Does lack of sleep lead to diabetes?

One hundred years ago, adults in the United States averaged 9 hours of sleep per night. Today, that average is less than 7 hours. Although researchers have shown that lack of sleep can impair mental function, they have yet to demonstrate any physical consequences of sleep deprivation.

A study from the University of Chicago now suggests the body's reaction to sleep loss resembles insulin resistance, a condition in which cells fail to efficiently use this sugar-processing hormone. Insulin resistance results in high blood-glucose concentrations and can lead to type II, or adult-onset, diabetes.

The researchers recruited 13 people chronically short of sleep, averaging less than 6.5 hours per night, and 14 other people who typically snoozed more than 7.5 hours nightly. The groups had matching ethnicity and medical histories.

For 8 days, participants wore a wrist device that monitored nighttime movement, which diminishes when people are asleep. They also kept a sleep diary. On the last night, each participant set aside a saliva sample, slept through the night, and then skipped breakfast.

Before their next meal, the participants received an intravenous dose of glucose to test how each one processed the sugar. People in the group that slept less during the preceding week needed to produce 50 percent more insulin to metabolize the glucose, said Bryce A. Mander of the University of Chicago. He presented these findings at an American Diabetes Association meeting late last month in Philadelphia.

Moreover, insulin sensitivity--the measure of how well cells take up the hormone and use it to process sugars--was only about 60 percent as efficient in those getting less sleep as in the more-rested group, said Mander. Even though these people averaged 28 years of age, Mander said, "the short-sleepers had an insulin profile of a 61-year-old."

Further tests showed that the saliva of short-sleepers contained excess cortisol, a. stress hormone, he noted.

"This is a nice study," says Peter Nilsson of Lund University in Malmo, Sweden. Still, the apparent link between sleep and diabetes will require more examination, he says.

For example, Nilsson says, some people in the study might have slept less because of factors that increase their blood-cortisol concentrations, such as job stress. "It's just difficult to disentangle these factors" from sleep in assessing the effects of each on diabetes risk, Nilsson says.