Defining thresholds for meaningful interpretation of patient-reported outcome scores

There are many different patient-reported outcome measures to assess how patients feel or function but they vary in how they collect responses and score results. Therefore, it can be hard to know what a change in score means for a patient or whether treatments compared in a trial result in different patient experiences. Methods exist to estimate thresholds to aid interpretation, but no single method is considered the best. This review considers recent advances in efforts to improve these methods.

New research has been published on the choice of appropriate anchors and the methods to establish thresholds based on them. There has been increased focus on the derivation of thresholds used to compare scores between groups of patients versus those used to interpret an individual's change in scores over time. Studies are also highlighting how baseline severity and other factors affect how we estimate thresholds.

Consensus on the optimal methods for the derivation of score interpretation thresholds is still required, although the field continues to grow in understanding biases and impacts of treatment and population factors on estimation.