Semeere, AS, Bbaale, DS, Castelnuovo, B Kiragga, A, Kigozi, J, Muganzi, A, Kambugu, A and Coutinho, AG, 2016. Assessing the impact of delivering messages through intimate partners to create demand for voluntary medical male circumcision in Uganda. 3ie Impact Evaluation Report 48. New Delhi: International Initiative for Impact Evaluation (3ie)
Link to Source
This study evaluates whether an information-based intervention among pregnant women in their third trimester attending antenatal care at three clinics in Kampala, Uganda would significantly increase VMMC demand among their male partners.
Key clinical-trials have demonstrated that circumcision is an effective and safe method of HIV prevention among men in Sub-Saharan Africa. Currently, only 27% of Ugandan men between the ages of 15 and 49 years are circumcised; it is a region with high HIV seroprevalence. In order to tackle the issue of low VMMC take-up, a pilot study was conducted with an intervention targeting male partners of pregnant women in the third trimester at three health centres in Kampala, Uganda as a means to increase VMMC uptake among men with the intimate female partners being the potential change agents.
In this evaluation, the primary research question is to assess whether empowering women, through an information-based intervention, to engage their male counterparts in a discussion on VMMC will lead to increased VMMC demand among their male partners.
The intervention was implemented at three municipal clinics in Kampala, Uganda offering free VMMC services. For the first three months, pregnant women who were 32 weeks or more pregnant were recruited at antenatal clinics into a control phase where no intervention (information or skills) were given to the women. They were informed of the availability of free VMMC services and those interested received transport refund vouchers to enable their male partners to access the free circumcision services. This was followed by a two-month washout period where no intervention activities were done to allow all the women contacted in the control phase to deliver. This was then followed by the three-month intervention phase where pregnant women were recruited into the study. In this phase, women were equipped with information and skills to pass on the information to their male partners as well as offered the transport refund vouchers.
Theory of change
The evaluation applied two theoretical models to inform the theory of change.
- If female partners of uncircumcised men were empowered to identify and address specific misperceptions of their partners and given an information leaflet to deliver to these partners to act as a cue-to-action, then this would lead to more instances of individualised VMMC-focused conversations.
- If increased conversations that identify misperceptions and provide answers and solutions to potential barriers related to VMMC occur between couples, it would lead to more male partners becoming more inclined to undertake VMMC, offer a cue to action to the men and in turn result in more male partners signing up for VMMC.
The evaluation was performed in a three-phase process: control phase (three months), washout (two months) and intervention phase (three months). This was done over a period of eight consecutive months. Using a natural experiment design, the study evaluated the intervention among pregnant women receiving antenatal services at three high volume clinics located in Kampala. The intervention used a before and after design natural experiment. Restriction and regression were used to try and preclude potential threats to validity of the estimates.
The intervention did not lead to a statistically significant increase in the demand over the project period. Though the intervention led to a 1.5 fold increase in the odds of a male partner presenting for VMMC, this was not statistically significant. Whereas the intervention did not significantly impact VMMC uptake, it increased the level of comfort of the female partner in communicating about VMMC with their male counterpart (0 per cent versus 35 per cent in the control and intervention populations respectively).
Cost efficiency analysis
The intervention was rolled out over a nine-month period, and on the basis the annual cost of the intervention, the total cost of implementing the intervention was determined as US$61, 009.7. Despite the non-significant impact of our intervention, it was calculated that for every US$5084.1 that was spent, one additional circumcision was achieved, on the basis of the US$61,009.7 annualised intervention cost and the projected 12 circumcisions.
Implications for further research
The findings suggest that the information-based partner-mediated intervention for generating demand for VMMC increased the level of comfort of female partners to engage their male counterparts in VMMC discussions, however, it did not increase VMMC uptake.
The intervention, while providing general information on circumcision and its benefits, did not directly provide incentives or a mechanism to alleviate this discomfort. This may have potentially contributed to the lack of impact of the intervention. Therefore, additional studies are needed to explore a modified intervention.