Factors Associated with Burden in Caregivers of Patients with End-Stage Kidney Disease (A Systematic Review)

Background: Caring for a patient with end-stage kidney disease (ESKD) is highly stressful and can impact negatively on the physical and psychological well-being of caregivers. To accurately assess caregiver burden (CB), health care providers (HCPs) need to identify characteristics associated with an increase in CB. Aim: The aim of this review is to explore CB in caregivers of adult patients with ESKD and to identify characteristics associated with any increase in CB. Method: A comprehensive literature search was completed using five electronic databases. Medline, Embase, CINHAL, PsycINFO, and Scopus. The Joanna Briggs Institute checklist (JBI) was used to quality appraise full text papers included in the review. No time limit for the date of publication of studies was employed, to enable the inclusion of more extensive literature. Results: A total of 38 relevant studies from 18 countries were identified and included in the review. A variety of patient and caregiver factors can impact positively or negatively on CB, including socio-demographic factors of patients and caregivers, disease-related factors, situational and relational factors, environmental factors, and psychological factors. Conclusion: This review provides awareness to HCPs of the important factors associated with CB, when assessing or targeting interventions for caregivers experiencing burden.


Introduction
The population with chronic kidney disease (CKD) is increasing due to the growing prevalence of hypertension, diabetes, and obesity. Globally, CKD has a prevalence of between 11 to 13.4% in the general population, which makes it one of the most common worldwide diseases [1]. When CKD is not properly managed, it can progress to end-stage kidney disease (ESKD), which is the final, permanent stage of CKD, where kidneys are no longer able to function properly to meet the patient's needs. Patients with ESKD require lifelong replacement of kidney function by dialysis or transplantation to survive, and many are unable to meet self-care needs and require support from family or friends who are caregivers. The major responsibilities undertaken by caregivers include managing patients' medical treatments, dietary requirements, and clinic and dialysis appointments [2]. Living with an individual in the advanced stages of CKD, and being the main caregiver is associated with challenges such as depression, anxiety, and increased use of medication for caregivers [3]. Caregiver burden (CB) can be defined as the extent to which caregivers perceive their emotional or physical health, social life, and financial status as deteriorating because of caring for their relative [4].
In the ESKD population, many factors are related to CB, such as the demographic characteristics of both caregivers and care recipients, ethnicity, comorbidity status, cohabiting with the patient, relationship with the patient, and duration of caring [3,5,6]. A recent mixed-methods systematic review explored the experiences of family members and friends who provided support for adults receiving haemodialysis (HD) or peritoneal dialysis (PD) [2]. This review considered the association between caregiver demographics and feelings of burden but did not address other factors associated with CB, such as patient characteristics, relationship duration, comorbidity, culture, and type of dialysis received by the care recipients. Consequently, a comprehensive and comparative overview is necessary to inform researchers and health care professionals of the impact these important factors can have on caregiver experiences. Understanding all of the factors is necessary to determine how health care professionals can provide appropriate and effective assistance to improve the quality of caregiving and reduce CB [7]. The aim of this review was to analyze studies that quantitatively measured the level CB among caregivers of adult patients with ESKD and to summarize the factors associated with CB.

Search Strategy
The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist was used to guide the reporting of the systematic reviews and to improve the quality and transparency of the data included [8]. The search was conducted in February 2019 and updated in December 2020 with assistance from the university librarian and was not limited to a specific time period. Included studies were identified following a search of electronic databases; Medline (1982-2020), Embase (1974-2020), CINAHL (1982-2020), PsycINFO (1809-2020), and Scopus (1985-2020). The following search terms were used: (Dialysis OR h*emodialysis OR Peritoneal Dialysis OR renal failure OR Renal Insufficiency OR kidney failure OR Kidney Diseases) AND (carer* OR caregiver*) AND (Burden OR strain). Terms were searched in the English language. The search also included reference lists contained within review studies and other relevant published reviews. Databases were searched individually and a combined search was subsequently completed. ProQuest RefWorks and Endnote were used to manage references electronically and to remove duplicate studies. The Boolean 'OR' featured aided in broadening the search, while the 'AND' helped narrow the search to identify relevant studies in each database.

1.
Studies including informal caregivers of adult patients with ESKD, either undergoing dialysis or receiving supportive/palliative care.

