The impact of childhood cancer on parents' socio‐economic situation—A systematic review

Taking care of children diagnosed with cancer may have considerable consequences on parents' socio‐economic situation. Our systematic review aimed to evaluate and synthesise the evidence on the impact of childhood cancer on parents' socio‐economic situation.


| BACKGROUND
Childhood cancer is a devastating experience for the whole family system with few life events being as far outside a family's routine. 1 This is particularly challenging for the parents who are confronted with the potential fatality of the disease and conflicting caregiving, emotional, and practical demands. 2 The child's acute treatment requires frequent hospitalisations, invasive procedures, and depending on the cancer type, a combination of surgery, chemotherapy, and radiotherapy. 3 Parents are often involved by providing and monitoring treatment or managing treatment-related symptoms, particularly in the outpatient setting. 4,5 Because of the increased risk for late effects after treatment, 6 long-term follow-up care is recommended for childhood cancer survivors. 7 Even years after treatment, many parents remain actively involved in the child's medical care. 8,9 The management of the child's disease alongside everyday responsibilities is highly challenging for the parents. Previous research indicates that parents of children with cancer experience substantial work and income disruptions during the child's treatment. 10 Moreover, many parents are confronted with medical and nonmedical expenditures. [11][12][13][14][15][16] Direct costs of childhood cancer have been evaluated in two reviews concluding that substantial financial toxicity may occur in paediatric oncology. 10,17 However, a comprehensive assessment of parents' socio-economic situation also including aspects related to financial assistance is currently lacking. It further remains unclear whether the socio-economic situation of mothers and fathers is differentially affected because of different parenting roles and tasks. 18 Moreover, evidence on temporal patterns after the diagnosis and socio-demographic or cancer-related determinants of adverse socio-economic consequences is lacking. There may be groups of parents that particularly struggle with their professional life during treatment or to re-establish and compensate work-related disruptions after the child's cure. Identifying parents who are at particular risk of adverse socio-economic consequences is crucial to provide targeted supportive services along the cancer trajectory to reduce these inequities from an individual and societal perspective.
The objective of this systematic review was to critically evaluate and synthesise the evidence on the impact of childhood cancer on parents' socio-economic situation. Specifically, we aimed to address the following research questions: 1. What are the consequences of childhood cancer for the parents' socio-economic situation regarding employment, income, financial situation, and financial assistance? 2. Are there differences in the consequences between mothers and fathers of children with cancer?
3. Are there temporal patterns in the consequences after the child's cancer diagnosis? 4. What are the main socio-demographic and cancer-related characteristics associated with adverse socio-economic consequences?

| METHODS
Our systematic review complies with the PRISMA statement regarding the reporting of systematic reviews and meta-analyses. 19 A review protocol was registered in PROSPERO (number: CRD42018096121).

| Search strategy
Our literature search was conducted on March 23, 2018, and included articles published in peer-reviewed journals after January 1, 2000, that were indexed in the databases PubMed, Scopus, and PsycINFO. This time frame was chosen to account for improvements in cancer treatment protocols over time. The search was updated on January 11, 2019. Our search included four individual blocks with search terms referring to socio-economic situation, parents, childhood, and cancer ( Figure S1). In PubMed, we additionally performed searches using medical subject headings (MeSH). We hand-searched reference lists of included studies to identify other relevant articles.

| Study selection
To select eligible articles, we hierarchically applied the following inclusion criteria: sample size greater than 20, quantitative methodology, parents of children with cancer as main study population, child's age at cancer diagnosis less than 20 years, parents' socio-economic situation as primary outcome. Editorials, commentaries, conference abstracts, and original articles without English full text were excluded.
We excluded studies solely focusing on costs or expenses as the respective literature has been previously reviewed. 10,17 Two reviewers (L.M. and K.R.) independently assessed eligibility by first screening titles and abstracts followed by the full texts of remaining studies. Discrepancies between reviewers were resolved by consensus or consulting a third reviewer (F.E.).

