2017年-世界发展银行全球_Impact_Evaluation_on_Improving_Voluntary_Medical_Male_Circumcision_Demand_in_Malawi_through_the_Use_of_Incentives_178页_4mb

Summary of Impact Evaluation on Improving VMMC Demand in Malawi through the Use of Incentives

Core Content

This document presents the findings of an impact evaluation study on the effectiveness of incentives in improving the uptake of Voluntary Medical Male Circumcision (VMMC) in two districts of Malawi: Rumphi and Mchinji. The study was conducted by the National AIDS Commission (NAC) in collaboration with the World Bank and the Ministry of Health (MoH), and supported by the United Kingdom's Department for International Development (DFID). The evaluation aimed to assess how incentives, both collective and individual, influenced VMMC demand among boys and young men aged 10 to 34, as well as their families and communities.

Main Objectives and Research Questions

The main objective was to evaluate the effect of incentives on increasing VMMC uptake. The study addressed the following key research questions:

1. Primary Research Question:
   Do incentives significantly increase VMMC uptake among in-school and out-of-school boys and young men aged 10 to 34 in Malawi?

2. Secondary Research Questions:

   • Does the intervention have spill-over effects that increase VMMC uptake in the study participants' households (brothers and parents)?
   • Are attitudes towards VMMC among school girls changed by the intervention?
   • Are women effective in motivating young men to seek VMMC?
   • Do school heads recruit more males for VMMC than do the Mothers' Groups, or vice versa?
   • What is the role of young men's informal social networks (peers, family, and guardians) in increasing VMMC uptake?

Intervention Overview

The intervention included two types of incentives:

• Part A: Collective incentives such as whiteboards and football equipment, distributed monthly to institutions (schools and Mothers' Groups) that met certain VMMC uptake criteria.
• Part B: Individual vouchers that could be redeemed by the person getting circumcised and their caregiver during three key events at the district hospital: the procedure, first follow-up, and second follow-up.

The voucher booklets were also used to distribute to family members and friends, encouraging wider VMMC uptake.

Methodology and Data Collection

• Design: A cluster randomised control trial was planned, with schools and their associated Mothers' Groups as clusters. However, due to implementation challenges, the study design was modified.
• Data Analysis: Three analytical approaches were used:
  • Before-after analysis in the two study districts.
  • Comparison with 17 other districts to assess the impact of the intervention.
  • Synthetic control analysis to estimate the effect of the intervention on VMMC uptake.
• Sampling: At baseline, 3,034 individuals aged 10 to 34 were surveyed in a 0–10 km radius of the district hospitals. The endline survey expanded to include 3,034 individuals in a 20 km radius.
• Qualitative Data Collection: Included ethnography, focus group discussions (FGDs), and key informant interviews (KIIs), conducted in Chichewa and later transcribed into English.
• Quantitative Data Collection: Focused on measuring VMMC uptake and associated knowledge and attitudes.

Key Findings

• Incentives Increased VMMC Uptake: All three analyses confirmed that incentives, combined with the MoH's communications campaign, significantly increased VMMC uptake in the two study districts.
• Rumphi and Mchinji:
  • In Mchinji, the monthly VMMC rate increased from 0.24 to 2.61 per 1,000 adult males during the 4-month intervention period.
  • In Rumphi, the rate increased from 0.48 to 3.78 per 1,000 adult males during the same period.
  • After the intervention ended, the rate in Mchinji dropped to 0, while it remained stable in Rumphi.
• Voucher Redemption Rates:
  • Voucher redemption rates were higher in Mchinji (40%) than in Rumphi (26%), with a statistically significant difference (P<0.01).
  • 93% of the redeemed vouchers reached the intended targets (ages 10–34).
  • Only 27 out of 2,241 circumcisions did not involve voucher redemption.

Implications of Implementation Changes

• The intervention was not fully implemented as planned, which affected the randomisation and data analysis.
• The MoH's communication campaign inadvertently created demand for vouchers in the control group, undermining the study's design.
• The collective incentives were delayed until after the study ended.
• The synthetic control method was adopted to account for these changes, using data from 17 other districts.

Limitations and Ethical Considerations

• The study faced challenges in maintaining the integrity of the control group due to the widespread distribution of voucher-related messages.
• Ethical compliance was ensured, and the study was conducted with the support of local and national authorities, including the MoH, MoEST, and district offices.
• The data collection tools were carefully designed and implemented, and the study adhered to ethical standards.

Conclusion

The study found that incentives significantly increased VMMC uptake in the two districts. The intervention's design, although modified during implementation, still yielded meaningful insights into the effectiveness of incentives in improving VMMC demand. The findings underscore the importance of integrating incentives with broader communication strategies to enhance the uptake of VMMC in communities with low baseline rates. The role of informal social networks and the involvement of community and family members were also highlighted as important factors influencing VMMC uptake.

Key Information

• Study Period: January 2015 to December 2016.
• Intervention Period: December 2015 to April 2016.
• Study Districts: Rumphi (Tumbuka ethnic group) and Mchinji (Chewa ethnic group).
• Sample Size: 3,034 individuals at baseline and endline.
• Incentives: Vouchers and collective rewards for institutions.
• Key Informants: Included head teachers, clinical service providers, community leaders, and Mothers' Group members.
• Methodology: Synthetic control analysis, before-after analysis, and comparison with other districts.
• Findings: Incentives had a measurable impact on increasing VMMC uptake, especially in Mchinji.

Tables and Figures

• Figure 1: Shows the number of VMMC procedures in Mchinji and Rumphi before, during, and after the intervention.
• Table 1: Lists the impact evaluation questions, outcomes, and analytical approaches.
• Table 4 and 5: Provide data on the number of voucher booklets distributed and VMMC procedures performed.
• Table 6 and 7: Show voucher redemption rates in Rumphi and Mchinji.
• Table 8 and 9: Compare the MSPE (Mean Squared Prediction Error) ratios for the study districts and their comparators.

Overall Impact

The study demonstrated that incentives, particularly individual vouchers, can effectively increase VMMC uptake among boys and young men in Malawi. The findings support the use of such strategies in conjunction with broader public health campaigns to address the barriers to VMMC adoption, especially in communities with low baseline rates.