Abstract
Sleep is increasingly recognized as a core determinant of health, yet the literature remains fragmented by organ system and is frequently reduced to a single metric, sleep duration. This integrative review synthesizes epidemiological, experimental, mechanistic, and intervention evidence to advance a multidimensional account of sleep health. Duration, regularity, timing, continuity and efficiency, architecture, daytime functioning, and sleep disorders are treated as partially distinct dimensions that converge on shared physiological systems. Habitual short sleep predicts obesity, type 2 diabetes, cardiovascular events, and mortality, while experimental restriction impairs insulin sensitivity and promotes excess energy intake; a randomized sleep-extension trial shows that increasing sleep can reduce objectively measured energy intake in habitual short sleepers, and sleep regularity has emerged as an independent, and in some models stronger, predictor of cardiometabolic risk and mortality than duration. Insufficient and fragmented sleep can weaken vaccine responses, and repeated partial sleep deprivation increases inflammatory signaling. Sleep also supports memory consolidation and attentional stability. By contrast, the proposed role of sleep in glymphatic clearance and neurodegeneration remains mechanistically promising but actively contested, with at least one direct experimental challenge to the central clearance hypothesis. Psychiatric evidence is comparatively stronger: insomnia prospectively predicts depression, and randomized trials now indicate that cognitive behavioral therapy for insomnia can improve mood and, in selected populations, reduce incident major depression. We propose a Sleep Systems Integrity model in which multidimensional sleep health regulates four interacting domains: metabolic and autonomic control, immune and inflammatory balance, neural plasticity and homeostasis, and affective and circadian stability. The evidence supports treating sleep as a foundational and modifiable health behavior while avoiding the stronger claim that optimizing sleep alone can prevent the diverse diseases with which poor sleep is associated.
Keywords: sleep health, sleep regularity, glymphatic system, inflammation, cardiometabolic health, circadian rhythm, insomnia, memory consolidation, depression prevention
1. Introduction
Human beings spend approximately one-third of their lives asleep, a behavior that, on its face, appears biologically extravagant. Sleep suspends foraging, mating, vigilance against predation, and social engagement, and it does so on a nightly, non-negotiable basis across nearly the entirety of the animal kingdom. That natural selection has preserved so costly a behavior across phylogenetically diverse taxa is itself strong evidence that sleep performs functions that cannot be readily accomplished during wakefulness (Robbins et al., 2024). Yet contemporary industrialized societies have increasingly treated sleep as a discretionary commodity, compressible in the service of work, technology use, or social demands. National surveillance data illustrate the scale of this trend: the most recent analysis of the U.S. National Health Interview Survey found that 30.5% of American adults slept less than seven hours per night in 2024, with prevalence highest among adults aged 50–64 and among Black adults relative to other racial and ethnic groups (Ng et al., 2026).
This review advances the thesis that sleep is not merely correlated with health but is a constitutive, causally implicated determinant of it — an active physiological process whose disruption produces measurable pathology across metabolic, immunological, and psychiatric domains. The review departs from earlier syntheses in one important respect: rather than treating sleep duration as a single summary exposure, it treats duration, regularity, timing, continuity, architecture, and daytime functioning as partially distinct dimensions of sleep health that converge on shared physiological systems (Buysse, 2014; St-Onge et al., 2025). Organized around this multidimensional framework, the review first develops the case for treating sleep health as more than duration; it then examines sleep as a metabolic and cardiovascular regulator, as a determinant of immune competence and cerebral homeostasis, and as a foundation of cognitive function and psychiatric stability. It closes by integrating these domains into a unified physiological account, considering the methodological limitations of the underlying evidence base, and outlining implications for clinical practice and public health policy.
2. Review Methodology and Scope
This is a narrative integrative review rather than a formal systematic review or meta-analysis; no protocol was pre-registered, and no formal quality-scoring or risk-of-bias synthesis was conducted across the entire literature, consistent with the review's aim of synthesizing converging evidence across disparate disciplines rather than exhaustively cataloguing a single body of work. Priority was given, wherever available, to systematic reviews, meta-analyses, randomized controlled trials, and large prospective cohort studies over single small studies, and to work published within the past decade, supplemented by foundational experimental studies of continuing mechanistic importance (e.g., Spiegel et al., 1999, 2004; Van Dongen et al., 2003). Evidence spans epidemiology, experimental physiology, immunology, neuroscience, and clinical psychiatry, reflecting the deliberately cross-disciplinary aim of the review. The resulting heterogeneity in study design — observational cohort, randomized experimental sleep restriction and extension, and mechanistic animal work — is treated as a strength for triangulating causal inference, but its implications for the strength of causal claims are addressed explicitly in the Discussion and Limitations sections below.
3. A Multidimensional Model of Sleep Health
Sleep duration remains important, but it is only one component of sleep health. Buysse (2014) conceptualized sleep health through six dimensions: regularity, satisfaction, alertness, timing, efficiency, and duration. More recent American Heart Association guidance expands this view to include continuity, architecture, daytime functioning, and the absence of sleep disorders (St-Onge et al., 2025), paralleling the addition of sleep as an eighth metric to the AHA's Life's Essential 8 cardiovascular health construct (Lloyd-Jones et al., 2022). These dimensions are correlated but not interchangeable. A person can obtain an apparently adequate number of hours while maintaining irregular timing, severe fragmentation, or untreated sleep apnea, and a duration-only model can therefore misclassify biologically important sleep disturbance.
Regularity is especially consequential. In a prospective UK Biobank cohort of 60,977 adults with more than ten million hours of accelerometer data, greater sleep regularity predicted lower all-cause, cancer, and cardiometabolic mortality, and regularity was a stronger mortality predictor than duration in comparative models (Windred et al., 2024). A 2025 review of the sleep-irregularity literature similarly found consistent evidence linking irregular sleep timing with depressive and anxiety symptoms, adiposity, insulin resistance, hypertension, cardiovascular events, and mortality, while cautioning that experimental and longitudinal data establishing a direct causal effect of irregularity remain limited (Hartstein, Grandner, & Diniz Behn, 2025). These findings do not prove that regularizing sleep schedules will prevent these outcomes, but they demonstrate that duration alone omits clinically relevant information.
