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.

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.

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.

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.

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.

Predicting Therapeutic Response to Unfractionated Heparin Therapy: Machine Learning Approach

Background

Unfractionated heparin (UFH) is an anticoagulant drug that is considered a high-risk medication because an excessive dose can cause bleeding, whereas an insufficient dose can lead to a recurrent embolic event. Therapeutic response to the initiation of intravenous UFH is monitored using activated partial thromboplastin time (aPTT) as a measure of blood clotting time. Clinicians iteratively adjust the dose of UFH toward a target, indication-defined therapeutic aPTT range using nomograms, but this process can be imprecise and can take ≥36 hours to achieve the target range. Thus, a more efficient approach is required.

Objective

In this study, we aimed to develop and validate a machine learning (ML) algorithm to predict aPTT within 12 hours after a specified bolus and maintenance dose of UFH.

Methods

This was a retrospective cohort study of 3019 patient episodes of care from January 2017 to August 2020 using data collected from electronic health records of 5 hospitals in Queensland, Australia. Data from 4 hospitals were used to build and test ensemble models using cross-validation, whereas data from the fifth hospital were used for external validation. We built 2 ML models: a regression model to predict the aPTT value after a UFH bolus dose and a multiclass model to predict the aPTT, classified as subtherapeutic (aPTT <70 seconds), therapeutic (aPTT 70-100 seconds), or supratherapeutic (aPTT >100 seconds). Modeling was performed using Driverless AI (H2O), an automated ML tool, and 17 different experiments were iteratively conducted to optimize model accuracy.

Results

In predicting aPTT, the best performing model was an ensemble with 4x LightGBM models with a root mean square error of 31.35 (SD 1.37). In predicting the aPTT class using a repurposed data set, the best performing ensemble model achieved an accuracy of 0.599 (SD 0.0289) and an area under the receiver operating characteristic curve of 0.735. External validation yielded similar results: root mean square error of 30.52 (SD 1.29) for the aPTT prediction model, and accuracy of 0.568 (SD 0.0315) and area under the receiver operating characteristic curve of 0.724 for the aPTT multiclassification model.

Conclusions

To the best of our knowledge, this is the first ML model applied to intravenous UFH dosing that has been developed and externally validated in a multisite adult general medical and surgical inpatient setting. We present the processes of data collection, preparation, and feature engineering for replication.

A whole-of-community program of advance care planning for end-of-life care

Since 2015 a whole-of-community program to promote advance care planning (ACP) within one Queensland Hospital and Health Service (HHS) catchment has spread statewide, financed by Queensland Health (QH) agencies and led by the Statewide Office of Advance Care Planning (SOACP). The program aims to identify ACP-eligible patients, invite and finalise ACP discussions, and ensure documented care preferences are easily retrievable by clinicians to guide future care if a person loses capacity. The SOACP established a digital infrastructure whereby quality-audited ACP documents are uploaded to a software platform accessible to all QH clinicians, private medical specialists, ambulance paramedics, general practitioners (GPs), and registered nurses, including those in residential aged care facilities (RACFs). The SOACP also hosts a website providing resources for clinicians and patients, delivers educational events and mentorship to GPs and hospital and RACF staff, and employs ACP facilitators working across all QH HHSs. The program has seen yearly increases in the numbers of ACP documents uploaded from around the state, with up to 79% of eligible patients in some hospitals receiving ACP, significant ACP uptake in RACFs, and acceptance by GPs to engage in ACP. Audits reveal high concordance between stated preferences and hospital care received, and ACP patients, compared to matched non-ACP controls, more frequently die out of hospital, have fewer inpatient days during their last 6 months of life, and receive less invasive care, with similar results seen among same-patient cohorts post-ACP. Barriers and enablers to ACP have been identified which will inform program evolution.

Research Activities in General Medicine: A Scoping Survey by the Internal Medicine Society of Australia and New Zealand (IMSANZ)

Background

In developing an effective framework for a collaborative research network (RN) that supports members involved in research, the Internal Medicine Society of Australia and New Zealand (IMSANZ) required a better understanding of the current level of research activity and engagement by general physicians, and factors influencing such engagement.

Aims

To explore the current research landscape amongst general physicians in Australia and Aotearoa New Zealand.

