Substantial changes in the organization, delivery, and financing of health care over the last decade, combined with data collection and methodological improvements in the 1996 Medical Expenditure Panel Survey (MEPS), pose special challenges in comparing expenditure estimates in MEPS with those in the 1987 National Medical Expenditure Survey (NMES). The 1987 NMES used charges as its fundamental expenditure concept, whereas the 1996 MEPS used actual payments as its expenditure measure. In spite of these differences, researchers and policymakers will want to be able to analyze trends in health care expenditures using these two surveys. We discuss these issues in detail and present a simple, straightforward adjustment method that can be applied to the 1987 NMES public use expenditure data to improve comparability to the MEPS. We base this adjustment method on an analysis of provider-reported payment data collected in NMES. We present several examples of the application of this method that illustrate the importance of the adjustments for analyses of trends in health care spending.