BACKGROUND:

Estimates of the incidence of HIV infection among persons testing for HIV can be derived by applying a newly available serologic test to the diagnostic specimen of HIV-positive persons. Such estimates would enhance the targeting of HIV prevention resources and provide a sensitive outcome measure for prevention program evaluation. The goal of this investigation was to estimate the incidence of HIV infection among persons testing for HIV in New York City.

METHODS:

The study population consisted of persons testing for HIV in public settings in New York City during 2001 (n = 114,703). We applied a less sensitive enzyme immunoassay (LS-EIA) (Vironostika, BioMerieux, Durham, NC) to the diagnostic blood specimen of 1022 persons in whom HIV (non-AIDS) had been diagnosed for the first time in 2001. The distribution of transmission risk among HIV-negative persons--men who have sex with men (MSM), injection drug users (IDUs), heterosexuals-from a large telephone health survey was used to generate denominators for transmission risk groups.

RESULTS:

The 1022 persons tested by the LS-EIA represented 27% of all persons in whom HIV (non-AIDS) had been diagnosed in New York City during 2001. The incidence of HIV was estimated to be 0.29% per year (95% CI: 0.20-0.38), and was significantly higher for men than women (rate ratio 3.6, 95% CI: 2.6-5.1), and HIV incidence increased with age. Male IDU and MSM testers had the highest HIV incidence rates: 2.7% per year (95% CI: 2.3-3.1) and 2.5% per year (95% CI: 2.1-2.8), respectively.

CONCLUSIONS:

Male IDUs and MSM may be good candidates for intensified targeting of HIV prevention resources in New York City.