End-stage renal disease (ESRD) is a health problem encountered worldwide, and, according to the international registries, the number of patients with ESRD is increasing globally (Belasco et al, 2006). In Europe, some 564 638 people received renal replacement therapy in 2016, with about 84% of these receiving haemodialysis (HD), 12% receiving peritoneal dialysis and 4% receiving renal transplant (Kramer et al, 2019). In the UK, 7759 persons started renal replacement therapy in 2016, and the prevalence of this therapy increased from 523 per million population in 2000 to 962 per million in 2016 (Byrne et al, 2018). For instance, there was a 14% increase in the number of patients starting renal replacement therapy in 2016 compared with the number of those who had started therapy in 2011 (Byrne et al, 2018). In Jordan, there were approximately 5000 patients with ESRD receiving either HD or peritoneal dialysis by the end of 2015, with a prevalence rate of 709 per million population and an incidence of 800 new cases each year (As-Sayaideh et al, 2013). Despite many significant advances in medical treatment, ESRD remains a serious illness and one of the main causes of morbidity and mortality in Jordan (As-Sayaideh et al, 2013).
Patients with ESRD receiving HD need help and support, which is mainly provided by family members or family caregivers (Belasco et al, 2006). Family caregiving—defined as help, aid and assistance provided to a relative person in need (Brown and Brown, 2014)—has been shown to be associated with many negative outcomes for caregivers, including poor sleep and quality of life (QoL) (Belasco et al, 2006; Çelik et al, 2012), and this, in turn, predicates poor QoL in patients (Al-Rawashdeh et al, 2017). Sleep disturbance and poor sleep quality are common complaints in family caregivers of patients with ESRD on HD (Çelik et al, 2012). Sometimes, the caregivers reported significantly more disturbed sleep and poorer sleep quality than the patients themselves (Çelik et al, 2012). Similarly, poor QoL is also a common complaint among family caregivers of patients with ESRD (Belasco et al, 2006; Shdaifat and Manaf, 2012).
Many studies have examined the associations between these two outcomes in several populations, including ESRD patients and their family caregivers. A correlation between poor sleep quality and poor QoL has been reported among patients with ESRD (Liu et al, 2015; He et al, 2019), family caregivers of patients with cancer (Chang et al, 2007) and older women (Moreno-Vecino et al, 2017). However, these associations remain controversial. For example, Wang et al (2019) found no significant differences in the psychological, social relationships and environmental domains of QoL (World Health Organization Quality of Life BREF; WHOQoL BREF) when they compared the QoL subscale scores of 681 older adults between those who reported sleep disturbance and those who did not. Similarly, de la Fuente-Tomás et al (2016) examined the impact of sleep disturbance on QoL as measured using the 36-item Short Form Survey (SF-36) in a sample of 119 euthymic bipolar disorder patients with and without sleep problems. They found a statistically significant difference only on the emotional role functioning subscale between those with and without sleep problems, in that those with sleep problems showed higher negative emotional role interference (de la Fuente-Tomás et al, 2016). In their study involving 80 family caregivers of patients with Alzheimer's disease and Parkinson's disease and 150 healthy controls, Cupidi et al (2012) found that good sleepers (global Pittsburgh Sleep Quality Index (PSQI) score ≤5) reported higher scores in total QoL (measured using the McGill QoL Questionnaire) and its domains of psychological symptoms (health-related domain), existential wellbeing and support (the non-health-related domains) compared with poor sleepers. However, they did not find differences in the health-related domains of physical wellbeing and physical symptoms (Cupidi et al, 2012).
Few studies have evaluated the associations between sleep quality and QOL for the family caregivers of ESRD patients receiving HD. To the authors' knowledge, this has been examined in only one study. Çelik et al (2012) found in Turkish family caregivers that poor sleep quality, particularly as indicated by the PSQI subscales of sleep latency, sleep disturbances, daytime dysfunction and subjective sleep quality, was a predictor of poor mental and physical components scores of QoL measured using the SF-36.
