Background Sexual assault (SA) is commonly reported in cross-sectional studies of men who have sex with men (MSM) worldwide, but there are few community-based longitudinal data.

Methods Participants in the Health in Men (HIM) cohort study reported lifetime history of SA at baseline and annual experience of SA at each interview. Predictors and outcomes of lifetime and incident SA were assessed using logistic and Cox regression analyses, respectively.

Results At study entry 319 of 1427 participants (22.5%, 95% CI 20.4 to 24.8%) reported past SA. Independent demographic predictors of past SA were being on pension or benefits (p<0.001), Asian v Anglo ethnicity (OR=0.37, p=0.015) and Indigenous vs non-Indigenous ethnicity (OR 3.79, p=0.024). Those reporting past SA had more lifetime male sexual partners (p<0.001) and more often reported past gonorrhoea infection (p=0.010). Past SA was associated with cigarette smoking (p<0.001), marijuana use (p<0.001), amyl nitrate use (p=0.034) but not use of alcohol or other recreational drugs. During 7392.7 person years of follow-up 55 incident SA were reported (incidence 0.74 per 100PY, 95% CI 0.57 to 0.97). Incidence was significantly higher among those with lower incomes (p=0.015) and those who reported a past history of SA at baseline (HR 2.61, p=0.009). In the previous 6 months, those suffering incident SA had more male sexual partners (p=0.056), more unprotected receptive anal sex with casual partners (p=0.004), but were no more likely to acquire HIV (p=0.313). There was no association of incident SA with current sex work, smoking, alcohol or other recreational drug use. After adjustment for age both past and incident SA were predictors of current erectile dysfunction (p<0.001 & p=0.005, respectively), low libido (p=0.050 & p=0.028, respectively) and other problems affecting sexual satisfaction (both p<0.001). Lifetime and incident SA were both also associated with a range of adverse psychological correlates.

Conclusions As far as we are aware, these are the first prospective data to examine sexual violence among MSM. A number of demographic and behavioural correlates suggest socioeconomic disadvantage and potentially heightened STI/HIV risk in these MSM victims. There was a strong association of SA with adverse markers of psychological and psychosexual health. Routine enquiry regarding SA experience of MSM should occur in sexual health clinics due to its association with a broad range of adverse sexual health outcomes.