Serosorting raises the risk of STIs and HIV for German gay men

A survey of German gay men has found that ‘serosorting’ – restricting unprotected sex to partners of the same HIV status – does not work as a safer-sex strategy. The survey found that serosorting in HIV positive men increased the risk of having a bacterial sexually transmitted infection (STI) like syphilis or gonorrhoea more than fivefold.

Serosorting was also associated with a five-times-greater risk of a recent HIV diagnosis than using condoms and/or monogamy as a strategy, and was even more risky than having no strategy. Here, though, the researchers were unable to determine if serosorting was the cause of the HIV-positive diagnosis or the result of it (i.e. newly positive men seeking out positive partners).

Serosorting did not raise the risk of STIs significantly in HIV-negative men, but then exclusive serosorting – having unprotected sex, but only with men known or assumed to also be negative – was a strategy only adopted by a small proportion (3%) of those who were HIV-negative.

The survey was conducted via gay magazines and the internet during 2007 by the Social Science Research Centre of Berlin and 8170 questionnaires were analysed. The findings were presented as a poster at the Sixteenth Conference on Retroviruses and Opportunistic Infections in Montreal last week.

Knowing or assuming partners’ status

Thirty-six per cent of the HIV-positive men and 41% of the HIV-negative men said they didn’t try and find out or guess their partner’s HIV status. The majority of these men always used condoms and said their partner’s status was irrelevant; they were successful in using condoms 95% of the time. But 9% of the positive men and 2% of the HIV-negative men said they never used condoms and didn’t ask their partners’ status.

This left 48% of the positive men and 44% of the negative men who said they did try and find out, or at least guess, their partner’s status. HIV-positive men who did this assumed their partner was negative 60% of the time and positive 40% of the time. Negative men only assumed their partner was positive 4.5% of the time (probably an underestimate: HIV prevalence in gay men in Germany was nearly 11% in 2008, according to the 2008 UNAIDS report on the global epidemic).

The questionnaire then unpicked these declarations to find out if the men who made assumptions about status used it to influence condom use the last time they had had sex. Here the researchers found that, “the general intention to have unprotected anal intercourse only with seroconcordant partners is not transferred into general sexual practice.”

With HIV-positive men, nonetheless, partners’ status made some difference. Two-thirds of HIV-positive men used condoms last time they had sex when they assumed their partner was negative, but only 28% when they assumed they were positive. With HIV-negative men it did not make much difference: 61% had used condoms at last sex when they assumed their partner was negative and 68% when they assumed they were positive (but remember that negative men rarely assumed their partners were positive).

How did they know?

So how did men assume that they ‘knew’ their partners’ status? Amongst the HIV-positive men, direct disclosure by the partner or reading it in an internet profile accounted for two-thirds of this knowledge when they assumed the partner was positive and 56% when they assumed they were negative, knowledge that could be pretty well relied on.

However a quarter of the time positive men’s assumption that their partner was also positive was based on the fact that they didn’t want to use condoms. When they assumed their partner was negative, a third based this on their partner’s appearance, or on verbal hints.

As for the HIV-negative men, on the relatively few occasions when they ‘knew’ their partner was positive this was usually due to direct disclosure: more than three-quarters of negative men who’d had a partner they assumed was positive made that assumption on the basis of disclosure in person or online, although 15% based it on appearance and 8% on the fact that the partner did not want to use condoms.

Similarly, 73% of the time negative men ‘knew’ their partner was negative because they said so. Here, however, we must remember that knowledge of one’s status is dependent on time since the last test and behaviour since then and, as the researchers point out, fully a third of the men in the survey had never had an HIV test and 22% had a test result older than 18 months.

Serosorting and risk

By analysing the questionnaire answers, the researchers arrived at estimates of the proportions of men who used specific risk-management strategies.

They estimated that: about a third of positive men and 60% of negative men used condoms and/or monogamy; one-in-five positive men and only 3% of negative men used ‘pure’ serosorting, i.e. had unprotected sex but strictly reserved it for partners they perceived to have the same status; that one-in-five positive men and a quarter of negative men used a ‘bit of both’, meaning they used condoms sometimes but also serosorted on occasion; and that one-in-eight negative men and a quarter of positive men didn’t try to use any risk-reduction strategy.

How risky were these different approaches? As noted above, only the negative men who didn’t try to reduce risk had a significantly increased chance of having a bacterial sexually transmitted infection (STI) – 2.1 times that of those who relied on condom use or monogamy 100% of the time. In HIV-positive men, however, choice of strategy made a big difference to sexual health. Serosorters were 4.3 times more likely to have a bacterial STI than those who used condoms or monogamy and this risk was greater than the 3.7-fold risk of those who used no strategy. Positive men who used both strategies had 2.2 times the risk of acquring a bacterial STI when compared to 100% condom or monogamy users.

The researchers then looked at the men who had recently been diagnosed with HIV – meaning in the last 18 months. Approximately 2.2 to 2.4% of men who used the condom/monogamy or ‘mixed’ strategies had recently seroconverted, 7.8% of men who did not try to use a strategy, and 12.5% of men who exclusively serosorted, though this could reflect post-diagnosis rather than pre-diagnosis practice.

The researchers conclude from this that “serosorting among HIV-positive MSM is more likely to be effective, but profoundly increases incidence and prevalence of bacterial STIs”.

With regard to HIV-negative men, they say: “Among MSM who believe themselves to be negative, ‘serosorting’ is highly ineffective as a strategy, resulting in increasing, not decreasing, the risk of HIV transmission.”