Methods

A questionnaire exploring research participation, scope, research enablers and barriers was disseminated to IMSANZ members over a 3‐month period. Core functions of IMSANZ‐RN, research priorities, potential solutions to perceived barriers and required level of support were also evaluated.

Results

A total of 82 members, mostly senior medical staff (74.4%), responded to the survey (11.8% response rate). More than 70% were involved in impactful research across multiple disciplines, encompassing a wide range of research themes and topics. However, there is limited support and resources available to conduct research, with most projects being self‐instigated and self‐funded. There is overwhelming support to increasing the profile of research in general medicine through the establishment of IMSANZ‐RN, whose principal purposes, as identified by respondents, are to foster collaboration, promote research, provide research education and training, and share information among general physicians. Quality improvement studies (56.1%) and clinical trials (41.5%) were also identified as priority research types.

Conclusions

This study has profiled the constraints faced by general physicians in conducting high‐quality collaborative research and provides insights into what is needed to support greater research engagement, through development of a discipline‐specific clinical RN.

Health literacy assessment in the clinic: benefits, pitfalls and practicalities

Approximately 60% of Australians have low or marginal health literacy, which is associated with poorer outcomes in patients with chronic disease. Patient-centred strategies (such as reduced medical jargon, use of pictograms, multimedia narratives) are effective in improving outcomes for many chronic diseases, with the impact being greatest in individuals with low health literacy. However, clinicians need a reliable and practical tool for assessing health literacy, the results of which help inform the choice of communication techniques best tailored to deliver information to patients. This article reviews the evidence of health literacy as an independent predictor of poor disease outcomes, describes feasible methods for assessing health literacy and presents communication strategies aimed at facilitating shared decision-making among those with low health literacy.

Scoping Review of Studies Evaluating Frailty and Its Association with Medication Harm

Introduction

Frailty is associated with an increased risk of death and morbid events. Frail individuals are known to have multiple comorbidities which are often associated with polypharmacy. Whilst a relationship between polypharmacy and frailty has been demonstrated, it is not clear if there is an independent relationship between frailty and medication harm.

Aims

This scoping review aimed to identify and critically appraise studies evaluating medication harm in patients with frailty.

Methods

PubMed, EMBASE, CINAHL and Cochrane databases were searched from inception until 1 February 2021 using key search terms that are synonymous with frailty (such as frail and frail elderly) and medication harm (such as adverse drug events and adverse drug reactions). To be included, studies must have identified medication harm as a primary or secondary outcome measure, and used a frailty assessment tool to determine frailty, or clearly defined how frailty was assessed. Data were narratively synthesised and presented in tables. The checklist from the Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies from the National Heart, Lung, and Blood Institute was used to assess the quality and risk of bias of studies that met the inclusion criteria.

Results

Of 2685 retrieved abstracts, 24 underwent full-text review and nine studies met the inclusion criteria. Three studies were retrospective cohort studies, and six were prospective observational studies. Six studies comprised two distinct groups of frail and non-frail individuals, and the remaining three studies evaluated medication harm in an entirely frail population. Seven studies used validated frailty tools such as the Clinical Frailty Scale, Fried Frailty Index, and Fried Frailty Phenotype. Two studies measured frailty using self-defined criteria. Overall, frail individuals were at risk of medication harm with rates ranging between 18.7 and 77% across the nine studies. However, whether frailty is an independent predictor of medication harm remains uncertain, as this was only evaluated in one study. The risk of bias assessment identified limitations in methods and reporting with all nine studies.

Conclusion

This scoping review identified nine studies evaluating medication harm in frail patients. However, all were limited by the methodological quality and inadequate reporting of study factors. There are few high-quality studies that described a relationship between medication harm and frailty. More robust studies are required that examine the independent relationship between frailty and medication harm, after adjusting for all possible confounders and in particular polypharmacy.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40266-022-00940-3.

Exploring stakeholder attitudes towards AI in clinical practice

Objectives

Different stakeholders may hold varying attitudes towards artificial intelligence (AI) applications in healthcare, which may constrain their acceptance if AI developers fail to take them into account. We set out to ascertain evidence of the attitudes of clinicians, consumers, managers, researchers, regulators and industry towards AI applications in healthcare.

