“It’s Been a Hard Day’s Night”: Sleep Problems in Caregivers for Older Adults

Abstract

Purpose of review

This paper critically evaluates literature from the last three years on sleep in caregivers for older adults. Research is evaluated in four main areas: factors related to caregivers having sleep problems, sleep health in various types of caregivers, how caregivers’ sleep and health inter-relate, and interventions to improve sleep.

Recent findings

A range of both care recipients’ and caregivers’ characteristics have been associated with caregiver sleep quality measured subjectively and objectively. Care recipient factors associated with caregiver sleep quality include fall risk and sleep quality, while caregiver factors include stress, duration, intensity, provision of medical/nursing care tasks, and even particular serotonin genotypes (e.g. short allele carriers for 5-HTTLPR). While the greatest focus has been on dementia caregivers, recent work suggests groups such as end-of-life caregivers, former caregivers, and caregivers with multiple care roles (e.g. child, older adult, paid care) have sleep disturbances that vary by their unique contexts. Caregiver’s sleep, particularly subjective assessments, relates to their mood and physiological well-being; and evidence suggests that sleep may be an important mediator linking care stressors with mental health. Factors such as mindfulness and sense of coherence may protect caregivers from sleep disturbances.

Summary

Significant evidence exists regarding which caregivers are at risk for sleep disturbances and how sleep is associated with health. Caregivers’ appraisals of sleep health were strongly associated with their mental health and well-being. Objective assessments of sleep have been associated with medical conditions such as cardiovascular disease. Yet more work needs to be done to target the caregiving dyad jointly and help intervene to maintain health over time and prevent declining health. Ultimately, intervening within the unique context of caregiving still presents a challenge of significant public health interest given global population aging.

Introduction

In the United States, an estimated 17.6 million family members provide care for an older relative with significant impairment [1]. This care may include supervision, assistance with activities of daily living (e.g. dressing, bathing), or support with medical care tasks (e.g. medication management, giving injections). The care context can vary by care dyad, for example, whether the caregiver and care recipient cohabitate, the medical condition(s) of the care recipient (e.g. dementia, cancer, frailty), and the duration of care. In general, the broad literature on family care suggests caregiving responsibilities can be a chronic stressor with the potential for negative health impact [2–4]. A recent report suggests that 66% of dementia caregivers stated that caregiving interfered with their ability to care for themselves and 27% delayed or did not do things for their own health due to care provision [5]. One such health factor impacted by care, and receiving growing public health interest, is sleep. Indeed, prior work found that 50–74% of caregivers for individuals with dementia reported sleep disturbances [6]. Caregivers also report lower sleep quality and shorter sleep duration in the vein of 23–33 minutes a night relative to age matched non-caregiving adults [7, 6].

Evidence suggests a wide range of factors could explain heightened prevalence of sleep problems in caregivers. Caregivers may be predisposed to sleep problems due to factors including age-related disease or female gender [8, 9]. They may experience precipitating events including the care recipient’s nighttime behaviors, stress, or new onset depression. Finally, managing care may lead caregivers to adopt new “perpetuating” factors that contrast with sleep hygiene recommendations such as irregular sleep and wake schedules, using caffeine or naps to compensate for lost sleep, and reduced activity outside of the home [8]. Whatever the cause, poor sleep in caregivers is consequential, having been linked with health outcomes including elevated coagulation and inflammation levels, increased care stress, and earlier relinquishing of the care role [6, 10].

Given the prevalence and impact of sleep disturbances in caregivers, we aimed to critically review a growing body of recent work on caregiver sleep. Recently, meta-analytic evidence demonstrated that dementia caregivers have shorter sleep duration and poorer quality compared to non-caregivers [7]. We extend on this finding by reviewing the most recently published evidence on which aspects of sleep health are affected in caregivers, for whom these problems are worse, and which health outcomes are related. Furthermore, we review results from current state of the art interventions aimed at improving sleep health in caregivers.