We therefore propose a Sleep Systems Integrity model as an organizing heuristic for the evidence that follows. Multidimensional sleep health is hypothesized to regulate four interacting domains: metabolic and autonomic control; immune and inflammatory balance; neural plasticity and homeostasis; and affective and circadian stability. The model predicts that cumulative impairment across sleep dimensions will often outperform any single sleep metric in explaining downstream disease risk, that different sleep disturbances will generate partially overlapping downstream phenotypes because they converge on shared pathways, and that interventions will produce the largest downstream benefit when the targeted sleep dimension is genuinely causal for the outcome of interest rather than merely correlated with it. This framework is offered as a heuristic for organizing a fragmented literature, not as an independently validated causal model; it is revisited and qualified in the Discussion.
4. Sleep as a Regulator of Metabolic and Cardiovascular Health
4.1 Epidemiological Evidence Linking Sleep Duration to Obesity
The relationship between sleep duration and obesity has been examined in dozens of prospective cohort studies and synthesized in successive meta-analyses. Pooled analyses of prospective cohorts have established a discernible dose-response relationship: a meta-analysis of eleven cohorts encompassing nearly 200,000 participants reported a pooled odds ratio of 1.45 (95% CI, 1.25–1.67) for incident obesity among short sleepers relative to normal-duration sleepers, with no comparable association for long sleep duration (Wu et al., 2014). A subsequent updated meta-analysis restricted to adult cohorts similarly found that only short sleep duration — not long sleep duration — predicted future obesity (OR = 1.41, 95% CI 1.18–1.69), and a dose-response meta-analysis modeling the relationship continuously found a reverse-J-shaped curve, with obesity risk rising by approximately 9% for each one-hour decrement below seven hours of nightly sleep. Focusing specifically on abdominal adiposity, a 2024 systematic review and meta-analysis of seven prospective cohorts found that short sleep duration was associated with an 8% increase in the relative risk of central obesity (RR = 1.08, 95% CI 1.04–1.12), whereas long sleep duration showed no such association (Kohanmoo et al., 2024). Together, these analyses converge on a consistent asymmetric pattern: it is sleep insufficiency, rather than sleep excess, that most reliably predicts adiposity gain over time.
This observational picture is tempered, however, by evidence from randomized controlled trials that experimentally manipulated sleep duration and objectively measured adiposity-related outcomes; a systematic review and meta-analysis of this experimental literature concluded that it does not yet provide as strong a basis for causal inference as the observational literature alone might suggest, a tension addressed further in the Discussion below.
4.2 Endocrine and Behavioral Mechanisms: Appetite Hormones and Hedonic Eating
The mechanistic case for a causal effect of sleep loss on adiposity rests substantially on evidence from tightly controlled experimental sleep-restriction studies. In an influential crossover study, Spiegel, Tasali, Penev, and Van Cauter (2004) restricted twelve healthy young men to two nights of sleep curtailed to four hours and compared their hormonal and subjective appetite profiles to two nights of ten-hour sleep extension under matched caloric intake. Sleep restriction was followed by an 18% reduction in circulating leptin, a 28% elevation in ghrelin, and corresponding increases in self-reported hunger (24%) and appetite (23%), an effect most pronounced for calorie-dense, high-carbohydrate foods (Spiegel et al., 2004). This built on earlier work by the same research group demonstrating that even modest sleep debt perturbs metabolic and endocrine function more broadly, including sympathovagal balance, cortisol, and thyrotropin regulation (Spiegel et al., 1999).
More persuasive evidence comes from intervention rather than restriction alone. In a randomized clinical trial of adults with overweight who habitually slept less than 6.5 hours, individualized sleep counseling increased sleep by approximately 1.2 hours per night and reduced objectively measured energy intake by about 270 kcal per day without a prescribed diet or exercise program (Tasali et al., 2022). This provides unusually direct evidence that increasing sleep can alter energy balance under free-living conditions, although longer trials are required to determine durability and effects on obesity incidence.
Subsequent studies employing ad libitum feeding paradigms — arguably more externally valid than tightly controlled intravenous-feeding protocols — have not always replicated the leptin and ghrelin findings, suggesting that the hormonal explanation, while real, may not be the dominant mechanism driving excess caloric intake under free-living conditions (Rogers et al., 2024). A complementary mechanism implicates hedonic and reward-related neural circuitry: sleep-restricted individuals show heightened behavioral and neural responsiveness to energy-dense food cues, a pattern consistent with the observation that ad libitum sleep-restriction experiments consistently show increased caloric intake even when appetite-hormone changes are absent or inconsistent (Rogers et al., 2024). The convergent behavioral outcome — a preferential increase in intake of energy-dense food without compensatory increases in energy expenditure — creates a sustained positive energy balance that, compounded nightly, plausibly accounts for a meaningful fraction of the epidemiological association between short sleep and weight gain.
4.3 Sleep and Glucose Metabolism: Experimental Evidence
Independent of its effects on appetite, sleep restriction directly impairs glucose homeostasis. In one of the earliest experimental demonstrations of this relationship, one week of sleep restricted to five hours in bed reduced intravenous-glucose-tolerance-test-derived insulin sensitivity by approximately 20% in healthy men, without a compensatory increase in insulin secretion. This finding has since been replicated and extended: a systematic review and meta-analysis of randomized controlled trials found that experimental sleep restriction — evaluated using oral and intravenous glucose tolerance testing and homeostatic model assessment of insulin resistance — consistently reduced insulin sensitivity, with comparable reductions observed following experimental suppression of slow-wave sleep and induced circadian misalignment, but not following isolated REM-sleep disturbance or sleep fragmentation alone. Even a single night of total sleep deprivation can reduce insulin sensitivity by more than 20% without a compensatory beta-cell response. Conversely, interventional studies extending sleep in chronically short-sleeping adults have demonstrated improvements in fasting insulin sensitivity, providing complementary evidence that the relationship between sleep and glucose metabolism is not merely correlational but bidirectionally modifiable. The population-level relevance of these experimental findings is corroborated by meta-analytic evidence that habitual short sleep duration (five to six hours per night) predicts incident type 2 diabetes with a pooled relative risk of approximately 1.28 across more than 100,000 participants, consistent with the broader epidemiological literature establishing sleep duration as an independent risk factor for type 2 diabetes (Antza et al., 2021).
