Risk assessment tools are increasingly used in policing to enhance decision-making accuracy and objectivity; yet their implementation has raised significant ethical concerns regarding issues of bias, transparency, and governance. This paper examines the ethical complexities of risk assessment tools through an analysis of four instruments: the harm assessment risk tool, previously developed and used by Durham Constabulary; the Active Risk Management System (ARMS), used across all police forces in England and Wales; the Offender Assessment System, used to profile risk of reoffending by probation services in the UK; and the Correctional Offender Management Profiling for Alternative Sanctions, a widely researched tool deployed in US corrections contexts whose ethical challenges are directly relevant to tools now entering policing practice. A thematic framework identifies 10 key challenges for the field, including disparities in accuracy metrics, fairness trade-offs, bias linked to demographics and social identity, and retraining. The paper contextualizes these issues within influential roles, including tool developers, decision-makers, and oversight committees. The credible risk of ethical harms arising from the use of risk assessment tools underscores the need for rigorous validation, transparency, and adaptive governance to minimize these risks. This paper arises from a meeting of an interdisciplinary working group convened at Ethox, University of Oxford, comprising academics in philosophy, law, psychology, psychiatry, and criminology, as well as police stakeholders.

BACKGROUND:Because confirmatory clinical trials are costly, large-scale endeavors, the choice of their design carries significant weight. While the current methodological landscape offers tools to address residual pre-trial uncertainty through prespecified adaptations to design elements, design choices remain frequently constrained by prevailing orthodoxies. METHODS:We examine how unacknowledged design uncertainties (e.g., sample size calculations based on incorrect effect size or variance estimates) impact the statistical, ethical, and resource-stewardship goals of confirmatory trials. We contrast how fixed and adaptive design strategies handle these uncertainties, detailing the operational safeguards, firewalls, and trade-offs (e.g., complexity, operational bias) required to preserve trial integrity. RESULTS:Strict adherence to default templates prevents stakeholders from matching the design strategy to the specific complexities of the research question. We observe that the gold standard status of fixed designs often obscures their limitations in handling design uncertainties. A rigid adherence to these conventions can ethically hinder a trial's ability to deliver conclusive results. Conversely, while adaptive designs can offer tools to efficiently reduce uncertainty, we acknowledge that adaptive elements are not without cost; their implementation requires rigorous safeguards to manage specific risks to trial integrity and potential operational biases. CONCLUSION:To better align clinical research with its ethical and scientific mandates, we issue two calls to action. First, trial designers, including statisticians, must clearly articulate how target and design uncertainties impact ethical obligations, promoting an open-minded evaluation of both fixed and adaptive methods. Second, stakeholders must foreground ethical considerations during design selection, requiring explicit justification for the use-or non-use-of adaptive elements. The ethical path forward is not to default to the old or blindly adopt the new, but to explicitly justify the chosen design based on its responsiveness to the specific uncertainties of the trial. We posit that trial design must transition from a habit-based process to one where pre-trial uncertainties are openly discussed and addressed.

The UK government has recently committed to adopting a new policy-dubbed 'Martha's Rule'-which has been characterised as providing patients the right to rapidly access a second clinical opinion in urgent or contested cases. Support for the rule emerged following the death of Martha Mills in 2021, after doctors failed to admit her to intensive care despite concerns raised by her parents. We argue that framing this issue in terms of patient rights is not productive, and should be avoided. Insofar as the ultimate goal of Martha's Rule is the provision of a clinical service that protects patient safety, an approach that focuses on the obligations of the health system-rather than the individual rights of patients-will better serve this goal. We outline an alternative approach that situates rapid clinical review as part of a suite of services aimed at enhancing and protecting patient care. This approach would make greater progress towards addressing the difficult systemic issues that Martha's Rule does not, while also better engaging with the constraints of clinical practice.

The availability of medical imaging data is indispensable for medical advancements such as the development of new diagnostic tools, improved surgical navigation systems, and profiling for personalized medicine through imaging biomarkers. A central challenge in data governance is balancing the need to protect patient privacy with the necessity of promoting scientific innovation. Restrictive data governance policies could limit access to the large, high-quality datasets needed for such advancements. Conversely, lenient policies could compromise patient trust and lead to potential misuse of sensitive information. We call for a deliberate and well-considered approach to data governance, highlighting important factors that patients and healthcare organizations should consider when making imaging data governance decisions around data sharing.