Methods

We searched for recent journal articles (between January 1^st 2016 and September 1^st 2019) on informal caregivers providing care for adults aged 65 and older or with an age-related disorder such as dementia. The PubMed and PsychInfo databases were searched using the following terms: sleep, sleep disturbance, caregiving, caregiver, carer, informal caregiver, caregiver burden, adult, older adult. Papers that did not report the age of the care recipient or with care recipients younger than 65 were excluded unless the care recipient sample had an age-related disorder such as dementia. The search resulted in 66 relevant articles of which 37 met inclusion criteria. As the aim of the current paper was a critical review of recent literature on sleep in caregivers for older adults, we focus specifically on papers which provide new empirical evidence regarding associations with sleep in caregivers which specifically consider the caregiving context in their analysis and discussion.

Factors related to sleep problems in caregivers

In the last three years, research has shown a wide range of characteristics, of both the caregiver and care recipient, are associated with caregivers’ sleep quality.

Care recipient characteristics

How well a care recipient sleeps is associated with a caregiver’s own sleep quality. Bartolomei and colleagues [11] found that poorer sleep quality in Parkinson’s disease patients (Parkinson’s disease Sleep Scale & Epworth Sleepiness Scale score) was related to their caregivers having poorer quality (Medical Outcomes Study-Sleep Scale score) even after accounting for motor disturbances. Similar associations between care recipient and caregiver self-reported sleep quality have been found in Japanese and American dementia caregiver samples [12, 13]. Examining a nationally representative sample of dementia caregivers, Leggett and colleagues [14] found that a care recipient’s fall risk was associated with more frequent self-reported nighttime awakenings in caregivers. Falls may be a particularly worrisome event for caregivers that may lead to increased nighttime vigilance and interrupted routines.

Caregiving characteristics and stress

The study by Leggett and colleagues [14] also found that greater emotional difficulty related to the care role was independently associated with caregiver’s self-reported nighttime awakenings. Caregivers reporting more emotional difficulty with care may be worrying or ruminating, which could serve to perpetuate sleep disturbances.

The intensity of the care role and associated care-related stress appear to be key predictors of sleep disturbances. Simón and colleagues [15] found two care-related factors significantly predicted insomnia (Diagnostic Interview in Mental Health) in female caregivers of patients with dementia who were classified as totally dependent (by an official register in Spain). Caregivers with more than 8 years of care provision had a two-fold increased risk of insomnia, and those who provided care for more than 12 hours a day had a five-fold increased risk. These findings suggest that more advanced stages of dementia may require nighttime behavioral care that disrupts sleep.

Song and colleagues [16] compared ceased caregivers and continuous non-caregivers with low-intensity caregivers (intensity based on ADL/IADL provision), and high-intensity caregivers based on prior two-year care status. In general, actigraphy and self-report sleep measurements did not differ between groups. One exception was that among caregivers in the high-intensity group, those reporting high stress spent more actigraphy measured minutes being awake after sleep onset when compared with those reporting low stress (82 vs 65 minutes). These sleep continuity issues (also known as sleep fragmentation) may be because high intensity caregivers reporting stress are involved in more nighttime care activities such as helping the care recipient to the restroom. Medical comorbidities and older age of the caregivers may explain poor sleep quality across groups, yet this study adds evidence that the combination of high care intensity and stress is particularly impactful on nighttime awakenings [16].

Wilson and colleagues [17] also found that among dementia caregivers, care-related distress (defined as intrusive thoughts regarding caregiving and avoidant behaviors) was associated with poorer sleep quality (Pittsburgh Sleep Quality Index; PSQI score). They extend this line of work by showing that caregiver’s serotonin-related genetic profiles help explain the relationship between care-distress and sleep quality. Relative to other genotypes, caregivers who were short-allele carriers for 5-HTTLPR that reported greater care-related distress had poorer sleep quality. C carriers for the rs6295 polymorphism also showed the strongest associations between care-distress and sleep [17]. In other words, care avoidant coping strategies put caregivers at risk for sleep problems, particularly if caregivers have serotonin genotypes including the short-allele. However, caregivers with the short allele reporting low care-distress had the least sleep problems. Thus Wilson and colleagues posit that caregivers who show genetic predisposition to stress reactivity may have strong positive response to interventions [17].