4.4 Cardiovascular Consequences and Mortality
The metabolic and neuroendocrine perturbations produced by sleep loss — elevated sympathetic tone, glucocorticoid excess, systemic low-grade inflammation, and endothelial dysfunction — converge mechanistically on cardiovascular risk. A dose-response meta-analysis of 71 cohorts encompassing 3.8 million participants found a significant U-shaped relationship between nighttime sleep duration and both cardiovascular and cerebrovascular disease, with the lowest risk observed at approximately 7.5 hours of sleep per night. A separate meta-analysis focused specifically on cardiovascular mortality, encompassing 19 studies and more than 800,000 individuals, similarly concluded that both short (<7 hours) and long (>9 hours) sleep durations increased cardiovascular mortality risk, an effect more pronounced among Asian populations and older adults. Using National Health and Nutrition Examination Survey data linked to the National Death Index, a U.S. population-based cohort study likewise found a U-shaped association between sleep duration and both all-cause and cardiovascular mortality, with the nadir of risk at approximately seven hours per night. The consistency of this U-shaped pattern across independent cohorts and outcome measures — obesity, diabetes, cardiovascular events, and all-cause mortality — suggests a shared physiological vulnerability at both extremes of the sleep-duration distribution, though the mechanisms underlying the long-sleep association, which may partly reflect reverse causation from occult illness, are less well characterized than those underlying the short-sleep association, a distinction discussed further below.
4.5 Circadian Misalignment and Shift Work as a Natural Experiment
Beyond sleep duration per se, circadian misalignment — the desynchronization between the endogenous circadian clock and the behavioral sleep-wake, feeding, and activity cycle — provides a complementary experimental and naturalistic model for isolating the cardiometabolic consequences of disrupted sleep-wake timing. In a landmark laboratory study, Scheer, Hilton, Mantzoros, and Shea (2009) subjected healthy adults to a forced-desynchrony protocol that misaligned behavioral cycles from the endogenous circadian pacemaker and observed resultant increases in postprandial glucose and insulin, blood pressure, and inflammatory markers, along with a reversal of the normal diurnal cortisol rhythm — effects that emerged from circadian misalignment independent of any reduction in total sleep time (Scheer et al., 2009). These experimental findings converge with epidemiological evidence from shift-worker populations: rotating and night-shift schedules are consistently associated with elevated rates of obesity, metabolic syndrome, type 2 diabetes, dyslipidemia, and cardiovascular disease, effects attributed mechanistically to circadian misalignment-induced sympathovagal imbalance, systemic inflammation, and impaired glucose metabolism, alongside elevated cancer risk and mood disturbance (James et al., 2017). Collectively, the shift-work literature demonstrates that the timing of sleep relative to the endogenous circadian clock constitutes a metabolic and cardiovascular risk factor that is at least partly independent of sleep quantity, a distinction with direct relevance for the roughly one-fifth of the workforce in industrialized economies engaged in shift work (James et al., 2017).
5. Sleep, Immune Function, and Cerebral Homeostasis
5.1 Sleep and Adaptive Immunity: Evidence from Vaccination Studies
Sleep and the immune system share a bidirectional regulatory relationship: infection and inflammatory challenge alter sleep architecture, while sleep, in turn, modulates the efficiency of the adaptive immune response (Bollinger et al., 2010; Ferreira et al., 2025). The clearest experimental demonstration of this relationship comes from vaccination studies, which use antibody titer as an objective, clinically meaningful readout of immune competence. Individuals reporting fewer hours of actigraphy-measured sleep in the nights surrounding hepatitis B vaccination have been shown to exhibit lower secondary antibody responses months later, independent of age, sex, and body mass index (Prather et al., 2012), and experimental work has shown that slow-wave sleep occurring the night following vaccination specifically boosts antigen-specific, T-cell-mediated immunological memory (Lange et al., 2011). A 2023 meta-analysis formally synthesizing this literature across seven experimental and observational studies of influenza and hepatitis vaccination found that habitual sleep duration below six hours in the days surrounding vaccination was associated with a substantially blunted antibody response, though the effect was more consistent in men than in women, a sex disparity the authors note remains mechanistically unexplained (Spiegel et al., 2023). Mechanistic work in animal models corroborates and extends these human findings: chronic sleep fragmentation prior to and during influenza vaccination in mice markedly impaired antibody responses, reduced neutralizing antibody titers, and diminished survival following lethal viral challenge, directly linking a specific pattern of sleep disruption — fragmentation, as distinct from simple curtailment — to compromised vaccine-conferred protection (Guan et al., 2026). Together, this body of evidence establishes sleep not merely as a correlate of immune status but as a modifiable determinant of the efficacy of one of medicine's most consequential public health interventions.
5.2 Sleep and Inflammation
Inflammation is a plausible cross-domain mechanism connecting sleep disturbance with cardiometabolic, immune, and psychiatric outcomes. An earlier meta-analysis of 72 studies and more than 50,000 participants found that sleep disturbance and long sleep duration were associated with higher C-reactive protein and interleukin-6, while findings for short sleep duration and for acute experimental deprivation were less consistent (Irwin et al., 2016). An updated meta-analysis of 35 experimental human studies found that multiple consecutive nights of partial sleep deprivation significantly increased interleukin-6 and C-reactive protein, whereas a single night of total or partial deprivation did not (Ballesio et al., 2025). Repeated partial restriction may therefore be more biologically informative than a single night of total deprivation because it more closely resembles common real-world sleep loss. Inflammation should nevertheless be treated as one mediator among several, and bidirectionality remains important because inflammatory signaling can itself disrupt sleep.
5.3 The Glymphatic System: Discovery, Mechanism, and Sleep-Dependence
Perhaps the most significant advance in sleep neuroscience of the past decade has been the discovery and mechanistic elaboration of the glymphatic system, a glial-dependent perivascular pathway proposed to clear interstitial solutes and metabolic waste products from the brain parenchyma. Cerebrospinal fluid flows through periarterial spaces into the brain parenchyma and exchanges with interstitial fluid, and early work reported that this exchange, and the associated clearance of solutes including amyloid-beta, is markedly more efficient during sleep or anesthesia than during wakefulness, a difference originally attributed to a substantial expansion of the interstitial space during sleep. A major recent mechanistic advance clarified a candidate physiological driver of this proposed sleep-dependent clearance: using fiber photometry in freely moving mice, Hauglund and colleagues demonstrated that infraslow oscillations in norepinephrine released from the locus coeruleus during non-REM sleep drive rhythmic vasomotion — coordinated constriction and dilation of cerebral blood vessels — that functions as a pump propelling cerebrospinal fluid through the brain; critically, the common sleep aid zolpidem suppressed these norepinephrine oscillations and correspondingly reduced glymphatic clearance, suggesting that not all pharmacologically induced sleep confers equivalent restorative, waste-clearing benefit (Hauglund et al., 2025). In humans, sleep deprivation has been reported to measurably impair molecular clearance from the brain, with glymphatic activity reported to be highest during slow-wave, N3-stage non-REM sleep, tracking cortical delta-wave activity (Shirolapov et al., 2024).
