BACKGROUND: Exercise in pregnancy is recommended in many countries, and swimming is considered by many to be an ideal activity for pregnant women. Disinfection by-products in swimming pool water may, however, be associated with adverse effects on various reproductive outcomes. We examined the association between swimming in pregnancy and preterm and postterm birth, fetal growth measures, small-for-gestational-age, and congenital malformations. METHODS: We used self-reported exercise data (swimming, bicycling, or no exercise) that were prospectively collected twice during pregnancy for 74,486 singleton pregnancies. Recruitment to The Danish National Birth Cohort took place 1996-2002. Using Cox, linear and logistic regression analyses, depending on the outcome, we compared swimmers with physically inactive pregnant women; to separate a possible swimming effect from an effect of exercise, bicyclists were included as an additional comparison group. RESULTS: Risk estimates were similar for swimmers and bicyclists, including those who swam throughout pregnancy and those who swam more than 1.5 hours per week. Compared with nonexercisers, women who swam in early/mid-pregnancy had a slightly reduced risk of giving birth preterm (hazard ratio = 0.80 [95% confidence interval = 0.72-0.88]) or giving birth to a child with congenital malformations (odds ratio = 0.89 [0.80-0.98]). CONCLUSIONS: These data do not indicate that swimming in pool water is associated with adverse reproductive outcomes.