Who benefits from health insurance? Uncovering heterogeneous policy impacts using causal machine learning.

TY - UNPB

T1 - Who benefits from health insurance?

T2 - Uncovering heterogeneous policy impacts using causal machine learning.

AU - Kreif, Noemi

AU - Mirelman, Andrew

AU - Moreno Serra, Rodrigo

AU - Hidayat, Taufik

AU - DiazOrdaz, Karla

AU - Suhrcke, Marc

PY - 2020/10/6

Y1 - 2020/10/6

N2 - To be able to target health policies more efficiently, policymakers require knowledge about whichindividuals benefit most from a particular programme. While traditional approaches for subgroupanalyses are constrained only to consider a small number of arbitrarily set, pre-defined subgroups,recently proposed causal machine learning (CML) approaches help explore treatment-effectheterogeneity in a more flexible yet principled way. This paper illustrates one such approach –‘causal forests’ – in evaluating the effect of mothers’ health insurance enrolment in Indonesia.Contrasting two health insurance schemes (subsidised and contributory) to no insurance, we findbeneficial average impacts of enrolment in contributory health insurance on maternal health careutilisation and infant mortality. For subsidised health insurance, however, both effects were smallerand not statistically significant. The causal forest algorithm identified significant heterogeneity in theimpacts of the contributory insurance scheme: disadvantaged mothers (i.e. with lower wealthquintiles, lower educated, or in rural areas) benefit the most in terms of increased health careutilisation. No significant heterogeneity was found for the subsidised scheme, even though thisprogramme targeted vulnerable populations. Our study demonstrates the power of CML approachesto uncover the heterogeneity in programme impacts, hence providing policymakers with valuableinformation for programme design.

AB - To be able to target health policies more efficiently, policymakers require knowledge about whichindividuals benefit most from a particular programme. While traditional approaches for subgroupanalyses are constrained only to consider a small number of arbitrarily set, pre-defined subgroups,recently proposed causal machine learning (CML) approaches help explore treatment-effectheterogeneity in a more flexible yet principled way. This paper illustrates one such approach –‘causal forests’ – in evaluating the effect of mothers’ health insurance enrolment in Indonesia.Contrasting two health insurance schemes (subsidised and contributory) to no insurance, we findbeneficial average impacts of enrolment in contributory health insurance on maternal health careutilisation and infant mortality. For subsidised health insurance, however, both effects were smallerand not statistically significant. The causal forest algorithm identified significant heterogeneity in theimpacts of the contributory insurance scheme: disadvantaged mothers (i.e. with lower wealthquintiles, lower educated, or in rural areas) benefit the most in terms of increased health careutilisation. No significant heterogeneity was found for the subsidised scheme, even though thisprogramme targeted vulnerable populations. Our study demonstrates the power of CML approachesto uncover the heterogeneity in programme impacts, hence providing policymakers with valuableinformation for programme design.

KW - policy evaluation

KW - machine learning

KW - heterogeneous treatment effects

KW - health insurance

M3 - Working paper

T3 - CHE Research Paper

BT - Who benefits from health insurance?

PB - Centre for Health Economics, University of York

CY - York, UK

ER -