Keith a lot of your 08:48 is good and fair enough.

But is unfair of you to keep quoting de Cock as saying "Behaviour is important..." without quoting the rest of that sentence, which continues

"...but it doesn't seem to explain [all] the differences between populations. Even if the total number of sexual partners [in sub-Saharan Africa] is no greater than in the UK, there seems to be a higher frequency of overlapping sexual partnerships creating sexual networks that, from an epidemiological point of view, are more efficient at spreading infection."

And I invite people to consider a reasonable contraction of that statement:

Behaviour is important...but doesn't explain the differences between populations...sexual networks...from an epidemiological point of view, are more efficient at spreading infection

Because that is my point. That the circumstances people faultlessly and blamelessly find themselves in, from an epidemological viewpoint, are more efficient at spreading infection, than just behaviour.

Behaviour is patently important - don't have sex and you certainly won't contract HIV unless you are staggeringly unlucky. But given that all groups share and practice the same sexual transmission vectors, the reasons for differing outcomes lie elsewhere

We can go further with the at-risk heterosexual groups. Let's leave the numbers for a moment, talk about common sense issues.

Chlamydia and other STI's spread by promiscuous sexual activity involving unprotected penetrative sex. They can also hit the unlucky person that has one encounter with a carrier.

The people most at risk of contracting STI's are the most sexually active people - typically adolescents and "young adults".

So that is consistent with, it explains, the prevalence (up to 10%) of some STI's amongst young women, for instance. They were either statistically unlucky on a one night stand in a sexual network of overlapping partners or they acuired it by being a regular member of such a network.

That is exactly the sexual network that propagates HIV. All it takes is for the virus to be introduced into the network. That the networks exist is proven by the prevalence of other STI's. All that's missing is a few HIV+ people in each town centre on a friday or saturday and we know the numbers of carriers are rising and that we are particularly singled out by UNAIDS for late-diagnosis of heterosexual carriers.

So it is clear that if we are to prevent pockets of HIV outbreaks in these at-risk heterosexual people - and quite a lot of them as well - then we have to be looking at a more concerted general education and outreach program, don't we? We need to persuade these at-risk young men and women to come and get tested, don't we? Or should we, as Ake says, bring in compulsory testing annually for them? As well as for gay men?

It is, of course, silly to send condoms and leaflets to Granny, there is no point blanket testing WI meetings, but all sexually active people outside of a monogamous relationship are at risk and there is no real differentiation between gay straight, black or white when it comes to prevention.

Incidentally, while the only safe sex is to abstain until married and then never "stray". the reality (and reality is all that matters) is that this does not happen universally. A couple, where one or both has a sexual history, could not declare themselves "safe" until 6 months of celibacy with an HIV test at each end and then a lifetime commitment to each other. Food for thought.