Health related quality of life and buffering factors in adult survivors of acute pediatric lymphoblastic leukemia and their siblings

Acute lymphatic leukemia (ALL) is the most common cancer in children with highest incidence among 2–4 year old’s. In many countries, the prognosis has improved substantially and continually during the last 60 years due to improved treatment protocols. This has resulted in a five-year survival rate up to 90% [1, 2] and an over-all life expectancy similar to the general population [3].

Treatment protocols have improved to reduce treatment-related complications both during treatment and later in life [4]. However, children with ALL do experience impairment in physical and mental health and in health-related quality of life (HRQoL) after completion of the up to 3-year-long treatment [5, 6]. A long-term disease burden remains decades post treatment [7]. While the long-term prevalence of physical health problems has been more widely investigated, knowledge on the long-term prevalence of mental health problems remains scarce. Results indicate a higher prevalence of mental ill health [8, 9] poorer self-image [9] and mild to severe anxiety and depression [10]. There are however studies that have found the opposite, that ALL survivors report better HRQoL healthy controls of similar age [11]. The latter study suggests a problem with report bias when assessing HRQoL among cancer survivors, and that future studies should include a wider range of measurements, e.g. assessment of coping mechanisms [11]. Among survivors of pediatric ALL a large variety in overall HRQoL has been described. The experience of low HRQoL is associated with treatment related factors, physical late effects, fatigue/insomnia, anxiety and depression [12, 13] whereas higher HRQoL is associated with protecting factors such as resiliency, coping skills, independence, and achievement of life goals [13].

The definition of HRQoL differs based on the instrument being used and for long-term follow-ups, there are advantages and disadvantages using generic respective disease specific HRQoL instruments [14]. The generic instruments may be inferior in capturing phenomena specific to a particular diagnostic group, but for long-term follow-up after pediatric ALL, where the study group is in remission and has a HRQoL comparable to the normal population, a disease-specific instrument may be less suitable, especially if the results should be compared with a control group without the specific disease history. The SF-36 is a generic scale based on eight subscales measuring different dimensions of HRQoL15. Apart from being a widely used instrument to assess HRQoL as an outcome, lower scorings on the SF-36 scale predict poor health outcomes, including development of chronic widespread pain [16], long term sickness absence [17] and mortality [18]. The late effects of pediatric ALL is often investigated from a perspective of pathology. Disorders with less distinct pathological representation, but which imposes substantial suffering for the affected, such as the mental and social late effects of previous illness, has been less thoroughly described. The siblings of survivors of pediatric ALL are sometimes used as a healthy control group, when estimating the long-term effects [19]. This is advantageous when the interest lies in separating effects of the disease and effects from the environment, such as the family [20].

Thus, the improvement in treatment of pediatric ALL has introduced new challenges for pediatric oncology care in understanding and handling treatment-related complications that requires long-term follow-up beyond treatment, remission and survival, and into adulthood. To meet these challenges, studies examining health and HRQoL post treatment are needed to support the development of care processes that identify problems early and address its consequences. Research has shown the importance of risk factors for the occurrence of HRQoL. But the care processes needed to promote HRQoL among cancer survivors should not only be based on identifying and managing risks in relation to HRQoL, but also support factors that may have a buffering effect on the HRQoL over time. Changes in social support and self-efficacy has been shown to be associated with changes in HRQoL over time in children and adolescents [21, 22], which supports the idea that these factors may be buffering also among young adults that have survived pediatric ALL. Prevention programs focusing on such buffering factors can be especially important for the HRQoL outcomes among young adult (YA) survivors of ALL where there is a high incidence of risk factors.

The aim of this study was to describe HRQoL and the relation to the potential buffering factors self-efficacy and social support among YA survivors of pediatric ALL and their siblings.

This cross-sectional study used data from a nationwide survey in Sweden 2012 including young adult survivors of pediatric ALL and their siblings in Sweden. The aim of the survey was to investigate potential long-term effects on different dimensions on health, HRQoL and comorbidities, and to assess care seeking behavior of the young adult survivors. The survey was carried out in 2012 in Sweden with approval from the regional ethical board (dnr 2010/579).

In all, 227 YA ALL survivors responded to the questionnaire and fulfilled the inclusion criteria. They were between 23 and 41 years old when they responded. In all, 147 did not respond to the questionnaire and 42 were excluded due to presence of mental health problems, disabilities (downs syndrome), emigration or longer stays abroad. There was no difference in current age or age at diagnosis or years since diagnosis between respondents and non-respondents/excluded, between 14 and 27 years had passed since diagnosis for both groups. There was a higher proportion of men among the non-respondents/excluded (Table 1). In all, 70 siblings out of 110 responded to the questionnaire. The included siblings were between 18 and 44 years old.

