Monday, February 13, 2012

Collecting Data Only to Explain it Away

The HIV industry loves the story of massive increases in 'unsafe' sexual behavior as an explanation for rapid increases in HIV prevalence in high and medium prevalence countries. If the author can even be bothered to say why unsafe sexual behavior should have increased at some period, they vaguely point to urbanization (which started long before HIV spread from its epicenter and continues in most high prevalence countries thirty years later) or some such phenomenon. As for the decrease, it is rarely mentioned, unless it is to say that it was a result of various policies, strategies and activities, with which the HIV industry of the day was fully engaged.

The story must assume or even include subsequent and equally massive decreases in unsafe sexual behavior when prevalence rapidly drops later. The problem is, the drop in prevalence is more likely to be a result of very high death rates, which tend to occur some years after the epidemic began. As for these increases and decreases in sexual behavior alluded to, there is no evidence that either of these ever took place. It's just that if you assume HIV is almost always transmitted sexually, things have to have happened as the orthodox account says they did.

Of course, sexual behavior may have changed in various ways; unsafe behaviors may have waxed and waned, just as they have probably done elsewhere and at various times. Since sexual behavior data has been collected with a view to getting HIV transmission under control, things have certainly waxed and waned in Uganda, which is often cited as the best example of a country where HIV prevalence dropped because the country did all the right things. But unfortunately, even though HIV prevalence has bumped along at a figure that still makes Uganda a medium prevalence country, sexual behavior appears to be thoroughly waxing at the moment, and not for the first time since their heroic success of the 1980s.

There's an article in the Daily Monitor about PEPFAR funded research which shows that plenty of the behavior considered to be 'unsafe' by that bastion of sexual safety still occurs in Uganda. "20% city students in sex trade - report" screams the headline. There are also figures about the number of young people conceiving, having abortions, having sex with relatives for money or gifts, underage sex, transactional sex, rape, oral sex, anal sex, alcohol, drugs and lots more.

These are all extremely worrying phenomena; but does anything about this research show whether they relate to HIV or how they might relate to HIV? Certainly, past research tends to show that a lot of HIV transmission does not correlate very closely with sexual behavior. And if the above trends in sexual and other behaviors among young Uganda people are increasing, why is HIV not also increasing? Perhaps that will follow. But there have been articles about increasing levels of unsafe sexual behavior in uganda going back many years.

There are the usual implications that HIV transmission is being 'driven' by things like higher vulnerability among girls/women, economic dependency, transactional sex, etc, and that people need education and what not. Which is all true, the bad things mentioned are indeed bad things and people need good things, many of which are mentioned in the Universal Declaration of Human Rights. But good things have not, despite the claims of the HIV orthodoxy, been unambiguously associated with lower HIV transmission and bad things have not been unambiguously associated with higher HIV transmission.

A report from Namibia shows that good and bad things also happen there but that, in contrast to Uganda, HIV prevalence increased among a large group of people being monitored for changes between 2006/7 and 2009. Figures were higher for females, as they are in most sub-Saharan African countries, but there was very little correlation with socioeconomic factors. And while HIV knowledge was 'strongly' associated with lower prevalence, most people didn't appear to know a great deal; 75% either felt they had no risk of being infected with HIV, or they didn't know or refused to answer the question.

The paper concludes that neither prevalence nor incidence are declining and that everybody in urban areas (the research was carried out in Windhoek) is at risk. But it is hard to know what to conclude from the data collected. Only 55% of the people surveyed in 2006/7 were surveyed again in 2009, suggesting that a lot of the conclusions could be quite biased (possibilities which are discussed in the paper). It's great that the research identified geographical areas "that would require prioritized HIV campaigning"; but do we know why so many of city dwelling Namibians, a lot of whom actually face fairly low HIV risks, are being infected?

The Ugandan research may show that high rates of 'unsafe' sexual behavior do not necessarily result in high rates of HIV transmission. But the Namibian research seems to show that researchers can't quite say why sexual behavior, whether high risk or low risk, seems to result in prevalence figures that are two or more times higher than those in Uganda. Ironically, the Namibian research was carried out with a view to "improving access to affordable health care". Access to health care in Namibia is very high compared to that in Uganda. It is despite very low access to poor quality health care in Uganda that HIV prevalence is medium, unlike in Namibia, where it is high.

Sexual risk is not the only phenomenon relevant to HIV transmission, neither in Namibia or Uganda or any other country where HIV is transmitted; there are also non-sexual risks, such as unsafe health care, intravenous drug use and unsafe cosmetic practices. Both of these accounts of HIV concentrate on heterosexual risk, as do most accounts of African HIV epidemics. Data collected generally relates either to sexual risk or relates to other phenomena that are thought to or are claimed to relate to sexual risk. Instead of tying themselves in knots to explain away data that doesn't support the orthodox view, it's time to reconsider the orthodox view. Wouldn't that be more scientific?



Anonymous said...

The Economist claims that a lot of HIV is spread through relationships between young women and much older men.

What do you think?

Simon said...

The connections between HIV and poverty have always been unclear: in many countries, HIV prevalence was higher in wealthier quintiles and among the better educated. Giving people money in the form of conditional or unconditional 'cash transfers' is just a more recent term for handouts, which are a lot less acceptable to Western donors now. The Economist is wrong about one thing: that HIV is spread through any one thing. Many years of research has shown that there are many factors involved, which has allowed journalists to skirt the issue of non-sexual HIV transmission. But hey, they get to write about sex, what more could a quality journal ask for?