HIV incidence among injection drug users in New York City, 1992-1997: Evidence for a declining epidemic

Objectives. We assessed recent (1992-1997) HIV incidence in the large HIV e pidemic among injection drug users in New York City. Methods. Data were compiled from 10 separate studies (N=4979), including 6 cohort studies, 2 "repeat service user" studies, and 2 analyses of voluntar y HIV testing and counseling services within drug treatment programs. Results. In the 10 studies, 52 seroconversions were found in 6344 person-ye ars at risk. The observed incidence rates among the 10 studies were all wit hin a narrow range, from 0 per 100 person-years at risk to 2.95 per 100 per son-years at risk. In 9 of the 10 studies, the observed incidence rate was less than 2 per 100 person-years at risk. The weighted average incidence ra te was 0.7 per 100 person-years at risk. Conclusions. The recent incidence rate in New York City is quite low for a high-seroprevalence population of injection drug users. The very large HIV epidemic among injection drug users in New York City appears to have entere d a "declining phase," characterized by low incidence and declining prevale nce. The data suggest that very large high-seroprevalence HIV epidemics may be "reversed.".