Women's views on using artificial intelligence in breast cancer screening: A review and qualitative study to guide breast screening services.

As breast screening services move towards use of healthcare AI (HCAI) for screen reading, research on public views of HCAI can inform more person-centered implementation. We synthesise reviews of public views of HCAI in general, and review primary studies of women's views of AI in breast screening. People generally appear open to HCAI and its potential benefits, despite a wide range of concerns; similarly, women are open towards AI in breast screening because of the potential benefits, but are concerned about a wide range of risks. Women want radiologists to remain central; oversight, evaluation and performance, care, equity and bias, transparency, and accountability are key issues; women may be less tolerant of AI error than of human error. Using our recent Australian primary study, we illustrate both the value of informing participants before collecting data, and women's views. The 40 screening-age women in this study stipulated four main conditions on breast screening AI implementation: 1) maintaining human control; 2) strong evidence of performance; 3) supporting familiarisation with AI; and 4) providing adequate reasons for introducing AI. Three solutions were offered to support familiarisation: transparency and information; slow and staged implementation; and allowing women to opt-out of AI reading. We provide recommendations to guide both implementation of AI in healthcare and research on public views of HCAI. Breast screening services should be transparent about AI use and share information about breast screening AI with women. Implementation should be slow and staged, providing opt-out options if possible. Screening services should demonstrate strong governance to maintain clinician control, demonstrate excellent AI system performance, assure data protection and bias mitigation, and give good reasons to justify implementation. When these measures are put in place, women are more likely to see HCAI use in breast screening as legitimate and acceptable.

Public values to guide childhood vaccination mandates: A report on four Australian community juries.

OBJECTIVE: Governments use vaccination mandates, of different degrees of coerciveness, to encourage or require childhood vaccination. We elicited the views of well-informed community members on the public acceptability of using childhood vaccination mandates in Australia. METHODS: Four community juries were conducted in Canberra, Launceston, Cairns and Melbourne, Australia between 2021 and 2022. We recruited 51 participants from diverse backgrounds, genders and ages through random digit dialling and social media. Two juries were held in metropolitan areas, and two in regional/rural settings. Outcome measures included jury verdicts and reasons in response to structured questions. RESULTS: All juries were concerned about collective protection and individual rights but prioritised the former over the latter. A majority in all juries supported mandates but juries disagreed with respect to the appropriate mandate types. All juries endorsed using the least restrictive or coercive means to encourage vaccination (providing incentives or education, e.g.) before imposing penalties such as financial losses and school exclusions. The overriding view was that it is fairer to place a direct burden on parents rather than children and that mandates should be designed to avoid inequitable impacts on less advantaged groups in society. Many jurors found conscientious objection acceptable as a controlled option for resolute refusers, provided that overall vaccination coverage remains high. CONCLUSION: This paper gives policymakers access to the reasons that Australians have for supporting or opposing different mandates under conditions of high knowledge, understanding and deliberation regarding policy options. Sustaining high rates of vaccination requires high levels of co-operation between governments, public health actors and the public. Our findings highlight the importance of considering public values in the design and implementation of vaccination mandates. PATIENT AND PUBLIC INVOLVEMENT: We sought input from individuals who did and did not vaccinate during the study design. The views and perspectives of nonvaccinating parents were presented in the evidence to juries. We deliberately excluded nonvaccinating individuals from participating, as the divisive and often hostile nature of the topic, and their minority status, made it difficult to ensure they would feel safe as members of the jury without overrepresenting their perspective in the sample. Two related projects engaged directly with these parents.

A qualitative study of financial concerns of mothers' of adults with intellectual and developmental disabilities.

Parents have long been concerned with the transition to adulthood of their children with intellectual and developmental disability (IDD) particularly with financial issues. The purpose of this study was to explore the financial concerns of mothers of adults with IDD. Five mothers of adults with IDD participated in a focus group to share their experiences, opinions, concerns, and challenges related to finances. Five themes emerged from the qualitative data analysis: (a) job-related difficulties, (b) living expenses, (c) access to essential services, (d), skills deficits and/or challenging behavior, and (e) long-term care. Mothers in general indicated tremendous financial burdens, loss of income, and expressed concerns about long-term care of their children with IDD. These themes are discussed and implications for practice and research are provided.

