The Physical Health of Caregivers of Children With Life-Limiting Conditions: A Systematic Review

Published empirical studies were included if (1) they reported the physical health of a caregiver and (2) the caregiver cared for a child with a LLC or LTI listed in the Hain Directory of LLCs and (3) the child was <25 years old.²⁹  Articles were excluded if (1) they did not describe the physical health of a caregiver; (2) caregiver was caring for a child who did not have a LLC or LTI; (3) caregivers were bereaved at the time of data collection; (4) the study was not peer-reviewed; (5) the publication did not describe primary data collection (ie, opinion and systematic and topical reviews were excluded); (5) the study scored below the mean on the Hawker quality checklist.³⁰ 

The search strategy was designed to identify research conducted on the health of caregivers caring for a child with a LLC or LTI. However, determining which conditions and diseases from which a child could die before reaching adulthood is not straightforward.³¹  Some conditions are not always life-limiting, or even life-threatening, but rather have a broad spectrum of severity, including the possibility that a child might die of issues related to the illness, as well as the possibility that the child will reach adulthood and their life expectancy will not be compromised. Because of the diversity and spectrum of severity of pediatric LLCs and LTIs, we developed and piloted a search strategy using terms such as “palliative,” “life-threatening,” and “life-limiting” coupled with disease categories from the Hain Directory. However, the results did not identify the preidentified studies that we knew should be identified by the search.^32–34  We therefore determined that this review required the development of a list of conditions based on disease categories.

The list of LLCs and LTIs was developed iteratively by using International Statistical Classification of Diseases and Related Health Problems, 10th Revision categories identified as pediatric LLC or LTIs^4,29  with the direct input of a pediatric palliative care clinician, as well as data from referrals to the largest UK-based tertiary palliative care team over the preceding 2 years. The search terminology was further confirmed by comparing the search findings to diseases cited in epidemiological research on the leading causes of death among children in the United States and United Kingdom who have LLCs and LTIs.^35–38  Our search strategy is listed in Supplemental Table 4.

The search terms and linking strategy were checked by a medical librarian and tested iteratively for each database to ensure the terminology used matched that used by each database. Once confirmed, the searches were conducted (October 8 and 9, 2018, and updated on August 18, 2020) in Ovid Medline, Ovid Embase, Ovid PsycINFO, and Cumulative Index of Nursing and Allied Health Literature Plus. All searches were limited to articles in English. The full list of search terms and linking strategy can be found in Supplemental Table 4.

As in any systematic review, there is a need to assess the quality and, therefore, the studies’ trustworthiness.³⁹  In this review, we excluded studies with serious methodologic flaws or which lacked full presentation of findings. To assess the quality of studies of different methodologies, we used the Hawker checklist, a 9-item checklist that was designed to accommodate studies of different research methodologies while taking into consideration the study quality.³⁰  This tool has been recommended for use in systematic reviews in which included studies come from various epistemological and methodologic perspectives, as in this review.³⁹  Although there is no accepted cutoff score for inclusion or exclusion of studies using this checklist, in the current review, we calculated the mean of the quality assessment scores for all included studies and used it as a cutoff for inclusion in the review.

Study characteristics including country of origin, sample size, method of sampling, method of data collection, data analysis, and results of the study as applied to physical health and any covariables were extracted from each included article. We also extracted data on methods for measuring caregivers’ physical health; child variables including age, diagnosis, illness characteristics, and stage of the child’s illness (when available); and caregiver variables such as ethnicity, marital status, sex, age, employment status, and household income. All extracted data were imported into an Excel file for further analysis.

Given the diverse and heterogeneous mix of study types and methodologies, in this review we use “narrative synthesis” to bring the study findings together.⁴⁰  As in previous reviews of this sort, we adopted a “textual approach to the process of synthesis, to bring out the commonalities and relationships between the studies and the health impacts across them.”⁴¹  Narrative synthesis is well suited to this review because it “relies primarily on the use of words and text to summarize and explain the findings of the synthesis.”⁴⁰ 

The database searches in 2018 retrieved 102 878 citations, of which 57 272 citations were unique. The database searches in 2020 returned 19 961 citations, of which 12 062 were unique. Titles and abstracts were screened for inclusion by 3 authors (J.H., E.H., and B.C.), resulting in 1249 citations (1054 in 2018 and 195 in 2020) that could not be excluded. Two authors (J.H. and E.H.) critically assessed the full texts of all 1249 studies against the inclusion and exclusion criteria. Any studies for which the condition’s life-limiting nature was not clear were assessed for inclusion by a clinician (J.D.). Ninety-seven studies were eligible and underwent quality assessment by using the Hawker checklist. The mean score on the checklist was 25 (SD = 10). As per the systematic review protocol, only the 81 studies scoring 25 or higher on the Hawker checklist were included in the review (see Fig 1).