2.
Studies measuring CB and the factors associated positively or negatively with CB. In all studies included in this review, caregiver burden was defined as the extent to which caregivers perceive their emotional or physical health, social life, and financial status as becoming worse because of caring for their relative [4] measured by burden-specific instruments such as the Zarit Burden Interview (ZBI).

3.
Studies examining informal caregiving of patients with renal transplant (RT) were excluded. 4.
Studies exploring informal caregiving of patients with early stages of CKD were excluded.

5.
Studies measuring CB in informal caregivers of patients with a range of chronic illnesses, which did not differentiate the burden experienced by informal caregivers of individuals with renal disease, were excluded. 6.
Studies published in English were included. 7.
No time limit was employed to exclude studies, to help maximise the number of studies included. 8.
Studies assessing the effectiveness of interventions in CB were excluded. 9.
Primary research studies were included. 10. Reviews, conference abstracts, dissertations, editorials, or researcher opinions were excluded.

Selection of Studies
After duplicate studies were removed, 4023 titles and abstracts from all databases were independently screened for eligibility by three authors: B.A., H.N., and P.O. After initial title and abstract screening, 94 full text articles were read and the content discussed with the review team, to assess suitability for inclusion and to resolve any disagreements concerning inclusion or exclusion. A total of 60 studies were excluded as they did not meet the eligibility criteria. Four additional studies were added, as they had been identified in other relevant reviews, dissertations, or reference lists of included studies [9][10][11][12]. Therefore, 38 articles are included in this review, as illustrated in flow diagram which is outlined by the Preferred Reporting Items for Systematic Reviews Figure 1

Data Extraction
The following data were extracted from each study: identification data (author's name, year of publication and setting), methodological data (study design, number of participants, aim and method of measurement), and key findings (CB levels and factors associated with burden). Data extraction from all the included studies was performed by BA, PO and FA, and results were compared for consistency. Any discrepancies were resolved by discussion and reappraisal.

Quality Assessment Tool
We assessed the quality of the included studies to evaluate their methodological rigor and strength of the evidence they provide. The Joanna Briggs Institute (JBI) quality assessment tool was used to evaluate all studies included in the review [13]. This tool is designed to be used in systematic reviews to assess the methodological quality of a study and to determine the extent to which a study has addressed the possibility of bias in its design, conduct, and analysis. This tool employs a range of criteria to assess a variety of research study designs. The JBI Critical Appraisal Checklist for Analytical Cross-Sectional Studies was used to screen the final list of cross-sectional studies in domains including clearance of inclusion and exclusion criteria, description of the sample, location of the study, validity and reliability of the outcome measures, appropriateness of statistical analysis, identification of confounding variables, and how these have been considered. Each question can be answered as "yes", "no", "unclear", or "NA" (not applicable).
The quality score was calculated by aggregating the number of ""yes" responses to all individual criteria, with a maximum score of 8. A score less than 3 indicates a low-quality study, a score between 3 to 5 a study of moderate quality, and a score of 5 or higher is a study of high quality. Two authors (BA and PO) performed the quality assessment of all included studies, which was subsequently reviewed and discussed by all authors. The authors assigned specific ratings of high, moderate, or low quality to all studies included. Some of the included studies failed to report the standard criteria used to measurement CB, and did not identify the setting in which the study was completed; however, all studies achieved an overall moderate to high quality score with a low risk of bias and none were excluded on this basis. (Table A1 in Appendix A).

Data Analysis, Synthesis, and Presentation
A descriptive narrative synthesis of the findings of each study was performed [14]. This decision was made because narrative synthesis provides a broad overview of relevant information, through a textual approach, which is appropriate when studies are too heterogeneous, to allow for a quantitative summary [15]. We assessed the studies to be too heterogeneous due to inconsistencies in methodologies, sample characteristics, and the instruments used to measure CB. The initial synthesis involved searching of studies, listing, and presenting the findings in tables. Then, the findings were sorted into five thematic categories based on their common characteristic with the frequency of studies within each theme presented. Subsequently, the included studies were summarised in a narrative synthesis, which was performed by one author and discussed and agreed by the review members.

