Abstract
The term cancer survivor originally designated a family member who lost a loved one to cancer (Leigh, 1996). With growing knowledge about cancer and more successful treatments, a long-term cancer survivor is now defined as a person who lives for 5 yr or more after the diagnosis of cancer (Mols, Vingerhoets, Coebergh, & van de Poll-Franse, 2005). The increased survival rate of approximately 70% of children with brain tumors (BTs) in North America (Canadian Cancer Society [CCS], 2011) is a major medical achievement. However, for many survivors, overcoming cancer is just the first of many hurdles to be surmounted after diagnosis (Turner, Rey-Casserly, Liptak, & Chordas, 2009).
Despite the advances in research and technology, patients with BT and survivors experience many negative physical, psychosocial, and neuropsychological late effects (Boman, Hovén, Anclair, Lannering, & Gustafsson, 2009). A late effect is a condition that appears after the acute phase of an earlier condition; it can be caused either directly, by the condition, or indirectly, by the treatment. Late effects often leave patients unnecessarily disabled or, at the very least, able to function only at a suboptimal level (Hewitt, Greenfield, & Stovall, 2005). The percentage of survivors of childhood BT with a medium, high, or severe burden of adverse health outcomes is estimated at more than 80% (Geenen et al., 2007). Furthermore, 60% have at least one impairment (Aarsen et al., 2006; Macedoni-Luksic, Jereb, & Todorovski, 2003), including visual, motor, cognitive, or neurological impairments; endocrine complications; and abnormalities found on computed tomography scans.
Ness et al. (2010) reported that survivors of childhood BT have decreased employment, less education beyond high school, and less independent living. However, because few researchers have addressed the functional outcome of patients with BT from a rehabilitation perspective (Kunin-Batson et al., 2011; Ness et al., 2005, 2010), the impacts of late effects on function and activities of daily living (ADLs) are poorly documented and quantified. ADLs are defined as essential activities for self-care, such as washing and feeding, and for independent living, such as home management and meal preparation (Lawton, 1987). Research that objectively measures ADL performance in survivors of childhood cancer is rare in the literature (Parks, Rasch, Mansky, & Oakley, 2009). In addition, little research has been conducted to examine how survivors of childhood BT cope with everyday living, and descriptions of their ADL performance are not available.
As life expectancy increases for patients with cancer, there is growing recognition that patient-reported outcome measures, including measures of health-related quality of life (HRQoL), convey important information to determine the overall burden of cancer (Lipscomb, Gotay, & Snyder, 2007). HRQoL is a multidimensional construct that assesses the impact of a disease and its treatment in terms of physical, social, emotional, cognitive, and overall well-being (Berra et al., 2007; World Health Organization, 1996). It is best evaluated with self-reported instruments (International Society for Quality of Life Research, 2015), because each person is the ultimate judge of his or her own health and treatment outcome. Survivors of childhood BT have one of the lowest HRQoL among survivors of childhood cancer as a result of a complex array of late effects (McDougall & Tsonis, 2009; Zeltzer et al., 2009). Past studies have suggested that associations exist among physical performance, executive function, and HRQoL outcomes in adult survivors of childhood cancer (Ness et al., 2008) but not specifically in survivors of childhood BT. Therefore, research is needed to understand the relationship between ADL performance and HRQoL in this population.
This study addresses ADL performance of youth and young adult survivors of childhood BT (ages 16–29 yr) and its association with HRQoL because these factors reflect the everyday lives of survivors beyond the acute diagnosis and treatment phases. The specific aims were to (1) estimate the extent to which youth and young adult survivors of childhood BT achieved a level of ADL performance that was comparable to that of their peers and (2) assess the association between observed limitations in motor and process skills and HRQoL.
Method
Participants
A cross-sectional study was conducted with a convenience sample of survivors of childhood BT recruited from a long-term follow-up clinic at Sainte-Justine University Hospital Centre (UHC) in Montreal. Participants had to meet the following inclusion criteria: treatment between 1987 and 2007 inclusively at CHU and age ≤21 yr during that period, age ≥16 yr at the time of the study, BT as the first cancer occurrence, at least 3 yr off treatment with no recurrence or progression, and at least 5 yr since diagnosis. Excluded from participation were people who had a spinal cord tumor, were treated by surgery only, had a preexisting comorbid condition not related to the cancer, and had treatments that might have affected their functional outcome.
