Background Previous work demonstrated that physical activity during pregnancy mediates the relationship between maternal insulin action and infant body composition. It could, therefore, be an important target for intervention to improve metabolic outcomes for pregnant women and reduce the risk of future obesity and diabetes in their offspring. This study examined the relationship between physical activity during pregnancy and offspring health.

Methods Participants were 183 women recruited to the pilot phase of the UK Pregnancy Better Eating and Activity Trial (UPBEAT). Physical activity was assessed objectively by accelerometer at recruitment (16–18 weeks’ gestation), at 27–28 weeks’ gestation and at 35–36 weeks’ gestation. The relationships between sedentary behaviour and moderate and vigorous physical activity (MVPA) and neonatal health were examined using Spearman partial correlations, adjusted for maternal age and body mass index.

Results Valid accelerometry data were available for 140 women at baseline, 75 women at 27–28 weeks’ gestation and 54 women at 35–36 weeks’ gestation. Maternal sedentary time at 35–36 weeks’ gestation was positively associated with infant abdominal circumference (Spearman correlation coefficient 0.435, p = 0.023) and head circumference (0.408, p = 0.035) and inversely associated with abdominal circumference at baseline (-0.333, p = 0.016). Sedentary time was not associated with these outcomes at 27–28 weeks’ gestation. MVPA at 35–36 weeks’ gestation was inversely associated with infant abdominal circumference (-0.466, p = 0.014) and head circumference (-0.443, p = 0.021). No associations between physical activity and birthweight were apparent.

Conclusion This analysis supports previous work suggesting that physical activity during pregnancy is associated with indicators of offspring health and that maternal physical activity is an appropriate target for intervention to improve infant outcomes.