Evidence gaps and challenges in maintaining and increasing vaccine uptake: A Delphi survey with Australian stakeholders.

ISSUE ADDRESSED: Increasing and maintaining vaccination uptake is crucial for preventing and managing infectious diseases. In the context of the post-coronavirus disease 2019 (COVID-19) pandemic landscape, this paper examines the perceptions of immunisation implementers and policymakers to uncover the challenges and evidence gaps in routine immunisation efforts. METHODS: We conducted an online two-round modified Delphi survey with immunisation experts, senior public servants, policymakers, policy advisory groups, and representatives from peak bodies from across Australia. We asked respondents to outline what they see as the greatest challenges to increasing and maintaining uptake of recommended vaccines in Australia; the most difficult aspects of their work in vaccination; the largest evidence gaps in vaccine uptake; and the kinds of social and behavioural research they would like to see prioritised. RESULTS: The two most important challenges for increasing and maintaining vaccine uptake were effectively communicating the benefits of vaccines to parents and the public and ensuring accessible and affordable vaccination services. Participants strongly agreed that 'communication about the importance of vaccination' was the most difficult aspect of their work. Consistently important was the need to better engage specific population groups, such as culturally and linguistically diverse people, pregnant people, at risk cohorts, and health care providers. Social and behavioural research about 'how to effectively address hesitancy' was ranked highly among participants. CONCLUSIONS: Findings from this project help provide an understanding of the behavioural, social, ethical, and policy knowledge needs for immunisation policy and implementation in Australia. To respond to vaccine challenges, increase coverage and build public trust in vaccination, policymakers and governments should incorporate social research into vaccination programmes. SO WHAT?: Australia is preparing to launch a Centre for Disease Control. This study demonstrates the importance of integrating social, behavioural, ethical, and policy research into the fabric of this new enterprise. It underlines the need to capacity-build a workforce able to deliver high-quality research in these areas, address the needs of immunisation implementers and policymakers, and achieve good outcomes for Australians.

Patients' perspectives on quality and patient safety failures: lessons learned from an inquiry into transvaginal mesh in Australia.

BACKGROUND: Transvaginal mesh (TVM) surgeries emerged as an innovative treatment for stress urine incontinency and/or pelvic organ prolapse in 1996. Years after rapid adoption of these surgeries into practice, they are a key example of worldwide failure of healthcare quality and patient safety. The prevalence of TVM-associated harms eventually prompted action globally, including an Australian Commonwealth Government Senate Inquiry in 2017. METHOD: We analysed 425 submissions made by women (n = 417) and their advocates (n = 8) to the Australian Senate Inquiry, and documents from 5 public hearings, using deductive and inductive coding, categorisation and thematic analysis informed by three 'linked dilemmas' from healthcare quality and safety theory. We focused on women's accounts of: a) how harms arose from TVM procedures, and b) micro, meso and macro factors that contributed to their experience. Our aim was to explain, from a patient perspective, how these harms persisted in Australian healthcare, and to identify mechanisms at micro, meso and macro levels explaining quality and safety system failure. RESULTS: Our findings suggest three mechanisms explaining quality and safety failure: 1. Individual clinicians could ignore cases of TVM injury or define them as 'non-preventable'; 2. Women could not go beyond their treating clinicians to participate in defining and governing quality and safety; and. 3. Health services set thresholds for concern based on proportion of cases harmed, not absolute number or severity of harms. CONCLUSION: We argue that privileging clinical perspectives over patient perspectives in evaluating TVM outcomes allowed micro-level actors to dismiss women's lived experience, such that women's accounts of harms had insufficient or no weight at meso and macro levels. Establishing system-wide expectations regarding responsiveness to patients, and communication of patient reported outcomes in evaluation of healthcare delivery, may help prevent similar failures.

