Bad HIV/AIDS science

The 'Spreading Heterosexual Epidemic'

During a recent interview it was put to me that AIDS was now a heterosexual disease too as equal numbers of heterosexuals were being diagnosed as gay men. The source wasn't quoted and I hadn't looked at those apparent figures in detail, but I thought it was an important point to answer thoroughly. So in addition to the verbal answer I gave I later on sat down and wrote a longer response utilising some figures I had already analysed more comprehensively.

As I think this is an important point to deal with, because the claim that  ‘HIV' is spreading among heterosexuals is used as confirmation of HIV diagnoses being primarily sexually transmitted, I decided to reproduce my detailed response here and to include more detail than the verbal response I had originally given. I'm not apologising for not having analysed the most recently available data, because the same kinds of misrepresentations in the presentation of HIV/AIDS related data has been occurring from the very beginning. Even the late, celebrated Mathematician Serge Lang resigned from the American Mathematical Society because he was so frustrated that his exposés of inappropriate and misleading statistical methods in HIV/AIDS figures were being ignored. The shit remains the same, only the year changes.

The important point about this is you can't really trust headline claims in the media at face value, because closer examination of the evidence from which those claims comes often reveals something different when analysed without a determination to come to a pre-determined conclusion. Even the conclusions from scientific papers can't be assumed to reliably represent the data from those studies as the data may be misrepresented or selected to fit a politically-acceptable outcome. So this is my overall response.

Are heterosexuals really being diagnosed HIV+ as often as gay men?

In answer to the point that equal numbers of heterosexuals are being diagnosed HIV+, that may be true. But what is the proportion of gay men in relation to the rest of the population? It depends on what your criteria are but for predominantly gay men it appears to be somewhere in the region of 5-8%. If in any given period this 5-8% of the population is still getting around 50% of the HIV+ diagnoses, then at that time HIV+ diagnoses are still heavily biased towards gay men. So, despite regular attempts to make it seem like a heterosexual sexually-transmitted epidemic in western countries, that isn't really happening to an appreciable degree that makes it credible as being predominantly sexually-transmitted, in the same way that other sexually transmitted infections are. (It is only superficially, not actually, a sexually transmitted epidemic in Africa).

I must make the point that some heterosexual, non-drug using women are still being diagnosed HIV+, and we must not pretend they don't exist. What I'm saying is that, for a variety of reasons, the notion of HIV+ diagnoses because predominantly or even significantly caused by unprotected sex is not tenable, then we must consider other explanations for diagnoses in those cases when it does happen.

But this doesn't explain why gay men would still be getting such a disproportionate number of HIV+ diagnoses, so what might the reasons be? I think there are a number of factors to consider.

Inherently hazardous behaviour

It still appears to me, from a variety of sources including some anecdotal evidence told to me personally, that gay men are still using recreational drugs to a much higher degree than the general population. These are not necessarily all injected drugs, but it is indicative of a degree of reduced care being taken over their own bodies by some gay men than among heterosexuals, and ‘reduced care' can mean more smoking, drinking, less eating properly, more nights out and less sleep, perhaps catching more others sexually transmitted infections, and a number of small factors that collectively may impinge on the body's ability to handle things. But why might that be, what with effectively gay marriage and everything in the UK?

Because even with greater freedoms and protections now, the vast majority of gay men have already grown up in an environment that has been extremely oppressive until recently. I think most straight people can't really comprehend the multitude of ways in which cultural norms have supported heterosexual ‘ways of being' while providing virtually no role models or acceptable paths for gay men. I'm pretty certain that there were both gay and straight people who were stunned at the number of older gay male couples who seemed to emerge from the woodwork when civil partnerships were made legal in the UK. Prior to that, they seemed not to exist, creating an urge for gay men to cling to youth (as one straight woman pointed out to me) in the absence of any apparent kind of gay life after 40.

I'm pretty certain that this wider unsupportive environment, let alone many gay people's own oppressive families that I've witnessed at first hand, has led to self-worth issues to some degree in a much higher proportion of gay men than in straight people. Low self-worth tends to result in behaviours that give a more immediate payback with less regard to long-term consequences, such as recreational drug use.

