Evaluating automated machine learning platforms for use in healthcare.

Objective: To describe development and application of a checklist of criteria for selecting an automated machine learning (Auto ML) platform for use in creating clinical ML models. Materials and Methods: Evaluation criteria for selecting an Auto ML platform suited to ML needs of a local health district were developed in 3 steps: (1) identification of key requirements, (2) a market scan, and (3) an assessment process with desired outcomes. Results: The final checklist comprising 21 functional and 6 non-functional criteria was applied to vendor submissions in selecting a platform for creating a ML heparin dosing model as a use case. Discussion: A team of clinicians, data scientists, and key stakeholders developed a checklist which can be adapted to ML needs of healthcare organizations, the use case providing a relevant example. Conclusion: An evaluative checklist was developed for selecting Auto ML platforms which requires validation in larger multi-site studies.

The adverse inpatient medication event and frailty (AIME-frail) risk prediction model.

BACKGROUND: Medication harm affects between 5 and 15% of hospitalised patients, with approximately half of the harm events considered preventable through timely intervention. The Adverse Inpatient Medication Event (AIME) risk prediction model was previously developed to guide a systematic approach to patient prioritisation for targeted clinician review, but frailty was not tested as a candidate predictor variable. AIM: To evaluate the predictive performance of an updated AIME model, incorporating a measure of frailty, when applied to a new multisite cohort of hospitalised adult inpatients. METHODS: A retrospective cohort study was conducted at two tertiary Australian hospitals on patients discharged between 1st January and April 31, 2020. Data were extracted from electronic medical records (EMRs) and clinical coding databases. Medication harm was identified using ICD-10 Y-codes and confirmed by senior pharmacist review of medical records. The Hospital Frailty Risk Score (HFRS) was calculated for each patient. Logistic regression analysis was used to construct a modified AIME model. Candidate variables of the original AIME model, together with new variables including HFRS were tested. Performance of the final model was reported using area under the curve (AUC) and decision curve analysis (DCA). RESULTS: A total of 4089 patient admissions were included, with a mean age ± standard deviation (SD) of 64 years (±19 years), 2050 patients (50%) were males, and mean HFRS was 6.2 (±5.9). 184 patients (4.5%) experienced one or more medication harm events during hospitalisation. The new AIME-Frail risk model incorporated 5 of the original variables: length of stay (LOS), anti-psychotics, antiarrhythmics, immunosuppressants, and INR greater than 3, as well as 5 new variables: HFRS, anticoagulants, antibiotics, insulin, and opioid use. The AUC was 0.79 (95% CI: 0.76-0.83) which was superior to the original model (AUC = 0.70, 95% CI: 0.65-0.74) with a sensitivity of 69%, specificity of 81%, positive predictive value of 0.14 (95% CI: 0.10-0.17) and negative predictive value of 0.98 (95% CI: 0.97-0.99). The DCA identified the model as having potential clinical utility between the probability thresholds of 0.05-0.4. CONCLUSION: The inclusion of a frailty measure improved the predictive performance of the AIME model. Screening inpatients using the AIME-Frail tool could identify more patients at high-risk of medication harm who warrant timely clinician review.

The new paradigm in machine learning - foundation models, large language models and beyond: a primer for physicians.

Foundation machine learning models are deep learning models capable of performing many different tasks using different data modalities such as text, audio, images and video. They represent a major shift from traditional task-specific machine learning prediction models. Large language models (LLM), brought to wide public prominence in the form of ChatGPT, are text-based foundational models that have the potential to transform medicine by enabling automation of a range of tasks, including writing discharge summaries, answering patients questions and assisting in clinical decision-making. However, such models are not without risk and can potentially cause harm if their development, evaluation and use are devoid of proper scrutiny. This narrative review describes the different types of LLM, their emerging applications and potential limitations and bias and likely future translation into clinical practice.

Reducing potentially inappropriate polypharmacy at a national and international level: the impact of deprescribing networks.

