ARTIFICIAL INTELLIGENCE IN PREVENTING DRUG INTERACTIONS: TOOLS FOR CLINICAL DECISION SUPPORT

Abstract

Introduction: Polypharmacy is an increasing challenge in clinical practice, particularly among older adults and patients with chronic diseases. The concomitant use of multiple medications raises the risk of drug interactions, compromising patient safety. In this context, digital technologies and artificial intelligence (AI) emerge as promising tools to support healthcare professionals in preventing such interactions. Objective: To assess the role of artificial intelligence as a clinical decision support tool in preventing drug interactions, identifying emerging technologies, benefits, and challenges for clinical implementation. Methodology: A bibliographic review was conducted between August 2024 and May 2025, using the BVS, SciELO, LILACS, and PubMed databases. The descriptors “drug interactions,” “polypharmacy,” “artificial intelligence,” “patient safety,” “clinical decision support systems,” and “health technology” were applied in Portuguese and English. Results and discussion: The reviewed studies highlight that AI-based clinical decision support systems demonstrate strong capacity to identify drug interactions with greater accuracy and speed. Machine learning algorithms and expert systems analyze clinical data and generate real-time, personalized alerts. Hybrid models, combining predefined rules with continuous learning, further enhance detection effectiveness. Reported benefits include reduced prescribing errors, fewer adverse events, optimized clinical workflow, and personalized therapy. However, challenges remain regarding algorithm interpretability, integration with existing systems, and clinical validation. Conclusion: Artificial intelligence shows transformative potential in preventing drug interactions and strengthening patient safety. Developing systems that balance technical accuracy with clinical usability represents a major step forward in digital health. Effective implementation requires a multidisciplinary approach, addressing technical, ethical, and organizational aspects.