| Data extraction
We extracted first author, publication year, country, study design, sample size(s), and response rate(s). For parents of children with cancer and comparison parents (if applicable), we extracted data on sex, age at study, and other available socio-demographic characteristics.
Socio-economic consequences of childhood cancer regarding employment, income, financial situation (financial burden and material hardship), and financial assistance (governmental and non-governmental) were extracted. We further extracted the following cancer-related characteristics of the child: diagnosis, treatment, diagnostic period,

| Quality assessment
Study quality was independently assessed by two reviewers (L.M. and F.E.) using the Newcastle-Ottawa Quality Assessment Scale (NOS) 20 as recently used in a review addressing childhood cancer survivors 21 (Table S1). NOS evaluates the quality of nonrandomised studies with a star rating system (maximum nine stars) based on three criteria: selection (four items, max. one star/item), comparability (one item, max. two stars), and outcome (three items, max. one star/item). The criterion selection refers to the representativeness of the study population(s) (parents of children with cancer and comparison parents) and the exposure ascertainment (childhood cancer diagnosis). According to the NOS' manual, 20 we defined education as the most important factor to adjust for in a comparison between study populations for the criterion comparability. An additional star was appointed to studies controlling for sex, age, or year of outcome assessment. The criterion outcome refers to type of outcome assessment, length of follow-up, and adequacy of follow-up. A follow-up rate of greater than 70% was considered unlikely to introduce bias. 22

| Data synthesis
Findings related to parents' socio-economic situation were narratively synthesised. A priori, we decided not to follow a meta-analytic approach because of expected heterogeneity related to study design, study period, and outcome definition between studies and differences in socio-economic context across geographical regions. 23 The narrative synthesis focused on the socio-economic consequences regarding employment, income, financial situation, and financial assistance, differences between mothers and fathers, temporal patterns after diagnosis, and characteristics associated with adverse socio-economic consequences. We further evaluated how the quality of included studies may have affected our synthesis.

| Literature search and study characteristics
We identified 3359 articles through literature searches and included 35 articles,  reporting on 29 individual studies ( Figure 1). Thirteen (37%) studies were conducted in Europe, 16             3.2 | Impact of childhood cancer on parents' socio-economic situation

| Employment
A high prevalence of disruptions in parental employment including job quitting, job loss, unemployment, changes in work hours, or extended leaves was reported. Most studies found more profound work disruptions among mothers compared with fathers. Twelve studies reported that mothers were more likely to quit work or to be unemployed after the child's diagnosis. 24,25,28,30,31,[35][36][37][38]44,46,49 Only one study from Indonesia showed higher work loss among fathers. 55

| Income
The majority of studies on income reported substantial income loss after the child's diagnosis. 26,28,31,36,[38][39][40]44,47,[49][50][51][52][53][54][55]58 The proportion of parents reporting income loss and the extent of these losses varied largely. Two studies from Norway 25 and Finland 24 found no effects on income, and one study from New Zealand 52 found more parents reporting income gain than loss. Evidence related to gender differences is limited with two studies showing similar effects in mothers and fathers 25,31 and one study each reporting higher income loss in mothers 53 or fathers 36 . A population-based study from Sweden 26,28 indicated that maternal income reductions persisted until 6 years after diagnosis compared with 3 years among fathers. Findings from longitudinal studies suggest that income losses are most pronounced in the first months after diagnosis, 24,31,39,40 but may persist into early survivorship. 26,28 Lower income at diagnosis, 28,40,44,47 younger age of child at diagnosis, 25,28,38 and diagnosis of leukaemia 25,50 were consistently associated with income loss.

| Financial situation
Parents' financial situation was affected by the child's diagnosis across all studies. The extent of the impact varied largely from 18% of parents reporting a great financial burden in Sweden 32 to 83% in Kenya. 58 Two-thirds of parents reported debts because of the child's disease in studies from Asia and Africa, 53,55,58 and two of these studies additionally reported that parents withheld treatment because of financial reasons. 55,58 Also in the United States, a study showed that 15% of families fell below the poverty level because of cancer-related financial strains. 44 Findings from three longitudinal studies 24,39,40 suggest that the impact on the financial situation peaks about 6 months after diagnosis, and one study from the United States reports that it persisted until 2.6 years after diagnosis. 39 The main characteristics associated with adverse consequences for parents' financial situation were rural residency or greater distance to hospital, 37,42,45,49,51,53 lower socio-economic position, 34,40,47,49,54 and single parenthood. 36,49

