Correlation of subjective and objective measures before and after shoulder arthroplasty

The degree to which subjective patient-reported measures reflects objective findings or how well subjective and objective measures reflect patient satisfaction is not well established. The purpose of this study was to determine the correlation between such measures before and after shoulder arthroplasty. A group of 174 patients (93 total shoulder arthroplasty and 81 reverse shoulder arthroplasty) were prospectively evaluated pre- and postoperatively (mean follow-up, 49 months) with the following subjective measures: American Shoulder and Elbow Surgeons score, Simple Shoulder Test, Short Form 36 (SF-36) summary scores, and patient satisfaction. Objective measures included Biodex isometric strength and videotaped range of motion. The objective measures were combined to derive 1 number representative of the overall shoulder function. All measures improved from their preoperative statuses except the SF-36 physical component summary in patients undergoing revision and the SF-36 mental component summary in patients undergoing primary reverse shoulder arthroplasty. A patient satisfaction score of 5 or more was seen in 89% of patients. Preoperatively, a significant (P<.05) correlation existed between the American Shoulder and Elbow Surgeons and the Simple Shoulder Test (ρ=0.546), American Shoulder and Elbow Surgeons score and SF-36 physical component summary (ρ=0.407), and Simple Shoulder Test and SF-36 physical component summary (ρ=0.479). Objective measures had lower correlations (ρ<0.4) with subjective scores. Postoperatively, the correlation improved among all measures. Patient satisfaction correlated more with subjective than objective measures. Subjective measures had relatively low correlations with objective measures. Improvements in the current measures are necessary to provide evidence-based comparisons of the effectiveness of shoulder arthroplasty.