Methodological properties of six shoulder disability measures in patients with rheumatic diseases referred for shoulder surgery.

The aim of this study was to explore the methodological properties in 6 commonly used shoulder disability measures (Dash, Spadi, Oxford, the Constant score, Shoulder Function Assessment Scale, Bostrom's shoulder movement impairment scale) in patients with inflammatory or degenerative diseases referred for shoulder surgery. One-hundred and six patients completed the measures preoperatively. Fifty-five (51.9%) were not able to carry out the assessment of the strength component of the Constant score. Pearson correlation coefficients between the measures varied between 0.22 and 0.87. The lowest correlation coefficients were found between performance-based and self-report measures. All measures, except the Dash, were able to differentiate significantly between patients who were more and less severely affected. Performance-based measures differentiated better (effect size, 0.68- 0.87) than self-report measures (effect size, 0.21-0.61) between the 2 patient groups. Performance-based and self-report assessments provide complementary information about shoulder status and should not be used interchangeably.