Econometric modeling of health care costs and expenditures: a survey of analytical issues and related policy considerations.

BACKGROUND: Econometric modeling of healthcare costs and expenditures has become an important component of decision-making across a wide array of real-world settings.
OBJECTIVES: The objective of this article is to provide a brief summary of important conceptual and analytical issues involved in econometric healthcare cost modeling. To this end, the article explores: outcome measures typically analyzed in such work; the decision maker's perspective in econometric cost modeling exercises; specific analytical issues in econometric model specification; statistical goodness-of-fit testing; empirical implications of "upper tail" (or "high cost") phenomena; and issues relating to the reporting of findings. DATA: Some of the concepts explored here are illustrated in light of samples drawn from the 2005 Medical Expenditure Panel Survey and the 2005 Nationwide Inpatient Sample. RESULTS AND
CONCLUSIONS: Analysts of healthcare cost data have at their disposal an increasingly sophisticated tool kit for analyzing such data that can in principle and in fact yield increasingly interesting insights into data structures. Yet for such analyses to usefully inform policy decisions, the manner in which such studies are designed, undertaken, and reported must accommodate considerations relevant to the decision-making community. The article concludes with some preliminary thoughts on how such bridges might be constructed.