Facilitating public involvement in research about healthcare AI: A scoping review of empirical methods.

OBJECTIVE: With the recent increase in research into public views on healthcare artificial intelligence (HCAI), the objective of this review is to examine the methods of empirical studies on public views on HCAI. We map how studies provided participants with information about HCAI, and we examine the extent to which studies framed publics as active contributors to HCAI governance. MATERIALS AND METHODS: We searched 5 academic databases and Google Advanced for empirical studies investigating public views on HCAI. We extracted information including study aims, research instruments, and recommendations. RESULTS: Sixty-two studies were included. Most were quantitative (N = 42). Most (N = 47) reported providing participants with background information about HCAI. Despite this, studies often reported participants' lack of prior knowledge about HCAI as a limitation. Over three quarters (N = 48) of the studies made recommendations that envisaged public views being used to guide governance of AI. DISCUSSION: Provision of background information is an important component of facilitating research with publics on HCAI. The high proportion of studies reporting participants' lack of knowledge about HCAI as a limitation reflects the need for more guidance on how information should be presented. A minority of studies adopted technocratic positions that construed publics as passive beneficiaries of AI, rather than as active stakeholders in HCAI design and implementation. CONCLUSION: This review draws attention to how public roles in HCAI governance are constructed in empirical studies. To facilitate active participation, we recommend that research with publics on HCAI consider methodological designs that expose participants to diverse information sources.

How should artificial intelligence be used in Australian health care? Recommendations from a citizens' jury.

OBJECTIVE: To support a diverse sample of Australians to make recommendations about the use of artificial intelligence (AI) technology in health care. STUDY DESIGN: Citizens' jury, deliberating the question: "Under which circumstances, if any, should artificial intelligence be used in Australian health systems to detect or diagnose disease?" SETTING, PARTICIPANTS: Thirty Australian adults recruited by Sortition Foundation using random invitation and stratified selection to reflect population proportions by gender, age, ancestry, highest level of education, and residential location (state/territory; urban, regional, rural). The jury process took 18 days (16 March - 2 April 2023): fifteen days online and three days face-to-face in Sydney, where the jurors, both in small groups and together, were informed about and discussed the question, and developed recommendations with reasons. Jurors received extensive information: a printed handbook, online documents, and recorded presentations by four expert speakers. Jurors asked questions and received answers from the experts during the online period of the process, and during the first day of the face-to-face meeting. MAIN OUTCOME MEASURES: Jury recommendations, with reasons. RESULTS: The jurors recommended an overarching, independently governed charter and framework for health care AI. The other nine recommendation categories concerned balancing benefits and harms; fairness and bias; patients' rights and choices; clinical governance and training; technical governance and standards; data governance and use; open source software; AI evaluation and assessment; and education and communication. CONCLUSIONS: The deliberative process supported a nationally representative sample of citizens to construct recommendations about how AI in health care should be developed, used, and governed. Recommendations derived using such methods could guide clinicians, policy makers, AI researchers and developers, and health service users to develop approaches that ensure trustworthy and responsible use of this technology.

'Integrating Ethics and Equity with Economics and Effectiveness for newborn screening in the genomic age: A qualitative study protocol of stakeholder perspectives.

