Reliability of maximal static strength measurements of the arms in subjects with hemiparesis

Objective: To evaluate the reliability of maximal static strength measurements of five arm muscle groups and of strength ratios (paretic/non-paretic) in subjects with poststroke hemiparesis. Design: The generalizability theory was used to estimate the reliability coefficients and standard errors of measurement of maximal strength for various combinations of trials and sessions and of the strength ratios for one and two sessions. Grip maximal voluntary force and/or maximal voluntary torques exerted by flexor and extensor muscles at both the elbow and shoulder joints were measured in 17 subjects with poststroke hemiparesis. Multiple trials were performed by subjects during two sessions. S etting: Rehabilitation centre. Subjects: A convenience sample of 17 subjects with poststroke hemiparesis. Results: The reliability coefficients for the strength measurements were in the range of 0.81-0.97 with standard errors of measurement accounting for 4% to 20% of the group means. For the strength ratios, the coefficients of generalizability ranged from 0.76 to 0.95 with standard errors of measurement equal to 6% to 19% of the group means. Conclusions: The maximal strength measurements of the arms in subjects with hemiparesis are reliable. The strength ratios are also reliable and can be used to quantify strength impairment.