Sleep duration and risk of cancer incidence and mortality: a pooled analysis of six population-based cohorts in Japan
Menée à partir des données de 6 cohortes japonaises portant sur 271 694 personnes, cette étude analyse l'association entre la durée du sommeil et l'incidence des cancers (40 751 cas) ou la mortalité spécifique (18 323 décès), en fonction du sexe
Sleep duration is emerging as an important modifiable risk factor for morbidity and mortality. We assessed the association between sleep duration and cancer incidence and mortality among Japanese adults using data from six population-based cohorts with 271 694 participants. During a total follow-up period of about 5.9 million person-years, we identified 40 751 incident cancer cases and 18 323 cancer deaths. We computed study-specific hazard ratios (HRs) and 95% confidence intervals (CIs) using Cox proportional hazards regression models and pooled the estimates using random-effects meta-analysis. Sleep duration of ≥ 10 h (vs. 7 h) was associated with increased risk of cancer incidence among women (HR 1.19, 95% CI 1.02-1.38), but not men, and increased risk of cancer mortality among men (HR 1.18, 95% CI 1.00-1.39) and women (HR 1.44, 95% CI 1.20-1.73). Sleep duration of ≤ 5 h (vs. 7 h) was not associated with cancer incidence and mortality. However, among post-menopausal women, sleep durations of both ≤ 5 h and ≥ 10 h (vs. 7 h) were associated with an increased risk of cancer mortality. Among Japanese adults, sleep duration of ≥ 10 h is associated with increased risk of cancer incidence and mortality among women and cancer mortality among men.