ORIGINAL RESEARCH article
Front. Pharmacol.
Sec. Drugs Outcomes Research and Policies
Volume 16 - 2025 | doi: 10.3389/fphar.2025.1614002
This article is part of the Research TopicMathematical Modeling of Medication NonadherenceView all 4 articles
Mathematical Modeling of Medication Nonadherence in Multimodal Learning Systems
Provisionally accepted
- 1Southeast University, Nanjing, China
- 2Nanjing University of Chinese Medicine, Nanjing, Jiangsu Province, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Medication nonadherence is a widespread challenge driven by behavioral, socioeconomic, and structural factors, imposing high costs on healthcare systems. Traditional methods of improving adherence, including manual reminders and behavioral interventions, have proven ineffective in addressing the complex and multifactorial nature of this issue. The growing integration of machine learning and multimodal learning systems presents a promising approach to better understanding and mitigating medication nonadherence. In this paper, we propose the Personalized Adherence Optimization Model(PAOM) and Adaptive Personalized Intervention System (APIS) that incorporate predictive analytics and dynamic intervention adaptation to optimize medication adherence—defined as the degree to which a patient's actual medication use aligns with prescribed instructions—in personalized settings. The model integrates patient-specific features such as medical history, behavior, and demographics with real-time data to predict adherence likelihood, while also tailoring interventions based on predicted adherence scores. Using machine learning algorithms, we demonstrate the potential for significant improvement in adherence outcomes by adapting interventions dynamically to individual patient needs and barriers. Our approach outperforms traditional models by continuously refining intervention strategies based on evolving patient data, providing a personalized, data-driven solution to medication nonadherence. Experimental results show that our model not predicts adherence more accurately but also significantly improves adherence over time compared to static interventions. This research contributes to the growing body of work on enhancing medication adherence through advanced mathematical modeling, offering a scalable and effective solution to a pervasive healthcare problem.
Keywords: Medication Adherence, machine learning, Multimodal Learning Systems, predictive analytics, Dynamic intervention
Received: 18 Apr 2025; Accepted: 02 Sep 2025.
Copyright: © 2025 Chang, Qin and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Jianming Chang, Southeast University, Nanjing, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.