Cappuccio · Sleep 2010 · systematic review and meta-analysis · n=16 studies (1,382,999 participants)

Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies.

Cited 2238 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis of prospective cohort studies for a prognostic question

PubMed 20469800 · doi:10.1093/sleep/33.5.585 · record verified 2026-08-31

What was done

The authors conducted a systematic review and meta-analysis of prospective cohort studies identified through MEDLINE (1966–2009), EMBASE (from 1980), the Cochrane Library, and manual searches without language restrictions. Studies were included if they were prospective, had more than 3 years of follow-up, assessed baseline sleep duration by questionnaire, and recorded all-cause mortality prospectively via death certification. Relative risks (RR) and 95% confidence intervals (CI) were pooled using random-effects models, with assessments of heterogeneity, publication bias, and sensitivity analyses.

What was found

Sixteen studies comprising 27 independent cohorts, 1,382,999 participants (followed for 4 to 25 years), and 112,566 deaths were analyzed. Pooled short sleep duration was associated with higher all-cause mortality (RR 1.12, 95% CI 1.06 to 1.18, P < 0.01; heterogeneity P = 0.02; publication bias P = 0.74). Long sleep duration was also associated with higher all-cause mortality (RR 1.30, 95% CI 1.22 to 1.38, P < 0.0001; heterogeneity P < 0.0001; publication bias P = 0.18).

Why it matters

This meta-analysis provides large-scale epidemiological evidence that both short and long habitual sleep durations are independent prospective predictors of all-cause mortality.

Limits

Sleep duration was exclusively self-reported via questionnaires rather than objectively measured by actigraphy or polysomnography. Significant statistical heterogeneity was observed across studies, and residual confounding or reverse causation from underlying subclinical comorbidities could not be completely excluded.

Cited by