The Shoulder Self-Administered Motion Evaluation Has Excellent Patient Reliability and Reproducibility on Both Physician and Repeat Follow-up Testing

Abstract

Long-term clinical evaluation is becoming more important, yet difficulties exist because there are substantial patient costs for extended follow-up. The goal of this study was to validate whether a new shoulder motion evaluation is reproducible and reliable with physician examination. Consecutive patients were administered a shoulder motion assessment during their clinic visit. Patients completed the same evaluation 2 weeks after their visit to determine test-retest reliability. Exact and approximate (within 20° or 4 spinal levels) agreements between patient and physician measurements were calculated. A total of 112 patients (224 shoulders) with an average age of 56.7 years were evaluated. The most common diagnoses were osteoarthritis (33.0%) and rotator cuff tear (32.1%). All motion questions had greater than 50% exact patient-physician agreement and greater than 70% approximate agreement. When agreement was off, patients more often underestimated their motion. There was substantial clinician-patient agreement for all questions and almost perfect agreement for forward elevation (intraclass correlation coefficient, 0.78) and internal rotation (intraclass correlation coefficient, 0.77). On test-retest reliability testing, patient-patient agreement was substantial for external rotation at the side (intraclass correlation coefficient, 0.71). There was near perfect agreement for internal rotation (intraclass correlation coefficient, 0.83) and abduction (intraclass correlation coefficient, 0.81) testing. This shoulder assessment tool had both good agreement to physician examination and substantial agreement on correlation testing. There was substantial agreement between follow-up visits, suggesting excellent reproducibility. This motion assessment shows effectiveness with most shoulder pathologies. It is recommended as a good screening tool for both clinical and research purposes requiring long-term follow-up in which in-person clinical examinations may be costly and time consuming. [Orthopedics. 2018; 41(6):e820-e826.].