I'm not ignoring lesbians, by the way, just that as far as HIV/AIDS is concerned, they seem to be almost completely off the radar. Perhaps this is because women's greater natural tendency towards caution has tended to hold them back from living some of the most physiologically hazardous lifestyles that some gay men have embraced, seemingly competitively and without apparent restraint. Lesbian academic and author Camille Paglia who had many gay male friends in New York at the onset of the AIDS crisis said, "I've constantly said of the gay men of my generation that they challenged nature and lost....I've always felt that gay men were pushing the limits of the human body throughout that period"

Testing, testing....

But gay men are also encourage to get HIV tested regularly, much more so than the rest of the population for whom there is no constant urging to get tested at all. In fact, GUM clinics will often dissuade straight people from getting tested unless they can be convinced that the person has been exposed to some significant risk - from their perspective - of catching HIV, which doesn't usually include unprotected heterosexual sex. As an example of the persuasive and pervasive efforts to get gay men to associate themselves with AIDS, the issue of Pink Paper issue 965 had two full-page adverts encouraging gay men to have HIV tests and engage with HIV/AIDS related services, plus a persistent proportion of AIDS-related stories, references to the forthcoming annual World AIDS Day and pictures with people wearing red ribbons. Gay dating websites frequently have a specific field for declarinig your HIV status - ‘positive', ‘negative', don't know', ‘havent' tested recently', for example, a feature usually often conspicuously absent from heterosexual-aimed dating services. If AIDS really is affecting straight people as much as gay men, you wouldn't know from the gay media or services aimed at gay men.

So the issue about this is that a much higher proportion of gay men end up going for these non-specific measures of antibody levels, laughingly called HIV tests. The point is, if there are already documented so many non-specific factors that can temporarily trip your relevant antibody level above the arbitrary threshold that will trigger an HIV+ diagnosis, if you are testing one section of the population much more than any other they will inevitably end up being diagnosed HIV+ more frequently.

Heterosexuals who rarely feel the need to get tested may well avoid being diagnosed HIV+ simply because they weren't getting tested when a temporary greater level of the relevant antibodies had time to go down again without them being any the wiser that it was ever over the threshold that would have been diagnosed HIV+ if they'd been tested at that time. And of course, once you've been diagnosed HIV+ testing stops dead. Because "you've got it for life", so there's theoretically no need to ever test again. But we know that it is quite a common occurrence for people to be diagnosed HIV+ at one time and some time later, even after having had more positive diagnoses in the meantime, to start testing HIV negative (listen to this podcast for one example). People have reported to HEAL London a number of occasions when this has happened to them or people they know.

"All  you'll feel is a little prick..."

Another insidious factor that edges gay men closer to the arbitrary threshold classified as HIV+ is vaccinations, and particularly Hepatitis B, which has been documented multiple times in medical literature to trigger false positives. at the time of writing I have personally known five people who were diagnosed HIV+ not long after their Hepatitis B vaccinations, including a previous partner at the time we were in a relationship. The time you will be most likely to be persuaded to get vaccinated is when you go to the clinic for your regular HIV test, of course. When they do their battery of blood tests for gay men it will often include Hep B if you haven't already been vaccinated. If there are no antibodies, they will ask if you'd like some.

What about the 'heterosexual' statistics?

So that covers some of the issues that may contribute to gay men being diagnosed HIV+ disproportionately to straight people, but what about the claims that HIV is spreading amongst the heterosexual population? Let's consider 2002, a year in which I've already analysed the data once. That was when it was considered  there was a big rise in the heterosexual epidemic.

Now it seems strange that it would take about 19 years to then suddenly spread to the heterosexual community, as if there is some firewall between gay and straight that had only just burnt down, but let's leave aside that epidemiological quirk for a moment. This claim was actually derived from a big increase in the number of women being diagnosed HIV+. Ok, so it's women as opposed to heterosexuals per se, ie, it wasn't including heterosexual men, on the assumption that if women are now being diagnosed HIV+ more frequently then it must have spread further into the heterosexual community. Let's accept that for a moment also. So IF HIV tests were valid then a significant increase in the proportion of HIV tests of women being positive would indicate a spread in the epidemic.

