Measurement Properties of Commonly Used Outcome Measures in Patients With Glenohumeral Instability

Purpose: The objective of this study was to compare the reliability and validity of the Western Ontario Shoulder Instability Index (WOSI), a disease-specific outcome measure, with two commonly used joint-specific measures, the American Shoulder and Elbow Surgery (ASES) score and the Constant Murley score (CMS), in patients who have undergone Bankart repair or capsular plication. Method: This was a secondary analysis of prospectively collected data. Cross-sectional validity involved internal consistency as part of reliability assessment and factorial validity, cross-sectional convergent, and known-group validity. Longitudinal validity involved assessment of sensitivity to change in patients with pre- and post-operative data. Results: The data of 206 patients were used for cross-sectional analysis and the data of 140 patients were used for longitudinal validity. Factor analysis of the WOSI produced three domains, with some questions being eliminated or loaded in a different domain. The Cronbach coefficient alphas for total WOSI were 0.91 and 0.82, 0.86, 0.70, and 0.87 for its respective four domains. There was a positive moderate association between WOSI and ASES ( r = 0.60) and CMS ( r = 0.49). All measures were able to differentiate between patients with different levels of pathology, with the ASES and CMS being superior to WOSI. Standardized response mean values were 1.07, 0.85, and 0.84, for WOSI, ASES, and CMS, respectively. Conclusions: The WOSI, ASES, and CMS had comparable convergent and known-group validity and sensitivity to change in patients with glenohumeral instability. The four-domain theory and clear distinction of each construct of the WOSI was not supported.