Methods

We undertook an exploratory analysis of articles whose titles or abstracts contained the terms ‘artificial intelligence’ or ‘AI’ and ‘medical’ or ‘healthcare’ and ‘attitudes’, ‘perceptions’, ‘opinions’, ‘views’, ‘expectations’. Using a snowballing strategy, we searched PubMed and Google Scholar for articles published 1 January 2010 through 31 May 2021. We selected articles relating to non-robotic clinician-facing AI applications used to support healthcare-related tasks or decision-making.

Results

Across 27 studies, attitudes towards AI applications in healthcare, in general, were positive, more so for those with direct experience of AI, but provided certain safeguards were met. AI applications which automated data interpretation and synthesis were regarded more favourably by clinicians and consumers than those that directly influenced clinical decisions or potentially impacted clinician–patient relationships. Privacy breaches and personal liability for AI-related error worried clinicians, while loss of clinician oversight and inability to fully share in decision-making worried consumers. Both clinicians and consumers wanted AI-generated advice to be trustworthy, while industry groups emphasised AI benefits and wanted more data, funding and regulatory certainty.

Discussion

Certain expectations of AI applications were common to many stakeholder groups from which a set of dependencies can be defined.

Conclusion

Stakeholders differ in some but not all of their attitudes towards AI. Those developing and implementing applications should consider policies and processes that bridge attitudinal disconnects between different stakeholders.

Systematic review of machine learning models for personalised dosing of heparin

AIM

To identify and critically appraise studies of prediction models, developed using machine learning (ML) methods, for determining the optimal dosing of unfractionated heparin (UFH).

METHODS

Embase, PubMed, CINAHL, Web of Science, International Pharmaceutical Abstracts and IEEE Xplore databases were searched from inception to 31 January 2020 to identify relevant studies using key search terms synonymous with artificial intelligence or ML, 'prediction', 'dose', 'activated partial thromboplastin time (aPTT)' and 'UFH.' Studies had to have used ML methods for developing models that predicted optimal dose of UFH or target therapeutic aPTT levels in the hospital setting. The CHARMS Checklist was used to assess quality and risk of bias of included studies.

RESULTS

Of 8393 retrieved abstracts, 61 underwent full text review and eight studies met inclusion criteria. Four studies described models for predicting aPTT, three studies described models predicting optimal dose of heparin during dialysis and one study described a model that used surrogate outcomes of clotting and bleeding to predict a therapeutic aPTT. Studies varied widely in reporting of study participants, feature characterisation and selection, handling of missing data, sample size calculations and the intended clinical application of the model. Only one study conducted an external validation and no studies evaluated model impacts in clinical practice.

CONCLUSION

Studies of ML models for UFH dosing are few and none report a model ready for routine clinical use. Existing studies are limited by low methodological quality, inadequate reporting of study factors and absence of external validation and impact analysis.

What is needed to mainstream artificial intelligence in health care?

Artificial intelligence (AI) has become a mainstream technology in many industries, but not yet in health care. Although basic research and commercial investment are burgeoning across various clinical disciplines, AI remains relatively non-existent in most healthcare organisations. This is despite hundreds of AI applications having passed proof-of-concept phase, and scores receiving regulatory approval overseas. AI has considerable potential to optimise multiple care processes, maximise workforce capacity, reduce waste and costs, and improve patient outcomes. The current obstacles to wider AI adoption in health care and the pre-requisites for its successful development, evaluation and implementation need to be defined.

Developing critical thinking skills for delivering optimal care

Healthcare systems across the world are challenged with problems of misdiagnosis, non-beneficial care, unwarranted practice variation and inefficient or unsafe practice. In countering these shortcomings, clinicians must be able to think critically, interpret and assimilate new knowledge, deal with uncertainty and change behaviour in response to compelling new evidence. Three critical thinking skills underpin effective care: clinical reasoning, evidence-informed decision-making and systems thinking. It is important to define these skills explicitly, explain their rationales, describe methods of instruction and provide examples of optimal application. Educational methods for developing and refining these skills must be embedded within all levels of clinician training and continuing professional development.

Pill testing at music festivals: is it evidence-based harm reduction?

