Paid to Fail: the Politics of HIV Research

I recently commented on a much hyped but relatively vacuous paper by Halperin, Mugurungi, Hallett, Muchini, Campbell, Magure, Benedikt and Gregson, entitled 'A Surprising Prevention Success: Why Did the HIV Epidemic Decline in Zimbabwe?'

But there was one item in the paper I didn't comment on that has stuck in my mind since. They write: "Many men in focus groups and interviews reported that having less disposable income has increasingly led to reduced ability to purchase sex or maintain multiple sexual relationships".

The sentence sounds reasonable enough if you accept the overall conclusion of the paper, that changes in sexual behavior driven by fear of HIV infection resulted in massive reductions in HIV transmission. But the data used by the paper only shows minor changes in sexual behavior. In truth, correlations between sexual behavior and HIV transmission are as unconvincing as they always have been, in Zimbabwe and elsewhere.

But even thinking about this sentence from an economic point of view and the little we know about commercial sex, does having less disposable income necessarily result in a reduced likelihood of purchasing sex? I think the real worry when money becomes scarce is that those who depend on providing sexual services for money have to settle for less money or provide more and/or riskier services.

Of course, such hypotheses need to be tested and many HIV researchers are reluctant to carry out rigorous research into sexual behavior. If sex turned out to be less relevant to HIV it's likely that funding would dry up. And if HIV prevention turned out to be a matter of providing decent health care, those currently selling drugs would also quickly lose interest.

There must be a lot of money in providing safe health care but it doesn't seem to attract the donors. But then, it's important to sell HIV drugs to developing countries because that's where the bulk of the market is. The same can't be said for other health care goods and services; not yet, anyhow.

Another problem with the idea that commercial sex becomes less common when money is scarce is that you'd think the trend for wealthier people to be infected in greater numbers than poorer people would reverse. This trend has reversed in some countries, but not noticeably as a result of economic changes.

The authors of the paper in question put a lot of credence in focus group discussions and that seems unwise. But it's their analysis of those findings that seem most disingenuous. The analysis appears to be independent of the data on sexual behavior, probably because the data is the same in high prevalence countries as it is in low prevalence countries. It's almost as if they are saying "here's the data, it contradicts our conclusions but everyone knows our conclusions are true anyway".

Because this paper doesn't even give a convincing pretense of having any substance, I'm still wondering what is behind it. The authors are very well established so the only reason I can think of that they would put their names to such rubbish is that they know no one really cares how they come to their conclusion as long as they come to the approved conclusion. But it seems like an expensive and circuitous way of failing to reduce HIV transmission.

Researchers Have No Opinion On Nosocomial HIV Infections in Zimbabwe

Non-sexual HIV transmission, when it's even discussed by the HIV orthodoxy, is usually dismissed with little argument and no evidence. What is most extraordinary is that one could hypothesize that both sexual and non-sexual transmission contribute to most epidemics and then try to work out the relative contribution of each. But they don't tend to do that.

Halperin, Mugurungi, Hallett, Muchini, Campbell, Magure, Benedikt and Gregson toe the party line in their recent paper 'A Surprising Prevention Success: Why Did the HIV Epidemic Decline in Zimbabwe?' They barely even mention non-sexual transmission and completely dismiss its significance.

But they do come to a very media friendly and quotable 'conclusion', that "fear of contracting the virus [is] the primary motivation for changes in sexual behavior". Journalists have pounced on this 'finding' and will continue spreading it for some time. Perhaps these researchers have recognized the value of media friendliness and found it to be more congenial than credible, enlightening research that could turn around the HIV pandemic.

Despite constant boasts about the number of people on antiretroviral treatment and the idea that you can contain an epidemic by throwing lots of drugs at it (which happens to be the current global treatment policy), these researchers even mention the very real possibility of drug resistance making mass treatment campaigns less sustainable than they currently are. They are in good company; Bill Gates recently said more less the same thing.

But what was their quotable conclusion based on? Well, they did a bit of mathematical modelling and read a few papers written by like-minded people (actually, the bibliography overlaps considerably with the list of authors), but they also give a lot of credence to a bunch of 'stakeholders', who certainly seemed to do a fair amount of agreeing with each other. Perhaps they see this as quantitative, their credence, the stakeholders' agreement, etc.

It's odd, when people say they have never had sex, never had unprotected sex or never had sex with anyone other than their partner (who is often HVI negative), they are unlikely to be believed, especially if they are African. But if they are like-minded people holed up in a hotel, their responses are treated at face value.

"[T]he unanimous conclusion from the stakeholders meeting held to assess, triangulate, and interpret the evidence assembled in the review was that a reduction in multiple sexual partnerships was the most likely proximate cause for the recent decline in HIV risk." What a surprise.

It goes on: "In assessing the underlying factors for the national prevalence decline, high AIDS mortality appears to have been the dominant factor for stimulating behavior change." Yet, high AIDS mortality has been a phenomenon in many countries that have had very high HIV prevalence. When lots of people become infected, lots of them die, widespread treatment regimes notwithstanding.

Similar claims used to be made about Uganda, though these researchers are also keeping Uganda at arm's length. Well, it's almost certainly true that some people were devastated by what they saw around them when huge numbers of people were dying terrible deaths. That would have some impact on anyone.

But the idea that it would be almost entirely responsible for levels of behavior change that resulted in a massive drop in rates of new infections in a short space of time in Zimbabwe, but nowhere else, is not credible. Nor is it even necessary to make such a foolish claim.

The economic decline experienced in Zimbabwe in the late 90s and early 2000s, we are told, played a considerable secondary role in amplifying patterns of behavior change. No doubt it did. But economic decline could also have resulted in fewer visits to the country's deteriorating health facilities, which would have reduced the number of nosocomial infections (infections resulting from medical treatment).

And what levels of behavior change occurred? From the figures cited, age of sexual debut and condom use barely changed. And multiple partnership indicators improved a bit, but these were never common enough to explain the almost umprecedented rates of transmission once found in the country. Most of these indicators wouldn't even look out of place in rich countries.

Interestingly, the researchers mention "the Zimbabwean government's early adoption of a home-based care policy [which] may inadvertently have accelerated the process of behavior change. It has been hypothesized that, when people die at home, this direct confrontation with AIDS mortality is more likely to result in a tangible fear of death among family and friends than when patients are primarily cared for in clinical facilities, such as in Botswana"

I'd interpret the effect of this policy rather differently. It could also have taken a lot of HIV positive people out of a health system that was not able to provide people with safe healthcare.

The authors conclude that significant changes in behavior are unlikely to have resulted from increasing levels of mortality alone. They also suggest that prevention programs provided people with information about the link between risky sexual behavior and HIV transmission. And they are probably right, to an extent. But why were these programs so successful in Zimbabwe when they failed so miserably elsewhere? The authors bluster on, unconvincingly.

Indeed, they don't even seem that convinced themselves. They can't really put their finger on anything much so they talk about "cumulative exposure" to prevention messages, as if that wouldn't have happened elsewhere. Similar claims have long been made to "explain" what happened in Uganda. After all, there must be some explanation, and if it has to be about sex this one is as good as any other.

I can understand a whole group of stakeholders churning out answers that would satisfy even a UNAIDS employee about the drivers of HIV. I have met few people who wouldn't say similar things. But I don't believe the authors could look on this this paper as a publishable piece of research. If they are all happy with it, then I am disgusted. Their own research screams for investigation of non-sexual transmission levels, but they carry on regardless.