Are UNAIDS Really Giving Good Advice to Malawians?

This could be a good time for condom manufacturers to get involved in large scale HIV prevention campaigns in African countries. After all, it must represent a huge market. But the figures for HIV prevalence among condom users are not always very encouraging. In fact, reported condom use often seems to be associated with far prevalence than those who claim not to use them.

The latest Demographic and Health Survey for Malawi is a case in point. Among women who had ever used condoms, HIV prevalence was 20.3%, whereas among those who never used them it was only 13%. Among men who had ever used them, prevalence was 11.8% but among those who never did was only 5.8%. Those are substantial differences.

Similar trends are found among men who have paid for sex in the last 12 months. Among those who ever used condoms, prevalence is 11.5% but among those who never did it was 4.1%. Even those who did not pay for sex in the last 12 months had far higher prevalence than those who did pay for sex and did not use condoms (or did not have sex in the last 12 months), at 9.3%.

Why does condom use seem to be associated with far higher HIV transmission rates in many African countries? We know they work to reduce sexual transmission of HIV and other sexually transmitted diseases, so what could be going on in Malawi? It seems surprising that condom manufacturers are not taking a careful look at these figures.

The report also finds that: "Among respondents who have never been married, the HIV prevalence [...] 2 percent for those who have never had sex. This suggests that some women and men incorrectly reported that they were not sexually active, or that there is some degree of nonsexual HIV transmission occurring (e.g., through blood transfusions or non-sterile injections)."

Granted, some people forget, lie, or misreport, in Malawi and everywhere else. But even those who have reported high levels of 'unsafe' sexual behavior are just as likely to have been infected through some non-sexual mode of transmission as those who have never had sex. After all, sexual transmission is less efficient than many non-sexual modes.

HIV rates by circumcision status also appears to be very mixed. In almost every age group, HIV prevalence is higher among circumcised men and overall prevalence stands at 10.3% for circumcised men and only 7.6% for uncircumcised. Given that a lot of Malawian men are not circumcised, let's hope the circumcision enthusiasts figure out exactly what's going on here before rushing in and carrying out an operation that doesn't decrease transmission and may increase it. There are already claims that about 250,000 Kenyan men have been circumcised to 'reduce' HIV transmission.rends

There are also the trends that you will find in other high prevalence African countries. For example, prevalence is far higher among women, roughly a 60-40 split. When you consider that most of the high risk groups that don't involve sexual transmission, intravenous drug users, men who have sex with men and prison populations, are almost all men, it makes you wonder how women can face such high risks.

Prevalence is twice as high in urban areas and varies a lot between different regions and different tribes. Prevalence also varies considerably among religions, with Anglicans having the highest prevalence, Catholics the lowest and Muslims somewhere in between.

Higher or lower levels of education are not clearly associated with HIV prevalence, with almost the same rates among those with no education and those with higher than secondary level. Employment status is a lot clearer, with rates being far higher among employed people than unemployed people.

As is often the case in high prevalence countries, wealth appears to be highly positively correlated with HIV prevalence. This is especially the case for women, where prevalence among the lowest quintile is 8.9%, rising steadily to 19.7% among the highest quintile. For men, the figures are 5.6% rising to 10.8% in the highest wealth quintile.

I'm all for promoting interventions that work and condom use is one that certainly does work to reduce sexual transmission of HIV. But there are always figures in Demographic and Health Surveys that make one wonder what proportion of HIV really is transmitted sexually and what proportion is transmitted through other routes. Many of the figures for Malawi, though, seem to suggest that a lot of HIV industry advice is really not very effective.