Investigating the MCID of health status tools for patients with COPD in pulmonary rehabilitation

Background: COPD is a major public health problem. Health status has become an important tool in its diagnostics and management with CCQ, CAT and SGRQ as the main instruments. Important is to what extent a change in its score is clinically relevant. The Minimal Clinically Important Difference (MCID) defines this. Current MCIDs are 0.4 (CCQ), 2.0 (CAT) and 4.0 (SGRQ). This study investigated the MCID of these tools using multiple approaches, since existing evidence is limited. Methods: In total 316 patients participated in a 3-week Pulmonary Rehabilitation programme (63% male, 51%FEV 1 pred., 41 mean packyears, 58 years mean age, 2/50/38/10% GOLD stage I/II/III/IV). Anchor-based approaches, including patient-referencing, criterion-referencing and questionnaire-referencing, and the distribution-based methods SEM and 0.5SD were used to determine the MCID. Results: Table 1: MCID estimates CCQ CAT SGRQ Patient-referencing 0.49 2.80 7.03 Criterion-referencing 0.42 - - Questionnaire-referencing SGRQ as anchor Scatter plots / regression 0.30 2.17 Error bar estimate 0.44 2.66 ROC Curves 0.45 3.00 CAT as anchor Scatter plots / regression 0.41 7.26 Error bar estimate 0.47 7.69 ROC Curves 0.45 7.92 CCQ as anchor Scatter plots / regression 2.81 7.58 Error bar estimate 2.86 7.85 ROC Curves 3.50 7.92 Distribution-based approaches SEM 0.51 3.02