Introduction

Older adulthood (generally defined as those aged 65 and older) is a unique developmental period, as the particular combination of circumstances and contexts in older adulthood involves distinct benefits and challenges. Older adults are more likely to have at least one chronic disease than other age groups and often have comorbidities in other aspects of health such as mental health problems, cognitive decline, and sleep disturbance [1, 2]. Largely due to these health challenges, older adults are much more likely to utilize healthcare services, with the cost of health care and the amount of healthcare utilization increasing as people age [1]. Moreover, the population of older adults is projected to sharply increase over the next several decades, with some calling it a “silver tsunami.” As such, research investigating the health and well-being of older adults is becoming increasingly salient as society prepares to meet the increased demands of this growing population.

Sleep plays an essential role in an individual’s health and well-being. Poor sleep is associated with increased risks of cardiovascular disease, hypertension, diabetes, depression, anxiety, and all-cause mortality [3,4,5,6,7,8]. Changes in sleep, including in quantity and quality, occur throughout the lifespan as individuals age. For example, the National Sleep Foundation recommends that teenagers get 8–10 h of sleep, young and middle-aged adults get 7–9 h of sleep, and older adults get 7–8 h of sleep [9]. Yet, short sleep is a prevalent problem, especially for older adults. One national survey in the USA reported that 24% of adults aged 65 and older were not getting at least 7 h of sleep per night [10]. Moreover, a nationally representative poll on healthy aging reported that 54% of older adults believe that poor sleep is a normal part of aging [11]. Older adults who have short sleep duration are more likely to report pain, frailty, cognitive decline, and lower quality of life [12,13,14,15,16]. Furthermore, 67.7% of older adults report difficulties falling asleep or waking up too early, with 46% reporting difficulties falling asleep at least 1–2 nights per week [17]. Insomnia symptoms in older adults are often more severe, chronic, and impairing than in younger samples. Insomnia in late life has been associated with health conditions such as pain, prostate cancer, heart disease, impaired cognition, obesity, and depression [18,19,20,21]. Sleep problems and depression are highly comorbid, generally, and in older adults as well, highlighting an important area of investigation.

Depression, a mental disorder characterized by low mood and anhedonia, has been identified as one of the most underfunded and pressing health issues [22,23,24]. Estimates of the global prevalence of depression in older adults range from 13 to 28% [25, 26]. Depression is seemingly less common among community dwelling older adults; however, this may be due to underdiagnosis [27] and changing clinical presentation [28,29,30,31,32,33]. Moreover, depression may be more chronic in older adults, as one study observed that 36% of young adults with an initial diagnosis of depression had a diagnosis 2 years later, while 51% of older adults with an initial diagnosis of depression had the diagnosis persist for 2 years [34]. Subclinical depressive symptoms are also common and impactful in late life [35].

Although sleep problems were long thought to be an epiphenomenon of depression (e.g., sleep disturbance being part of the diagnostic criteria of major depressive disorder), it is now widely recognized that sleep and mood have a bi-directional association [36]. In older adults, those who have comorbid insomnia and depression have been observed to have a higher prevalence of Alzheimer’s disease and related dementias, as well as greater limitations on activities of daily living (ADLs), than those with either insomnia or depression [37, 38, 12, 39, 40]. Due to the high comorbidity between sleep problems and depression, underlying factors, such as shared neural pathways or altered brain structure and activity, may elucidate important aspects about the nature of this association. For example, it has been observed that depression can be associated with abnormalities and dysregulation in brain areas that overlap with those observed in the pathophysiology of insomnia [22, 41]. Given the impacts that sleep problems and depression symptoms can have on the well-being of older adults, elucidating the nature and underlying factors of the bi-directional association between sleep problems and depression remains a pressing research priority for the field [42].

This review aims to synthesize and report on current research findings (i.e., last 1–4 years) regarding the link between sleep duration and depression and the link between insomnia and depression in older adults. A review of recent literature was conducted using PubMed, including only articles that were published from 2019 to 2023. Articles were included if at least 95% of the sample comprised adults aged 60 years or older and if the study included measurement of depression and sleep duration or insomnia as variables of interest. The review is organized by first providing a summary of findings regarding sleep duration and depression and then summarizing findings regarding insomnia and depression. Refer to Table 1 for a summary of studies focused on sleep duration and depression. Refer to Table 2 for a summary of studies focused on insomnia and depression.

