Introduction:
Professional medical errors represent one of the most critical challenges facing healthcare systems worldwide, contributing annually to millions of injuries, preventable deaths, and substantial economic burdens. According to reports by the World Health Organization, medical errors are among the leading causes of mortality and adverse events in hospital settings. Despite advances in clinical knowledge and medical technology, diagnostic, medication, and surgical errors remain prevalent, particularly in high-demand and resource-constrained healthcare environments. In Iran, hospitals affiliated with medical sciences universities—especially in less-developed regions such as Kermanshah—face compounded challenges including workforce shortages, high workload, inadequate technological infrastructure, and weak error-reporting cultures. These conditions increase the likelihood of human and system-based errors and limit the effectiveness of traditional risk management approaches.Recent advancements in artificial intelligence (AI) have created new opportunities for improving patient safety through real-time data analysis, predictive modeling, and clinical decision support. However, the adoption of AI in healthcare systems like Iran’s remains fragmented and largely experimental, lacking a comprehensive and context-sensitive strategic framework. This study addresses this gap by developing a strategic model for reducing professional medical errors through AI application in hospitals affiliated with Kermanshah University of Medical Sciences.
Theoretical Background and Research Gap The theoretical foundations of patient safety emphasize that medical errors are rarely the result of isolated individual failures; rather, they emerge from complex interactions among human, organizational, technological, and regulatory factors. Previous studies—both domestic and international—have documented the prevalence and causes of medical errors, highlighting issues such as fatigue, insufficient training, poor communication, and flawed organizational processes. While Iranian studies have predominantly focused on identifying error types and contributing factors, they have paid limited attention to advanced technological interventions.Conversely, international research underscores the potential of AI-driven systems—such as machine learning algorithms, clinical decision support systems (CDSS), and predictive analytics—to detect patterns of risk, reduce cognitive overload, and minimize preventable harm. Nevertheless, these studies are often conducted in technologically advanced healthcare systems and do not adequately consider contextual barriers such as legal ambiguity, cultural resistance, and infrastructural limitations. The primary research gap, therefore, lies in the absence of a localized, strategic, and integrated model that aligns AI capabilities with organizational, human, and regulatory realities in Iranian hospitals.
Methodology:
This study employed an applied qualitative research design using the grounded theory approach developed by Strauss and Corbin (1998). The research population included physicians, pharmacists, and managerial staff working in hospitals affiliated with Kermanshah University of Medical Sciences in 2025. Purposeful and snowball sampling techniques were applied, resulting in 15 participants (8 physicians, 4 pharmacists, and 3 managers) selected until theoretical saturation was achieved.Data were collected through an extensive literature review and in-depth semi-structured interviews lasting between 45 and 60 minutes. Interviews focused on experiences with medical errors, perceptions of AI capabilities, implementation barriers, and strategic requirements for error reduction. All interviews were transcribed verbatim and analyzed using MAXQDA 11 software. Data analysis followed three systematic stages: open coding, axial coding, and selective coding. Credibility was enhanced through member checking and data triangulation, while reliability was confirmed through inter-coder agreement, yielding a coefficient of 77%.
Findings:
The analysis resulted in the development of a paradigmatic model with the core phenomenon identified as “strategic reduction of professional medical errors through artificial intelligence.” Causal conditions included early-stage errors related to weak predictive mechanisms and late-stage errors associated with human mistakes and process failures. Contextual conditions encompassed managerial factors (risk management practices, resource allocation), cultural factors (error-reporting culture, acceptance of technology), and human factors (digital literacy and professional training).Intervening conditions were identified as inadequate supervision, weaknesses in executive structures, legal and ethical ambiguities, and technological complexity. In response to these conditions, several strategic actions emerged, including the development of AI-based infrastructure, systematic training of healthcare professionals, strategic policymaking, and operational interventions such as the implementation of CDSS, intelligent alert systems, and real-time monitoring tools.The anticipated outcomes of the model include a reduction in diagnostic and medication errors, enhanced patient safety, improved organizational efficiency, higher quality of care, and increased patient satisfaction. Importantly, participants emphasized that AI should function as a decision-support tool rather than a replacement for clinical judgment.
Conclusion:
The findings demonstrate that reducing medical errors requires a strategic, system-oriented approach rather than isolated technological solutions. The proposed model provides an integrated framework that enables hospital managers to leverage AI as a preventive and predictive mechanism for managing clinical risks. By aligning technological innovation with human capacity building, organizational culture, and regulatory support, AI can transform structural challenges into opportunities for healthcare improvement. Policy implications include the need for targeted investment in digital infrastructure, the establishment of clear legal and ethical guidelines, and continuous professional education in AI literacy. This study contributes to the literature by offering a context-specific strategic model that bridges the gap between AI potential and practical implementation in healthcare systems. It also lays a theoretical and empirical foundation for future research on AI-driven patient safety and strategic healthcare management in developing healthcare contexts.