Interrater reliability of isokinetic measures of knee flexion and extension

The purpose of this investigation was to determine the interrater reliability of peak torque and total work values obtained with isokinetic measures of knee flexion and extension. Eight male and eight female students were evaluated on four occasions by four different examiners (range of isokinetic test experience: 0 to 10 yrs) using a standardized isokinetic measurement protocol. Subjects were randomly assigned to participate in a test sequence determined by a 4 x 4 balanced Latin square. Peak torque and total work values at 60 degrees /sec and 180 degrees /sec were obtained for the concentric measures of knee extension and flexion. The measures of peak torque and total work were corrected for the effects of gravity. Intraclass correlation coefficients and standard error of measurement estimates were used to estimate the interrater reliability for each test condition (test speed x muscle group). Intraclass correlation coefficient values ranged from .90 to .96 for peak torque and .90 to .95 for total work. Standard error of measurement estimates ranged from 8.9 to 13.3 Nm for peak torque and 11.3 to 16.8 Nm for total work. The results of this investigation demonstrate that reliable measures of isokinetic muscle performance of knee extension and flexion may be obtained by four clinicians with varied experience when following a standardized measurement protocol.