Parent-reported health-related quality of life in pediatric childhood cancer survivors and factors associated with poor health-related quality of life in aftercare

Childhood cancer is a serious disease that affects more than 2000 children and adolescents each year in Germany [1]. Among children, the most frequent cancer diagnoses are leukemias and CNS tumors [1]. Thanks to advances in cancer treatment and improved diagnostic methods, the 15-year survival rate is 82% across all pediatric cancer diagnoses in developed countries [1, 2]. Despite the improved survival rates, cancer and the corresponding multifaceted treatment can have far-reaching consequences for a child, causing physical and mental burden, even after the end of treatment [3]. In Germany, every pediatric childhood cancer survivor continues to attend regular oncological aftercare until the age of 18, even after the acute aftercare has already been completed. Especially with more time after the end of treatment, many clinics provide late effect-oriented aftercare. Implementation is facilitated by children's hospitals, which integrate many specialties in addition to pediatric oncologists [4].

To measure children’s health outcomes, general well-being and everyday functions, the multidimensional construct health related quality of life (HRQoL) was utilized [5]. According to the international guidelines, it is common to assess HRQoL in childhood cancer survivors’ aftercare [6]. Nevertheless, HRQoL is less focused in aftercare than in acute care and empirical evidence is not entirely consistent. While a review by Shin and colleagues reported similar or even higher HRQoL scores for most survivors compared to healthy controls [7], various other studies illustrated the remarkably complex and powerful impact of childhood cancer on psychosocial and physical HRQoL [8‐11], even in long-term childhood cancer suvivors [12, 13]. Additionally, a systematic review showed that childhood cancer survivors experience lower levels of mental well-being, reduced positive mood, and lower self-esteem as well as increased anxiety and physical and sleeping difficulties compared to healthy controls and population norms [9].

In Germany, as in most Western countries, parents are surrogates for their children as long as they are not old enough to make autonomous decisions [14]. In aftercare of childhood cancer, the parental role is important, and further implications for children’s health care often result from parental involvement [15]. Due to the different roles of mothers and fathers in coping with the disease, they might perceive their children’s HRQoL during aftercare differently. Heterogenous perceptions may result in different implications for aftercare. Therefore, it is necessary to identify how mothers and fathers assess their children’s HRQoL.

Early identification of vulnerable children with a high risk for decreased HRQoL during aftercare is essential to allow health care professionals (HCPs) to offer preventative care and help where it is needed. In Germany, pediatric cancer survivors and their family members can participate in a 4-week family-oriented rehabilitation (FOR) stay to recover from disease and to regain psychological and physical well-being [16]. Several risk factors for poor HRQoL in children and adolescent long-term childhood cancer survivors after completion of cancer treatment have already been identified: Type of cancer and a higher number of applied treatment modalities, chronic symptoms, late effects, and demographic differences including higher age, female gender, lower education, and lower household income [7, 13]. Survivors of central nervous system (CNS) tumors reported poorer physical and psychological functioning compared to leukemia survivors [17]. However, the studies primarily focused on either long-term survivors or patients immediately after and during treatment, while studies focusing on the aftercare period (up to 5 years after end of treatment) are scarce.

Hence, this study aims to investigate how mothers and fathers assess their child’s HRQoL, assuming that fathers rate their child’s HRQoL higher than mothers [18]. Additionally, the objective of the study was to evaluate risk factors for poor HRQoL in children 2.5 years after being diagnosed with cancer and during aftercare when survivors are followed-up regularly. To provide information on vulnerable patients at an early stage after completion of cancer treatment and without high additional effort, we focused on sociodemographic and disease-specific factors. Moreover, we focused on the most frequent cancer diagnoses in children, leukemias and CNS tumors [1]. Thus, the present study contributes to identifying high-risk patients at an early stage. At the same time, it clarifies potential differences in parental assessments regarding child HRQoL after surviving cancer.

