Abstract P085: Extremes of Sleep Duration and Type II Diabetes in Older Women Participating in the Women’s Health Study

Abstract

Background: Extreme sleep duration defined as either short (<6 hours) or long (>9 hours) sleep is positively associated with cardiovascular disease (CVD). Because sleep disturbance is prevalent in older women, we evaluated the relationship between extremes of sleep duration and type II diabetes (diabetes) in older women. Methods and Results: In a cross-sectional analysis, we examined the association between self-reported short and long sleep duration and diabetes among 21, 769 older women (mean age = 71. 2 ± 6.0 years old) participating in the ongoing follow-up cohort of the Women’s Health Study. Compared to women with diabetes (N=2539) who slept an average of 6-9 hours, diabetic women who reported short sleep were younger, more likely to be current smokers, less likely to use 1+ drinks/day, were less physically active, had lower annual household income and were less likely to possess a bachelor’s degree. Diabetic women who reported long sleep were older, less likely to be a current smoker, more likely to drink 1+ drinks/day, more likely to have hypertension, depression and anxiety symptoms, were less likely to have a bachelor’s degree, and had a lower annual household income. Logistic regression analyses adjusted for age and race/ethnicity revealed that both short [OR, 95% CI: 1.86 (1.48-2.34)] and long [OR, 95% CI: 1.49 (1.34-1.66)] sleep were significantly associated with a higher odds of diabetes (reference = 6 to 9 hrs sleep). These significant relationships persisted but were somewhat attenuated after control for CVD risk factors and psychological status (i.e., depression and anxiety symptoms) [Short: OR, 95% CI: 1.38 (1.07-1.70); Long: 1.28 (1.14-1.45)]. Conclusion: Extremes of sleep duration were significantly associated with higher odds of diabetes in older women participating in the Women’s Health Study. These hypothesis-generating data suggest that short and long sleep duration might confer key heightened cardiometabolic risk even in older women.