From Principles to Practice: An Actionable Framework for AI Governance in Healthcare Organisations

From Principles to Practice: An Actionable Framework for AI Governance in Healthcare Organisations | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article From Principles to Practice: An Actionable Framework for AI Governance in Healthcare Organisations Sam Freeman, Amy Wang, Sudeep Saraf, Erica Potts, Amy McKimm, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8302125/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract New AI technologies are being rapidly adopted by healthcare professionals, yet existing organisational governance often lacks the processes necessary to oversee their safe and responsible use. Previous AI governance frameworks have focused on high-level AI ethics principles, leaving healthcare organisations struggling to translate them into practice, assess risk, and embed AI oversight into existing processes. We developed a practice-oriented AI governance framework that unifies ethics and governance principles, delivers tiered oversight aligned with digital maturity, and incorporates a structured review checklist for assessing AI tools. Our multimethod approach drew on a scoping review, document analysis, and semi-structured interviews. The resulting framework was validated through stakeholder workshops and applied to exemplar AI tools, demonstrating the checklist’s usability, relevance, and capacity to identify risks and guide decision-making. The framework offers an actionable approach linking principles to operational governance, enabling proportionate oversight, integration with existing processes, and ongoing monitoring, addressing a gap in a field where guidance remains largely conceptual. Scientific community and society/Business and industry Health sciences/Health care Health sciences/Medical research Scientific community and society/Scientific community artificial intelligence governance healthcare patient safety ethics oversight framework implementation digital health Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 INTRODUCTION AI technologies are being rapidly adopted in healthcare organisations across clinical and non-clinical tasks, from predicting patient care needs and supporting diagnosis to optimising administrative processes. [ 1 ] Yet, their integration into healthcare also raises persistent ethical and safety concerns, including bias in decision-making, unequal access to benefits, lack of transparency, and risks from errors due to data drift. [ 2 , 3 ] Despite rapid progress, many organisations still lack the governance structures and decision-making processes to identify and manage these risks effectively. [ 4 ] Concerns around transparency, accountability, privacy, and fairness can overshadow the potential benefits of AI, particularly when harm can occur at scale or go undetected. [ 5 ] Generative AI tools such as ambient scribes, virtual assistants, and discharge planners are already being used in clinical settings, despite unresolved issues around accuracy, accountability, and workforce impact. [ 6 ] In this evolving context, healthcare organisations increasingly face a growing need for AI governance or formal processes to ensure the safe and responsible use of AI technologies. [ 5 , 7 ] Even as demand grows, there is limited guidance on governing the safe and responsible development, implementation, and use of AI. [ 7 ] Although many theoretical frameworks exist, most provide broad ethics and governance principles, leaving a gap between principle and practical application. [ 8 ] Few frameworks in the current literature have been tested or implemented in healthcare settings, and even fewer are aligned with the complex workflows and decision-making processes of healthcare organisations. It also remains unclear how ethics and governance responsibilities should be operationalised across distinct stages of the AI lifecycle or embedded into existing processes for patient safety and digital health. [ 9 ] Additionally, limited attention has been given to the varying levels of digital maturity needed to assess, implement, and monitor AI effectively across diverse healthcare contexts. [ 5 ] This study addresses the gap between high-level AI ethics principles and their practical implementation in healthcare. To close this gap, we developed and validated an actionable governance framework that enables healthcare organisations to adopt and use AI tools safely and responsibly. The framework was designed to address known challenges in AI governance while providing a practical approach to support management across the AI lifecycle. It includes ethics and governance principles, a review checklist for healthcare AI tools, and suggested oversight mechanisms that align with existing organisational processes. The framework was validated by applying it to exemplar AI tools to assess its practical applicability and relevance in real-world healthcare delivery contexts. METHODS Study design and setting We used an exploratory multimethod approach [ 10 ] to develop and test a governance framework for the safe and responsible use of AI in healthcare organisations. The study was designed in accordance with the consolidated criteria for reporting qualitative research (COREQ). [ 11 ] It comprised four stages as outlined in the published protocol [ 12 ]: (1) understanding AI governance needs through a document analysis and a scoping review of the literature; (2) stakeholder interviews; (3) co-developing a draft framework through synthesis of findings; and (4) validating and refining the framework through stakeholder workshops and by application to AI case studies involving tools that were in use or under consideration at the study site (Fig. 1 ). The study was conducted at a major public healthcare organisation in Australia serving a catchment of over 770,000 people in Melbourne’s inner south. The organisation operates across three hospital campuses, including a tertiary and quaternary referral hospital providing specialised care. Its additional campuses focus on community-based care including general medicine, geriatric medicine, rehabilitation, and mental health. It also delivers statewide services across the state of Victoria (e.g. Victorian Adults Burns Service) and national services across Australia (e.g. paediatric lung transplant service). [ 13 ] As a large, complex, and digitally enabled health service, this setting provided a robust environment for developing and testing a practice-oriented AI governance framework for real-world care delivery settings. Data collection took place between April and July 2024. Ethics approval was obtained from the Alfred Health Human Research Ethics Committee (HREC) (ID: 171/24) and Macquarie University HREC (ID: 16508). Participants Participants were recruited for the semi-structured interviews (Stage 2) and stakeholder workshops (Stage 4). Individuals with professional experience relevant to the design, implementation, or oversight of healthcare AI tools were eligible. These included healthcare professionals, researchers, data and analytics specialists, legal and ethics experts, governance leads, executives, board members, and health consumer representatives. For the interviews, two key informant groups were targeted: (1) healthcare staff and consumers from the study site, and (2) subject matter experts from academia (ethics, governance, digital health, technology, AI), government, healthcare administration, clinical settings, and professional healthcare associations. A purposive sampling strategy was used to identify individuals meeting the inclusion criteria, followed by snowball sampling to allow participants to refer additional relevant stakeholders. Healthcare staff were recruited in consultation with the study site’s project team and selected based on their experience in domains such as clinical operations, digital health, data governance, legal, ethics, research, and executive leadership. Subject matter experts were identified through the scoping review, [ 14 ] referrals, and professional networks. Interviews were conducted with participants from Australia, New Zealand, United States, United Kingdom, and Hong Kong. Stakeholder workshop participants were recruited at the study site. The first workshop included a multidisciplinary cohort with diverse institutional representation, while the second involved members of the board of directors. All workshop participants were identified and invited in consultation with the study site’s project team. Written informed consent was obtained from all participants prior to participation. Procedures The framework was developed and refined in four stages: Stage 1: Understanding governance needs : An initial structure for the framework was developed and refined based on the results of the scoping review, which is reported separately. [ 14 ] The scoping review examined 77 frameworks for healthcare organisations implementing AI tools for clinical or operational purposes. From these, we derived four components that form a practical AI governance framework: Set of guiding principles. Organisational oversight mechanism (e.g. governance committee). Method to review or assess AI tools (e.g. questions, checklist items, supporting materials). Review timeline or consideration of the AI life cycle stages. We identified the theoretical and practical components of AI governance, synthesising ethics and governance principles, oversight mechanisms, and checklist elements drawn from best practices, operational guidance, and ethics standards relevant to acute care. As the framework developed, the AI lifecycle stages (component 4) were absorbed into the review method, this was to streamline the process and ensure each assessment item was grounded in a specific stage of the AI lifecycle. To ensure alignment with existing organisational governance, a document analysis was conducted at the study site. This involved a comprehensive review of internal policies, governance documents, and procedural guidelines relevant to digital health and patient safety. The objective was to establish a baseline of current processes for oversight, identify strengths and gaps, and avoid duplication in the proposed AI governance framework. Stage 2: Gathering insights on AI governance : Forty-three semi-structured interviews were conducted with two key informant groups, 23 with healthcare staff and 20 with subject matter experts. Thematic saturation was reached for both groups. Additional input was sought through consultations with two hospital-based consumer reference groups, one focused on digital health and the other on diversity and inclusion. These consultations explored consumer perspectives on ethics concerns, safety, and acceptable uses of AI, and discussed how consumers should be represented within AI governance processes. Interview guides were developed based on a literature review and input from the multidisciplinary project team, which brought expertise in healthcare, qualitative research, AI safety, and governance. Interviews took place either in person or via videoconferencing and were audio recorded. Recordings were de-identified and professionally transcribed using a secure transcription service. Two researchers independently coded the transcripts using grounded theory and open coding techniques (SF, AW). A grounded theory approach enabled the inductive development of concepts that were grounded in the experience and professional expertise of participants. This method was selected because it provides a systematic procedure for generating theory from empirical data where existing models may be limited or underdeveloped. [ 15 ] The analysis informed the framework by identifying practical governance needs, common challenges, and priority areas for safe and responsible AI in healthcare. Key themes extracted from the interviews included data governance, ethical and legal accountability, workforce capability, risk stratification, clinical integration, and equity. These findings shaped the structure and content of the ethics and governance principles, oversight mechanisms, and review checklist components. Findings from the 20 subject matter expert interviews are reported separately. [ 16 ] Stage 3: Co-developing draft framework : Findings from the scoping review, document analysis, and interviews were synthesised to develop a draft AI governance framework tailored to the operational context of healthcare delivery organisations. The framework incorporated regulatory requirements, international best practices, and stakeholder insights to support applicability. The draft framework outlined integration pathways with existing clinical and operational processes. It was refined in Stage 4 through stakeholder workshops and validated by applying it to three AI case studies. Stage 4: Validating and refining framework : Structured workshops were conducted with internal stakeholders to validate and refine the draft governance framework. Participants were identified in consultation with the study site (n = 23). Workshops followed a guided facilitation approach, incorporating structured discussion, and scenario-based evaluation to assess feasibility, usability, and alignment of the framework with existing governance structures and processes. Feedback was thematically analysed and used to iteratively revise the framework. The draft framework was then validated using three case studies of AI tools in use or under consideration at the study site, which included: (1) Home-based Eligibility Analysis & Recommendation Tool (HEART) Summary: This in-house developed AI tool was developed to help identify hospital patients who may be suitable for home-based care. The tool supports an existing process by which clinicians identify patients who may be suitable based on certain criteria. It reviews patient records twice a day and suggests those most likely to meet eligibility criteria. This helps clinical teams make faster decisions, improves patient outcomes, and frees up hospital