ESTIMASI DAMPAK PROGRAM ASURANSI KESEHATAN PADA JUMLAH KUNJUNGAN RAWAT JALAN DI INDONESIA

Abstract

Background and method: This research aimed to select
the best methods to predict the effect of health insurance
program on the numbers of outpatient visits in Indonesia. The
analysis was applied to the second round of the Indonesian
Family Life Survey data (IFLS2).
Result: The author compares the estimation results derived
from 6(six) econometrics technique count data model and select
the best alternatives based on several statistics tests. The
results confirm that Generalized Method of Moments (GMM)
estimator is best to model the number of visits to public outpatient,
whilst Hurdle Negative Binomial (HNB) is superior to model the
number of visits to private one. It is proved that the insured
have higher probability in the number of visits for outpatient
services then uninsured (p<1%). Supplies induce demand
phenomena was not detected among the insured, however
this behaviour was likely happen where provider’s competition
are relatively high.
Conclusions: This study concludes that estimates of health
care demand given insurance have been shown to depend on
the empirical specification used in the analysis. Not controlling
the existence endogeneity of insurance leads to lower the
parameter estimates. This study supports a national health
insurance policy as an instrument to increase access to formal
health care services.
Keywords: health insurance, modeling, demand for health care
services