The 81 studies represent the experiences of 9381 parental and familial caregivers, of which 7165 were mothers, 979 were fathers, and 89 were other family members (grandmothers and siblings). There were 1148 caregivers for whom the familial relationship was not described (see Table 1 for a full description of included studies).

The children had a range of diagnoses (see Table 1). The most common category was neurologic (cerebral palsy, n = 20; epilepsy, n = 5; muscular dystrophies, n = 4; Rett’s syndrome, n = 4; neuromuscular disorders, n = 1; neurologic disorders of unknown etiology, n = 1; encephalopathy, n = 1; and Angelman Syndrome, n = 1). The second-most common category was oncological and hematologic cancers (n = 29). The remaining 12 studies explored the health of caregivers of children with heart disease (n = 3); studies in which children had a variety of conditions (n = 7); congenital and chromosomal abnormalities such as tuberous sclerosis complex (n = 1) and neurofibromatosis (n = 1); 1 study of epidermolysis bullosa (n = 1) and 2 studies of caregivers of children with cystic fibrosis (n = 2). The search did not identify any relevant studies of caregivers of children with gastrointestinal, renal, or metabolic conditions.

Thirty-one studies (38.2%) were conducted in North and South America. There were 15 studies of European populations (18.51%) and 11 studies originating from Western Asia (13.5%). Eleven studies originated from Eastern Asia (13.5%), 7 studies from Australia (8.6%), and 4 from Africa (4.9%). Two studies explored the experiences of caregivers across multiple countries (2.4%). See Table 1 for further information on the countries included in each study.

We find the methodologies used to measure physical health varied between studies (see Table 2). The majority of the studies (n = 62) explored physical health via prevalidated, self-reported questionnaires designed to measure health-related quality of life among the general population.

Qualitative methods such as interviews, focus groups, and participant observation were the second-most common group of methodologies included in this review (n = 14) (see Table 1). Researchers in 5 studies employed objective measures of health; those in 2 used activity trackers^42,43  and those in 3 studies explored immune function through salivary or blood cortisol measurement.^44–46  Of the aforementioned studies, researchers in 2 evaluated interventions^42,47  and those in 8 used mixed methods.^{15,43,44,47–51}  Table 2 outlines the different measures of caregiver health used by the included studies.

An overwhelming number of studies found caregiver health was directly and negatively impacted by the nature of the care they provided to their child (n = 68, 84%). When explored according to the health measure, 28 studies^{26,48,49,51–76}  employing the 36-Item Short-Form Health Survey (SF-36) demonstrated that physical health scores were below average population norms, control scores or, in the case of longitudinal studies, showed a reduction in health over time.

Of the 11 studies employing the 12-Item Short-Form Health Survey (SF-12),^{14,47,77–85}  all but 3 studies showed a reduction in at least 1 domain of physical health,^47,79,84  2 studies showed a longitudinal reduction in physical health,^80,82  and 5 showed a lower health score compared with population norms.^{14,47,77,78,85}  The remaining quantitative studies measured caregiver’s health using other questionnaires, activity trackers, or cortisol. Of them, 19 studies^{15,42–46,50,83,86–96}  demonstrated that the physical health of caregivers who care for a child with a LLC or LTI was negatively affected.

All qualitative studies (n = 14) demonstrated that caregivers felt their health was impacted negatively by a caregiver.^{8,24,87,97–107} 

One of the most common effects of caregiving was pain. Researchers in 32 studies found caregivers reported pain arising from caregiving.^* With the exception of 7 studies which specifically explored back pain, the location of pain was nonspecific. The studies that specifically explored back pain and its severity found caregivers experienced pain daily.^{58,71,89,90,94,96,98}  In 5 of these studies, researchers suggested pain was a result of the weight of the child,^{58,89,90,94,96}  their high levels of dependency,⁸⁹  and/or the physical nature of the care caregivers were required to provide.^8,48,95,96  Despite the possible connection between caregiver pain and lifting and moving their child, no researchers explored the role of training in lifting and handling. Scant attention was paid to supportive equipment for lifting or moving the child for alleviating pain.

Sleep disturbance was reported as an effect of caregiving in 25 studies.^† Caregiver sleep disturbance was caused by a range of issues, including the child’s nighttime seizures,⁵⁰  abnormal sleep patterns characteristic of the child’s condition,^80,104,106  and/or the need for nighttime care.^8,91,99 

Given the nature of the ways these studies were conducted, it was not possible to determine how much of the sleep disturbance reported by these caregivers was associated with caregiving’s psychological burden. However, a significant minority of researchers found correlations between anxiety, depression, or stress and caregiving, which was sometimes related to sleep disturbance.^49,68,102 

Health-seeking behaviors including attention to a healthy diet, taking exercise, and attendance at the GP are known to lessen ill-health in the general population.⁴⁸  However, these behaviors are underexplored in the literature in this systematic review.

