Validation of the Charité Mobility assessment (CHARMI) in older adults across a wide range of functional levels - the prospective longitudinal MobiTest cohort study

Comprehensive mobility assessment is crucial in geriatrics, yet existing tools may lack feasibility, or applicability for older adults with varying functional levels. We evaluated the construct validity, interrater reliability, responsiveness, and minimal important change (MIC) of the Charité Mobility Index (CHARMI), in older adults. A prospective longitudinal cohort study was conducted with older adults ≥ 70y undergoing rehabilitation in acute care, inpatient or outpatient rehabilitation. CHARMI assessments were performed at admission and discharge and correlated with established mobility measures – the De Morton Mobility Index (DEMMI), Short Physical Performance Battery (SPPB), Timed Up and Go (TUG), and Barthel Index (BI) mobility items. Construct validity and responsiveness were evaluated using a hypothesis-testing approach based on a priori-hypotheses; interrater reliability (IRR) was assessed in a subsample. The MIC was determined in an anchor-based approach via ROC analysis. 93 participants (mean age 82.8y) were included. Evidence for validity was supported by meeting 7 of 8 a priori hypotheses including strong correlations with DEMMI and SPPB and a moderate inverse correlation with TUG. IRR was 0.97 (95% CI: 0,91; 0,99). Evidence for responsiveness was supported by meeting 6 of 8 a priori hypotheses. The MIC was established at 1.5 CHARMI points. This study provides evidence supporting the CHARMI as a valid, responsive and reliable outcome measure with a useful MIC, making it a valuable option for tracking mobility in older adults across different settings and timepoints. Further studies should confirm generalizability and evaluate feasibility.