beds. However, staff remain in control, with the tool supporting not replacing clinical judgement. (2) Aidoc CT Tool [ 17 ]. Summary: This commercially developed AI tool is used in radiology to help flag urgent findings on CT scans, such as brain bleeds, spine fractures, or blood clots, so radiologists can review them faster. The goal is to reduce delays, improve patient safety, and support quicker treatment decisions. Radiologists remain responsible for reviewing scans and all related decisions, with the AI tool serving as an additional review to help detect serious issues earlier. (3) Clostridioides difficile (C. diff) Classification Tool. Summary: This co-developed (university and hospital) AI tool helps infection prevention teams to classify C. diff infections. It uses hospital data to automatically apply national definitions, saving time and reducing the risk of manual errors. The tool supports surveillance and reporting, but final decisions rest with the clinical team. These case studies were selected to reflect the different pathways through which healthcare delivery organisations acquire AI capabilities including in-house tool development, procurement of commercial AI products or via co-development in partnership with external organisations. For each case study, project teams completed the checklist using a think-aloud protocol. [ 18 ] This allowed real-time feedback on the usability, relevance, and clarity of the checklist items. Feedback was analysed thematically and used to revise checklist content, clarify definitions, and improve logical flow. The case studies tested the framework’s applicability across clinical, operational, and research contexts and demonstrated its flexibility and utility in surfacing risks, clarifying roles, and informing decision-making. Findings from this stage, combined with earlier insights, informed the final version of the governance framework. RESULTS A practice-oriented AI governance framework for healthcare organisations The AI governance framework presented here provides a structured, organisational-level approach to the safe and responsible integration of AI in healthcare (Fig. 2 ). It comprises three components: (1) a set of ethics and governance principles; (2) oversight mechanism (based on digital maturity); and (3) a review checklist to support consistent decision-making about the implementation and operation of AI tools across clinical and operational contexts. In the following sections, we present each component, first describing it and then outlining its application at the study site. Ethics and governance principles Our analysis defined 30 ethics and governance principles for healthcare organisations to consider when implementing AI (see Supplementary file 1, Appendix 5). Each organisation will need to review and prioritise these principles based on their local context, risk profile, and existing governance structures. At the study site, the first step involved mapping their existing organisational principles, followed by aligning each one with corresponding checklist questions (see Supplementary file 1, Appendix 4 – mapping template; Supplementary file 2 – principle/checklist mapping). This process supported the operationalisation of the principles within the framework and ensured that they were addressed either through the checklist or through existing governance mechanisms. It also ensured that these principles were not treated as abstract elements, but embedded into decision-making, implementation, and evaluation. During the workshop (Stage 4), stakeholders reviewed a consolidated list of ethics principles and prioritised those that were either poorly represented in existing policies or that posed new risks in the context of AI. These included accountability, consent, cultural appropriateness, fairness, transparency, sustainability, and trust. Some, such as dignity and wellbeing, were considered to already align well with existing organisational principles. Participant discussions highlighted the challenge of distinguishing between principles already embedded in governance processes and those requiring AI-specific adoption. Participants stressed the need to differentiate what can be managed at the organisational level (e.g. privacy) from broader societal risks (e.g. regulation). The process was considered valuable in understanding these considerations and identifying practical pathways for operationalising high-level principles within an AI governance framework tailored to a local context. Oversight mechanism Three dominant mechanisms for organisational oversight emerged from our analysis. These included: (1) creating a dedicated AI governance committee; (2) expanding an existing governance committee to include AI oversight; and (3) integrating AI governance into existing non-AI specific governance processes. To account for the varying levels of digital maturity and resourcing across healthcare organisations, we developed a tiered approach for organisational oversight. A. Digitally mature organisations For digitally mature organisations with adequate infrastructure and workforce capabilities, a dedicated AI governance committee was suggested (Fig. 3 ). This high-level, multidisciplinary body would provide centralised oversight of AI evaluation, approval, and monitoring, supported by a broader advisory group and an operations group responsible for implementation. The advisory group provides subject matter expertise as required for specific AI tools and makes recommendations to the governance committee. The operations group supports applicants to complete the checklist, conducts risk stratification for AI tools, responds to queries, and provides the administrative functions that enable the governance committee and advisory group to fulfil their oversight roles (Supplementary file 1, Appendix 2). The advisory group is shown with a dotted line in Fig. 3 to indicate its advisory function when required. An AI use register can also be established (Supplementary file 1, Appendix 6) as part of this oversight function to promote transparency and accountability, and to support ongoing monitoring, evaluation, and decommissioning of AI tools. B. Developing digital capacity For healthcare organisations with developing digital capacity, AI oversight can be delegated to an existing multidisciplinary team, such as a digital health or clinical innovations committee, with clearly defined roles and domain-specific advisors (Fig. 4 ). This committee would use the AI review checklist to guide evaluations and provide formal recommendations to an existing governance body, such as an executive or digital health committee. An AI advisory group could also be established to support the existing governance committee. C. Digitally mature organisations For healthcare organisations with limited digital capacity, where both infrastructure and workforce capabilities are constrained, AI governance can be initially embedded within existing oversight mechanisms such as clinical governance, digital health, ethics, and procurement processes (Fig. 5 ). A gap analysis can then be conducted to identify any unaddressed responsibilities, with the AI review checklist serving as the central evaluation tool. This approach would support practical, distributed governance by leveraging existing resources, while maintaining alignment with safety, ethical, and regulatory requirements. Stakeholders noted that healthcare organisations vary widely in their digital capability, which influences both the scope of AI they can safely adopt and the governance structures they can sustain. For organisations with limited digital capacity, embedding AI oversight within existing processes offered a pragmatic starting point. These organisations were viewed as often having constrained infrastructure and limited specialist expertise and were best positioned to focus on targeted AI use cases that were operationally critical or lower in risk. This approach enabled proportionate oversight while avoiding the significant administrative and workforce investment required to govern more complex AI tools. It also recognised that, at early stages of maturity, organisations may not yet be equipped to make consistent and well-informed decisions about higher-risk AI applications. In contrast, organisations with more mature digital capability were seen as being better equipped to support a broader and more proactive approach to AI governance. These services typically had stable infrastructure, defined digital governance systems, and access to specialist expertise across data governance, privacy, cybersecurity, clinical governance, and informatics. As they transitioned to a more embedded governance model, a dedicated AI advisory or governance committee could continue to set organisational priorities, assess strategic fit, and provide structured oversight of safety, ethics, and implementation. Mature organisations were also viewed as having the resources required to ensure that AI tools progressed efficiently through approval pathways, were monitored appropriately, and were integrated into existing governance functions over time. This was viewed as enabling a longer-term framework in which AI governance became embedded across executive and operational roles, supported by an executive-level function guiding prioritisation and organisational alignment. D. AI oversight mechanism at the study site At the study site, internal stakeholders supported establishing a high-level, multidisciplinary AI governance committee during the early stages of AI integration. This committee would provide centralised oversight for ethical, clinical, operational, legal, and technical considerations, with a phased plan to embed responsibilities within existing structures over time. A direct reporting line to the Executive Committee (the main body exercising managerial responsibility for the performance of all activities within the organisation) was favoured to ensure visibility, organisational alignment, and resourcing. Two alternative options, reporting to the Digital Health Steering Committee (accountable for overseeing and setting the overall strategic direction across all digital health programs of work) or directly to the Board (effectively bypassing the Executive Committee), were considered less effective due to perceived gaps in focus or proximity to organisation-wide operational decision-making. While a dedicated AI governance committee was considered suitable in the early stages of adoption, it was recognised that over time these responsibilities should transition into existing governance and executive functions as organisational capability, confidence, and experience in reviewing AI tools increases. To support implementation, a smaller ‘AI operations group’ was proposed. This group would provide technical and logistical support, assist with completion of the AI review checklist, and manage the day-to-day coordination of AI governance activities. A broader advisory group of subject matter experts (e.g. clinical, legal, cybersecurity, consumer representatives, etc.) would be engaged per use case. Stakeholders noted that many existing governance committees lacked the necessary breadth of expertise to manage AI-related risks, particularly in data science, ethics, and emerging and evolving regulation. Our governance framework aims to address this gap by supporting AI governance literacy, creating structured approval pathways, and enabling robust monitoring. To support this, organisations are recommended to develop tailored internal policies and terms of reference that clearly define the governance committee’s scope, functions, and decision-making authority. The committee’s proposed functions, reporting lines, and membership serve as a practical foundation for governing clinical and operational AI tools within a care delivery setting (see Supplementary file 1, Appendix 1). To support proportionate oversight and avoid unnecessary administrative burden, several participants recommended introducing a preliminary screening or “gating” process. This approach enables organisations to distinguish between high and low risk AI activities based on a small number of objective criteria. For example, a project may be referred for formal AI governance review if it involves identifiable data, requires integration into hospital IT infrastructure, or requires organisational resources such as ongoing IT support. Tools that do not meet these criteria can generally proceed through existing research or quality improvement processes. Integrating a short gating step into AI intake processes can help ensure that governance attention is directed to areas of greatest organisational risk and impact (see Supplementary file 1, Appendix 3). AI review checklist A central component of the framework is the AI review checklist, a structured evaluation tool designed to guide consistent, transparent, and evidence-informed oversight of AI tools across clinical and operational domains. It supports both research and operational pathways and aims to ensure that the use of AI aligns with ethical standards, organisational priorities, and regulatory requirements, while being adaptable and flexible in its application. The checklist was developed through the synthesis of 10 AI governance and implementation frameworks, [ 19 – 28 ] identified through the scoping review [ 14 ], and key domains from the informant interviews and document analysis. Each question was mapped to ensure broad coverage of key checklist items (Supplementary file 1, Appendix 3). Checklist items were iteratively refined based on informant interviews, mapping of ethics and governance principles (Supplementary file 2), and validation on the AI case studies. The final checklist comprises seven core domains and includes items relevant to both clinical and operational applications of AI (Table 1 ). These domains reflect key stages in the AI lifecycle, to support comprehensive oversight from conception through to decommissioning (Fig. 6 ). While the checklist was