Beyond the Algorithm: Exploring Risk, Trust, and Value in the Adoption of AI-Driven Clinical Decision Support Systems

The advent of Artificial Intelligence (AI) within the healthcare sector represents one of the most revolutionary developments of the past few decades. Despite its potential to enhance diagnostic accuracy, treatment plans, and organizational efficiencies, the adoption of AI-powered Clinical Decision Support Systems (AI-CDSSs) remains low. This doctoral research aims at exploring the behavioural, organizational, and institutional issues that influence adoption, trust, and purchasing of AI solutions in healthcare, combining perspectives lenses from behavioural economics, innovation, and governance research streams. Theoretically grounded in Prospect Theory (Kahneman & Tversky, 1979), this work investigates how perceptions of risk, uncertainty, and trust influence decision-making processes among clinicians and institutions. To understand this complex phenomenon at various levels of analysis, a multi-method research design was implemented. The initial study offers a systematic review of the literature with the aim of clarifying the impact of cognitive biases, risk attitudes, and trust perceptions on the implementation of AI in the clinical setting. The second piece of work looks at the European Commission’s official communications to examine the main policy narratives – human oversight, risk classification, and accountability – that characterize the changing governance of “trustworthy AI” in the healthcare sector. In the third study, over 8,000 AI-healthcare patents (2014-2024) were subjected to patentometric and network analysis to locate worldwide innovation trajectories and thus demonstrate the central role of diagnostics, predictive analytics, and medical imaging ecosystems. The fourth piece of this work consists of a cross-sectional survey of healthcare professionals in Italy and the UK. This is based on the Unified Theory of Acceptance and Use of Technology (UTAUT- Venkatesh, 2003) to identify the psychological and contextual factors that drive AI adoption. The final study (ongoing) employs a Discrete Choice Experiment (DCE) with NHS procurement specialists to understand institutional decision-making and trade-offs between technical performance, fairness, and regulatory ​‍​‌‍​‍‌compliance. Overall, the findings highlight that AI adoption in healthcare is influenced by a multitude of factors, including technical, behavioural and managerial dimensions. Additionally, this research offers actionable insightful perspectives for managers and policymakers seeking to align innovation, accountability and ethical standards in the implementation of AI in healthcare.