Prediction methods to account for the effect of gait speed on lower limb angular kinematics.

Altering gait speed appears to have an effect on lower limb kinematics during gait. This is a potential source of inaccuracy when data recorded from healthy volunteers are compared with those of slow-walking patients. The aim of this study was to assess the accuracy of several prediction methods which could be used to predict appropriate kinematic data obtained at various walking speeds. Seventeen healthy adults were video-recorded while walking at three different speeds (slow, normal and fast). Five lower limb angles (knee, ankle, thigh, shank and foot) were assessed at 26 points in the gait cycle. Significant correlations were observed between angles and gait speed. The relationship between gait speed and angles was also dependent on the point in the gait cycle that was used for comparison. Results indicate that using prediction methods based on group mean data or regression data from the same speed range on which the prediction methods are being applied is significantly more accurate than prediction methods based on other speed ranges. These results indicate that 'ideal' angle data for slow-walking speeds should be based on data recorded from slow-walking healthy adults or regressed using suitable regression equations.