ABSTRACT
Background: Machine learning algorithms for predicting 30-day stroke readmission are rarely discussed. The aims of this study were to identify significant predictors of 30-day readmission after stroke and to compare prediction accuracy and area under the receiver operating characteristic (AUROC) curve in five models: artificial neural network (ANN), K nearest neighbor (KNN), random forest (RF), support vector machine (SVM), naive Bayes classifier (NBC), and Cox regression (COX) models. Methods: The subjects of this prospective cohort study were 1,476 patients with a history of admission for stroke to one of six hospitals between March, 2014, and September, 2019. A training dataset (n = 1,033) was used for model development, and a testing dataset (n = 443) was used for internal validation. Another 167 patients with stroke recruited from October, to December, 2019, were enrolled in the dataset for external validation. A feature importance analysis was also performed to identify the significance of the selected input variables. Results: For predicting 30-day readmission after stroke, the ANN model had significantly (P < 0.001) higher performance indices compared to the other models. According to the ANN model results, the best predictor of 30-day readmission was PAC followed by nasogastric tube insertion and stroke type (P < 0.05). Using a machine learning ANN model to obtain an accurate estimate of 30-day readmission for stroke and to identify risk factors may improve the precision and efficacy of management for these patients. Conclusion: Using a machine-learning ANN model to obtain an accurate estimate of 30-day readmission for stroke and to identify risk factors may improve the precision and efficacy of management for these patients. For stroke patients who are candidates for PAC rehabilitation, these predictors have practical applications in educating patients in the expected course of recovery and health outcomes.
ABSTRACT
Few studies have compared how rehabilitative post-acute care affects recovery of walking ability and other functions after stroke in different age groups. After propensity score matching (1:1), 316 stroke patients were separated into an aged group (age ≥65 years, n=158) and a non-aged group (age <65 years, n=158). Both groups significantly improved in Barthel index, EuroQol-5 dimension, Berg balance scale, 6-minute walking distance and 5-meter walking speed (P<0.001). The non-aged group had significantly larger improvements in Berg balance scale, instrumental activities of daily living, EuroQol-5 dimension and 6-minute walking distance (P<0.001) compared to the aged group. The two groups did not significantly differ in Barthel index, 5-meter walking speed, length of stay, and cost. The aged group had poorer walking ability and poorer instrumental activities of daily living compared to the non-aged group. After intensive rehabilitative post-acute care, however, the aged group improved in walking ability, functional performance and mental health. Intensive strength training for unaffected lower limbs in the stroke patients achieved good recovery of walking ability and other functions. Overall, intensive rehabilitative post-acute care improved self-care ability and decreased informal care costs. Rehabilitative PAC under per-diem reimbursement is efficient and economical for stroke patients in an aging society.
