Abstract: We examined the effect of methadone treatment, duration of Medicaid enrollment during pregnancy, and other maternal characteristics on receipt of prenatal care by 2,254 women infected with human immunodeficiency virus (HIV) delivering a singleton in New York state from 1985 through 1990. Data were obtained from the New York State Medicaid HIV/AIDS Research Data Base and vital statistics records. Adequacy of the number of prenatal visits reported by the mother on vital statistics records was assessed with use of the Kessner Index, which adjusts for gestational age at delivery. Too few visits were reported by 65% of the study population. Illicit drug users had higher odds of having too few visits [1.64, 95% confidence interval (CI) 1.24-2.17] than methadone-treated women but the odds were similar for non-drug users and methadone-treated women (0.79, 95% CI 0.60-1.25). Women with brief Medicaid enrollment (< or = 25% of pregnancy) had 45% higher odds of having too few visits than did longer enrollees. Treatment for drug addiction and longer Medicaid enrollment during pregnancy may offer important interventions to increase prenatal care of HIV-infected women. Approaches to increase prenatal care of HIV-infected women are especially important given trials showing a reduction in vertical transmission from zidovudine treatment during pregnancy.