Treatment as Prevention: Treating People to Death

There was an attack on 'treatment as prevention' in March which came from a person you wouldn't expect to oppose a technological quick fix, Elizabeth Pisani. Despite the fact that she disagrees with UNAIDS in some ways, she is an adherent of the behavioral paradigm. It seems a pity to hold views that challenge the mainstream and yet still cling to the mainstream's central premise about HIV: that it is almost always transmitted through heterosexual sex in African countries.

But it's worth citing her opposition to a strategy which has a lot in common with PrEP. Firstly, Pisani points out that "HIV is most infectious in the few months after a person is first infected. Even if everyone got tested annually, we’d miss most of these new infections." I hope the 'modelling' work that is said to support treatment as prevention includes this point, but I doubt it.

Pisani also notes that there are a number of circumstances under which viral load (infectiousness) can spike, such as contracting another sexually transmitted infection (or perhaps other diseases) or failure to take medication correctly, which can occur for many reasons. Such a spike would increase infectiousness in people who may well be engaging in unprotected sex.

Pisani refers to findings relating to treatment becoming more widely available in rich countries. Apparently rates of unprotected sex increases as a result of 'disinhibition', engaging in unprotected sex in the belief that the risk is now low. Many have claimed that disinhibition does not happen to any great extent in African countries. The 'model' used by proponents of treatment as prevention believe that disinhibition will not significantly contribute to HIV transmission and that adherence to drug regimes will be extremely high in African countries.

Pisani casts doubt on both of these claims. I have to say, I agree. I would suggest that the finding that disinhibition is low in African countries is more likely to indicate that HIV is not as closely related to sexual behavior as we have been led to believe.

As for claims about high levels of adherence, I'm not sure if figures for treatment in countries like Kenya and Tanzania are very complete or credible. Death rates among HIV positive people seem to be high enough to keep prevalence steady and there is no evidence that sexual behavior has been influenced greatly by behavior change programs.

I'd say UNAIDS, and Pisani herself, are over-optimistic about a lot of things. Treatment as prevention sounds, on the surface, like a good idea. But it's not going to be enough on its own, especially if only sexually transmitted HIV is being targeted. Waiting till people become infected and then treating them, hoping that they will all become less infectious and therefore slowing down the epidemic, is ludicrous.

Even if HIV is 100% sexually transmitted this would not work. We must know by now how hard it is to influence people's sexual behavior or, indeed, any other kind of behavior. But HIV is also transmitted non-sexually. It is vital to establish the contribution of non-sexual HIV transmission to serious HIV epidemics, otherwise sexual transmission will continue to be overestimated. As long as we overestimate sexual transmission, HIV will continue to spread.