Back in 2004 I summarized an editorial in the journal Sleep summarized the view, based on numerous studies, that the association between sleep duration and mortality can be represented by a J- or U-shaped curve. The lowest mortality rate was apparently found with sleep duration of about 7 hours a night.

Finnish searchers have now brought confirmation of the not-too-little, not-too-much concept. Results of their study have been published in the journal Sleep. Over 10,000 sets of twins responded to questionnaires sent out in 1975 and 1981; they were followed for a total of 22 years, on average, with death rates assessed between 1982 and 2003.

Short sleep was defined as less than 7 hours, and long sleep as more than 8 hours. There was significantly increased mortality both for short-sleep men and women (+26% and +21%, respectively) and for long-sleep men and women (+24% and +17%, respectively, as well as for users of hypnotics or tranquilizers (+31% and +39%, respectively). This effect was more pronounced in young men.

The effect of sleep on mortality is modest, compared with smoking, obesity, and physical inactivity, say the researchers. Interestingly, they report that snoring didn't influence the results. (Of course, they meant in the subject, not in the sleeping partner!)