Understanding the factors that influence the performance of India’s community nutrition workers: Anganwadi workers of the integrated child development services scheme in Bihar

My PhD thesis is an enquiry into the factors that influence the performance of community nutrition workers known as Anganwadi workers (AWWs) employed by the Integrated Child Development Services (ICDS) scheme in the Indian state of Bihar. My research question asks: what factors influence the performance of AWWs and how does the addition of a technology augmented intervention influence AWW performance in the context of a state with a high burden of child undernutrition. I use qualitative and quantitative methods to answer my research question. To explore the concept of performance in the AWW context, I developed a conceptual framework informed by a review of frameworks on the performance of community and facility-based health workers. In my research I utilise the context of a pilot programme – the Bihar Child Support Programme – that introduced mobile phone technology as a job aid for the AWW, combined with a monetary incentive. As part of the qualitative research, I conducted 30 semi-structured interviews with AWWs including 15 AWWs who received the mobile phone technology and monetary incentives intervention of the BCSP. I used a hybrid method of inductive and deductive thematic analysis to analyse the data. In the quantitative research, I employed a Difference?in?Difference estimation strategy to assess the influence of the mobile phone technology and monetary incentives intervention on the uptake of ICDS services linked to the intervention. I found a range of factors that impact on AWW performance. My research identified four new factors to add to the starting framework: family support, beneficiary and AWW service preferences, seasonal migration, and corruption. The technology augmented intervention examined in this thesis would have been expected to be successful based on the existing frameworks for community and facility-based health worker performance. However, it had no positive impact on household level service delivery outcomes. One of the new factors identified in this thesis – beneficiary and AWW service preferences – is the primary explanatory for this. The intervention sought to strengthen information-oriented nutrition services (weighing and counselling) but this was not a preference for either the beneficiaries (who prefer product-oriented services) or AWWs (who prefer education related services due to their self-identification as pre-school teachers) and as such did not lead to impact. This has implications for the understanding of the motivation and performance of AWWs and similar community health workers and the design of interventions aimed at improving their performance.