Seven studies explored exercise,^{24,42,52,98,101,104,105}  and 1 study examined the caregiver’s diet.²⁴  One study attributed less exercise and poorer diet to caregivers having less time or fewer cognitive resources to devote to their health or feeling that their child’s needs were more important than their own health needs.⁹⁸ 

Caregiver attendance at primary care for regular check-ups, screening, or immunization was explored in 3 studies.^65,66,98  One study reported that caregivers were aware that they should attend regular health checks but reported caregiving was a barrier to attending these appointments.⁹⁸  There were only 2 studies of interventions (daily walking and yoga)^42,47  to improve physical health in this population, despite recognition that public health initiatives may need to be modified for caregivers.²² 

Authors of 12 studies explored coping behaviors detrimental to health, including self-harm,⁸⁷  the use of alcohol,^{45,92,101,104,105}  illicit drugs,^92,101,105  and tobacco.^92,100,105  Use of alcohol, illicit drugs, and tobacco were explored as coping mechanisms, but the studies did not elaborate on the frequency of use, and no studies reported on interventions to reduce these behaviors. Six studies acknowledged the use of antidepressants.^{14,15,24,44,101,105} 

In 39 studies, researchers examined the larger context (familial, social, health) in which caregiving occurs.^‡ The majority of these studies explored the assistance that caregivers had received either informally from a spouse or other family member or formally through support services for their child. Some studies suggest a positive impact of respite care for caregivers.^§

Researchers in 9 studies reported whether the caregiver had a previously diagnosed chronic medical condition,^ǁ whereas those in 8 studies explored whether caregivers had preexisting health issues requiring regular prescription medication.^¶ Some studies postulate that caregivers who care for their child for a longer time are at increased risk of developing chronic conditions such as diabetes, arthritis, and hypertension.^15,72,95  These findings are in keeping with research with caregivers of children with chronic conditions.²⁵ 

In this systematic review, we synthesized the literature around the physical health of parental caregivers of children with LLCs. Eighty-one studies were included. Eighty-four percent of studies reported caregiving negatively impacted caregivers’ physical health, with sleep deprivation and pain (especially lower back pain) the most commonly reported issues. These physical effects appeared to be mediated by support for the caregiver’s role, including support from a spouse, access to good quality respite care, and professional support for their child’s medical needs.

We note the absence of a straightforward association between the condition with which the child is diagnosed and caregiver health status. Instead, the physical health impact on caregivers was related foremost to the physical characteristics and acuity of the child, including their physical limitations and the child’s age and weight. However, our analysis was complicated by the lack of information presented in some articles as to the severity of the child’s condition. For instance, that some diseases (eg, cerebral palsy) present with a spectrum of severity, and that children’s needs may differ on the basis of this spectrum.

An aim with this review was to identify the methodologies used to assess the physical health of caregivers. We found the tools used to be disparate and varied, using quantitative, qualitative, and mixed methods. Quantitative measures generally explored caregiver health as an aspect of “quality of life” or “well-being.” As such, caregivers’ physical health as a discrete and independent focus of research remains largely under-explored. Uniformly, the qualitative studies revealed that caregivers discussed the negative physical toll that caring for their child took on them, either in the form of sleep problems or physical pain.

The studies included in this review demonstrate that many factors affect the health of caregivers. Despite differences between and across the included studies, when taken as a whole, they highlight the act of caregiving as intense mental and physical work involving complex medical technologies, taking place in the home and often throughout the night. However, no study in this review looked at the multiplicity of factors impacting caregiver physical health completely and comprehensively. Table 3 describes what is known about the contextual factors leading to reduced caregiver physical health, as described in this systematic review. This article confirms that caregivers of children with LLCs and LTIs report higher than average levels of physical pain and disrupted sleep.

The economic, familial, and social contexts in which caregiving takes place have been shown to shape the caregiving experience for caregivers of adults.²²  Our findings resonate with a recent study of parents caring for children with life-limiting neurologic conditions. The provision of formal respite care mitigates the negative impacts of caregiving, particularly as the intensity of their child’s care needs make maintaining casual social support networks difficult.¹¹⁶ 

However, this review finds a paucity of literature around societal, economic, and policy factors influencing the physical health of caregivers (including family size, religion, ethnicity, language, socioeconomic status and social class, and availability of targeted services to improve caregiver health, including public health interventions) despite the clear need for attention to these factors in the development of health improvement initiatives.^117,118  Also, in this review, we found few studies that considered the caregivers’ preexisting, diagnosed physical health. We have set out these factors to demonstrate the interconnected relationship between various aspects of the caregiving experience in Table 3. These contexts include the presence or absence of family or spousal support, access to health insurance, respite care for the child, caregiver age, acuity, and child behavior.