In addition to care intensity and stress, particular care tasks may interfere with caregivers’ sleep routines and take time away from self-care. Polenick and colleagues [18] examined medical/nursing tasks (e.g. injections, medication management) carried out by spousal caregivers residing with their partner with dementia. Caregivers who performed a greater number of medical/nursing tasks experienced more self-reported care-related sleep disturbances independent of other contextual characteristics such as caregiving stressors. Wound care, in particular, was independently associated with care-related sleep disturbances. Thus care recipient’s sleep quality and fall risk, and caregiver’s care intensity, stress, and medical care task provision are important contextual factors that are associated with caregiver sleep quality, nighttime awakenings, and insomnia.

Which caregivers experience sleep problems?

As sleep quality in caregivers has been well detailed, recent studies have tended to explore specific types of caregivers which may be experiencing greater sleep problems.

Dementia

As dementia poses a particularly challenging care context with behavioral and psychological symptoms which may interfere with sleep, a great extent of the literature has focused on dementia caregivers. Liu and colleagues [19] compared caregivers for individuals with Frontotemporal lobar degeneration (FTD), dementia with Lewy bodies (LBD), and Alzheimer’s disease (AD) on their PSQI scores. Both caregivers for LBD and FTD had higher PSQI scores than caregivers for individuals with AD, with LBD caregivers having the highest scores. In the sample, individuals with FTD and LBD had more neuropsychiatric symptoms than individuals with AD, which could lead to greater caregiver burden and sleep disrupting care challenges. For example, sleep disorders (e.g. rapid eye movement sleep behavior disorder) and hallucinations are cardinal symptoms of LBD which may lead to greater nighttime interferences for caregivers. A small sample of dementia caregivers showed particularly high prevalence disturbed sleep patterns. For example, 67.5% reported taking past month sleep medication, on average caregivers slept for only 5.9 hours (based on actigraphy), and 91% fell above the cut-point indicating poor sleep quality on the PSQI [13]. Cross-cultural work presents complimentary findings, with Japanese caregivers showing poor sleep quality, and in particular caregivers younger than 65 years had significantly more sleep problems (39%) relative to same age community residents (17%), though the measure of sleep problems was imprecise (presence vs. absence of prior 2 weeks sleep problems) [20, 12]. Thus caregivers for individuals with dementia show particularly poor levels of sleep quality, likely due to the unique challenge of behavioral and psychological symptoms which may lead to nighttime disturbances.

Double-duty and triple-duty caregivers

Innovative new work by DePasquale and colleagues [21] compared double-duty-elder caregivers (providing formal and informal care to older adults) and triple-duty caregivers (providing formal care as well as informal care for a child and an older adult) with long-term care only caregivers and double-duty-child caregivers. This is key as these caregivers are overseeing the well-being and safety of care recipients in professional long-term care settings as well as their familial care recipient(s). Inadequate sleep may put individuals at risk for medical errors, irritability, or other outcomes that could reduce care quality. Using both objective actigraphy measurements and caregiver’s self-reports, caregivers were followed over 18 months. At baseline, double-duty-elder caregivers had shorter self-reported sleep durations relative to workplace-only caregivers (approximately 13 minutes less per day). Triple-duty caregivers reported poorer self-reported sleep quality as well as a shorter sleep duration relative to workplace-only caregivers (29-minute shorter daily sleep duration). Over 18-months, entering a double-duty-elder care role was related to worsening perceptions of sleep quality and quantity (10-minute loss of time in bed). Of note, differential findings between elder-carers and child-carers suggest that qualitative differences in the care roles impact sleep quality, not just the number of care roles one fulfills. Further, a number of different sleep outcomes, particularly objective ones (e.g. actigraphy measured total sleep time and waking after sleep onset) did not show significant differences among caregivers. While subjective sleep quality is robustly affected by caregiving role, more work is needed to show whether underlying sleep physiology is related to the caregiving role.

