US HIV incidence 40% higher than previous estimates, new test shows

This article is more than 16 years old. Click here for more recent articles on this topic

Based on a new testing technique that can distinguish recent from long-standing HIV infections, the U.S. Centers for Disease Control and Prevention (CDC) now estimates that there were 56,300 new HIV infections in the U.S. in 2006—about 40% higher than the previous estimate. The figures were released on August 2, in advance of the XVII International AIDS Conference in Mexico City, and published in the August 6 edition of the Journal of the American Medical Association.

The incidence of HIV in the U.S. had not previously been measured directly. Standard HIV antibody tests show whether a person has the virus, but cannot determine how recently infection occurred. Furthermore, due to confidentiality concerns, several U.S. states have in past years reported only AIDS cases by name, not HIV infections. Based on these limited data, CDC has claimed for several years that the annual number of new infections was 40,000.

CDC investigators used the BED HIV-1 capture enzyme immunoassay, a modified antibody test that can distinguish infections that occurred in the previous five months, to analyze leftover serum specimens from individuals age 13 or older who were newly diagnosed with HIV during 2006 in 22 states with name-based HIV reporting. Together, these states accounted for about three-quarters of all AIDS cases diagnosed that year.



Clear, non-cellular portion of the blood, containing antibodies and other proteins and chemicals.



A protein which speeds up a chemical reaction.


A test used to measure something.

peer review

The process of subjecting a scientist’s research to the scrutiny of other scientists working in the same field. Studies published in medical journals are usually peer reviewed, whereas conference presentations are not.

sexually transmitted diseases (STDs)

Although HIV can be sexually transmitted, the term is most often used to refer to chlamydia, gonorrhoea, syphilis, herpes, scabies, trichomonas vaginalis, etc.

Out of the total 39,400 persons diagnosed with HIV in these areas, serum samples from 6,864 were analysed using the BED assay. Of this subset, 2,133 samples—or about one-third—were classified as recent infections.

The researchers then extrapolated from these numbers to estimate the number of new infections in all 50 states and the District of Columbia, coming up with a total of 56,300. This represents an incidence rate of 22.8 per 100,000 people for the U.S. population as a whole.

CDC Director Julie Gerberding called this “the most reliable estimate we’ve had since the beginning of the epidemic.” Advocates have criticised the agency for failing to release the updated numbers sooner, but Gerberding said the CDC wanted the data to be thoroughly vetted by a peer-reviewed medical journal before it was released to the public.

The updated estimates confirm that the largest proportion of new infections—53%—is occurring among gay and bisexual men. HIV incidence has steadily increased among this group since the early 1990s, while the rate of new infections among heterosexuals and injection drug users has fallen. Women accounted for 25% of all new infections.

By race/ethnicity, African-Americans account for 45% of new infections, compared with 35% among whites and 17% among Latinos. According to the revised estimates, blacks have an infection rate nearly three times higher than that of Latinos and seven times higher than that of whites (83.7, 29.3, and 11.5 per 100,000 people, respectively).

Calculating backwards, the investigators then applied the data to adjust historical incidence estimates from prior years (1977–2006). They derived a new estimate of 55,400 new infections per year for the 2003–2006 period. For the population as a whole, the annual number of new infections peaked at around 130,000 in the mid-1980s, fell to a low of about 49,000 in the early 1990s, increased again to about 58,000 in the late 1990s, then declined slightly after 1999 and has since remained stable at around 55,000; the number probably never was as low as 40,000.

In announcing the findings, the CDC acknowledged that the new figures “reveal that the U.S. epidemic is—and has been—worse than previously estimated.” But Kevin Fenton, Director of the National Center for HIV/AIDS, Viral Hepatitis, STD and TB Prevention, emphasised that the updated 2006 estimate “does not represent an actual increase in the number of new infections, but reflects our ability to more precisely measure HIV incidence and secure a better understanding of the epidemic.”


Hall HI et al. Estimation of HIV incidence in the United States. J Amer Med Assoc 300: 520 – 529. 2008.

Centers for Disease Control and Prevention. New technology reveals higher number of new HIV infections in the United States than previously known (media release). August 3, 2008.