For those of us who work in the HIV/AIDS field, the month of July was dominated by exciting HIV prevention news coming out of the International AIDS Society meeting in Rome. Results from the HPTN 052 study showed that early, compared to delayed, antiretroviral treatment resulted in a 96% reduction in HIV transmission to uninfected partners. The TDF2 study conducted by CDC in partnership with the Botswana Ministry of Health, found that a once-daily pill containing two anti-HIV drugs reduced the risk of acquiring HIV infection by about 63% in a study population of healthy, heterosexual men and women. These and other study findings continue to add weight to the notion that HIV treatment is prevention. All of us are encouraged when we think about how these findings could be translated into real world settings in a way that would bring us closer to achieving the goals of the National HIV/AIDS Strategy.
Without minimizing the tremendous enthusiasm that rightly attends the prevention breakthroughs that were presented in Rome, I would like to talk about another scientific discussion that took place in July. As it turns out, this meeting was also held in a world capital, although to a much smaller audience. And while the results of this two-day meeting didn’t garner media attention the same way as the Rome meeting did, the topics under discussion were no less consequential. In mid-July, I was very fortunate to attend a two-day workshop on “Modeling and Evidence-Based Decision Making” sponsored by amfAR, the Foundation for AIDS Research and cosponsored by the Kaiser Family Foundation , the National Alliance of State and Territorial AIDS Directors , and the Urban Coalition for HIV/AIDS Prevention Services . Meeting participants included colleagues from state and local departments of health, academia, federal government, and professional and community-based organizations.
Colleagues from Los Angeles, San Francisco, Maryland, and New York City shared with us their experiences with using various models to assist in making decisions about “optimizing” HIV prevention investments. Using different approaches, each of these health departments was trying to answer the same question, “What combination of prevention services and activities will result in the greatest reduction of the number of new HIV infections?”
At the onset of the meeting, we were reminded that modeling is used in other areas of health and public policy decision-making, especially when leaders are trying to combine diverse information from a variety of sources in order to make sound decisions at a population level. However, even the biggest fans of modeling reminded us that a model is not a “crystal ball” nor is it infallible. Instead, what models do is provide a tool to help us make better decisions about complex realities. Good models should always be clear about the inputs and assumptions that were used to generate the results. And perhaps most importantly, they should be used to guide rather than to conclude any discussions about how best to allocate resources.
The meeting ended with a great deal of enthusiasm about the promise of models to help to support sound decision-making. Several participants noted that the discussion of models to improve HIV/AIDS decision-making only served to further underscore the need for federal agencies to agree on a set of core, standard program measures that can be used as standard inputs. Others noted that it would be helpful to identify a small subset of usable models that could be shared with colleagues who are interested in using them to help guide program decision-making. All participants noted the need to provide technical assistance to facilitate the understanding and uptake of models in local and community settings.