Former caregivers

In related mixed-methods work, Corey and colleagues [22, 23] found that concerns with sleep quality continued in the post-caregiving experience. Caregivers described difficulty adjusting back to prior routines or waking with nightmares following the death of a loved one, even as many as 10 years post-caregiving. Poor sleep quality (PSQI) in the former caregivers was associated with pre-loss depression, dysfunctional coping patterns, and being less extroverted [22, 23]. Ultimately, dementia caregivers have received the greatest focus and indeed report poor sleep quality on average. Innovative investigations into women holding multiple care roles show that juggling care tasks has negative implications for sleep. Further sleep problems may persist following termination of the care role.

Associations of sleep disturbances on caregivers’ health and well-being

Over the past three years, research has extended understanding of the mental and physical health effects of poor sleep on caregivers.

Cognition

Lathan and colleagues [24] compared dementia caregivers against matched controls on a web survey on cognition (Brain Performance Test web battery of cognitive tests) and self-reported sleep duration. Each additional hour of sleep was associated with better performance on the Digit Symbol Coding task, a measure of processing speed, and the Forward Memory Span test, a measure of working memory storage and maintenance. Despite the significant main effects, sleep and stress interacted contrary to a linear effect where more sleep duration was always better. Findings showed that longer sleep duration was associated with poorer cognitive performance at higher levels of stress. Potentially, an unmeasured confounder such as depression explains the association of long sleep in the context of stress with poorer cognitive performance.

Burden, affect, and mood

Recent studies also found associations between self-reported sleep problems and greater caregiver burden, anxiety and depressive symptoms [17, 25, 26]. Several studies have suggested that sleep may explain why care-related stressors such as memory and behavioral problems of dementia are associated with caregiver stress. Among working caregivers for older adults in Japan, self-reported sleep quality attenuated the association between a care-recipient’s behavioral and psychological symptoms of dementia and caregiver well-being, suggesting that one impact of stressful symptoms on well-being is their effect on caregiver sleep [27]. Further, a study of Taiwanese family caregivers for individuals with Alzheimer’s disease explored the mediating role of sleep quality (PSQI) and self-efficacy on the association between forms of care stress and caregiver depression [28]. Both the controlling upsetting thoughts facet of self-efficacy and sleep quality partially mediated the association between stress measures with both dementia and caregiver depression [28]. This suggests that worry and rumination over caregiving-related stressors reduces sleep quality, which further exacerbates depressive symptoms.

Associations with mental health are also found using objective measures of sleep quality. Using polysomnography and actigraphy, Smagula and colleagues [29] explored how 24-hour sleep-wake patterns (rest-activity rhythms) were associated with depressive symptom severity in strained dementia caregivers. In multivariate analyses, two facets of rest-activity rhythm were independently associated with caregiver’s depression: high sleep fragmentation (more minutes awake after sleep onset) and less distinct resting and active periods (reflecting slower transitions between active and resting states). The association between this specific rest-activity rhythm characteristic and depression symptoms was attenuated by 40% when adjusting for the care recipient’s frequency of dementia-related behavioral disruptions; this suggests that the relationship between this aspect of rest-activity rhythms and depressive symptoms is partially due to behavioral problems in care-recipients.

McCrae and colleagues, on the other hand, considered daily associations of self-reported and actigraphy assessed sleep quality and waking after sleep onset with positive and negative affect in caregivers for individuals with dementia [30]. Nights with better than average self-reported sleep quality and less waking after sleep onset were associated with lower next day negative affect ratings. However, no associations with actigraphy measured wake time or with positive affect were found. This suggests that caregivers’ own assessments of sleep quality and timing are important to their affect, and thus again suggests that self-report sleep measures are important to collect alongside objective measurements to fully characterize sleep and understand caregiver well-being.

