|
13th International AIDS ConferenceDurban, South Africa - July 9-July 14, 2000 |
Int Conf AIDS 2000 Jul 9-14; 13:(abstract no. MoOrC194)
Serwadda D, Gray R, Kiwanuka N, Sewnkambo N, Kelly R, Wawer M
D. Serwadda, Makerere University, Institute of Public Health, Faculty of Medicine, Kampala, Uganda, Tel.: +256 415 300 09, Fax: +256 415 450 01, E-mail: dserwada@imul.com
BACKGROUND: Male circumcision has been proposed for HIV prevention. We estimate the potential impact of circumcision on male HIV acquisition.
METHODS: A cohort of 5,507 HIV- men were followed for 10,321 person years (py). HIV incidence per 100 py was determined in relation to circumcision status, sociodemographic and behavioral characteristics, to estimate the potential impact of a circumcision program for prevention of male HIV acquisition, after adjustment for confounding by religion and behaviors.
RESULTS: Circumcision prevalence was16.4% overall, 99% in Muslims (mainly pre-pubertal) and 3.7% in non-Muslims (pre- and post-pubertal). HIV incidence was 1.1/100 py in the circumcised and 1.8/100 py in the uncircumcised (adjusted RR = 0.53). Among non-Muslims, incidence was 1.6/100 py in circumcised, and 1.8/100py in uncircumcised men (adjusted RR = 0.80). Thus, subtle differences in risk behaviors between Muslim and non-Muslim men may confound the association between circumcision and HIV risk. Assuming that circumcision of non-Muslim men could reduce HIV acquisition to the same level as in circumcised Muslim men, then a universal program of prophylactic circumcision could, theoretically, prevent 44.8% of HIV acquisition. Assuming 50% program coverage, the reduction in HIV acquisition would be ~ 22%. However, if a circumcision program reduced HIV acquisition to the level empirically observed among circumcised non-Muslims in this study, the intervention might reduce HIV acquisition by only 11.6%.
CONCLUSION: The utility of pre- or post-pubertal circumcision for HIV prevention is difficult to predict on the basis of observational studies due to confounding by differential characteristics and behaviors. Randomized trials may be needed to assess the efficacy of prophylactic circumcision.
Copyright © 2000 - International AIDS Society (IAS). Reproduction of this abstract (other than one copy for personal reference) must be cleared through the IAS.