Characteristics of Included Studies
The characteristics of the final studies are shown in Table 1. The search identified 38 studies that met the eligibility criteria. Of the included studies, 35 used a cross sectional design [5,[9][10][11][12] and three provided cross sectional data within other designs; one used a quasi-experimental design [46], one used a longitudinal design [47], and one a mixed method design [6]. Studies included in the review were completed in a variety of countries, including: seven in Turkey [11,[22][23][24]27,33,46], four in the USA [5,18,29,44], four in Iran [31,34,37,39], two in Brazil [25,26], two in Japan [34,41,43], two in Saudi Arabia [9,21], three in China [12,42,45], two in Nigeria [6,17], two in Jordan [16,19], two in India [32,35]; and a single study in Nepal [10], Canada [38], Pakistan [40], Greece [36], Spain [20], Singapore [47], Vietnam [30], and Indonesia [28]. CB was assessed in a variety of individuals with ESKD, including: HD in 21 studies, PD in four studies, both HD and PD in four studies, PD and RT in one study, HD and RT in one study; and PD, HD, and RT in three studies. Three studies investigated ESKD without specifying if subjects are dialysis dependent or not, and a single study included caregivers of both dialysis-dependent and non-dependent patients. All the reviewed studies were published between 1997 and 2020, see Table 1.  CB score was found to be higher in caregivers who reported higher level of psychological symptoms (anxiety and depression). CB score was found to be highest in caregivers of patients receiving HD, and it was significantly higher than PD and RT groups. The gender, age, occupation, marital status, and education level of caregivers and patients were not associated with ZBI score. ZBI score was also not associated to caregiver relationships to the patient, living in the same house, or whether caregivers take all responsibility alone or not. Longer duration of HD affected the ZBI scores positively.   High CB before and after transplantation using CSI CB was not associated with age, sex, or relationship duration even though older females reported high CB but did not reach statistical significance. High levels of patient physical health associated with lower CB. CB was higher when the patient was on HD than not yet on dialysis. Higher CB was associated with worse patient health and lower mental QOL. No differences between CB of patients before and after RT.   The most common tools used to measure CB was the ZBI [48] in 29 studies, Caregiver Burden Scale (CBS) [49] in four studies, Novak and Guest Care Burden Questionnaire in three studies [50], with single studies using the Oberst Caregiving Burden Scale (OCBS) [51], Caregiving Stress Appraisal (CSA) [52], and the Caregiver Strain Index (CSI) [53].

Level of CB
In this review, 38 studies investigated the level of CB in advanced stages of CKD patients and reported contrasting findings. Three studies reported high levels of burden [5,6,16]. The majority of the included studies reported moderate to severe levels of burden in caregivers of patients undergoing HD [9,10,18,20,21,27,30,32,34,37,39,46]. Several studies reported moderate CB [11,19,25,26,31,36]. Three studies reported moderate CB in the caregiver dialysis group and low CB in caregivers of RT patients [22,23,35]. Mild to moderate burden was reported in caregivers of patients receiving HD [12,17,33,40,42,43,45,47] and patients receiving PD [27]. One study reported mild to moderate CB in caregivers of patients receiving PD, HD, and RT [24]. Several studies reported mild to no burden in caregivers of patients receiving HD [28,38,41] and PD [44]. Low burden was also reported in one study of dialysis dependent patients, without specifying whether patients were HD dependent or PD dependent [29].

Factors Associated with CB
The factors associated with CB were sorted into five thematic categories based on their common characteristic, with the frequency of studies within each theme presented, see Table 2. The content of each category is described as follows: 1.
Caregiver's and patient's socio-demographical factors including age, gender, ethnicity, marital status, education, employment, income, ethnicity and race, and religion/spirituality.

2.
Disease-related factors including treatment modality, frequency of weekly dialysis sessions, duration and frequency of dialysis, duration of illness, comorbidity, level of patient's dependency, and quality of Life (QoL).

3.
Situational and relational factors including cohabiting status, relationship to the patient, the duration of caregiving, contact time per week, duration of relationship with patients, number of children, smoking and alcohol consumption, and caregiver performing exercise.

4.
Environmental factors including social support.

5.
Psychological factors including depression and anxiety.    HRQoL = health-related quality of life, HD = haemodialysis, PD = peritoneal dialysis, RT = renal transplant, n = factor explored but did not show a significant association to CB, s = factor explored and showed a significant association to CB.