It turns out that the rise in women being diagnosed HIV+ was specifically pregnant women - ok, fair enough, although we'll keep one eye on the fact that pregnancy is already know to trigger false positives on HIV tests. So, what is the proportional increase? Mysteriously, that figure was not given, nor was the raw data from which to calculate the proportional increase. After all, it is only a proportional increase in the number of pregnant women's HIV tests being diagnosed HIV+ that is meaningful.

Dodgy data

But wait a minute - on another page on the website at the time there was the proud announcement that there was a big success in increasing the number of pregnant women being given HIV tests. So, there was a big increase in the number of pregnant women being given HIV tests and a big increase in the number of pregnant women being diagnosed HIV+, but - very mysteriously - there is no claim or data available to suggest or prove either way that the proportion of pregnant women's HIV tests turning out positive had increased at all.

One thing we can be sure of is that if there really was an increase in the proportion of women's HIV tests coming up HIV+, that evidence would have been easily accessible and probably at the forefront of the claim. This suggests that in reality, there was no increase in the proportion of pregnant women being diagnosed HIV+, or else it decreased. It would not be the first time that AIDS statistics have been deceptively presented: For example, there was a case in the US where the total number of men being freshly diagnosed had gone down, and so had the number of women being freshly diagnosed - although, the number of women being freshly diagnosed had not gone down quite fast proportionally as the number of men.

The headline from these results was "Proportion of women being diagnosed HIV+ Increasing" - technically true, but pretty deceptive because the missing information would lead most people to assume that the rate at which women were being diagnosed HIV+ was increasing faster than men. In other words, they (and the accompanying text) managed to make a drop in the number of both men and women being diagnosed HIV+ sound as though it was a galloping epidemic spreading further into previously unpenetrated sections of the population. (I'm sorry, you'll have to excuse that pun, it just sort of...slipped in).

Racial overtones

But that's not all, because if we look even closer at the statistics from 2002 we find 71% of the women of this successful increase in testing pregnant women in the UK, were classified as 'sub-Saharan'.  The document containing the detailed statistics at the link I stored is no longer available, so I can't double-check the criteria for inclusion in this category (I'm taking these figures from a book I got part way through writing a few years ago from when I'd originally gone through that document). So, does 'sub-Saharan' mean literally women who have come from sub-Saharan Africa, or does it mean 'black'?

Let's consider both possibilities: If it means literally women who have come from sub-Saharan Africa, then this is a tiny proportion of the total women in the UK. If this tiny proportion has such a high proportion of the HIV diagnoses, then if HIV tests were reliable and HIV does cause AIDS, sub-Saharan women who've made it to Britain must be surely dropping like flies monumentally out of all proportion to their numbers. I do not recall coming across any such evidence, anecdotal or otherwise.

It also fails to take into account any of the factors that we already know can trigger false positives that are more widely prevalent in Africa, such as Malaria, TB, leprosy, bacterial infections, living in close proximity with cattle and drinking unpasteurised milk, and so-on (some tribes drink a mixture of blood and milk taken straight from cow and mixed and drunk / eaten while still warm. What proportion of them are likely to be diagnosed HIV+?). But in any case, if this category literally only includes women who have come from sub-Saharan Africa, then it is not indicative of a spread in the UK indigenous population either, and it would be somewhat deceptive to present it in such as way as to mislead people into thinking that is what it represented.

What if 'sub-Saharan' is effectively a code word for 'black', which would also include British-born, Caribbean-born and European-born black people living here? If I recall correctly, black people are still only around 6% of the UK population despite being more highly represented in London and other major cities. So, by my calculations, if women in this 6% of the population were actually representing 71% of the new HIV+ diagnoses, black women would have be diagnosed HIV+ at a rate of 38 times as frequently as white women, pro-rata. This is a truly massive difference, and surely would indicate a hugely visible pandemic of AIDS amongst black people in the UK that is mysteriously absent.