Research has shown that up to 40% of dementia incidence can be accounted for by 12 modifiable lifestyle risk factors. However, the predictive value of these risks factors at an individual level remains uncertain. Ethical considerations that are typically invoked with respect to the disclosure of individual research results-beneficence and non-maleficence, respect for autonomy, and justice-do not provide conclusive justification for, or against, disclosing modifiable risk factors for future dementia to cognitively unimpaired research participants. We argue for a different approach to evaluating the disclosure of individual-level modifiable risk factors for dementia. Rather than focusing on individual-level disease prediction and prevention, we suggest that disclosure should be evaluated based on the impact of behavioral and lifestyle changes on current brain health.

INTRODUCTION:Diabetic retinopathy is one of the leading causes of avoidable blindness among adults globally, and screening programmes can enable early diagnosis and prevention of progression. Artificial intelligence (AI) diagnostic solutions have been developed to diagnose diabetic retinopathy. The aim of this review is to identify ethical concerns related to AI-enabled diabetic retinopathy diagnostics and enable future research to explore these issues further. METHODS:This is a narrative review that uses thematic analysis methods to develop key findings. We searched two databases, PubMed and Scopus, for papers focused on the intersection of AI, diagnostics, ethics, and diabetic retinopathy and conducted a citation search. Primary research articles published in English between 1 January 2013 and 14 June 2024 were included. From the 1878 papers that were screened, nine papers met inclusion and exclusion criteria and were selected for analysis. RESULTS:We found that existing literature highlights ensuring patient data has appropriate protection and ownership, that bias in algorithm training data is minimised, informed patient decision-making is encouraged, and negative consequences in the context of clinical practice are mitigated. CONCLUSIONS:While the technical developments in AI-enabled diabetic retinopathy diagnostics receive the bulk of the research focus, we found that insufficient attention is paid to how this technology is accessed equitably in different settings and which safeguards are needed against exploitative practices. Such ethical issues merit additional exploration and practical problem-solving through primary research. AI-enabled diabetic retinopathy screening has the potential to enable screening at a scale that was previously not possible and could contribute to reducing preventable blindness. It will only achieve this if ethical issues are emphasised, understood, and addressed throughout the translation of this technology to clinical practice.

In our recent paper ‘Trust and the Goldacre Review: Why TREs are not about trust’ we argue that trusted research environments (TREs) reduce the need for trust in the use and sharing of health data, and that referring to these data storage systems as ‘trusted’ raises a number of concerns. Recent replies to our paper have raised several objections to this argument. In this reply, we seek to build on the arguments presented in our original paper, address some of the misunderstanding of our position expressed in these replies, and sketch out where further research is needed.

Powered by ‘big health data’ and enormous gains in computing power, artificial intelligence and related technologies are already changing the healthcare landscape. Harnessing the potential of these technologies will necessitate partnerships between health institutions and commercial companies, particularly as it relates to sharing health data. The need for commercial companies to be trustworthy users of data has been argued to be critical to the success of this endeavour. I argue that this approach is mistaken. Our interactions with commercial companies need not, and should not, be based on trust. Rather, they should be based on confidence. I begin by elucidating the differences between trust, reliability, and confidence, and argue that trust is not the appropriate attitude to adopt when it comes to sharing data with commercial companies. I argue that what we really should want is confidence in a system of data sharing. I then provide an outline of what a confidence-worthy system of data sharing with commercial companies might look like, and conclude with some remarks about the role of trust within this system.

Research has shown that up to 40% of dementia incidence can be accounted for by 12 modifiable lifestyle risk factors. However, the predictive value of these risks factors at an individual level remains uncertain. Ethical considerations of beneficence and non-maleficence, respect for autonomy, and justice —on which most ethical guidelines for disclosing individual research results are based— fail to provide conclusive justification for, or against, disclosing modifiable risk factors for future dementia to cognitively unimpaired research participants. We argue for a different approach to evaluating the disclosure of individual-level modifiable risk factors for Alzheimer’s disease. Rather than focussing on individual-level disease prediction and prevention, we suggest that disclosure should be evaluated based on the impact of behavioural and lifestyle changes on current brain health.