Overall, there seems to be a lack of information about the relationship between sleep quality and general wellbeing and QoL in family caregivers of those with ESRD. This population has received very little attention in research and in the clinical practice of nursing in many countries around the world.
Accurate understanding of the associations between sleep quality and wellbeing and QoL may assist health professionals, including community nurses, in enhancing family caregivers' sleep, wellbeing and QoL, which may, in turn, improve these aspects in the patients. Family caregiving has substantial economic value in most healthcare systems. For example, the estimated economic value of the family caregivers' contributions in the US in 2017 was approximately $470 billion (Susan et al, 2019). Thus, the importance of supporting family caregivers cannot be overemphasised. The present study aimed to: (1) describe the sleep quality, general wellbeing and QoL in family caregivers of community-dwelling patients with ESRD; and (2) examine the association between sleep quality and general wellbeing and QoL in this group of individuals.
Methods
Design, setting and subjects
This was a descriptive cross-sectional comparative study of 120 primary family caregivers of patients with ESRD receiving HD treatment in Jordan. The subjects were recruited from different HD units across Jordan (units in hospitals affiliated with the Ministry of Health, the Royal Medical Services, and the King Abdullah University Hospital). A convenience sampling method was used. All subjects in these units were considered eligible for participation. Those included were family caregivers aged ≥18 years, with no terminal illness and providing home care to an adult relative who was undergoing HD treatment for at least 3 months and who had no terminal illnesses.
Procedure
The study was approved by the institutional review boards of Hashemite University, as well as the participating hospital units. A trained research assistant screened and approached potential subjects at the participating sites to obtain initial approval. Eligible participants were given information about the study, and consent to participate was obtained from both the family caregivers and patients prior to data collection. Participants were asked to complete the study questionnaires and return them at the end of the HD session or in the next session. They were assured that their participation was voluntary and that they had the right to withdraw from the study at any time without any consequences.
Measures
In addition to the self-reported questionnaires covering the main variables of sleep, general wellbeing and QoL, a checklist developed by researchers was used to collect demographic and background data for both the family caregivers and patients.
Sleep quality
Sleep quality was measured using the PSQI, a self-rated measure that assesses sleep quality in the past month (over a 1-month period). Sleep quality is reflected by measuring seven components related to sleep: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbance, use of sleeping medications and daytime dysfunction (Buysse et al, 1989). Each component score ranges from 0 (no difficulty) to 3 (severe difficulty). The total scores on sleep quality (global PSQI score) are calculated by summing the scores of the seven components, ranging between 0 and 21, with high scores indicating poor sleep quality (Buysse et al, 1989). The PSQI can distinguish between good sleepers (global score<5), poor sleepers (5–8) and those with severe sleep disturbance (>8) (Buysse et al, 1989). The reliability and validity of the PSQI have been established in healthy subjects, poor sleepers and cardiac patients (Buysse et al, 1989; Suleiman et al, 2010). Cronbach's alpha for PSQI in the present study was .81.
General wellbeing
General wellbeing was measured using the 28-item General Health Questionnaire (GHQ-28). The items are rated on a 4-point scale (0= ‘less than usual’ to 3= ‘much more than usual’). Four subscale scores are calculated by summing the responses of seven items for each subscale of somatic symptoms, anxiety and insomnia, social dysfunction and severe depression (Goldberg and Williams, 1988). Each subscale score ranges between 0 and 21, with high scores indicating worse general wellbeing in that particular subscale. The scores of the subscales are summed to calculate the total score, which ranges from 0 to 84; higher scores indicate worse general wellbeing. The GHQ-28 has shown good reliability and validity in different Arabic-speaking populations (El-Rufaie and Daradkeh, 1996; Alhamad and Al-Faris, 1998; Farhood and Dimassi, 2015). Cronbach's alphas for GHQ-28 in the present study were .89, .88, .83, .88 and .94 for the subscales of somatic symptoms, anxiety and insomnia, social dysfunction, depression and the total scores, respectively.
