Validity, inter-rater reliability, and feasibility of the Chelsea Physical Assessment Tool for assessing physical function in post-acute COVID-19 patients: A cross-sectional study

BACKGROUND: Various tools have been created to measure physical function during intensive care unit (ICU) stay and after ICU discharge, but those have not been validated in coronavirus 2019 (COVID-19) patients. There is a need for a reliable, valid and feasible tool to define the rehabilitation needs of post-ICU COVID-19 patients entering the acute wards and then rehabilitation clinics. OBJECTIVE: This study aims to investigate the validity, inter-rater reliability and feasibility of Chelsea Physical Assessment Tool (CPAx) in assessing the functional status of COVID-19 patients after discharge from the ICU. METHODS: Demographic and clinical characteristics of the patients were recorded. Patients were evaluated using the modified Medical Research Council (MRC) dyspnea scale, Functional Oral Intake Scale, Glasgow Coma Scale, CPAx, Barthel Index, Katz Index and MRC sum score, measurements of grip strength obtained by dynamometer, the 5 time sit-to-stand test and 30 seconds and sit-to-stand test. CPAx and the other functional assessment tools were administered to 16 patients within 48 hours following ICU discharge. For inter-rater reliability, another physiatrist independently re-assessed the patients. MRC sum score, Barthel and Katz indexes were used to assess construct validity of CPAx. The discriminative validity of CPAx was determined by its ability to differentiate between patients with and without ICU acquired muscle weakness based on MRC sum score. The intra-class correlation coefficients (ICC) were calculated to determine inter-rater reliability for total scores of the functional assessment tools. Cohen's Kappa (kappa) coefficient and weighed Kappa (kappa(w)) were calculated to determine inter-rater reliability of individual CPAx items. Ceiling and flooring effects were calculated by percentage frequency of lowest or highest possible score achieved. The number and percentages of the patients who were able to complete each tool were calculated to assess feasibility. RESULTS: The CPAx score was strongly correlated with MRC sum score (rho: 0.83), Barthel Index (rho: 0.87) and Katz Index (rho: 0.89) (p < 0.001) showing construct validity. Area under the ROC curve demonstrated that cut off score for CPAx was <= 12 to discriminate patients with MRC sum score < 48, with a sensitivity and a specificity of 100% and 63%, respectively (AUC = 0.859, p < 0.001). ICC was high for CPAx, MRC sum score, Barthel and Katz indexes, Glasgow Coma Scale, and hand grip strength measurement, with the highest value observed for CPAx (ICC, 0.96; 95% confidence interval (CI), 0.71-0.98). kappa and kappa(w) analysis showed good to excellent inter-rater reliability for individual CPAx items. No floor or ceiling effect was observed at CPAx while floor effect was observed at Barthel Index scores (25%) and Katz Index scores (37.5%). All patients could be evaluated using CPAx while less were physically able to complete the 5 time sit-to-stand, 30 seconds sit-to-stand tests (n = 4) and MRC sum score (n = 14). CONCLUSION: CPAx is a valid, reliable, and feasible tool to assess the physical functional state in COVID-19 patients following discharge from the ICU.