BACKGROUND: Newborn bloodspot screening is a well-established population health initiative that detects serious, childhood-onset, treatable conditions to improve health outcomes. With genomic technologies advancing rapidly, many countries are actively discussing the introduction of genomic assays into newborn screening programs. While adding genomic testing to Australia's newborn screening program could improve outcomes for infants and families, it must be considered against potential harms, ethical, legal, equity and social implications, and economic and health system impacts. We must ask not only 'can' we use genomics to screen newborns?' but 'should we'?' and 'how much should health systems invest in genomic newborn screening?'. METHODS: This study will use qualitative methods to explore understanding, priorities, concerns and expectations of genomic newborn screening among parents/carers, health professionals/scientists, and health policy makers across Australia. In-depth, semi-structured interviews will be held with 30-40 parents/carers recruited via hospital and community settings, 15-20 health professionals/scientists, and 10-15 health policy makers. Data will be analysed using inductive content analysis. The Sydney Children's Hospital Network Human Research Ethics Committee approved this study protocol [2023/ETH02371]. The Standards for Reporting Qualitative Research will guide study planning, conduct and reporting. DISCUSSION: Few studies have engaged a diverse range of stakeholders to explore the implications of genomics in newborn screening in a culturally and genetically diverse population, nor in a health system underpinned by universal health care. As the first study within a multi-part research program, findings will be used to generate new knowledge on the risks and benefits and importance of ethical, legal, social and equity implications of genomic newborn screening from the perspective of key stakeholders. As such it will be the foundation on which child and family centered criteria can be developed to inform health technology assessments and drive efficient and effective policy decision-making on the implementation of genomics in newborn screening.

Making recommendations to subsidize new health technologies in Australia: A qualitative study of decision-makers' perspectives on committee processes.

OBJECTIVES: To explore experiences of, and perspectives on, health technology assessment (HTA) processes used to produce recommendations about subsidizing new medicines, and medical technologies in Australia, from the perspectives of those experienced in these processes. METHODS: Semi-structured interviews with a diverse group of 18 informants currently or previously members of the Pharmaceutical Benefits Advisory Committee (PBAC) or the Medical Services Advisory Committee (MSAC). Participants were interviewed September 2021-February 2022. Transcripts were analyzed using reflexive thematic analysis. RESULTS: 3 major themes were identified: contrasting technical and decision-making stages, resisting reductionist approaches, and navigating decision-making trade-offs. Participants discussed the complexities of the evaluative HTA process, especially when considering uncertainty in the evidence. As part of the current process, a deliberative decision-making stage was considered essential, allowing a flexible approach to decision making to consider factors beyond strength and quality of quantifiable data in the technical evaluation. Participants acknowledged these less-quantifiable factors were sometimes considered implicitly or were difficult to describe and this, paired with commercial in confidence requirements, presented challenges with respect to the desire to increase transparency. CONCLUSION (S): As HTA processes for new medicines and medical technologies in Australia continue to be reviewed, the balance between retaining flexibility during deliberation, confidentiality for sponsors and the public's desire for greater transparency may be a fruitful area for continuing research.

Direct-to-consumer tests advertised online in Australia and their implications for medical overuse: systematic online review and a typology of clinical utility.

OBJECTIVES: The objective of this study is to map the range and variety of direct-to-consumer (DTC) tests advertised online in Australia and analyse their potential clinical utility and implications for medical overuse. DESIGN: Systematic online search of DTC test products in Google and Google Shopping. DTC test advertisements data were collected and analysed to develop a typology of potential clinical utility of the tests at population level, assessing their potential benefits and harms using available evidence, informed by concepts of medical overuse. RESULTS: We identified 484 DTC tests (103 unique products), ranging from $A12.99 to $A1947 in cost (mean $A197.83; median $A148.50). Using our typology, we assigned the tests into one of four categories: tests with potential clinical utility (10.7%); tests with limited clinical utility (30.6%); non-evidence-based commercial 'health checks' (41.9%); and tests whose methods and/or target conditions are not recognised by the general medical community (16.7%). Of the products identified, 56% did not state that they offered pretest or post-test consultation, and 51% did not report analytical performance of the test or laboratory accreditation. CONCLUSIONS: This first-in-Australia study shows most DTC tests sold online have low potential clinical utility, with healthy consumers constituting the main target market. Harms may be caused by overdiagnosis, high rates of false positives and treatment decisions led by non-evidence-based tests, as well as financial costs of unnecessary and inappropriate testing. Regulatory mechanisms should demand a higher standard of evidence of clinical utility and efficacy for DTC tests. Better transparency and reporting of health outcomes, and the development of decision-support resources for consumers are needed.

Childhood vaccine refusal and what to do about it: a systematic review of the ethical literature.