The results from this study show that the YA ALL survivors scored significantly lower on the HRQoL parameters general health and role emotional, than a control group consisting of their siblings. The general health score from SF-36 primarily assess perceived physical health, rather than mental health [33] whereas the role emotional score reflects role impairment due to emotional distress. Low role emotional scores may thus represent an increased vulnerability to poor outcome in working life [34] and an increased risk of sickness absence [17]. The YA ALL survivors also reported significantly lower general health, vitality, social functioning, role emotional and mental health than normative values from the general population. This is in line with the results from a French study, where ALL survivors scored lower on social functioning and role emotional than the general population, however that study had a wide age range of participants, including both younger adults and children [35].

In this study, both YA ALL survivors and the siblings reported lower vitality (higher fatigue) and worse mental health than the general population [27], which imply that fatigue and mental health are important factors to observe in both groups. Low vitality, or fatigue, includes both physical and mental fatigue and has high impact on the quality of life and disease burden in individuals with chronic conditions [36, 37] and is highly correlated with depression [38]. Scoring low vitality has also shown to be a predictor for chronic widespread pain [39]. It should be noted however, that the normative values from the general population are based on data assessed in the early 1990’s [27] whereas data for this study was assessed in 2012, and some of the deviation from the normative values may thus reflect a cultural or societal change during this time period.

The YA ALL survivors seemed less likely to have “normal levels” of symptoms of depression, anxiety and stress, than the siblings, although no statistical differences were seen between the groups. It was almost twice as common among YA ALL survivors compared to the siblings to report moderate to extremely severe depression, however the difference was not significant. A similar trend was seen for scoring moderate to extremely severe anxiety whereas levels of distress in YA ALL survivors and siblings were comparable. A prevalence for reporting depression, anxiety and stress in YA ALL survivors has been shown in studies from other contexts [10] and in comparison with siblings [40]. However, the levels of anxiety and distress found in our study were lower and were based on self-report rather than parent report.

A significant association between self-efficacy and all the HRQoL parameters as well as the mental health parameters were found among the YA ALL survivors. The strongest associations were seen for general health, vitality, social functioning, mental health, depression and stress. Among the siblings however, self-efficacy was associated only with the HRQoL parameters general health and mental health. The findings are supported by previous research investigating quality of life and self-efficacy in adult cancer patients [41, 42]. Social support showed to be of importance among both YA ALL survivors and the siblings regarding general health, vitality, mental health and depression. The associations were somewhat stronger among the siblings than among the YA ALL survivors. The R² values were low overall in all associations found in the regression analysis, indicating that social support and self-efficacy (respectively) only explain a smaller part of the variability of the HRQoL parameters, depression, anxiety and stress.

In summary, the differences and similarities between YA ALL survivors and their siblings, regarding HRQoL, and the associations between the buffering factors and HRQoL and mental health parameters suggests that (1) self-efficacy may be of greater importance as a buffering factor among adults who have had pediatric ALL than among the siblings; (2) the siblings rate poorer vitality and mental health than the norm, suggesting an increased vulnerability to ill health than the norm; (3) social support is potentially a more important buffering factor for poor mental health and low vitality than self-efficacy among the siblings. Previous studies support our finding of lower HRQoL among YA ALL survivors, and the importance of self-efficacy as a potential buffering factor. However, our study adds knowledge about the importance of the buffering factors in relation also to the siblings.

Between 14 to 27 years had passed since diagnose among the YA ALL survivors included in this study and they were at the time of the study 23 to 41 years old. Even though many years had passed, differences in some of the health parameters, and differences in the importance of buffering factors were seen between the YA ALL survivors and the siblings. These results recognize the relevance of developing and providing support to these individuals long-time after the end of treatment and even into adulthood to promote their health outcomes and successful introduction to adult life.

A strength of this study is that all participants were adults when this follow-up was conducted. Most other studies do not have this long-term perspective but focus on HRQoL in younger survivors, where shorter period of time has passed from diagnosis [9, 13, 35, 43]. Further, the group consist of survivors among whom a mere part had received radiation therapy (only 4.4% had radiation therapy). The results from this study adds knowledge on how YA ALL survivors may be vulnerable to factors related to ill health also in their adult life. Another strength of this study was the possibility to relate the HRQoL of YA ALL survivors not only to a group of siblings, but also to normative values from a Swedish population, thereby allowing for comparisons to the HRQoL in the general population that is not potentially affected by experiences from growing up together with a brother or sister with a serious illness.

There are however also a few limitations with this study that needs to be addressed. Due to practical reasons for data collection, the siblings were much fewer than the YA ALL survivors which complicates the comparison between the two groups and the possibility to reach significance in the regression analysis. Further, in a cross-sectional study like this, the results can not reveal if any of the factors investigated have any causative relations or predictive value. Future longitudinal studies should further investigate the buffering effect self-efficacy and social support and their relevance as predictors of long-term health outcomes and causal relationships.

The results from this study suggest that buffering factors, like social support and self-efficacy, may play an important role for psychosocial outcomes and HRQoL among YA ALL survivors later in life and that this group could benefit from continuous support in adult life to handle consequences of their pediatric disease. The study highlights the importance of further studies with long-term follow-up of psychosocial outcomes and HRQoL among YA ALL survivors and investigation of potential buffering and predictive effects of self-efficacy and social support.