Assessing primary care performance in Indonesia: An application of frontier analysis techniques

Hafidz, Firdaus Ensor, Tim Tubeuf, Sandy [UCL]

Despite increased national health expenditure in health facilities in Indonesia, health outcomes remain low. The aim of our study is to examine the factors determining the relative efficiency of public primary care facilities. Using linked national data sources from facility-, households, and village-based surveys, we measure the efficiency of 185 primary care facilities across fifteen provinces in Indonesia with output oriented data envelopment analysis (DEA) and stochastic frontier analysis (SFA). Inputs include the number of doctors, midwife and nurses, and other staff while outputs are the number of outpatients and maternal child health patients. We run truncated regression in second stage DEA and one stage SFA analysis to assess contextual characteristics influencing health facilities performance. Our results indicate a wide variation in efficiency between health facilities. High-performing primary care facilities are in affluent areas. Primary care facilities located in urban areas, in Java and Bali Island, with high coverage of insurance scheme for the poor perform better than other geographical location. We find an inconclusive impact of quality of care, patient mix, and availability of inpatient services on efficiency. This paper concludes by highlighting the characteristics of primary care facilities that have the potential to increase efficiency.