Integrating clinical decision support systems, nursing vigilance, and physician prescribing patterns to reduce preventable adverse drug events: a structured evidence-based narrative review on human-AI interface in medication safety

[Technology-Enabled Nursing Practice: Clinical Decision Support, Early Warning Analytics, and Human-Centered Care].

A Multidisciplinary Model for Risk Management and Detection of Ageist Bias in Healthcare Systems in the Era of Artificial Intelligence

Background: The rapid integration of Artificial Intelligence (AI), specifically Clinical Decision Support Systems (CDSS), into healthcare offers substantial efficiency but introduces critical ethical and legal challenges, particularly the perpetuation of systemic bias against older adults (“Digital Ageism”). While technological advances may improve care, they can violate fundamental bioethical principles when models are trained on unrepresentative data. Aim: This article argues that traditional clinical risk-management models are structurally insufficient to address opaque algorithmic bias and presents a conceptual, multidimensional governance framework designed to prevent the codification of human ageism into AI infrastructure. Methods: Drawing on systemic failures observed during the COVID-19 pandemic, the normative model integrates legal and governance standards aligned with the EU AI Act, Explainable AI (XAI) tools, and a three-phase implementation protocol. Results: To illustrate potential application without overburdening medical staff, the article introduces a theoretical Targeted Escalation Protocol and an Autonomous High-Load Safety Mode. The latter applies deterministic hardcoded constraints to contain age-dominant outputs during acute surges while preserving attending-clinician authority. The framework is explored through an Intensive Care Unit (ICU) thought experiment. Conclusions: The framework provides a structured roadmap for policymakers, ethicists, and healthcare administrators to move from reactive defensive medicine toward proactive ethical safety, safeguarding the dignity of the aging population while aiming to mitigate institutional legal exposure.

Artificial Intelligence-Driven Clinical Decision Support in Pakistan: Implications for Primary Care Practice

This research proposes a framework for the sustainable integration of AI-CDSS into Pakistan's healthcare system, with a focus on scaling solutions to primary care environments such as Basic Health Units and rural health centers.

From Evidence to Value: Integrating Evidence-Based Medicine, Patient Safety, Quality Improvement, Learning Health Systems, and Value-Based Healthcare.

RATIONALE Evidence-Based Medicine (EBM) has strengthened clinical decision-making, but evidence alone cannot ensure that care is delivered safely, consistently, or in a manner that produces outcomes meaningful to patients. Patient safety, Quality Improvement (QI), Learning Health Systems (LHS), and Value-Based Healthcare (VBHC) have emerged to address these limitations. Understanding their relationships is important for healthcare systems seeking improvement and patient-centred care. AIMS AND OBJECTIVES To examine the historical development and conceptual relationships among EBM, patient safety, QI, LHS, and VBHC, and to propose an integrated framework for understanding their complementary roles in healthcare quality. METHOD A narrative review was conducted using PubMed as a database, supplemented by reference-list searching. Literature published from approximately 1990 to June 2026 was considered. Searches combined terms related to EBM, patient safety, QI, LHS, VBHC, implementation science, systems thinking, shared decision-making, quality of care, and healthcare transformation. Landmark publications, conceptual papers, guidelines, consensus statements, systematic reviews, and original studies were narratively synthesized. RESULTS The review identifies a Five-Stage Evolution Framework: EBM establishes what works; patient safety addresses how effective care can be delivered safely; QI enables reliable and continuous improvement; LHS creates feedback loops that allow healthcare systems to learn from routine clinical practice; and VBHC evaluates whether these efforts generate outcomes that matter to patients relative to the resources required. These paradigms are complementary rather than competing approaches. Their integration also requires systems thinking, data literacy, interprofessional collaboration, shared decision-making, and attention to patient values and equity. Challenges include the evidence-to-practice gap, organizational barriers, data quality, algorithmic bias, privacy, and the difficulty of measuring individualized value. CONCLUSION Modern healthcare should move beyond isolated quality initiatives toward an integrated system that generates evidence, delivers care safely, continuously learns and improves, and creates meaningful value for patients. The proposed framework is conceptual and requires empirical evaluation.

Advancing Trustworthy Clinical Decision Support Through Explainable Artificial Intelligence: A Framework for Transparent, Interpretable, and Ethical Healthcare Analytics

Artificial intelligence (AI) has become an integral component of clinical decision support systems, improving diagnostic accuracy, risk prediction, treatment planning, and healthcare resource management. But, the "black box" nature of many successful machine learning models has brought up concerns around trust and accountability, fairness and regulatory acceptance in the clinical setting. To address these challenges, Explanatory Artificial Intelligence (XAI) has become a promising method that aims to provide the explainability of the results predicted by the models without compromising the performance of the analysis. The aim of this narrative review is to provide an overview of the evolving image of XAI as a tool to create trustworthy clinical decision support systems and how it can be performed based on the concepts of transparency, interpretability and moral decision-making. It provides a summary of the current literature on essential explainability techniques, how they can be applied to different health care-related problems, and how they help increase trust and transparency in health care decision making. It also addresses human-AI collaboration, model validation, bias mitigation, privacy protection and governance frameworks to enable responsible use of AI. The new emerging developments, such as federated learning, multimodal explainable models, causal reasoning, and generative AI, are also discussed to emphasize future opportunities for clinically reliable and scalable intelligent healthcare systems. The review concludes that explainability is no longer a choice of technical attribute but rather a fundamental component to the use of AI in everyday clinical practice. For safe, equitable and trusted clinical decision support to benefit everyone in the health care sector, transparency, accountability and ethical governance will play a pivotal role.

Enhancing diagnostic safety: addressing knowledge gaps for using human factors tools in the safe and effective use of AI – a proposed research agenda

Abstract Introduction Identify knowledge gaps in applying artificial intelligence in clinical settings, using medical imaging as a primary use case to enhance diagnostic efficacy, efficiency, and patient and provider safety. Methods We convened a two-day workshop with 18 interdisciplinary experts from three countries. Experts represented quality and patient safety, human factors and systems engineering, radiology and other medical specialties, nursing, medical informatics, cognitive and perceptual psychology, psychometrics and statistics, and machine learning, drawn from academia, industry, health systems, and government. Results We identified by consensus six major knowledge-gap domains, with specific research questions for each domain: development, validation, integration and sustainability; redesign of existing healthcare systems; human and team augmentation; deployment of adaptive-learning “Foundation Models;” and balancing innovation, standardization, and regulatory oversight. Conclusions We recommend employing a multidisciplinary collaborative approach in future research to leverage transformational AI capabilities anticipated in the next 5–7 years for each of the identified knowledge-gap domains, including ensuring that clinical AI supports diagnostic decision-making, integrates into clinical workflows, and mitigates risks related to automation bias, overreliance, fragmentation of care, and unintended consequences.