Prediction of Gait Independence of Stroke Patients through Signal Detection Analysis in the Convalescent Period

Abstract

[Purpose] This study developed a chart for predicting stroke patients' independence of gait during the convalescent period. [Subjects and Methods] Two hundred fifty-one first-time stroke patients were enrolled in this study. Physical and cognitive functions, basic movement ability, balance ability and activities of daily living were evaluated on admission to a rehabilitation ward for the convalescent stage. Based on this information, a decision tree for predicting independence of gait three months later was developed using signal detection analysis. [Results] Independence in rising motion, balance ability, and cognitive function were extracted as significant predictors. A valid and reliable decision tree was developed. [Conclusion] A chart for predicting independence of gait was developed. This chart is appropriate for the convalescent period after stroke.