INTRODUCTION: Over the past decade, polypharmacy has increased dramatically. Measurable harms include falls, fractures, cognitive impairment, and death. The associated costs are massive and contribute substantially to low-value health care. Deprescribing is a promising solution, but there are barriers. Establishing a network to address polypharmacy can help overcome barriers by connecting individuals with an interest and expertise in deprescribing and can act as an important source of motivation and resources. AREAS COVERED: Over the past decade, several deprescribing networks were launched to help tackle polypharmacy, with evidence of individual and collective impact. A network approach has several advantages; it can spark interest, ideas and enthusiasm through information sharing, meetings and conversations with the public, providers, and other key stakeholders. In this special report, the details of how four deprescribing networks were established across the globe are detailed. EXPERT OPINION: Networks create links between people who lead existing and/or budding deprescribing practices and policy initiatives, can influence people with a shared passion for deprescribing, and facilitate sharing of intellectual capital and tools to take initiatives further and strengthen impact.This report should inspire others to establish their own deprescribing networks, a critical step in accelerating a global deprescribing movement.

Achieving large-scale clinician adoption of AI-enabled decision support.

Computerised decision support (CDS) tools enabled by artificial intelligence (AI) seek to enhance accuracy and efficiency of clinician decision-making at the point of care. Statistical models developed using machine learning (ML) underpin most current tools. However, despite thousands of models and hundreds of regulator-approved tools internationally, large-scale uptake into routine clinical practice has proved elusive. While underdeveloped system readiness and investment in AI/ML within Australia and perhaps other countries are impediments, clinician ambivalence towards adopting these tools at scale could be a major inhibitor. We propose a set of principles and several strategic enablers for obtaining broad clinician acceptance of AI/ML-enabled CDS tools.

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.

Monoclonal antibodies for treating early Alzheimer disease-a commentary on recent 'positive' trials.

Recent phase 3 randomised controlled trials of amyloid-targeting monoclonal antibodies in people with pre-clinical or early Alzheimer disease have reported positive results, raising hope of finally having disease-modifying drugs. Given their far-reaching implications for clinical practice, the methods and findings of these trials, and the disease causation theory underpinning the mechanism of drug action, need to be critically appraised. Key considerations are the representativeness of trial populations; balance of prognostic factors at baseline; psychometric properties and minimal clinically important differences of the primary efficacy outcome measures; level of study fidelity; consistency of subgroup analyses; replication of findings in similar trials; sponsor role and potential conflicts of interest; consistency of results with disease causation theory; cost and resource estimates; and alternative prevention and treatment strategies. In this commentary, we show shortcomings in each of these areas and conclude that monoclonal antibody treatment for early Alzheimer disease is lacking high-quality evidence of clinically meaningful impacts at an affordable cost.

Systematic review and longitudinal analysis of implementing Artificial Intelligence to predict clinical deterioration in adult hospitals: what is known and what remains uncertain.

OBJECTIVE: To identify factors influencing implementation of machine learning algorithms (MLAs) that predict clinical deterioration in hospitalized adult patients and relate these to a validated implementation framework. MATERIALS AND METHODS: A systematic review of studies of implemented or trialed real-time clinical deterioration prediction MLAs was undertaken, which identified: how MLA implementation was measured; impact of MLAs on clinical processes and patient outcomes; and barriers, enablers and uncertainties within the implementation process. Review findings were then mapped to the SALIENT end-to-end implementation framework to identify the implementation stages at which these factors applied. RESULTS: Thirty-seven articles relating to 14 groups of MLAs were identified, each trialing or implementing a bespoke algorithm. One hundred and seven distinct implementation evaluation metrics were identified. Four groups reported decreased hospital mortality, 1 significantly. We identified 24 barriers, 40 enablers, and 14 uncertainties and mapped these to the 5 stages of the SALIENT implementation framework. DISCUSSION: Algorithm performance across implementation stages decreased between in silico and trial stages. Silent plus pilot trial inclusion was associated with decreased mortality, as was the use of logistic regression algorithms that used less than 39 variables. Mitigation of alert fatigue via alert suppression and threshold configuration was commonly employed across groups. CONCLUSIONS: : There is evidence that real-world implementation of clinical deterioration prediction MLAs may improve clinical outcomes. Various factors identified as influencing success or failure of implementation can be mapped to different stages of implementation, thereby providing useful and practical guidance for implementers.

