Abstract
BACKGROUND: This study aimed to develop alternative prediction equations to predict isokinetic muscle strength at 60°/s based on anthropometric characteristics, including body mass, height, age, and sex for young and middle-aged non-athlete populations. METHODS: Three hundred and thirty-two healthy non-athletic participants (174 females, 158 males) between 20 and 59 years underwent a 60°/s isokinetic knee joint concentric contraction test. Forty people were randomly selected for retesting to assess the reliability of the isokinetic instrument. Multivariate linear regression was used to establish extension peak torque (EPT) and flexion peak torque (FPT) prediction equations. Sixty extra participants were used individually to validate the prediction equations, and Bland Altman plots were constructed to assess the agreement of predicted values with actual measurements. RESULTS: The result demonstrated that the instrument we used has excellent reliability. The multivariable linear regression model showed that body mass, age, and sex were significant predictors of PT (EPT: Adjusted R(2) = 0.804, p < 0.001; FPT: Adjusted R(2) = 0.705, p < 0.001). Furthermore, the equations we established had higher prediction accuracy than those of Gross et al. and Harbo et al. CONCLUSION: The equations developed in this study provided relatively low bias, thus providing a more suitable reference value for the knee isokinetic strength of young and middle-aged non-athletes.