Long sleep duration and cause‐specific mortality according to physical function and self‐rated health: the <scp>O</scp>hsaki <scp>C</scp>ohort <scp>S</scp>tudy

Although several studies have examined the association between sleep duration and all-cause or cause-specific mortality, it is unclear whether long sleep duration might merely reflect decreased physical strength and poorer health status. We therefore examined the association between sleep duration and all-cause and cause-specific mortality, and conducted stratified analysis based on physical function and self-rated health. This study used prospective data from the Ohsaki Cohort Study, conducted in Miyagi Prefecture, in northern Japan. This study population comprised 49 256 subjects aged 40-79 years at the baseline survey. We estimated hazard ratios (HRs) and 95% confidence intervals (CIs) of all-cause and cause-specific mortality according to the five categories of sleep duration (≤6, 7, 8, 9, ≥10 h day(-1) ), treating 7 h as the reference group, employing Cox's proportional hazard regression analysis. We found that long sleep duration was associated with mortality. The HRs (95% CIs) of subjects who slept more than 10 h were 1.37 (1.27-1.47), 1.49 (1.30-1.71) and 1.53 (1.36-1.73) for mortality due to all causes, total cardiovascular disease and other causes of death mortality, respectively. The association between long sleep duration and stroke mortality was especially marked among subjects with limited physical function and poorer health status. However, we did not observe such a trend for mortality due to all causes, total cardiovascular disease, ischaemic heart disease, cancer or other causes of death. We conclude that, with the exception of stroke mortality, the association between long sleep duration and mortality is not modified by physical function or health status.