involved in psychos ocial functioning, particularly optimism, resilience and social support, had strong positive relationships with both posttraumatic growth and long term adjustment. The second model pres ented in Figure 4.2 represents our current understanding of the relatio nships between these factors, particularly that long term adjustment is not positively or negatively associated with any specific survivor characteristic. Limitations There were a number of limitations to our study, one of which was a small sampl e size. This significantly limited the type of analyses that could be done. In particular, we were unable to perform a hierarchical cluster analysis on the responses to the Impact of Cancer Scale, and also were unable to use multiple regression and step wi se strategies to determine the specific amounts of variance in long term adjustment that was accounted for amongst these different variables. This is an important analysis in order to determine the biggest predict ors of long term adjustment, and be able to conclude causation. Sample size also impacted our ability to detect medium to small effect size s and there may have been other effects that we were unable to measure because they were not large effects. The pool of eligible participants was a unique group of young adults with a history of pediatric cancer who attended a one time consultative clinic for childhood
54 cancer survivors (TACTIC clinic) It was a small population and only a select number of the survivors treated at Children's Hospita l Colorado or other surrounding pediatric cancer centers attended the TACTIC clinic. The population of our study was therefore a convenience sample, rather than a probability sample, since all participants volunteered to be part of the clinic and were not chosen at random. Select ion bias is therefore a concern, and those who attended the clinic are not representative of the adult survivors of childhood cancer population as a whole for a number of reasons. First, our sample lacked ethnic or racial diversity. Second, the survivors at the TACTIC clinic may have had different levels of functional impairment and number of late effects compared to the general survivor population. It is hard to know if participants in the TACTIC clinic were higher or lower function ing than those who did not attend, and we did not have access to other variables that might have helped improve our understanding of the differences between the TACTIC participants and other adult survivors of childhood cancer. Additionally, the clinic req uired that the participants have adequate means to cover the cost of the ir care (insurance or self pay), which likely decreased the generalizability of the sample Therefore, we have limited ability to generalize our results to the larger population of adu lt survivors of childhood. Another limitation was the number of non respondents in our pool of possible participants. Data collection was limited by the review board's restrictions on the number of contacts allowed between researchers and participants fo r any one project. We were allowed contact only those participants who have attended the clinic over the past five years a maximum of 3 times. Although we were disappointed in the low number of responders to the survey the response rate of 41% is not unus ual in Internet based
55 psychosocial research A review of available literature on response rate for emailed surveys shows that average response rates are around 30% (Medway & Fulton, 2012; Israel, 2011; Archer, 2008; Sheehan, 2006 ). Differences between thos e who chose to participate in the study and those who did not were compared. Previous data from participants' first visit to the TACTIC clinic were analyzed using independent samples t tests. When looking at date of their initial appointment for the TACTIC clinic, distance traveled to the TACTIC clinic, and cancer type we found no significant differences when comparing those who chose to completed the survey and those who did not. Unfortunately we were not able to compare the groups on other variables such as psychosocial distress or health status so it is unclear if these factors may have differed between completers and non completers. There are also some limitations as a result o f the nature of the self report Internet based assessment. First, relying on self report can introduce additional error, especially as some of the participants were diagnosed at a young age and may not have an accurate memory for the events during and immediately following their treatment. Second, there is a risk when using intern et based assessment that the intended participant was not the one who completed the survey. Conclusions and Future Directions Long term adjustment is a multi faceted phenomenon that includes quality of life, health related quality of life, and satisfacti on with life, and involves the intersection of many different areas of functioning, including resilience, optimism, social support, and posttraumatic growth. The more the medical and psychological community understands about factors that contribute to impr oved long term adjustment in adult survivors of
56 childhood cancer, the earlier they will be able to identify and intervene in order to improve long term adjustment It will also allow providers to focus their efforts on areas of a survivor's life where chan ge would provide the most benefit over time. Finally, it will allow providers the opportunity to provide a positive outlook on life after cancer for survivors of different types of cancer, and those who endure multiple treatment regimens. Future investig ation into long term adjustment of adult survivors of childhood cancer could focus on specific affects of childhood cancer on meeting milestones in adulthood. In particular, education and employment, romantic relationships and having children. Investigatin g potential hindrance to goals as a result of having childhood cancer is an important step to better understanding the life challenges that may affect long term adjustment. Also, re evaluating using the same set of questionnaires with a larger, more repres entative sample would provide validity, reliability and generalizability for the conclusions that resulted from this data.
