Universal health coverage (UHC) requires not only insurance enrolment but also effective use of coverage. In Indonesia, despite near-universal enrolment in the National Health Insurance programme (Jaminan Kesehatan Nasional, JKN), many insured individuals do not use their insurance when seeking care. Evidence on the extent and determinants of this underutilisation remains limited. We analysed nationally representative data from the 2023 National Socioeconomic Survey (SUSENAS), including insured individuals who reported outpatient care in the past month or inpatient care in the past 12 months. Insurance underutilisation was defined as the non-use of JKN during a healthcare encounter among those insured. Guided by Andersen’s Behavioral Model, we estimated random-effects logistic regression models for outpatient and inpatient care, incorporating individual, household, and district-level health system factors and accounting for clustering at the district level. Among insured individuals, 48.3% did not use JKN for outpatient care, compared with 11.0% for inpatient care. In adjusted models, outpatient non-use was higher among self-employed (aOR 1.43), casual (aOR 1.54), and unpaid family workers (aOR 1.42) than among those not working, and increased with income (highest vs. lowest quartile: aOR 1.34). Rural residents were more likely to underutilise outpatient insurance (aOR 1.47), while higher education was protective (university vs. elementary: aOR 0.77). Greater hospital (aOR 0.28) and primary care density (aOR 0.86) reduced outpatient non-use. For inpatient care, non-use remained associated with self-employment (aOR 1.29), high income (aOR 1.56), rural residence (aOR 1.25), and lower hospital density (aOR 0.21). Between-district differences explained 21% of outpatient and 14% of inpatient variation. Substantial inequities persist in the effective use of JKN, particularly for outpatient care. Addressing labour-market constraints, opportunity costs, administrative barriers, and local service capacity is essential to translate coverage into equitable access.
Sujarwoto et al. (Wed,) studied this question.