Why Can't We Track HIV At Home?

I make a brief return to the Washington Post’s pages this weekend with an Outlook section piece about the AIDS problem in D.C. As I say there, and in a previous posting here, I think some of the important trends got lost in the furor after last weekend’s report about a 3 percent HIV rate in the city. And there’s not much evidence of an AIDS problem that’s actually getting worse in Washington, or one that resembles problem spots in Africa.
That said, there is something elusive about what’s happening in Washington. We simply lack the surveillance tools to know if, for example, there is a new spike of infection in the city. The report offers no evidence of one, but it also doesn’t offer much in the way of contrary evidence. Because HIV takes many years to develop in AIDS, and because the D.C. surveillance systems passively rely on reports rather than seeking out trends at sentinel testing sites as is common in Africa, we’re left guessing at the most important question: How many people are getting newly infected with HIV today?
In scientific terms, this is called HIV “incidence,” as opposed to the more commonly quoted “prevalence” that merely tracks how many people are alive with the virus, no matter whether they got it last year, or during the peak of transmission in the 1980s.
Amazingly, it was easy for me to find a pretty good estimate of HIV incidence for urban Uganda. Reasonably good studies of incidence exist for many other African countries. Yet it was impossible for me to get one for Washington, D.C.
As much as I’m wary of comparisons between AIDS epidemics in Africa and the ones here, I do think that some of the tools used to guide valuable programming there should be considered for deployment here. The new D.C. HIV/AIDS report is much better than anything we’ve had for the city before, but the risk is that because that data based on old infections, we still might miss the important trends today. If we are serious about a renewed commitment to knocking down the rate of new infections, we need to know when, where and how they are happening.