Navigating the machine learning pipeline: a scoping review of inpatient delirium prediction models.

OBJECTIVES: Early identification of inpatients at risk of developing delirium and implementing preventive measures could avoid up to 40% of delirium cases. Machine learning (ML)-based prediction models may enable risk stratification and targeted intervention, but establishing their current evolutionary status requires a scoping review of recent literature. METHODS: We searched ten databases up to June 2022 for studies of ML-based delirium prediction models. Eligible criteria comprised: use of at least one ML prediction method in an adult hospital inpatient population; published in English; reporting at least one performance measure (area under receiver-operator curve (AUROC), sensitivity, specificity, positive or negative predictive value). Included models were categorised by their stage of maturation and assessed for performance, utility and user acceptance in clinical practice. RESULTS: Among 921 screened studies, 39 met eligibility criteria. In-silico performance was consistently high (median AUROC: 0.85); however, only six articles (15.4%) reported external validation, revealing degraded performance (median AUROC: 0.75). Three studies (7.7%) of models deployed within clinical workflows reported high accuracy (median AUROC: 0.92) and high user acceptance. DISCUSSION: ML models have potential to identify inpatients at risk of developing delirium before symptom onset. However, few models were externally validated and even fewer underwent prospective evaluation in clinical settings. CONCLUSION: This review confirms a rapidly growing body of research into using ML for predicting delirium risk in hospital settings. Our findings offer insights for both developers and clinicians into strengths and limitations of current ML delirium prediction applications aiming to support but not usurp clinician decision-making.

Digital health competencies for the next generation of physicians.

As health care continues to change and evolve in a digital society, there is an escalating need for physicians who are skilled and enabled to deliver care using digital health technologies, while remaining able to successfully broker the triadic relationship among patients, computers and themselves. The focus needs to remain firmly on how technology can be leveraged and used to support good medical practice and quality health care, particularly around resolution of longstanding challenges in health care delivery, including equitable access in rural and remote areas, closing the gap on health outcomes and experiences for First Nations peoples and better support in aged care and those living with chronic disease and disability. We propose a set of requisite digital health competencies and recommend that the acquisition and evaluation of these competencies become embedded in physician training curricula and continuing professional development programmes.

Deployment of machine learning algorithms to predict sepsis: systematic review and application of the SALIENT clinical AI implementation framework.

OBJECTIVE: To retrieve and appraise studies of deployed artificial intelligence (AI)-based sepsis prediction algorithms using systematic methods, identify implementation barriers, enablers, and key decisions and then map these to a novel end-to-end clinical AI implementation framework. MATERIALS AND METHODS: Systematically review studies of clinically applied AI-based sepsis prediction algorithms in regard to methodological quality, deployment and evaluation methods, and outcomes. Identify contextual factors that influence implementation and map these factors to the SALIENT implementation framework. RESULTS: The review identified 30 articles of algorithms applied in adult hospital settings, with 5 studies reporting significantly decreased mortality post-implementation. Eight groups of algorithms were identified, each sharing a common algorithm. We identified 14 barriers, 26 enablers, and 22 decision points which were able to be mapped to the 5 stages of the SALIENT implementation framework. DISCUSSION: Empirical studies of deployed sepsis prediction algorithms demonstrate their potential for improving care and reducing mortality but reveal persisting gaps in existing implementation guidance. In the examined publications, key decision points reflecting real-word implementation experience could be mapped to the SALIENT framework and, as these decision points appear to be AI-task agnostic, this framework may also be applicable to non-sepsis algorithms. The mapping clarified where and when barriers, enablers, and key decisions arise within the end-to-end AI implementation process. CONCLUSIONS: A systematic review of real-world implementation studies of sepsis prediction algorithms was used to validate an end-to-end staged implementation framework that has the ability to account for key factors that warrant attention in ensuring successful deployment, and which extends on previous AI implementation frameworks.

