Targeted Testing for Acute HIV Infection in North Carolina

Because people with acute HIV infection contribute disproportionately to transmission of the virus, identifying these persons is a critical public health challenge. The current cross-sectional study outlines targeted approaches the authors developed for detecting HIV RNA in persons with negative serological tests.

Study participants included those undergoing publicly funded HIV testing in North Carolina between October 2002 and April 2005. Logistic regression was used to develop targeted testing approaches. Simple approaches were assessed based on clinic type and geography. Algorithm development used persons with recent HIV infection, determined by a detuned enzyme-linked immunosorbent assay. Validation used persons with acute HIV infection identified with an HIV RNA pooling procedure.

Of the 215,528 eligible persons, 232 had recent HIV infection and 44 had acute HIV infection. Five indicators (testing site, sexual preference, sex with an HIV-positive person, county HIV incidence, and race) combined identified 92 percent of recent infections when testing 50 percent of the population. In validation among persons with acute infection, this combination had sensitivities of 98 percent in years one and two and 88 percent in year three. A simple combination of testing site and county performed similarly [development (recent infections): sensitivity=95 percent; validation (acute infections): sensitivity=86 percent in years one and two; 81 percent in year three; cut-off established for testing 50 percent of population].

Targeted testing can accurately identify acute HIV infection, the authors concluded, suggesting "simple approaches for identifying the types of clinics and geographical areas where infections are concentrated may be logistically feasible and cost-efficient."