Data-Driven Strategies for Cardiovascular Medication Management: A Dual Approach Using Decision Tree Modeling and Process Mining

Effective management of cardiovascular medication is crucial for improving patient outcomes and optimizing treatment strategies. As healthcare systems increasingly rely on data-driven approaches, integrating advanced analytical techniques can enhance medication management. Despite advancements in cardiovascular care, challenges remain in determining the most effective medication for patients based on individual health metrics, highlighting the need for robust methodologies to analyze and interpret large datasets that can inform treatment decisions. This study aims to develop data-driven strategies for managing cardiovascular medication through two distinct techniques: decision tree modeling and process mining. The dataset utilized in this study includes various patient health metrics, such as blood pressure, sodium levels, cholesterol levels, age, and gender, collected from a diverse population. These features are essential for understanding the relationships between health metrics and medication outcomes. The study employed decision tree analysis (both normal and radial styles) to classify patient responses to medications based on health metrics, while process mining techniques were used to analyze age group variations in medication consumption and identify trends within the data. The findings reveal that blood pressure is the most significant predictor of medication response, followed by sodium and cholesterol levels. The results highlight distinct medication preferences across different age groups and blood pressure categories, emphasizing the need for tailored treatment strategies. Based on our results, the presence of high blood pressure and high cholesterol in individuals aged up to 20 years is concerning. The analysis also revealed commonalities between the decision tree and process mining results, particularly in identifying critical health metrics that influence medication management. Both methodologies underscored the importance of physiological factors over demographic variables in predicting medication efficacy. This study offers valuable insights for healthcare practitioners, enabling them to make informed decisions regarding cardiovascular medication management. By leveraging data-driven strategies, practitioners can enhance patient outcomes, reduce adverse effects, and optimize treatment protocols based on individual health metrics.