OBJECTIVE To determine the prevalence of HIV-1 infection in the general community by using a target population (antenatal patients) as an indicator of infection, to monitor any change in the prevalence of HIV infection in this population and provide baseline information on heterosexual spread of HIV infection into this low prevalence population. METHODS Between January 1989 and January 1992, 4537 unlinked anonymous antenatal sera were tested in two study groups at Westmead Hospital. Repeatedly reactive sera were confirmed by western blot and other supplementary assays as appropriate. RESULTS No significant change in the seroprevalence of HIV-1 infection was detected between the two study periods. Of the 2208 sera tested in 1989-1990, one (0.05%) was confirmed as positive and one (0.05%) gave a non-specific reaction by enzyme immunoassay (EIA). Of the 2329 sera tested in 1991-1992, there was one (0.04%) HIV-1 antibody positive serum, and two gave non-specific EIA reactions. These results were compared with three other sample populations tested at Westmead Hospital during the same period: linked antenatal patients, antenatal methadone clinic attendees and all women tested. CONCLUSIONS Periodic anonymous testing of a sample antenatal population (combined with screening of high risk patients) is useful for monitoring the prevalence of HIV infection in these populations and estimating any future need for generalised screening.