Using discrete choice experiments to inform the design of complex interventions

Table 4 Overview of the routine VMMC service (control arm) and the complex intervention arm as characterized and influenced by the DCE findings, by DCE attribute

Attribute^a	Key DCE finding	Control	VMMC service intervention (approach to address issue)	Community complement
Out-of-hours service	Not significant	No	Upon request
Service separation by age	Age separated preferred	Joint waiting areas	Separate waiting areas
Partner counselling	Available preferred	Not available	Available	Female-friendly information booths placed in community
HIV testing	Available preferred	Opt-out	Opt-out (staff re-training to emphasize HIV testing not required)	Reiteration of HIV testing not required for VMMC in messaging
Service provider genders	Male providers preferred	Both male and female	Both male and female (staff re-training for client friendliness)	Media emphasised staff professionalism

DCE discrete choice experiment, VMMC voluntary medical male circumcision
^aIncentives were not considered as part of this intervention and were purely included in the DCE for exploratory analysis

ISSN: 1745-6215