Sample size calculations were based on the main outcome measure, the Assessment of Motor and Process Skills (AMPS; Fisher, 2012). For the effect size, a significant change for motor or process skill measures was at least 0.3 logit (Magalhaes, Hamonet, de Sousa, Matos, & da Costa, 1996; Silverstein, Fisher, Kilgore, Harley, & Harvey, 1992). For a level of significance of ɖ (one-tailed) = .05, β = .20, and the standard error of the sample = 0.9, sample sizes of 36 for motor skills and 26 for process skills were required. Therefore, 36 participants were recruited. The norms used were healthy people of the same age (16–29 yr) from the AMPS standardization sample (Fisher & Merritt, 2012). The study protocol was approved by the UHC research ethics board. Written informed consent was obtained from participants ages ≥18 yr and from parents of participants ages <18 yr.
Data Collection
Every month, a list of survivors of childhood BT attending their annual appointment at the long-term follow-up clinic at UHC was provided by the research nurse. The list of transferred patients was also consulted on a regular basis. The main researcher (Catherine Demers) contacted patients who met the eligibility criteria by telephone. Patients who agreed to participate were asked to give their written, informed consent. Medical charts were reviewed and clinical data were abstracted to obtain relevant material on participants’ medical condition, treatment information, and past history.
Participants completed the assessments during a follow-up visit or at another convenient time. For participants living in the Montreal area, the possibility of conducting the assessments at home was also proposed for the patient’s convenience. During the appointment, the participants were first asked to fill out two questionnaires, one to assess their HRQoL and another to document their current occupation, level of education, and living situation. The functional assessment was administered to study participants by the main researcher, who is a calibrated AMPS rater. The evaluation took, on average, 1–2 hr to complete, and all participants were assessed over a 10-mo period (from February to December 2013).
Measures
The quality of ADL task performance was evaluated by a standardized objective measure, the AMPS. The AMPS assesses 16 motor and 20 process skills using a 4-point ordinal scale (1 = unacceptable performance, 2 = ineffective, 3 = questionable, 4 = competent) according to explicit, specific criteria. ADL motor skills are observable actions used to move oneself and task objects (e.g., reaches, bends, calibrates), and ADL process skills are observable actions used to organize and adapt task actions to prevent or overcome problems (e.g., initiates, chooses, and organizes). The AMPS assessment is designed to measure observable goal-directed skills related to enacting an ADL task performance, not to test specific physical or cognitive impairments. During AMPS administration, the rater observes a participant performing two culturally relevant ADL tasks, such as preparing a sandwich and making a bed. Studies have supported the reliability and validity of the AMPS across age groups and gender (Hayase & Thimmaiah, 2004; Merritt & Fisher, 2003) and its sensitivity as an outcome measure (Cooke, Fisher, Mayberry, & Oakley, 2000; Graff et al., 2006; Kottorp, Bernspång, & Fisher, 2003).
HRQoL was evaluated by having each participant complete the Medical Outcomes Study 12-Item Short-Form Health Survey (SF–12; Ware, Kosinski, & Keller, 1996). The SF–12 is a generic health profile that produces two overall summary scores, the physical component summary (PCS) and the mental component summary (MCS). The PCS is based on the domains of Physical Functioning, Role Limitations Due to Physical Health Problems, Bodily Pain, and General Health Perception, and the MCS is based on the domains of Vitality, Social Functioning, Role Limitations Due to Emotional Problems, and Mental Health (Ware & Sherbourne, 1992). The SF–12 is widely used with numerous health conditions and has previously been used with survivors of childhood cancer (Bober et al., 2013; O’Leary, Diller, & Recklitis, 2007).
Data Analysis
Descriptive statistics were calculated for the sociodemographic and clinical characteristics of the eligible participants (Table 1). The AMPS computer-scoring software was used to generate ADL motor and process skill scores for each study participant. Many-faceted Rasch analysis was then used to convert the raw, ordinal scores to logits, which are linear measures expressed as equal-interval, log odds probabilities. Means and 95% confidence intervals (CIs) for ADL motor and process skill scores were calculated and expressed as logits (Table 2). Differences in mean ADL motor and process skill scores between participants and healthy norms were determined using a one-sample t test (p < .05). Raw scores from the SF–12 were converted to t scores (range = 0–100), and differences in mean MCS and PCS scores between participants and the general population were determined using a one-sample t test (p < .05; Table 3).