Marital Status and CB
Twelve studies examined the relationship between caregiver marital status (being single, married, widow or divorced) and reported CB [6,10,11,16,24,30,31,33,34,37,39,46]. Two studies reported that single caregivers experienced higher levels of burden when caring for ESKD patients than married caregivers [34,46]. These findings are discussed by Mashayekhi et al. [34], who reported that being single played a significant role in some CB components, including disappointment, lack of freedom, financial difficulties, and feelings that life is unfair. However, younger caregivers reported being confident in their ability to provide care and cope with patient problems [34]. Shakya et al. [10] reported that widowed caregivers reported higher CB levels than those who were married, with the lowest CB in single caregivers.
The majority of studies reported that there was no relationship between CB and the marital status of caregivers [6,11,16,24,31,33,37,39]. No association between CB and the patient's marital status was found [16,24,34,37,39]. One study investigated marital satisfaction levels reported by caregivers, with no relationship identified [34].
Twelve studies examined caregiver income as a predictor of CB [11,12,16,22,27,28,30,31,34,42,45,46]. Lower caregiver income was found to be a factor in higher CB [11,28,34,42], while no association was reported in the remaining studies. Additionally, five studies explored patient's income as a contributor to CB, two of these studies reported that low patient income was associated with higher CB [12,28], while the other studies reported no association [16,26,39].
Thirteen studies examined the association between CB and the employment status of caregivers, including employed, unemployed, retired, full-time, and part-time work [6,11,12,16,24,26,30,31,34,36,37,44,46]. Two studies suggested that feelings of bur-den were greater in unemployed caregivers [12,37]. However, Cagan et al. [11] reported that higher CB was more evident in caregivers who were employed. Four studies examined the employment status of both caregivers and patients and found no correlation with CB [16,24,26,34], while one study reported no association between employment status of caregivers and CB [30].

Ethnicity, Race and Religion
A small number of studies have investigated ethnicity and the race of caregivers in patients with ESKD and reported no correlation with perceived CB [6,44]. Only one study examined caregivers who identified with specific religious groups (Islam and Christianity). No correlation with CB was found [6]. The spiritual well-being of caregivers was negatively associated with CB [37].
CB is likely to increase when a patient's physical health deteriorates and patient functional independence declines [5,31,37,40,42,45]. In addition, CB appears to increase when caregivers also experience poor physical and mental health [18,44]. Two studies investigated the duration of illness and reported no relationship between illness duration in patients with ESKD and CB [9,21]. Two studies explored the relationship between CB and HRQoL in caregivers and reported a negative correlation to CB [5,41]. Three studies reported a negative association between CB and QoL [31,44,45], and one study reported no association [17].

Duration and Frequency of Dialysis, and Treatment Modality
The duration of dialysis in care recipients has been explored and found to be positively correlated with the level of CB [11,24,40,45]. A positive association between the frequency of dialysis and CB also was found [37]. However, other studies reported no association between CB and dialysis duration [10,16,26,31,34,37,39] or dialysis frequency [10,11,34].
Several studies investigated treatment modality and whether patients were dialysis dependent or not, in relation to reported CB [5,[20][21][22][23][24][25]27,[35][36][37]44]. CB scores were found to be higher in caregivers of patients receiving HD compared with caregivers of patients receiving PD [24,27] and RT groups [23,24,35], as well as for patients who were not yet receiving dialysis [5]. Caregivers of patients with RT reported lower CB than patients receiving peritoneal dialysis [22]. The CB level was lower in caregivers whose patients had a kidney transplant history than the ones who did not [37]. On the other hand, several studies reported that CB levels did not differ by dialysis dependency (dialysis dependent or not) [36,44], type (PD, HD, home or in-centre HD) [20,21,25,44] or before and after RT [5].
In five studies, caregivers who were spouses of patients were found to have a comparatively high level of CB compared to those with other types of relationships [10,12,26,39,43]. Three of them have reported that high levels of CB can also be observed in parents of patients [10,12,43].
On the other hand, two studies of caregivers of patients receiving HD in Turkey reported that spouse caregivers have less burden compared to other family relatives such as sons, daughters, siblings, grandmothers and grandfathers [33], daughters, daughtersin-law, and siblings [46]. However, a number of studies suggested that the nature of the relationship with the patient did not predict CB [6,11,19,24,34,44]. A single study reported a positive correlation between the duration of the relationship between caregivers and care recipients and caregiving benefits [5].
Only one study highlighted that caregivers living with patients, irrespective of the relationship, experienced greater CB than relatives who live in a separate residence [19,45], while four studies reported no effect [6,11,24,26,42]. The place of residency (either city or rural area) has been investigated in patients receiving in-centre HD and found that those living in a city location had lower CB scores than those in rural areas [33,46]. However, one study reported no association between the place of residence and CB [11].
A single study investigated a variety of variables including: the number of children, smoking habits, and alcohol consumption, and reported that caregivers having three or more children have higher CB, with no changes in CB associated with smoking and alcohol consumption [11]. One study identified a positive association between CB and the number of medications used by patients [20]. One study found a lower CB in the caregivers who exercised for more than one hour a week [45].

