Associations between Indonesia's national health insurance, effective coverage in maternal health and neonatal mortality: a multilevel interrupted time-series analysis 2000-2017.

Background: We assessed the effect of Indonesia's national health insurance programme ( ()) on effective coverage for maternal and child health across geographical regions and population groups.

Methods: We used four waves of the Indonesia Demographic and Health Survey from 2000 to 2017, which included 38 880 women aged 15-49 years and 144 000 birth records. Key outcomes included antenatal and delivery care, caesarean section and neonatal and infant mortality. We used multilevel interrupted time-series regression to examine changes in outcomes after the introduction of the JKN in January 2014.

Findings: JKN introduction was associated with significant level increases in (1) antenatal care (ANC) crude coverage (adjusted OR (aOR) 1.81, 95% CI 1.44 to 2.27); (2) ANC quality-adjusted coverage (aOR 1.66, 95% CI 1.38 to 1.98); (3) ANC user-adherence-adjusted coverage (aOR 1.80, 95% CI 1.45 to 2.25); (4) safe delivery service contact (aOR 1.83, 95% CI 1.42 to 2.36); and (5) safe delivery crude coverage (aOR 1.45, 95% CI 1.20 to 1.75). We did not find any significant level increase in ANC service contact or caesarean section. Interestingly, increases in ANC service contact and crude coverage, and safe delivery crude coverage were larger among the poorest compared with the most affluent. No statistically significant associations were found between JKN introduction and neonatal and infant mortality (p>0.05) in the first 3 years following implementation.

Interpretation: Expansion of social health insurance led to substantial improvements in quality of care for maternal health services but not in child mortality. Concerted efforts are required to equitably improve service quality and child mortality across the population in Indonesia.