Fatigue

DePasquale and colleagues [31] examined the convergence of double-duty and triple-duty care roles (as previously described) on fatigue experiences in daily diary reports. On average, women with better self-reported sleep quality had lower odds of reporting fatigue. Longer-than-usual self-reported sleep duration and reports of better-than average sleep quality were associated with lower odds of reporting fatigue the next day and of reporting less intense fatigue. These symptoms were similar regardless of double or triple-duty caregiver status with one distinction. On nights where a woman slept longer than their usual duration, the odds of next day fatigue only decreased for double-duty but not triple-duty caregivers. DePasquale and colleagues (2019) posit that extra sleep may not be enough of a resource to offset the fatigue debt that triple-duty caregivers are experiencing due to the added demands of informal care on top of formal and childcare.

Self-rated health

Washington and colleagues examined family caregivers of hospice patients on self-reported sleep problems, anxiety, and self-rated health [32]. Sleep and anxiety may be strongly linked in end of life caregivers due to anxiety of sleeping through a medical emergency or ruminating about other stressors. Approximately one third of hospice caregivers reported moderate sleep problems and/or anxiety, and these symptoms were highly correlated. Further, caregivers with greater sleep problems and anxiety symptoms reported poorer self-reported health [32]. As both sleep and anxiety may be facets of how one considers their global self-rated health, it is important to consider this question with more objective measures of sleep and health in the future.

Physiological well-being

Both prolonged and short sleep duration have been associated with chronic medical conditions such as diabetes, cardiovascular disease, and even mortality risk [33–35]. In exploring potential mechanisms, Leggett and colleagues examined self-reported sleep duration in association with the cortisol awakening response, a Hypothalamus-Pituitary-Adrenal Cortex axis marker of the stress response, in caregivers for individuals with dementia whose care recipient was attending adult day services [36]. The cortisol awakening response when adaptive is posited to increase energy for the day, arousal, and have anticipatory effects, whereas a blunted awakening response has been associated with fatigue and burnout in older adults. When caregivers slept longer, they had a smaller awakening response. However, Leggett and colleagues found this association was moderated by adult day service use. On adult day service days, regardless of sleep duration, caregivers showed cortisol increases, whereas on non-day service days the association varied by sleep duration (blunted cortisol patterns were found when caregivers slept longer than average, whereas a normal cortisol awakening response was found if they slept shorter than average). Respite interventions may offer physiological recovery to caregivers, overriding the prolonged sleep and maladaptive cortisol pattern.

Of note, aside from cognitive tests and cortisol reactivity, in the literature on the association between sleep quality and a variety of health indicators, most measures of well-being were self-rated and how sleep is associated with diagnosed health conditions and other objective measures of health should be explored. Further, effects of caregiver sleep on care recipient health and well-being were not explored in recent work.

Improving sleep

Improving sleep health is important given the high prevalence of sleep concerns among caregivers and the associated consequences reviewed above. Recent work has predominantly focused on modifiable protective factors that may be potential targets for such efforts. For example, Portier and colleagues [25] found that caregivers who reported a greater sense of coherence (comprehending, managing, and recognizing the meaning of life events) had better self-reported sleep duration. Thus, similar concepts of mastery and efficacy may be promoted to help caregivers better manage their circumstances and reduce stress. It may be that existing interventions focusing on caregiver management, such as supports for handling behavioral and psychological symptoms of dementia, already bolster caregiver’s mastery, and thus sleep health should be considered as an outcome measure for such studies.

Similarly, Hicks and colleagues [37] examined dispositional mindfulness, broadly defined as non-judgmental awareness of the present moment, in Parkinson’s disease caregivers. Dispositional mindfulness is believed to be protective against stress as it helps individuals live in the moment. This may be particularly fruitful for caregivers and patients with Parkinson’s as new challenges arise daily. Sleep disturbances are common in individuals with Parkinson’s disease with resulting physical exhaustion being one of those most common non-motor manifestations of the disease which can lead to reduced mental health and quality of life [37]. Hicks and colleagues found that greater dispositional mindfulness was associated with better sleep quality (PSQI) in both caregivers and their care recipients with Parkinson’s. Mindfulness practices have been associated with numerous health benefits and thus may serve as a stress reducing and sleep enhancing intervention that is easily trained and practiced among caregivers [38, 39].