designed to be comprehensive, it is not intended as a one-size-fits-all solution. Rather, it should be applied in a way that reflects and compliments the specific governance structures, policies, and decision-making processes already established within the local context of the healthcare organisation. The use of the checklist should be guided by the assessed risk level of the AI tool, with higher-risk applications subject to more rigorous scrutiny and lower-risk tools using a proportionate approach. This flexible, context-specific use of the checklist is intended to support integration into existing workflows without introducing unnecessary duplication or burden. Each item in the checklist is linked to relevant ethical and governance principles (e.g. transparency, fairness, accountability), ensuring these concepts are embedded in the assessment process rather than treated as separate considerations. Applicants are asked to complete only the questions relevant to their proposed pathway (research, clinical/operational, or both), with clear instructions and subheadings for readability. Feedback from the three case studies led to refinements including clearer definitions, improved logic flow, and the ability to reference existing documentation rather than duplicating content. The checklist now serves as both a practical tool for governance and an educational mechanism to build AI literacy among developers and decision-makers (see Supplementary file 1, Appendix 3 for complete AI review checklist). Table 1 AI governance checklist core domains. Domain Purpose Background details Outlines AI tool/ company information. Problem identification Defines rationale, value, and strategic alignment. Design and development Details model type, intended use, IP, stakeholder involvement. Validation and safety Covers performance metrics, risk, clinical safety, regulatory compliance. Implementation and Integration Reviews workflow, UI, technical integration, training, and contingency planning. Monitoring and evaluation Ensures ongoing review, equity analysis, safe removal, and lifecycle tracking. Funding Details funding for development, implementation, and ongoing costs. DISCUSSION This study describes the design, testing, and refinement of a practical AI governance framework tailored to the needs of healthcare delivery organisations. The framework addresses a persistent gap between high-level AI ethics and governance principles and the real-world processes required to govern the safe and responsible use of AI in healthcare. It includes a mapped set of ethics and governance principles, possible committee structures with defined oversight, and a practical checklist for reviewing AI tools aligned to key stages in the AI lifecycle. By combining a comprehensive scoping review with stakeholder input and real-world testing, the framework connects normative and procedural approaches to AI governance, supporting translation into practice. Our findings align with previous research that has characterised AI governance as fragmented, heavily principle-based, and largely lacking in operational tools. [ 29 , 30 ] While many current frameworks present high-level concepts such as transparency, fairness, and accountability, few provide a clear path for applying these concepts in healthcare delivery environments. The limited uptake of governance tools may reflect this disconnect, as well as uncertainty among healthcare leaders about where governance responsibility lies. Prior work has highlighted the challenge of operationalising governance in health systems, particularly when governance expectations are vague or insufficiently contextualised. Pyone et al., in their systematic review of health systems governance, highlighted many governance frameworks that lacked practical guidance for implementation at different levels of the health system. [ 31 ] Similarly, Brinkerhoff and Bossert emphasised the disparity between formal governance structures “and the extent to which these are actually put into practice.” [ 32 ] This study responds to those concerns by defining a structured governance framework with aligned tools and processes that can be integrated into existing organisational processes. Accounting for AI lifecycle stages (Fig. 6 ) enables continuous oversight throughout the development, implementation, and use of AI tools. Each stage is supported by relevant principles, ensuring that governance is both continuous and proportional to risk. This lifecycle-based approach supports iterative, context-specific governance models that respond to the evolving nature of AI use in clinical and operational settings. [ 33 , 34 ] Our stakeholder consultations confirmed that current governance structures do not routinely account for AI-specific considerations. Participants reported that project approval pathways vary across clinical, research, and digital domains, with little clarity around who is responsible for assessing AI-related risks. These insights support those of Morley et al. and Jobin et al., who have highlighted the challenges of operationalising ethics principles across complex healthcare systems. [ 35 , 36 ] Key gaps identified by stakeholders include insufficient attention to bias mitigation, post-deployment safety, integration with clinical workflows, and the ability to contest or override AI-generated outputs. The review checklist was developed to address these needs and was validated on three diverse AI tools with relevant stakeholders, including frontline and operational users who assessed its usability and relevance within real-world organisational conditions. This validation process demonstrated the checklist’s flexibility across different acquisition pathways (internal development, commercial procurement, and co-development) and its ability to elicit detailed, structured discussion about model purpose, performance, implementation challenges, and ongoing monitoring. Unlike more conceptual self-assessment tools, this checklist directly supports project-level review and approval. Participants found it useful both as a decision-making aid and as a tool to build AI literacy within multidisciplinary teams. Importantly, the framework includes defined roles and committee structures that align with existing healthcare governance models. The AI governance committee oversight should function similar to that of quality, safety, and innovation governance committees. This design reflects a deliberate strategy to embed AI oversight within existing organisational processes, reducing duplication and ensuring alignment with broader service priorities. Here, the framework aims to promote both integration and validity. As organisations gain experience with AI and build capability to assess and manage associated risks, there is growing support for AI governance responsibilities to transition from standalone committees into existing executive and governance functions. With appropriate upskilling, staff across domains such as clinical governance, research ethics, data governance, privacy, and cybersecurity will be able to incorporate AI considerations within their usual decision-making processes. An executive role, such as a Chief AI Officer or equivalent, could oversee the prioritisation, need, and suitability of AI tools, working alongside other executive functions to ensure consistent, efficient, and accountable decision-making across the organisation. The framework also addresses equity and inclusion through multiple mechanisms. The inclusion of cultural appropriateness, fairness, and transparency as core principles was prioritised by stakeholders, particularly in relation to diverse data governance and health equity considerations. The framework recommends consulting Aboriginal and/or Torres Strait Islander advisors and equity experts during project planning, recognising that ethical risks and opportunities are not evenly distributed across populations. This aligns with strength-based, community-led approaches to digital health governance. [ 37 ] While the framework is grounded in the Australian health system, its components are designed to be adaptable across diverse healthcare contexts. Its development draws on widely accepted ethics and governance principles, and its tools can be localised to reflect different regulatory requirements and institutional structures and processes. The modular design of the checklist and committee structure allows for phased implementation or adaptation to resource-constrained settings. This flexibility is essential, given the rapidly evolving nature of AI technologies and the varied levels of digital maturity across healthcare organisations. [ 38 ] Several limitations warrant discussion. The framework was developed and tested within a single healthcare organisation and will likely require modification in other contexts. Although the checklist was validated on three AI use cases, a larger sample across multiple settings would strengthen generalisability. Future research should evaluate the framework’s implementation over time, including its impact on decision-making quality, project outcomes, and staff engagement. Notwithstanding these limitations, this study provides one of the first practice-oriented frameworks for AI governance in healthcare that has been developed and validated within a real-world healthcare organisation. The framework is designed to be adapted by other healthcare organisations seeking to enhance their governance processes to ensure the safe and responsible use of AI. It contributes a practical approach to the emerging field of digital health governance, grounded in systems thinking, real-world constraints, and inclusive design. As organisations continue to explore the potential of AI, frameworks such as this one will be essential to guide adoption and manage risk. AI governance must move beyond principles and into systems, processes, and shared accountabilities. This study contributes an applied governance approach that can support transformation in healthcare without compromising safety or responsibility. Declarations Competing Interests SS, AM, and EP were employed by Alfred Health and participated in the project steering committee but were not involved in data collection, analysis, or interpretation. Their involvement in recruitment was limited to forwarding the study invitation to potential participants. SS participated in the board and chief executive officer interviews. SF, AW, FM, and EC have no professional or personal affiliation with Alfred Health. Funding declaration The study was supported by the Digital Health CRC Limited, which is funded under the Australian Commonwealth’s Cooperative Research Centres program, an Australian Government initiative for collaboration between university researchers and the industry. Author Contribution FM and EC conceptualised the study. SF, FM, and AW designed and conducted the study, with AM, SS and EP assisting with participant recruitment at the study site. SF and AW undertook data analysis. SF, FM, and AW drafted the manuscript with input from all authors. All authors provided revisions for intellectual content. All authors have approved the final manuscript. Acknowledgement We wish to acknowledge the participants from the study site as well as key informants from healthcare, academia, government, and professional associations who gave their time to participate in the study. Data Availability The datasets generated during this study will not be made not publicly available for privacy reasons but requests to the corresponding author will be considered on a case-by-case basis. References Magrabi, F., et al. (2023). "Automation in Contemporary Clinical Information Systems: A Survey of AI in Healthcare Settings." Yearb Med Inform 32(1): 115–126. Challen, R., et al. (2019). "Artificial intelligence, bias and clinical safety." BMJ Qual Saf 28(3): 231–237. Lyell, D., et al. (2023). "More than algorithms: an analysis of safety events involving ML-enabled medical devices reported to the FDA." J Am Med Inform Assoc 30(7): 1227–1236. Davenport TH, Glaser JP. Factors governing the adoption of artificial intelligence in healthcare providers. Discov Health Syst. 2022;1(1):4. Solanki P, Grundy J, Hussain W. Operationalising ethics in artificial intelligence for healthcare: A framework for AI developers. AI and Ethics. 2023;3(1):223–40. Coiera E, Fraile Navarro D. AI as an ecosystem: ensuring generative AI Is safe and effective. NEJM AI. 2024 Sept;1(9):1–4. doi: 10.1056/AIp2400611 Australian Commission on Safety and Quality in Health Care. Literature Review and Environmental Scan Final Report: AI Implementation in Hospitals: Legislation, Policy, Guidelines and Principles, and Evidence about Quality and Safety.; 2024 29 May 2024. Ihaddouchen I, Buijsman S, Pozzi G, Van de Sande D, Reis A, Townsend R, et al. Responsible artificial intelligence in healthcare: a systematic review on the use of ethical principles in the development and deployment of artificial intelligence. BMJ Digital Health & AI. 2025;1:e000086. Farah Magrabi LB, Kalissa Brooke-Cowden, Tamasha Jayawardena, Amy Wang, Enrico Coiera, Yves Saint James Aquino, Emma Frost, Stacy Carter, Carolyn Adams, Alaa Almohanna. AI Implementation in Hospitals: Legislation, Policy, Guidelines and Principles, and Evidence about Quality and Safety. 2024. Davies P. The SAGE Dictionary of Social Research Methods. 2006 2025/03/18. London: SAGE Publications, Ltd. Available from: https://methods.sagepub.com/dict/edvol/the-sage-dictionary-of-social-research-methods/toc . Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care. 2007;19(6):349–57. Freeman S, Wang A, Saraf S, Potts E, McKimm A, Coiera E, Magrabi F. Developing an AI Governance Framework for Safe and Responsible AI in Health Care Organizations: Protocol for a Multimethod Study. JMIR Res Protoc. 2025;14:e75702. doi: 10.2196/75702 . PMID: 40720809; PMCID: PMC12340452. Alfred Health. Alfred Health Annual Report 2023-24. 