Models for Health Care

Abstract

Abstract This article, which provides an outline of the methods that are typically used to model individual health care costs, reviews the literature on the comparative performance of the methods, especially in the context of forecasting individual health care costs, and concludes with an empirical case study. It is organized as follows. Section 2 begins with linear regression on the level of costs and on transformations of costs. Section 3 moves on to nonlinear regressions that are specified in terms of an exponential conditional mean. Many recent studies of nonlinear specifications are embedded within the generalized linear model (GLM) framework. The language of the GLM approach is commonplace in the statistics literature, but is less used in econometrics and is outlined in Section 4. Recent research has seen the development of more flexible parametric and semiparametric approaches, and some of the key methods are described in Section 5. Section 6 reviews evidence on the comparative performance of methods that are most commonly used to model costs and for some of the recent methodological innovations. This is reinforced in Section 7, which presents an illustrative application of the methods with data from the US Medical Expenditure Panel Study. Section 8 suggests some further reading.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 101 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.