Artificial Intelligence-Enabled Adaptive Patient Education in Pharmacy: A Workflow-Centered Model For Patient Engagement, Medication Adherence, and Clinical Outcomes

Abstract

Non-adherence to medicines is a widespread health problem worldwide and is a major contributor to poor clinical outcomes and to health care costs when managing chronic non-communicable diseases. Although effective, traditional pharmacist-led interventions are often limited by significant workflow challenges, staffing issues and the static nature of traditional teaching materials. However, artificial intelligence offers a sophisticated, data-driven alternative to optimize these traditional strategies in a proactive, personalised and highly scalable way in medication management. The present study aims to examine the possibility of implementing an adaptive patient education platform integrated into the clinical pharmacy workflow that could increase patient engagement, adherence, and improve specific clinical parameters using AI technology. A twelve-month-long longitudinal randomized controlled trial was carried out in adults with both type 2 diabetes and hypertension. The intervention was based on the use of natural language processing and machine learning algorithms for tailored learning content based on the Patient Activation Measure. Analysis of the results showed that the use of AI was statistically significantly associated with higher Patient Activation Measure (PAM) scores and greater Proportion of Days Covered (PDC) when compared with standard care. In addition, significant reductions in glycated hemoglobin and SBP measures were seen. In conclusion, the use of AI models that are workflow oriented can significantly complement current clinical practice in pharmacies, with the aim of fostering ongoing behavioral change and improved clinical results.