Abstract Introduction Women commonly experience sleep disruption during the perinatal period, which has been associated with adverse outcomes for the mother and infant. Exercise has been associated with improved sleep quality in the general population. However, the impact of exercise on sleep in the perinatal period remains unclear. The purpose of this analysis was to examine the relationship between subjective sleep duration and quality with self-reported exercise during the perinatal period. Methods Forty-four pregnant women (31±5 years) were enrolled in a longitudinal observational study. Visits took place at 16-20 weeks gestation, 28-31 weeks gestation, and 3 months postpartum. At each timepoint participants completed the Pittsburgh Sleep Quality Index (PSQI) and reported weekly exercise frequency. To examine if there were any differences between those who exercised and those who didn’t at each timepoint, independent t-tests were performed. One-way ANOVAs were performed to examine the association of exercise frequency with sleep duration and PSQI global scores at each timepoint (α=0.05). Data are presented as mean±SD. Results Participants self-reported sleeping 7.4±1.4, 7.2±1.2, and 6.6±1.1 hours per night at each timepoint. PSQI scores were 7±4, 7±4, and 6±3, respectively. The number of participants reporting exercise at each timepoint were 29, 28, and 18, respectively. There were no differences in sleep duration or PSQI scores between those who exercised and those who did not (p0.05). Furthermore, there were no significant associations between weekly exercise frequency and sleep duration or PSQI (p0.05). Conclusion In this sample, participants experienced poor sleep quality throughout the perinatal period. Additionally, the number of participants exercising decreased from pregnancy to postpartum. However, exercising was not associated with subjective sleep duration or quality during the perinatal period. Future studies identifying strategies for improving sleep duration and quality during the perinatal period are needed. Support (if any) NICHD R15HD102904 (Okun)
KISHMAN et al. (2026) studied this question.