A general obligation to make aggregate research results available to participants has been widely supported in the bioethics literature. However, dementia research presents several challenges to this perspective, particularly because of the fear associated with developing dementia. The authors argue that considerations of respect for persons, beneficence, and justice fail to justify an obligation to make aggregate research results available to participants in dementia research. Nevertheless, there are positive reasons in favor of making aggregate research results available; when the decision is made to do so, it is critical that a clear strategy for communicating results is developed, including what support will be provided to participants receiving aggregate research results.

UNSTRUCTURED Patients' health data are routinely collected and stored in hospitals for care delivery. If the data is made available for researchers, it can be secondarily used for artificial intelligence research and development. Data sharing and reuse brought new ethical concerns to healthcare organizations and the medical and research communities alike. In this article, we discuss the main ethical aspects that underpin data sharing decisions: health data ownership, patient consentment, privacy, personal data anonymization, and the purpose of data reuse. Three pillars should guide organizations in data sharing decisions: regulations, ethics, and support for increasing patient engagement. Efforts should be undertaken to explain the risks, benefits, and opportunities of data sharing to patients and society at large. There is a concern that algorithms' outputs are biased, which may be a result of biased datasets. To prevent this, datasets must include a representative cross-section of the population so that discoveries and innovations are generalizable to all people. A deeper understanding of health data reuse by different stakeholders are decisive for elevating population health informatics research.

To cite: Graham M, Milne R, Fitzsimmons P, et al. J Med Ethics Epub ahead of print: [please include Day Month Year]. doi:10.1136/ medethics-2022-108435 Wellcome Centre for Ethics and Humanities, University of Oxford, Oxford, UK Wellcome Connecting Science, Wellcome Genome Campus, Hinxton, UK Kavli Centre for Ethics, Science and the Public, Faculty of Education, University of Cambridge, Cambridge, UK Ethox Centre, Nuffield Department of Population Health, University of Oxford, Oxford, UK NIHR Oxford Biomedical Research Centre, Oxford, UK

[See related article at www.cmaj.ca/lookup/doi/10.1503/cmaj.212063][1] KEY POINTS Recent years have seen a dramatic increase in the collection, storage and curation of human genomic data for biomedical research. These data sets hold great promise for research into the genetic basis of disease, and

Objective: Functional neuroimaging may provide a viable means of assessment and communication in patients with Guillain-Barre Syndrome (GBS) mimicking the complete locked-in state. Functional neuroimaging has been used to assess residual cognitive function and has allowed for binary communication with other behaviourally non-responsive patients, such as those diagnosed with unresponsive wakefulness syndrome. We evaluated the potential application of functional neuroimaging using a clinical-grade scanner to determine if individuals with severe GBS retained auditory function, command following, and communication. Methods: Fourteen healthy participants and two GBS patients were asked to perform motor imagery and spatial navigation imagery tasks while being scanned using functional magnetic resonance imaging. The GBS patients were also asked to perform additional functional neuroimaging scans to attempt communication. Results: The motor imagery and spatial navigation task elicited significant activation in appropriate regions of interest for both GBS patients, indicating intact command following. Both patients were able to use the imagery technique to communicate in some instances. Patient 1 was able to use one of four communication tasks to answer a question correctly. Patient 2 was able to use three of seven communication tasks. However, two questions were incorrectly answered while a third was non-verifiable. Conclusions: GBS patients can respond using mental imagery and these responses can be detected using functional neuroimaging. Furthermore, these patients may also be able to use mental imagery to provide answers to `yes' or `no' questions in some instances. We argue that the most appropriate use of neuroimaging-based communication in these patients is to allow them to communicate wishes or preferences and assent to previously expressed decisions, rather than to facilitate decision-making.

Disorders of consciousness (DOC) continue to profoundly challenge both families and medical professionals. Once a brain-injured patient has been stabilized, questions turn to the prospect of recovery. However, what "recovery" means in the context of patients with prolonged DOC is not always clear. Failure to recognize potential differences of interpretation-and the assumptions about the relationship between health and well-being that underlie these differences-can inhibit communication between surrogate decisionmakers and a patient's clinical team, and make it difficult to establish the goals of care. The authors examine the relationship between health and well-being as it pertains to patients with prolonged DOC. They argue that changes in awareness or other function should not be equated to changes in well-being, in the absence of a clear understanding of the constituents of well-being for that particular patient. The authors further maintain that a comprehensive conception of recovery for patients with prolonged DOC should incorporate aspects of both experienced well-being and evaluative well-being.