57 REFERENCES American Cancer Society (2014). Cancer Facts and Figures 2014. Atlanta: American Cancer Society; 2014. Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173 1182. Bibinger, M. (2013). Notes on the sum and maximum of independent exponentially distributed random variables with different scale parameters. Institute for Mathematics, Humboldt University 6, 99 100. Bruce, M., 2006. A systematic and conceptual review of posttraumatic stress in childhood cancer survivors and their parents. Clinical Psychology Review 26, 233 256. Byrne J, Fears TR, Steinhorn SC et al., 1989. Marriage and divorce after childhood and adolescent cancer. Journal of the American Medical Association 262: 2693 2699. Cancer 49, 17 7 182. Cancer Trends Progress Report 2011/2012 Update National Cancer Institute, NIH, DHHS, Bethesda, MD, August 2012, http://progressreport.cancer.gov. Chiarelli, A. M., Marrett, L. D., & Darlington, G. (1999). Early menopause and infertility in fema les after treatment for childhood cancer diagnosed in 1964 1988 in Ontario, Canada. American Journal of Epidemiology, 150, 245 254. Christ, G. H., Lane, J. M., & Marcove, R. (1995). Psychosocial adaptation of long term survivors of bone sarcoma. Journal o f Psychosocial Oncology, 13, 1 22. Cohen, S., & Hoberman, H. (1983). Positive events and social supports as buffers of life change stress. Journal of Applied Social Psychology, 13, 99 125. Cohen, S., Mermelstein, R., Kamarck, T., & Hoberman, H. (1985). Measuring the functional components of social support. In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory, research, and application. The Hague, Holland: Martinus Nijhoff. disorder (PTSD) in young adult survivors of childhood cancer. Pediatri c Blood & Dolgin, M. J., Somer, E., Buchvald, E., & Zaizov, R. (1999). Quality of life in adult survivors of childhood cancer. Social Work in Health Care, 28, 31 43. Eiser C, Hill JJ, Vance YH. 2000. Examining the psychosocial consequences of surviving childhood cancer: Systematic review as a research method in pediatric psychology. Journal of Pediatric Psychology 25: 49 60.
58 Eiser C. 1990a. Chronic childhood disease. An introduction to psychological theory and research. Eiser C. 1990b. Psychological eff ects of chronic disease. Journal of Child Psychology and Psychiatry 31: 85 98. Elkin, T.D., Phipps, S., Mulhern, R.K., Fairclough, D., 1997. Psychological functioning of adolescent and young adult survivors of pediatric malignancy. Medical and Pediatric O ncology 29, 582 588. Evans, S. E., & Radford, M. (1995). Current lifestyle of young adults treated for cancer in childhood. Archives of Disease in Childhood, 72, 423 426. Formann, A.K., 1984. Die Latent Class Analyse: EinfÂŸhrung in die Theorie und Anwendung. Weinheim: Beltz. Garber J. 1984. Classification of childhood psychopathology: A developmental perspective. Child Development. 55: 30 48. Griffith, K. C., & Hart, L. K. (2000). Characteristics of adult cancer survivors who pursue postsecondary education. Cancer Nursing, 23, 468 476. Hauke, J, & Kossowski, T. (2011) Comparison of values of Pearson's and Spearman's correlation coefficients on the same sets of data. Quaestiones Geographicae, 30( 2 ), 87 93. Haupt R, Byrne J, Connelly RR et al. 19 92. Smoking habits in survivors of childhood and adolescent cancer. Medical Pediatric Oncology 20: 301 306. Hobbie,W. L., Stuber, M., Meeske, K.,Wissler, K., Rourke, M. T., Ruccione, K., et al. (2000). Symptoms of posttraumatic stress in young adult survi vors of childhood cancer. Journal of Clinical Oncology, 18, 4060 4066. Kinehan, K.E., Sharp, L.K., Seidel, K., Leisenring, W., Didwania, A., Lacouture, M.E., Stovall, M., Haryani, A., Robison, L.L., Krull, K.R. (2012). Scarring, disfigurement, and quality of life in long term survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. Journal of Clinical Oncology 30(20), 2466 2474. Langeveld, N.E., Grootenhuis, M.A., Vou te, P.A., de Haan, R.J., van den Bos, C., 2004. Quality of lif e, self esteem and worries in young adult survivors of childhood cancer. Psycho oncology 13, 867 881. Langeveld, N.E., Stam, H., Last, B.F., 2002. Quality of life in young adult survivors of Quality of life, self esteem and worries in young adult survivo rs of childhood cancer. Psycho oncology 13, 867 881. Lett H.S., Blumenthal J.A., Babyak M.A., Strauman T.J., Robins C., Sherwood A. (2005). Social support and coronary heart disease: Epidemiologic evidence and implications for treatment. Psychosomatic Med icine, 67(6).
