ObjectiveTo report the experience with state-mandated voluntary newborn human immunodeficiency virus (HIV) screening in a large university hospital.MethodsRetrospective analysis of delivery records and newborn HIV screening data for infants born between May 4, 1996, and January 31, 1997.ResultsOf 3111 women, 40% had HIV testing during pregnancy, 28% had testing before pregnancy, 30% never had testing, and 2% gave no information. Most (91%) consented to postpartum newborn HIV testing, although rates varied by HIV counselor (range 81–96%,P< .001) and maternal HIV testing history (range 72–94%,P< .001). Maternal antepartum HIV testing rates varied by delivering physician (range 8–100%,P< .001) and by antepartum care site (clinics, 83%; faculty practices, 72%; private practices, 57%;P< .001). Fourteen HIV-exposed infants (0.46% of infants tested) were identified, 13 of whom were born to women known to be HIV positive before delivery. These women had received zidovudine prophylaxis according to Adult AIDS Clinical Trials Group 076 guidelines. All 14 infants subsequently tested negative for HIV. Human immunodeficiency virus–positive women were more likely than HIV-negative women to have received antepartum care in clinics (93% compared with 11%), have care paid by Medicaid (93% compared with 28%), and reside in high-risk areas (72% compared with 11%) (P< .001).ConclusionNewborn HIV screening has little incremental impact in settings with aggressive antepartum screening and/or low HIV infection rates. Interventions to increase antepartum HIV screening in the private sector should be implemented.