Quality of life
QoL was measured using the World Health Organization (WHO) Quality of Life BREF (WHOQoL BREF), a well-known multicultural measure of QoL. It is a 26-item tool that measures four domains of QoL: physical health (7 items), psychological health (6 items), social relationships (3 items) and environmental (8 items) domains (WHO Quality of Life Group, 1998). The remaining two items determine the overall QoL and general health. The scores of the domains are transformed into a scale of 0–100, with high scores indicating better QoL (WHO, 1996). The reliability and validity of the WHOQoL BREF have been supported in Arabic-speaking family caregivers (Dalky et al, 2017). Cronbach's alphas were .89, .81, .75 and .80 for the physical, psychological, social relationships and environmental domains, respectively, while that for all 26 items was .93.
Data analysis
The Statistical Package for the Social Sciences v 22 was used for data analysis, and the significance level was set at <0.05. Descriptive statistics including frequency, percentages, means and standard deviations were used as appropriate to describe the subjects' demographic and clinical variables. Paired-sample t-test and chi-square test were used to compare age and gender, respectively, between included and excluded family caregivers. For the first aim of the study, sleep quality, general wellbeing and QoL scores were presented as means, standard deviations and possible and actual ranges of the scores. For the second aim, the correlations between total sleep quality scores and the total and subscales scores of general wellbeing and the domains of QoL were correlated using the Pearson r correlation coefficients. Then, family caregivers were categorised into three groups, namely, good, poor and very poor sleepers, based on their global (total) PSQI scores. The associations of sleep quality with general wellbeing and QoL were examined by comparing the total and subscales or domains scores of general wellbeing and QoL among the three groups of sleepers using one-way ANOVA with the Scheffe test as a post-hoc test.
Results
Participant characteristics
The family caregivers had a mean age of 39 years. Some 33% of them were single, had less than high school level of education, full-or part-time jobs, a history of chronic illness (hypertension was the most prevalent) or were the patients' spouses. The mean age of the patients was 49 years, 40% were female and 67% were married. Detailed descriptions of the subjects, including caregiving-related variables, are presented in Table 1. There were no significant differences between those who included and excluded family caregivers (excluded caregivers' data not shown), neither in regard to their age (39.18±14.36 vs. 38.94± 15.92 years, p=.932) nor in the proportion of either male or female participants in each group (p=.361).
| Characteristic | Caregivers (n=120) | Patients (n=120) |
|---|---|---|
| Mean±SD** or n (%) | ||
| Age, years | 39.18 ±14.36 | 49.28 ±15.4 |
| Gender* | ||
| Female | 65 (54.2) | 49 (40.8) |
| Male | 54 (45) | 66 (55) |
| Marital status* | ||
| Married | 79 (65.8) | 81 (67.5) |
| Single | 37 (30.8) | 29 (24.2) |
| Widowed/divorced | 3 (2.5) | 5 (4.2) |
| Education* | ||
| <High school | 41 (34.2) | 54 (45) |
| High school | 36 (30) | 38 (31.7) |
| >High school | 40 (33.3) | 22 (18.3) |
| Employment* | ||
| Full-or part-time | 46 (38.3) | 25 (20.7) |
| Unemployed | 72 (60) | 88 (73.3) |
| Health insurance* | ||
| None | 30 (25) | 20 (16.7) |
| Public health insurance | 52 (43.3) | 53 (44.2) |
| Military health insurance | 34 (28.3) | 41 (34.1) |
| University health insurance | 4 (3.3) | 2 (1.7) |
| History of chronic illness, yes | 35 (29.2) | 84 (72.4)† |
| History of hypertension | 19 (15.8) | 63 (52.5) |
| History of diabetes | 11 (9.2) | 30 (25) |
| Type of hospital | ||
| Public | -- | 52 (43.3) |
| Military | -- | 49 (40.9) |
| University | -- | 19 (15.8) |
| Relationship with the patient* | ||
| Spouse | 36 (30) | -- |
| Child | 42 (35) | -- |
| Sibling | 15 (12.5) | -- |
| Parent | 10 (8.3) | -- |
| Aunt/uncle | 3 (2.5) | -- |
| Caregiving duration, years | 6.58±1.5 | -- |
| Caregiving days per week | 5.1±2.7 | -- |
| Caregiving hours per day | 9.1±1.2 | -- |
Sleep quality, general wellbeing and QoL
Table 2 shows the PSQI, GHQ-28 and WHOQoL-BREF findings. The highest (worst) reported mean score was on the sleep latency component and the lowest (best) score was on the use of sleeping medication component. The mean global PSQI score of the family caregivers (6.57±4.42) corresponded with the category of poor sleep.