5.4 Glymphatic Clearance, Amyloid-Beta, and Neurodegenerative Risk
The clinical significance of glymphatic physiology derives substantially from its proposed role in clearing amyloid-beta and tau, the proteinopathies central to Alzheimer's disease pathogenesis. Using positron emission tomography, Shokri-Kojori and colleagues (2018) demonstrated that a single night of total sleep deprivation was sufficient to produce measurable increases in amyloid-beta burden in brain regions vulnerable to Alzheimer's disease pathology, providing direct human evidence that acute sleep loss transiently increases the brain's amyloid burden. A subsequent randomized crossover trial using a novel multimodal device to trace glymphatic outflow found that normal sleep, relative to experimentally enforced sleep deprivation, increased the overnight efflux of amyloid-beta and tau from cerebrospinal fluid into peripheral plasma, directly linking sleep-dependent physiological processes to the clearance of Alzheimer's disease biomarkers in living humans (Dagum et al., 2026). Consistent with a bridging role between sleep disruption and neuropsychiatric and neurodegenerative disease, impaired glymphatic clearance has also been implicated in neuroinflammatory processes relevant to depression, bipolar disorder, and schizophrenia, and reduced glymphatic function has been reported in patients with chronic insomnia relative to controls (Shirolapov et al., 2024; Xu et al., 2026).
The glymphatic literature, however, remains actively contested and should not be treated as settled physiology. Although several studies support greater cerebrospinal fluid dynamics or metabolite clearance during sleep, a 2024 Nature Neuroscience study directly measuring clearance and molecular movement in mice reported that brain clearance was, in fact, markedly reduced rather than increased during both sleep and anesthesia — a finding that directly challenges the central version of the sleep-enhanced-clearance hypothesis (Miao et al., 2024). Differences in tracer properties, anatomical compartments, anesthetic regimen, and the distinction between fluid movement and net solute clearance may account for part of this disagreement. Accordingly, the association between disrupted sleep and neurodegenerative pathology — which is reasonably well supported by human PET and biomarker studies — should be regarded as better established than the specific claim that impaired glymphatic clearance is the principal causal mediator of that association, a claim that remains genuinely unresolved as of this writing.
6. Sleep, Cognition, and Mental Health
6.1 Neural Oscillations and Memory Consolidation
Sleep is not merely permissive of cognitive recovery but is mechanistically implicated in the active consolidation of memory. According to active systems consolidation theory, information encoded during wakefulness is progressively transferred from hippocampal short-term stores to distributed cortical networks during non-REM sleep, a process orchestrated by the temporal coupling of three characteristic sleep oscillations: cortical slow oscillations (0.16–4 Hz), thalamocortical sleep spindles (8–16 Hz), and hippocampal sharp-wave ripples (80–300 Hz) (Ng et al., 2025). A recent Bayesian meta-analysis synthesizing 23 studies and 297 effect sizes provided quantitative confirmation of this model, finding that the precision and strength of coupling between slow oscillations and fast sleep spindles — particularly over frontal cortex — reliably and significantly predicted subsequent memory retention, moderated by age and memory type (Ng et al., 2025). Although the proportion of variance in memory performance explained by any single coupling metric was modest (on the order of 0.5%), the consistency of the association across studies, populations, and analytic approaches supports slow-oscillation–spindle coupling as a genuine, generalizable physiological mechanism underlying sleep-dependent memory consolidation, rather than an artifact of any single experimental paradigm (Ng et al., 2025).
6.2 Sleep Restriction and Cognitive Performance: Dose-Response Evidence
Beyond memory consolidation specifically, chronic partial sleep restriction produces cumulative, dose-dependent deficits across multiple domains of waking cognitive performance. In a now-classic laboratory study, Van Dongen, Maislin, Mullington, and Dinges (2003) restricted healthy adults to 4, 6, or 8 hours of nightly sleep opportunity for fourteen consecutive days and found that both the 4-hour and 6-hour restriction conditions produced significant, cumulative, dose-dependent deficits in sustained attention and cognitive performance that continued to worsen across the two-week period without evidence of a plateau, with the 6-hour condition eventually producing performance deficits equivalent to those observed after up to two nights of total sleep deprivation. Notably, subjective sleepiness ratings did not track these accumulating objective deficits, plateauing early even as performance continued to decline — a dissociation with troubling practical implications, as it suggests that chronically sleep-restricted individuals may be poor judges of their own cognitive impairment (Van Dongen et al., 2003). Subsequent work has confirmed that sustained attention is reliably impaired in a sleep-dose-dependent manner across numerous independent laboratories, with executive-function deficits somewhat less consistently observed.
6.3 Obstructive Sleep Apnea as a Natural Model of Fragmented Sleep
Obstructive sleep apnea (OSA), a disorder characterized by repeated upper-airway obstruction, intermittent hypoxia, and consequent sleep fragmentation, offers a valuable naturalistic model for isolating the cognitive consequences of chronically disrupted sleep architecture. A 2025 systematic review and meta-analysis of 23 studies encompassing more than 33,000 individuals with OSA found a pooled cognitive-impairment prevalence of 36.9%, increasing monotonically with apnea severity from 32.2% in mild cases to 44.5% in severe cases. A focused meta-analysis of executive function specifically found that OSA reliably impairs multiple executive subdomains, effects that showed only partial reversal following continuous positive airway pressure treatment (Olaithe & Bucks, 2013). These findings converge with, and extend, the broader sleep-restriction literature by demonstrating that fragmentation and intermittent hypoxia — rather than reduced sleep duration alone — constitute an independent pathway to cognitive impairment, with attention and memory deficits shared between OSA and simple sleep deprivation, while visuospatial deficits appear relatively unique to OSA and its hypoxic component.