Environmental Factors
Caregivers who seek social support from family and friends experience less CB than caregivers without strong support networks [10,19,20,42]. Social support diminished the impact of emotional CB and stress by providing a solution to the problem, by providing distraction from the problem or by facilitating healthy behaviours needed [10,19,20]. The studies did not report how social support can reduce the physical burden of care.

Psychological Factors
All studies that investigated the relationship between caregiver depression and anxiety and CB reported that higher levels of psychological symptoms are correlated with increased CB [10,17,24,36,38,43]. Caregivers of patients who experience depressive symptoms reported higher levels of CB [42].

Discussion
This review synthesised the current evidence related to CB and the factors associated with CB in patients with ESKD. Informal caregiving research has largely focused on patients with cancer and mental illness, including dementia and Alzheimer's disease, with limited research on caregiving in patients with renal disease. This systematic review includes research from a range of countries. Cultural values and social patterns in these countries are presented in the review, which allows judicious generalisation of the findings to relevant cultural contexts. However, there are several limitations of this review that should be acknowledged. Some potentially relevant studies were unavailable in English [54][55][56][57][58][59]. Additionally, the majority of studies included in the review used a convenience sample, which means study participants may not be representative of their populations. Moreover, due to the range of different instruments used to measure CB, a variety of aspects of CB have been reported. Each instrument used different components to measure CB, for example, the CSA comprises two components: social constraints and physical exhaustion [52]. The ZBI includes items related to consequences of caregiving, guilt or self-criticism, patient's dependence, frustration/embarrassment or anger, emotional reactions and psychological burden, personal strain, and role strain [60], while the CBS contains isolation, disappointment, and emotional involvement items [49]. These differences in the measurement of CB are due to the lack of a conceptualisation or agreed definition of CB in the literature [7,61], which may contribute to differences in study findings [62].
Caregivers of patients with ESKD experience a significant burden, regardless of country of residence. Burden levels in caregivers are likely to be regulated by a wide range of factors. These factors include socio-demographic characteristics of caregivers and patients, disease-related factors, caregiving-related factors, environmental factors, and psychological factors. However, some factors are relatively consistent across studies, whilst for others, findings are inconsistent or inconclusive.