2024. Amy Wang, Sam Freeman, Farah Magrabi et al. Governance for safe and responsible AI in healthcare organisations: A scoping review of frameworks, 19 August 2025, PREPRINT (Version 1) available at Research Square [ https://doi.org/10.21203/rs.3.rs-7209096/v1] Halaweh M, Fidler C, McRobb S. Integrating the Grounded Theory Method and Case Study Research Methodology Within IS Research: A Possible 'Road Map'2008. 165 p. Freeman S, Wang A, Magrabi F. Key Considerations for Governing Safe and Responsible Use of AI in Healthcare. Stud Health Technol Inform. 2025;329:1367–1371. doi: 10.3233/SHTI251062 . PMID: 40776080. Aidoc. Aidoc 2025 [Available from: https://www.aidoc.com/solutions/radiology/ . Ericsson KA, Simon HA. Protocol Analysis: Verbal Reports as Data: The MIT Press; 1993. Liu X, Cruz Rivera S, Moher D, Calvert MJ, Denniston AK. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Nat Med. 2020;26(9):1364–74. Vasey B, Nagendran M, Campbell B, Clifton DA, Collins GS, Denaxas S, et al. Reporting guideline for the early stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI. Bmj. 2022;377:e070904. Lekadir K, Frangi AF, Porras AR, Glocker B, Cintas C, Langlotz CP, et al. FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare. Bmj. 2025;388:e081554. U.S. Department of Health and Human Services. Trustworthy AI (TAI) Playbook 2021 [Available from: https://digitalgovernmenthub.org/library/trustworthy-ai-tai-playbook/ . Coalition for Health AI. Responsible AI Guide 2024 [Available from: https://chai.org/wp-content/uploads/2025/03/CHAI_Responsible-AI-Guide_03202025.pdf . Economou-Zavlanos NJ, Bessias S, Cary MP, Jr, Bedoya AD, Goldstein BA, Jelovsek JE, et al. Translating ethical and quality principles for the effective, safe and fair development, deployment and use of artificial intelligence technologies in healthcare. Journal of the American Medical Informatics Association. 2023;31(3):705–13. Whittaker R, Dobson R, Jin CK, Style R, Jayathissa P, Hiini K, et al. An example of governance for AI in health services from Aotearoa New Zealand. NPJ Digit Med. 2023;6(1):164. Collins GS, Dhiman P, Andaur Navarro CL, Ma J, Hooft L, Reitsma JB, et al. Protocol for development of a reporting guideline (TRIPOD-AI) and risk of bias tool (PROBAST-AI) for diagnostic and prognostic prediction model studies based on artificial intelligence. BMJ Open. 2021;11(7):e048008. Liao F, Adelaine S, Afshar M, Patterson BW. Governance of Clinical AI applications to facilitate safe and equitable deployment in a large health system: Key elements and early successes. Front Digit Health. 2022;4:931439. Reddy S, Rogers W, Makinen VP, Coiera E, Brown P, Wenzel M, et al. Evaluation framework to guide implementation of AI systems into healthcare settings. BMJ Health Care Inform. 2021;28(1). Mittelstadt B. Principles alone cannot guarantee ethical AI. Nature Machine Intelligence. 2019;1(11):501–7. Mesko B. The role of artificial intelligence in precision medicine. Expert Review of Precision Medicine and Drug Development. 2017;2(5):239–41. Pyone T, Smith H, van den Broek N. Frameworks to assess health systems governance: a systematic review. Health Policy and Planning. 2017;32(5):710–22. Brinkerhoff DW, Bossert TJ. Health governance: principal–agent linkages and health system strengthening. Health Policy and Planning. 2013;29(6):685–93. Papagiannidis E, Mikalef P, Conboy K. Responsible artificial intelligence governance: A review and research framework. The Journal of Strategic Information Systems. 2025;34(2):101885. Greenhalgh T, Wherton J, Papoutsi C, Lynch J, Hughes G, A'Court C, et al. Beyond Adoption: A New Framework for Theorizing and Evaluating Nonadoption, Abandonment, and Challenges to the Scale-Up, Spread, and Sustainability of Health and Care Technologies. J Med Internet Res. 2017;19(11):e367. Morley J, Floridi L, Kinsey L, Elhalal A. From What to How: An Initial Review of Publicly Available AI Ethics Tools, Methods and Research to Translate Principles into Practices. Science and Engineering Ethics. 2020;26(4):2141–68. Jobin A, Ienca M, Vayena E. The global landscape of AI ethics guidelines. Nature Machine Intelligence. 2019;1(9):389–99. Cordes A, Bak M, Lyndon M, Hudson M, Fiske A, Celi LA, et al. Competing interests: digital health and indigenous data sovereignty. NPJ Digit Med. 2024;7(1):178. Kim GJ, Namkoong K. Developing the Digital Health Communication Maturity Model: Systematic Review. J Med Internet Res. 2025;27:e68344. Additional Declarations Competing interest reported. SS, AM, and EP were employed by Alfred Health and participated in the project steering committee but were not involved in data collection, analysis, or interpretation. Their involvement in recruitment was limited to forwarding the study invitation to potential participants. SS participated in the board and chief executive officer interviews. SF, AW, FM, and EC have no professional or personal affiliation with Alfred Health. Supplementary Files Supplementaryfile1.docx Supplementaryfile2Principleandchecklistmapping.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8302125","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":557839210,"identity":"1cebc600-03d3-45d4-9db8-a1c191e3e678","order_by":0,"name":"Sam Freeman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIiWNgGAWjYLCCCgjF+ABMHSBGyxkIxWxAshY2CaK0yLefTmA4uMdGTrf97LGqG2UMcnw3Ehg/8+DRYnAmdwPDgWdpxmZn8tJu55xjMJa8kcAsjVcLQ+4G5g8HDiduO5Bjdju3jSFxw40EBrxa5PvfAm058L9+2/k3ZsVALfVALcy/8WlhuAFy2IEDCWY3csyYgVoSDG4ksOF32I23G4A6kg233XhjLJ1zTsJw5pmHbZZz8Dosd+ODAwfs5M3O5xh+zimzkec7nnz4xht8DmNAiQhw1DA2ENCAAthIUTwKRsEoGAUjBQAARe9U5IjZSkwAAAAASUVORK5CYII=","orcid":"","institution":"Macquarie University","correspondingAuthor":true,"prefix":"","firstName":"Sam","middleName":"","lastName":"Freeman","suffix":""},{"id":557839211,"identity":"c8e7a77c-bd64-4abd-ad68-9dda2900cedc","order_by":1,"name":"Amy Wang","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Amy","middleName":"","lastName":"Wang","suffix":""},{"id":557839212,"identity":"bfa30775-03f7-4db4-a475-59845f7d3bfb","order_by":2,"name":"Sudeep Saraf","email":"","orcid":"","institution":"Alfred Hospital","correspondingAuthor":false,"prefix":"","firstName":"Sudeep","middleName":"","lastName":"Saraf","suffix":""},{"id":557839213,"identity":"208f4716-94c3-4320-b203-6cd32a9b4756","order_by":3,"name":"Erica Potts","email":"","orcid":"","institution":"Alfred Hospital","correspondingAuthor":false,"prefix":"","firstName":"Erica","middleName":"","lastName":"Potts","suffix":""},{"id":557839214,"identity":"71394b99-f90f-4420-a231-cb9a9accaa0a","order_by":4,"name":"Amy McKimm","email":"","orcid":"","institution":"Alfred Hospital","correspondingAuthor":false,"prefix":"","firstName":"Amy","middleName":"","lastName":"McKimm","suffix":""},{"id":557839215,"identity":"c2786ecc-1740-4126-abd2-5ecb970dba74","order_by":5,"name":"Enrico Coiera","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Enrico","middleName":"","lastName":"Coiera","suffix":""},{"id":557839216,"identity":"6e53f552-da43-455b-85d5-1cffec160e92","order_by":6,"name":"Farah Magrabi","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Farah","middleName":"","lastName":"Magrabi","suffix":""}],"badges":[],"createdAt":"2025-12-07 22:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8302125/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8302125/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98309896,"identity":"1245ea88-7d47-4017-8265-053a90091e63","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":666684,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptFromPrinciplestoPracticeAnActionableFrameworkforAIGovernanceinHealthcareOrganisations.docx","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/80321c8c39fdc706a4c6a8d4.docx"},{"id":98435730,"identity":"fe422515-8467-4794-a273-530389a90cdb","added_by":"auto","created_at":"2025-12-17 16:54:19","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8546,"visible":true,"origin":"","legend":"","description":"","filename":"ec292be1ef274253bd58552d4046d4f0.json","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/debe0bc54bae7dfccf3e64b6.json"},{"id":98309894,"identity":"4c4436d2-296e-423d-8575-93074f66cc59","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":301000,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/5873111b9f6e899444dcb7c7.docx"},{"id":98309888,"identity":"8fbd4cb8-1b52-41a5-8188-29c4b99f599b","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31814,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfile2Principleandchecklistmapping.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/66bf3af4b5b88ea95ef45cdc.xlsx"},{"id":98435369,"identity":"86ae7856-b6d5-404f-af5b-0a1d25ad3b39","added_by":"auto","created_at":"2025-12-17 16:53:37","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105351,"visible":true,"origin":"","legend":"","description":"","filename":"ec292be1ef274253bd58552d4046d4f01enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/1278bb74cdee25cbe311a8b1.xml"},{"id":98437050,"identity":"8179f701-4a30-43ce-b05b-d78db1790391","added_by":"auto","created_at":"2025-12-17 16:56:51","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1074,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/71c5bd9f336ffff3b785a65b.jpeg"},{"id":98436998,"identity":"73979834-0db1-495f-901a-73f24c30a8c6","added_by":"auto","created_at":"2025-12-17 16:56:42","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/f611284925d176cf06c24973.png"},{"id":98309897,"identity":"27cb5663-3584-4bfa-80cf-587f4c828a39","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/5b55bc34a386a8989aaf3860.png"},{"id":98434512,"identity":"9fcde02f-3bf6-497e-931f-1d67bc99c112","added_by":"auto","created_at":"2025-12-17 16:52:13","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33535,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/bee47c64eaa01e2350d1bc7b.png"},{"id":98436666,"identity":"904cbb0e-78b5-40b2-ac54-6a7103f21e3b","added_by":"auto","created_at":"2025-12-17 16:56:02","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30227,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/e0455c78a7ee765f3d358e06.png"},{"id":98309899,"identity":"d33fa692-e36d-4e4f-9b75-6471a537a221","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":74632,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/d1600a634ffcf6bd4b7b6441.png"},{"id":98309900,"identity":"7fefce37-411e-4a85-a1fe-f9923ef9d7e1","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11972,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinegroupimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/6a57b913eb828185920e9837.png"},{"id":98437659,"identity":"535c402c-c643-4508-9c10-95edcfd8c862","added_by":"auto","created_at":"2025-12-17 16:57:32","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98809,"visible":true,"origin":"","legend":"","description":"","filename":"ec292be1ef274253bd58552d4046d4f01structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/0418951f4aaa1b4393bbcaf1.xml"},{"id":98309902,"identity":"af46a310-143b-406b-96e9-8b54bf87e733","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118265,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/2a7ae4827997a445a1bf7b78.html"},{"id":98309881,"identity":"001cd4b1-5599-4023-93ed-2a7e422c66d9","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":74862,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eStudy process diagram\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/de646a0e3aba08ef337bd63e.png"},{"id":98309882,"identity":"ad97f43f-44b2-43ce-98db-198fca687c1c","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":16057,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eA practice-oriented AI governance framework for healthcare organisations\u003c/em\u003e\u003c/p\u003e","description":"","filename":"groupimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/fae421ecbf8a79be161d571f.jpeg"},{"id":98437450,"identity":"6aa9dbad-06d9-4721-8a97-12e42efe0f22","added_by":"auto","created_at":"2025-12-17 16:57:21","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":84016,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAI oversight in a digitally mature healthcare organisation through a dedicated AI governance committee\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/4c6bb4fecd8d35947f1244be.png"},{"id":98437271,"identity":"92aa91d5-aabb-4256-bcd8-ed876f48566d","added_by":"auto","created_at":"2025-12-17 16:57:10","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":70399,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAI oversight in a healthcare organisation with developing digital capacity by expanding an existing multidisciplinary committee\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/1c925304867131f7fbe19e72.png"},{"id":98309892,"identity":"16660249-8d61-47d8-b970-7fed61ce1f32","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":414690,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAI oversight in a healthcare organisation with limited digital capacity by integrating governance into existing clinical governance, digital health, ethics, and procurement processes\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/f65c2857841eaa4c6418addc.png"},{"id":98309887,"identity":"5c528ecf-211d-4dc9-bb93-036c458cbc61","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":16913,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAI lifecycle stages\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/6cfaa0871abf6dde1b62bc09.png"},{"id":99316185,"identity":"99a1bb37-9c83-40f5-ae30-ecfaa731474b","added_by":"auto","created_at":"2025-12-31 16:27:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1110399,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/977a47b9-2f40-4b35-9e2f-a734b6fe4f91.pdf"},{"id":98309884,"identity":"e7d7c9b7-62ab-4424-95fc-47ec0b794dbd","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":301000,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/9dab9e1113f38691e7154ef2.docx"},{"id":98309890,"identity":"a801a4f2-23bf-4fa2-bf9c-563161c2d137","added_by":"auto","created_at":"2025-12-16 12:02:51","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":31814,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfile2Principleandchecklistmapping.