Estimations of HIV incidence in the US from 2006 to 2009, based on data from 16 states extrapolated to the country as a whole, show stability over the period in respect to rates overall yearly incident cases (2006: 48,000; 2009: 48,100) and stability in respect to its ethnic distribution (incidence for Blacks and Hispanics respectively 7.4 and 2.8 times incidence for Whites in 2006, and 7.7 and 2.9 times in 2009). However, this stability in overall incidence conceals a significant shift in age distribution over the four years 2006-2009 – notably a 21% increase for the 13-29 age group, with a concomitant fall in other groups. The authors show this increase in the youngest age group to be driven by a 48% increase of incidence for young Black MSM, with no other ethnic sub-groups of the 13-29 age group showing a significant increase.
All this would suggest that, in US as elsewhere, the HIV problem is increasingly located in specific populations. Prevention and treatment will need to attend to the needs of these populations – as indicated in the UNAIDS strategy and 2011 report.
Estimating yearly HIV incidence is not straightforward due to the long latency between infection and symptom development. These estimates, originating from the HIV Incidence Surveillance Group, set up by the US Centers for Disease Control and Prevention (CDC), rely on systems funded in selected localities by the CDC to submit remnant HIV-positive blood specimens to serological testing and collect supplementary data on HIV testing and antiretroviral use required for the estimation. They also rely – given the less than universal coverage of these systems, and the logistical challenges of securing blood specimens even in selected areas – on development of extrapolation methods. The research paper, containing the estimates, provides a full account of the extrapolation method employed.
Since 2008 HIV incidence surveillance areas have included 25 health departments, including 18 state and 7 city/county health departments. Only localities meeting certain criteria (including 15% completion of the serological test) were included in the analysis – amounting in all to 16 states and 2 cities. It would be interesting to learn whether and how far the US experience of developing systems of HIV incidence surveillance replicates that of other countries.
Joseph Prejean, H. Irene Hall et al., for the HIV Incidence Surveillance Group, “Estimated HIV Incidence in the United States, 2006–2009”, Public Library of Science (PLoS) – one, 3rd August 2011