BACKGROUND: The economics of universal antenatal human immunodeficiency virus (HIV) screening should be explored if mother-to-child transmission of HIV occurs, the health-service infrastructure for universal screening exists, and optimal risk-reducing treatments can be supplied. METHODS: We evaluated a hypothetical cohort of the antenatal population of Australia during 2001-2002, to examine whether universal antenatal HIV screening is cost-effective in this setting. A quasi-societal perspective was adopted, secondary data sources were used, and sensitivity analyses were undertaken. Costs and benefits incurred in the future were discounted to their present value. RESULTS: The intervention would be cost-effective if the prevalence of undiagnosed HIV in the currently unscreened Australian antenatal population was >or=0.004372%. We predict 6.95 new diagnoses of HIV, 1.73 infections avoided, and 46.97 discounted-life-years gained. Applying favorable and unfavorable values for key variables suggests that the prevalence at which the intervention would be cost-effective is 0.0016%-0.0106%. CONCLUSIONS: Universal antenatal HIV screening would be cost-effective at a very low prevalence and would generate net benefits under many scenarios; accurate statistics on the true prevalence of HIV in the currently unscreened antenatal population are required.