However, intervention studies targeting caregiver sleep are rare. Based on a meta-analysis, behavioral sleep interventions had larger effect sizes on caregivers’ sleep quality relative to pharmacological interventions though results were statistically similar [7]. While behavioral interventions such as Brief Behavioral Treatment for Insomnia (BBT-I) [40, 41] are evidence based and effective for adults, aspects of the specific caregiving context such as nighttime awakenings of the care recipient, fall risk, medication regimen, stage or type of dementia, etc. need to be addressed for an intervention to be effective. Further, dyadic interventions targeting the patient and caregiver jointly have not been studied [16]. As sleep in caregiving dyads is bi-directional, particularly among spousal pairs and those living together, dyadic interventions may show greater impact [16]. A recent study by Gibson [42] and colleagues targeted both members of the care dyad with an intervention that included bright light (30 minutes of daily scheduled light box of 10,000 lux for morning use), exercise (30–40 minutes of yoga stretches or walking), and bibliotherapy (handbook on sleep hygiene and dementia). However, the intervention was not adapted to the specific situational context of each dyad, and some dyads improved while others declined over the study period in terms of actigraphy measured sleep parameters and PSQI score.

To determine how to improve dyadic interventions, Song and colleagues [16] asked caregivers their perspectives on sleep interventions in a focus group and caregivers reported attempting use of sleep hygiene strategies but not finding them helpful (though it is unclear what those attempts looked like). While they often utilized sleep medications, they stated a preference for not using pharmacological treatment yet also described increasing physical activity as challenging given physical and functional limitations of care recipients. Caregivers stated a preference for adapting any behavioral modifications themselves rather than including care recipients in the intervention protocol due to cognitive limitations [16]. These insights suggest that an intervention for caregivers might be most effective if it targets the dyadic context of the sleep, care recipient specific challenges, and also helps caregivers maintain behavioral modifications over time as health conditions may decline.

Future Directions and Conclusions

The past three years have advanced our understanding of caregiver sleep from exploring how care provision impacts sleep, for whom sleep disturbances are worse (e.g. former caregivers, double- or triple-duty caregivers), unique risk and protective factors (e.g. serotonin related genetic profiles, mindfulness, medical care activities), and health impacts of poor sleep (e.g. depression, blunted cortisol patterns). Further, qualitative, quantitative, and intervention studies have been conducted to understand caregivers’ perceptions of sleep interventions, determine modifiable factors on which to intervene, and test behavioral modifications to improve sleep for the care dyad.

Yet some limitations of these studies lead to directions for future research. Many reviewed studies used poor quality, often one-item, self-reported measures of sleep quality, and self-reported sleep quality (e.g. PSQI total score) was the most common sleep construct studied. While this may be a measure of convenience in a study that is assessing numerous characteristics, if we want to better understand caregiver’s sleep patterns and concerns, more precise and objective measures are needed (e.g. actigraphy assessments of waking after sleep onset and other parameters). Most studies are cross-sectional in design limiting causal determinations and health comorbidities are regularly unaccounted for in analysis. Further, studies showed caregivers’ own perceptions of their sleep quality may be as meaningful as objective indicators, with several showing associations among self-reported sleep indicators but not objective indicators. Thus, interventions that change perceived sleep quality may improve mental health beyond measurable change in sleep constructs (latency, waking after sleep onset). This is in line with the Stress Process Model of Caregiving [43] which shows how caregiver’s appraisals of stress matter more to well-being than the specific stressors themselves. Caregivers had preconceived notions about the utility of sleep hygiene, light therapy and other interventions and thus understanding values and preferences of this group may lead to better treatment adherence and utilization. As caregiving challenges can change daily, adapting interventions to the dynamic process of care are needed for long-term maintenance of sleep hygiene practices and sustained effects.

