Introduction: Male circumcision (MC) reduces the risk of HIV infection. However, the risk reduction effect of MC can be modified by type of circumcision (medical, traditional and religious) and sexual risk behaviours post-circumcision. Understanding the risk behaviours associated with HIV infection among circumcised men (regardless of form of circumcision) is critical to the design of comprehensive risk reduction interventions. This study assessed risk factors for HIV infection among men circumcised through various circumcision approaches.

Methods: This was a case-control study which enrolled 155 cases (HIV-infected) and 155 controls (HIV-uninfected), all of whom were men aged 18–35 years presenting at the AIDS Information Center for HIV testing and care. The outcome variable was HIV sero-status. Using SPSS version 17, multivariable logistic regression was performed to identify factors independently associated with HIV infection.

Results: Overall, 83.9% among cases and 56.8% among controls were traditionally circumcised; 7.7% of cases and 21.3% of controls were religiously circumcised while 8.4% of cases and 21.9% of controls were medically circumcised. A higher proportion of cases than controls reported resuming sexual intercourse before complete wound healing (36.9% vs. 14.1%; p<0.01). Risk factors for HIV infection prior to circumcision were:being in a polygamous marriage (AOR: 6.6, CI: 2.3–18.8) and belonging to the Bagisu ethnic group (AOR: 6.1, CI: 2.6–14.0). After circumcision, HIV infection was associated with: being circumcised at >18 years (AOR: 5.0, CI: 2.4–10.2); resuming sexual intercourse before wound healing (AOR: 3.4, CI: 1.6–7.3); inconsistent use of condoms (AOR: 2.7, CI: 1.5–5.1); and having sexual intercourse under the influence of peers (AOR: 2.9, CI: 1.5–5.5). Men who had religious circumcision were less likely to have HIV infection (AOR: 0.4, 95% CI: 0.2–0.9) than the traditionally circumcised but there was no statistically significant difference between those who were traditionally circumcised and those who were medically circumcised (AOR: 0.40, 95% CI: 0.1–1.1).

Conclusions: Being circumcised at adulthood, resumption of sexual intercourse before wound healing, inconsistent condom use and having sex under the influence of peers were significant risk factors for HIV infection. Risk reduction messages should address these risk factors, especially among traditionally circumcised men.