AEGiS-15IAC: HIV incidence among injecting drug users in New York City, 1990 to 2002: Use of STARHS to assess expansion of HIV prevention services.

15th International AIDS Conference

Bangkok, Thailand - July 11-16, 2004

HIV incidence among injecting drug users in New York City, 1990 to 2002: Use of STARHS to assess expansion of HIV prevention services.

Int Conf AIDS 2004 Jul 11-16; 15:(abstract no. ThOrC1423)

Des Jarlais DC, Perlis TE, Torian LV, Arasteh K, Beatrice S, Milliken J, Mildvan D, Yancovitz S, Friedman S
Beth Israel Medical Center, New York, United States

OBJECTIVE: Evaluation of community-level HIV prevention is typically very difficult and very expensive. The Serologic Test Algorithm for Recent HIV Seroconversion (STARHS, "detuned assay") permits estimating HIV incidence from cross-sectional seroprevalence surveys. STARHS was used to measure HIV incidence among injecting drug users (IDU) in New York City from 1990 to 2002. This period saw rapid growth in HIV prevention programming for IDUs in New York City, with the expansion of syringe exchange from 250,000 to 3,000,000 syringes exchanged annually.

METHODS: Serial cross-sectional surveys were used to obtain serum samples from 3651 IDUs entering Beth Israel Medical Center drug abuse detoxification program from 1990 to 2002. Samples were screened for antibody to HIV by replicate EIA with confirmatory Western Blot. After removal of all personal identifiers, confirmed HIV positive samples were tested for incident HIV infection using the STARHS methodology.

RESULTS: HIV incidence was 3.61/100 person-years at risk (PYAR) in 1990-1992, then fell to 2.64/100 PY in 1993-95, to 1.05/100 PYAR in 1996-98 and to 0.77/100 PYAR in 1999-2002 (p<.001 by Cochran-Armitage test). There was a strong linear relationship between the increase in syringes exchanged and the decrease in HIV incidence, r = -.99, p<.001. Cordray's "rule in" logic and Bayesian analysis were used to assess the likely causal contribution of expanded exchange to reduced HIV incidence.

CONCLUSIONS: The legalization and expansion of syringe exchange in New York City was associated with a very substantial (80%) reduction in HIV incidence among injecting drug users in New York City. STARHS testing of samples from large serial cross-sectional surveys can be very useful in the evaluation of community-level HIV prevention programs.

Keywords: AEGIS, New York City, Incidence, HIV Seropositivity, Acquired Immunodeficiency Syndrome, HIV Infections, Substance-Related Disorders, Syringes, Cross-Sectional Studies, Seroepidemiologic Studies, Israel, epidemiology, prevention & control, utilization

040711
ThOrC1423