BACKGROUND: Parental refusal of routine childhood vaccination remains an ethically contested area. This systematic review sought to explore and characterise the normative arguments made about parental refusal of routine vaccination, with the aim of providing researchers, practitioners, and policymakers with a synthesis of current normative literature. METHODS: Nine databases covering health and ethics research were searched, and 121 publications identified for the period Jan 1998 to Mar 2022. For articles, source journals were categorised according to Australian Standard Field of Research codes, and normative content was analysed using a framework analytical approach. RESULTS: Most of the articles were published in biomedical journals (34%), bioethics journals (21%), and journals that carry both classifications (20%). Two central questions dominated the literature: (1) Whether vaccine refusal is justifiable (which we labelled 'refusal arguments'); and (2) Whether strategies for dealing with those who reject vaccines are justifiable ('response arguments'). Refusal arguments relied on principlism, religious frameworks, the rights and obligations of parents, the rights of children, the medico-legal best interests of the child standard, and the potential to cause harm to others. Response arguments were broadly divided into arguments about policy, arguments about how individual physicians should practice regarding vaccine rejectors, and both legal precedents and ethical arguments for vaccinating children against a parent's will. Policy arguments considered the normative significance of coercion, non-medical or conscientious objections, and possible reciprocal social efforts to offset vaccine refusal. Individual physician practice arguments covered nudging and coercive practices, patient dismissal, and the ethical and professional obligations of physicians. Most of the legal precedents discussed were from the American setting, with some from the United Kingdom. CONCLUSIONS: This review provides a comprehensive picture of the scope and substance of normative arguments about vaccine refusal and responses to vaccine refusal. It can serve as a platform for future research to extend the current normative literature, better understand the role of cultural context in normative judgements about vaccination, and more comprehensively translate the nuance of ethical arguments into practice and policy.

Multi-stakeholder preferences for the use of artificial intelligence in healthcare: A systematic review and thematic analysis.

INTRODUCTION: Despite the proliferation of Artificial Intelligence (AI) technology over the last decade, clinician, patient, and public perceptions of its use in healthcare raise a number of ethical, legal and social questions. We systematically review the literature on attitudes towards the use of AI in healthcare from patients, the general public and health professionals' perspectives to understand these issues from multiple perspectives. METHODOLOGY: A search for original research articles using qualitative, quantitative, and mixed methods published between 1 Jan 2001 to 24 Aug 2021 was conducted on six bibliographic databases. Data were extracted and classified into different themes representing views on: (i) knowledge and familiarity of AI, (ii) AI benefits, risks, and challenges, (iii) AI acceptability, (iv) AI development, (v) AI implementation, (vi) AI regulations, and (vii) Human - AI relationship. RESULTS: The final search identified 7,490 different records of which 105 publications were selected based on predefined inclusion/exclusion criteria. While the majority of patients, the general public and health professionals generally had a positive attitude towards the use of AI in healthcare, all groups indicated some perceived risks and challenges. Commonly perceived risks included data privacy; reduced professional autonomy; algorithmic bias; healthcare inequities; and greater burnout to acquire AI-related skills. While patients had mixed opinions on whether healthcare workers suffer from job loss due to the use of AI, health professionals strongly indicated that AI would not be able to completely replace them in their professions. Both groups shared similar doubts about AI's ability to deliver empathic care. The need for AI validation, transparency, explainability, and patient and clinical involvement in the development of AI was emphasised. To help successfully implement AI in health care, most participants envisioned that an investment in training and education campaigns was necessary, especially for health professionals. Lack of familiarity, lack of trust, and regulatory uncertainties were identified as factors hindering AI implementation. Regarding AI regulations, key themes included data access and data privacy. While the general public and patients exhibited a willingness to share anonymised data for AI development, there remained concerns about sharing data with insurance or technology companies. One key domain under this theme was the question of who should be held accountable in the case of adverse events arising from using AI. CONCLUSIONS: While overall positivity persists in attitudes and preferences toward AI use in healthcare, some prevalent problems require more attention. There is a need to go beyond addressing algorithm-related issues to look at the translation of legislation and guidelines into practice to ensure fairness, accountability, transparency, and ethics in AI.