Consistency between Studies
The studies included in this systematic review consistently report that gender, caregiver and patient income, time providing daily care, duration of caregiving, the relationship to patients, and cohabiting arrangements are associated with increased CB. Being a female caregiver is frequently reported as a contributor to higher CB, in patients with ESKD. This finding was consistent with the findings of a systematic review conducted on caregivers of patients with dementia [7]. Several factors explain this finding. For example, gender differences and burden may be due to gender roles, with women still largely regarded as the primary caregiver in social and informal care situations [63]. Therefore, women may self-impose the duty of caregiving more readily than their male counterparts. Alternatively, women may more readily voice their experiences of caregiving versus men, which may be aligned to gender roles that typically result in women being more expressive and vocal about their emotions [64]. Accordingly, women are more likely to express negative feelings than men [65,66]. In contrast, in Middle Eastern countries, men are less willing to complain and express weakness due to cultural norms that reinforce that men are strong and able to tolerate stress [67]. Consequently, differences between men and women may in part reflect differences in their willingness to report CB.
This review suggests that individuals within lower socio-economic populations (the unemployed or on low incomes) may experience higher CB. The explanation for this finding is that caregivers with limited income may have a lack of adequate facilities to meet patient requirements, limited access to suitable care and medication, a lack of transportation, difficulty accessing medical facilities, and limited contact with social support organisations [26]. Caregivers of patients with dementia in lower socio-economic groups have also been found to experience higher CB [68].
A longer duration of caregiving and spending a longer time providing daily care were found to be associated with increased CB. These findings may be due to the poor health status of patients that require more time to care for. This is consistent with the findings of Serrano-Aguilar et al. [69] and Conde-Sala et al. [70], who explained that when patients have lower levels of wellbeing, caregivers would be expected to assist in providing further hours of care. Spending hours in caregiving responsibilities may lead to limitations in carrying out daily personal duties [71], and restrictions in participating in social activities [72]. Furthermore, a longer time spent on caregiving might be due to the difficulty of caregiving tasks associated with CB [30].
Living with patients at the same residence was found to increase CB. The same results were also stated by Raccichini et al. [73] and Viñas-Diez et al. [74], who explained that living with the patients, most likely to be spouses, involved constant patient care leading to a greater physical, emotional, and social burden, which increase over time.
Living in a peripheral district was associated with increased CB. Notably, studies reported that caregivers living in rural areas may report higher levels of CB, which may be due to the increased probability that these patients require in-centre dialysis treatment, or the additional physical and financial burden of travelling to dialysis centres. Social support was found to help reduce CB [10,19,20,42]. According to Alnazly [19], in the Kingdom of Jordan, spouses, children, and siblings are all involved physically and emotionally in caregiving, which helps to minimise feelings of burden in primary caregivers. Additional studies are required to identify the association between the numbers of individuals involved in providing care and the level of CB.
In this systematic review, studies reported that disease-related factors are consistently associated with increased CB. Burden for caregivers of those with receiving HD was significantly higher than for caregivers of patients within PD and RT groups. These results can be explained by the essential differences between treatment modalities. Patients undergoing HD typically attend a dialysis centre three times a week and spend a minimum of three hours per session. Unlike patients on HD, patients receiving PD experienced fewer symptoms, less pain, and are better able to maintain their personal lives and social interaction actively [75], which decreases the amount of support required from caregivers and therefore reduces CB.
Higher numbers of comorbidities in patients and caregivers were associated with higher levels of CB. These findings are similar to those reported by informal caregivers of patients with dementia [76]. Caregiving responsibilities and the amount of time needed to provide care were likely to increase when a patient's physical health deteriorated, reducing their ability to perform daily activities and increasing their functional dependence. Caregivers may feel overwhelmed with managing the complex needs of patients with comorbidities, while at the same time they are dealing with the presence of CKD and dialysis treatment leading to increase CB. Furthermore, this review revealed that the number of medications used by patients was associated with CB [20]. Patients may take medications due to the presence of other comorbidities that increase CB.
Psychological symptoms such as depression and anxiety were significant contributors to CB. These findings support earlier research with caregivers of those with dementia, which reported that psychological symptoms, experienced by caregivers, were the main factors that contribute to CB [7].
Zhang et al. [45] suggested that performing exercise for more than one hour a week can reduce CB. Limited studies focus on the caregivers doing exercise to relieve CB. One study focused on the effectiveness of the involvement of care recipients in intradialytic exercise and found that it successfully reduced CB [77].

