Global priorities for HIV prevention: are these the right ones?
As there has been much media coverage lately on HIV-AIDS including concerns around funding, it appears that there is a clear need to reassess the directions of global HIV prevention. The International AIDS Society meeting in July may provide an opportunity to create some consensus on what the priorities should be. I sought the opinions of Dr. Daniel Halperin, at the Harvard University School of Public Health, and wanted to share his responses to questions regarding the top priorities currently in HIV prevention for the community's feedback and reflections:
DH: "Well, funny you should ask, as in recent weeks I've actually been trying to distill down, concisely, regarding what the core priorities for HIV prevention should be now. (And David Wilson of the World Bank and I had a paper in The Lancet, for the Mexico AIDS Conference two years ago, which attempted to lay out essentially the same points;
http://www.thelancet.com/journals/lancet/article/PIIS0140673608608831/fulltext). In summary, I've proposed the following three sets of recommendations which I believe would result in more effective HIV prevention and thereby increase the return we get on investments into AIDS prevention, both in terms of infections averted and potentially for other health outcomes as well. They are:
1. We need once and for all to eliminate the confusion around the types of epidemic we are addressing (specifically, whether concentrated or more generalized), and move away from issuing blanket recommendations for HIV prevention across all settings to using more targeted recommendations that account for the specific epidemiological context. Hence, in those epidemics where sex work, needle sharing, and male-male sex are the main sources of infection (which is the case in the large majority of countries in the world), it's ludicrous to keep putting large investments into diffuse interventions for "youth and women in the general population," when it's clear that we could avert far more infections, and at less cost, by focusing more closely on "100% condom" and other effective approaches for sex work, on needle exchange programs for IDUs (and also on methadone and 12-step type of programs to help get people off of drugs altogether), and on more effective strategies for MSM (promoting consistent condom use, partner reduction, addressing alcohol and drug use, which often is also an issue for sex workers, and probably we also need to consider male circumcision availability for insertive MSM partners). Luckily, under Obama the PEPFAR program seems to have improved considerably in terms of recognizing the need to focus more on high-risk populations ("MARPs").
However, there also is a danger of perhaps OVERLY focusing on such high risk groups -- in those relatively very few countries which have truly generalized epidemics (especially in the some dozen countries in southern Africa and parts of eastern Africa which are the world's most affected regions, but to some degree also in some other parts of sub-Saharan Africa, the Caribbean, and probably in the island of Papua/New Guinea). Even there it would, however, be a mistake to ignore these classic high-risk populations, but in the most affected countries they usually account for only a small % of overall infections so -- unlike in most other places -- the large majority of attention and investments needs to focus on risky sex (i.e. multiple and especially concurrent sexual partnerships) in the general population.
2. Along these lines, we need to do a much better job of identifying and deciding on the highest impact prevention strategies, for a given epidemic, and making these the central priority of national HIV prevention strategies. As I mentioned, for most epidemics (the concentrated ones) this would include targeted interventions for high risk groups such as promoting consistent condom usage for commercial sex work. And in generalized epidemics, the most promising approaches are making safe and affordable voluntary male circumcision much more available, and in promoting social norms to reduce multiple and concurrent sexual partnerships, as Jim Shelton of USAID emphasized in a recent short paper in The Lancet (http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2810%2960416-3...).
Tachi Yamada of the Gates Foundation had a thought-provoking opinion piece in the New England Journal of Medicine a couple years ago that made a very important point, which is that one of the main reasons prevention programs have been so dismally ineffective is because they have generally employed what he called a "consensus based" approach, instead of a truly evidence-based one. In my view, the latest incarnation of the consensus based approach is what many people are calling "comprehensive" or "combination" prevention. They claim it's a mistake to propose (as some people like Wilson and I have, and 10 researchers in a policy paper in the journal Science did two years ago) that there should be a (re)prioritization of prevention approaches (i.e., to select the two or perhaps three top priorities for each type of epidemic), but they believe it's instead better to just "throw everything but the kitchen sink" at the problem. These critics have, in various opinion papers and conference presentations, etc lately, analogized HIV prevention to ARV treatment: where using only one or two medications doesn't' work, but when the (correct) three ARV medications are given to the patient, then the effects are positive. Hence they argue that although it's true that most of the standard approaches (like HIV testing and counseling, treatment of other STIs, and mass promotion of condoms or abstinence in generalized epidemics) have not been shown to be effective in randomized trials or other evidence, that in COMBINATION such approaches do in fact work. However, and as I tried to explain in a short piece last year in The Lancet (http://www.lancet.com/journals/lancet/article/PIIS0140-6736%2809%2960200-2/fu...), the difference is that whereas for triple ARV therapy there is very clear (RCT, etc) evidence that this works, for "combination prevention" there is so far only theoretical or speculative evidence that it may work, and in fact there have already been some randomized trials which attempted to test just this notion, but none of these trials reduced HIV incidence. (And in one of the most important such studies, conducted in Manicaland, Zimbabwe, HIV incidence was probably even higher among those who received
the "combination prevention" intervention.)
Of course, no one, including myself, is arguing for only ONE (or even only two) approaches to prevention! (However, recently some people have been arguing, quite passionately, that "testing and treating" entire populations in Africa could eliminate the epidemic over time, but this is based on very optimistic modeling assumptions and some other rather dubious notions.) That said, there is an important difference between looking for a magic bullet (as some of our critics have accused us of pursuing) and in PRIORITIZING which interventions are most likely to have the largest impact (and at the lowest cost). This is what the ten of us wrestled with/tried to do in our 2008 Science paper, and came to the conclusion that -- in the generalized epidemics -- male circumcision and addressing multiple concurrent partnerships are likely to have the largest prevention impact. (We also wanted to include, for reducing mother-to-child-transmission, the importance of increasing access to family planning services and of educating populations about the importance of practicing exclusive breast feeding, but were hemmed in by the journal's strict space requirements: we had to limit the entire policy piece to only two pages!).
By the way, the recent acknowledgment by President Zuma of South Africa that he has been circumcised and has recommended it not only for his own sons but for the entire country, is an especially promising development, one mirrored by the Zulu King's recent public advocacy of the procedure (which was abandoned by the Zulus almost two centuries ago), as well as the open discussions around multiple concurrent partnerships sparked, perhaps ironically, in part by Zuma's own recent behavior.
3. Finally, we need to do a better job at considering HIV-AIDS in the context of the larger global health picture, by looking for example at how HIV prevention relates to overall health system strengthening and to some of the recently more neglected areas of health, like child and maternal mortality and family planning -- where relatively small but smartly targeted investments could produce considerable results in health and other social benefits. At the end of the day, our objective should be healthy people, and HIV is just one disease of many, and in fact is not at all the overriding health problem in most poor countries. Luckily, the Obama administration and some others are increasingly realizing that finding ways to leverage HIV-AIDS activities and funding to improve overall health delivery and outcomes is really a no-brainer."
What do others think? I'm curious as to whether these are indeed the right priorities for the HIV prevention community, and any other reactions to Dr. Halperin's comments, and/or other ideas for how to improve HIV prevention efforts globally.