Participant Sociodemographic and Clinical Characteristics
Note. N = 36. PNET = primitive neuroectodermal tumor.
Preuniversity college in Quebec province education system.
Participants could have more than one condition or type of service.
Motor and Process Skill Scores, Participants Versus Age Group Norms
Note. AMPS = Assessment of Motor and Process Skills; M = mean; SD = standard deviation.
SF–12 MCS and PCS Scores, Participants Versus Age Group Norms
Note. M = mean; MCS = mental component summary; PCS = physical component summary; SD = standard deviation; SF–12 = Medical Outcomes Study 12-Item Short-Form Health Survey. Raw scores from the SF–12 were converted to t scores for this analysis.
Multiple linear regression was used to evaluate the relationship between ADL motor and process skill scores and MCS and PCS scores for HRQoL. Simple linear regressions were first performed to evaluate the association between each SF–12 summary score and each AMPS ADL performance skill score. The potential for confounding for age, gender, time since treatment, age at diagnosis, tumor type and location, treatment type, radiation dose, and chronic health condition was then tested using simple linear regression. Each potential confounding variable was assessed for its relationship with the outcome (HRQoL) and with each of the predictor variables (motor and process skills). Confounding variables found to have a significant effect on both the outcome and one of the predictor variables were added to the model. Statistical significance was set at the 5% level. Data analyses were performed using IBM SPSS Statistics (Version 20; IBM Corp., Armonk, NY).
Results
Participant Characteristics
The sociodemographic and clinical characteristics of our study participants (see Table 1) differed from CCS (2008, 2011) statistics. According to the CCS (2008), the incidence of BT is approximately 20% higher in boys than in girls; however, only 39% of our participants were boys. In addition, the CCS (2011) reported that in the 16- to 29-yr age group 44% of BTs are astrocytomas and 20% are intracranial or intraspinal; however, in our study they accounted for 14% and 64%, respectively.
Activities of Daily Living and Health-Related Quality of Life
Participants’ motor and process skill scores were significantly lower than the norms in the 16- to 29-yr age group (p < .05; see Table 2). Participants’ MCS and PCS scores indicated that their mental and physical health fall within the average of the general population, with means of 50.27 and 48.16, respectively (see Table 3). The associations between each ADL performance skill (motor and process) score and each HRQoL summary score (MCS and PCS) are shown in Figure 1. For the multiple linear regression models, after adjusting for gender and chronic health conditions, the level of performance in ADL skills (motor and process) was significantly associated with overall physical health (r 2 = .5), and after adjusting for time since treatment and chronic health conditions, with mental health (r 2 = .3).
Associations between AMPS ADL motor and process skills scores and SF–12 MCS and PCS scores. (A) ADL motor skills vs. PCS scores; (B) ADL motor skills vs. MCS scores; (C) ADL process skills vs. PCS scores; (D) ADL process skills vs. MCS scores. Higher motor and process skill scores represent higher performance; higher MCS and PCS scores represent higher health-related quality of life.
Discussion
The purpose of this study was to examine ADL functioning and HRQoL among youth and young adult survivors of childhood BT. To our knowledge, this study is the first to document specific deficits in ADLs using an objective, standardized, and performance-based measure. Our study demonstrated that ADL functioning of survivors of childhood BT is significantly lower than that of age norms but that HRQoL is similar to the general population. Furthermore, we showed that the functional outcome is positively associated with survivors’ self-reported physical and mental health.
Over the past 2 decades, the cancer literature has reported variable results regarding functional abilities of long-term survivors of childhood BT (Aarsen et al., 2006; Armstrong et al., 2009; Bhat et al., 2005; Boman et al., 2009; Fouladi et al., 2005). The selection of functional assessment tools for patients with cancer has been particularly problematic, especially because many questionnaires completed by these patients were not designed specifically to elicit information about function. Because we used an objective, standardized, and performance-based measure in this study, we were able to provide more accurate, real-life information on the functional status of survivors of childhood BT. The findings from this study highlight that the late effects experienced by survivors of childhood BT affect their performance in their basic day-to-day life and activities, such as cooking and cleaning, and in their capacity to live independently.