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8302125/v1/2ff92dede5009cc8916f6feb.xlsx"}],"financialInterests":"Competing interest reported. SS, AM, and EP were employed by Alfred Health and participated in the project steering committee but were not involved in data collection, analysis, or interpretation. Their involvement in recruitment was limited to forwarding the study invitation to potential participants. SS participated in the board and chief executive officer interviews. SF, AW, FM, and EC have no professional or personal affiliation with Alfred Health.","formattedTitle":"From Principles to Practice: An Actionable Framework for AI Governance in Healthcare Organisations","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eAI technologies are being rapidly adopted in healthcare organisations across clinical and non-clinical tasks, from predicting patient care needs and supporting diagnosis to optimising administrative processes. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] Yet, their integration into healthcare also raises persistent ethical and safety concerns, including bias in decision-making, unequal access to benefits, lack of transparency, and risks from errors due to data drift. [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] Despite rapid progress, many organisations still lack the governance structures and decision-making processes to identify and manage these risks effectively. [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] Concerns around transparency, accountability, privacy, and fairness can overshadow the potential benefits of AI, particularly when harm can occur at scale or go undetected. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] Generative AI tools such as ambient scribes, virtual assistants, and discharge planners are already being used in clinical settings, despite unresolved issues around accuracy, accountability, and workforce impact. [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] In this evolving context, healthcare organisations increasingly face a growing need for \u003cem\u003eAI governance\u003c/em\u003e or formal processes to ensure the safe and responsible use of AI technologies. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eEven as demand grows, there is limited guidance on governing the safe and responsible development, implementation, and use of AI. [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] Although many theoretical frameworks exist, most provide broad ethics and governance principles, leaving a gap between principle and practical application. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] Few frameworks in the current literature have been tested or implemented in healthcare settings, and even fewer are aligned with the complex workflows and decision-making processes of healthcare organisations. It also remains unclear how ethics and governance responsibilities should be operationalised across distinct stages of the AI lifecycle or embedded into existing processes for patient safety and digital health. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] Additionally, limited attention has been given to the varying levels of digital maturity needed to assess, implement, and monitor AI effectively across diverse healthcare contexts. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e This study addresses the gap between high-level AI ethics principles and their practical implementation in healthcare. To close this gap, we developed and validated an actionable governance framework that enables healthcare organisations to adopt and use AI tools safely and responsibly. The framework was designed to address known challenges in AI governance while providing a practical approach to support management across the AI lifecycle. It includes ethics and governance principles, a review checklist for healthcare AI tools, and suggested oversight mechanisms that align with existing organisational processes. The framework was validated by applying it to exemplar AI tools to assess its practical applicability and relevance in real-world healthcare delivery contexts.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design and setting\u003c/h2\u003e \u003cp\u003eWe used an exploratory multimethod approach [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] to develop and test a governance framework for the safe and responsible use of AI in healthcare organisations. The study was designed in accordance with the consolidated criteria for reporting qualitative research (COREQ). [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] It comprised four stages as outlined in the published protocol [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]: (1) understanding AI governance needs through a document analysis and a scoping review of the literature; (2) stakeholder interviews; (3) co-developing a draft framework through synthesis of findings; and (4) validating and refining the framework through stakeholder workshops and by application to AI case studies involving tools that were in use or under consideration at the study site (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe study was conducted at a major public healthcare organisation in Australia serving a catchment of over 770,000 people in Melbourne\u0026rsquo;s inner south. The organisation operates across three hospital campuses, including a tertiary and quaternary referral hospital providing specialised care. Its additional campuses focus on community-based care including general medicine, geriatric medicine, rehabilitation, and mental health. It also delivers statewide services across the state of Victoria (e.g. Victorian Adults Burns Service) and national services across Australia (e.g. paediatric lung transplant service). [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] As a large, complex, and digitally enabled health service, this setting provided a robust environment for developing and testing a practice-oriented AI governance framework for real-world care delivery settings. Data collection took place between April and July 2024.\u003c/p\u003e \u003cp\u003eEthics approval\u003c/strong\u003ewas obtained from the Alfred Health Human Research Ethics Committee (HREC) (ID: 171/24) and Macquarie University HREC (ID: 16508).\u003c/p\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eParticipants\u003c/h3\u003e\n\u003cp\u003eParticipants were recruited for the semi-structured interviews (Stage 2) and stakeholder workshops (Stage 4). Individuals with professional experience relevant to the design, implementation, or oversight of healthcare AI tools were eligible. These included healthcare professionals, researchers, data and analytics specialists, legal and ethics experts, governance leads, executives, board members, and health consumer representatives.\u003c/p\u003e \u003cp\u003eFor the interviews, two key informant groups were targeted: (1) healthcare staff and consumers from the study site, and (2) subject matter experts from academia (ethics, governance, digital health, technology, AI), government, healthcare administration, clinical settings, and professional healthcare associations. A purposive sampling strategy was used to identify individuals meeting the inclusion criteria, followed by snowball sampling to allow participants to refer additional relevant stakeholders. Healthcare staff were recruited in consultation with the study site\u0026rsquo;s project team and selected based on their experience in domains such as clinical operations, digital health, data governance, legal, ethics, research, and executive leadership. Subject matter experts were identified through the scoping review, [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] referrals, and professional networks. Interviews were conducted with participants from Australia, New Zealand, United States, United Kingdom, and Hong Kong.\u003c/p\u003e \u003cp\u003e Stakeholder workshop participants were recruited at the study site. The first workshop included a multidisciplinary cohort with diverse institutional representation, while the second involved members of the board of directors. All workshop participants were identified and invited in consultation with the study site\u0026rsquo;s project team. Written informed consent was obtained from all participants prior to participation.\u003c/p\u003e\n\u003ch3\u003eProcedures\u003c/h3\u003e\n\u003cp\u003eThe framework was developed and refined in four stages:\u003c/p\u003e \u003cp\u003e\u003cem\u003eStage 1: Understanding governance needs\u003c/em\u003e: An initial structure for the framework was developed and refined based on the results of the scoping review, which is reported separately. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] The scoping review examined 77 frameworks for healthcare organisations implementing AI tools for clinical or operational purposes. From these, we derived four components that form a practical AI governance framework:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eSet of guiding principles.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eOrganisational oversight mechanism (e.g. governance committee).\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eMethod to review or assess AI tools (e.g. questions, checklist items, supporting materials).\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eReview timeline or consideration of the AI life cycle stages.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e We identified the theoretical and practical components of AI governance, synthesising ethics and governance principles, oversight mechanisms, and checklist elements drawn from best practices, operational guidance, and ethics standards relevant to acute care. As the framework developed, the AI lifecycle stages (component 4) were absorbed into the review method, this was to streamline the process and ensure each assessment item was grounded in a specific stage of the AI lifecycle.\u003c/p\u003e \u003cp\u003eTo ensure alignment with existing organisational governance, a document analysis was conducted at the study site. This involved a comprehensive review of internal policies, governance documents, and procedural guidelines relevant to digital health and patient safety. The objective was to establish a baseline of current processes for oversight, identify strengths and gaps, and avoid duplication in the proposed AI governance framework.\u003c/p\u003e \u003cp\u003e \u003cem\u003eStage 2: Gathering insights on AI governance\u003c/em\u003e: Forty-three semi-structured interviews were conducted with two key informant groups, 23 with healthcare staff and 20 with subject matter experts. Thematic saturation was reached for both groups. Additional input was sought through consultations with two hospital-based consumer reference groups, one focused on digital health and the other on diversity and inclusion. These consultations explored consumer perspectives on ethics concerns, safety, and acceptable uses of AI, and discussed how consumers should be represented within AI governance processes.\u003c/p\u003e \u003cp\u003eInterview guides were developed based on a literature review and input from the multidisciplinary project team, which brought expertise in healthcare, qualitative research, AI safety, and governance. Interviews took place either in person or via videoconferencing and were audio recorded. Recordings were de-identified and professionally transcribed using a secure transcription service. Two researchers independently coded the transcripts using grounded theory and open coding techniques (SF, AW). A grounded theory approach enabled the inductive development of concepts that were grounded in the experience and professional expertise of participants. This method was selected because it provides a systematic procedure for generating theory from empirical data where existing models may be limited or underdeveloped. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] The analysis informed the framework by identifying practical governance needs, common challenges, and priority areas for safe and responsible AI in healthcare. Key themes extracted from the interviews included data governance, ethical and legal accountability, workforce capability, risk stratification, clinical integration, and equity. These findings shaped the structure and content of the ethics and governance principles, oversight mechanisms, and review checklist components. Findings from the 20 subject matter expert interviews are reported separately. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e \u003cem\u003eStage 3: Co-developing draft framework\u003c/em\u003e: Findings from the scoping review, document analysis, and interviews were synthesised to develop a draft AI governance framework tailored to the operational context of healthcare delivery organisations. The framework incorporated regulatory requirements, international best practices, and stakeholder insights to support applicability. The draft framework outlined integration pathways with existing clinical and operational processes. It was refined in Stage 4 through stakeholder workshops and validated by applying it to three AI case studies.