A Wald test comparing medical costs based on log-normal distributions withzero valued costs

Medical cost data often exhibit strong skewness and sometimes contain large proportions of zero values. Such characteristics prevent the analysis of v ariance (ANOVA) F-test and other frequently used standard tests from provid ing the correct inferences when the comparison of means is of interest. One solution to the problem is to introduce a parametric structure based on lo g-normal distributions with zero values and then construct a likelihood rat io test. While such a likelihood ratio test possesses excellent type I erro r control and power, its implementation requires a rather complicated itera tive optimization program. In this paper, we propose a Wald test with simpl e computation. We then conduct a Monte Carlo simulation to compare the type I error rates and powers of the proposed Wald test with those of the likel ihood ratio test. Our simulation study indicates that although the likeliho od ratio test slightly outperforms the Wald test, the performance of the Wa ld test is also satisfactory, especially when the sample sizes are reasonab ly large. Finally, we illustrate the use of the proposed Wald test by analy sing a clinical study assessing the effects of a computerized prospective d rug utilization intervention on in-patient charges. Copyright (C) 1999 John Wiley & Sons, Ltd.