Sexual behaviors and other HIV risk factors in circumcised and uncircumcised men in Uganda

Objective: To study the differences in sexual practices, hygienic behaviors , and other HIV risk factors between circumcised and uncircumcised men. Design: A cross-sectional study of men >17 years of age selected by single stage cluster sampling in the Industrial Borough, Mbale, Uganda. Methods: Using a structured questionnaire, 188 circumcised and 177 uncircum cised consenting Ugandan men were interviewed in one of four native languag es during April and May, 1997. Results: Among non-Muslims, circumcised men had a higher risk profile than uncircumcised men in that they were more Likely to drink alcohol in conjunc tion with sex (odds ratio [OR], 1.86; 95% confidence interval [CI], 1.09-3. 16), to have sexual contacts with women on the first day of meeting (OR, 2. 37; 95% CI, 1.39-4.04), to have had sexual contacts in exchange for money o r gifts (OR, 2.08; 95% CI, 1.21-3.09), to have experienced episodes of pain on urination or to have experienced penile discharge (OR, 1.68; 95% CI, 1. 07-2.64), had an earlier age at sexual debut (15.7 versus 16.9 years), and had more extramarital sex partners in the last year (1.13 versus 0.62). Cir cumcised men also reported a preference for nonwet sex. Muslims generally h ad a lower risk profile than other circumcised men except they were less li kely to have ever used a condom (OR, 0.34; 95% CI, 0.15-0.78) or to have us ed a condom during the last sex encounter (OR, 0.37; 95% CI, 0.14-0.87). Conclusions: These results suggest that differences between circumcised and uncircumcised men in their sex practices and hygienic behaviors do not acc ount for the higher risk of HN infection found among uncircumcised men. Fur ther consideration should be given to male circumcision as a prevention str ategy in areas of high prevalence of HIV and other sexually transmitted dis eases. Studies of the feasibility and acceptability of male circumcision in traditionally noncircumcising societies are warranted.