The relationship between physical activity during pregnancy, preterm birth, and gestational age-adjusted birth weight was investigated prospectively in a cohort of 7101 women. This study is one of few to evaluate both employment- and non-employment-related physical activity. Prolonged periods of standing were associated with a modestly increased risk of preterm delivery (adjusted odds ratio for greater than or equal to 8 hours/day of standing = 1.31). Heavy work or exercise was not associated with preterm delivery (adjusted odds ratio for greater than or equal to 4 hours per day of heavy work = 1.04). The proportion of infants born preterm did not differ among women working in predominantly standing, active, and sedentary occupations. Physical activity was not associated with gestational age-adjusted birth weight after controlling for confounding variables. These data suggest that unmeasured socioeconomic differences among women reporting different levels of activity may account for previously described associations between physical activity and pregnancy outcome. Most pregnant women who report increased levels of physical activity are not at increased risk of preterm delivery or reduced intrauterine growth. However, these data do not address the role of activity restriction in the management of selected women at high risk for adverse pregnancy outcome.