OBJECIVE: Observational data often contain retrospective patient-specific and population-based information, and have been widely used for health services and outcomes research. The objective of this study is to review and compare characteristics of public and proprietary health care databases available for research in Pharmacoeconomics, epidemiology, and outcomes research. METHODS: Published information and literature on health care databases other than clinical trials were searched using Medline, International Pharmaceutical Abstracts, Internet, as well as other commercially available sources. RESULTS: Over 51 public health data sources or proprietary health care databases were selected for review. This paper discussed major characteristics of these selected healthcare databases by categories, including: 1) electronic claims databases, like US Medicare and Medicaid, Canadian Saskatchewan health database, US Veterans Affairs clinic databases, regional Blue Cross/Blue Shield database; 2) medical record databases, like Harvard Pilgrim health care, United Health Group, UK General Practice Research Database, Kaiser Permantente; 3) spontaneous adverse drug reaction reporting systems, like Food Drug Administration Adverse Event Reporting Systems, Canadian Adverse Drug Reaction Monitoring Program, World Health Organization Adverse Reaction database, and UK Yellow Card; 4) health care survey data, like US Medical Expenditure Panel Survey (MEPS), and National Ambulatory Medical Care Survey; and 5)disease registries and other health data, like Surveillance Epidemiology and End Results, Metropolitan Atlanta Congenital Defect Program, and National Death Index. In addition, the linkage capability of health data files, data validity, and health privacy issues were also discussed. CONCLUSION: Many public and proprietary healthcare databases are available for research in Pharmacoeconomics and Pharmacoepidemiology. This review article on healthcare databases provides useful information to compare and identify potential appropriate databases for specific research objectives.