Recent pill-related deaths of young people at music festivals in Australia have led to a concerted push for on-site pill testing as a means for preventing such events. However, whether pill testing (also termed 'safety checking') is an effective harm reduction strategy remains uncertain. This narrative review concludes that pill testing currently lacks evidence of efficacy sufficient to justify publicly funded national roll-out of on-site pill-testing programmes. Australian governments, addiction specialists and public health experts should collaborate in conducting properly designed field studies aimed at confirming clear benefits from such programmes in reducing pill-related harm.

Development and dissemination of the national strategic action plan for reducing inappropriate polypharmacy in older Australians

A cohesive, national approach is needed to address inappropriate polypharmacy in older adults and promote deprescribing. We describe the dissemination of the Quality Use of Medicines to Optimise Ageing in Older Australians: Recommendations for a National Strategic Action Plan to Reduce Inappropriate Polypharmacy, and the initiatives taken to date that align with, and assist in operationalising this plan.

Demystifying machine learning: a primer for physicians

Machine learning is a tool for analysing digitised data sets and formulating predictions that can optimise clinical decision-making. It aims to identify complex patterns in large data sets and encode them into models that can then classify new unseen cases or make predictions on new data. Machine learning methods take several forms and individual models can be of many different types. More than 50 models have been approved for use in routine healthcare, and the numbers continue to grow exponentially. The reliability and robustness of any model depends on multiple factors, including the quality and quantity of the data used to develop the models, and the selection of features in the data considered most important to maximising accuracy. In ensuring models are safe, effective and reproducible in routine care, physicians need to have some understanding of how these models are developed and evaluated, and to collaborate with data and computer scientists in their design and validation. This narrative review introduces principles, methods and examples of machine learning in a way that does not require mastery of highly complex statistical and computational concepts.

Going digital: a narrative overview of the effects, quality and utility of mobile apps in chronic disease self-management

Objective Smartphone health applications (apps) are being increasingly used to assist patients in chronic disease self-management. The effects of such apps on patient outcomes are uncertain, as are design features that maximise usability and efficacy, and the best methods for evaluating app quality and utility. Methods In assessing efficacy, PubMed, Cochrane Library and EMBASE were searched for systematic reviews (and single studies if no systematic review was available) published between January 2007 and January 2018 using search terms (and synonyms) of 'smartphone' and 'mobile applications', and terms for each of 11 chronic diseases: asthma, chronic obstructive lung disease (COPD), diabetes, chronic pain, serious mental health disorders, alcohol and substance addiction, heart failure, ischaemic heart disease, cancer, cognitive impairment, chronic kidney disease (CKD). With regard to design features and evaluation methods, additional reviews were sought using search terms 'design', 'quality,' 'usability', 'functionality,' 'adherence', 'evaluation' and related synonyms. Results Of 13 reviews and six single studies assessing efficacy, consistent evidence of benefit was seen only with apps for diabetes, as measured by decreased glycosylated haemoglobin levels (HbA1c). Some, but not all, studies showed benefit in asthma, low back pain, alcohol addiction, heart failure, ischaemic heart disease and cancer. There was no evidence of benefit in COPD, cognitive impairment or CKD. In all studies, benefits were clinically marginal and none related to morbid events or hospitalisation. Twelve design features were identified as enhancing usability. An evaluation framework comprising 32 items was formulated. Conclusion Evidence of clinical benefit of most available apps is very limited. Design features that enhance usability and maximise efficacy were identified. A provisional 'first-pass' evaluation framework is proposed that can help decide which apps should be endorsed by government agencies following more detailed technical assessments and which could then be recommended with confidence by clinicians to their patients. What is known about the topic? Smartphone health apps have attracted considerable interest from patients and health managers as a means of promoting more effective self-management of chronic diseases, which leads to better health outcomes. However, most commercially available apps have never been evaluated for benefits or harms in clinical trials, and there are currently no agreed quality criteria, standards or regulations to ensure health apps are user-friendly, accurate in content, evidence based or efficacious. What does this paper add? This paper presents a comprehensive review of evidence relating to the efficacy, usability and evaluation of apps for 11 common diseases aimed at assisting patients in self-management. Consistent evidence of benefit was only seen for diabetes apps; there was absent or conflicting evidence of benefit for apps for the remaining 10 diseases. Benefits that were detected were of marginal clinical importance, with no reporting of hard clinical end-points, such as mortality or hospitalisations. Only a minority of studies explicitly reported using behaviour change theories to underpin the app intervention. Many apps lacked design features that the literature identified as enhancing usability and potential to confer benefit. Despite a plethora of published evaluation tools, there is no universal framework that covers all relevant clinical and technical attributes. An inclusive list of evaluation criteria is proposed that may overcome this shortcoming. What are the implications for practitioners? The number of smartphone apps will continue to grow, as will the appetite for patients and clinicians to use them in chronic disease self-management. However, the evidence to date of clinical benefit of most apps already available is very limited. Design features that enhance usability and clinical efficacy need to be considered. In making decisions about which apps should be endorsed by government agencies and recommended with confidence by clinicians to their patients, a comprehensive but workable evaluation framework needs to be used by bodies assuming the roles of setting and applying standards.