Inconsistent Findings
A number of socio-demographic factors reported inconsistent findings in relation to CB. The age of caregivers and patients was found to affect CB both negatively and positively. Younger caregivers may be more vulnerable to the challenges imposed when caregiving, and this may result in greater social isolation and financial insecurity [78], and reduced problem-solving skills [46]. On the other hand, older caregivers may be limited physically and mentally, which may influence their caregiving abilities [79]. Both explanations are credible, and these differences may be due to the multinational research included in the review, which considers this issue in a diverse range of cultures. It is clear that culture shapes caregiving attitudes; this conclusion is supported by a number of authors [80,81], but only two studies explored the ethnicity, race, and religion of caregivers, and these found no effect on CB. Further research is required to examine cultural differences and the effect on CB.
Caregivers with low education levels or caring of patients with low education had higher CB scores, with the exception of one study, which reported contradictory findings with more educated caregivers reporting higher CB [46]. The authors of this study suggest that caregivers with higher educational attainment may commit to other responsibilities and have higher expectations for their lives rather than dedicating themselves totally to caring. Additionally, caregivers with adequate health literacy [16] and receive higher levels of education in how to support the care recipients experience less CB [46]. Caregivers who have the ability to gain access to and understand relevant information may use this to help maintain and promote patients' health, and so experience lower CB. Caregivers with adequate health knowledge are prepared more to support their patients and having less concern about performing tasks such as dealing with disease symptoms, preventing infection, and assisting with side effects [82]. This review highlights that single caregivers experience greater feelings of burden, in contrast to one study, which suggested that single caregivers reported the lowest CB [10]. Burnley [83] argues that having a spouse (who is not the patient) who can provide support in times of stress can reduce feelings of burden. Married caregivers may receive support from partners, a resource that single caregivers cannot avail of [83]. The author suggests that marriage can be a source of solace and support and can help to reduce burden [83]. In this review, only one study reported that being married was a significant predictor of higher CB in patients with ESKD [30], which was similarly demonstrated in patients with dementia [84]. Marital status has been investigated extensively in CB literature, but marital satisfaction levels may be a key to positive caregiving outcomes, as more cooperative couples may engage in more adaptive behaviours and positive self-management [85]. In this review, only one study investigated marital satisfaction in relation to CB, and found no correlation, although this may be due to a small sample size [36]. Additional research is required to identify if any association exists between marital satisfaction and CB.
The study of caregiver relationship to patients reported some varied findings between studies. Five studies report that spouses experience higher, while two studies suggested that parent caregivers reported high CB. The findings in this review and in additional studies demonstrate that close family members are more likely to experience higher CB than other relatives or unrelated individuals [86,87]. This may be because family members are more emotionally involved in their duties as caregivers, and so feel obliged to take care of the family member, even when their personal well-being is compromised [87].
Spouses in particular may experience a greater sense of responsibility, which intensifies the emotional content of caregiving, leading to a greater burden [88]. The challenges facing the spouses who are being primary caregivers and the expectations from them are different from any other relative caregivers. The challenges included managing a drop in income, dealing with additional parenting responsibilities, and the lack of intimacy or reciprocity within the marital relationship [89]. On the other hand, two studies conducted in Turkey reported lower CB in spouses [33,46] and suggested that this is influenced by Turkish culture where spouses perceiving the caring role as a normal task rather than a burden [33]. It is clear from these studies that findings cannot be generalised to different contexts, in light of the differences in caregiver relationships across different regions. For example, more traditional values make it clear that daughters are the preferred caregivers in Middle Eastern countries [46], while the spouse is judged to be the most appropriate caregiver in Canada [38]. This suggests that there may be a disparity in CB in patients across a wide range of cultural settings.

Conclusions
Increased reported CB is associated with female sex; carer anxiety, depression, and ill-health; caring for patients receiving HD and with poorer health; spending longer time giving care; lower socio-economic status; and living a significant distance from a dialysis centre. A longer duration of relationship between caregiver and patient, increased marital satisfaction, and social support from other family members may provide protective effects. Understanding all the of the factors is necessary to determine how health care professionals can provide appropriate and effective assistance to improve the quality of caregiving and reduce CB. The current evidence suggests that the differences in the levels of CB between studies might be due to the influence of cultural variables, which need further investigation. Further research is needed to clarify the relationship between CB and caregiver age, educational level, marital and relationship status, together with the influence of cultural norms on the caregiver role and experience. Those supporting caregivers of patients with ESKD should recognise that most caregivers will experience at least moderate CB, and that this may increase as the patient's condition deteriorates. Assessing CB should be a regular component of care for people with ESKD and caregivers with characteristics associated with increased CB may need targeted additional support. This is necessary to avoid early nursing home placement, reduce the adverse health outcomes for patients, and prevent the deterioration of caregivers' health. HCPs should inform caregivers to consider seeking social support, performing exercise, and obtaining an adequate education in how to support the care recipients when experiencing burden. Acknowledgments: The authors thank Hugh O'Donnell for refining the writing style and improving the manuscript significantly.

Conflicts of Interest:
The authors have no conflict of interest to declare.