| Financial assistance
The different types and extent of financial assistance across studies precluded an overall synthesis. Only one Australian study emphasised that families received no assistance for most cancer-related expenses. 51 Two US studies reported that more than 50% of parents used individual fundraising as a financial coping strategy. 44,47 In Sweden, parents of children with cancer were more likely to rely on sickness or childcare benefits than comparison parents. 26,27 Sick leave was more often used by mothers than fathers. 26,27,31,36 Findings from longitudinal studies suggest that the uptake of such benefits is highest in the first months after diagnosis 24,26,27,31,38 and decreases in early survivorship. 26,27,30,31,38 Non-governmental assistance appeared to be more often received by families with higher expenses or rural residency, 51 whereas the uptake of social security benefits was mainly associated with parents' education, income, and cohabitation status. 26,27

| Study quality
The average quality rating was 5.0 (Table S1). Quality ratings were higher for studies reporting on income (mean = 5.2) than for studies on employment (mean = 4.7), financial situation (mean = 4.6), and financial assistance (mean = 4.7). We identified no conclusive patterns in the reported findings according to study quality. Quality ratings were higher for European studies (mean = 6.5) compared with studies from North America/Australia (mean = 4.3) and Asia/Africa (mean = 4.0).

| DISCUSSION
Our systematic review of 35 articles indicates that having a child with cancer may have a considerable impact on the parents' socioeconomic situation supporting conclusions from earlier reviews. 10,17 We found a high prevalence of disruptions in parental employment, particularly among mothers. The associated income losses contributed to families' perceived financial burden. Socio-economic consequences were most pronounced shortly after diagnosis, however, persisted into early survivorship for certain groups of parents.
We identified families with lower socio-economic position, parents of children diagnosed with haematological cancers, and diagnosed at younger age to be at particular risk for adverse socio-economic consequences.
Disruptions in parents' socio-economic situation varied largely by geographical region. Differences in regional labour market and economic circumstances, social welfare systems including health care services, and the extent of psycho-social support provided may account for this finding. A Swiss study estimated that parents need on average 240 working days for caretaking during the child's treatment. 59

| Study limitations
A limitation refers to the large variations in outcome definition and methodological approaches across studies, which limited betweenstudy comparisons. Our findings mainly apply to high-income countries as studies from middle-or low-income countries are underrepresented. However, a family's socio-economic situation may be of higher concern and public health relevance in such countries in regard to treatment access and health outcomes. Another limitation refers the self-reported information in many studies. This may have resulted in biased responses caused by social desirability with parents tending to present a more favourable image. 65 Finally, the explanatory power of the NOS for appraising study quality is limited as sample size is not considered. 21 This aspect is critical to identify characteristics associated with parents' socio-economic situation as smaller studies may be underpowered.
However, the comprehensive literature search enabled the inclusion of studies from various countries with different socio-economic contexts. The search terms used ensured that a broad range of socio-economic consequences that parents of children with cancer may experience are captured. A major strength of our review refers to the scientifically rigorous methodological approach with searching relevant databases, performing an extensive hand search, and updating our search to include recent articles. Study selection and quality appraisal were performed independently by two researchers.

| Clinical implications
Family poverty has been described as a negative prognostic indicator in paediatric oncology. 40 In 2015, standards for psychosocial care of children with cancer were published 66 including a recommendation for assessing family financial hardship. 67 A follow-up study from the United States outlined that while most paediatric oncology programs could implement some of these standards, lack of monetary resources precludes a comprehensive implementation, 68 and only half of paediatric oncologists and psychosocial leaders agreed that their psychosocial care is state of the art. 69 However, a recent study from the United States evaluating the feasibility of poverty screening in paediatric oncology revealed promising results by assessing household material hardship with a short screening tool in routine care. 70 From a global perspective, further efforts are needed to develop, implement, and systematically evaluate cost-and time-effective screening tools for family financial hardship. Ideally, such screening tools lead to referral to targeted financial counselling and supportive services according the families' risk profile. 67 This is of particular relevance as our review revealed that a majority of parents received financial assistance.
Increasing the awareness of existing support services and guidance in navigating through potential administrative barriers may reduce the parents' burden in the life-threatening context of having a child with cancer.