Abstract
We can experience a large difference in the value of first and second measurements in clinical measurements when evaluating joint position sense. This can bring about a misinterpretation of the true value of joint position sense and it is often thought that reliability of the measurement value is low. This study was designed to examine the number of measurements that are necessary to guarantee a high reliability for measuring joint position sense. Using Barrett's method, we measured the joint position sense at three different set angles for each knee of 32 healthy elderly subjects (i.e. 64 knees). The set angles were set at 10°, 40° and 70°, and measurements were taken as part of our objective at two angle speeds approximately set at 10°/sec and 70°/sec. First, the coefficient of reliability at each set angle was obtained from the measurement values gained through a number of measurements undertaken. Second, we set the coefficient of reliability at 0.81 according to our objectives and we substituted the Spearman-Brown formula for the smallest value of coefficient of reliability that was obtained. We measured the number of times the values that were given due to application of the substitution to obtain the reliability temporarily and subsequently confirmed the angle measurements. As a result, it was understood that in order to guarantee high reliability, it is necessary to use the average values of 3 separate measurements. From this we can assume that it is unreliable to make judgments whether abnormalities exist on just one measurement value and it is therefore necessary to use the average value of three measurements in order to guarantee high reliability for measurements of joint position sense. We consider that it will be necessary to adopt these numbers of measurements for the elderly and for the handicapped in general.
