External validity, generalisability, applicability and directness: a brief primer.

Abstract

External validity is a construct that attempts to answer the question of whether we can use the results of a study in patients other than those enrolled in the study. External validity consists of two unique underlying concepts, generalisability and applicability. When the concern is about extending the results from a sample to the population from which the sample was drawn, the problem is one of generalisability. When the concern is about using inferences drawn from study participants in the care of specific patients belonging to any population, the problem is one of applicability. Clinicians, guideline developers and policymakers do not struggle with generalisability, but often struggle with applicability. When applicability is deemed to be low for a certain population, certainty in the supporting evidence becomes low due to indirectness.