Abstract
Aim: We aimed to determine whether the Trail Making Test Part A (TMT-A) is a useful tool for predicting home discharge in patients with acute cerebral infarction.Methods: This was a retrospective observational study. The subjects were acute stroke patients. Logistic regression analysis was used to estimate the influence of the TMT-A on determining the home discharge. We created three models to predict home discharge. Age, NIHSS, BRS of the upper extremity, and number of people living with the patient in addition the MMSE-J was used for Model 1, the TMT-A for Model 2, and both for Model 3 as the cognitive assessment. NRI and IDI analyses were conducted to compare the predictive accuracy of each model for home discharge.Results: The analysis included 164 patients, and 119 were discharged home. Logistic regression analysis showed that the TMT-A influenced home discharge in a model that includes the TMT-A (Model 2: p = 0.009, Model 3: p = 0.024). NRI did not lead to an improvement in home discharge predictive accuracy. On the other hand, IDI showed that models containing TMT-A (Model 2 & 3)has significantly improved the prediction accuracy than a model that does not contain TMT-A (Model 1 vs Model 2: IDI = 0.034, p = 0.041; Model 1 vs Model 3: IDI = 0.033, p = 0.034).Conclusions: TMT-A helps predict discharge home in patients with acute cerebral infarction. The addition of TMT-A to the model for predicting home discharge may lead to an improvement in the accuracy of predicting home discharge.
