Abstract
Background
Concurrent evaluation of outcome measures is necessary to determine the most suitable measure(s) for a particular population. This property was evaluated in patients with rheumatic diseases undergoing shoulder surgery.
Methods
Three patient-reported measures (the Disabilities of the Arm, Shoulder and Hand Score; the Oxford Score; and the Shoulder Pain and Disability Index); three measures combining patient-reported and clinical outcomes (the Bostrom Shoulder Impairment Scale; the Constant Shoulder Score; and the Shoulder Function Assessment Scale); and one clinical measure (Range of Motion) were included. Responsiveness was expressed by the Standardized Response Mean (SRM).
Results
The study included 123 patients: 54 patients with shoulder arthroplasty (mean age 65.3 years) and 69 patients with non-arthroplasty (mean age 58.9 years). At 1 year, the Shoulder Function Assessment Scale was the most responsive measure in the arthroplasty group with SRM = 2.03 [95% confidence interval (CI) = 1.38 to 2.66] and Range of Motion (passive flexion) the least responsive measure with SRM = 0.54 (95% CI = 0.14 to 0.92). In the non-arthroplasty group, the Constant was the most responsive measure with SRM = 1.35 (95% CI = 0.87 to 1.83) and Range of Motion (passive flexion) the least responsive measure with SRM = 0.53 (95% CI = 0.16 to 0.90).
Conclusion
Measures comprising patient-reported outcomes were more responsive than clinical measures, and should be included when evaluating shoulder surgery in patients with rheumatic diseases.
Get full access to this article
View all access options for this article.
