Many authors have suggested that administrative health data files could provide the solution to the well-known research problem of obtaining data about volunteer subjects. However, many authors have failed to achieve satisfactory results because the use of these data files leads to special difficulties. Thus, unless the proper methodology is applied, use of health data files does not produce valuable results. In this paper we propose a conceptual framework which can be used to examine the suitability of health data files. More specifically, five aspects are discussed: (1) research team knowledge and skills; (2) identification and documentation of an accessible file; (3) file adequacy to answer the research question; (4) data quality; and (5) data processing. This conceptual framework should help health policy planners and researchers to better identify the problems they will meet and to find adequate solutions to these problems before embarking on the venture of using public data files.