AI Summary & Key Takeaways
AI InsightThe numerical results indicate that the model performed acceptably on the training dataset, with a sensitivity of 70.3% and an overall accuracy of 62.2%.
📌 3 Key Takeaways
Abstract
Patient attendance status (Show-ups and No-shows) is a significant problem that seriously impacts healthcare efficiency and resource planning. The primary objective of this study is to comprehensively evaluate the performance of the Classification and Regression Trees (CART) model, a highly interpretable machine learning algorithm, in predicting patient attendance. Four independent variables (Patient Age, Appointment Time, Gender, and Chronic Disease Status) were used to predict attendance status (show-up/no-show), and the CART model's hyperparameters were optimized using 10-fold cross-validation to select the optimal tree structure. The numerical results indicate that the model performed acceptably on the training dataset, with a sensitivity of 70.3% and an overall accuracy of 62.2%. The relative importance of the independent variables revealed that Appointment Time (AppTime) and Age were the strongest predictors, while Gender and Chronic Disease contributed significantly less to the model's predictions. In conclusion, the potential of the CART model in predicting patient absences has been demonstrated.