OBJECTIVES: The suitability and the practicality of using different data sources in economic evaluation have been well documented. The objective of this study was to identify any trends in the types of data sources used in these studies over a 6-year period (1995 - 2000) METHODS: For this preliminary study, the area of heart disease was selected to create a sample of papers for analysis. A simple search strategy was constructed and run on the NHS EED database. The retrieved studies were then sorted by year and type of economic evaluation. The sources of data used to estimate effectiveness, resource use and costs for the interventions or programmes in the studies were then identified for each study. RESULTS: The literature search retrieved 326 economic evaluations published between 1995 and 2000. Data used in the evaluations ranged from solely expert opinion to pragmatic trials where cost and resource use information were collected alongside one another. There is a downward trend in the proportion of studies using a retrospective analysis to establish the effectiveness of interventions and programmes. A review of the literature (with no meta-analysis) has remained a constant data source for effectiveness, accounting for approximately a quarter of all evaluations each year. The type of resource utilisation data used has not changed over the years, with the majority of information collected alongside trials, with either the whole study population or with smaller cohorts, or a retrospective analysis of case notes. The documentation of cost data sources has improved over the six years, and the use of previously published data has decreased also. CONCLUSIONS: This preliminary study reveals definite trends in the different data sources used in economic evaluation over this six-year period. Whether these trends can be seen regardless of disease area and type of intervention remains to be investigated.