Predictive Sub-items and Cutoff Scores of the Functional Assessment for Control of Trunk (FACT) for Gait Independence in Patients with Acute Stroke

抄録

Objectives: In this study, we aimed to identify the predictive subitems and cutoff values of the Functional Assessment for Control of the Trunk (FACT) for independent gait in patients with acute stroke. Methods: This retrospective observational study included 110 stroke patients. The FACT was assessed early after mobilization. Logistic regression analysis was conducted with independent gait at discharge as the dependent variable and 11 independent variables, including FACT subitems 3–10 and potential confounding factors. Receiver-operating characteristic (ROC) analysis was used to determine cutoff values for the total FACT score in predicting independent gait. Results: For level-ground gait, FACT items 4 and 5 (odds ratios: 4.71 and 2.89, respectively) were significant predictors. For uneven-ground gait, FACT item 8 (odds ratio: 1.87) was significant. The cutoff values were 11 points (sensitivity: 84.0%; specificity: 93.3%; area under the curve [AUC]: 0.95) for level-ground gait and 15 points (sensitivity: 77.8%; specificity: 95.0%; AUC: 0.89) for uneven-ground gait. Conclusions: Early FACT scores predicted independent gait. For level-ground gait, coordinated movement of the lower limbs and trunk during forward and lateral weight shifts and extensive lateral weight-shifting ability (items 4 and 5) were key. For uneven-ground gait, the rotational ability of the bilateral lower trunk during extensive lateral weight shifts (item 8) was crucial. The present findings suggest that these predictors and cutoff values can assist in guiding the prioritization of trunk assessment and devising tailored intervention plans aimed at functional gait acquisition during the early physical therapy phase following stroke onset.