Australian women's judgements about using artificial intelligence to read mammograms in breast cancer screening.

Objective: Mammographic screening for breast cancer is an early use case for artificial intelligence (AI) in healthcare. This is an active area of research, mostly focused on the development and evaluation of individual algorithms. A growing normative literature argues that AI systems should reflect human values, but it is unclear what this requires in specific AI implementation scenarios. Our objective was to understand women's values regarding the use of AI to read mammograms in breast cancer screening. Methods: We ran eight online discussion groups with a total of 50 women, focused on their expectations and normative judgements regarding the use of AI in breast screening. Results: Although women were positive about the potential of breast screening AI, they argued strongly that humans must remain as central actors in breast screening systems and consistently expressed high expectations of the performance of breast screening AI. Women expected clear lines of responsibility for decision-making, to be able to contest decisions, and for AI to perform equally well for all programme participants. Women often imagined both that AI might replace radiographers and that AI implementation might allow more women to be screened: screening programmes will need to communicate carefully about these issues. Conclusions: To meet women's expectations, screening programmes should delay implementation until there is strong evidence that the use of AI systems improves screening performance, should ensure that human expertise and responsibility remain central in screening programmes, and should avoid using AI in ways that exacerbate inequities.

Measures of socioeconomic advantage are not independent predictors of support for healthcare AI: subgroup analysis of a national Australian survey.

Objectives: Applications of artificial intelligence (AI) have the potential to improve aspects of healthcare. However, studies have shown that healthcare AI algorithms also have the potential to perpetuate existing inequities in healthcare, performing less effectively for marginalised populations. Studies on public attitudes towards AI outside of the healthcare field have tended to show higher levels of support for AI among socioeconomically advantaged groups that are less likely to be sufferers of algorithmic harms. We aimed to examine the sociodemographic predictors of support for scenarios related to healthcare AI.Methods: The Australian Values and Attitudes toward AI survey was conducted in March 2020 to assess Australians' attitudes towards AI in healthcare. An innovative weighting methodology involved weighting a non-probability web-based panel against results from a shorter omnibus survey distributed to a representative sample of Australians. We used multinomial logistic regression to examine the relationship between support for AI and a suite of sociodemographic variables in various healthcare scenarios.Results: Where support for AI was predicted by measures of socioeconomic advantage such as education, household income and Socio-Economic Indexes for Areas index, the same variables were not predictors of support for the healthcare AI scenarios presented. Variables associated with support for healthcare AI included being male, having computer science or programming experience and being aged between 18 and 34 years. Other Australian studies suggest that these groups may have a higher level of perceived familiarity with AI.Conclusion: Our findings suggest that while support for AI in general is predicted by indicators of social advantage, these same indicators do not predict support for healthcare AI.

Ethical issues in big data: A qualitative study comparing responses in the health and higher education sectors.

INTRODUCTION: The health and higher education sectors are increasingly using large administrative datasets for secondary purposes. Both sectors experience ethical challenges in the use of big data. This study identifies and explores how these two sectors are responding to these ethical challenges. OBJECTIVES AND APPROACH: Through in-depth qualitative interviews, we asked 18 key Australian stakeholders using or sharing big data in the health and higher education sectors to identify the ethical, social and legal issues associated with big data use and their views on how to build ethical policies in this area. RESULTS: There was strong agreement between participants in the two sectors in a number of areas. All participants believed in the benefits of data usage and recognised the importance of privacy, transparency and consent, and the duties for data custodians which followed from these principles. However, there were also significant differences. The participants in the two sectors took different views on what data are for, what benefits data should provide, who should benefit and how, and the imagined unit of analysis for working with data. Broadly, participants from the higher education sector approached these questions with individual students in mind, while health sector informants approached these questions with collectives, groups, or publics in mind. In deciding what to do, the health participants drew principally on a shared toolkit of legislative, regulatory and ethical instruments, and higher education participants on a culture of duties towards individuals. CONCLUSION / IMPLICATIONS: The health and higher education sectors are responding to ethical challenges in the use of big data in different, but potentially complementary, ways.

