Bedtime Variability and Metabolic Health in Midlife Women: The SWAN Sleep Study

Study objectives: Circadian misalignment, as seen in shift workers, can disrupt metabolic processes. Associations between sleep timing in nonshift workers and metabolic health are unknown. We examined sleep timing and indices of metabolic health in a community sample of midlife women.

Methods: Caucasian (n = 161), African American (n = 121) and Chinese (n = 56) non-shift-working women aged 48-58 y who were not taking insulin-related medications, participated in the Study of Women's Health Across the Nation (SWAN) Sleep Study and were subsequently examined approximately 5.39 (standard deviation = 0.71) y later. Daily diary-reported bedtimes were used to calculate four measures of sleep timing: mean bedtime, bedtime variability, bedtime delay and bedtime advance. Body mass index (BMI) and insulin resistance (homeostatic model assessment-insulin resistance, HOMA-IR) were measured at two time points. Linear regressions evaluated whether sleep timing was associated with BMI and HOMA-IR cross-sectionally and prospectively.

Results: In cross-sectional models, greater variability in bedtime and greater bedtime delay were associated with higher HOMA-IR (β = 0.128; P = 0.007, and β = 0.110; P = 0.013, respectively) and greater bedtime advance was associated with higher BMI (β = 0.095; P = 0.047). Prospectively, greater bedtime delay predicted increased HOMA-IR at Time 2 (β = 0.152; P = 0.003). Results were partially explained by shifted sleep timing on weekends.

Conclusion: Frequent shifts in sleep timing may be related to metabolic health among non-shift working midlife women.

Commentary: A commentary on this article appears in this issue on page 269.