RESUMEN
Objective To determine the intra-and inter-tester reliability of a dynamometer in assessment of knee joint position senses (JPS).Methods Sixteen healthy volunteers (intra-rater N =6,inter-rater N =10) were examined with regard to intra-and inter-tester reliability of knee JPS by using Biodex System 3 dynamometer,measured by passive replication test,with subjects passively returning to the initial positions at 30°,45° and 60° knee flexion.The absolute error (AE) angle was recorded.Both intra-and inter-tester reliability studies involved two testers.Results The intra-tester reliability of the AE were moderate reliable at 30°,45°and 60° knee flexion,and the values of intra-class correlation coefficients (ICC) were 0.728,0.780 and 0.807,respectively.The values of Pearson's correlation coefficients for inter-tester reliability were 0.676,0.610 and 0.705,respectively (P ＜ 0.05).There was no significant difference between the AE values of dominance and non-dominant knees,test and retest,and in the AE values of 30°,45°and 60° knee flexion (P ＞ 0.05).The 95％ confidence upper limits of all subjects' AE values were less than 3°.Conclusion The major findings of this study suggest that the test-retest reliability of the knee JPS were moderate using Biodex System 3 dynamometer,and selecting a target angle of passive replication test can achieve the same purpose in 30° to 60°knee flexion.The clinical significance should be taken into consideration when the AE changes were greater than 3 degrees.