6.4 Insomnia and Depression: Bidirectionality and Prospective Risk
The relationship between sleep disturbance and depression is bidirectional, but the evidence for insomnia as an independent prospective risk factor for incident depression — rather than merely a prodromal symptom — is now substantial. A meta-analysis of 34 cohort studies encompassing more than 170,000 participants found that insomnia was associated with more than double the risk of subsequently developing depression (pooled RR = 2.27, 95% CI 1.89–2.71). An earlier meta-analytic evaluation restricted specifically to non-depressed individuals at baseline similarly found an approximately twofold increased risk of incident depression among those with insomnia (Baglioni et al., 2011), a finding that has since been replicated and extended with additional cohorts published through 2022, with the updated pooled effect estimate, if anything, larger than in the original synthesis (Hertenstein et al., 2023). This consistency across independent research groups and expanding participant samples has led sleep researchers to characterize insomnia as a general transdiagnostic risk factor for psychopathology, rather than a depression-specific phenomenon (Hertenstein et al., 2023).
6.5 Circadian Rhythms and Bipolar Disorder
Circadian and sleep-wake instability constitute a defining, and possibly causally central, feature of bipolar disorder rather than a mere epiphenomenon of mood state. Alterations in sleep pattern — decreased need for sleep as a hallmark of mania and insomnia during depressive episodes — have been shown to precede corresponding shifts in mood, and circadian disturbances (delayed chronotype, irregular sleep-wake timing, and abnormalities of melatonin secretion) persist even during euthymic, mood-stable periods, supporting their characterization as a trait marker rather than solely a state-dependent symptom. The social zeitgeber theory offers one influential mechanistic account of how this vulnerability is triggered: life events that disrupt daily social routines and sleep-wake schedules are proposed to desynchronize the endogenous circadian pacemaker, in turn precipitating mood episodes in vulnerable individuals (Grandin, Alloy, & Abramson, 2006). Consistent with this model, therapies that explicitly stabilize daily social and sleep-wake rhythms, such as interpersonal and social rhythm therapy, have demonstrated efficacy in improving the longitudinal course of bipolar disorder, providing indirect but clinically meaningful support for circadian instability as a modifiable contributor to the disorder's pathophysiology rather than solely its downstream consequence.
6.6 Treating Sleep to Treat Mental Illness: Evidence from Cognitive Behavioral Therapy for Insomnia
If insomnia is a causal contributor to psychiatric symptomatology rather than merely a correlate, treating insomnia directly should, at least in principle, confer benefit for comorbid psychiatric conditions — a prediction with direct clinical and public-health significance. Cognitive behavioral therapy for insomnia (CBT-I) is a structured, multi-component intervention combining stimulus control, sleep-restriction therapy, and cognitive restructuring, and is endorsed as a first-line treatment for insomnia by major clinical practice guidelines. A meta-analysis of long-term outcomes across 30 randomized controlled trials found that CBT-I produced moderate-to-large, durable improvements in insomnia severity, sleep-onset latency, and sleep efficiency that persisted, albeit with some attenuation, at three, six, and twelve months post-treatment (van der Zweerde et al., 2019). In populations with comorbid insomnia and depression, systematic evidence indicates that CBT-I is reliably superior to control conditions for improving insomnia outcomes, with corresponding improvements in comorbid depression symptoms observed in several, though not all, included trials.
Recent randomized prevention trials strengthen the causal interpretation considerably. In 291 older adults with insomnia but without current major depression, two months of CBT-I reduced incident and recurrent major depressive disorder relative to an active sleep-education control across 36 months of follow-up, with sustained insomnia remission associated with an 82.6% lower likelihood of depression (Irwin et al., 2022). In a randomized trial of 708 young people in Hong Kong and mainland China with insomnia disorder and subclinical depressive symptoms, six weeks of app-based CBT-I reduced new-onset major depression over 12 months from 18% in the health-education group to 10% in the intervention group, corresponding to a hazard ratio of 0.58 (Chen et al., 2025). These trials do not imply that insomnia is a sufficient cause of depression, but they provide stronger evidence than prospective association alone that treating sleep disturbance can alter psychiatric risk in selected high-risk populations.
7. Integrative Discussion: Toward a Unified Physiological Model
The evidence reviewed above, though drawn from disciplines that rarely cite one another — endocrinology, cardiology, immunology, neuroscience, and psychiatry — converges on a small number of shared physiological mechanisms, suggesting that sleep functions as a unified upstream regulator rather than as a set of independent, domain-specific processes. Three mechanistic threads recur across domains. First, sympathetic nervous system activation and hypothalamic-pituitary-adrenal axis dysregulation — elevated cortisol and catecholamine tone following sleep loss — appear implicated in both the insulin resistance documented in the metabolic literature and the endothelial dysfunction and hypertension documented in the cardiovascular literature (Rogers et al., 2024; Scheer et al., 2009). Second, low-grade systemic and neuroinflammation recur as a mechanistic node linking sleep loss to cardiovascular disease, to blunted vaccine-induced immunity, and — via the emerging, though contested, connection between glymphatic clearance and neuroinflammatory processes — to psychiatric disorders including depression and schizophrenia (Bollinger et al., 2010; Irwin et al., 2016; Shirolapov et al., 2024). Third, the norepinephrine-driven vasomotor mechanism recently identified as a candidate proximate driver of glymphatic clearance (Hauglund et al., 2025) offers a plausible physiological bridge between the autonomic-nervous-system disturbances documented in the metabolic and cardiovascular literatures and the neurodegenerative risk documented in the glymphatic literature — a connection that, to date, has not been directly tested and that must now be read alongside direct experimental evidence against the underlying clearance hypothesis (Miao et al., 2024).