Albrecht, G. L., and P. J. Devlieger. 1999. The disability paradox: High quality of life against all odds. Social Science & Medicine 48 (8):977–88. doi:10.1016/S02779536(98)00411-0. Bernham, J. L. 1999. How to get serious answers to the serious question: ‘How have you been?’: Subjective Quality of Life (QOL) as an individual experiential emergent construct. Bioethics 13 (3–4):272–87. doi:10.1111/1467-8519. 00156. Bruno, M.-A., J. L. Bernheim, D. Ledoux, F. Pellas, A. Demertzi, and S. Laureys. 2011. A survey on self-assessed well-being in a cohort of chronic locked-in syndrome patients: happy majority, miserable minority. BMJ Open 1 (1):e000039–9. doi:10.1136/bmjopen-2010-000039. Bruno, M.-A., M.-C. Nizzi, S. Laureys, and O. Gosseries. 2015. Consciousness in the locked-in syndrome. In The neurology of consciousness: Cognitive neuroscience and neuropathology, ed. S. Laureys, O. Gosseries, and G. Tononi, 2nd ed., 181–98. Oxford: Elsevier Academic Press. Cummins, R. 2003. Normative life satisfaction: Measurement issues and a homeostatic model. Social Indicators Research 64 (2):225–56. doi:10.1023/ A:1024712527648. Greenidge, K. 2020. My mother’s garden. The New York Times, Opinion, January 28, 2020. Majerus, S., H. Gill-Thwaites, K. Andrews, and S. Laureys. 2005. Behavioral evaluation of consciousness in severe brain damage. Progress in Brain Research 150:397–413. doi:10.1016/S0079-6123(05)50028-1. Nizzi, M.-C., A. Demertzi, O. Gosseries, M.-A. Bruno, F. Jouen, and S. Laureys. 2012. From armchair to wheelchair: How patients with a locked-in syndrome integrate bodily changes in experienced identity. Consciousness and Cognition 21 (1):431–7. doi:10.1016/j.concog.2011.10.010. Nizzi, M.-C., S. Tasigiorgos, M. Turk, C. Moroni, E. Bueno, and B. Pomahac. 2017. Psychological outcomes in face transplant recipients: A literature review. Current Surgery Reports 5 (10):26. doi:10.1007/s40137-017-0189-y. Nizzi, M.-C., V. Blandin, and A. Demertzi. 2018. Attitudes towards personhood in the locked-in syndrome: From thirdto first-person perspective and to inter-personal significance. Neuroethics 13 (2):193–201. doi:10.1007/ s12152-018-9375-6. Peterson, A., S. Aas, and D. Wasserman. 2021. What justifies the allocation of health care resources to patients with disorders of consciousness? AJOB Neuroscience 12 (2-3):127–139.

Many are calling for concrete mechanisms of oversight for health research involving artificial intelligence (AI). In response, institutional review boards (IRBs) are being turned to as a familiar model of governance. Here, we examine the IRB model as a form of ethics oversight for health research that uses AI. We consider the model's origins, analyze the challenges IRBs are facing in the contexts of both industry and academia, and offer concrete recommendations for how these committees might be adapted in order to provide an effective mechanism of oversight for health-related AI research.

A rapidly growing proportion of health research uses 'secondary data': data used for purposes other than those for which it was originally collected. Do researchers using secondary data have an obligation to disclose individual research findings to participants? While the importance of this question has been duly recognised in the context of primary research (ie, where data are collected from participants directly), it remains largely unexamined in the context of research using secondary data. In this paper, we critically examine the arguments for a moral obligation to disclose individual research findings in the context of primary research, to determine if they can be applied to secondary research. We conclude that they cannot. We then propose that the nature of the relationship between researchers and participants is what gives rise to particular moral obligations, including the obligation to disclose individual results. We argue that the relationship between researchers and participants in secondary research does not generate an obligation to disclose. However, we also argue that the biobanks or data archives which collect and provide access to secondary data may have such an obligation, depending on the nature of the relationship they establish with participants.

Neuroimaging research regularly yields "incidental findings": observations of potential clinical significance in healthy volunteers or patients, but which are unrelated to the purpose or variables of the study.