Predicting Patient Adherence in Healthcare using Artificial Intelligence and Machine Learning Techniques: A Narrative Review

Patient adherence to prescribed treatment regimens is crucial for achieving optimal health outcomes, particularly in managing chronic diseases such as heart failure (HF) and hepatitis C virus (HCV) infections. Non-adherence can lead to worsening of the disease, increased hospitalizations, and higher healthcare costs. This narrative review explores the use of artificial intelligence (AI) and machine learning (ML) techniques in predicting and improving patient adherence. Key studies demonstrate the effectiveness of these technologies in enhancing adherence in HF and HCV patients. For instance, one study used various classification algorithms, including Random Forest and Support Vector Machines, achieving up to 91% accuracy in predicting medication adherence in HF patients. Another study evaluated the AiCure platform, which uses AI for visual confirmation of medication ingestion, reporting a mean adherence rate of 91.3% and sustained virologic response in 88.2% of HCV patients. The review highlights the methodologies used in these studies, such as feature selection and handling imbalanced datasets, and addresses technical, clinical, and ethical challenges. Future research should focus on advancing AI and ML techniques, integrating new data sources, and developing supportive policies to ensure the successful deployment of these technologies in healthcare. By harnessing the power of AI, we can improve patient outcomes, reduce healthcare costs, and move closer to achieving optimal health for all.