Improving Medication Prescription Strategies for Discordant Chronic Comorbidities Through Medical Data Bench-Marking and Recommender Systems

Patients with Discordant Chronic Comorbidities (DCCs) face the challenge of managing evolving treatment needs and conflicting regimens, necessitating multiple medical appointments. The complexity arises from conflicting treatment plans and potential drug interactions, making it difficult for both patients and healthcare providers to set an optimal treatment plan. To address this, Machine Learning-powered tools offer significant solution by providing treatment recommendations and detecting potential interactions. Despite the potential, there is a lack of comprehensive documentation on how these algorithms effectively handle conflicting recommendations and evolving patient needs in the context of DCCs. This research seeks explore and improve the how Support Vector Machines (SVM), Neural Networks, and Random Forests algorithms set and prioritize treatment regimens for DCCs. We conducted survey to understand healthcare providers’ strategies for prioritizing medication regimens, we constructed a dataset from survey data and deployed machine learning algorithms to predict optimal medication regimens for DCC patients with specific treatment concerns. Following benchmarking, SVM emerged as the top performer, achieving an impressive accuracy rate of 0.95.