Random Forest-Based Classification and Identification of Major Risk Factors for Comorbidities in Type 2 Diabetes Mellitus

Type 2 Diabetes Mellitus (T2DM) is a chronic condition frequently accompanied by comorbidities such as hypertension, obesity, and dyslipidemia, increasing the risk of severe complications. Early identification of high-risk patients is essential to improve treatment outcomes and guide preventive interventions. This research presents a Random Forest (RF)-based classification approach to distinguish T2DM patients with and without comorbidities, while also identifying key clinical predictors associated with comorbidity risk. Using real-world hospital data from AlIhsan Hospital, Bandung, the model was evaluated across three train-test splits (70:30, 80:20, and 90:10), with the 80:20 split yielding the highest validation accuracy of 89% and a recall of 100% for comorbid cases. The feature importance analysis revealed HbA1c, fasting blood glucose (FBG), and hematological parameters such as hemoglobin and hematocrit as the most influential factors for classifying comorbidities. This RF-based model demonstrates strong predictive performance and interpretability, making it a valuable tool for early risk detection and supporting personalized clinical decision-making in the management of T2DM patients with comorbidities.