A survey of gay men in the European cities of Barcelona, Bratislava, Bucharest, Ljubljana, Prague and Verona has found that in all cities but Barcelona the majority of gay men with HIV did not know they had it – despite a majority of them having taken an HIV test in the last year. Indeed, having taken an HIV test in the previous year was associated positively with undiagnosed HIV infection.
The survey was conducted in 2008-2009 but its results have only just been published in print. The situation may have changed considerably in the last six years and the study is best read as a snapshot of the epidemic in gay men and other men who have sex with men (MSM) in central and southern Europe at the time.
What it shows is that, at that time, while HIV prevalence was relatively lower in the central European cities (Bratislava, Bucharest, Ljubljana and Prague, respectively the capitals of Slovakia, Romania, Slovenia and the Czech Republic) than it was in the Spanish and Italian cities (Barcelona and Verona), the high levels of recently undiagnosed HIV suggested that the HIV epidemic in MSM in parts of central Europe was poised to expand considerably.
Later evidence from central Europe has found an increasing epidemic of HIV in gay men and a 2013 survey from Poland, for instance, found that new HIV infections in gay men there have been increasing exponentially in the last decade.
The survey
The SIALON Project was an EU-funded collaboration between Spain and the Veneto region of Italy. Between 2008 and 2009, it used what is called time-location sampling (TLS) to gather approximately 400 respondents from each of the six cities studied. This means that a list of venues where local gay men met was compiled and then researchers went there at random times to invite eligible participants to take part in the study. This is in order to try and get a representative sample of venue users. The list of venues, compiled in consultation with local gay organisations, included saunas, private parties, cruising areas and sex shops as well as bar and club venues.
Participants were given a questionnaire to complete on personal data, access to HIV prevention services, and sexual risk behaviour. At the same time, an oral HIV test was taken. Tests results were anonymised and could not be traced back to individuals: this same methodology had previously been used in a sequence of similar surveys in London, Brighton and Manchester – which also uncovered a high proportion of gay men unaware that they had HIV.
Participant characteristics
In the study, 2285 men both completed the questionnaire and took an HIV test. There were significant differences between the men in the different cities. The average age was 32 but was considerably lower in the central European cities, from 26 in Bucharest to 31 in Ljubljana, compared to Barcelona and Verona (38 and 36 respectively).
A majority were employed and the cities with the lowest employment rates had the highest proportion of students. A majority self-defined as gay but in Bucharest and Prague a higher proportion than elsewhere defined as bisexual or heterosexual – a third and a quarter of men in Bucharest and Prague respectively.
One interesting statistic, given the recent controversy stirred up in the UK election about HIV-positive immigrants, was whether a high proportion of men surveyed were native to the local country. The highest proportion who were migrants was in Barcelona (28%) while the lowest was in Ljubljaana (3%): the average was 12%.
The participants were asked whether there were positive or negative attitudes to homosexuality in their city and also whether they had access to HIV prevention programmes. In most cities, only a few men said they thought there were positive attitudes: in Bratislava only 6%, in Ljubljana 8% and in Verona 9%. Only in Barcelona did a high proportion of respondents consider that there were positive attitudes towards homosexuality in their city, and even then only 40% thought this.
Access to HIV prevention programmes only correlated moderately with social and political attitudes. While the lowest rate of access to HIV programmes and resources was in Bratislava (38%), the highest rates were not only in Barcelona (83.5%) but also Ljubljana (85%).
A majority of men had tested for HIV at some point and the proportion who had never done so – ranging from 12% in Barcelona to 38% in Bratislava – correlated very well with the availability of HIV prevention services. In Barcelona and Verona, a majority of men (56% and 53%) – had tested for HIV in the last year. In the central European cities it was lower – Bucharest 43%, Prague 40%, Ljubljana 38% and Bratislava 32%.
HIV infection and risk
Overall, 8% of the men in the survey had HIV. Local prevalence ranged widely, from 17% (one in six) in Barcelona to 2.6% (one in 40) in Prague.
