OBJECTIVE To evaluate the potential impact of ACOG and Centers for Disease Control and Prevention (CDC) consensus strategies for the prevention of perinatal group B streptococcal disease. METHODS We evaluated cases of early-onset group B streptococcal disease identified by active surveillance during 1995, in four areas in North America with an aggregate 186,000 births per year. We reviewed the hospital records of mothers and infants and any prenatal records available on site. Cases were determined to be preventable based on whether group B streptococcal screening could have been performed prenatally, sensitivity of screening, presence of obstetric complications, and opportunity to administer antibiotics. RESULTS We reviewed records for 245 of 246 infants with early-onset group B streptococcal disease in the surveillance areas. Most of the 53 case-mothers who delivered preterm and 192 who delivered full-term had had at least one prenatal visit (83% and 99%, respectively). Few case-mothers had prenatal group B streptococcal screening cultures, although compliance was high for other prenatal screening tests. Fifty-four percent of case-mothers had a recognized obstetric risk factor for group B streptococcal disease: labor or rupture of membranes at less than 37 weeks, rupture of membranes for 18 hours on longer, or temperature 38C or greater. The estimated preventable portion of early-onset group B streptococcal cases was 78% for the screening-based approach (range 74% to 82% by area), compared with 41% for the risk-based approach (range 39% to 53% by area). CONCLUSIONS Comprehensive implementation of either of the recommended prevention strategies could potentially prevent a substantial proportion of early-onset group B streptococcal disease.