Condom avoidance and determinants of demand for male circumcision in Johannesburg, South Africa.

BACKGROUND Circumcision is efficacious in reducing HIV acquisition in heterosexual males. The South Africa government has been reluctant to adopt a national circumcision programme, possibly due to concerns that circumcision may result in decreased condom use. OBJECTIVE To identify the determinants of demand for male circumcision, to examine variations by ethnicity, and to determine whether it is demanded to avoid condom use. METHODS 403 parents and 237 sons in Johannesburg, South Africa, were recruited through a randomized household survey, with oversampling to balance between blacks (n = 220), 'coloured' (mixed ethnicity) (n = 202) and whites (n = 218). The demand for male circumcision was estimated using a conjoint analysis, with each respondent randomly receiving four tasks comparing seven possible benefits-six identified through key informant interviews and one for condom avoidance. Respondents' choices were analysed using logistic regression, including stratified analyses to test for homogeneity. RESULTS Overall, circumcision's beneficial effects on HIV transmission (P < 0.001), sexually transmitted infection (STI) transmission (P < 0.001), hygiene (P < 0.05) and sex (P < 0.05) were identified as determinants of demand, but the condom avoidance hypothesis was rejected as it was 'repulsive' to respondents (P < 0.001). Consistent results were found for blacks (P < 0.001) and coloured (P < 0.001), but not for whites who found condom avoidance attractive (P < 0.04), a result not explained by variations in wealth, age or paternal circumcision status. CONCLUSIONS Male circumcision programmes should be tailored to accommodate variations in the determinants of demand across the target population. We find that circumcision's protective effect against HIV acquisition in men is the only determinant to be found consistently across all ethnic groups in Johannesburg. We also find that concerns over condom avoidance may have been overstated. This said, male circumcision strategies should reinforce a range of HIV prevention strategies, including condom use, as we find evidence that whites may view circumcision as a means to avoid condom use.

Utilization, outcomes, and retransplantation of liver allografts from donation after cardiac death: implications for further expansion of the deceased-donor pool.

OBJECTIVE: Utilization, outcomes, and retransplantation (ReTx) of liver allografts obtained by donation after cardiac death (DCD) are examined to identify mechanisms to optimize donation. SUMMARY AND BACKGROUND DATA: DCD for liver transplantation (LTX) has immediate potential to expand the donor pool but application is limited. METHODS: Retrospective analysis of the Scientific Registry of Transplant Recipients (SRTR) from January 2002 to April 2007 identified 855 DCD and 21,089 donation after brain death (DBD) adult, initial, whole-organ, liver-only LTX. Donor, recipient, and transplant characteristics were compared. Outcome measures were listing for ReTx within 1 year and graft survival determined as death or ReTx. RESULTS: DCD donors were younger (P < 0.001), with fewer African American and non-white race (P < 0.001), and fewer deaths secondary to stroke (P < 0.001). DCD recipients were older (P < 0.001), with lower Model for End-Stage Liver Disease (MELD) scores (P < 0.001), and less likely in intensive care (P = 0.02) or high-urgency status (P < 0.001). DCD allografts were more frequently imported from another allocation region (12% vs. 7%; P < 0.001). Cox regression analysis of time to DCD graft failure demonstrates higher DCD graft failure within the first 180 days (20.5% DCD vs. 11.5% DBD; P < 0.001) with convergence thereafter. DCD listing for ReTx and graft failure progressed continuously over 180 days versus 20 days in DBD. At ReTx, DCD recipients waited longer and received higher risk allografts (P = 0.039) more often from another region. More DCD recipients remain waiting for ReTx with fewer removed for death, clinical deterioration, or improvement. CONCLUSIONS: DCD utilization is impeded by early outcomes and a temporally different failure pattern that limits access to ReTx. Allocation policy that recognizes these limitations and increases access to ReTx is necessary for expansion of this donor population.