Minimal clinically important differences: review of methods.

Determining a minimal clinically important difference to be detected in a clinical trial is a critical methodological step in the design of a study. In this review, the different methods that have been used in detecting important changes or differences are considered and categorized according to a classification system consisting of which groups are contrasted, the setting of the results, and the type of change of difference being quantified. It was found that most methods and procedures for deriving minimal clinically important differences considered important changes from the viewpoint of a group of patients. Development of methods that focus on individuals should be a goal of the future.