Abstract: The decision about whether the findings from analytical epidemiological studies can be extrapolated to another setting is an important one. We discuss some of the issues involved in the process of generalisation. A critical systematic qualitative approach should be used. This involves an assessment of the internal validity of the study and of the nature of the study base. The study subject matter and choice of epidemiologic measure should be considered. A study base similar to the target population is essential if the study is descriptive or if there is concern about effect modifiers whose effects cannot be predicted. In addition, data from quantitative tests of study findings in different populations should be considered.