Sociocognitive approach to behaviour change for reducing low-value care

Social and cognitive factors that predispose to low-value care (LVC), and strategies for countering them, may be underarticulated in campaigns aimed at reducing LVC. A sociocognitive approach, in addition to traditional knowledge translation strategies, may augment understanding and changing clinician behaviour underpinning LVC.

Frequency, trends and institutional variation in 30-day all-cause mortality and unplanned readmissions following hospitalisation for heart failure in Australia and New Zealand

AIMS

National 30-day mortality and readmission rates after heart failure (HF) hospitalisations are a focus of US policy intervention and yet have rarely been assessed in other comparable countries. We examined the frequency, trends and institutional variation in 30-day mortality and unplanned readmission rates after HF hospitalisations in Australia and New Zealand.

METHODS AND RESULTS

We included patients >18 years hospitalised with HF at all public and most private hospitals from 2010-15. The primary outcomes were the frequencies of 30-day mortality and unplanned readmissions, and the institutional risk-standardised mortality rate (RSMR) and readmission rate (RSRR) evaluated using separate cohorts. The mortality cohort included 153 592 patients (mean age 78.9 ± 11.8 years, 51.5% male) with 16 442 (10.7%) deaths within 30 days. The readmission cohort included 148 704 patients (mean age 78.6 ± 11.9 years, 51.7% male) with 33 158 (22.3%) unplanned readmission within 30 days. In 392 hospitals with at least 25 HF hospitalisations, the median RSMR was 10.7% (range 6.1-17.3%) with 59 hospitals significantly different from the national average. Similarly, in 391 hospitals with at least 25 HF hospitalisations, the median RSRR was 22.3% (range 17.7-27.1%) with 24 hospitals significantly different from the average. From 2010-15, the adjusted 30-day mortality [odds ratio (OR) 0.991/month, 95% confidence interval (CI) 0.990-0.992, P < 0.01] and unplanned readmission (OR 0.998/month, 95% CI 0.998-0.999, P < 0.01) rates declined.

CONCLUSION

Within 30 days of a HF hospitalisation, one in 10 patients died and almost a quarter of those surviving experienced an unplanned readmission. The risk of these outcomes varied widely among hospitals suggesting disparities in HF care quality. Nevertheless, a substantial decline in 30-day mortality and a modest decline in readmissions occurred over the study period.

Going digital: a checklist in preparing for hospital-wide electronic medical record implementation and digital transformation

