Study Design: Systematic review.
Objective . To identify, evaluate, and compare validated outcome scales measuring dysfunction of the upper extremity.
Summary of Background Data. Dysfunction of the upper extremity can be measured by many different outcome scales. Due to the lack of a systematic, comparative review, choosing the appropriate scale represents a challenge to clinicians and researchers.
Methods. Articles including the psychometric evaluation of outcome scales measuring dysfunction of the upper extremity were identified by a database search using nine different on-line databases. Additional references were obtained from lists of bibliographic references, contact with experts, and journal hand searches. All original scales were tracked with Citation Index to identify studies performing a re-validation of the original scale. Data was extracted in a systematic fashion to account for validity, reliability, responsiveness, factor analysis, and alternative methods of psychometric validation.
Results. Of the 1,753 articles initially identified, 22 scales were found to measure dysfunction of the shoulder (n=8), elbow (n=2), forearm (n=1), hand/wrist (n=6), and the whole upper extremity (n=6). Factor analysis was only performed in 5 scales to determine if dysfunction was the only measured construct. Of importance, three scales associated pain and dysfunction in the same scale without generating separate scores. All scales presented have been tested for reliability and validity, but only 11 were evaluated for responsiveness to change. In a comparative analysis, the Disabilities of the Arm Shoulder and Hand (DASH) and the Shoulder Pain and Disability Index (SPADI) had the most extensive psychometric property validation across different patient populations and are recommended for upper extremity and shoulder conditions, respectively.
Conclusions. Scales for measurement of dysfunction of the upper extremity have been validated with a wide variety of methodological quality. Frequent problems in the validation were found, thus claiming for a standardization of the methodology to validate outcome scales. The final choice of a scale should be tailored according to the target population and the purpose of evaluation.