| Measure | Possible range | Mean±SD | Actual range |
|---|---|---|---|
| Sleep quality (PSQI) | |||
| Subjective sleep quality | 0–3 | .65±.95 | 0–3 |
| Sleep latency | 0–3 | 1.25±.98 | 0–3 |
| Sleep duration | 0–3 | .78±1.08 | 0–3 |
| Sleep efficiency | 0–3 | .67±1.05 | 0–3 |
| Sleep disturbance | 0–3 | 1.4±.69 | 0–3 |
| Use of sleep medication | 0–3 | .63±.89 | 0–3 |
| Daytime dysfunction | 0–3 | 1.17±.83 | 0–3 |
| Total | 0–21 | 6.57±4.42 | 0–20 |
| General wellbeing (GHQ-28) | |||
| Somatic symptoms | 0–21 | 5.56±4.62 | 0–19 |
| Anxiety and insomnia | 0–21 | 6.03±4.67 | 0–18 |
| Social dysfunction | 0–21 | 8.43±3.67 | 0–19 |
| Depression | 0–21 | 3.0±3.89 | 0–18 |
| Total | 0–84 | 23.04±14.21 | 0–70 |
| QoL (WHOQoL BREF) | |||
| Physical health | 0–100 | 64±22 | 6–100 |
| Psychological | 0–100 | 54.89±17.75 | 13–100 |
| Social relationships | 0–100 | 59.71±22.73 | 0–100 |
| Environment | 0–100 | 52.85±16.45 | 13–88 |
The caregivers had a good mean total score on the GHQ-28 (Table 2). The respondents had mean scores ranging between 3.0 (depression subscale) and 8.43 (social dysfunction subscale). They had good or near-normal scores on all but the social dysfunction subscale, for which the score was slightly high. With regard to QoL, the caregivers reported means greater than 50 on the four dimensions of QoL but none were higher than 64, which is considered low (Hawthorne et al, 2006). The highest and lowest reported mean scores were on the physical and the environment domains, respectively.
Association of sleep quality with general wellbeing and QoL
The total and subscale scores of general wellbeing had low to moderate significant positive correlations (r-coefficients =.421–.610, all p-values<.001) with sleep quality scores. There were significant negative correlations (r-coefficients=-.320 to -.512, all p-values<.001) between total sleep quality scores and QoL domains. According to the sleep quality cutoff scores, 43.3%, 30% and 26.7% of the subjects were classified as good, poor and very poor sleepers, respectively. Thus, about 56.7% of the subjects had poor sleep quality (>5 on global PSQI). The general wellbeing and QoL scores differed significantly among the groups of sleepers (Table 3). Scheffe's post-hoc test results indicated that very poor sleepers had higher (poorer) total and subscale scores on GHQ-28 and lower (poorer) scores on the physical and psychological domains of QoL compared with both good and poor sleepers. Very poor sleepers also had lower (poorer) scores on the social relationships and environment domains of QoL compared with good sleepers.
