By Randy Dotinga
The study found that women who added more than two hours of shuteye a night showed a 15 percent higher risk of developing type 2 diabetes.
The researchers also suggested that women who regularly slept six hours or less a night might have higher odds of developing type 2 diabetes. But after adjusting the data for other factors such as obesity, this link was not considered statistically significant, the researchers said.
Women who were chronically short on sleep who then tried to catch up were the ones who fared the worst in the study. In fact, the researchers found that short sleepers who added two hours of sleep a night actually increased their odds of diabetes by 21 percent.
"Increasing sleep duration after previous years of short sleep may not be a panacea," said study author Elizabeth Cespedes, a research postdoctoral fellow at Kaiser Permanente Division of Research in California.
It's important to note, however, that this research doesn't confirm a cause-and-effect connection between type 2 diabetes and the amount of sleep women get a night. Although the study found an association between these factors, it isn't clear whether changes in sleep patterns contribute to diabetes or vice versa, the researchers said.
Cespedes said previous research has shown that people who sleep too little -- or too much -- and eat poorly and exercise less are more likely to develop obesity and type 2 diabetes. But there's been little research into the role of long-term changes in sleep patterns. What happens when people begin to sleep more or less over time, the researchers wondered.
To see if they could answer that question, the researchers tracked almost 60,000 American women. The women were nurses between the ages of 55 and 83. The researchers looked for changes in sleep patterns from 1986 to 2000. Then they looked for any connections between sleep changes and cases of type 2 diabetes diagnosed between 2000 and 2012. Just over 3,500 women were diagnosed with diabetes in that time period.
After the researchers adjusted their statistics to account for changes in factors such as obesity, they found the only statistically significant relationship was in those who added 2 or more hours of sleep each night. Women whose sleep time grew by 2 or more hours had 15 percent greater odds of developing type 2 diabetes, the study concluded.
It's possible that simply having diabetes disrupts sleep, though Cespedes said that's probably not a major factor for this group of women because there wasn't a lot of undiagnosed diabetes in this group.
"Some scientists argue that long sleep is a symptom of underlying sleep disorders, depression or ill health," Cespedes said, "and that it is these factors, and not long sleep, that increase the risk of diabetes." But, the researchers tried to account for those factors and still saw "a relationship between large increases in sleep duration and increased risk of diabetes," she said.
For now, it's not clear whether changing sleep patterns -- sleeping more or less -- could prevent diabetes, she said. However, she added, several studies in children and adults are seeking to answer this question.
What about men? Some research has suggested that extremes in sleep affect men and women differently, Cespedes said. But research similar to this study included men and found similar results, she said.
Jane Ferrie, a senior research fellow at the University of Bristol in England, who's worked on similar research, praised the study and said it's "the best evidence we have to date" on this topic. She speculated that short sleep may disrupt the way the body processes blood sugar. It's also possible that people who sleep longer may have undiagnosed sleep apnea. (The study only included sleep apnea diagnosed by a doctor.)
For now, Ferrie said, "women whose sleep duration changes by two or more hours per night should mention this to their doctor."
The study appeared Nov. 2 in the journal Diabetologia.