There were 899 families potentially eligible to participate in the study. If possible, we included both parents within a family. Five hundred and twenty-seven families that had been identified as eligible via the study registries could not participate either because they met the exclusion criteria or because the HCPs in the clinics could not inform them about the study. Reasons for non-participation for the remaining 60 families that had been identified as potentially eligible via the rehabilitation clinic were: No interest (n = 21), insufficient German language skills (n = 14), physical/and or mental burden (n = 12), cognitive limitations (n = 3), or not specified (n = 10). A total of 312 families participated in the survey. Due to missing signed consent forms (n = 2), missing parental questionnaires (n = 2), a wrong diagnosis (n = 2), or incorrectly answered questionnaires because of limited German language skills (n = 1), 7 families were subsequently excluded from the analyses. Therefore, due to robust estimators of mixed models, a total of 516 valid parental questionnaires of 305 families at baseline could be included in the mixed models (initial participation rate: 57.4%). In 211 families, both mother and father completed the baseline questionnaires. Two hundred and ninety-four parents of 172 childhood cancer survivors also completed the 12–18 month questionnaires at follow-up and could be included in the t-test analysis.

The purpose of this study was to investigate parent-reported health-related quality of life (HRQoL) in children and adolescents after surviving childhood cancer.

Firstly, we wanted to identify differences between mothers and fathers in their assessment of their children’s HRQoL. Previously, only a few studies investigated parent-reported child HRQoL [26‐28]. However, parent-reported HRQoL of early childhood cancer survivors has not been examined before. Many health-related decisions are made on parental reports because they can provide HCPs with useful information about their children. Thus, especially for younger children, parents’ assessment play a crucial role in health care and health-related decisions [29]. Evidence shows that reports of a child’s HRQoL from parents and children are highly consistent [30, 31]. Therefore, parental reports as applied in the current study can be obtained with reasonable confidence in case a child is too young or too sick to provide self-report. Our results show that fathers rate their children’s HRQoL total score and the condition-specific domains family, friends, and disease higher than mothers. Results are partly in line with research on other chronic conditions such as epilepsy and esophageal atresia: While Kalyva and Melonashi did not find any differences between maternal and paternal reports of child HRQoL, Witt and collegues reported that fathers rated child HRQoL scores in the condition-specific social domain higher than mothers [26, 28]. It is important to consider that mothers of children with cancer might experience higher levels of distress than fathers [32, 33]. Different roles of mothers and fathers in coping with the disease might affect their parental experience and their perception of childs’ HRQoL. Mothers seem to be more involved in supporting their child by accompanying the child during hospital stays, whereas fathers are more likely to stay at home or return to work after a little while [34]. At the same time, mothers’ own well-being and worries may impact their ratings of their child’s HRQoL being lower [33, 35]. Consequently, it can be expected that factors such as parental functioning also have an impact on our findings. Current literature identified modifiable factors associated with parental ratings of their child’s HRQoL. Fear of cancer relapse and fear of late effects, as well as parental resilience, may play a role in reporting HRQoL in pediatric survivors [36]. Parents take an important role in aftercare of childhood cancer and implications for children’s health care often result from parental report [15]. Therefore, it is necessary keeping in mind the differences in parental perceptions when it comes to children’s aftercare after surviving childhood cancer because heterogeneous perceptions could result in different implications for aftercare.

