Increasing Sleep Time Raises T2D Risk in Older Women.
ENDOCRINOLOGY |
Consistently adequate sleep duration over time appears best.
Consistently getting too little sleep each night or increasing nightly sleep times over a period of several years were both associated with modest, long-term increases in type 2 diabetes risk in an analysis of women enrolled in the Nurse’s Health Study.
Changes in diet, physical activity, and body mass index did not explain the finding of a small, but significant association with type 2 diabetes risk in middle-aged and older women whose sleep duration increased by more than 2 hours over the 14-year analysis.
Regularly sleeping 6 hours or less a night over the study period was associated with a nonsignificant (hazard ratio of 1.05, 95% CI 0.96-1.16) increase in diabetes risk after adjusting for body mass index (BMI), according to Elizabeth Cespedes and colleagues from the Harvard T.H. Chan School of Public Health in Boston, reporting online in the journal Diabetologia.
In an email exchange with MedPage Today, Cespedes noted that a new finding from the study is that the adverse influence of sleeping too little does not appear to be ameliorated by increasing sleep duration later in life. Compared to women who reported consistently sleeping 7 to 8 hours, those who once slept 6 hours or less daily but later increased their sleep duration to 7 to 8 hours had an increased risk for diabetes. They also experienced more weight gain than consistently short sleepers.
“Our results support the message that long-term maintenance of healthy sleep duration is a pillar of health and chronic disease prevention,” Cespedes noted. “However, simply increasing sleep duration after previous years of short sleep may not be a panacea.”
Impact of Sleep Changes Over Time Unknown
The study represents the largest examination ever reported of long-term changes in self-reported sleep duration on diabetes risk and energy balance.
“Typically (in past studies), sleep was only measured once, so existing studies cannot tell us about long-term sleep habits or changes over time. Yet, from a public health perspective this may be what matters most: what happens when people change their sleep duration? What is the influence of chronic short or long sleep on metabolic health?” Cespedes noted.
The analysis included data on close to 60,000 Nurses Health Study (NHS) enrollees without a diabetes diagnosis in the year 2000. Changes in sleep duration were identified by comparing self-reported 24-hour total sleep time in 1986 and 2000.
Diet, physical activity and covariates were updated every 2 to 4 years, and self-reported diabetes was confirmed using validated questionnaires.
About half of women (49%) reported sleeping between 7 and 8 hours a night, with little change in sleep duration between 1986 and 2000, while 15% of women consistently slept 6 hours or less a night and 2% consistently slept 9 hours or more a night.
Few women reported dramatic decreases (5%) or increases (7%) in sleep duration of 2 or more hours a day. Compared to other women, these women tended to have higher body masses and report lower levels of physical activity and diet quality. They were also more likely to snore frequently, smoke, have hypertension, hypercholesterolemia or a family history of diabetes, take antidepressants, or have a ≥ 5 year history of shift work.
Cox regression models were adjusted for reported 1986 sleep duration and 1986 values of diabetes risk factors, including BMI and subsequently for changes in covariates from 1986 to 2000.
Consistent Sleep Duration May Be Key
The researchers documented 3,513 incident diabetes cases occurring by 2012 among the study cohort.
“Compared with no change, decreases in sleep duration were adversely associated with changes in diet quality and physical activity, while increases were associated with greater weight gain,” the researchers wrote.
They reported the following hazard ratios, adjusted for covariates identified at the study’s outset, for various changes in sleep duration seen during follow-up:
- ≤-2 hours/day: 1.09 (95% CI 0.93-1.28)
- >-2 to <0 hours/day: 1.10 (95% CI 1.001-1.12)
- >0 to <2 hours/day: 1.09 (95% CI 1.00-1.18)
- ≥2 hours/day: 1.30 (95% CI 1.14-1.46)
Additional adjustment for diet and exercise did not meaningfully alter the results, the researchers indicated.
Cespedes and colleagues cited the observational design of the study as a limitation, along with the fact that sleep duration in NHS was self-reported. Previous epidemiologic studies have linked very short and very long sleep duration to lower diet quality, lower physical activity, and higher rates of obesity and diabetes.
They noted that several ongoing interventional studies should provide further insight into the role of sleep duration in diabetes risk.