Implementation frameworks for end-to-end clinical AI: derivation of the SALIENT framework.

OBJECTIVE: To derive a comprehensive implementation framework for clinical AI models within hospitals informed by existing AI frameworks and integrated with reporting standards for clinical AI research. MATERIALS AND METHODS: (1) Derive a provisional implementation framework based on the taxonomy of Stead et al and integrated with current reporting standards for AI research: TRIPOD, DECIDE-AI, CONSORT-AI. (2) Undertake a scoping review of published clinical AI implementation frameworks and identify key themes and stages. (3) Perform a gap analysis and refine the framework by incorporating missing items. RESULTS: The provisional AI implementation framework, called SALIENT, was mapped to 5 stages common to both the taxonomy and the reporting standards. A scoping review retrieved 20 studies and 247 themes, stages, and subelements were identified. A gap analysis identified 5 new cross-stage themes and 16 new tasks. The final framework comprised 5 stages, 7 elements, and 4 components, including the AI system, data pipeline, human-computer interface, and clinical workflow. DISCUSSION: This pragmatic framework resolves gaps in existing stage- and theme-based clinical AI implementation guidance by comprehensively addressing the what (components), when (stages), and how (tasks) of AI implementation, as well as the who (organization) and why (policy domains). By integrating research reporting standards into SALIENT, the framework is grounded in rigorous evaluation methodologies. The framework requires validation as being applicable to real-world studies of deployed AI models. CONCLUSIONS: A novel end-to-end framework has been developed for implementing AI within hospital clinical practice that builds on previous AI implementation frameworks and research reporting standards.

Coping with uncertainty in clinical practice: a narrative review.

Clinicians must make decisions amid the uncertainty that is ubiquitous to clinical practice. Uncertainty in clinical practice can assume many forms depending on its source, such as insufficient personal knowledge or scientific evidence, limited practical understanding or competence, challenging interpersonal relationships, and complexity and ambiguity in clinical encounters. The level and experience of uncertainty varies according to personal traits, clinical context, affective factors and sociocultural norms. Clinicians vary in their tolerance of uncertainty, and maladaptive responses may adversely affect patient care and clinician wellbeing. Various strategies can be used to minimise and manage, but not eliminate, uncertainty and to share uncertainty with patients without compromising the clinician-patient relationship or clinician credibility.

Recognition and Management of Hospital-Acquired Sepsis Among Older General Medical Inpatients: A Multi-Site Retrospective Study.

Purpose: To assess accuracy of early diagnosis, appropriateness and timeliness of response, and clinical outcomes of older general medical inpatients with hospital-acquired sepsis. Methods: Hospital abstracts of inpatient encounters from seven digital Queensland public hospitals between July 2018 and September 2020 were screened retrospectively for diagnoses of hospital-acquired sepsis. Electronic medical records were retrieved and cases meeting selection criteria and classified as confirmed or probable sepsis using pre-specified criteria were included. Investigations and treatments following the first digitally generated alert of clinical deterioration were compared with a best practice sepsis care bundle. Outcome measures comprised 30-day all-cause mortality after deterioration, and unplanned readmissions at 14 days after discharge. Results: Of the 169 screened care episodes, 59 comprised probable or confirmed cases of sepsis treated by general medicine teams at the time of initial deterioration. Of these, 43 (72.9%) had no mention of sepsis in the differential diagnosis on first medical review, and only 38 (64%) were managed as having sepsis. Each care bundle component of blood cultures, serum lactate, and intravenous fluid resuscitation and antibiotics was only delivered in approximately 30% of cases, and antibiotic administration was delayed more than an hour in 28 of 38 (73.7%) cases. Conclusion: Early recognition of sepsis and timely implementation of care bundles are challenging in older general medical patients. Education programs in sepsis care standards targeting nurses and junior medical staff, closer patient monitoring, and post-discharge follow-up may improve patient outcomes.

