Stakeholder Perspectives on Early Feasibility Studies for Digital Health Technologies in the European Union: Qualitative Interview Study.

Background: Early feasibility studies (EFSs) are small-scale clinical investigations conducted during the early development of medical devices to assess initial safety and performance, especially when bench or in-silico testing is insufficient. While EFSs are well established for hardware devices, their application to digital health technologies (DHTs) including artificial intelligence (AI)-enabled medical devices remains limited. The rapidly evolving regulatory landscape, including the European Union Medical Device Regulation (EU MDR 2017/745) and the phased introduction of the European Union Artificial Intelligence (EU AI) Act, creates additional complexity for DHT developers. Despite the recognized potential of EFSs to support iterative, user-centered innovation, little is known about how European DHT companies and contract research organizations (CROs) perceive and implement EFSs, or what barriers and opportunities exist for broader adoption. Objective: This study aimed to explore stakeholder perspectives on the use, barriers, and opportunities of EFSs for DHTs in the European Union, and to generate stakeholder-driven recommendations for a harmonized EU-wide EFS framework. Methods: A qualitative descriptive study was conducted using semistructured interviews with representatives from 12 DHT companies and 3 CROs across a range of company sizes, MDR device risk classes, and clinical domains. Participants were recruited through purposive maximum-variation sampling until saturation was reached to capture diverse experiences in regulatory and clinical evidence generation. Interviews, conducted in November 2024 and January 2025, were transcribed and analyzed using thematic analysis, combining deductive and inductive coding. Results: Interviews revealed that while EFSs are valued for providing early human-factor feedback and facilitating iterative design improvements, their current use in DHT development is limited. Key barriers include unclear and hardware-centric regulatory requirements under MDR, fragmented and inconsistent interpretations across EU member states, resource and expertise constraints, and limited dialog with regulatory authorities. The anticipated introduction of the EU AI Act is expected to further increase regulatory complexity, with stakeholders expressing uncertainty about overlapping obligations and the risk of slowed innovation. Some companies, particularly larger or AI-focused ones, have proactively prepared for these changes, while others, especially small and medium-sized enterprises, face significant resource challenges. Several companies reported prioritizing the US Food and Drug Administration pathway due to clearer guidance for DHTs and structured timelines. Stakeholders advocated for a harmonized EU EFS program with DHT-specific guidelines, standardized documentation, predictable timelines, and improved communication channels. Several international models were highlighted as best practices. Conclusions: EFSs remain underused in the EU DHT sector, primarily due to regulatory complexity, fragmentation, and a lack of tailored guidance. A harmonized, DHT-specific EFS framework featuring clearer definitions, standardized processes, and structured dialog between innovators and regulators could accelerate safe, effective, and user-centric digital health innovation. As the MDR and AI Act converge, coordinated regulatory approaches will be critical to balancing innovation, safety, and patient benefit in Europe.