Palm-up test and range of motion in flexion and external rotation provide best correlation with disability and perceived pain in patients with shoulder complaints.

The aim of this study was to find a reliable set of clinical tests to predict pain and disability in patients with shoulder pain. Shoulder pain is the second most frequent musculoskeletal complaint in the primary care setting and has a great impact in work and leisure activities. Patient reported outcomes measuring pain and disability are available, but they are time-consuming, often biased by psychological and sociological factors and depend on patient collaboration. This was an observational, cross-sectional study, including patients with shoulder pain aged 18 to 70 years. Patients filled in the questionnaires Disabilities of the Arm, Shoulder and Hand outcome measure (DASH) and Shoulder Pain and Disability Index (SPADI) pain scale. A Visual Analogue Scale (VAS) for current pain was applied and sociodemographic and clinical data were collected. Physical examination included the Jobe, Neer and palm-up signs, range of motion (ROM) of shoulder abduction, flexion and rotation (internal and external), both active and passive. An independent t-test to compare differences between groups and the Spearman's coefficient for evaluation of bivariate correlation were used. Linear regression analysis was applied to relevant predictors. Tests were two-tailed and p values < 0.05 were considered significant. A total of 127 patients were included. Female patients and those with a positive Jobe, Neer or palm-up tests, complaints on dominant side, no leisure activity involving shoulder effort, a history of previous shoulder tendinopathy or taking analgesics had significantly higher DASH scores. Age and all range of motion variables significantly correlated with DASH scores. A linear regression model with six dependent variables (Palm-up test, range of motion in active flexion and external rotation, age, gender and complaints on dominant side) produced the highest correlation (R = 0.665), explaining 44% of the variance of DASH score. A model based on few physical examination items and individual objective data like age, gender and dominancy, can help predict disability and perceived pain in shoulder disorders. Palm-up test and range of motion in active flexion, abduction and external rotation showed best correlation with the outcome, but abduction was found redundant for the obtained prediction model.