Many methods for modeling skewed health care cost and use data have been suggested in the literature. This talk will compare the performance of eight alternative estimators, including OLS and GLM estimators and one- and two-part models, in predicting different types of health care costs including Medicare costs, physicians' services costs, and costs for persons with employer-sponsored insurance. I find many of the alternatives produce very similar results in practice. I then suggest an efficient method for researchers to use when selecting estimators of health care costs.