Table 1 Studies investigating sleep duration and depression in older adults (n = 18)
Full size table
Table 2 Studies investigating insomnia and depression in older adults (n = 34)
Full size table

Sleep Duration and Depression

Sleep duration is one of the most recognizable characteristics of sleep and is associated with various aspects of an individual’s health and well-being [3, 4]. For example, individuals with depression often experience sleep disturbances such as hypersomnia or insomnia. Recent research has highlighted that short (less than 6–7 h) and long (more than 8–9 h) sleep durations are risk factors for depression and increased depression symptoms in older adults. For example, one recent meta-analysis, and a recent study in Taiwanese older adults, observed a u-shaped association between nighttime sleep duration and risk of depression, such that the lowest risk was at 7 h of sleep and higher risks were at both short and long sleep durations [43, 44]. Hu and colleagues found that short and long sleep durations were associated with increased depression symptoms which were then associated with poorer quality of life in older adults [12•]. The u-shaped association between sleep duration and depression symptoms may, in part, be reflective of individual differences. Zuidersma and colleagues did a series of 8 single subject studies that tracked sleep and depression symptoms over 63 consecutive days in older adults and found that individuals displayed unique patterns of associations between sleep duration and subsequent depression symptoms. For example, for some older adults, long sleep duration was related to depression symptoms; however, in other older adults, short sleep duration was associated with increased depression symptoms. Interestingly, changes in depression symptoms were not predictive of subsequent changes in sleep duration for any of the participants [45]. One potential explanation for these disparate results may be that different combinations of depression symptoms can lead to either more or less sleep. For example, ruminative thinking and psychomotor agitation may lead to less sleep while increased fatigue and low mood may lead to more sleep.

Sleep Duration, Depression, and Other Clinical Factors

Short or long sleep duration combined with depression can potentially lead to an individual feeling more fatigued and less motivated during the day, which can heighten adverse health consequences. Several recent studies investigated changes in sleep duration and depression symptoms during the transition from pre- to peri-pandemic in older adults. Mishra and colleagues observed that depression symptoms significantly increased during the initial phases of the COVID-19 pandemic and that this increase was associated with a decline in both time in bed and physical activity [46]. The increase in depression symptoms, with an associated decrease in total sleep time, was also observed by a recent study utilizing a representative sample of older adults from Italy [47], which found that the prevalence of depression symptoms more than doubled in older adults, increasing from 16.5 to 27.1%, and that the prevalence of insufficient sleep rose as well, increasing from 35.2 to 39.5%. The transition to pandemic life was associated with a variety of changes, and one robust observation was that decreases in physical activity were associated with increases in depression symptoms and decreases in sleep duration.

Several recent studies have highlighted clinical characteristics that are associated with both sleep duration and depression. For example, a meta-analysis demonstrated that depression and sleep duration (long and short) were consistent and robust predictors of subsequent cognitive decline [15•]. This finding was further clarified by a cross-sectional study of 4417 older adults that found that both short and long sleep durations were associated with increased depression symptoms, sharper cognitive decline, and increased amyloid-β burden [48•]. In another meta-analysis that pooled over 6000 participants, depression symptoms and short sleep duration were risk factors of cognitive frailty, while a separate study observed that long sleep duration and depression symptoms were also associated with cognitive frailty [40, 49]. A longitudinal, prospective cohort study done in 5026 Chinese older adults observed that short sleep duration accelerated the transition to frailty in older adults [14•]. Moreover, a study that utilized a random sample of older adults observed that both short and long sleep duration were associated with decreased functional capacity [50]. Two studies, using a statistical method to model probabilities, reported that independent hypothetical increases in sleep duration and decreases in depression symptoms were associated with significantly lower risks of falls and long-term risk of cognitive impairment [51, 52]. Moreover, these studies indicated that the effects of hypothetical interventions could be additive, such that multicomponent interventions targeting both sleep and depression could be more effective than just treating one or the other. Overall, these results suggest that treatments aimed at improving sleep duration and depression may have potential to improve other clinical characteristics, such as falls and cognitive impairment. However, more studies need to be conducted to directly test this hypothesis.

