Quantifying the burden of caregiving in Duchenne muscular dystrophy

Throughout the Western world, including the US, Australia, Japan, and the European Union, long-term care of disabled or chronically ill patients is predominately provided at home by untrained, unpaid family members [1‐4]. For many caregivers, assisting their spouse, parent, offspring, or other relative in their day-to-day life may be a positive experience, but can at times also be tremendously challenging. Numerous studies have shown that the provision of informal care is associated with serious adverse health effects for the caregiver, including anxiety and depression, impaired immune system function, and coronary heart disease, as well as social isolation, financial deprivation, and even premature death [1, 5‐7]. Existing research, however, focus on the burden on caregivers to elderly patients, predominantly with dementia, with less attention devoted to genetic pediatric conditions, including Duchenne muscular dystrophy (DMD).

DMD is a rare, X-linked, neuromuscular disease characterized by progressive muscle weakening, diminishing functional ability, and serious multisystem complications, with a mean life expectancy of 25 years [8‐10]. As a result of the devastating disease progression, patients inevitably transition towards a state of total dependency, requiring wheelchairs for mobility from their early teens and ventilation support for survival in more advanced stages of the disease. Many patients with DMD also suffer from mental health comorbidities, such as autism spectrum disorder (ASD) and obsessive–compulsive disorder (OCD). The complexity of the disease necessitates multidisciplinary management including regular visits to neuromuscular, cardiac, and respiratory specialists, physiotherapists, and other healthcare practitioners [8, 9].

We have previously reported that caring for a child with DMD is both time-consuming and financially burdensome [11]. Many DMD caregivers terminate their employment or reduce their working hours to find the time needed to care for their sons, and those who do continue to work have markedly impaired productivity with high levels of absenteeism. In addition to forgone income, depending on national policies, affected households also carry substantial costs associated with, e.g., insurance premiums and co-payments for healthcare.

The caregiver burden has been studied in DMD only in small samples from a single clinic or country, in combination with other muscle dystrophies, and/or without stratifying results by disease stage or patient health status [12‐22]. The objective of this multinational study was to complement our previous data on the objective caregiver burden in DMD with estimates of the subjective burden: that is, the impact on health-related quality of life (HRQL). A specific aim was to investigate mental distress among caregivers.

Demographic statistics of the participating caregivers (n = 770) are presented in Table 1. Patients had a mean age of 14 years (median 12 years; interquartile range 9–17 years), 47 % (359 of 770) were wheelchair dependent, 16 % (126 of 770) required ventilation support, and 63 % (486 of 770) were currently taking glucocorticoids. Additional details of the study sample have been previously published [11, 23].

Caregiver burden has received considerable attention in the gerontology literature during recent decades, but relatively few studies have investigated the impact on HRQL in caregivers to children with chronic diseases. Compared with adult caregivers of patients with diseases associated with advanced age, such as dementia or Parkinson’s, parents raising a child with a chronic illness may face even greater challenges as they normally live together and have no choice but to fully take on the caregiver role [1]. In addition, as a result of improved survival in many childhood diseases, including DMD, the duration of informal caregiving has increased considerably, in some indications from years to several decades, with increased morbidity and care needs in later stages of the patient's life [30, 31]. For incurable, terminal, progressive conditions, awareness of the devastating disease trajectory would also be expected to have serious adverse effects on caregiver mental well-being, and for genetic diseases, there may also be aspects of guilt.

