Estimating the Proportion of Undiagnosed HIV Cases

Keith Henry, MD reviewing Fellows IE et al. PLoS One 2015.

Using novel methodology, researchers estimated that in King County/Seattle, Washington, 6% of HIV infections in men who have sex with men are undiagnosed, with wide variation by race/ethnicity.

The first step in the HIV care cascade is determining the proportion of HIV infections that remain undiagnosed within a given jurisdiction. This task is difficult and often requires either testing a representative sample of high-risk individuals or obtaining an indirect measure by back-calculation to estimate HIV incidence. Now, researchers have used two new methods to estimate the undiagnosed fraction among men who have sex with men (MSM) in King County, Washington.

The first method, a back-calculation approach, requires no assumptions about HIV-incidence trends; the second is a simple formula used when HIV incidence is stable. Both methods use data on HIV testing history for newly diagnosed cases to estimate the time of last negative test, the possible infection interval, and the distribution of time from infection to diagnosis.

Between 2006 and 2012, 1522 MSM in King County received a diagnosis of HIV infection; the date of the last negative HIV test was known for 1233 of them. By the authors' best estimate, 6.0% of HIV-infected MSM were unaware of their infection, with levels almost twice as high in Hispanics (9.3%) and blacks (8.6%) as in whites (4.9%). Use of the most conservative assumptions yielded an upper bound overall estimate of 11.0%. These figures are well below the CDC's nationwide estimates of undiagnosed HIV infections in MSM.

Citation(s):

Fellows IE et al. A new method for estimating the number of undiagnosed HIV infected based on HIV testing history, with an application to men who have sex with men in Seattle/King County, WA. PLoS One 2015; 10:e0129551.