Certain levels of financial incentive may be effective in increasing adult men’s uptake of medical circumcision, according to a randomized trial conducted in western Kenya.¹ Among the 1,502 uncircumcised participants aged 25–49 randomly selected to receive food vouchers of varying amounts (or no compensation) if circumcised within two months, uptake was higher among those who received vouchers worth US$15 or US$8.75 (9% and 7%, respectively) than among those who received vouchers worth US$2.50 or no compensation (2% each). In logistic regression analyses, men in the two higher compensation groups were more likely than those in the no compensation group to get circumcised (odds ratios 6.2 and 4.3, respectively); no difference was found between men in the US$15 group and those in the US$8.75 group.

Male circumcision has been shown to reduce men’s risk of HIV; however, the prevalence of circumcision generally remains low among adult African men, who commonly cite loss of wages during and after the procedure as a barrier to uptake. To investigate whether small incentives could partially offset this barrier and increase circumcision among adult men, researchers conducted a study between June 2013 and February 2014 in three districts of Kenya’s Nyanza region, where HIV prevalence is 15%. They used a two-stage sampling technique to randomly select a sample of uncircumcised men aged 25–49. Consenting participants were interviewed and given information about medical male circumcision; they were then randomly assigned to one of four study groups and referred to area clinics for a free circumcision procedure. Men assigned to the three intervention groups were told that they would receive about US$2.50, US$8.75 or US$15 worth of food vouchers (amounts chosen to approximate transportation costs and 0–3 days’ worth of lost wages) if they underwent circumcision within two months of study enrollment; men in the control group would not be compensated if they underwent circumcision.

The analytic sample consisted of 1,502 men, approximately 375 in each of the four study groups. Descriptive and logistic regression analyses were conducted to examine circumcision uptake by study group. Additional subgroup analyses examined uptake by individual characteristics among men who received high compensation (US$8.75–$15) or low compensation (US$0–2.50).

Virtually all men in the sample were from the Luo ethnic group, and 84% were married; the men’s mean age was 34. On average, participants worked 47 hours during the week prior to the interview, earned the equivalent of US$5 a day and lived 6 km from the nearest clinic. Ninety-two percent of men reported having a primary sex partner, and 15% had had a sex partner other than a primary partner in the past year. When asked about their likelihood of getting circumcised in the future, 25% said "definitely yes," 61% "maybe," 7% "unlikely" and 8% "definitely not." In general, men’s characteristics were similar across the four study groups.

The two-month circumcision uptake rate was 2% for men in the US$2.50 group, 7% for those in the US$8.75 group and 9% for those in the US$15 group; the rate for the control group was 2%. In logistic regression analyses, men in the US$15 and US$8.75 groups had greater odds than those in the control group of having undergone circumcision within two months of enrollment (odds ratios, 6.2 and 4.3, respectively); the likelihood of circumcision uptake did not differ between the two higher compensation groups.

In all subgroup analyses by individual characteristics, men offered high compensation (US$15 or US$8.75) were more likely than those offered low compensation (US$2.50 or US$0) to get circumcised. For example, men older than 33 who were offered high compensation had 4.5 times the odds of their peers offered low compensation of having undergone circumcision; among men who reported having had a sex partner other than a primary partner in the past year, those who were offered high compensation had 11.8 times the odds of their peers offered low compensation of having undergone circumcision.

The authors mention several limitations of their study, such as the intervention’s potential lack of generalizability and the small sample sizes and limited statistical power in the subgroup analyses. Despite these limitations, they suggest that their study "adds to an as yet small evidence base on effective strategies to create demand for male circumcision," and conclude that "the effects of more intense promotion or longer implementation require further investigation."—J. Rosenberg

REFERENCE

1. Thirumurthy H et al., Effect of providing conditional economic compensation on uptake of voluntary medical male circumcision in Kenya: a randomized clinical trial, Journal of the American Medical Association, 2014, 313(7):703–711.