HIV infection and risk behaviours in young gay and bisexual men

[The authors respond:]

We recalculated the HIV infection rate based on all years and on whether a subject had reported that they had ever injected drugs. These new person-time estimates of HIV incidence are based on 18 subjects who became HIV positive after their baseline seronegative test, 8 more than in our published study1 and excluding the person with the baseline indeterminate result. The incidence rate has been revised to 1.2 per 100 person-years (95% CI 0.6-1.7) (Table 1) since the paper was written. Gay and bisexual men who injected drugs have a higher incidence rate (3.1 per 100 person-years [95% CI 0.6-6.1]) than those who did not (0.9 per 100 person-years [95% CI 0.4-1.4]). Robert Remis and colleagues are correct in assuming that our rate of HIV incidence among noninjecting participants is similar to that reported for men under 30 years in the Omega cohort (1.0 per 100 person-years from 1996 to 1999). However, the annualized incidence rates presented in Table 1 for noninjecting gay and bisexual men indicate that HIV infection appears to be increasing in this population.

Remis and colleagues felt that we should have reported follow-up risky sexual behaviour among participants who had safe as well as unsafe sexual practices at baseline. Of the 285 men with regular partners, 89 (31.2%) reported having unprotected insertive anal sex and 100 (35.0%) reported having unprotected receptive anal sex in the year before the baseline visit. At 1-year follow-up, 66 (74.1%) of the 89 subjects and 71 (71.0%) of the 100 subjects reported having unprotective insertive and receptive anal sex respectively. Of the 279 men with casual partners, 46 (16.5%) reported having unprotected insertive and 36 (12.4%) unprotected receptive anal sex in the year before the baseline visit. Of these men, 21 (46.6%) and 16 (44.4%) reported having had unprotected insertive and receptive anal intercourse respectively by the time of their first follow-up visit. In combining these data with other information presented in our paper,1 the odds of relapseamong men with regular partners increased 2-fold for both unprotected insertive intercourse (odds ratio 2.2, 95% CI 1.4-3.7) and receptive anal intercourse (odds ratio 1.9, 95% CI 1.4-3.0). Among men with casual partners, similar odds were observed for unprotected insertive intercourse (odds ratio 1.7, 95% CI 1.0-2.8), but the odds for receptive anal intercourse were not significantly increased (odds ratio 1.3, 95% CI 0.7-2.3). These new data along with the findings originally presented in our article confirm the high HIV rates and sexual risk behaviour in our cohort.