Objective. To investigate the duration of effects and health consequences of earlier antenatal corticosteroid exposure in infants born late preterm or term. Design. Observational cohort study. Setting. Children born after gestational week 34 in Sweden, 1976–1997, whose mothers were hospitalized for imminent preterm delivery. The children were followed to their 11th birthday. Sample. The cohort consisted of 11 873 infants, of whom 8620 were exposed. Methods. Exposure was estimated at hospital level. Infants born at a hospital practicing antenatal corticosteroid administration were classified as exposed. Estimation of hospital routines was based on questionnaire data, telephone interviews with physicians and pharmacy sales, validated in a random sample of medical records. Logistic regression was used to assess associations with adjustments for pregnancy length, birth year and hospital level. Main outcome measures. Rates and odds ratios of mortality, respiratory distress syndrome, bronchopulmonary dysplasia, epilepsy, cerebral palsy, childhood diabetes, birthweight, length and head circumference for all infants, and for preterm and term infants, respectively. Results. Exposed infants had reduced risks of respiratory distress syndrome (odds ratio 0.54, 95% confidence interval 0.35–0.83) and small head circumference (odds ratio 0.47, 95% confidence interval 0.36–0.61), and an increased risk of low Apgar scores (odds ratio 1.40, 95% confidence interval 1.01–1.94), most pronounced in infants born after gestational week 37. Conclusions. Infants born after gestational week 34 seem to benefit from earlier antenatal corticosteroid administration, with reduced risks of respiratory distress syndrome. However, the treatment was less beneficial for term infants, because they also had increased risk of low Apgar scores.