An organisational approach to improving diagnostic safety.

Diagnostic error affects up to 10% of clinical encounters and is a major contributing factor to 1 in 100 hospital deaths. Most errors involve cognitive failures from clinicians but organisational shortcomings also act as predisposing factors. There has been considerable focus on profiling causes for incorrect reasoning intrinsic to individual clinicians and identifying strategies that may help to prevent such errors. Much less focus has been given to what healthcare organisations can do to improve diagnostic safety. A framework modelled on the US Safer Diagnosis approach and adapted for the Australian context is proposed, which includes practical strategies actionable within individual clinical departments. Organisations adopting this framework could become centres of diagnostic excellence. This framework could act as a starting point for formulating standards of diagnostic performance that may be considered as part of accreditation programs for hospitals and other healthcare organisations.

Quality of observational studies of clinical interventions: a meta-epidemiological review.

BACKGROUND: This meta-epidemiological study aimed to assess methodological quality of a sample of contemporary non-randomised clinical studies of clinical interventions. METHODS: This was a cross-sectional study of observational studies published between January 1, 2012 and December 31, 2018. Studies were identified in PubMed using search terms 'association', 'observational,' 'non-randomised' 'comparative effectiveness' within titles or abstracts. Each study was appraised against 35 quality criteria by two authors independently, with each criterion rated fully, partially or not satisfied. These quality criteria were grouped into 6 categories: justification for observational design (n = 2); minimisation of bias in study design and data collection (n = 11); use of appropriate methods to create comparable groups (n = 6); appropriate adjustment of observed effects (n = 5); validation of observed effects (n = 9); and authors interpretations (n = 2). RESULTS: Of 50 unique studies, 49 (98%) were published in two US general medical journals. No study fully satisfied all applicable criteria; the mean (+/-SD) proportion of applicable criteria fully satisfied across all studies was 72% (+/- 10%). The categories of quality criteria demonstrating the lowest proportions of fully satisfied criteria were measures used to adjust observed effects (criteria 20, 23, 24) and validate observed effects (criteria 25, 27, 33). Criteria associated with ≤50% of full satisfaction across studies, where applicable, comprised: imputation methods to account for missing data (50%); justification for not performing an RCT (42%); interaction analyses in identifying independent prognostic factors potentially influencing intervention effects (42%); use of statistical correction to minimise type 1 error in multiple outcome analyses (33%); clinically significant effect sizes (30%); residual bias analyses for unmeasured or unknown confounders (14%); and falsification tests for residual confounding (8%). The proportions of fully satisfied criteria did not change over time. CONCLUSIONS: Recently published observational studies fail to fully satisfy more than one in four quality criteria. Criteria that were not or only partially satisfied were identified which serve as remediable targets for researchers and journal editors.

A whole-of-health system approach to improving care of frail older persons.

The population is aging, with frailty emerging as a significant risk factor for poor outcomes for older people who become acutely ill. We describe the development and implementation of the Frail Older Persons' Collaborative Program, which aims to optimise the care of frail older adults across healthcare systems in Queensland. Priority areas were identified at a co-design workshop involving key stakeholders, including consumers, multidisciplinary clinicians, senior Queensland Health staff and representatives from community providers and residential aged care facilities. Locally developed, evidence-based interventions were selected by workshop participants for each priority area: a Residential Aged Care Facility acute care Support Service (RaSS); improved early identification and management of frail older persons presenting to hospital emergency departments (GEDI); optimisation of inpatient care (Eat Walk Engage); and enhancement of advance care planning. These interventions have been implemented across metropolitan and regional areas, and their impact is currently being evaluated through process measures and system-level outcomes. In this narrative paper, we conceptualise the healthcare organisation as a complex adaptive system to explain some of the difficulties in achieving change within a diverse and dynamic healthcare environment. The Frail Older Persons' Collaborative Program demonstrates that translating research into practice and effecting change can occur rapidly and at scale if clinician commitment, high-level leadership, and adequate resources are forthcoming.