Overall, a majority (60%) of the men who had HIV did not know it, ranging from 47% in Barcelona to 85% in Bucharest. Of those unaware of their HIV status, 85% had taken an HIV test at some point and 60% had tested in the last year. While it is understandable that those who were unaware of their HIV infection were more likely to have ever tested for HIV than to have never tested (because they presumably are at higher risk), it is perhaps more surprising that those who had HIV but were undiagnosed were more likely to have tested in the last year (60%) than those who did not have HIV (43%).
Men with HIV (diagnosed or undiagnosed) had the same number of steady partners in the last six months as HIV-negative men (2.6 on average). The overall rates of condomless anal sex (measured as sex without a condom on the last occasion) were actually higher in HIV-negative men (59%) than in undiagnosed HIV-positive (44%) or diagnosed (38%) men. Diagnosed HIV-positive men had a higher than average number of casual partners in the last six months (19 versus 14 for undiagnosed and 9 HIV-negative). However, they were no more likely to have had condomless sex with them.
HIV status was associated with the use of poppers (nitrites): 36% of undiagnosed HIV-positive men had used them in the last six months compared with 14% of HIV-negative men. It was also associated with using Viagra (11.5% versus 3%, with diagnosed HIV-positive men at 7%) and cocaine (17% in undiagnosed and 15.5% in diagnosed men with HIV compared with 5% in HIV-negative).
Undiagnosed HIV status was also associated with higher rates of syphilis co-infection (9%, versus 5% in diagnosed men and 2% in negative). DiagnosedHIV-positive status was strongly associated with anal warts (14.4% compared with only 1% in undiagnosed men), as was hepatitis B infection.
Having HIV, diagnosed or otherwise, was also associated significantly with being recruited for the survey in a ‘sex-focused venue’: 59% of diagnosed and 52% of undiagnosed HIV-positive men were recruited in this way compared with 28.5% of HIV-negative men. Lack of access to HIV testing programmes was associated with being HIV-negative (32% of HIV-negative men said they had no access to HIV prevention help compared with 12% of HIV-positive men, both diagnosed and undiagnosed).
Some of these associations were mediated by age: diagnosed HIV-positive men were on average 39 years old compared with 34 for undiagnosed and 32 for HIV negative. When age was controlled for, factors that remained independent predictors of undiagnosed HIV infection were sex-focused venue recruitment (associated with 2.5 times the risk of having undiagnosed HIV), syphilis (2.6 times), poppers during last sex (3.4 times), and having taken an HIV test in the last year (2.0 times).
Conclusions
While some factors may have changed since this survey was done, the most striking finding remains clear: an HIV-negative test result during the last year did nothing to reduce the likelihood that someone now had HIV: indeed, it was positively associated with having it. This, of course, does not mean that testing ‘caused’ HIV but that testing is associated with high-risk behaviour and does not induce behaviours that reduce that risk.
As the researchers comment: “These results…highlight the limit of strategies aimed at promoting HIV status disclosure as a means of reducing transmission risk.”
They comment that their findings reinforce the need to continue promoting condom use but, given that reported condom use was no lower with either steady or casual partners in undiagnosed HIV-positive men than HIV-negative ones, the study also seems to point to the limitations of condom use as the sole risk-reduction measure available to gay men. The fact that the strongest associations of positive status were with sexually transmitted infections, sex parties and drugs may point to the then-emerging phenomenon of 'chemsex’ as a contributory factor to HIV in gay men in southern and central Europe, though drugs specifically associated with this (methamphetamine, mephedrone etc.) were not asked about.
However, the high rates of undiagnosed HIV and the association, rather than lack of association with high rates of recent testing suggest that a ‘fast-transmission’, high-incidence epidemic of HIV was emerging in the gay communities in hitherto lower-prevalence central European countries around this time.
Ferrer L et al. Undiagnosed HIV infection in a population of MSM from six European cities: results from the SIALON Project. European Journal of Public Health 25(3):494-500. 2015. See abstract here.