\u003c/p\u003e \u003cp\u003e \u003cem\u003eStage 4: Validating and refining framework\u003c/em\u003e: Structured workshops were conducted with internal stakeholders to validate and refine the draft governance framework. Participants were identified in consultation with the study site (n\u0026thinsp;=\u0026thinsp;23). Workshops followed a guided facilitation approach, incorporating structured discussion, and scenario-based evaluation to assess feasibility, usability, and alignment of the framework with existing governance structures and processes. Feedback was thematically analysed and used to iteratively revise the framework. The draft framework was then validated using three case studies of AI tools in use or under consideration at the study site, which included:\u003c/p\u003e \u003cp\u003e(1) Home-based Eligibility Analysis \u0026amp; Recommendation Tool (HEART)\u003c/p\u003e \u003cp\u003eSummary: This in-house developed AI tool was developed to help identify hospital patients who may be suitable for home-based care. The tool supports an existing process by which clinicians identify patients who may be suitable based on certain criteria. It reviews patient records twice a day and suggests those most likely to meet eligibility criteria. This helps clinical teams make faster decisions, improves patient outcomes, and frees up hospital beds. However, staff remain in control, with the tool supporting not replacing clinical judgement.\u003c/p\u003e \u003cp\u003e(2) Aidoc CT Tool [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSummary: This commercially developed AI tool is used in radiology to help flag urgent findings on CT scans, such as brain bleeds, spine fractures, or blood clots, so radiologists can review them faster. The goal is to reduce delays, improve patient safety, and support quicker treatment decisions. Radiologists remain responsible for reviewing scans and all related decisions, with the AI tool serving as an additional review to help detect serious issues earlier.\u003c/p\u003e \u003cp\u003e(3) Clostridioides difficile (C. diff) Classification Tool.\u003c/p\u003e \u003cp\u003eSummary: This co-developed (university and hospital) AI tool helps infection prevention teams to classify C. diff infections. It uses hospital data to automatically apply national definitions, saving time and reducing the risk of manual errors. The tool supports surveillance and reporting, but final decisions rest with the clinical team.\u003c/p\u003e \u003cp\u003eThese case studies were selected to reflect the different pathways through which healthcare delivery organisations acquire AI capabilities including in-house tool development, procurement of commercial AI products or via co-development in partnership with external organisations. For each case study, project teams completed the checklist using a think-aloud protocol. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] This allowed real-time feedback on the usability, relevance, and clarity of the checklist items. Feedback was analysed thematically and used to revise checklist content, clarify definitions, and improve logical flow. The case studies tested the framework\u0026rsquo;s applicability across clinical, operational, and research contexts and demonstrated its flexibility and utility in surfacing risks, clarifying roles, and informing decision-making. Findings from this stage, combined with earlier insights, informed the final version of the governance framework.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003ch2\u003e A practice-oriented AI governance framework for healthcare organisations\u003c/h2\u003e\u003cp\u003eThe AI governance framework presented here provides a structured, organisational-level approach to the safe and responsible integration of AI in healthcare (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). It comprises three components: (1) a set of ethics and governance principles; (2) oversight mechanism (based on digital maturity); and (3) a review checklist to support consistent decision-making about the implementation and operation of AI tools across clinical and operational contexts. In the following sections, we present each component, first describing it and then outlining its application at the study site.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eEthics and governance principles\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e Our analysis defined 30 ethics and governance principles for healthcare organisations to consider when implementing AI (see Supplementary file 1, Appendix 5). Each organisation will need to review and prioritise these principles based on their local context, risk profile, and existing governance structures.\u003c/p\u003e \u003cp\u003eAt the study site, the first step involved mapping their existing organisational principles, followed by aligning each one with corresponding checklist questions (see Supplementary file 1, Appendix 4 \u0026ndash; mapping template; Supplementary file 2 \u0026ndash; principle/checklist mapping). This process supported the operationalisation of the principles within the framework and ensured that they were addressed either through the checklist or through existing governance mechanisms. It also ensured that these principles were not treated as abstract elements, but embedded into decision-making, implementation, and evaluation.\u003c/p\u003e \u003cp\u003e During the workshop (Stage 4), stakeholders reviewed a consolidated list of ethics principles and prioritised those that were either poorly represented in existing policies or that posed new risks in the context of AI. These included accountability, consent, cultural appropriateness, fairness, transparency, sustainability, and trust. Some, such as dignity and wellbeing, were considered to already align well with existing organisational principles. Participant discussions highlighted the challenge of distinguishing between principles already embedded in governance processes and those requiring AI-specific adoption. Participants stressed the need to differentiate what can be managed at the organisational level (e.g. privacy) from broader societal risks (e.g. regulation). The process was considered valuable in understanding these considerations and identifying practical pathways for operationalising high-level principles within an AI governance framework tailored to a local context.\u003c/p\u003e \u003cp\u003e \u003col start=2\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eOversight mechanism\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThree dominant mechanisms for organisational oversight emerged from our analysis. These included: (1) creating a dedicated AI governance committee; (2) expanding an existing governance committee to include AI oversight; and (3) integrating AI governance into existing non-AI specific governance processes. To account for the varying levels of digital maturity and resourcing across healthcare organisations, we developed a tiered approach for organisational oversight.\u003c/p\u003e\n\u003ch3\u003eA. Digitally mature organisations\u003c/h3\u003e\n\u003cp\u003eFor digitally mature organisations with adequate infrastructure and workforce capabilities, a dedicated AI governance committee was suggested (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This high-level, multidisciplinary body would provide centralised oversight of AI evaluation, approval, and monitoring, supported by a broader advisory group and an operations group responsible for implementation. The advisory group provides subject matter expertise as required for specific AI tools and makes recommendations to the governance committee. The operations group supports applicants to complete the checklist, conducts risk stratification for AI tools, responds to queries, and provides the administrative functions that enable the governance committee and advisory group to fulfil their oversight roles (Supplementary file 1, Appendix 2). The advisory group is shown with a dotted line in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e to indicate its advisory function when required. An AI use register can also be established (Supplementary file 1, Appendix 6) as part of this oversight function to promote transparency and accountability, and to support ongoing monitoring, evaluation, and decommissioning of AI tools.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eB. Developing digital capacity\u003c/h2\u003e \u003cp\u003eFor healthcare organisations with developing digital capacity, AI oversight can be delegated to an existing multidisciplinary team, such as a digital health or clinical innovations committee, with clearly defined roles and domain-specific advisors (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This committee would use the AI review checklist to guide evaluations and provide formal recommendations to an existing governance body, such as an executive or digital health committee. An AI advisory group could also be established to support the existing governance committee.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eC. Digitally mature organisations\u003c/h3\u003e\n\u003cp\u003eFor healthcare organisations with limited digital capacity, where both infrastructure and workforce capabilities are constrained, AI governance can be initially embedded within existing oversight mechanisms such as clinical governance, digital health, ethics, and procurement processes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). A gap analysis can then be conducted to identify any unaddressed responsibilities, with the AI review checklist serving as the central evaluation tool. This approach would support practical, distributed governance by leveraging existing resources, while maintaining alignment with safety, ethical, and regulatory requirements.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eStakeholders noted that healthcare organisations vary widely in their digital capability, which influences both the scope of AI they can safely adopt and the governance structures they can sustain. For organisations with limited digital capacity, embedding AI oversight within existing processes offered a pragmatic starting point. These organisations were viewed as often having constrained infrastructure and limited specialist expertise and were best positioned to focus on targeted AI use cases that were operationally critical or lower in risk. This approach enabled proportionate oversight while avoiding the significant administrative and workforce investment required to govern more complex AI tools. It also recognised that, at early stages of maturity, organisations may not yet be equipped to make consistent and well-informed decisions about higher-risk AI applications.\u003c/p\u003e \u003cp\u003eIn contrast, organisations with more mature digital capability were seen as being better equipped to support a broader and more proactive approach to AI governance. These services typically had stable infrastructure, defined digital governance systems, and access to specialist expertise across data governance, privacy, cybersecurity, clinical governance, and informatics. As they transitioned to a more embedded governance model, a dedicated AI advisory or governance committee could continue to set organisational priorities, assess strategic fit, and provide structured oversight of safety, ethics, and implementation. Mature organisations were also viewed as having the resources required to ensure that AI tools progressed efficiently through approval pathways, were monitored appropriately, and were integrated into existing governance functions over time. This was viewed as enabling a longer-term framework in which AI governance became embedded across executive and operational roles, supported by an executive-level function guiding prioritisation and organisational alignment.\u003c/p\u003e\n\u003ch3\u003eD. AI oversight mechanism at the study site\u003c/h3\u003e\n\u003cp\u003e At the study site, internal stakeholders supported establishing a high-level, multidisciplinary AI governance committee during the early stages of AI integration. This committee would provide centralised oversight for ethical, clinical, operational, legal, and technical considerations, with a phased plan to embed responsibilities within existing structures over time. A direct reporting line to the Executive Committee (the main body exercising managerial responsibility for the performance of all activities within the organisation) was favoured to ensure visibility, organisational alignment, and resourcing. Two alternative options, reporting to the Digital Health Steering Committee (accountable for overseeing and setting the overall strategic direction across all digital health programs of work) or directly to the Board (effectively bypassing the Executive Committee), were considered less effective due to perceived gaps in focus or proximity to organisation-wide operational decision-making. While a dedicated AI governance committee was considered suitable in the early stages of adoption, it was recognised that over time these responsibilities should transition into existing governance and executive functions as organisational capability, confidence, and experience in reviewing AI tools increases.\u003c/p\u003e \u003cp\u003eTo support implementation, a smaller \u0026lsquo;AI operations group\u0026rsquo; was proposed. This group would provide technical and logistical support, assist with completion of the AI review checklist, and manage the day-to-day coordination of AI governance activities. A broader advisory group of subject matter experts (e.g. clinical, legal, cybersecurity, consumer representatives, etc.) would be engaged per use case.\u003c/p\u003e \u003cp\u003eStakeholders noted that many existing governance committees lacked the necessary breadth of expertise to manage AI-related risks, particularly in data science, ethics, and emerging and evolving regulation. Our governance framework aims to address this gap by supporting AI governance literacy, creating structured approval pathways, and enabling robust monitoring. To support this, organisations are recommended to develop tailored internal policies and terms of reference that clearly define the governance committee\u0026rsquo;s scope, functions, and decision-making authority. The committee\u0026rsquo;s proposed functions, reporting lines, and membership serve as a practical foundation for governing clinical and operational AI tools within a care delivery setting (see Supplementary file 1, Appendix 1).