Estimating the minimum important change in the 15D scores

Abstract

Purpose: To facilitate the interpretation of empirical results produced by the 15D, a generic, preference-based instrument for measuring health-related quality of life (HRQoL), a subjective five-category global assessment scale (GAS) was used as an external anchor to determine the minimum important change (MIC) in the 15D scores.

Methods: Altogether 4,903 hospital patients representing sixteen disease entities and having the 15D score at baseline repeated the HRQoL assessment at 6 months after treatment and answered the question: compared to the situation before treatment, my overall health status is now (1) much better, (2) slightly better, (3) much the same, (4) slightly worse, (5) much worse. Regression analysis was used to estimate the MIC for improvement/deterioration, defined as the lower/upper limit of 99.9 % confidence interval of the regression coefficient, standardized for baseline HRQoL, for categories (2) and (4), respectively, and confirmed by ROC curve analysis.

Results: The limits or intervals for classifying the changes of 15D scores into GAS categories were >.035 for (1), .015-.035 for (2),>-.015 and<.015 for (3), -.035--.015 for (4) and <-.035 for (5). The lower/upper limits of ±.015 for categories (2) and (4) can be regarded as the generic MIC thresholds for improvement/deterioration, respectively.

Conclusions: The generic MICs for the change of 15D scores are ±.015. Follow-up studies using the 15D should report the mean change in the 15D score, its statistical significance, relationship to the MIC, and the distribution of the changes of the 15D scores into the five categories.