Saturday, October 9, 2010

Meaningless HIV Hypotheses Result in Meaningless Predictions

The Lancet has published a report on the "aids2031 project" which "modelled long-term funding needs for HIV/AIDS in developing countries with a range of scenarios and substantial variation in costs: ranging from US$397 to $722 billion globally between 2009 and 2031, depending on policy choices adopted by governments and donors." The world's press has dutifully repeated choice bits of the report without comment, question or (God forbid!) criticism.

Indeed, the report has the usual veneer of confidence that you find with long term predictions, lots of big figures, a range of scenarios, a typology of country epidemics and some pretty diagrams. If this was about bailing out a handful of rich bankers, the amounts of money would be far higher, the period of time would be shorter and the money would be handed over without much fuss. But as there are tens of millions of people involved and they are mostly living in poor countries, serious questions need to be asked about which spending scenario is 'best'.

The figures are, of course, entirely meaningless. HIV was discovered at a time when it had probably only infected a few million people. It is not known why the virus spread so rapidly in a few countries, all in sub-Saharan Africa, yet it only infected people in very specific risk groups in other countries. In many countries, transmission of the virus peaked around about ten or 15 years after it arrived and then declined. Therefore, in some countries the virus may well be peaking now, in some countries transmission rates continue to drop and in others it has already dropped to a low level.

But the question of why HIV struck, spread, peaked and declined has never been answered. As a result, the virus appears to have waxed and waned independently of what any country did to reduce its spread, which was in most cases nothing, anyhow. In countries like Kenya, Uganda and Tanzania, prevalence is stuck at between 6 or 8%, suggesting a steady stream of new infections. Many are being treated, but how many and how successfully is anyone's guess. Body counts are good for fundraising but not for reassuring funders that everything that can be done is being done.

The four spending scenarios all assume that a steady stream of new infections may continue in some countries and that transmission rates in higher prevalence countries, such as South Africa, Namibia, Botswana, Swaziland and others, will continue to decline. And this may be true, new infections may decline in every country. If, as is assumed by the HIV orthodoxy, HIV is mostly transmitted sexually, rates could never go particularly high anyhow. There is a limit to how much sex people can have. The fact that sexual behavior seems to have only a small effect on HIV transmission rates is irrelevant to the orthodox view.

But the authors of the report don't seem to consider the possibility that HIV transmission rates in some, perhaps many countries, could suddenly spike. Of course, there is unlikely to be a spike in sexual activity, especially a level of activity that goes beyond what is humanly possibly. But such a spike must have occurred at some time, if HIV is almost always transmitted sexually (and I'm talking about African countries).

If you don't accept that HIV is almost always transmitted sexually, a spike of non-sexual transmission, perhaps many spikes, are always a possibility. Such spikes, perhaps occurring in health facilities, involving unsafe injections and other procedures, could occur at any time and are the most reasonable explanation for rates of HIV transmission that has never been accounted for by levels of sexual activity that are beyond what humans are capable of.

The problem with the four different scenarios discussed in the report is that they are not based on any real epidemics, instead they are based on a very faulty paradigm. One can easily generate an infinite number of equally meaningless scenarios based on that same paradigm. None of the scenarios deal with improving health systems or health infrastructure, which could reduce transmission considerably.

Poor conditions in health facilities, which can quite easily explain otherwise inexplicable rates of transmission, remain in all high prevalence countries (or ones that once had high prevalence). As long as they remain, planning for the future is a shot in the dark. But as the HIV industry have long been demonstrating, you don't develop a better aim by practicing shooting in the dark. It's only advisable if you don't know what you've hit, or don't much care.

The authors argue that "substantial reductions in incidence (≥50%) will only occur with introduction of a vaccine or curative treatment." I disagree. Substantial reductions in incidence can be achieved by ensuring that nosocomial transmission is eradicated in every country, especially those which have ever had high rates of transmission. As long as health facilities and services remain underdeveloped and inaccessible, further serious outbreaks of nosocomial HIV transmission can not be ruled out. The authors of the report are not even in a position to diagnose what has been happening in the past, let alone make predictions 20 years hence.

(For discussions of Pre-exposure Prophylaxis (PrEP), see my other blog)


No comments: