OBJECTIVES: The concept of minimally important difference (MID) is defined as the smallest change on a patient-reported outcome (PRO) questionnaire that a patient considers meaningful. There is currently little guidance on selecting the appropriate MID method or for interpreting MID results. In addition, studies rarely differentiate between effect-size-based (MID_E) estimates, which can be used for sample size determination, and observed-change-based (MID_C) estimates which can be used for interpreting changes in scores. This study compared MID estimation methods in two PROs to provide empirical evidence on the 1) efficiency of different MID estimation methods, 2) the difference between MID_E and MID_Cestimates, and 3) the relative magnitude of MID estimates across different PROs.

METHODS: This study was a retrospective analysis of observational data from patients undergoing elective surgery for chronic rhinosinusitis or end-stage ankle arthritis. The PROs investigated were the Sino-Nasal Outcome Test (SNOT-22) and the Ankle Osteoarthritis Scale (AOS). The study applied three distribution-based MID methods and three anchor-based MID methods. Bootstrap 95% confidence intervals were used to assess efficiency.

RESULTS: