Physical activity and sleep among midlife women with vasomotor symptoms

Abstract

Objective: Few studies have examined the relationship between physical activity and sleep among women with vasomotor symptoms. We examined the associations of habitual leisure time and household physical activity with sleep characteristics among women reporting vasomotor symptoms.

Methods: A subcohort of women (N = 52: white, 27; African American, 25) in the Study of Women's Health Across the Nation was assessed for self-reported and actigraphic sleep measurements for four nights. Women were between 54 and 63 years, were currently experiencing vasomotor symptoms, and were not taking selective serotonin reuptake inhibitors or hormone therapy. The associations between physical activity (leisure time and household) and sleep (diary, actigraphy, and global sleep quality), as well as interactions by race and body mass index, were assessed using logistic regression and generalized estimating equations.

Results: Greater leisure time physical activity was associated with higher odds of rating global sleep quality as good (OR, 8.08, 95% CI, 1.5, 44.5, P = 0.02). Greater household physical activity was associated with more favorable diary-reported sleep characteristics, including fewer awakenings during the night (B = -0.14, SE = 0.06, P = 0.01). Exploratory analyses suggest that household physical activity was associated with more favorable sleep characteristics primarily among white and nonobese women.

Conclusions: Greater levels of habitual physical activity, particularly non-leisure time physical activity, are associated with more favorable sleep characteristics. Considering the potential impact of physical activity on sleep, even at the relatively modest levels characteristic of household physical activity, may be important for women with vasomotor symptoms, a subgroup at high risk for sleep problems.

Figure 1

Unadjusted relationship between household physical activity and diary-reported sleep characteristics (top: number of awakenings; bottom: compared to a usual night) by race. Fitted lines for White (solid) and African American (dashed) women were derived from bivariate correlations between household physical activity and sleep variables. Significance testing (including of interaction terms) conducted with general estimating equations.

Figure 2

Unadjusted relationship between household physical activity and actigraphy-assessed wake after sleep onset (WASO) in non-obese and obese women. Fitted lines for non-obese (solid) and obese (dashed) women were derived from bivariate correlations between household physical activity and WASO. Significance testing (including of interaction term) conducted with general estimating equations.