This study was designed to overcome some of the methodological limitations of previous work and investigate the impact of prenatal cocaine use in an understudied population: women using rural county public health units who had minimal access to drug rehabilitation. Through maternal history, interviews, and urine screens, 172 cocaine users were identified. Using an independently collected perinatal data base, 168 nonusers were matched for six variables known to affect pregnancy outcome and chosen a priori: race, age, parity, prenatal care, alcohol, and nicotine use. To avoid chance findings, 10 adverse perinatal outcome variables were identified prospectively. Cocaine-exposed neonates experienced significantly more of the adverse events than the matched controls and were more likely to be preterm, low birthweight, resuscitated at birth, and to remain in the hospital after their mothers were discharged. We conclude that prenatal cocaine use can be a contributor to adverse perinatal outcome in this population. An understanding of the effects of prenatal cocaine use and the needs of these women and infants is important for designing appropriate prenatal care, treatment, and follow-up programs.