Young survivors of childhood BT face additional challenges as they adapt to the roles, responsibilities, and challenges of adulthood. Late effects may influence the completion of developmental tasks, such as establishing independence from parents, starting a career, developing intimate relationships, and planning a family (Zebrack & Walsh-Burke, 2004). For example, in our study, only a few participants lived independently. In addition, only about half of the participants who had finished school had a full-time job. Results of this study identify survivors of childhood BT as a group who are still in need of services to optimize their health and functional outcome many years after the completion of treatment.
Overall, the results indicate that survivors of childhood BT report good HRQoL. These findings are consistent with those of another study concluding that survivors of childhood BT did not differ from population norms for the majority of HRQoL domains (Aukema, Schouten-van Meeteren, Last, Maurice-Stam, & Grootenhuis, 2013). Reports by these survivors of similar HRQoL compared with population norms could be explained by the response shift phenomenon (Sprangers & Schwartz, 1999); that is, experiencing childhood cancer inoculates survivors against other negative life experiences and provides them with feelings of life satisfaction and overall psychological well-being (Zeltzer et al., 2008). Previous studies have raised the possibility that some survivors not only adapt to the negative impact of cancer but also thrive as a result, therefore reporting a better quality of life after cancer (Zebrack, 2000). However, other studies have reported no differences in psychological outcomes between survivors of childhood cancer and norms, a result that may be influenced by response bias (Recklitis et al., 2006). Another possible explanation is the limitation of normative data; therefore, using a control group such as siblings of long-term survivors may provide a more accurate assessment of HRQoL in survivors of childhood BT (Zeltzer et al., 2009).
Our results showed a statistically significant association between performance limitations and HRQoL. Consequently, interventions aimed at improving the performance of ADLs in survivors of childhood cancer may eventually improve HRQoL. However, the relatively weak relationship between ADL performance and HRQoL suggests that factors other than functional outcome must be taken into consideration to explain survivors’ perceived health. For example, common challenges identified by young survivors of cancer, such as social support and satisfaction with physical appearance (D’Agostino & Edelstein, 2013), were not addressed in this study.
Limitations and Future Research
Our study has potential limitations. Its cross-sectional nature provides only a snapshot of participants’ ADL performance and HRQoL; therefore, the impact of late effects over time is unknown for this sample. Moreover, the small number of study participants did not allow us to perform additional analyses between subgroups. Other limitations are the convenience sample; the absence of a proper control group; and the restriction of the data to one hospital, which may limit the generalizability of the results.
For future studies, we recommend methodological enhancement (e.g., conduct a case-control study). Rehabilitation programs to optimize both function and quality of life should be developed for adult survivors of cancer and for children currently undergoing therapy. A knowledge translation strategy should also be planned to ensure clinicians working in the area of childhood cancer are informed of the potential vulnerabilities in this population to enable them to help patients and families as much as possible.
Implications for Occupational Therapy Practice
Our findings have the following practical implications for occupational therapy practice:
Occupational therapy practitioners can use objective, performance-based measures to gain a more comprehensive understanding of performance of ADLs in survivors of childhood cancer.
Timely evaluation and intervention in occupational therapy could help survivors of childhood BT achieve optimal independence, participation, and HRQoL.
Increased awareness of the potential for functional limitations of BT survivors may increase access and utilization of occupational therapy services.
Conclusion
Our study is the first to document specific deficits in ADLs in survivors of childhood BT using an objective, standardized, and performance-based measure. The results underline the impact of childhood BTs on the functional outcome of survivors. On average, adult survivors of childhood BT have significantly lower motor and process skills than age norms but good HRQoL. Moreover, this research shows a positive association between performance of ADLs, from an objective perspective, and HRQoL, from the participant’s perspective.
Footnotes
Acknowledgments
This research was supported by the Coast-to-Coast Foundation. We would like to acknowledge the contribution of the CHU oncology rehabilitation team and to thank the participants for their very precious time and generosity.