Men who have sex with men may now be the highest-risk group for HIV in Africa, IAVI study suggests

This article is more than 8 years old. Click here for more recent articles on this topic

Men who have sex with men may now be at considerably higher risk of acquiring HIV than other at-risk groups such as female sex workers or young people of either sex, if findings by the International AIDS Vaccine Initiative (IAVI) of HIV incidence at two centres in Kenya can be generalised to other populations.

The study, which compared the Kenyan populations with a largely heterosexual group from South Africa, also found lower-than-expected HIV incidence amongst female sex workers and their clients. The researchers also found that recruiting men who have sex with men (MSM) into the study was easier than expected, but note that there was a particularly high drop-out rate in MSM.

They comment that, while MSM “need urgent risk reduction interventions, and may be a suitable cohort for future HIV prevention studies”, because African MSM face considerable legal and social hurdles in coming forward, “careful consideration of the counselling and clinical needs, follow-up schedule and social support is vital to ensure continuing research participation.”

The study

The aim of the study was to collect data on HIV and sexually transmitted infection (STI) incidence and risk factors in three populations in Kilifi, a district north of Mombasa, and the Kangemi district of Nairobi, both in Kenya, and from Gugulethu township in Cape Town in South Africa, the better to target HIV vaccine trials.



Insertive anal intercourse refers to the act of penetration during anal intercourse. The insertive partner is the ‘top’. 

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.

not significant

Usually means ‘not statistically significant’, meaning that the observed difference between two or more figures could have arisen by chance. 


Receptive anal intercourse refers to the act of being penetrated during anal intercourse. The receptive partner is the ‘bottom’.


Expresses the risk that, during one very short moment in time, a person will experience an event, given that they have not already done so.

The researchers recruited 716 people in Mombasa, 653 in Nairobi and 465 in Cape Town. The researchers primarily used participants to recruit their peers in South Africa, where background HIV prevalence (at 28%) is ten times higher than in Kenya. In Kenya they recruited attendees at HIV testing centres, via outreach work in bars and brothels, and via ‘snowball’ sampling (asking members of a particular group to recruit others from the same group). The original idea had been to collect data on high-risk heterosexuals including sex workers but, as the researchers comment, “it quickly became apparent that MSM were willing to come forward and participate in HIV prevention research”.

Somewhat different monitoring and follow-up criteria were used in the three centres. In Cape Town, participants were monitored monthly and followed up for one year, while in the two Kenyan cohorts participants were monitored quarterly for two to four years. In Mombasa, participants were examined for STIs at every visit but in Nairobi and Cape Town only examined if they had symptoms. As a result, annual STI incidence was much higher in Mombasa (23%) than in the other two centres (3.7 and 4.4%).

The average age of participants was mid-20s (slightly older in Nairobi); nearly 70% were women in Cape Town, 50% in Nairobi and 36% in Mombasa. Participants in Cape Town were almost entirely heterosexual men and women and were not sex workers.

In Mombasa, 56% of men (36% of the study population) were MSM; 63% of men said they had sold sex (mainly to other men) and 54% had bought it. Three-quarters of female participants said they were sex workers, while one-in-20 women said they had bought sex.

In Nairobi, nearly all women defined as a sex worker and 85% of the men had bought sex; 22.5% of the men had had sex with other men and 33% defined as a male sex worker.

There was a high drop-out rate in the study: 13% did not return after their enrolment visit; 37% altogether left the study prematurely. Annual attrition rates were 22% in Cape Town, 20% in Mombasa and 10% in Nairobi.

The results

HIV incidence was high in MSM in the Kenyan centres: annual incidence in MSM was 9.7% in Nairobi and 6.1% in Mombasa (there were only three individuals who said they were MSM in Cape Town, and none contracted HIV).

Annual HIV incidence in women was 3% in Cape Town, 2.7% in female sex workers and 2.3% in non-sex-workers in Mombasa, and only 0.4% – much lower than expected – in Nairobi. Annual HIV incidence in non-MSM men was 0.9% in Mombasa and zero in the other two centres.

In a multivariate analysis, predictors of HIV infection included:

  • No secondary education versus some: hazard ratio (HR): 3.34
  • Genital ulcers, yes versus no: HR 4.48
  • Paid for sex versus not: HR 0.17
  • Receptive-only anal sex versus no anal sex (in men and women): HR 8.19
  • Receptive and insertive anal sex versus none: HR 3.55
  • Insertive-only anal sex versus none: HR 0.88 (non-significant).

Thus, while receptive anal sex was very strongly associated with HIV infection, insertive anal sex was not. The finding that people who paid for sex were more than five times less likely to acquire HIV than people who did not was described as ‘unexpected’; the researchers suggest that people having paid-for sex may be more wary of HIV and STIs and more likely to use condoms.   

The fact that HIV incidence in female sex workers was far lower than expected, especially in Nairobi, is likely due to decreasing background HIV prevalence and possibly more use of antiretrovirals. Higher condom use is a less likely explanation, because annual pregnancy rates remained high: the annual pregnancy rate was 18% in women in Nairobi, 14% in Cape Town and 11% in Mombasa.

This is some of the first data on HIV incidence in MSM in Africa, a continent where, as the researchers say, “The focus of prevention trials in adult Africans has largely been on heterosexual transmission.” They add that a recent UNAIDS report highlights the deficiencies in addressing the needs of MSM and comment that it “reinforces the importance of closing this gap from both a human rights and public health perspective”.


Price MA et al. Identifying high-risk populations in Kenya and South Africa: HIV incidence in cohorts of men who report sex with men, sex workers, and youth. JAIDS 59(2):185-193, 2012. Abstract can be viewed here.