Monday, June 25, 2012
It's an oft used metaphor about development that if all you have is a hammer, every problem looks like a nail. Health and education related development are no exceptions. At different times and by different organizations, they have been dominated by the likes of birth control and related issues, water, nutrition and, of course, sex. If all you talk about is sex and all you ask about is sex, you will end up with a very rarefied view of your research area. Which is why HIV policy in African countries is mostly about sex and mostly not about any non-sexual modes of transmission.
A recently published paper by Samuel Manda, Carl Lombard and Thabang Mosala is a case in point. The title is explicit enough: "Divergent spatial patterns in the prevalence of the human immunodeficiency virus (HIV) and syphilis in South African pregnant women". A map that represents HIV prevalence in a country looks very different from a map that represents syphilis because the areas where HIV prevalence is high, in general, are places where syphilis prevalence is low, and vice versa. In fact, "HIV was more highly prevalent in deprived and populated urban areas than elsewhere, while syphilis had a high prevalence in less deprived and less populated rural areas. Spatially, the HIV prevalence was lowest in the southwestern and highest in the northeastern parts of the country".
Though this sort of phenomenon has been demonstrated before, in research going back many years, the researchers assumed that syphilis is a "proxy for sexual behaviour and increased HIV transmission". They conclude that the "differing disease-specific spatial prevalence patterns may point to inconsistent successes in interventions between the two diseases". Syphilis prevalence may indeed be low in some areas as a result of successful eradication programs. But is there a hint that successful syphilis eradication programs may have inadvertently increased HIV prevalence? Well, I'm sure that's not one of the conclusions the authors are hinting at.
But their contention that the "discordant ecological and spatial effects between HIV and syphilis may suggest that the syphilis prevalence is not suited as a predictor of HIV prevalence" sounds spot on. Not only that, but sexual behavior is not suited as a predictor of HIV risk or HIV prevalence; and HIV status is not a reliable indicator of the sexual behavior of HIV positive and HIV negative people. HIV can be transmitted sexually, sure, but it can also be transmitted non-sexually. The question is, to what extent is the virus spread non-sexually, because prevention interventions for various modes of transmission should be very different; interventions should not be limited to hammers.
Or scalpels; because if all you have is a scalpel (or a PrePex device or some other gadget invented to perform 'bloodless' circumcision, not sure if bloodless is metaphorical) you may be tempted to circumcise a lot of men at great expense. There are better ways of spending a few billion dollars, whether it's spent on development more generally, health or even HIV, specifically. And mass male circumcision could do a lot of damage; it may give rise to an increase in unprotected sex, which would result in higher prevalence of syphilis and other STIs, not just HIV, and an increase in unplanned pregnancies. In a setting where healthcare is not safe, or where standards are not very high, HIV and other diseases might even be transmitted during the operation itself, or shortly after.
The metaphor seems to be bountiful in the HIV field; if all you have is contraception, every problem looks like an unplanned pregnancy or an unmet need for birth control. Statistics for every health issue affecting women and children seem to be used to press home the need for greater access to 'modern' methods, such as condoms, implants or, the big one in African countries with high HIV prevalence, Depo Provera and other forms of injectable hormonal contraception. The fact that Depo Provera may increase HIV transmission, from women to men and from men to women, is evidently not enough to stop this from being the number one birth control method in some countries, such as Kenya.
If you wanted to kill someone with a hammer you'd probably have to bludgeon them on the head with it or force it down their throat. And that's pretty much what UNAIDS does with its 'messages' about HIV prevention, sex and safe sex, which might be better referred to as sentences. In countries where sex is taboo, insisting that someone has a sexually transmitted infection which their partner doesn't have can be condemning them to ostracization, persecution, violence or even death. Why not tell the truth; that we don't yet know the relative contribution of non-sexual and sexual modes of HIV transmission? And then start to answer that question.
[For more about non-sexual HIV transmission and mass male circumcision, see the Don't Get Stuck With HIV site.]