Understanding incentive preferences of community health workers using discrete choice experiments: a multicountry protocol for Kenya, Uganda, Bangladesh and Haiti. Issue 12 (11th December 2019)

• Record Type: Journal Article
• Authors: Agarwal, Smisha; Anaba, Udochisom; Abuya, Timothy; Kintu, Richard; Casseus, Alain; Hossain, Sharif; Obadha, Melvin; Warren, Charlotte E
• Is Part Of: BMJ open. Volume 9:Issue 12(2019)
• Publication Date: 2019-12-11
• Subjects: community health workers -- community health systems -- incentives -- discrete choice experiment; Medicine -- Research -- Periodicals; 610.72
• DOI: 10.1136/bmjopen-2019-033601
• Languages: English
• ISSNs: 2044-6055

Abstract

Abstract : Introduction: There is a renewed global interest in improving community health worker (CHW) programmes. For CHW programmes to be effective, key intervention design factors which contribute to the performance of CHWs need to be identified. The recent WHO guidelines recommends the combination of financial and non-financial incentives to improve CHW performance. However, evidence gaps remain as to what package of incentives will improve their performance in different country contexts. This study aims to evaluate CHW incentive preferences to improve performance and retention which will strengthen CHW programmes and help governments leverage limited resources appropriately. Methods and analysis: A discrete choice experiment (DCE) will be conducted with CHWs in Bangladesh, Haiti, Kenya and Uganda with different levels of maturity of CHWs programmes. This will be carried out in two phases. Phase 1 will involve preliminary qualitative research including focus group discussions (FGDs) and key informant interviews to develop the DCE design which will include attributes relevant to the CHW country settings. Phase 2 will involve a DCE survey with CHWs, presenting them with a series of job choices with varying attribute levels. An orthogonal design will be used to generate the choice sets for the surveys. The surveys will be administered in locally-appropriate languages to at least 150 CHWs from each of the cadres in each country. Conditional and mixed multinomial logit (MMNL) models will be used for the estimation of stated preferences. Ethics and dissemination: This study has been reviewed and approved by the Population Council's Institutional Review Board in New York, and appropriate ethics review boards in Kenya, Uganda, Bangladesh and Haiti. The results of the study will be disseminated through in-country dissemination workshops, meetings with country-level stakeholders and policy working groups, print media, online blogs and peer-reviewed journals.