Physical Wellness

An orientation to how much sleep adults need and why sleeping outside that range in either direction is tied to higher mortality and disease risk.

Roughly one-third of US adults report short sleep, only about 15% consistently sleep 7–9 hours on most nights by objective monitoring, whether duration is declining is a genuine self-report-versus-objective conflict, and short and extreme sleep are consistently more prevalent among Black, American Indian, Alaska Native, Native Hawaiian and Pacific Islander adults and those with lower education or income.

Cognitive behavioural therapy for insomnia is a structured programme delivered by a trained clinician, not something you assemble yourself. van Straten 2017 (meta-analysis, 87 randomised trials) reports it effective in treating insomnia, whether as components or the full package, and Baglioni 2020 (network meta-analysis) reports that none of the interventions it compared were “superior to CBT-I”.

Soh 2020 (meta-analysis, 33 studies) reports that digital CBT-I reduced the Insomnia Severity Index by 5.00 points against control after treatment and by 3.48 at one year, with face-to-face CBT-I producing greater improvement; Hwang 2025 reports fully automated digital CBT-I less effective than therapist-assisted CBT-I; Linardon 2024 reports higher dropout for apps; and Le 2025 reports healthcare and societal cost savings for digital CBT-I against face-to-face care.

Sella 2022 reports that only physical activity improved objectively measured sleep in older adults, Riedel 2024 reports exercise improved objective and subjective sleep in insomnia with larger effects at higher intensity, and the reviews of other non-drug approaches report results ranging from effective (acupuncture, Hu 2023) to mixed (nutrition, Knowlden 2016) and inconclusive (single-plant herbal aids, Kim 2018).

Duffy 1996 (randomised trial) concludes that light is the primary synchronizer of the human circadian pacemaker; van Maanen 2016 (meta-analysis, 53 studies) reports light therapy effective for sleep problems overall, with most effect sizes small to medium; Xu 2022 reports higher light at night associated with a 22% higher prevalence of sleep problems; and the reviews of blue-blocking glasses differ, from substantial evidence of reduced sleep-onset latency (Hester 2021) to no significant difference from standard lenses (Lawrenson 2017).

Cohort syntheses report long daytime napping of an hour or more associated with cardiovascular risk and mortality (Sun 2022, which describes the evidence for most meta-analyses as weak or suggestive), coronary heart disease (Salari 2022) and all-cause mortality (Liu 2015), and daytime napping associated with stroke in a dose-dependent manner (Jin 2021), while Du 2026 reports that napping under 60 minutes maximised cognitive enhancement (SMD 0.69) and napping over 60 minutes correlated with a 30% higher risk of coronary heart disease.

Dinis 2018 (systematic review) reports poor sleep quality affecting 10-50% of college students, and pooled prevalence meta-analyses report poor sleep quality in 51% of police officers (Garbarino 2019), 69% of military personnel and veterans (Bai 2023) and 52.30% of pregnant women in Ethiopia (Zeleke 2024). Meta-analyses report CBT-I effective in menopausal women (Moon 2025), pregnant women (Zheng 2022; Visvanathan 2025) and daytime workers (Takano 2023), and possibly effective in older adults (Huang 2022); in shift workers Slanger 2016 could not determine whether interventions reduce sleepiness, and Nickel 2025 reports student sleep-education programs improved knowledge and sleep quality but not sleep duration.

Zhao 2013 “suggests a positive association between long sleep duration and colorectal cancer”, Chen 2018 reports long sleep associated with colorectal cancer in categorical but not dose-response analysis, and Li 2022 concludes that abnormal sleep duration “might be linked to several adverse cancer-related outcomes”; long sleep is also associated with cancer mortality (Ma 2016; Wilunda 2022); short sleep with cancer risk in Asian populations in categorical but not dose-response analysis (Chen 2018); and for breast cancer He 2015 reports circadian disruption associated with higher risk and no significant effect for short sleep, Yang 2014 reports light at night associated with higher risk and no significant sleep-duration dose-response, Pagkalidou 2026 reports evening chronotype associated with higher risk, and Zhao 2013 reports long sleep associated with a 21% lower risk of hormone-related cancers such as breast cancer.