59 Lewis M, Miller SM. 1990. Handbook of Developmental Psychopathology. Plenum: New York. Madan Swain A, Brown RT, Foster MA et al. 2000. Identity in adolescent survivors of childhood cancer. Journal of Pediatric Psychology 25: 105 115. Meeske, K. A., Ruccione, K., Globe, D. R., & Stuber, M. L. (2001). Posttraumatic stress, quality of life, and psychological distress in young adult survivors of childhood cancer. Oncology Nursing Forum, 28, 481 488. Oeffinger, K. C., Eshelman, D. A., Toml inson, G. E., Buchanan, G. R., & Foster, B. M. (2000). Grading of late effects in young adult survivors of childhood cancer followed in an ambulatory adult setting. Cancer, 88, 1687 1695. Ostroff, J., & Steinglass, P. (1996). Psychosocial adaptation follo wing treatment: A family systems perspective on childhood cancer survivorship. In L. Baider, C. L. Cooper, & A. Kaplan De Nour (Eds.), Cancer and the family. New York: John Wiley. Paul A. Harris, Robert Taylor, Robert Thielke Jonathon Payne, Nathaniel Gonzalez, Jose G. Conde, Research electronic data capture (REDCap) A metadata driven methodology and workflow process for providing translational research informatics support, J Biomed Inform. 2009 Apr;42(2):377 81. Pemberger S., Jagsch, R., Frey, E., Felder Puig, R., Gadner, H., Kryspin Exner, I., Topf, R., 2005. Quality of life in long term childhood cancer survivors and the r elation of late effects and subjective well being. Support Care Cancer 13, 45 56. Prouty, D., War d Smith, P., & Hutto, C.J. (2006). The lived experience of adult survivros of childhood cancer. Journal of Pediatric Oncology Nursing. 23(3), 143 151. Rauck AM, Green DM, Yasui Y, Mertens A, Robinson LL. 1999. Marriage in the survivors of childhood cancer : A preliminary description from childhood cancer survivor study. Medical Pediatric Oncology 33: 60 63. Richardson, R. D., Nelson, M. B., & Meeske, K. (1999). Young adult survivors of childhood cancer:Attending to emerging medical and psychosocial needs. Journal of Pediatric Oncology, 16, 136 144. Ries, L. A. G., Eisner, M. P., Kosary, C. L., Hankey, B. F., Miller, B. A., Clegg, L., et al. (2004). SEER cancer statistics review, 1975 2001. Bethesda, MD: National Cancer Institute. http://seer.cancer.gov/csr/1975_2002/ Robinson KE, Kuttesch JF, Champion JE, et al.: A quantitative meta analysis of neurocognitive sequelae in survivors of pediatric brain tumors. Pediatr ic Blood Cancer 55 (3): 525 31, 2010. Robison, L.L., Ness, K.K., 2008. Psychosocial outcomes and health related quality
60 of life in adult childhood cancer survivors: a report from the childhood cancer survivor study. Cancer Epidemiology Biomarkers & Prevention 17, 435 446. Rourke, M.T., Hobbi e, W.L., Schwartz, L., Kazak, A.E. ( 2007 ) Posttraumatic stress Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism from neuroticism (and trait anxiety, self mastery, and self esteem): A re evaluation of the Life Orientation Test. Journal of Personality and Social Psychology, 67, 1063 1078. Smith, B.W., Dalen, J., Wiggin s, K., Tooley, E., Christopher, P., Bernard, J. (2008). The brief resilience scale: assessing the ability to bounce back. International Journal of Behavioral Medicine; 15(3), 194 200. Stam, H., Grootenhuis, M.A., & Last, B.F. (2005). The course of life o f survivors of childhood cancer. Psycho Oncology. 