Abstract
BACKGROUND: Due to ongoing insecurity, the government of Afghanistan delivers health care to the country's population by contracting out service delivery to non-governmental organization service providers (SPs). In 2018, major changes to SP contracts were introduced, resulting in a new pay-for-performance service delivery model. This model, called "Sehatmandi", pays SPs based on the volume of 11 key services they provide. METHODS: A narrative review of Sehatmandi's key features is presented, as well as lessons learned during implementation. Counterfactual comparisons of service delivery data for 10 payment-related service indicators are made. The first comparison is between the rate of change in the volume of services delivered from 2018 to 2019 (ie, the first year of Sehatmandi implementation) relative to the rate change from 2017 to 2018 (ie, prior to the program). The second comparison is between the rate of change in the volume of services delivered in provinces under the pay-for-performance mechanism relative to provinces which were not financed using pay-for-performance. Time trends in non-payment service indicators and service quality are also examined. RESULTS: The increase in service volume in Sehatmandi provinces from 2018 to 2019 was higher than the increase from 2017 to 2018 for 8 out of 10 indicators. The median increase in the rate of change was 10 percentage points. Similar results were obtained when comparing pay-for-performance provinces to those not financed using pay-for-performance. Improvements were also observed for services that were not directly compensated by the pay-for-performance approach. Payment for service volume was not associated with reduced service quality. The narrative review suggests that the pay-for-performance system has stimulated more effective oversight of SPs by the government of Afghanistan and has incentivized innovative service delivery strategies by SPs. Sehatmandi may benefit from re-structuring its financial incentives to stimulate improved service quality and accelerate delivery of lagging services. CONCLUSIONS: The available evidence - though subject to some limitations - suggests that the introduction of a pay-for-performance system was associated with an expanded volume of service delivery in Afghanistan. This approach may be beneficial in other conflict-affected countries.