The optimal clinical strategy to prevent neonatal group B streptococcal infection has not been established. The Maternal-Fetal Medicine Units Network assessed the feasibility of a randomized clinical trial to compare the effectiveness of two methods of prevention of streptococcal infection. The attack rates of early-onset group B streptococcal disease were estimated for the following: 1) a hypothetical program of universal prenatal screening and selective intrapartum chemoprophylaxis, and 2) a hypothetical program of selective intrapartum chemoprophylaxis based on clinical risk factors. Sample size requirements were calculated for a clinical trial to detect a significant difference in attack rates between the two hypothetical programs. Similar low attack rates using both strategies would require extremely large sample sizes to detect differences between programs. Until a satisfactory, rapid diagnostic test for intrapartum group B streptococcal carrier status is developed and clinical strategies are tested in populations with varying carriage frequencies, institutions will need to individualize their approaches to prevent early-onset neonatal group B streptococcal disease.