Transvaginal mesh in Australia: An analysis of news media reporting from 1996 to 2021.

INTRODUCTION: Transvaginal mesh (mesh) surgeries have been used to treat stress urinary incontinence (incontinence) and/or pelvic organ prolapse (prolapse). In Australia, as in many other countries, the harms caused by mesh eventually prompted individual and collective attempts to achieve redress. The rise of mesh surgery as a procedure, the experience of mesh-affected women and the formal inquiries and legal actions that followed all occurred in social, cultural and discursive contexts. One strategy to understand these contexts is to track how the mesh and key actors in the mesh stories have been portrayed in mass media sources. We conducted a media analysis of the most highly read Australian newspapers and online news media platforms, focusing on how mesh and the interaction of stakeholders in mesh stories were presented to the Australian public. METHOD: We searched systematically in the top 10 most-read print and online media outlets in Australia. We included all articles that mentioned mesh, from the date of first use of mesh in Australia to the date of our final search (1996-2021). RESULT: After early scant media reporting focusing on the benefits of mesh procedures, major Australian medicolegal processes created a hook to shift reporting about mesh. The news media then played a significant role in redressing women's experienced epistemic injustice, including by amplifying previously ignored evidence of harm. This created an opportunity for previously unreported suffering to be revealed to powerful actors, in settings beyond the immediate control and epistemic authority of healthcare stakeholders, validating women's testimony and creating new hermeneutic resources for understanding mesh. Over time, media reports show healthcare stakeholders responding sympathetically to these new understandings in public discourse, contrasting with their statements in earlier media coverage. CONCLUSION: We argue that mass media reporting, in synergy with medicolegal actions and the Australian Senate Inquiry, appears to have provided women with greater epistemic justice, giving their testimony privileged epistemic status such that it was considered by powerful actors. Although medical reporting is not recognised in the hierarchy of evidence embedded in the medical knowledge system, in this case, media reporting appears to have contributed to shaping medical knowledge in significant ways. PATIENT OR PUBLIC CONTRIBUTION: We used publicly available data, print and online media outlets, for our analysis. Therefore, this manuscript does not contain the direct contribution of patients, service users, caregivers, people with lived experience or members of the public.

Practical, epistemic and normative implications of algorithmic bias in healthcare artificial intelligence: a qualitative study of multidisciplinary expert perspectives.

BACKGROUND: There is a growing concern about artificial intelligence (AI) applications in healthcare that can disadvantage already under-represented and marginalised groups (eg, based on gender or race). OBJECTIVES: Our objectives are to canvas the range of strategies stakeholders endorse in attempting to mitigate algorithmic bias, and to consider the ethical question of responsibility for algorithmic bias. METHODOLOGY: The study involves in-depth, semistructured interviews with healthcare workers, screening programme managers, consumer health representatives, regulators, data scientists and developers. RESULTS: Findings reveal considerable divergent views on three key issues. First, views on whether bias is a problem in healthcare AI varied, with most participants agreeing bias is a problem (which we call the bias-critical view), a small number believing the opposite (the bias-denial view), and some arguing that the benefits of AI outweigh any harms or wrongs arising from the bias problem (the bias-apologist view). Second, there was a disagreement on the strategies to mitigate bias, and who is responsible for such strategies. Finally, there were divergent views on whether to include or exclude sociocultural identifiers (eg, race, ethnicity or gender-diverse identities) in the development of AI as a way to mitigate bias. CONCLUSION/SIGNIFICANCE: Based on the views of participants, we set out responses that stakeholders might pursue, including greater interdisciplinary collaboration, tailored stakeholder engagement activities, empirical studies to understand algorithmic bias and strategies to modify dominant approaches in AI development such as the use of participatory methods, and increased diversity and inclusion in research teams and research participant recruitment and selection.