Abstract
This study assessed the test-retest reliability of a protocol designed to measure isokinetic average torque, peak torque, and peak torque angle. The measurements were obtained during concentric shoulder and elbow flexion and extension and concentric and eccentric shoulder adduction, at 60 and 120 deg/sec, using the Kin-Com (II) dynamometer. The test and retest of 30 (18 men and 12 women) healthy subjects (mean age = 25 yr) were separated by exactly 1 week. The results demonstrated that the protocol can be used to measure average torque [intraclass correlation coefficient (ICC) range 0.921–0.982] and peak torque (ICC range 0.916–0.980) with high reliability at the two angular velocities. Peak torque angle reliability results (ICC range 0.019–0.754) were considered unacceptable, suggesting that angle of peak torque is not a reliable measure of muscle performance. This protocol has advantages for both clinical and research application in that it is time efficient, can be conducted by one examiner, and uses positioning methods suitable for both able-bodied and lower limb-disabled subjects.