The objective of this study was to investigate the subjective caregiver burden associated with DMD. Overall, half of all DMD caregivers in our sample indicated being moderately or extremely anxious or depressed, and we also found the mental health summary score from the SF-12 to be significantly lower than general population reference data. These result confirms previous findings of elevated risks of depression and distress reported in caregivers to children with DMD and Becker muscular dystrophy [15‐17], type 1 diabetes [32], and autism [33], as well as parenting stress in pediatric chronic illnesses [34]. We found the prevalence of anxiety and depression to be comparable over the course of disease progression but strongly associated with the health and mental state of the patient as perceived by the caregiver. Specifically, caregivers to patients in fair/poor compared to excellent health had a sixfold risk increase of anxiety and depression. Comparing patients perceived as very unhappy and happy, we noted a sevenfold risk increase. These data suggest that it may be relevant to screen for anxiety and depression in caregivers to patients with DMD, and that patients’ health and mental status may be helpful predictors of caregiver distress.

To our knowledge, only one study has previously estimated HRQL in DMD caregivers using the EQ-5D, and in contrast to our findings, this study found the prevalence of anxiety and depression comparable to the general population [20]. Possible reasons for this discrepancy include differences in sample size and patient demographics, as the previous study was based on parents to 57 adult patients with DMD (mean age 27 years).

We found overall caregiver HRQL, as measured by EQ-5D utilities representing general population preferences, to be markedly impaired and closely associated with patient health and mental status, but not ambulatory class. Specifically, the mean caregiver utility was recorded at 0.81, significantly lower than the only previous report in DMD (0.87, p < 0.001) [20], and we estimated the mean age- and sex-matched disutility at between 0.06 and 0.30 across the investigated strata (mean in the pooled sample was 0.11). This implies that the loss in HRQL among DMD caregivers was similar to or greater than published disutility estimates for patients suffering from very serious and sometimes rapidly fatal diseases, including lung cancer and schizophrenia (0.11), systemic lupus erythematosus (0.08), and epilepsy (0.07) [35]. Given how insensitive the EQ-5D appears to be to capture impairment in HRQL in many conditions (e.g., breast and prostate cancer, asthma, and myocardial infarction, all with disutilities below the minimally important difference threshold of 0.074) [35, 36], the considerable loss in utility observed in our study is both surprising and noteworthy, and captures the exquisite stresses associated with caring for a child with a chronically disabling and progressive condition, with an invariably fatal outcome. The small change in utility and VAS scores across ambulatory classes indicates that caregivers may find ways to learn to cope with the disease and the increasing levels of disability and morbidity associated with the progression of DMD, and adjust their perception of their own HRQL over time. As shown in Fig. 2, the mean utility and VAS scores were in fact slightly higher among caregivers to late compared to early non-ambulatory patients (although the differences were not statistically different, p = 0.254 and 0.113, respectively), possibly related to additional assistance from, e.g., nurses and other healthcare practitioners when patients become more disabled. Comparing results from the SF-12 and the EQ-5D, it is also worth noting that the former indicated normal physical health, whereas EQ-5D domains relating to pain and discomfort and usual activities were found to be significantly impaired in relation to the general population. These inconsistent results may be explained by differences in the design and scoring of the two instruments, but warrants further research, in particular considering the recognized notion that without mental health there can be no true physical health [37].

Kenneson et al. show that employment outside of the home is a predictor of stress in caregivers to boys with DMD and Becker muscular dystrophy [16]. In line with this finding, across countries, we have previously reported that between 27 and 49 % of caregivers in our sample reduced their working hours or stopped working completely due to their son’s DMD, and we have also estimated the mean number of hours of leisure time devoted to informal caregiving per week at between 33 and 44 [11]. Still, in the present study, depending on their rating of patient well-being, between 12 and 40 % of caregivers replied that they frequently or always felt that they should be doing more for their son. We also found that between 57 and 86 % of caregivers frequently or always felt worried about the future of their child, 26 and 69 % were stressed between the demands of caring for the relative and trying to meet other responsibilities for family or work, and 17 and 62 % that they did not have enough money to take care of their son. Together with our findings that a high annual household cost burden and >25 h of leisure time devoted to informal care per week is associated with anxiety and depression, these data emphasize that lack of resources (i.e., time and money) is an important source of distress in DMD caregivers.