14, 227 238. Sundberg, K.K., Lampic, C., Bjork, O., Arvidson, J., & Wettergreen, L. (2009). Positive and negative consequences of childhood cancer influencing the lives of young adults. European Journal o f Oncology Nursing, 13, 164 170. Tedeschi, R.G., Calhoun, L.G., 1996. The posttraumatic growth inventory: measuring the positive legacy of trauma. Journal of Traumatic Stress 9, 455 471. Teeter MA, Holmes GE, Holmes FF, Baker AB. 1987. Decisions about ma rriage and family among survivors of childhood cancer. Journal of Psychosocial Oncology 5: 59 68. Vann, J. C. J., Biddle, A. K., Daeschner, C. W., Chaffee, S., & Gold, S. H. (1995). Health insurance access to young adult survivors of childhood cancer in North Carolina. Medical and Pediatric Oncology, 25, 389 395. Ware, J. E., Davies Avery, A., & Brook, R. H. (1980). Analysis of relationships among health status measures. Santa Monica, CA: Rand Corporation. Weigers, M.E., Chesler, M.A., Zebrack, B.J., Go ldman, S., 1998. Self reported worries among long term survivors of childhood cancer and their peers. Journal of Psychosocial Oncolgy 16, 1 23. Wyatt, G.K.H., & Friedman, L.L. (1996). Development and testing of a quality of life model for long term female cancer survivors. Quality of Life Research. 5, 387 394. Zebrack B, Yi J, Petersen L, Ganz P. ( 2013 ). The impact of cancer and quality of life for long term survivors. Psycho Oncology.
61 Zebrack BJ, Ganz PA, Bernaards CA, Petersen L, Abraham L. Assessing the impact of cancer: development of a new instrument for long term survivors. Psycho Oncology. 15(5):407 421. Zebrack, B.J. (2000). Quality of life in long term survivors of leukemia and lymphoma. Journal of Psychosocial Oncology, 18(4), 39 59. Zebrack B.J., Chesler, M.A., 2002. Quality of life in childhood cancer survivors. Psycho oncology 11, 132 141. Zevon MA, Neubauer NA, Green DM. 1990. Adjustment and vocational satisfaction of patients treated during childhood or adolescence for acute lymphoblas tic leukemia. The American Journal of Pediatric Hematology/Oncology 12: 454 461.
62 A PPENDIX A Psychosocial Questionnaires Impact of Events Scale Revised (IES R) Below is a list of difficulties people sometimes have after stressful life events. Please read each item, and then indicate how distressing each difficulty has been for you DURING THE PAST SEVEN DAYS with respect to (your problem), how much were you distre ssed or bothered by these difficulties? 0 = Not at all 1 = A little bit 2 = Moderately 3 = Quite a bit 4 =Extremely 1 Any reminder brought back feelings about it 2 I had trouble staying asleep 3 Other things kept making me think about it 4 I felt irritable and angry 5 I avoided letting myself get upset when I thought about it or was reminded of it 6 I thought about it when I didn't mean to 7 I felt as if it hadn't happened or wasn't real 8 I stayed away from reminders about it 9 Pictures about it popped into my mind 10 I was jumpy and easily startled 11 I tried not to think about it 12 I was aware that I still had a lot of feelings about it, but I didn't deal with them 13 My f eelings about it were kind of numb 14 I found myself acting or feeling like I was back at that time 15 I had trouble falling asleep 16 I had waves of strong feelings about it 17 I tried to remove it from my memory 18 I had trouble concentrating 19 Reminders of it caused me to have physical reactions, such as sweating, trouble breathing, nausea, or a pounding heart 20 I had dreams about it 21 I felt watchful and on guard 22 I tried not to talk about it !!