Cappuccio 2010 (meta-analysis, 1,382,999 participants) reports both short and long sleep associated with greater risk of death (RR 1.12 and 1.30); Yin 2017 reports each hour below 7 hours increasing all-cause mortality risk by 6% and each hour above by 13%; Lauderdale 2015 (a cross-sectional study of adults aged 62-90) reports that the long-sleep association with fair or poor health may be an artefact of survey-based sleep measurement; and Hall 2015 reports the short-sleep hazard attenuated after adjustment for inflammation and health status.

Ruan 2025 (meta-analysis, 42 randomised trials) reports that sleep hygiene education as a single therapy improves insomnia but is inferior to CBT-I, partial CBT-I, exercise and acupressure, and notes that 85.71% of trials had high risk of bias, so robust evidence from well-designed trials is needed; Chung 2018 reports it significantly less efficacious than CBT-I; and Irish 2014 reports the direct effects of its individual recommendations in the general population largely untested.

When and how consistently you sleep carries health signal beyond total hours: irregular timing tracks with worse cardiometabolic markers, higher dementia risk and higher mortality, and social jet lag with obesity and depression.

Short and long sleep both track with cardiovascular risk, with the lowest risk near seven hours; hypertension associates chiefly with short sleep, stroke risk rises most steeply at the long end, and the one causal-inference test of atrial fibrillation found no effect.

Short sleep is linked with higher risk of type 2 diabetes, obesity and metabolic syndrome, with a U-shape for diabetes and the lowest risk at seven to eight hours.

Scott 2021 (meta-analysis, 65 trials) reports that improving sleep produced medium effects on composite mental health, depression, anxiety and rumination, and Takano 2023 reports CBT-I improved depression and anxiety in daytime workers; CBT-I's effect on depressive symptoms is reported as significant by Palagini 2024 and as a possible positive effect by Ballesio 2018.

Wu 2018 (meta-analysis of cohorts) reports both short and long sleep associated with higher risk of cognitive disorders, with the lowest risk at 7 to 8 hours; Liang 2018 reports a pooled risk ratio per additional hour of 0.99, with 7 hours tending to carry the least risk; and cross-sectional studies report cognitive performance peaking at about 7 hours (Tai 2022; Richards 2017).

Hirshkowitz 2015 reports an expert-panel consensus of 7 to 9 hours for adults and 7 to 8 for older adults; Dzierzewski 2026 reports 74% of 133 meta-analyses consistent with those recommendations and none contradictory; Chaput 2020 reports 7 to 8 hours a day most favourably associated with health; and Li 2021 reports higher mortality and disease risk with each hour of deviation in either direction.

Lower indoor light in the hours before bed: one way to carry out what the review summary this entry was written from reports: that nocturnal light is disruptive, raising sleep problems and lowering overnight melatonin.

Daylight or bright light soon after waking: one way to carry out what the review summary this entry was written from reports: that appropriately timed bright light phase-shifts the clock and advances delayed sleep timing.

If you nap, keep it under 30 to 60 minutes: one way to carry out what the cohort syntheses on napping report, with naps of an hour or more associated with cardiovascular risk and mortality (Sun 2022, which describes the evidence for most meta-analyses as weak or suggestive; Salari 2022; Liu 2015) and Du 2026 reporting that napping under 60 minutes maximised cognitive enhancement (SMD 0.69).

Going to bed and waking at about the same time every day: one way to carry out what the review summary this entry was written from reports: that sleep timing and regularity carry health signal independent of total sleep time.

Regular physical activity: one way to carry out what the reviews of non-drug approaches to sleep studied; Riedel 2024 reports exercise improved objective and subjective sleep in insomnia with larger effects at higher intensity, and Sella 2022 reports physical activity the only intervention to improve objectively measured sleep in older adults.

Scheduling 8 to 9 hours in bed to get 7 to 9 hours of sleep: one way to carry out the 7 to 9 hour consensus for adults (Hirshkowitz 2015), which Dzierzewski 2026 reports as consistent with 74% of 133 meta-analyses and contradicted by none.

Using the bed mainly for sleep and getting up when awake and frustrated: stimulus control, a CBT-I component that Steinmetz 2024 and Verreault 2023 report effective for insomnia in component and network meta-analyses, with Verreault 2023 adding that results should be interpreted cautiously given study quality; on its own here as a habit, not as insomnia treatment.