Objective
The minimal clinically important difference (MCID) represents the smallest change in treatment outcome deemed clinically meaningful. This study estimates the MCID for 2 widely used tinnitus measures: the Tinnitus Handicap Inventory (THI) and the Tinnitus Functional Index (TFI), using anchor-based approaches while accounting for baseline severity and time interval.
Study Design
A multi-center randomized clinical trial.
Setting
European tinnitus centers.
Methods
Anchor-based approaches, including the effect size, receiver-operating characteristics, and ΔTHI/TFI methods, were employed to determine the MCID. The “minimally improved” category of the Clinical Global Impression Scale-Improvement (CGI-I) served as the anchor. The standard error of measurement was used to assess random variation.
Results
For the THI, MCID estimates ranged from 7.8 to 12, with a point estimate of 11 after 12 weeks of treatment (N = 364). For the TFI, MCID estimates ranged from 7.3 to 9.4, with a point estimate of 9 points after 12 weeks (N = 359). Both measures indicated that higher baseline severity and longer time intervals required greater score reduction for clinical relevance.
Conclusion
This study highlights the context-specific nature of MCID values for tinnitus measures and emphasizes the need for consensus on optimal anchor-based approaches to improve comparability.