Background We evaluated the informed consent (IC) process for male circumcision (MC) as services were being scaled up in Swaziland and Zambia. A primary objective was to assess male clients' comprehension of key concepts in the informed consent process to improve, standardise and streamline consent procedures as service delivery was expanded.

Methods In a post-test study design, we asked adult (18 years and older) and adolescent (13–17 years) clients who had completed counselling, but had not yet undergone MC, to respond to a 10-question, interviewer-administered, true-false comprehension test. We also collected data on demographics, recruitment for MC, and type of counselling received (group, one-on-one, VCT), and administered literacy and numeracy tests. Participants represented a convenience sample recruited from selected sites in urban Lusaka, and urban and rural sites in Swaziland. We used T-tests to test the null hypothesis that 90% of clients could pass the test (could score ≥80%), χ² tests to compare scores between sub-groups (eg, adults vs adolescents), and multivariate logistic regression models to explore factors associated with passing the comprehension test.

Results Between November 2009 and August 2010, 228 MC clients in Lusaka (n=159 adults, 69 adolescents) and 953 MC clients in Swaziland (n=756 adults, 197 adolescents) participated in the comprehension assessment. Most clients (90%, Swaziland; 89%, Zambia) passed the comprehension test; however, the pass rate in Swaziland was significantly higher among adults than adolescents (96% and 85%, respectively; p<0.001). The question that posed the greatest difficulty to clients was about surgical risks; 67% of Zambian clients (71% adults; 56% adolescents) and 87% of Swazi clients (adults and adolescents) incorrectly responded that MC surgery has no risks. Factors significantly (p<0.05) associated with passing the test included: literacy (both countries); advanced education (Zambia); counselling language (Zambia); and age (Swaziland). In addition, not passing the test was significantly associated with hearing about MC from a friend (vs any other source) in Swaziland.

Conclusions Service providers must place greater emphasis on potential risks of MC surgery to ensure that clients are fully informed before they consent to the procedure. Counselling messages should be tailored for adolescent clients, who may have lower levels of education and less sexual experience than adults undergoing male circumcision.