In January 1983, the New York City Department of Health initiated an active surveillance program for acquired immunodeficiency syndrome (AIDS) in 19 hospitals and a modified-active surveillance program in the remaining 69 hospitals. We reviewed hospital laboratory and autopsy records in 12 active surveillance hospitals and three modified-active surveillance hospitals six months later. Patients who had opportunistic diseases characteristic of AIDS diagnosed in 1982 (before active surveillance) and 1983 (after implementation of active surveillance) were matched against health department AIDS surveillance reports. For the 16 months we evaluated, 96% of patients identified with AIDS in the 12 active surveillance hospitals and 100% of those in the three modified-active surveillance hospitals had been reported to the health department. The delay between diagnosing a case and reporting it to the health department significantly decreased between 1981 and the first six months of 1983 in all hospitals. The proportion of cases reported within one month of diagnosis increased from 45% to 69% during this period. We conclude that the current surveillance program for AIDS in New York City is effective and that case reporting is sufficiently complete for accurate analysis of disease trends.