The relative and absolute reliability of leg muscle strength testing by a handheld dynamometer.

To examine the relative and absolute interrater reliability of handheld dynamometers (HHD) for assessing the lower extremity muscle strength, maximal voluntary contractions (MVCs) of 16 young adults for bilateral hip and knee muscles were tested using the break method. Three MVCs of each muscle group were required for obtaining the muscle strength. Participants' muscle strengths were tested by 2 raters. The intraclass correlation coefficients (ICCs) and the smallest real differences (SRD) were used to examine the relative and absolute reliabilities, and the Bland-Altman analyses were used to check whether systematic bias exists. The results showed that the relative reliabilities of all muscle groups were excellent (ICCs = 0.83-0.92) except for the knee extensors (ICC = 0.60). The SRD represents the smallest difference that indicates a real change for a single subject. The SRD% of all muscle groups was acceptable (ranging from 8.4 to 22.8 %), with the hip extensors being the smallest and knee extensors being the largest. The reliability of the knee extensors was unsatisfactory because of poor relative and absolute reliabilities and systematic bias. In addition to assessing the relative reliability in strength measurement, the absolute reliability provides the data of the measurement error, which is useful information in clinical practice to know whether the change in strength of a subject is real. Hand-held dynamometer is a reliable tool for quantifying most of the hip and knee strength except for the knee extensors. Modifying the measuring technique for knee extension is needed in future studies to improve the reliability.