Antenatal corticosteroids decrease the incidence of severe intraventricular hemorrhages (grades 3,4) in preterm infants. It is unclear whether their beneficial effects on intraventricular hemorrhage wane with time (as occurs in neonatal respiratory distress) and if repeat courses can restore this effect. Prior randomized controlled trials of betamethasone retreatment found no benefit on severe intraventricular hemorrhage rates. However, the trials may have included an insufficient number of infants at risk for intraventricular hemorrhage to be able to adequately address this question. Severe intraventricular hemorrhages occur almost exclusively in infants born <28 weeks, whereas only 7% (0%-16%) of the retreatment trials' populations were <28 weeks. To determine if the risk of severe intraventricular hemorrhage in infants delivered <28 weeks increases when the betamethasone treatment-to-delivery interval increases beyond 9 days and to determine if betamethasone retreatment prior to delivery decreases the rate of hemorrhage. Observational study examining the incidence of intraventricular hemorrhage before (epoch 1) and after (epoch 2) a practice change that encouraged obstetricians to retreat pregnant women still at high risk of delivery before 28 weeks' gestation when >9 days elapsed from the first dose of betamethasone. Multivariable analyses with logistic regression using generalized estimating equations techniques were conducted to examine the rates of intraventricular hemorrhage among 410 infants <28 weeks' gestation who either delivered between 1-9 days (n=290) after the first 2-dose betamethasone course or delivered ≥10 days (and eligible for retreatment) (n=120). After adjusting for potential confounding variables, infants who delivered ≥10 days after a single betamethasone course had an increased risk of either severe intraventricular hemorrhage alone or the combined outcome severe intraventricular hemorrhage or death before 4 days (OR (95%CI): 2.8 (1.2, 6.6)) compared with infants who delivered between 1-9 days after betamethasone. Among the 120 infants who delivered ≥10 days after the first dose of betamethasone, 64 (53%) received a second/retreatment course of antenatal betamethasone. The severe intraventricular hemorrhage rate in infants whose mothers received a second/retreatment course of betamethasone was similar to the rate in infants who delivered within 1-9 days and significantly lower than in those who delivered ≥10 days without retreatment (OR (95%CI): 0.10 (0.02, 0.65). Following the change in guidelines, the rate of retreatment in infants who delivered ≥10 days after the first betamethasone course (and before 28 weeks) increased from epoch 1 to epoch 2 (25% to 87%, p<0.001) and the rate of severe intraventricular hemorrhage decreased from 22% to 0% (p<0.001). In contrast, the rate of severe intraventricular hemorrhage in infants who delivered 1-9 days after the initial betamethasone dose (who were not eligible for retreatment) did not change between epochs 1 and 2 (12% and 11%, respectively). Although betamethasone's benefits on severe intraventricular hemorrhage appear to wane after the first dose, retreatment with a second course appears to restore its beneficial effects. Encouraging earlier retreatment of women at high risk of delivery before 28 weeks was associated with a lower rate of severe intraventricular hemorrhages among infants delivering <28 weeks.