A decade ago, Newhouse (1987) assessed the balance of trade between imports from the econometrics literature into health economics, and exports from health economics to a wider audience. While it is undoubtedly true that imports of concepts and techniques still dominate the balance, the literature reviewed in this chapter shows that the range and volume of applied econometric work in health economics has increased dramatically over the past ten years. Examples of good practice in health econometrics make extensive use of tests for misspecification and explicit model selection criteria. Robust and distribution-free estimators are of increasing importance, and the chapter gives examples of nonparametric, and semiparametric estimators applied to sample selection, simultaneous equations, count data, and survival models. Published replications of empirical results remain relatively rare. One way in which this deficit may be remedied is through the appearance of more systematic reviews of econometric studies. The use of experimental data remains an exception and most applied studies continue to rely on observational data from secondary sources. However applied work in health economics is likely to be influenced by the debate concerning the use of data from social experiments. The chapter illustrates the impressive diversity of applied econometric work over the past decade. Most of the studies reviewed here use individual level data and this has led to the use of a wide range of nonlinear models, including qualitative and limited dependent variables, along with count, survival and frontier models. Because of the widespread use of observational data, particular attention has gone into dealing with problems of self-selection and heterogeneity bias. This is likely to continue in the future, with the emphasis on robust estimators applied to longitudinal and other complex datasets.