New HIV Infections in U.S. 40% Higher Than Previously Estimated
Approximately 56,300 Americans became newly infected with HIV, the virus that causes AIDS, in 2006 according to the findings of the first study to directly measure the incidence of new HIV infections in the United States. That estimate, which is based on new, more precise surveillance methods, indicates that the annual rate of new infections is higher than previously believed; it represents a 40% increase over the Centers for Disease Control and Prevention's (CDC) former estimate of 40,000 new infections per year.
The study "Estimation of HIV incidence in the United States," conducted by a CDC research team including Edward H. Kaplan, the William N. and Marie A. Beach Professor of Management Sciences at the Yale School of Management, professor of public health at the Yale School of Medicine, and professor of engineering at the Yale School of Engineering and Applied Science, is published in the August 6 issue of JAMA.
"It is important not to mistake the message of this study," said Kaplan. "We have not discovered that the rate of new HIV infections increased over time. Rather, with new laboratory and statistical methods, we can now state that the annual rate of new HIV infections in this country has been higher than believed all along."
The study's analysis of historical trends finds that the incidence of new HIV infections per year increased in the mid-1990s, then slightly declined after 1999, and has remained relatively stable since that time, with estimates ranging between 55,000 and 58,500 new incidents annually.
Men who have sex with men and African-Americans accounted for the greatest proportion of estimated new HIV infections in 2006, at 53% and 45%, respectively.
The new estimate was calculated using a new laboratory test that uses a blood sample from a person who has received a positive diagnosis for HIV to detect a biomarker, known as the BED assay, which distinguishes new infections-those that occurred within approximately the last five months-from long-standing infections. Using new statistical models, the number of new infections identified in a given year was used to estimate the number of new infections in the national population.
"The idea of estimating the rate of new HIV infections from biomarker data has more than a ten year history, but several obstacles had to be overcome to enable our study to proceed," said Kaplan. "For example, a better biomarker than those available ten years ago was designed specifically for surveillance use. As another example, the original statistical methods were designed for use with researcher-selected samples from the population-such samples could be made representative using standard sampling methods. New models were required for application to the CDC's national HIV surveillance system, where individuals self-select when and how frequently to get tested. So, quite a bit of basic research was required before we could even begin to produce the estimated new rates of infection."
A paper on the new methodology, "Estimating HIV incidence in the United States from HIV/AIDS surveillance data and biomarker HIV test results," also co-authored by Kaplan, is published in the August 2008 issue of Statistics in Medicine.
Using the new method, the CDC will be able to provide an annual updated estimate of HIV incidence in the U.S., which will lead to improved targeting and evaluation of prevention efforts.
"Monitoring HIV incidence is perhaps the first step in targeting effective prevention programs and allocating resources efficiently. We cannot prevent more infections than those that would have occurred in the absence of any incremental prevention effort," said Kaplan.
Media coverage of this study appeared in the New York Times, Wall Street Journal, and Associated Press.