Insomnia and Depression

One of the diagnostic criteria for depression is sleep disturbance, specifically experiencing symptoms of hypersomnia or insomnia. Insomnia and depression are highly comorbid in older adults and across the lifespan. One epidemiologic study in individuals aged 16–74, estimated that 40.5% of individuals with depression meet criteria for a DSM-IV insomnia disorder diagnosis, with an estimated 83.2% having difficulty falling asleep or staying asleep [53]. In older adults, symptoms of insomnia are consistently associated with depression symptoms across a variety of samples and settings [8, 54,55,56,57,58,59]. However, there is still uncertainty about the specific details of this association. To elucidate underlying factors of this association, one study investigated how brain structures were associated with symptoms of insomnia and depression and found that lower cortical thickness in various brain areas, including the left insula, were associated with increased insomnia symptoms when adjusting for depression symptoms [60]. Other recent efforts have sought to examine how types of insomnia symptoms are associated with depression. For example, one study of Black adults found that depression symptoms were associated with the daytime symptoms of insomnia on the Insomnia Severity Index (ISI) and not the nighttime symptoms [61]. Another study observed that maintenance insomnia endorsed on the ISI (i.e., primary difficulty of staying asleep) was the only insomnia type associated with depression symptoms after controlling for anxiety [62]. However, a longitudinal study which investigated whether insomnia types predicted the onset of depression at 2-, 4-, and 6-year follow-ups found that maintenance insomnia did not predict the onset of depression symptoms at any follow-up, while early morning awakening (i.e., primary difficulty of waking up too early and not falling back to sleep) and sleep-onset (i.e., primary difficulty of falling asleep) types predicted depression onset at the 2-year follow-up, while nonrestorative sleep predicted the onset of depression at all follow-up time points [32•]. Another study investigated whether changes in sleep disturbance predicted a diagnosis of major depressive disorder and suicidal ideation 1 year later in older adult primary care patients and found that patients with worsening sleep disturbances were much more likely to have MDD and suicidal ideation and less likely to be in remission than those whose sleep disturbance persisted at a similar level [63•]. Taken together, these results suggest that insomnia symptoms play a significant role in subsequent depression symptoms and depressive episodes. However, there are mixed findings about the role that specific insomnia symptom types play in the subsequent development of depression. Moreover, a study that utilized a prospective, cross-lagged panel design observed that higher depression symptom burden was a significant predictor of increased insomnia symptoms the following week, thereby demonstrating that depression symptoms can also precede insomnia symptoms [64]. Thus, the association between insomnia and depression appears to be bi-directional in late life.

Insomnia, Depression, and Other Clinical Characteristics in Older Adults

Several studies have observed that both insomnia and depression are independent risk factors for cognitive decline and post-operative delirium, with those who have comorbid insomnia and depression having the highest prevalence of Alzheimer’s disease and related dementias [37, 65, 66]. Moreover, insomnia and depression have been associated with physical and cognitive frailty, with one longitudinal study showing that depression, but not insomnia, was predictive of later falls and another study demonstrating that individuals with insomnia had a higher prevalence and fear of falls [54, 67,68,69,70]. Other studies have observed that insomnia and depression are associated with poorer quality of life, with one study observing that those with comorbid insomnia and depression reported poorer quality of life than those with just insomnia or just depression [71,72,73,74]. Another study observed that older adults with asthma who had comorbid insomnia and depression were also much more likely to have ADL limitations than those who just had insomnia or depression [38•]. Moreover, one study of 2976 primary care patients with osteoarthritis found that individuals with increased symptoms of depression and insomnia were more likely to have higher utilization of healthcare services and increased costs associated with health care [75]. Taken together, these results demonstrate that insomnia and depression can have deleterious impacts on multiple aspects of older adult’s health and well-being, including physical health and frailty, cognition, and dementias, particularly if the conditions are comorbid. These impacts may lead to increased healthcare utilization and increased costs associated with health care.

Treatment for Insomnia and Depression in Older Adults

Several studies have investigated the effects of treatments for insomnia and depression, with a particular emphasis on whether treatments for insomnia also reduce symptoms of depression or vice versa. For example, one systematic review examined whether cognitive-behavioral interventions in depressed and anxious older adults would also improve sleep disturbances and only found two studies that investigated this issue [76]. The two studies demonstrated evidence that Cognitive-Behavioral Therapy for Insomnia (CBT-I) in depressed and anxious older adults reduced both depression symptoms and sleep disturbances, although more research is clearly needed [76]. In another study, a group that was randomized to just 2 months of CBT-I had significantly less likelihood of incident depression than a group which received sleep education therapy [77]. One randomized controlled trial of 327 older adults with comorbid insomnia and osteoarthritis pain found that those who got six 30-min telephone sessions of CBT-I had significant decreases in insomnia symptoms over a 12-month period, but no significant decreases in depression symptoms, while a different study utilizing telephone CBT-I found significant decreases in both insomnia and depression across 6 months in the CBT-I group [78, 79]. A secondary analysis of these same data showed that those who improved their insomnia or pain in the first 2 months were more likely to have improved in their sleep, pain, depression, and fatigue at the 12-month follow-up [80•]. Yet, another randomized controlled trial with 47 older adults found that a brief CBT-I intervention was effective in reducing depression symptoms but not effective in producing a change in global sleep quality [81]. A systematic review from McLaren and colleagues investigated whether behavioral interventions for insomnia, such as stimulus control and sleep restriction, were effective in reducing insomnia and mood symptoms [82]. This review found that all behavioral interventions reviewed improved self-reported aspects of sleep and also improved depression symptoms [82]. Overall, a greater preponderance of evidence suggests that interventions targeting sleep, including CBT-I and behavioral interventions, may also be efficacious in reducing depression symptoms, although this also may depend on treatment modality (e.g., in-person vs. online vs. telephone). Future research that investigates which treatments, and which components of those treatments, are most effective in reducing insomnia and depression when they are comorbid would help elucidate when treatments are most effective.

