抄録
Although various methods have been proposed for quantitative evaluation of gait, they are not yet widely used as routine clinical tools. The reasons are: 1) measuring systems are too expensive and complicated, and 2) the validity of the methods is not fully examined in a statistical manner.
In this paper, the validity of a new evaluation method based on a simple foot-force measuring device is investigated in terms of the correlation between quantitative indices and visually-rated grades.
Fourty-eight hemiplegic patients due to cerebral vascular accident were asked to walk on a 15m straight level floor. Four raters independently evaluated each subject in 8 gait characteristics or items in 1-4 grades on the basis of visual inspection. These items include asymmetry, variability, instability, and dependence on cane. At the same time, vertical ground-reaction-force exerted on the right and left foot and on the cane was measured, and then quantitative indices representing the above 8 items were derived.
All but one indices show reasonably good correlation with the mean grades of the 4 raters in corresponding items, thus demonstrating the feasibility of these indices. Statistical analysis also reveals several problems inherent in visual rating, such as inconsistency in the interpretation of evaluation criteria among raters, and degradation of independence of items due to rater's subjectivity.
