The objective of the study was to analyze possible adverse effects of peripartum cocaine use on maternal and fetal outcomes. Informed consent was given by 720 (97%) of 740 women who delivered consecutively at a large urban public hospital to test an umbilical cord blood sample for the presence of non‐medically administered drugs of abuse and alcohol and to be interviewed for the study. Samples were tested for the presence of a cocaine metabolite (benzoylecgonine‐BZE) by radioimmunoassay. The presence of other substances of abuse (alcohol, methamphetamine, opiates) resulted in exclusion from the sample of 143 subjects. Thus, in this cohort analysis, drug‐free controls (N = 469) were compared to those positive for cocaine only (N = 108). Peripartum exposure to cocaine only, and no other substances of abuse, was associated with an increased frequency of abruptio placentae (1.9% vs 0% for control, P < 0.004), thick meconium stained amniotic fluid (3.9% vs 0.7% for controls, P < 0.006), premature rupture of membranes (P < 0.02), genitourinary anomalies (OR = 3.6, P < 0.05), abdominal wall defects (OR = 4.4, P < 0.01) and increased frequency of low birth weight (OR = 2.0, P < 0.02). These are important findings because previous studies have been complicated by the confounding effects of other substances of abuse. Am. J. Hum. Biol. 11:598–602, 1999. © 1999 Wiley‐Liss, Inc.