Responsiveness and Minimal Important Change of the Arabic Disabilities of the Arm, Shoulder and Hand (DASH) in Patients with Upper Extremity Musculoskeletal Disorders

The aim of this study was to examine the responsiveness of the Arabic Disabilities of the Arm, Shoulder and Hand (DASH) and to quantify its minimal important change (MIC) for improvement. People with upper extremity musculoskeletal problems who were receiving physical therapy were evaluated at baseline and again during a follow-up appointment, with a median time frame of 7 days between the two testing sessions (range of 6 to 72 days). The participants completed the Arabic DASH, Global Assessment of Function (GAF), Numeric Pain Rating Scale (NPRS) and Global Rating of Change Scale (GRC). The responsiveness of the Arabic DASH was assessed by examining the pre-specified hypotheses. The MIC for improvement was determined using the receiver operating characteristic method (MIC(ROC)) and the predictive modeling method (MIC(pred)). As hypothesized, a change in the Arabic DASH demonstrated a significant positive correlation with changes in the GAF (r = 0.69), NPRS (r = 0.68) and GRC (r = 0.73). Consistent with our hypotheses, the DASH change scores could be used to differentiate between participants who improved and those who did not improve (area under the receiver operating characteristic curve = 0.87), and they showed a large magnitude of change (effect size = 1.53, standardized response mean = 1.42) in patients who improved. All the hypotheses specified a priori were supported by the results. The Arabic DASH MIC(ROC) and MIC(pred) were estimated to be 14.22 and 14.85. The interaction between the DASH change and baseline score was not a significant predictor of status (improved vs. not improved) (p = 0.75), indicating that the DASH MIC was not baseline-dependent. The Arabic DASH demonstrated sufficient responsiveness, supporting the idea that the Arabic DASH is capable of detecting changes in upper extremity function over time. The value of the Arabic DASH MIC was similar when estimated using the predictive modeling and ROC methods, and the MIC was not dependent on baseline status.