At the same time, several methodological considerations temper the strength of causal inference that can be drawn from this literature and warrant explicit discussion. First, the majority of the epidemiological evidence relies on self-reported sleep duration, which correlates only moderately with objectively measured (actigraphic or polysomnographic) sleep and is subject to recall and social-desirability bias; associations observed with self-report measures may therefore either overstate or understate true effect sizes. Second, the U-shaped relationship observed between sleep duration and numerous outcomes — obesity, cardiovascular disease, and mortality — is considerably better understood on its short-sleep arm, where experimental sleep-restriction studies provide convergent mechanistic support, than on its long-sleep arm, where reverse causation from undiagnosed illness, depression, or sleep-disordered breathing (which can paradoxically increase reported time in bed) remains a plausible, and likely substantial, alternative explanation. Third, the relationship between insomnia and depression, and between sleep and psychiatric illness more generally, is genuinely bidirectional, and while the CBT-I intervention literature — including two independent randomized depression-prevention trials (Chen et al., 2025; Irwin et al., 2022) — now provides some of the strongest available causal evidence that treating sleep improves psychiatric outcomes, the mixed results for depression-specific outcomes in some comorbid-population trials indicate that insomnia is best conceptualized as one causal contributor among several, rather than a sole or sufficient cause. Fourth, much of the experimental sleep-restriction literature, for practical and ethical reasons, has been conducted in young, healthy, predominantly male samples over relatively short durations (days to two weeks), raising open questions about generalizability to women, older adults, and populations with existing metabolic or psychiatric vulnerability, as well as to the years-long timescale over which real-world chronic sleep restriction typically operates.
A useful way to interpret the field is to separate three evidentiary levels. Association identifies clinically important relationships but remains vulnerable to confounding. Mechanistic experiments establish that sleep manipulation can alter proximal physiology, but their short duration and selected samples limit direct extrapolation to chronic disease. Intervention trials provide the strongest test of modifiability, yet they remain relatively scarce for hard clinical endpoints. Claims in sleep medicine should therefore be strongest where all three levels converge — as they now do for insomnia and depression — and more cautious where mechanistic plausibility substantially exceeds intervention evidence, as is currently the case for the glymphatic-clearance account of neurodegenerative risk.
8. Clinical and Public Health Implications
The convergent evidence reviewed here carries direct implications for clinical practice and public health policy. Clinically, the evidence supports incorporating routine, multidimensional sleep assessment — not duration alone, but regularity, timing, continuity, daytime function, and screening for sleep disorders such as OSA and insomnia — into primary care and psychiatric practice on par with the well-established practice of screening for diet and physical activity, given that sleep disturbance functions in the reviewed literature as both a presenting complaint and a modifiable contributor to the metabolic, cardiovascular, and psychiatric conditions clinicians are already treating (Robbins et al., 2024; St-Onge et al., 2025). The demonstrated efficacy and durability of CBT-I, together with the new depression-prevention trial evidence, positions it as a scalable, non-pharmacological intervention with plausible benefit that extends beyond sleep outcomes narrowly construed (Chen et al., 2025; Irwin et al., 2022; van der Zweerde et al., 2019). At the population level, the persistently high prevalence of short sleep duration documented by national surveillance — 30.5% of U.S. adults in the most recent (2024) cycle of the National Health Interview Survey (Ng et al., 2026) — alongside evidence that shift work, circadian misalignment, and sleep irregularity carry cardiometabolic risks that are at least partly independent of sleep duration (Scheer et al., 2009; Windred et al., 2024), suggests that purely individual-level interventions are unlikely to be sufficient. Structural determinants of sleep health, including school and work start times, shift-scheduling practices, and environmental light and noise exposure, represent public-health targets that are, at present, addressed far less systematically than analogous structural determinants of diet (e.g., the food environment) or physical activity (e.g., the built environment).
9. Limitations of the Present Review
As a narrative rather than systematic review, this synthesis did not employ a pre-registered search strategy, dual independent screening, or formal risk-of-bias assessment across the full body of the literature, and it is therefore subject to selection bias in the studies emphasized. The review draws heavily on English-language, predominantly high-income-country literature, which may limit generalizability to populations with differing sleep ecology, work structures, and access to sleep medicine. The Sleep Systems Integrity model proposed in Section 3 is offered as an organizing heuristic rather than an independently tested causal model, and its specific predictions have not yet been formally evaluated against competing frameworks. Finally, because sleep science is a rapidly advancing and, in places, actively contested field — the glymphatic-clearance mechanism identified by Hauglund and colleagues (2025) was published only in the past two years, and a direct experimental challenge to the underlying clearance hypothesis appeared within the same period (Miao et al., 2024) — readers should anticipate that specific mechanistic claims in the neuroscience sections of this review are considerably more likely to be revised by subsequent research than the more mature epidemiological literature on sleep duration, regularity, and cardiometabolic outcomes.
10. Conclusion
Sleep is not a unitary exposure, and health is not determined by sleep duration alone. Across epidemiology, laboratory experimentation, mechanistic neuroscience, and clinical intervention, the strongest synthesis is that sleep functions as a multidimensional regulatory system: duration, regularity, timing, continuity, architecture, and sleep disorders influence overlapping metabolic, autonomic, immune, neural, and affective pathways. Short sleep and circadian misalignment have substantial mechanistic and, increasingly, interventional support in cardiometabolic research; sleep fragmentation and insufficient sleep can impair adaptive immunity and promote inflammatory signaling; sleep supports memory and sustained attention; and insomnia is a prospective, and in selected populations now demonstrably modifiable, risk factor for depression. At the same time, several prominent claims require restraint. Long-sleep associations are particularly vulnerable to reverse causation, laboratory sleep restriction cannot be equated with years of real-world exposure, and the glymphatic explanation for sleep-related neurodegenerative risk remains genuinely contested rather than established. The Sleep Systems Integrity model proposed here therefore treats sleep as a foundational but heterogeneous health system rather than a single behavior measured in hours. Clinically, this supports routine assessment of sleep alongside diet and physical activity, with attention to regularity, timing, continuity, daytime function, and sleep disorders. Scientifically, the next step is not merely to accumulate more associations, but to determine which sleep dimensions are causal for which outcomes, in whom, through which mechanisms, and whether modifying those dimensions produces durable reductions in disease risk.