| Good sleepers (n=52) | Poor sleepers (n=36) | Very poor sleepers (n=32) | F test (df 2, 117) | p-value* | Significant Scheffe's test | |
|---|---|---|---|---|---|---|
| General wellbeing (GHQ-28) | ||||||
| Somatic symptoms | 3.71±3.7 | 4.67±3.1 | 9.59±4.9 | 23.39 | <.001 | V>G, V>P |
| Anxiety and insomnia | 3.84±3.2 | 5.3±3.9 | 10.4±4.5 | 29.92 | <.001 | V>G, V>P |
| Social dysfunction | 7.32±3.3 | 7.72±3.1 | 11.03±3.5 | 13.32 | <.001 | V>G, V>P |
| Depression | 1.51±2.1 | 2±2.5 | 6.56±5 | 25.96 | <.001 | V>G, V>P |
| Total | 16.4±9.6 | 19.69±9 | 37.59±15 | 38.02 | <.001 | V>G, V>P |
| QoL (WHOQoL BREF) | ||||||
| Physical health | 73.59±19.3 | 64.41±19.3 | 47.93±19.9 | 17.12 | <.001 | V<G, V<P |
| Psychological | 60.09±17.2 | 57.41±17.5 | 43.59±13.8 | 10.53 | <.001 | V<G, V<P |
| Social relationships | 64.17±22.5 | 60.08±18.6 | 50.43±24.8 | 4.40 | .014 | V<G |
| Environment | 59.48±14.7 | 51.72±16.8 | 43.34±13.9 | 11.31 | <.001 | V<G |
GHQ-28=General Health Questionnaire-28; QoL= Quality of Life; WHOQoL BREF= World Health Organization Quality of Life BREF; G, P, and V = good, poor, and very poor sleepers, respectively; df=degrees of freedom
Discussion
To the authors' knowledge, this is one of the few studies examining the association between sleep, general wellbeing and QoL in family caregivers, including those of patients with ESRD receiving HD treatment. One of the main findings of this study is that more than half of the subjects had poor sleep quality, and half of these individuals had very poor sleep quality (>8 on global PSQI). The mean global score corresponded to poor sleep quality. However, the proportion of respondents who reported poor sleep quality in this study is considerably lower than that in other caregiver populations. Çelik et al (2012) and Avşar et al, 2015) reported that 88% and 63% of the family caregivers of patients on HD in their study had poor sleep quality, and this number was 89% in Chang et al's (2007) study on the family caregivers of those with breast cancer. The differences in these findings could be explained by cultural differences, as well as the fact that the caregivers in the present study were mostly young or middle-aged relatively healthy women (about 70% had no history of chronic illness) who were looking after middle-aged patients of whom 25% had no history of other illnesses other than ESRD.
The family caregivers who participated in our study had mainly good or near-normal total scores on the GHQ-28 and its subscales. However, they had relatively low scores on all domains of QoL. This finding is in line with those of other studies, where family caregivers of patients on HD had poorer QoL compared with the general population (Shdaifat and Manaf, 2012). In addition, in most articles included in Sajadi et al's (2017) review on the QoL of family caregivers of patients on HD, the QoL was lower in all or some aspects, compared with the healthy general population.
Another important finding of the present study is that the caregivers' sleep quality had strong associations with their scores in general wellbeing (total and subscales scores) and all QoL domains. In particular, those with worse sleep quality had poor scores on both general wellbeing and QoL. Poor sleep may influence a person's general wellbeing and QoL by causing fatigue, elevating blood pressure and affecting cognitive ability and mental health, for example, by causing depression (World Health Organization, 2004). In their review, Gayomali et al (2008) reported that the QoL of family caregivers for patients with chronic diseases is poor, since caregiving is linked with anxiety and depression.
This finding is consistent with those of a study on a similar population of family caregivers (142 family caregivers of patients with ESRD on HD), in which the researchers reported that the mean scores of the SF-36 components were negatively correlated with the global PSQI scores (Çelik et al, 2012). Chang et al (2007) found that poor scores on the four domains of the WHOQoL BREF were predicted by sleep-related difficulties in the family caregivers of breast cancer patients. More specifically, scores on the physical domain of the WHOQoL BREF could be predicted by those on the daytime dysfunction subscale of the PSQI; those on the psychological domain could be predicted by the scores on the daytime dysfunction and sleep disturbance subscales; scores on the social relationships domain could be predicted by scores on the subjective sleep quality and sleep disturbance subscales; and scores on the environmental domain could be predicted by scores on the subjective sleep quality and daytime dysfunction subscales. Using a dyadic analysis approach in family caregivers of patients with heart failure, Al-Rawashdeh et al (2017) found that sleep disturbance in this group of people negatively influenced their physical and mental wellbeing.
