HIV infection and risk behaviours in young gay and bisexual men

We have several concerns about the interpretation of the findings of a recent cohort study of sexual behaviour and HIV infection among young men who have sex with men in Vancouver.1 Of the 11 subjects who became seropositive, 3 reported having injected drugs and 1 having shared needles. The other 2 may also have shared needles; this practice is often under-reported because of its social undesirability and poor recall related to the effect of the drugs. Thus, the 3 subjects may have been infected through injection rather than through sex with other men. In fact, injection was significantly associated with HIV infection (p < 0.001) whereas sharing needles was not (p = 0.06), and HIV incidence among injection drug users during this period was extremely high (18.6 per 100 person-years2).

We also question the inclusion of the man who had an indeterminate result at baseline in the seroconverter group; subjects in a cohort study should be susceptible at study entry. Excluding the 3 subjects who injected drugs and the seroconverter yields an HIV incidence of 1.1-1.3 per 100 person-years (depending on whether the seroconverter also injected drugs). This is similar to the HIV incidence of 1.05 per 100 person-years we observed in men under 30 years old who have sex with men in Montreal from 1996 to 1999 (unpublished data). We believe HIV incidence among men who have sex with men should be calculated excluding those with other risk factors or, alternatively, calculations should be made separately for subjects with and without other risk factors.

Finally, the authors concluded that levels of unsafe sex increased over time on the basis of the proportion of subjects reporting safe sex at baseline who reported unsafe sex at follow-up. In Montreal we found that sexual behaviour is dynamic; a large proportion (51%) of those who practised unsafe sex at baseline practised only safe sex at follow-up,3 which resulted in similar proportions of subjects reporting unsafe sex at baseline and follow-up despite the fact that about 10% of those reporting safe sex at baseline reported unsafe sex at follow-up. Therefore, risky sexual behaviour among both those with safe and those with unsafe sexual practices at baseline must be examined at follow-up.