\u003c/p\u003e \u003cp\u003eTo support proportionate oversight and avoid unnecessary administrative burden, several participants recommended introducing a preliminary screening or \u0026ldquo;gating\u0026rdquo; process. This approach enables organisations to distinguish between high and low risk AI activities based on a small number of objective criteria. For example, a project may be referred for formal AI governance review if it involves identifiable data, requires integration into hospital IT infrastructure, or requires organisational resources such as ongoing IT support. Tools that do not meet these criteria can generally proceed through existing research or quality improvement processes. Integrating a short gating step into AI intake processes can help ensure that governance attention is directed to areas of greatest organisational risk and impact (see Supplementary file 1, Appendix 3).\u003c/p\u003e \u003cp\u003e \u003col start=3\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eAI review checklist\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eA central component of the framework is the AI review checklist, a structured evaluation tool designed to guide consistent, transparent, and evidence-informed oversight of AI tools across clinical and operational domains. It supports both research and operational pathways and aims to ensure that the use of AI aligns with ethical standards, organisational priorities, and regulatory requirements, while being adaptable and flexible in its application. The checklist was developed through the synthesis of 10 AI governance and implementation frameworks, [\u003cspan additionalcitationids=\"CR20 CR21 CR22 CR23 CR24 CR25 CR26 CR27\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] identified through the scoping review [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], and key domains from the informant interviews and document analysis. Each question was mapped to ensure broad coverage of key checklist items (Supplementary file 1, Appendix 3). Checklist items were iteratively refined based on informant interviews, mapping of ethics and governance principles (Supplementary file 2), and validation on the AI case studies.\u003c/p\u003e \u003cp\u003eThe final checklist comprises seven core domains and includes items relevant to both clinical and operational applications of AI (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These domains reflect key stages in the AI lifecycle, to support comprehensive oversight from conception through to decommissioning (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhile the checklist was designed to be comprehensive, it is not intended as a one-size-fits-all solution. Rather, it should be applied in a way that reflects and compliments the specific governance structures, policies, and decision-making processes already established within the local context of the healthcare organisation. The use of the checklist should be guided by the assessed risk level of the AI tool, with higher-risk applications subject to more rigorous scrutiny and lower-risk tools using a proportionate approach. This flexible, context-specific use of the checklist is intended to support integration into existing workflows without introducing unnecessary duplication or burden.\u003c/p\u003e \u003cp\u003e Each item in the checklist is linked to relevant ethical and governance principles (e.g. transparency, fairness, accountability), ensuring these concepts are embedded in the assessment process rather than treated as separate considerations. Applicants are asked to complete only the questions relevant to their proposed pathway (research, clinical/operational, or both), with clear instructions and subheadings for readability.\u003c/p\u003e \u003cp\u003eFeedback from the three case studies led to refinements including clearer definitions, improved logic flow, and the ability to reference existing documentation rather than duplicating content. The checklist now serves as both a practical tool for governance and an educational mechanism to build AI literacy among developers and decision-makers (see Supplementary file 1, Appendix 3 for complete AI review checklist).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eAI governance checklist core domains.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDomain\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePurpose\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBackground details\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOutlines AI tool/ company information.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProblem identification\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDefines rationale, value, and strategic alignment.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDesign and development\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDetails model type, intended use, IP, stakeholder involvement.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValidation and safety\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCovers performance metrics, risk, clinical safety, regulatory compliance.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eImplementation and Integration\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eReviews workflow, UI, technical integration, training, and contingency planning.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonitoring and evaluation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnsures ongoing review, equity analysis, safe removal, and lifecycle tracking.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFunding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDetails funding for development, implementation, and ongoing costs.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study describes the design, testing, and refinement of a practical AI governance framework tailored to the needs of healthcare delivery organisations. The framework addresses a persistent gap between high-level AI ethics and governance principles and the real-world processes required to govern the safe and responsible use of AI in healthcare. It includes a mapped set of ethics and governance principles, possible committee structures with defined oversight, and a practical checklist for reviewing AI tools aligned to key stages in the AI lifecycle. By combining a comprehensive scoping review with stakeholder input and real-world testing, the framework connects normative and procedural approaches to AI governance, supporting translation into practice.\u003c/p\u003e \u003cp\u003eOur findings align with previous research that has characterised AI governance as fragmented, heavily principle-based, and largely lacking in operational tools. [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] While many current frameworks present high-level concepts such as transparency, fairness, and accountability, few provide a clear path for applying these concepts in healthcare delivery environments. The limited uptake of governance tools may reflect this disconnect, as well as uncertainty among healthcare leaders about where governance responsibility lies. Prior work has highlighted the challenge of operationalising governance in health systems, particularly when governance expectations are vague or insufficiently contextualised. Pyone et al., in their systematic review of health systems governance, highlighted many governance frameworks that lacked practical guidance for implementation at different levels of the health system. [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] Similarly, Brinkerhoff and Bossert emphasised the disparity between formal governance structures \u0026ldquo;and the extent to which these are actually put into practice.\u0026rdquo; [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eThis study responds to those concerns by defining a structured governance framework with aligned tools and processes that can be integrated into existing organisational processes. Accounting for AI lifecycle stages (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) enables continuous oversight throughout the development, implementation, and use of AI tools. Each stage is supported by relevant principles, ensuring that governance is both continuous and proportional to risk. This lifecycle-based approach supports iterative, context-specific governance models that respond to the evolving nature of AI use in clinical and operational settings. [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eOur stakeholder consultations confirmed that current governance structures do not routinely account for AI-specific considerations. Participants reported that project approval pathways vary across clinical, research, and digital domains, with little clarity around who is responsible for assessing AI-related risks. These insights support those of Morley et al. and Jobin et al., who have highlighted the challenges of operationalising ethics principles across complex healthcare systems. [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e] Key gaps identified by stakeholders include insufficient attention to bias mitigation, post-deployment safety, integration with clinical workflows, and the ability to contest or override AI-generated outputs. The review checklist was developed to address these needs and was validated on three diverse AI tools with relevant stakeholders, including frontline and operational users who assessed its usability and relevance within real-world organisational conditions. This validation process demonstrated the checklist\u0026rsquo;s flexibility across different acquisition pathways (internal development, commercial procurement, and co-development) and its ability to elicit detailed, structured discussion about model purpose, performance, implementation challenges, and ongoing monitoring. Unlike more conceptual self-assessment tools, this checklist directly supports project-level review and approval. Participants found it useful both as a decision-making aid and as a tool to build AI literacy within multidisciplinary teams.\u003c/p\u003e \u003cp\u003eImportantly, the framework includes defined roles and committee structures that align with existing healthcare governance models. The AI governance committee oversight should function similar to that of quality, safety, and innovation governance committees. This design reflects a deliberate strategy to embed AI oversight within existing organisational processes, reducing duplication and ensuring alignment with broader service priorities. Here, the framework aims to promote both integration and validity.\u003c/p\u003e \u003cp\u003eAs organisations gain experience with AI and build capability to assess and manage associated risks, there is growing support for AI governance responsibilities to transition from standalone committees into existing executive and governance functions. With appropriate upskilling, staff across domains such as clinical governance, research ethics, data governance, privacy, and cybersecurity will be able to incorporate AI considerations within their usual decision-making processes. An executive role, such as a Chief AI Officer or equivalent, could oversee the prioritisation, need, and suitability of AI tools, working alongside other executive functions to ensure consistent, efficient, and accountable decision-making across the organisation.\u003c/p\u003e \u003cp\u003eThe framework also addresses equity and inclusion through multiple mechanisms. The inclusion of cultural appropriateness, fairness, and transparency as core principles was prioritised by stakeholders, particularly in relation to diverse data governance and health equity considerations. The framework recommends consulting Aboriginal and/or Torres Strait Islander advisors and equity experts during project planning, recognising that ethical risks and opportunities are not evenly distributed across populations. This aligns with strength-based, community-led approaches to digital health governance. [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eWhile the framework is grounded in the Australian health system, its components are designed to be adaptable across diverse healthcare contexts. Its development draws on widely accepted ethics and governance principles, and its tools can be localised to reflect different regulatory requirements and institutional structures and processes. The modular design of the checklist and committee structure allows for phased implementation or adaptation to resource-constrained settings. This flexibility is essential, given the rapidly evolving nature of AI technologies and the varied levels of digital maturity across healthcare organisations. [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eSeveral limitations warrant discussion. The framework was developed and tested within a single healthcare organisation and will likely require modification in other contexts. Although the checklist was validated on three AI use cases, a larger sample across multiple settings would strengthen generalisability. Future research should evaluate the framework\u0026rsquo;s implementation over time, including its impact on decision-making quality, project outcomes, and staff engagement.\u003c/p\u003e \u003cp\u003eNotwithstanding these limitations, this study provides one of the first practice-oriented frameworks for AI governance in healthcare that has been developed and validated within a real-world healthcare organisation. The framework is designed to be adapted by other healthcare organisations seeking to enhance their governance processes to ensure the safe and responsible use of AI. It contributes a practical approach to the emerging field of digital health governance, grounded in systems thinking, real-world constraints, and inclusive design. As organisations continue to explore the potential of AI, frameworks such as this one will be essential to guide adoption and manage risk. AI governance must move beyond principles and into systems, processes, and shared accountabilities. This study contributes an applied governance approach that can support transformation in healthcare without compromising safety or responsibility.