A Pilot study on Prediction of Pouchitis in Ulcerative Colitis Patients by Decision Tree Method Versus Logistic Regression Analysis

Background: Pouchitis is a non-specific inflammation of the ileal reservoir and the most frequent complication that patients experience in long time periods. Diagnosis should be made on the basis of clinical, endoscopic, and histological aspects. Prediction of pouchitis is an important issue for the physician. Objectives: Identifying the predictive factors of pouchitis and their importance is the study’s objective. Patients and Methods: In the present study, two classifier techniques namely decision trees method and logistic regression analysis are used to help the physician for prediction of pouchitis in ulcerative colitis (UC) patients. These patients are submitted to a specific surgery. The ability of these two methods in prediction is achieved by comparison of the accuracy of the correct predictions (the minimum error rate) and the interpretability and simplification of the results for clinical experts. Results: The accuracy rate in prediction is 0.6 for logistic regression method and 0.45 for decision tree algorithm. In addition, the mean squared error is lower for logistic regression (0.41 versus 0.48). However, the area under the ROC is more for decision tree than logistic regression (0.52 and 0.45 respectively). Conclusions: The results are not in favor of none of these two methods. However, the simplicity of decision tree for clinical experts and theoretical assumptions of logistic regression method make the choice clear. But more sample size may be needed to choose the best model with more confident.