To assess teratogenic effects of cocaine exposure and maternal psychological distress on birth outcomes, we conducted a longitudinal prospective study of 415 infants (218 cocaine-exposed--CE, 197 nonexposed--NE). Drug exposure was determined through a combination of maternal self-report, urine, and meconium screens. Maternal psychological distress postpartum was evaluated through a standardized, normative, self-report assessment. An extensive set of confounding variables was controlled, including severity of exposure to alcohol, tobacco, marijuana and other drugs, maternal age, race, parity, number of prenatal care visits, educational, marital, and socioeconomic status, and verbal and nonverbal intelligence. CE infants were smaller on all birth parameters and more likely to be preterm, small for gestational age, and microcephalic than NE infants. Forty-one percent of cocaine users had clinically significant psychological symptoms, compared to 20% of a high-risk comparison group of noncocaine users. Consistent with a teratologic model, cocaine exposure independently predicted offspring birthweight, length, and head circumference. Maternal psychological distress self-reported postnatally also independently predicted head circumference. Tobacco, alcohol, and marijuana exposures were also significant independent predictors of some fetal growth parameters. In addition, maternal distress symptoms, which may be reflective of maternal mental health disorders or responses to stress, added significantly to the risk for poorer fetal growth.