In conclusion, recent work shows that factors of both the care recipient (sleep quality, fall risk) and the caregiver’s care provision (care stress, intensity and duration, serotonin genotypes) impact caregiver’s sleep. While dementia care has received the greatest focus, research on other types of caregiving groups- former caregivers, double-or triple-duty caregivers, show that sleep problems don’t always remit following care and even a good night’s sleep doesn’t solve fatigue debt for caregivers in multiple care roles. Caregiver burden and depression are the most commonly considered caregiver health outcomes, and recent work uniquely shows how sleep may partially mediate the effect of care stressors on mood. Further, as self-report measures of sleep showed the strongest associations with affect, appraisals of sleep may be a key facet to how caregivers rate their overall health and well-being. However, it may also be that depressed mood is coloring caregiver’s perceptions of their sleep. Fortunately, the field seems to be moving from a focus on risk to resilience and encouraging such factors as sense of coherence and mindfulness that may protect caregivers against disturbed sleep. Future work can focus on helping caregivers find adaptive solutions to care-related challenges in their environment and bolstering their resilience.

Table 1.

Reviewed Articles and Key Findings

Article	Sample	Sleep measure	Main finding
FACTORS RELATED TO SLEEP PROBLEMS IN CAREGIVERS
Bartolomei et al. 2018	55 individuals with Parkinson’s disease and their CGs	CRs- Parkinson’s disease Sleep Scale & Epworth Sleepiness Scale CG- Medical Outcomes Study- Sleep Scale	CRs sleep quality associated with CG sleep quality
Okuda et al. 2019	496 CGs of individuals with Alzheimer’s disease and insomnia symptoms	CRs- Sleep Disorders Inventory CG- PSQI	CRs sleep problems associated with CG sleep quality
Peng et al. 2019	43 CGs of PLWD	CRs- Sleep Disorders Inventory, CG- PSQI, Sleep Hygiene Index, sleep diary and actigraphy	on average CGs slept for only 5.9 hours, and 91% fell above the cut-point indicating poor sleep quality
Leggett et al. 2018	528 same household CGs for PLWD	Likert scale item- in last month when you woke up before you wanted to how often did you have trouble falling back asleep	Greater nighttime awakenings were associated with CRs fall risk and CG reported emotional difficulty of care
Simón et al. 2019	134 female CGs of dependent PLWD	Primary Insomnia diagnosis from the Diagnostic Interview in Mental Health	CGs with more than 8 years of care provision had a two-fold increased risk and those who provided care for more than 12 hours a day had a five-fold increased risk of insomnia
Song et al. 2017	800 older women non-CGs and CGs	PSQI and actigraphy measured sleep time, efficiency, waking after sleep onset, and latency	In general, sleep measures did not differ between non-CGs, ceased CGs, low-intensity and high-intensity CGs
Wilson et al. 2018	130 CGs for PLWD	PSQI	CGs who were short-allele carriers for 5-HTTLPR and who reported greater care-related distress had worse sleep quality. Also C carriers for the rs6295 polymorphism showed the strongest associations between care-distress and sleep.
Polenick et al. 2018	104 spousal CGs of PLWD	Likert scale item of how frequently in last month sleep was interrupted due to caregiving	More medical/nursing task care provision associated with greater care-related sleep disturbances
WHICH CGs EXPERIENCE WORSE SLEEP
Liu et al. 2018	492 PLWD and their CGs	PSQI	CGs of individuals with FTD and LBD had worse sleep quality than CGs for individuals with AD
Koyama et al. 2017	104 CGs for PLWD	Presence vs. absence of sleep problems in past 2 weeks	Sleep problems were more common in younger (<65) CGs
DePasquale et al. 2019	1,135 long-term care employees	Items adapted from the PSQI	Double-duty and triple-duty CGs had shorter sleep duration and poorer sleep quality than those not in informal caregiving roles
