OBJECTIVE: To assess the measurement properties of a simple index of symptom severity in osteoarthritis (OA) of the hips and knees. METHODS: Both the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the proposed new Comprehensive Osteoarthritis Test (COAT) instrument were completed weekly by 125 subjects in the context of a randomized, 12-week, 3 parallel-arm clinical trial. The reliabilities of the various scales were assessed on a weekly basis by use of Cronbach's alpha coefficients. The validity of the COAT total scale was assessed by correlation with the WOMAC total scale on a weekly basis with correlation coefficients, and in terms of the correlations between subject-level intercepts and slopes over time. The relative responsiveness of the WOMAC and COAT total scales was assessed using a multilevel (longitudinal) multivariate (WOMAC, COAT) linear model. RESULTS: The WOMAC and COAT total scales were highly reliable (mean over weeks: WOMAC alpha = 0.98; COAT alpha = 0.97). The correlations between the WOMAC and COAT scales were very high (mean over weeks = 0.92; subject-level intercepts = 0.91, slopes = 0.88). The COAT total scale was significantly more responsive than the WOMAC total scale in the active treatment (34.8% improvement vs 26.8%; p = 0.002). CONCLUSION: The COAT total scale is simple to administer, reliable, valid, and responsive to treatment effects.
RCT Entities:
OBJECTIVE: To assess the measurement properties of a simple index of symptom severity in osteoarthritis (OA) of the hips and knees. METHODS: Both the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the proposed new Comprehensive Osteoarthritis Test (COAT) instrument were completed weekly by 125 subjects in the context of a randomized, 12-week, 3 parallel-arm clinical trial. The reliabilities of the various scales were assessed on a weekly basis by use of Cronbach's alpha coefficients. The validity of the COAT total scale was assessed by correlation with the WOMAC total scale on a weekly basis with correlation coefficients, and in terms of the correlations between subject-level intercepts and slopes over time. The relative responsiveness of the WOMAC and COAT total scales was assessed using a multilevel (longitudinal) multivariate (WOMAC, COAT) linear model. RESULTS: The WOMAC and COAT total scales were highly reliable (mean over weeks: WOMAC alpha = 0.98; COAT alpha = 0.97). The correlations between the WOMAC and COAT scales were very high (mean over weeks = 0.92; subject-level intercepts = 0.91, slopes = 0.88). The COAT total scale was significantly more responsive than the WOMAC total scale in the active treatment (34.8% improvement vs 26.8%; p = 0.002). CONCLUSION: The COAT total scale is simple to administer, reliable, valid, and responsive to treatment effects.
Authors: Stephen P Myers; Joan O'Connor; J Helen Fitton; Lyndon Brooks; Margaret Rolfe; Paul Connellan; Hans Wohlmuth; Phil A Cheras; Carol Morris Journal: Biologics Date: 2010-03-24
Authors: Stephen P Myers; Ann M Mulder; Don G Baker; Shelley R Robinson; Margaret I Rolfe; Lyndon Brooks; J Helen Fitton Journal: Biologics Date: 2016-05-26