63 ISEL 12 Instructions: This scale is made up of a list of statements each of which may or may not be true about you. For each statement circle "definitely true" if you are sure it is true about you and "probably true" if you think it is true but are not absolutely certain. Similarly, you should circle "definitely false" if you are sure the statement is false and "probably false" if you think it is false but are not absolutely certain. 1. If I wanted to go on a trip for a day (for example, to the country or mountai ns), I would have a hard time finding someone to go with me. 2. I feel that there is no one I can share my most private worries and fears with. 3. If I were sick, I could easily find someone to help me with my daily chores. 4. There is someone I can turn to for advice about handling problems with my family. 5. If I decide one afternoon that I would like to go to a movie that evening, I could easily find someone to go with me. 6. When I need suggestions on how to deal with a personal problem, I know some one I can turn to. 7. I don't often get invited to do things with others. 8. If I had to go out of town for a few weeks, it would be difficult to find someone who would look after my house or apartment (the plants, pets, garden, etc.). 9. If I wanted to have lunch with someone, I could easily find someone to join me. 10. If I was stranded 10 miles from home, there is someone I could call who could come and get me. 11. If a family crisis arose, it would be difficult to find someone who could give me goo d advice about how to handle it. 12. If I needed some help in moving to a new house or apartment, I would have a hard time finding someone to help me. LOT R Next are some questions on your opinions about things in life. The items are presented as statements. Think about each one, and decide whether it says something you agree or disagree with. Be as honest and accurate as you can, but remember these are opi nion items, so there are no "right" or "wrong" answers. Answer according to your own feelings. Reply using these 4 choices: 1 = I agree a LOT 2 = I agree a little 3 = I DISagree a little 4 = I DISagree a LOT __ _ 1. In uncertain times, I usually expect the best. _____ 2. If something can go wrong for me, it will _____ 3. I'm always optimistic about my future. _____ 4. I hardly ever expect things to go my way. _____ 5. I rarely count on good things happening to me. _____ 6. Overall, I expect more good things to happen to me than bad.
64 Post Traumatic Growth Inventory Before answering the following questions, focus on one traumatic or life altering event that has occurred in your life. 0 = I did not experience this change as a result of my crisis. 1 = I experienced this change to a very small degree as a result of my crisis. 2 = I experienced this change to a small degree as a result of my crisis. 3 = I experienced this change to a moderate degree as a result of my crisis. 4 = I experien ced this change to a great degree as a result of my crisis. 5 = I experienced this change to a very great degree as a result of my crisis. 1. I changed my priorities about what is important in life. 2. I have a greater appreciation for the value of my ow n life 3. I developed new interests. 4. I have a greater feeling of self reliance. 5. I have a better understanding of spiritual matters. 6. I more clearly see that I can count on people in times of trouble. 7. I established a new path for my life. 8 I have a greater sense of closeness with others. 9. I am more willing to express my emotions. 10. I know better that I can handle difficulties. 11. I am able to do better things with my life. 12. I am better able to accept the way things work out. 1 3. I can better appreciate each day. 14. New opportunities are available which wouldn't have been otherwise. 15. I have more compassion for others. 16. I put more effort into my relationships. 17. I am more likely to try to change things which need cha nging. 18. I have a stronger religious faith 19. I discovered that I'm stronger than I thought I was. 20. I learned a great deal about how wonderful people are. 21. I better accept needing others.
65 Brief Resilience Scale Please indicate the extent to which you agree with each of the following statements by using the following scale: 1 = strongly disagre e, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree. "# $!%&'(!%)!*)+',&!*-,.!/+0,.12!-3%&4!5-4(!%06&7 8# $!5-9&!-!5-4(!%06&!6-.0':!0%!%54)+:5!7%4&773+1!&9&'%7! ; <=>== @# $%!()&7!')%!%-.&!6&!1)':!%)!4&,)9&4!34)6!-!7%4&773+1!&9&'%# A# $%!07!5-4(!3)4!6&!%)! 7'-B!*-,.!C5&'!7)6&%50':!*-(!5-BB&'7! ; <=>== D# $!+7+-112!,)6&!%54)+:5!(0330,+1%!%06&7!C0%5!10%%1&!%4)+*1& E# $!%&'(!%)!%-.&!-!1)':!%06&!%)!:&%!)9&4!7&% ; *-,.7!0'!62!103&! ; <=>== Satisfaction With Life Scale Below are five statements that you may agree or disagree with. Using the 1 7 scale below, indicate your agreement with each item by placing the appropriate number on the line preceding that item. Please be open and honest in your responding. 7 Strongly agree 6 Agree 5 Slightly agree 4 Nei ther agree nor disagree 3 Slightly disagree 2 Disagree 1 Strongly disagree ____ In most ways my life is close to my ideal. ____ The conditions of my life are excellent. ____ I am satisfied with my life. ____ So far I have gotten the important thi ngs I want in life. ____ If I could live my life over, I would change almost nothing.
66 Demographic and Follow Up Health Status and Emotional Health Questions (from the original TACTIC survey)