FIND researchers highlight issues, answers in tracking HIV incidence
How do you measure the impact of DREAMS?
The goals of the U.S. and partner-backed initiative to confront conditions fueling the spread of HIV among teenage girls and young women and to help the target population “develop into Determined, Resilient, Empowered, AIDS-free, Mentored, and Safe women,” seemed hard to quantify, but then the White House set a target. In high-burden areas across the 10 countries where the initiative was launched, rates of new infections would drop — specifically by 25 percent this year, and by 40 percent by 2017.
The challenge then, people who study HIV incidence – or rates of new infections — say, is how to find out if that happens.
Counting numbers of newly infected people to track the spread of an epidemic and to see if efforts to fight it are effective is a basic tenet of public health. But in the global HIV arena, where incidence is a critical indicator used to gauge the population level impact of a number of ongoing U.S. supported interventions — but where years can lapse between infection and diagnosis — getting accurate and rapid measures of incidence is both critical and complicated. One way to do it is through modeling: take numbers currently known, add information that can include the known impacts on those numbers, and do the necessary math. But that method can leave out unknowns, introduce bias, and produce a number that does not reflect a full reality. The findings of a recent HIV incidence household survey in Rwanda gave a glimpse of the gap that can open between estimated and actual incidence. That study showed how a two-stage household survey there found incidence that was higher than that estimated by modeling,while yielding more specific information on where new infections were occurring. Longitudinal studies, in which members of a population are tested and retested over time is considered a “gold standard” of incidence measuring, but require multiple tests among large sample populations — potentially thousands of people for representative results — for upwards of two years.
A reliable way to measure incidence, say researchers for FIND (Foundation for Innovative New Diagnostics), a nonprofit seeking appropriate diagnostic tools for resource-limited environments, factors in viral load and immune response test results to determine the recency of infections. The smaller the number of test results needed, the more efficient and affordable the process — or “RITA” — for recent infection testing algorithm — would be. In a study funded by the Bill and Melinda Gates Foundation (also partners with PEPFAR on the DREAMS initiative) and presented at the recent Conference on Retroviruses and Opportunistic Infections, researchers found the currently used RITA could work — producing reasonably precise estimates of incidence, and among some specific populations (or “key populations”). They also noted that the longer the period of time used to define “recent” infection, the smaller the sample size was that was needed to yield an accurate estimate of incidence.
The search continues. New measuring procedures — or assays — are needed, researchers say, to make possible broader and more cost-effective measures of incidence possible in both national and key population surveys, to assess the impacts of population-level prevention interventions.