Mahler, H and Bazant, E, 2016. Using smartphone raffles to increase demand for voluntary medical male circumcision in Tanzania, 3ie Impact Evaluation Report 51. New Delhi: International Initiative for Impact Evaluation (3ie)
Link to Source
This study evaluates the effectiveness of a material incentive through a game of chance to increase voluntary medical male circumcision (VMMC) uptake among men aged 20 and above in Tanzania.
In 2010, the Ministry of Health and Social Welfare (MOHSW) of Tanzania set regional targets for VMMC services uptake in uptake in Iringa, Njombe, and Tabora regions. The efforts targeted men aged 20-34 years to reduce the incidence of HIV in Tanzania. Due to the lack of uptake within this age group, the use of financial or in-kind incentives in the form of lotteries or raffles was used as an innovative approach, which was well-suited to the promotion of VMMC services to men aged 20 years and older.
The primary outcome for this study was evaluate the effectiveness of a smartphone raffle incentive to increase uptake of VMMC services by men aged 20 years and above at fixed service delivery sites in Iringa, Njombe, and Tabora regions of Tanzania.
For this intervention, MOHSW and Jhpiego developed financial or in-kind incentives in the form of raffles to promote VMMC services at fixed delivery sites in Iringa, Njombe, and Tobora. During a three-month promotional period, men aged 20 and above who visited the fixed delivery sites were offered the opportunity to enter smartphone raffles.
Theory of change
The theory of change is largely based on the health belief model which suggests that an individual’s beliefs are shaped by their perceptions about both the benefits and barriers of the proposed action and that their ability to act is a function of self-efficacy. Using this model, Jhpiego proposed that a lottery affects these barriers by replacing a perceived barrier (loss of social status due to the stigma of VMMC) with a benefit (increase in social status due to owning an expensive smart phone).
The study used a matched-pair, cluster-randomised study design comprised of 14 health facilities (7 pairs) matched on facility type, region, and client volume.
In the quantitative data analysis of VMMC counts, Jhpiego generalised the linear modeling with Poisson distribution and exchangeable working correlation structure and robust variance estimates. They then compared it to the total VMMCs performed for each study group during the intervention period and in the same three-month period in the previous year, using a difference-in-difference approach. Similar analysis was carried out for the Iringa region.
The study produced mixed results that make the relative benefit of a smartphone raffle promotion to increase uptake of VMMC complicated to interpret. The intervention and control groups showed a 47 per cent and 8 per cent increase respectively for VMMC among men aged 20 and above during the study period as compared to the same period in the previous year. None of this changes were statistically significant in models accounting for clustering.
However, the findings in the Iringa region was different, when examining Iringa’s 3 matched pairs of facilities in models that account for clustering of data in facilities. The intervention group has a statistically significant increase (2.16-fold increase) between the intervention period and the same period the previous year, while the control group had a non-significant increase.
Cost efficiency analysis
Seventy-nine phones were raffled to clients, 12 to peer promoters and six to providers, making a total of 97 phones. Each phone cost $85.60 and raffle costs were $827.59 and the cost per additional client was $88.65. The 2012 study Cost and Impact of Scaling up VMMC in Tanzania notes that for every infection averted, there is a cost savings of US$11,300. Considering all facilities, there are 10.8 infections averted for a total savings of $122,416. This includes the decrease in the number of male circumcisions for Iringa, which represents an expense of $3,766.
Implications for policy and practice
Based on the overall study results, there is no clear mandate to recommend or not recommend the use of a smartphone raffle to increase uptake of VMMC services. However, placed in the right geographic location at the right time of their VMMC scale-up, implementing such a raffle may help late adaptors move intention to action.
The qualitative research suggested that the introduction of a financial or in-kind incentive to an already ‘free’ service may have the potential for negative effects. Formative research should inform the selection of appropriate incentives for attracting older men to VMMC and funding should be made available specifically for formative research into any demand-generation innovation for VMMC and any specific considerations for local contexts.