The trust and insurance models of healthcare purchasing in the Ayushman Bharat Pradhan Mantri Jan Arogya Yojana in India: early findings from case studies of two states

BACKGROUND: The Pradhan Mantri Jan Arogya Yojana (PMJAY), a publicly funded health insurance scheme for the poor in India, was launched in 2018. Early experiences of states with various purchasing arrangements can provide valuable insights for its future performance. We sought to understand the institutional agencies and performance of the trust and insurance models of purchasing with respect to; a) Provider contracting b) Claim management c) Implementation costs. METHODS: A mixed methods case study design was adopted. Two states, Uttar Pradesh (representing a trust model) and Jharkhand (representing the insurance model) were purposively selected. Data sources included document reviews, key informant interviews, quantitative scheme data from the provider empanelment and claims database, and primary data on costs. Descriptive statistics were reported for quantitative data, content analysis was used for thematic reporting of qualitative data. RESULTS: In both models, the state was the final authority on empanelment decisions, with no significant influence of the insurance company. Private hospitals constituted the majority of empanelled providers, with wide variations in district-wise distribution of bed capacities in both states. The urgency of completing empanelment in the early days of the scheme created the need for both states to re-review hospitals and de-empanel those not meeting requirements. Very few quality- accredited private hospitals were empaneled. The trust displayed more oversight of support agencies for claim management, longer processing times, a higher claim rejection rate and numbers of queries raised, as compared to the insurance model. Support agencies in both states faced challenges in assessing the clinical decisions of hospitals. Cost-effectiveness showed mixed results; the trust cost less than the insurance model per beneficiary enrolled, but more per claim generated. CONCLUSIONS: Efforts are required to enable a better distribution and ensure quality of care in empanelled hospitals. The adoption of standard treatment guidelines is needed to support hospitals and implementing agencies in better claim management. The oversight of agencies through enforcement of contracts remains vital in both models. Assessing the comparative performance of trusts and insurance companies in more states at later stages of scheme implementation, would be further useful to determine their cost-effectiveness as purchasers. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12913-022-08407-2.