Corey & McCurry 2018	8 former family CGs	Qualitative discussion	Sleep problems persisted even 10 years post-care
Corey et al. 2018	171 former family CGs for PLWD	PSQI	poor sleep quality was associated with pre-loss depression, dysfunctional coping patterns, and being less extroverted
SLEEP AND HEALTH
Lathan et al. 2016	527 CGs for PLWD and 527 matched non-caregiving controls	Self-reported how many hours of sleep you typically get each night	Additional hours of sleep were associated with increases in processing speed and working memory performance
Portier et al. 2018	79 spousal CGs	Sleep duration more or less than 8 hours- yes/no	Less sleep associated with CG burden
Coffman et al. 2017	165 Alzheimer’s disease CGs	PROMIS sleep disturbance form	Sleep was correlated with burden and anxiety
Honda et al. 2017	105 working family CGs in Japan	Self-rating of sleep as good, intermediate or poor	CG sleep attenuated the association of behavioral and psychological symptoms of dementia and CG well-being
Wang et al. 2016	72 Alzheimer’s disease CGs	PSQI	self-efficacy and sleep quality partially mediated the association between care stress and CG depression
Smagula et al. 2017	57 strained CGs for PLWDs	Rest-activity rhythm from actigraphy and polysomnography	high sleep fragmentation (waking after sleep onset) and less distinct resting and active periods (low rest activity rhythm beta) independently associated with CG depression
McCrae et al. 2016	55 Alzheimer’s disease CGs	Sleep diaries- total wake time and quality; actigraphy- total wake time	nights with better sleep quality and self-reported waking time were associated with lower negative affect ratings
DePasquale, Crain, et al. 2019	166 women working in nursing homes	Sleep duration and quality items adapted from PSQI	Longer sleep duration and better sleep quality were associated with less fatigue
Washington et al. 2018	395 family CGs of hospice patients	PHQ-9 likert scale item on trouble falling or staying asleep or sleeping too much	More frequent sleep problems associated with worse overall health
Leggett et al. 2016	158 CGs for PLWD	Self-reported daily sleep duration	When CGs slept longer, they had a smaller cortisol awakening response; this association was moderated by adult day service use.
INTERVENTION AND MODIFIABLE PROTECTIVE FACTORS
Hicks et al. 2019	18 Parkinson’s disease patients and their CGs	PSQI	Dispositional mindfulness was associated with better sleep quality in CGs and CRs
Gao et al. 2019	3268 CGs for PLWD	Systematic review and meta-analysis of sleep duration and sleep quality	CGs had shorter sleep durations and worse sleep quality than non-CGs; CGs improved with intervention
Song et al. 2017	5 CGs for Veterans	Focus group on behavioral sleep intervention	CGs described acceptability and challenges of sleep interventions
Gibson et al. 2017	15 CGs and PLWDs; 9 dyads completed the trial	Actigraphy, sleep diary, and PSQI	Light therapy, exercise, and sleep hygiene intervention had mixed effects on sleep

Note. PSQI= Pittsburgh Sleep Quality Index, CG= Caregiver, CR= Care Recipient, PLWD= person living with dementia, AD= Alzheimer’s disease, FTD= Frontotemporal dementia, LBD= Lewy Body Dementia

Table 2.

Summary of Recent Research

Key Findings
1.	Both care recipient (sleep quality, fall risk) and caregiver (caregiver stress, intensity of caregiving, medical tasks, serotonin genotypes) are associated with caregiver sleep quality.
2.	Recent work has examined groups beyond dementia caregivers who may be at risk for sleep disturbances including former caregivers, end-of-life caregivers, and caregivers with multiple roles (long-term care and informal care roles).
3.	Caregiver sleep helps explain the association between caregiving stressors and quality of life.
4.	Subjective assessments of sleep, in particular, show strong associations with caregiver mood, health, and well-being.
5.	Interventions need to consider the caregiving context in their approach and help caregivers maintain behavioral modifications over the changing course of a caregiving role.