References
Antza, C., Kostopoulos, G., Mostafa, S., Nirantharakumar, K., & Tahrani, A. (2021). The links between sleep duration, obesity and type 2 diabetes mellitus. Journal of Endocrinology, 252(2), 125–141. https://doi.org/10.1530/JOE-21-0155
Baglioni, C., Battagliese, G., Feige, B., Spiegelhalder, K., Nissen, C., Voderholzer, U., Lombardo, C., & Riemann, D. (2011). Insomnia as a predictor of depression: A meta-analytic evaluation of longitudinal epidemiological studies. Journal of Affective Disorders, 135(1–3), 10–19. https://doi.org/10.1016/j.jad.2011.01.011
Ballesio, A., Fiori, V., & Lombardo, C. (2025). Effects of experimental sleep deprivation on peripheral inflammation: An updated meta-analysis of human studies. Journal of Sleep Research. Advance online publication. https://doi.org/10.1111/jsr.70099
Bollinger, T., Bollinger, A., Oster, H., & Solbach, W. (2010). Sleep, immunity, and circadian clocks: A mechanistic model. Gerontology, 56(6), 574–580. https://doi.org/10.1159/000281827
Buysse, D. J. (2014). Sleep health: Can we define it? Does it matter? Sleep, 37(1), 9–17. https://doi.org/10.5665/sleep.3298
Chen, S.-J., Que, J.-Y., Chan, N. Y., Shi, L., Li, S. X., Chan, J. W. Y., Huang, W., Chen, C. X., Tsang, C. C., Ho, Y. L., Morin, C. M., Zhang, J.-H., Lu, L., & Wing, Y. K. (2025). Effectiveness of app-based cognitive behavioral therapy for insomnia on preventing major depressive disorder in youth with insomnia and subclinical depression: A randomized clinical trial. PLOS Medicine, 22(1), e1004510. https://doi.org/10.1371/journal.pmed.1004510
Dagum, P., Elbert, D. L., Giovangrandi, L., Singh, T., Venkatesh, V. V., Corbellini, A., Kaplan, R. M., Levendovszky, S. R., Ludington, E., Yarasheski, K., Lowenkron, J., VandeWeerd, C., Lim, M. M., & Iliff, J. J. (2026). The glymphatic system clears amyloid beta and tau from brain to plasma in humans. Nature Communications, 17(1), Article 715. https://doi.org/10.1038/s41467-026-68374-8
Ferreira, R. C. M., Ruiz, F. S., & de Mello, M. T. (2025). Human sleep and immunity: The role of circadian patterns. Handbook of Clinical Neurology, 206, 93–103.
Grandin, L. D., Alloy, L. B., & Abramson, L. Y. (2006). The social zeitgeber theory, circadian rhythms, and mood disorders: Review and evaluation. Clinical Psychology Review, 26(6), 679–694. https://doi.org/10.1016/j.cpr.2006.07.001
Guan, M., Gu, W., Gileles-Hillel, A., Khalyfa, A., Balamalaliyage, P., Tunterak, W., Li, M., Huang, Y., Wen, Y., Driver, J., Gozal, D., & Wan, X. (2026). Poor sleep impairs immune responses and influenza vaccine protection. Nature Communications, 17(1).
Hartstein, L. E., Grandner, M. A., & Diniz Behn, C. (2025). Sleep irregularity and circadian rhythmicity: Implications for health and well-being. Current Sleep Medicine Reports. Advance online publication.
Hauglund, N. L., Andersen, M., Tokarska, K., Radovanovic, T., Kjaerby, C., Sørensen, F. L., Bojarowska, Z., Untiet, V., Ballestero, S. B., Kolmos, M. G., Weikop, P., Hirase, H., & Nedergaard, M. (2025). Norepinephrine-mediated slow vasomotion drives glymphatic clearance during sleep. Cell, 188(3), 606–622.e17. https://doi.org/10.1016/j.cell.2024.11.027
Hertenstein, E., Trinca, E., Wunderlin, M., Schneider, C. L., Züst, M. A., Fehér, K. D., Su, T., van Straten, A., Berger, T., Baglioni, C., Johann, A., Spiegelhalder, K., Riemann, D., Feige, B., & Nissen, C. (2023). Insomnia—A risk factor for mental disorders. Journal of Sleep Research, 32(6), Article e13930. https://doi.org/10.1111/jsr.13930
Irwin, M. R., Carrillo, C., Sadeghi, N., Bjurstrom, M. F., Breen, E. C., & Olmstead, R. (2022). Prevention of incident and recurrent major depression in older adults with insomnia: A randomized clinical trial. JAMA Psychiatry, 79(1), 33–41. https://doi.org/10.1001/jamapsychiatry.2021.3422
Irwin, M. R., Olmstead, R., & Carroll, J. E. (2016). Sleep disturbance, sleep duration, and inflammation: A systematic review and meta-analysis of cohort studies and experimental sleep deprivation. Biological Psychiatry, 80(1), 40–52. https://doi.org/10.1016/j.biopsych.2015.05.014
James, S. M., Honn, K. A., Gaddameedhi, S., & Van Dongen, H. P. A. (2017). Shift work: Disrupted circadian rhythms and sleep—Implications for health and well-being. Current Sleep Medicine Reports, 3(2), 104–112. https://doi.org/10.1007/s40675-017-0071-6
Kohanmoo, A., Akhlaghi, M., Sasani, N., Nouripour, F., Lombardo, C., & Kazemi, A. (2024). Short sleep duration is associated with higher risk of central obesity in adults: A systematic review and meta-analysis of prospective cohort studies. Obesity Science & Practice, 10(3), Article e772. https://doi.org/10.1002/osp4.772
Lange, T., Dimitrov, S., Bollinger, T., Diekelmann, S., & Born, J. (2011). Sleep after vaccination boosts immunological memory. The Journal of Immunology, 187(1), 283–290. https://doi.org/10.4049/jimmunol.1100015
Lloyd-Jones, D. M., Allen, N. B., Anderson, C. A. M., Black, T., Brewer, L. C., Foraker, R. E., Grandner, M. A., Lavretsky, H., Perak, A. M., Sharma, G., & Rosamond, W. (2022). Life's Essential 8: Updating and enhancing the American Heart Association's construct of cardiovascular health. Circulation, 146(5), e18–e43. https://doi.org/10.1161/CIR.0000000000001078
Meyer, N., Lok, R., Schmidt, C., Kyle, S. D., McClung, C. A., Cajochen, C., Scheer, F. A. J. L., Jones, M. W., & Chellappa, S. L. (2024). The sleep-circadian interface: A window into mental disorders. Proceedings of the National Academy of Sciences, 121(9), Article e2214756121. https://doi.org/10.1073/pnas.2214756121
