Despite reductions in perinatal HIV transmission, cases continue to occur. To determine factors associated with zidovudine (ZDV) receipt among HIV-infected pregnant women we merged three data sets for women in New Jersey in 1995-1997, identifying 395 HIV-infected pregnant women. Half received two arms of ZDV prophylaxis. Attendance at five or more prenatal visits was the strongest independent factor related to ZDV receipt (OR 6.37, 95% CI 3.84, 10.57). Half (49.0%) had limited prenatal care. AIDS diagnosis, race/ethnicity, and drug use were also independently related to ZDV receipt. Post hoc analysis revealed that being unmarried, Black, multiparous, having no insurance, and illegal drug use were associated with limited prenatal care. Although the U.S. has seen reductions in HIV perinatal transmission, our research showed that HIV-infected women who did not get prenatal care were less likely to receive two arms of ZDV prophylaxis. A wide public health net that brings all women into care is necessary to reduce perinatal transmission further.