HIV prevalence for New York State using Newborn Seroprevalence and AIDS Surveillance Data

Abstract

The objective of this work is to estimate human immunodeficiency virus (HIV) prevalence in the adult populations of New York City, and of New York State (NYS) excluding New York City. Evidence is presented that not all HIV-infected persons live long enough to satisfy an acquired immunodeficiency syndrome (AIDS) diagnosis. The term preAIDS mortality refers to the mortality of HIV-infected persons, due to any cause, before they can be diagnosed with AIDS. The effect of preAIDS mortality on the estimate of HIV incidence, using back calculation of AIDS Surveillance Data, is to produce low-biased estimates of HIV incidence and prevalence. Estimates of HIV prevalence in women of childbearing age were obtained from the Newborn Seroprevalence Survey [Lessner L. The estimation of HIV prevalence for women of childbearing age in New York City, presently under review]. HIV-positive women in this population typically do not have an opportunistic infection. Thus estimates of HIV prevalence in this population occurred earlier in the spectrum of HIV disease then when they finally satisfied the definition of AIDS that depends on incidence of an opportunistic disease. These earlier estimates of HIV prevalence in women were found to be substantially larger then those obtained from back calculation using female AIDS Surveillance Data. This comparison was used to obtain a quantitative measure R of the effect of preAIDS mortality on HIV prevalence: where 1/ R equals the proportion of HIV-infected intravenous drug users that become AIDS cases The preAIDS mortality adjustment factor R was used in the estimation of HIV prevalence and incidence for NYS. We believe that the use of the adjustment factor resulted in more accurate estimates. Using this factor, our estimated HIV incidence for New York City in 1991 was nearly 26% larger than without the adjustment while projected HIV prevalence for 1996 was nearly 27% larger. The paper estimates HIV prevalence. The AIDS Surveillance operation uses AIDS cases to back calculate HIV incidence and prevalence. Since many HIV-infected intravenous drug using (IDU) population die before they become an AIDS case, the estimates from the back calculation are low biased. Low biased because there were many HIV-infected people who died before they ever satisfied the AIDS protocol. The totals presented here adjust for preAIDS mortality and are necessarily larger than what was recorded by the NYS Dept. of Health. After the change in the definition of AIDS to simply HIV prevalence and the change to electronic reporting, the totals were mostly HIV cases reported in the given year and not AIDS. Thus there is no comparison. This paper represents the last best look at the natural history of AIDS in NYS.