Studies report caregivers approach their own health in a variety of ways, including health-seeking and preventive health behaviors (eg, exercising, healthy eating, attendance for routine screening, etc) and behaviors that, if performed to excess, could lead to poorer health outcomes (smoking, drinking, alcohol use, the use of illicit drugs, etc).¹¹⁹  A small number of studies indicate that lack of time may be a barrier to engaging with health-seeking and preventive health behaviors.^99,100  Further research is needed for the specific circumstances of caregiving that create these barriers.

This study has several limitations. First, we designed a search strategy to capture the many illnesses and conditions from which children could die before reaching adulthood. This required a search sensitive enough to capture the studies on LLCs and LTIs that might not have been tagged as palliative in the research databases while ensuring the number of results returned was manageable. We used broad disease categories in our search, but in doing so, we may have missed studies of rarer conditions or single case reports.

Second, it was not possible within the time frame of this review, or indeed possible given the sheer volume of studies we screened, to perform forward and backward searching for further inclusions. It was also not possible to systematically explore the gray literature for other possible inclusions.

Third, because of the homogeneous nature of the studies’ designs included in this review, identifying a suitable quality appraisal tool was challenging, and indeed it is debated within the literature.³⁹  We used the Hawker checklist to score the studies’ quality to decide if they were methodologically sound enough for inclusion in this review.³⁰  We excluded only those studies that did not meet the quality threshold and thus attempted to limit bias from poorly conducted studies.^39,120  However, excluding articles by quality because of the under-reporting of study finding means that our review may have missed studies with relevant findings. Table 1 presents the Hawker scores to be as transparent as possible to our reader. Of note, only 16 studies were excluded on the basis of poor quality, which is a strength of this body of literature on caregiver health.

We recognize that caregivers of children with chronic illnesses share many of the same challenges faced by caregivers of a child with an LLC or LTI. This review has uncovered that the level and intensity of care directly impacts a caregiver’s health. We also know from previous research that children who are at the end of life require more intense, physically demanding, around-the-clock care, which places intense physical demands on the caregiver. Understanding the nature of the work required to care for children at home as they approach the end of life is a direction for future research.

Few studies discussed in this review provide data on the intensity and nature of caregiving, the child’s place in their illness trajectory, or whether the illness leads to their death. Thus, it is not possible to pinpoint the exact difference that the life-limiting nature of the child’s condition has on the health of the caregiver. In studies in which these issues are discussed, the literature is too heterogeneous methodologically, and in clinical category, to ascertain the impact of the life-limited nature of the conditions on caregiver health. More research is needed to look at this issue specifically.

A further limitation of the included studies was a lack of information to include in our review on the specific care activities these caregivers were undertaking. We would recommend an explicit statement about the type of work being conducted in any future studies of caregiver physical health and the caregiver’s level of involvement in caring for their child.

We suggest further examination of preventive health care initiatives in this population of caregivers. The majority of the studies in this review were conducted by teams engaged in caring for a child with an LLC rather than the teams involved in the care of the caregivers (eg, primary care settings). A lack of discussion of preventive health care initiatives may be the result of who is conducting this research. Furthermore, we do not know how the various sources of help for caregiver physical health may work together to monitor the health of caregivers providing long-term, around-the-clock care to their child with a LLC.¹²¹ 

For the most part, many of the studies included in this review were conducted in resource-rich countries. Although it was not within the scope of this review to explore the larger socioeconomic context of the families in each study, the socioeconomic context may have a bearing on the ability of a family, or parent, to access and use services, particularly formal support for caregiving, which can potentially mitigate the burdens and adverse effects of caring for a child with a LLC or LTI.

The majority of studies included in this review reported that caregiving had a negative impact on caregiver physical health, specifically related to sleep and physical pain. Support for caregiving (both emotional and practical) seems to mitigate caregiving’s negative impact on caregiver physical health. The rate of caregiver health-seeking behaviors, preventive health care, and screening for health conditions is understudied despite recommendations that caregivers engage in these behaviors.²²  Most crucially, we find interventions such as the national provisioning of assistance for caregivers such as respite care, mechanical and technological aids, and public health and policy initiatives aimed at aiding caregivers with the increased demands of care within the home remain largely unexplored.

As informal caregiving is expected to continue to grow,¹²²  the number of caregivers caring for children with LLCs or LTIs will consequently increase and supporting those caregivers to continue to care for their children in a way that maintains their health will become all the more critical.

LLC

life-limiting condition

LTI

life-threatening illness

SF-12

12-Item Short-Form Health Survey

SF-36

36-Item Short-Form Health Survey