If those HIV+ diagnoses were all the result of promiscuous sex and drug-taking, we are implying that black people have massively and widely more reckless and irresponsible behaviour than white people. Does evidence really exist of such a dramatic difference in behaviour? From the significant number of black people I know or have known I find that hard to believe, and essentially it implies the same kind of racist stereotypes from the Victorian era would have to be revived to explain the disparity.

But...does sex even make any difference?

Wait a minute - even if black people were having wildly more unprotected sex than white people - which, also knowing white people too, I also find hard to believe - then surely studies would show that behaviour change would result in a noticeable difference in fresh HIV+ diagnoses? That doesn't appear to be the case either. For example, there's the study of 'unsafe' sexual behaviour in Trinidad (where we can hazard a guess that a significant proportion of the people involved in the study were black) between sero-discordant people (one person HIV negative, one person diagnosed HIV+). They never mentioned in the study that none of the HIV negative people were diagnosed HIV+ after unsafe sex in relationships with people diagnosed HIV+, I only found out because a colleague of mine was talking to a doctor who had been close to the research - but it essentially indicates that 'unsafe' sex between black people where one person is diagnosed HIV+ and the other negative, was not shown to cause any sexual transmission at all.

Additionally, there was a study a few years ago in Uganda (where we can hazard a guess that a significant proportion of those in the study were black) where people were encouraged to have fewer sexual partners and use condoms more in the hope of reducing HIV+ diagnoses. The result was that for other sexually transmitted infections there was indeed a noticeable drop, but there was absolutely no effect on new HIV+ diagnoses - the lead researcher said there was "not even a hint of an effect". When taking into account that there was a significant difference in the incidences of other sexually transmitted infections indicating that behaviour change actually did happen, this substantially undermines the notion that that massively higher rate of HIV+ diagnoses in 'sub-Saharans' in the UK could be due to rampant promiscuous unsafe sex.

This is not even taking into account the Padian study of sero-discordant heterosexual couples (of which 11% of both sexes were classified as African-American). They followed couples for a total of 282 couple-years, of which only 32% of the couples were consistently using condoms at the start of the study, and yet there was not even a single apparent transmission of HIV to someone previously HIV negative - indicating that condoms could not be shown to have made a difference because there were no transmissions despite the lack of condom use. So, if changes in levels of condom use and number of partners made no difference, and unprotected sex between sero-discordant couples resulted in no transmissions of HIV either, what could the explanation be for such a large disparity between the HIV+ diagnosis rates of white pregnant women and black pregnant women?

It appears black people in Britain are proportionally more black than black people in America (where the 'one-drop' rule still seems to apply to a large extent in self-identification and classification - take Colwyn Powell for instance) and a significant proportion of people classified as black have a significant amount of european genes. Given that in the US black pregnant women are 12 times as likely to be diagnosed HIV+ than white pregnant women, pro-rata, this is consistent with the notion that HIV tests are heavily biased against African genes, which we would expect: HIV tests are non-specific and Black people have a higher antibody level generally and typically produce a stronger antibody response to antigens, that some research indicates offers better protection against parasitic infections. It seems the higher the proportion of African genes you have, the more likely you are to be diagnosed HIV+, especially when pregnant. This is also consistent with people's experiences in Africa where white Africans seem to be hardly ever diagnosed HIV+, by comparison with the proportion of black Africans diagnosed HIV+.

"Twisting, twisting the night away..."

So, in the final analysis, taking just one year as an example, instead of evidence of a spread of HIV among the heterosexual population, closer examination of the figures actually provides more evidence that there is NOT a general spread into the heterosexual population. Instead, it provides more epidemiological support for the notion that HIV tests are heavily biased against black people. Yes, there are non-black, non-pregnant heterosexual women who don't use drugs who are nonetheless still being diagnosed HIV+. But from examination of the figures in 2002 at least, it's not the 'spreading heterosexual epidemic' that is suggested when statistics are presented in the media the way they are. In other words, there is actually no firm evidence it is spreading, it certainly isn't an epidemic, and the evidence is not consistent with the idea of it even being sexually transmitted.

So, there are lies, damn lies, and AIDS statistics. As always with AIDS, the devil is truly in the detail.