Sleep Duration as a Predictor of All-Cause Mortality: Results of the Japan Collaborative Cohort Study

By Dr. Gregg D. Jacobs

This study, published in the February (27:51-54), 2004 issue of Sleep by Drs. A. Tamakoshi and Y. Ohno, represents the third major study in the past two years to assess the relationship between sleep duration and mortality. Two previous studies involving a million people and 85,000 people found that sleeping seven hours per night was associated with the lowest death rates, while sleeping eight or more hours per night was associated with elevated death rates while short sleep was not. The present study by Tamakoshi and Ohno found a similar relationship.

The study involved 104,010 subjects in Japan . All subjects reported their average sleep duration during the week at baseline and then were followed for an average of 10 years. A strength of this study was that the researchers measured perceived mental stress as a potential covariate of mortality and sleep duration. The researchers also adjusted for many other factors known to be associated with mortality such as smoking, alcohol use, physical activity, marital status, and education.

They found that the average reported sleep duration was between 7 and 7.5 hours for men and women. The highest mortality rate was found in people who slept 10+ hours, followed by those who slept less than four hours, then those who slept 9 hours, and finally those who slept 5 hours. When the researchers adjusted of mental stress, the relationship between elevated mortality and sleeping less than 7 hours disappeared in men (meaning that elevated death rate for those who slept 4 or 5 hours per night was due to mental stress and not shortened sleep) and in women sleeping between and 4 and 7 hours.

However, the relationship between increased mortality and sleeping eight or more hours did not disappear when adjusting for mental stress, meaning that higher mortality rates were not due to mental stress or any other factors such as smoking that are associated with mortality. Furthermore, women sleeping less than four hours per night still exhibited increased mortality rates despite adjustments for covariates such as mental stress.

These findings confirm that sleeping longer than 7 hours per night is associated with increased mortality risk but that sleeping less than seven hours per night is not a risk factor for increased mortality in men or in women who sleep more than four hours per night. Sleeping seven hours per night is associated with the lowest mortality risk. The authors suggest that strategies should be developed to address longer sleep from a public health perspective.