It is widely accepted that clinical trials have to be carefully reviewed for internal validity. In addition, aspects of external validity, which is also known as 'generalizability' or 'directness', must be considered. The question of whether the study results can be applied to clinical practice under different conditions than the study itself is of major importance. In contrast to internal validity, external validity has to be viewed as an aspect of the situation, not of a study per se. Assessment of external validity addresses the question of whether effects (comparisons between treatments) are different between patient groups or clinical situations. It is not sufficient and may even not be important whether the patients differ. In epidemiology this situation is well known as 'effect modification.' External validity can be assessed according to the PICO scheme. However, empirical data about effect modifiers are scarce. Consequently, external validity is merely a matter of clinical judgement.