The researchers recommend getting seven to eight hours of nightly sleep on a regular basis.
Here's a quick look at the studies, published in the latest edition of the journal Sleep.
Sleep and Death Rate
In one of the studies, death rates were higher for people who spent too much or too little time sleeping.
That study included some 10,300 British government workers who were followed for 12-17 years.
At the study's start and midpoint, the workers got checkups and completed a health survey that included a question about their typical hours of nightly sleep.
During the study, the death rate was higher for these groups:
- People who consistently got less than five hours of nightly sleep
- People who consistently got more than nine hours of nightly sleep
- People who started out with six to eight hours of nightly sleep but later got less sleep
- People who started out with seven to eight hours of nightly sleep but later got more sleep
"In terms of prevention, our findings indicate that consistently sleeping 7 or 8 hours per night is optimal for health," write the researchers.
They included Jane Ferrie, PhD, of the University College London Medical School.
Diabetes and Sleep
The second study shows that sleeping too much or too little may make diabetes more likely.
Data came from nearly 9,000 U.S. adults.
When the study began, none of the participants reported having diabetes. Ten years later, 430 of them had diabetes.
The people most likely to develop diabetes were those who reported getting less than five hours or more than nine hours of nightly sleep at the study's start.
Sleeping more than nine hoursor less than five hoursper night raised the odds of getting diabetes by about 50%, compared with sleeping seven hours per night.
Those patterns weren't affected by other diabetes risk factors.
Getting too little sleep may hamper how the body handles blood sugar (glucose), and getting too much sleep may be a sign of other health problems, note the researchers.
They included James Gangwisch, PhD, of New York's Columbia University.
Both studies were observational, so they don't prove cause and effect. The researchers weighed many factors, but they note that there's always a chance that they missed influential factors.