Indonesian National Health Insurance scheme longitudinal sample data 2015-2020: overview and potential uses for health policy analysis.

BACKGROUND: The Indonesian National Health Insurance Agency (BPJS-K) administers one of the largest single-payer healthcare systems in the world, with 95% of Indonesia's population registered by December 2023. Since 2019, BPJS-K has provided sample data representing 1% of insured individuals. Despite its potential, the BPJS-K sample data remains underutilised in research. METHODS: This study provides an overview of the BPJS-K dataset, including research that has used it, its structure, contents, and key variables from 2015 to 2020. We present descriptive statistics for the sample, including age and gender distributions, insurance membership type, healthcare visits, diagnoses, referrals, and associated tariffs. We illustrate the dataset's potential applications for health policy analysis and its strengths and limitations. RESULTS: The BPJS-K sample data broadly represents the Indonesian population, as evidenced by comparisons to census data. Regional disparities in healthcare utilisation were observed, with lower service access in Eastern Indonesia. Key variables include diagnoses, such as acute respiratory infections (6% of the visits, the highest in primary healthcare), and reimbursement information for visits to referral healthcare providers and for non-capitation services to primary healthcare providers. The data has the potential to facilitate health policy analysis, given its longitudinal nature and possible linkage to other data. However, current shortcomings, such as limited socio-economic information and quality of diagnostic information, should be considered. CONCLUSION: The BPJS-K sample data offers potential for longitudinal and cross-sectional health policy research. However, further improvements in data quality, diagnostic recording, accessibility, and linkage to socio-economic data are recommended to optimise its usefulness for research.