Agreement and reliability of a symptom modification test cluster for patients with subacromial pain syndrome.

OBJECTIVES: To identify if a plausible theoretical construct exists for a test cluster in a group of patients with subacromial pain syndrome presenting with glenohumeral internal rotation deficit (GIRD); determine the intertester agreement and reliability of the proposed test cluster; determine if there are any meaningful relationships between the test cluster and the component tests for the entire sample; and determine if there are any differences in disability on the Dutch version of the shoulder pain and disability index between participants with a positive and negative test cluster. METHODS: This study is a retrospective secondary analysis of data that were collected to determine the interrater agreement and reliability of clinical tests for assessment of patients with shoulder pain in primary care. RESULTS: The test cluster total agreement and negative specific agreement was 87.8 and 90.4%, respectively. The prevalence-adjusted bias-adjusted kappa for the test cluster was substantial at 0.76. There were statistically significant meaningful relationships (>/=0.50) between GIRD and the test cluster for Tester A (Phi = 0.71, p < .01) and Tester B (Phi = 0.82, p < .01). No differences in disability were identified between those with a positive and negative test cluster. CONCLUSION: The test cluster described in this study may be a reliable means of identifying a subgroup of patients with subacromial pain syndrome related to GIRD. Future research should look to validate this test cluster prospectively.