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eCompeting Interests\u003c/h2\u003e\u003cp\u003eSS, AM, and EP were employed by Alfred Health and participated in the project steering committee but were not involved in data collection, analysis, or interpretation. Their involvement in recruitment was limited to forwarding the study invitation to potential participants. SS participated in the board and chief executive officer interviews. SF, AW, FM, and EC have no professional or personal affiliation with Alfred Health.\u003c/p\u003e\u003ch2\u003eFunding declaration\u003c/h2\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eThe study was supported by the Digital Health CRC Limited, which is funded under the Australian Commonwealth\u0026rsquo;s Cooperative Research Centres program, an Australian Government initiative for collaboration between university researchers and the industry.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eFM and EC conceptualised the study. SF, FM, and AW designed and conducted the study, with AM, SS and EP assisting with participant recruitment at the study site. SF and AW undertook data analysis. SF, FM, and AW drafted the manuscript with input from all authors. All authors provided revisions for intellectual content. All authors have approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe wish to acknowledge the participants from the study site as well as key informants from healthcare, academia, government, and professional associations who gave their time to participate in the study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated during this study will not be made not publicly available for privacy reasons but requests to the corresponding author will be considered on a case-by-case basis.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eMagrabi, F., et al. (2023). \"Automation in Contemporary Clinical Information Systems: A Survey of AI in Healthcare Settings.\" Yearb Med Inform 32(1): 115\u0026ndash;126.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChallen, R., et al. (2019). \"Artificial intelligence, bias and clinical safety.\" BMJ Qual Saf 28(3): 231\u0026ndash;237.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLyell, D., et al. (2023). \"More than algorithms: an analysis of safety events involving ML-enabled medical devices reported to the FDA.\" J Am Med Inform Assoc 30(7): 1227\u0026ndash;1236.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavenport TH, Glaser JP. Factors governing the adoption of artificial intelligence in healthcare providers. Discov Health Syst. 2022;1(1):4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSolanki P, Grundy J, Hussain W. Operationalising ethics in artificial intelligence for healthcare: A framework for AI developers. AI and Ethics. 2023;3(1):223\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoiera E, Fraile Navarro D. AI as an ecosystem: ensuring generative AI Is safe and effective. NEJM AI. 2024 Sept;1(9):1\u0026ndash;4. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/AIp2400611\u003c/span\u003e\u003cspan address=\"10.1056/AIp2400611\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAustralian Commission on Safety and Quality in Health Care. Literature Review and Environmental Scan Final Report: AI Implementation in Hospitals: Legislation, Policy, Guidelines and Principles, and Evidence about Quality and Safety.; 2024 29 May 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIhaddouchen I, Buijsman S, Pozzi G, Van de Sande D, Reis A, Townsend R, et al. Responsible artificial intelligence in healthcare: a systematic review on the use of ethical principles in the development and deployment of artificial intelligence. BMJ Digital Health \u0026amp; AI. 2025;1:e000086.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarah Magrabi LB, Kalissa Brooke-Cowden, Tamasha Jayawardena, Amy Wang, Enrico Coiera, Yves Saint James Aquino, Emma Frost, Stacy Carter, Carolyn Adams, Alaa Almohanna. AI Implementation in Hospitals: Legislation, Policy, Guidelines and Principles, and Evidence about Quality and Safety. 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavies P. The SAGE Dictionary of Social Research Methods. 2006 2025/03/18. London: SAGE Publications, Ltd. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://methods.sagepub.com/dict/edvol/the-sage-dictionary-of-social-research-methods/toc\u003c/span\u003e\u003cspan address=\"https://methods.sagepub.com/dict/edvol/the-sage-dictionary-of-social-research-methods/toc\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care. 2007;19(6):349\u0026ndash;57.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreeman S, Wang A, Saraf S, Potts E, McKimm A, Coiera E, Magrabi F. Developing an AI Governance Framework for Safe and Responsible AI in Health Care Organizations: Protocol for a Multimethod Study. JMIR Res Protoc. 2025;14:e75702. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/75702\u003c/span\u003e\u003cspan address=\"10.2196/75702\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 40720809; PMCID: PMC12340452.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlfred Health. Alfred Health Annual Report 2023-24. 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmy Wang, Sam Freeman, Farah Magrabi et al. Governance for safe and responsible AI in healthcare organisations: A scoping review of frameworks, 19 August 2025, PREPRINT (Version 1) available at Research Square [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21203/rs.3.rs-7209096/v1]\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-7209096/v1]\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalaweh M, Fidler C, McRobb S. Integrating the Grounded Theory Method and Case Study Research Methodology Within IS Research: A Possible 'Road Map'2008. 165 p.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreeman S, Wang A, Magrabi F. Key Considerations for Governing Safe and Responsible Use of AI in Healthcare. Stud Health Technol Inform. 2025;329:1367\u0026ndash;1371. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3233/SHTI251062\u003c/span\u003e\u003cspan address=\"10.3233/SHTI251062\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 40776080.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAidoc. Aidoc 2025 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.aidoc.com/solutions/radiology/\u003c/span\u003e\u003cspan address=\"https://www.aidoc.com/solutions/radiology/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEricsson KA, Simon HA. Protocol Analysis: Verbal Reports as Data: The MIT Press; 1993.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu X, Cruz Rivera S, Moher D, Calvert MJ, Denniston AK. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Nat Med. 2020;26(9):1364\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVasey B, Nagendran M, Campbell B, Clifton DA, Collins GS, Denaxas S, et al. Reporting guideline for the early stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI. Bmj. 2022;377:e070904.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLekadir K, Frangi AF, Porras AR, Glocker B, Cintas C, Langlotz CP, et al. FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare. Bmj. 2025;388:e081554.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eU.S. Department of Health and Human Services. Trustworthy AI (TAI) Playbook 2021 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://digitalgovernmenthub.org/library/trustworthy-ai-tai-playbook/\u003c/span\u003e\u003cspan address=\"https://digitalgovernmenthub.org/library/trustworthy-ai-tai-playbook/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoalition for Health AI. Responsible AI Guide 2024 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://chai.org/wp-content/uploads/2025/03/CHAI_Responsible-AI-Guide_03202025.pdf\u003c/span\u003e\u003cspan address=\"https://chai.org/wp-content/uploads/2025/03/CHAI_Responsible-AI-Guide_03202025.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEconomou-Zavlanos NJ, Bessias S, Cary MP, Jr, Bedoya AD, Goldstein BA, Jelovsek JE, et al. Translating ethical and quality principles for the effective, safe and fair development, deployment and use of artificial intelligence technologies in healthcare. Journal of the American Medical Informatics Association. 2023;31(3):705\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhittaker R, Dobson R, Jin CK, Style R, Jayathissa P, Hiini K, et al. An example of governance for AI in health services from Aotearoa New Zealand. NPJ Digit Med. 2023;6(1):164.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollins GS, Dhiman P, Andaur Navarro CL, Ma J, Hooft L, Reitsma JB, et al. Protocol for development of a reporting guideline (TRIPOD-AI) and risk of bias tool (PROBAST-AI) for diagnostic and prognostic prediction model studies based on artificial intelligence. BMJ Open. 2021;11(7):e048008.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiao F, Adelaine S, Afshar M, Patterson BW. Governance of Clinical AI applications to facilitate safe and equitable deployment in a large health system: Key elements and early successes. Front Digit Health. 2022;4:931439.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReddy S, Rogers W, Makinen VP, Coiera E, Brown P, Wenzel M, et al. Evaluation framework to guide implementation of AI systems into healthcare settings. BMJ Health Care Inform. 2021;28(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMittelstadt B. Principles alone cannot guarantee ethical AI. Nature Machine Intelligence. 2019;1(11):501\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMesko B. The role of artificial intelligence in precision medicine. Expert Review of Precision Medicine and Drug Development. 2017;2(5):239\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePyone T, Smith H, van den Broek N. Frameworks to assess health systems governance: a systematic review. Health Policy and Planning. 2017;32(5):710\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrinkerhoff DW, Bossert TJ. Health governance: principal\u0026ndash;agent linkages and health system strengthening. Health Policy and Planning. 2013;29(6):685\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePapagiannidis E, Mikalef P, Conboy K. Responsible artificial intelligence governance: A review and research framework. The Journal of Strategic Information Systems. 2025;34(2):101885.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreenhalgh T, Wherton J, Papoutsi C, Lynch J, Hughes G, A'Court C, et al. Beyond Adoption: A New Framework for Theorizing and Evaluating Nonadoption, Abandonment, and Challenges to the Scale-Up, Spread, and Sustainability of Health and Care Technologies. J Med Internet Res. 2017;19(11):e367.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorley J, Floridi L, Kinsey L, Elhalal A. From What to How: An Initial Review of Publicly Available AI Ethics Tools, Methods and Research to Translate Principles into Practices. Science and Engineering Ethics. 2020;26(4):2141\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJobin A, Ienca M, Vayena E. The global landscape of AI ethics guidelines. Nature Machine Intelligence. 2019;1(9):389\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCordes A, Bak M, Lyndon M, Hudson M, Fiske A, Celi LA, et al. Competing interests: digital health and indigenous data sovereignty. NPJ Digit Med. 2024;7(1):178.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim GJ, Namkoong K. Developing the Digital Health Communication Maturity Model: Systematic Review. J Med Internet Res. 2025;27:e68344.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"artificial intelligence, governance, healthcare, patient safety, ethics, oversight, framework, implementation, digital health","lastPublishedDoi":"10.21203/rs.3.rs-8302125/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8302125/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eNew AI technologies are being rapidly adopted by healthcare professionals, yet existing organisational governance often lacks the processes necessary to oversee their safe and responsible use. Previous AI governance frameworks have focused on high-level AI ethics principles, leaving healthcare organisations struggling to translate them into practice, assess risk, and embed AI oversight into existing processes. We developed a practice-oriented AI governance framework that unifies ethics and governance principles, delivers tiered oversight aligned with digital maturity, and incorporates a structured review checklist for assessing AI tools. Our multimethod approach drew on a scoping review, document analysis, and semi-structured interviews. The resulting framework was validated through stakeholder workshops and applied to exemplar AI tools, demonstrating the checklist\u0026rsquo;s usability, relevance, and capacity to identify risks and guide decision-making. The framework offers an actionable approach linking principles to operational governance, enabling proportionate oversight, integration with existing processes, and ongoing monitoring, addressing a gap in a field where guidance remains largely conceptual.\u003c/p\u003e","manuscriptTitle":"From Principles to Practice: An Actionable Framework for AI Governance in Healthcare Organisations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-16 12:02:46","doi":"10.21203/rs.3.rs-8302125/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"355178ac-7806-4222-9d39-2014fb8a5b9f","owner":[],"postedDate":"December 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":59375778,"name":"Scientific community and society/Business and industry"},{"id":59375779,"name":"Health sciences/Health care"},{"id":59375780,"name":"Health sciences/Medical research"},{"id":59375781,"name":"Scientific community and society/Scientific community"}],"tags":[],"updatedAt":"2025-12-28T18:53:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-16 12:02:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8302125","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8302125","identity":"rs-8302125","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}