LAUSR

Sleep duration and mortality have been a frequent topic in the media and mind of the public. “How long should I sleep?” shows up more than one million times in Google searches. Consequently, it has attracted a substantial amount of research during recent years, and a number of meta-analyses on sleep and health have been published [1–3] showing a U-shaped association, that is, both short sleep and a long sleep are associated with increased risk of premature death, but at rather modest levels, after adjustment for confounders. Some studies showed more of a linear pattern with a high risk for long sleep [4, 5]. The lowest risk is usually found at a sleep duration of 7 h, but also at 6–7 or 7–8 h. Short sleep is usually defined as <6 h and long sleep as >8 h, but studies vary considerably. Also, cause-specific mortality, like cardiovascular mortality, shows a similar pattern, whereas cancer mortality is less established. The association between disturbed sleep and mortality lacks conclusive findings [6]. Sleepiness [7], snoring [8], and evening chronotype [9] show indications of a link with mortality, but the number of studies are few.