A Nonparametric Matching Method for Covariate Adjustment with Application to Economic Evaluation

In health economic studies that use observational data, a key concern is how to adjust for imbalances in baseline covariates due to the non-random assignment of the programs under evaluation. Traditional methods of covariate adjustment such as regression, depend on correct model specification. Alternatives such as propensity score matching depend on covariate balance being achieved. We demonstrate a nonparametric matching method, Genetic Matching, which uses a search algorithm to optimize covariate balance. Genetic Matching is a generalization of propensity score and Mahalanobis distance matching. We apply Genetic Matching to an economic evaluation of a clinical intervention, Pulmonary Artery Catheterization. Our results show that Genetic Matching achieves better covariate balance than propensity score matching. Genetic Matching gives different estimates of incremental effectiveness and cost-effectiveness compared to propensity score matching. We conduct Monte Carlo simulations that show that Genetic Matching reduces bias and root mean squared error, compared to propensity score matching. We conclude that Genetic Matching improves covariate balance, and it can lead to less biased estimates than propensity score matching.