Objective To evaluate the cross-sectional association of physical activity during pregnancy with gestational weight gain and verify if associations are modified by levels of sleep duration and efficiency. Methods: Mothers who were participants of the 2015 Pelotas (Brazil) Birth Cohort study were analyzed. Physical activity (PA) and sleep were measured using the nondominant wrist accelerometer model wGT3X-BT (ActiGraph, Pensacola, FL). Moderate-to-vigorous activities were measured in bouts of 5 minutes (MVPA 5-min bout). Activities were characterized as sustained and overall physical activity was expressed in mg, and MVPA (non-bouted) activities were characterized as not sustained. Sleep was presented as minute bouts, sleep time window, total sleep time, and sleep efficiency. Gestational weight gain (GWG) was calculated by a woman's weight at the end of gestation minus the pre-pregnancy weight. Linear regression models were used to evaluate the association between PA and GWG. Stratification by sleep characteristics in tertiles was performed. Results: The analytical sample included 1,938 women with information for PA, sleep, and GWG. The mean GWG was 12,030 g (SD = 6658.8). After adjustments, for every 15 minutes of sustained physical activity (MVPA 5-min bout), the GWG decreased by 591.3 g [β= -591.3 (95%CI: -1,147.1; -35.5). A statistically significant association was observed between 15 minutes of sustained physical activity (MVPA 5-min bout) and GWG in the second and third tertiles of healthier sleep efficiency. Conclusion: MVPA with 5-min bouts is associated with lower GWG. The GWG reduction was greater in the higher sleep efficiency strata.
Keywords: Accelerometry; cohorts studies; gestational weight gain; physical activity; sleep.