Strengths and limitations
The present study has many strengths, including the use of well-validated measures of all investigated variables, having a fairly large sample size and classification of family caregivers according to well-examined cutoff global PSQI scores. Additionally, to the best of the authors' knowledge, this is the first study to examine both general wellbeing and QoL simultaneously among family caregivers of patients receiving HD treatment. Further, the results contribute to the body of literature related to sleep quality, wellbeing and QoL by demonstrating the relationship between these variables in this population of family caregivers. The findings highlight the importance of considering behavioural, pharmacological, educational and/or other interventions to improve sleep quality in family caregivers. They may also increase awareness among health practitioners, including community nurses, about the condition of family caregivers and their needs.
However, the study also has some limitations that need to be addressed. Factors other than sleep quality may influence the outcomes of general wellbeing and QoL, and these were not examined here. It is possible that some of the family caregivers' own characteristics, such as having chronic illnesses or having depression and anxiety, may have influenced their wellbeing or QoL or at least affected the relationship of sleep quality with general wellbeing and QoL (Lee et al, 2015). Therefore, studies examining these factors in this population are warranted. However, some researchers, such as Hanson and Ruthig (2012), have found that sleep-related problems are independent predictors of wellbeing after controlling for demographic variables and other characteristics, such as physical health, stress, physical activity and functional ability. This implies that sleep-related problems play a part in poor wellbeing, and that sleep is an important consideration for family caregivers. Nevertheless, future studies should control for such variables.
Self-report questionnaires, especially those that ask individuals to report on the main study variables retrospectively, sometimes affect findings, and objective measurement of sleep would be more reliable. Furthermore, the sample in the present study may not represent this population of family caregivers, as it mostly comprised healthy, middle-aged individuals. Thus, the findings may not be generalisable. Moreover, the cross-sectional nature of the study may hinder the inference of causality in the relationships of sleep quality with wellbeing and QoL. These relationships may be better examined using longitudinal interventional studies. Future studies should also investigate the potential improvements in wellbeing and QoL with improvements in sleep quality.
Practical implications
The findings of this study have many implications for the practice of community nurses and other care providers who deal with patients with ESRD in the community. First, the sleep quality of family caregivers should be frequently and regularly assessed, and interventions should be planned accordingly, because poor sleep quality is associated with poor general wellbeing and QoL. Second, nurses and other health providers should frequently assess the factors that might negatively influence the family caregivers' sleep quality, including caregiver- and patient-related factors. Third, health practitioners, in collaboration with other specialised providers (if required), should identify ways to implement individualised interventions for improving family caregivers' sleep quality. Finally, considering care for the family caregivers should be systematised or at least encompassed in the comprehensive care provided for patients themselves. The barriers to providing care for family caregivers should be investigated and appropriately managed. Caring for family caregivers should be institutionalised, whereby the involved health practitioners can receive training for this purpose. Health providers should spend time with caregivers to discuss their issues. Caregivers should be encouraged to talk about their experiences and problems, and they should be supported in identifying and using available resources.
Community health nurses are well placed to support family caregivers, as they often come in contact with this group in primary healthcare settings and patients' own homes. Having the opportunity to visit people at home enables accurate assessment of the needs and planning for care that fits with this populations' actual needs.
Conclusion
This study demonstrated that family caregivers of community-dwelling patients with ESRD receiving HD have poor sleep quality, and this is significantly associated with poor general wellbeing and QoL. Therefore, frequent assessment of sleep quality (and wellbeing and QoL) and implementation of interventions when needed is vital. This ought to become a routine part of the care and can be done in many settings, including primary care, patients' homes and HD units. To achieve successful outcomes, factors influencing the sleep quality of family caregivers should be identified, and referral should be offered if necessary. Family caregivers should be given the opportunity to express their feelings and talk about their caregiving experience.