Examining the minimum important difference

The increasing use of constructed scales and indices in clinical science ha s preceded in many cases a clear understanding of how to appraise the impor tance of the differences or changes that are thereby observed. For example, in the design of clinical trials which employ such scales as outcome measu res it may be difficult to specify what constitutes a clinically significan t shift in means, a key factor in sample size calculations. Determination o f the minimum important difference relative to specific outcome measures ha s historically been based on informal and/or intuitive arguments. In this p aper we propose a formal statistical framework for these considerations, ba sed on a previously published validation study design which captures patien ts' perceptions in comparative self-reported assessments. We begin by adopt ing a mixed-effect model to represent the comparative assessments as compos ites of individual self-ratings on an underlying continuous scale. We then present two basic approaches for assessing the relation between the hypothe sized latent scale and the outcome scale(s) under consideration, taking the latent scale as a plausible benchmark against which observable changes on the outcome scale can be judged. Copyright (C) 1999 John Wiley & Sons, Ltd.