RESUMO
OBJECTIVES: This study uses observational data collected as part of Zambia's male circumcision program to measure sexual behavior in the postsurgical healing period and to model its influence on HIV transmission for men and their partners. DESIGN/METHODS: From August to October 2010, 248 men about to undergo male circumcision were recruited to participate. Interviews were conducted at baseline and 6 weeks after circumcision, the recommended period of sexual abstinence for complete wound healing. Logistic and Poisson regression models were used to assess risk factors for early resumption of sex. A static-network deterministic transmission model was constructed to assess the impact of the program given early resumption of sex. RESULTS: Twenty-four percent of circumcised men reported resuming sex prior to 6 weeks after surgery. Of men having sex, 46% had sex in the first 3 weeks, 82% reported at least one unprotected sex act, and 37% reported sex with two or more partners. The model estimates that of the 61,000 men circumcised in 1 year, early resumption of sex leads to 69 extra HIV infections (32 among men, 37 among women), but it estimates a net effect of 230 fewer HIV infections in 1 year, predominantly among men. CONCLUSION: Prevalence of risky sexual behavior during wound healing is high and more likely among those reporting risky sexual behaviors at baseline. Nonetheless, the net effect of the male circumcision intervention is beneficial. The impact on women, however, is very sensitive to the prevalence of early resumption of sex.