Background:  Because of the long period between HIV seroconversion and clinical disease, most persons with HIV are diagnosed ≥1 years into the chronic phase of infection. Therefore, surveillance of HIV diagnoses provides limited insight into current incidence and patterns of transmission. Our objective was to estimate HIV incidence in New York City using a new Centers for Disease Control and Prevention (CDC) laboratory algorithm and statistical method.

Methods:  Remnant serum from diagnostic Western blots was salvaged and retested by a laboratory method that distinguishes recent from long-standing HIV infection. Results were combined with local surveillance data, missing data were imputed, and a stratified extrapolation approach was used to estimate incidence. The estimate includes persons who are newly infected but not diagnosed because they have never been tested.

Results:  In 2006, an estimated 4762 (95%CI 4208 to 5315) persons were newly infected with HIV in New York City. The overall incidence rate was 72/100,000 population. Three-quarters of new infections occurred in men (117/100,000) and 25% in women (33/100,000). Persons aged ≥30 years made up 72%. Blacks accounted for 46% (142/100,000), Hispanics for 32% (91/100,000), and whites for 21% (41/100,000). Among blacks, incidence was 231/100,000 among men and 78/100,000 among women. Men who have sex with men (MSM) accounted for half of new infections, among which 65% were in MSM aged ≥30. The 2 age groups of MSM presented different racial profiles, with blacks and Hispanics predominating among MSM <30 and whites predominating among MSM ≥30.  

Conclusions:  Overall incidence in New York City was 3 times the national rate of 23/100,000. New York City rates for various subpopulations were similarly elevated over national rates. The new estimates indicate that black New Yorkers, persons aged 30 to 50 years, and men, especially MSM, were at greatest risk for incident HIV in 2006. This new CDC method provides a baseline incidence estimate and will be used to monitor future epidemic trends as well as to evaluate expanded testing and prevention initiatives.