Miao, A., Luo, T., Hsieh, B., Edge, C. J., Gridley, M., Wong, R. T. C., Constandinou, T. G., Wisden, W., & Franks, N. P. (2024). Brain clearance is reduced during sleep and anesthesia. Nature Neuroscience, 27(6), 1046–1050. https://doi.org/10.1038/s41593-024-01638-y
Ng, A. E., Black, L. I., & Adjaye-Gbewonyo, D. (2026). Short sleep duration and sleep difficulties among adults: United States, 2024 (NCHS Data Brief No. 559). National Center for Health Statistics. https://www.cdc.gov/nchs/products/databriefs/db559.htm
Ng, T., Noh, E., et al. (2025). Bayesian meta-analysis reveals the mechanistic role of slow oscillation–spindle coupling in sleep-dependent memory consolidation. eLife, 13, Article e101992. https://doi.org/10.7554/eLife.101992
Olaithe, M., & Bucks, R. S. (2013). Executive dysfunction in OSA before and after treatment: A meta-analysis. Sleep, 36(9), 1297–1305. https://doi.org/10.5665/sleep.2950
Prather, A. A., Hall, M., Fury, J. M., Ross, D. C., Muldoon, M. F., Cohen, S., & Manuck, S. B. (2012). Sleep and antibody response to hepatitis B vaccination. Sleep, 35(8), 1063–1069. https://doi.org/10.5665/sleep.1990
Robbins, R., Quan, S. F., & Weaver, M. D. (2024). Sleep health. NEJM Evidence, 3(8), Article EVIDra2300269. https://doi.org/10.1056/EVIDra2300269
Rogers, E. M., Banks, N. F., & Jenkins, N. D. M. (2024). The effects of sleep disruption on metabolism, hunger, and satiety, and the influence of psychosocial stress and exercise: A narrative review. Diabetes/Metabolism Research and Reviews, 40(2), Article e3667. https://doi.org/10.1002/dmrr.3667
Scheer, F. A. J. L., Hilton, M. F., Mantzoros, C. S., & Shea, S. A. (2009). Adverse metabolic and cardiovascular consequences of circadian misalignment. Proceedings of the National Academy of Sciences, 106(11), 4453–4458. https://doi.org/10.1073/pnas.0808180106
Shirolapov, I. V., Zakharov, A. V., et al. (2024). The role of the glymphatic clearance system in the mechanisms of the interactions of the sleep–waking cycle and the development of neurodegenerative processes. Neuroscience and Behavioral Physiology. https://doi.org/10.1007/s11055-024-01585-y
Shokri-Kojori, E., Wang, G. J., Wiers, C. E., Demiral, S. B., Guo, M., Kim, S. W., Lindgren, E., Ramirez, V., Zehra, A., Freeman, C., Miller, G., Manza, P., Srivastava, T., De Santi, S., Tomasi, D., Benveniste, H., & Volkow, N. D. (2018). β-Amyloid accumulation in the human brain after one night of sleep deprivation. Proceedings of the National Academy of Sciences, 115(17), 4483–4488. https://doi.org/10.1073/pnas.1721694115
Spiegel, K., Leproult, R., & Van Cauter, E. (1999). Impact of sleep debt on metabolic and endocrine function. The Lancet, 354(9188), 1435–1439. https://doi.org/10.1016/S0140-6736(99)01376-8
Spiegel, K., Tasali, E., Penev, P., & Van Cauter, E. (2004). Brief communication: Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Annals of Internal Medicine, 141(11), 846–850. https://doi.org/10.7326/0003-4819-141-11-200412070-00008
Spiegel, K., Rey, A. E., Cheylus, A., Ayling, K., Benedict, C., Lange, T., Prather, A. A., Taylor, D. J., & Van Cauter, E. (2023). A meta-analysis of the associations between insufficient sleep duration and antibody response to vaccination. Current Biology, 33(6), 998–1005.e2. https://doi.org/10.1016/j.cub.2023.02.017
St-Onge, M. P., et al. (2025). Multidimensional sleep health: Definitions and implications for cardiometabolic health: A scientific statement from the American Heart Association. Circulation: Cardiovascular Quality and Outcomes, 18(5), e000139. https://doi.org/10.1161/HCQ.0000000000000139
Tasali, E., Wroblewski, K., Kahn, E., Kilkus, J., & Schoeller, D. A. (2022). Effect of sleep extension on objectively assessed energy intake among adults with overweight in real-life settings: A randomized clinical trial. JAMA Internal Medicine, 182(4), 365–374. https://doi.org/10.1001/jamainternmed.2021.8098
Van Dongen, H. P. A., Maislin, G., Mullington, J. M., & Dinges, D. F. (2003). The cumulative cost of additional wakefulness: Dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep, 26(2), 117–126. https://doi.org/10.1093/sleep/26.2.117
van der Zweerde, T., Bisdounis, L., Kyle, S. D., Lancee, J., & van Straten, A. (2019). Cognitive behavioral therapy for insomnia: A meta-analysis of long-term effects in controlled studies. Sleep Medicine Reviews, 48, Article 101208. https://doi.org/10.1016/j.smrv.2019.08.002
Watson, N. F., Badr, M. S., Belenky, G., Bliwise, D. L., Buxton, O. M., Buysse, D., Dinges, D. F., Gangwisch, J., Grandner, M. A., Kushida, C., Malhotra, R. K., Martin, J. L., Patel, S. R., Quan, S. F., & Tasali, E. (2015). Recommended amount of sleep for a healthy adult: A joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep, 38(6), 843–844. https://doi.org/10.5665/sleep.4716
Windred, D. P., Burns, A. C., Lane, J. M., Saxena, R., Rutter, M. K., Cain, S. W., & Phillips, A. J. K. (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep, 47(1), zsad253. https://doi.org/10.1093/sleep/zsad253
Wu, Y., Zhai, L., & Zhang, D. (2014). Sleep duration and obesity among adults: A meta-analysis of prospective studies. Sleep Medicine, 15(12), 1456–1462. https://doi.org/10.1016/j.sleep.2014.07.018
Xu, Y., et al. (2026). The glymphatic system in sleep: A nexus of waste clearance, brain homeostasis, and disease intervention. Molecular Psychiatry. Advance online publication.