We estimated the mean global ZBI score at 29, ranging from 23 to 41 across investigated strata, which may be compared with estimates from studies in other diseases, for example neuromuscular diseases in general (23) [13], irritable bowel syndrome (22) [38], Alzheimer’s disease (29) [39], obsessive–compulsive disorder (29) [40], and Parkinson’s disease (24) [41]. Although these findings indicate that caring for a patient with DMD is burdensome, it is difficult to further interpret the results as there is no link between ZBI scores and subjective caregiver burden or well-being. For example, to what extent are caregivers burdened by the feeling that their sons are dependent on them, or that they feel embarrassed over their sons’ behavior, or by not having as much privacy as they would like? This would be expected to vary across caregivers, by age, sex, cultural setting, and a range of other factors. In other words, the ZBI scoring algorithm does not take into account that the questions and response categories included in the instrument may have a different impact on the self-perceived burden of each caregiver.

Despite the negative impact on well-being, there is also evidence that being a caregiver to an individual with DMD can be a positive, rewarding experience. For example, Pangalila et al. found that 95 % of DMD caregivers in their sample regarded caring as enjoyable, and Magliano et al. found that 88 % of caregivers to patients with muscle dystrophies had got something positive out of the situation [18, 20]. We did not specifically measure positive aspects of caregiving in DMD, but we found some support for these previous results in the ZBI, where 82 % of caregivers never/rarely wished that they could leave the care of their child to someone else, indicative of the devotion with which these caregivers take on their responsibilities, despite the levels of stress and demand.

Although the observational nature of our data prevents us from drawing conclusions of causality, our results have several implications for health policy. First, given the association between number of hours spent providing informal care and caregiver mental health, respite care and similar initiatives is urgently needed to help reduce the family burden and improve caregiver well-being. Second, our results suggest that many families caring for a person with DMD require increased financial support to help shoulder the considerable cost burden associated with the disease, of particular importance given the association between annual household costs and caregiver anxiety and depression demonstrated in this study. It should be noted however that some family caregivers may increase their spending because they are feeling depressed or anxious (to help improve the situation of their child), not vice versa. Third, psychosocial support for caregivers and families must be improved, especially given the poor coverage of these types of services among patients as described in our previous work [23].

Primary strengths of our study include a multinational sample, comprising patients with DMD only (as compared to mixed cohorts of neuromuscular disorders), of sufficient power to enable meaningful stratification across both disease stage and patients’ perceived health and mental status. We found patient clinical data to be characteristic for the different ambulatory classes and the distribution of age similar for responders and non-responders (not reported), indicating that any discrepancy between our patient sample and the general DMD population would be limited. We chose to recruit patients via the TREAT-NMD national DMD registries, which accept registration on a voluntary basis from patients and families with a mutation-confirmed diagnosis of DMD. Although participation in the registries is family initiated, and therefore more likely to be sought by motivated families, the registry-based approach allowed an unbiased sample to be obtained with respect to attendance at any one clinic or restriction to any one domain with potentially different care practices. Our questionnaire was returned with a mean response rate of 42 % which is comparable to other surveys sent out via this kind of route: indeed it is also worth pointing out that the response rate among those who actually received a study invitation would be expected to be notably higher as a result of, e.g., lost invitations due to recent changes to email addresses and spam filters. This also means that our burden estimates should be viewed as conservative, as very distressed caregivers may not have had time nor energy to participate. Despite some limitations in the methodology therefore we believe that this strategy has given us the best possible chance of a generalizable view of burden of care in DMD and an unprecedented glimpse into a large sample size in a rare disease.

We show that caring for a person with DMD can be associated with a substantial burden and markedly impaired HRQL. Our data underscore the need for healthcare practitioners involved in the medical management of DMD to also pay attention to caregiver mental health, in particular when the health and mental status of the patient is perceived as poor. Our findings emphasize the need for a holistic approach to family mental health in the context of chronic childhood disease.