Secondly, we wanted to identify baseline factors associated with a risk for poor HRQoL during aftercare, about 2.5 years after diagnosis. Compared to reference data from the German Child and Adolescent Health Survey (KiGGS), childhood cancer survivors in this study sample were slightly below the average of healthy peers (healthy controls: M = 76.9; childhood cancer survivors: M = 74.1) [24]. In our linear model, family affiliation as random effect explained a moderate level of information on group level (ICC = 0.54). Moreover, our analyses identified the diagnosis of CNS tumor, an older age at diagnosis, and non-participation in a rehabilitation program as risk factors associated with poor HRQoL in children approximately 2 years after the end of intensive cancer treatment. These factors may help HCPs to identify patients at risk early in the disease trajectory. Our findings are in line with previous findings reporting a lower overall HRQoL in survivors of CNS tumors compared to leukemia and other childhood cancer survivors [37, 38]. Authors of the Childhood Cancer Survivor Study also identified adult brain tumor survivors as vulnerable patients for poor HRQoL. Especially, high intense treatments such as cranial radiation and/or surgery are associated with lower HRQoL after end of treatment [10, 36, 39]. However, most of these studies investigated HRQoL of (young) adult childhood cancer (AYA) survivors instead of pediatric survivors [40‐44]. Still, it becomes clear that especially the HRQoL of CNS tumor survivors needs to be more protected. We advocate more early interventions that support survivors individually and on multiple levels, e.g., cognitively, socially, emotionally. Our results indicate that lower HRQoL already shows during the first 2 years after the end of treatment. Additionally, our analyses showed that age was associated with HRQoL. An older age at diagnosis predicted poorer HRQoL at follow-up. It is possible that childhood cancer survivors who were diagnosed in their first years of life hardly remember anything and thus have less burden. Additionally, the treatment of childhood cancers tends to be shorter and less intensive than that of adolescents, and there are developmental and psychosocial aspects of adolescence that the treatment might affect [39, 45]. Our findings display the need for age-related support offers and are in line with previous research that focused on HRQoL and the positive and negative impact of cancer in AYA survivors [39, 46]. Moreover, previous research has found that female sex, lower socioeconomic status, and more time since diagnosis are associated with poor HRQoL in aftercare, whereas we found no association between these predictors and outcome [44, 47‐49]. Our findings indicate that most partcipants have participated in a rehabilitation program. Participation in a FOR was associated with better HRQoL at follow-up in our sample, indicating that FOR is beneficial for survivors and their families. FOR supports survivors and their families in reintegrating into daily life, gaining physical strength, and having peer contact [50]. We can assume that survivors and their families, particularly those at high risk for negative long-term consequences, may also benefit from FOR. However, several barriers can hinder the families’ participation in FOR, including concerns about work or school disruption, the effort involved in submitting an application, cultural and language barriers, as well as psychological barriers [51‐53].

Therefore, it is important to make information about FOR easily accessible and to inform parents and families about the potential benefits of a rehabilitation program for improving HRQoL after intensive treatment ends.

When interpreting the results of our analyses the following limitations should be considered. Firstly, due to data protection regulations, a personal contact of the parents was not possible, and parents were recruited via study registries and a rehabilitation clinic. Within this study design, we could not conduct a non-responder analysis which might limit the representativeness of the sample. Secondly, the current study is a secondary analysis of a main study which focused on the role of rehabilitation measures supporting families in the reintegration into daily life, school, and work. It should be kept in mind that all families were offered to participate in a FOR. Therefore, we cannot control for any selection bias that might affect our results. Thirdly, our outcome HRQoL in childhood cancer survivors was calculated only by parental proxy report. For our second aim, it would be even more valid to obtain data from both informants, the children themselves and their parents. Therefore, it is possible that we under- or overestimated the children’s true HRQoL. However, it can also be considered as a strength of our study that we included children of all age groups and thus also young children who are too young to self-report. Although obtaining reports from two parents is preferable, this was not always possible. Nevertheless, in most cases, both the maternal and the paternal report was available so that we included family affiliation as random effect. Additionally, the study design was predominantly questionnaire-based and appropriate German language skills were necessary to enter the study. Additionally, there were less CNS tumor survivors in our sample who tend to have an increased risk for higher burden compared to leukemia survivors because of cranial radiation [54]. At the same time, the review by Cantrell and collegues shows that female childhood cancer survivors have a higher risk for poor HRQoL compared to male childhood cancer survivors [54]. Therefore, we cannot account for these selection biases and over- or underestimation of burden is possible [55]. Lastly, due to the small number of families who did not participate in a FOR, results regarding the impact of the rehabilitation program can only be interpreted to a limited extent.

Despite the several limitations, the findings of the present study are important in the context of providing health care for pediatric childhood cancer survivors in an early phase of survivorship. Due to the comparatively large sample size, the longitudinal design, and recruitment from both a rehabilitation clinic and study registries, our results can be considered as reliable. They showed differences in parental reports of their childs’ HRQoL, with fathers rating their child’s HRQoL higher than mothers. Additionally, our sample provides evidence based on a large sample of childhood cancer survivors that several risk factors such as the diagnosis of CNS tumor, an older age at diagnosis, and non-participation in a rehabilitation program are associated with a poorer HRQoL in children after surviving childhood cancer.

It is necessary to keep in mind identified differences in parental perception, because parents take an important role in aftercare of childhood cancer and their perceptions can implicate children’s health care decisions. Early identification of patients with several risk factors is necessary to initiate adequate support offers during aftercare. HCPs and families should be educated and made aware of risk factors. Moreover, support should be offered to enable informed decisions in favour of pediatrics HRQoL after surviving childhood cancer.