Summary

Sleep problems and depression are prevalent phenomena in older adults, especially in clinical settings. While changes in sleep, including decreases in sleep duration and quality, can be a normal part of the aging process, poor sleep does not have to be part of the aging experience. Short and long sleep duration are associated with adverse impacts on the health and well-being of older adults and are particularly important risk factors for depression in older adults. Sleep duration and depression are also risk factors for other important clinical characteristics including cognitive and physical frailty, cognitive decline, and poorer quality of life. Importantly, sleep duration and depression symptoms are modifiable and treatable, with recent hypothetical interventions demonstrating that improvements in sleep duration and depression symptoms should lead to decreased risk of falls and cognitive impairment. Public health interventions aimed at promoting healthy sleep opportunity (i.e., setting aside enough time to get 7–8 h of sleep at night) and practices (e.g., regular bedtime and waketime, sleeping in a dark, quiet, and cool environment) may help improve the public’s sleep health and mental health as well. Insomnia and depression are highly comorbid phenomena in older adults, and when these phenomena are comorbid, the deleterious impacts on health appear to be additive. Because of the high patient burden and healthcare costs of insomnia and depression, there have been efforts to investigate potential interventions that reduce symptoms of both disorders at once. Several studies demonstrated that CBT-I and multicomponent behavioral interventions are effective in reducing both insomnia and mood symptoms, while other research has not had similar success in demonstrating these effects. Future research that investigates whether different treatment modalities (e.g., in-person, online, and telephone) affect the outcomes of these interventions, including the efficacy of improving both insomnia and depression symptoms, may have particular implications in expanding the accessibility of treatments to older adults. Moreover, investigating participant characteristics, such as various health comorbidities that are especially prevalent in older adults (e.g., osteoarthritis and dementia), may have important implications on the generalizability of treatment studies.

Limitations and Future Directions

While exciting work has been and continues to be done in the field, it is important to consider the recent research findings in light of a few limitations. Firstly, many studies are cross-sectional in nature. While cross-sectional studies are important and informative in their own right, more temporal research designs (e.g., longitudinal and cohort studies) would benefit the field, as these designs allow researchers to examine directionality in the associations between sleep problems and depression. Investigating directionality in these associations will have clear implications for treatment considerations. Another limitation of recent research is that sleep duration is frequently measured using self-report data. Utilizing behavioral measures of sleep duration (e.g., actigraphy and other accelerometers) would have the benefit of assessing sleep–wake activity rhythms in a person’s natural environment across several days which may have the potential to elucidate how changes in activity rhythms across both sleep and wake are associated with changes in depression [83]. Moreover, prospectively measuring sleep duration could also enhance our understanding of the complex interplay between sleep and depression, as it would allow for researchers to examine whether inter-individual and intra-individual fluctuation in sleep duration is associated with changes in depression symptoms. A final limitation of current research is that many intervention studies have small sample sizes, which limits the statistical power of the studies and the generalizability of the effects observed. Including more participants, especially those with comorbid conditions, could help determine which interventions are most effective for a variety of older adults.

Conclusions

This review summarized and synthesized recent research investigating the link between sleep duration and depression and the link between insomnia and depression in older adults. Recent studies demonstrated that both short (< 7 h) and long (> 8–9 h) sleep durations and insomnia are independent and important risk factors for depression. Moreover, short and long sleep durations, insomnia, and depression are risk factors for poorer health in older adults, including increased risk of cognitive decline and falls, and poorer quality of life. Importantly, sleep and depression are modifiable, and studies have suggested that interventions that improve sleep duration or insomnia (e.g., CBT-I) may also improve depression symptoms. Future research that examines whether inter-individual and intra-individual fluctuations in sleep duration and insomnia are associated with changes in depression symptoms may have important implications regarding the directionality of the association between sleep and depression in older adults, which may then lead to refinements in current treatments.