Objective In an era of rapid digitisation of Australian hospitals, practical guidance is needed in how to successfully implement electronic medical records (EMRs) as both a technical innovation and a major transformative change in clinical care. The aim of the present study was to develop a checklist that clearly and comprehensively defines the steps that best prepare hospitals for EMR implementation and digital transformation. Methods The checklist was developed using a formal methodological framework comprised of: literature reviews of relevant issues; an interactive workshop involving a multidisciplinary group of digital leads from Queensland hospitals; a draft document based on literature and workshop proceedings; and a review and feedback from senior clinical leads. Results The final checklist comprised 19 questions, 13 related to EMR implementation and six to digital transformation. Questions related to the former included organisational considerations (leadership, governance, change leaders, implementation plan), technical considerations (vendor choice, information technology and project management teams, system and hardware alignment with clinician workflows, interoperability with legacy systems) and training (user training, post-go-live contingency plans, roll-out sequence, staff support at point of care). Questions related to digital transformation included cultural considerations (clinically focused vision statement and communication strategy, readiness for change surveys), management of digital disruption syndromes and plans for further improvement in patient care (post-go-live optimisation of digital system, quality and benefit evaluation, ongoing digital innovation). Conclusion This evidence-based, field-tested checklist provides guidance to hospitals planning EMR implementation and separates readiness for EMR from readiness for digital transformation. What is known about the topic? Many hospitals throughout Australia have implemented, or are planning to implement, hospital wide electronic medical records (EMRs) with varying degrees of functionality. Few hospitals have implemented a complete end-to-end digital system with the ability to bring about major transformation in clinical care. Although the many challenges in implementing EMRs have been well documented, they have not been incorporated into an evidence-based, field-tested checklist that can practically assist hospitals in preparing for EMR implementation as both a technical innovation and a vehicle for major digital transformation of care. What does this paper add? This paper outlines a 19-question checklist that was developed using a formal methodological framework comprising literature review of relevant issues, proceedings from an interactive workshop involving a multidisciplinary group of digital leads from hospitals throughout Queensland, including three hospitals undertaking EMR implementation and one hospital with complete end-to-end EMR, and review of a draft checklist by senior clinical leads within a statewide digital healthcare improvement network. The checklist distinguishes between issues pertaining to EMR as a technical innovation and EMR as a vehicle for digital transformation of patient care. What are the implications for practitioners? Successful implementation of a hospital-wide EMR requires senior managers, clinical leads, information technology teams and project management teams to fully address key operational and strategic issues. Using an issues checklist may help prevent any one issue being inadvertently overlooked or underemphasised in the planning and implementation stages, and ensure the EMR is fully adopted and optimally used by clinician users in an ongoing digital transformation of care.

Frailty, a multisystem ageing syndrome

The management of frail older people is a key component of aged care. There has been a plethora of tools developed for the diagnosis and screening of frailty. Some of these tools are entering routine clinical practice at a time when the higher healthcare costs involved in caring for older people who are frail have become a potential target for cost-cutting. Yet there is still only limited evidence to support the widespread adoption of frailty tools, and foundational factors impact on their accuracy and validity. Despite the acceptance of frailty as a valid term in research and clinical practice, older people believe the term carries stigma. Such issues indicate that there may be a need to reconsider current approaches to frailty. Recent advances in the science of ageing biology can provide a new framework for reconfiguring how we screen, diagnose, treat and prevent frailty. Frailty can be considered to be a multisystem ageing syndrome of decreased physiological and functional reserve, where the biological changes of ageing are seen in most tissues and organs and are the pathogenic mechanism for frailty. Likewise age-related chronic disease and multimorbidity are syndromes where ageing changes occur in one or multiple systems, respectively. This model focusses diagnostic criteria for frailty onto the biomarkers of ageing and generates new targets for the prevention and treatment of frailty based on interventions that influence ageing biology.

SAFety, Effectiveness of care and Resource use among Australian Hospitals (SAFER Hospitals): a protocol for a population-wide cohort study of outcomes of hospital care

INTRODUCTION

Despite global concerns about the safety and quality of health care, population-wide studies of hospital outcomes are uncommon. The SAFety, Effectiveness of care and Resource use among Australian Hospitals (SAFER Hospitals) study seeks to estimate the incidence of serious adverse events, mortality, unplanned rehospitalisations and direct costs following hospital encounters using nationwide data, and to assess the variation and trends in these outcomes.

METHODS AND ANALYSIS

SAFER Hospitals is a cohort study with retrospective and prospective components. The retrospective component uses data from 2012 to 2018 on all hospitalised patients age ≥18 years included in each State and Territories' Admitted Patient Collections. These routinely collected datasets record every hospital encounter from all public and most private hospitals using a standardised set of variables including patient demographics, primary and secondary diagnoses, procedures and patient status at discharge. The study outcomes are deaths, adverse events, readmissions and emergency care visits. Hospitalisation data will be linked to subsequent hospitalisations and each region's Emergency Department Data Collections and Death Registries to assess readmissions, emergency care encounters and deaths after discharge. Direct hospital costs associated with adverse outcomes will be estimated using data from the National Cost Data Collection. Variation in these outcomes among hospitals will be assessed adjusting for differences in hospitals' case-mix. The prospective component of the study will evaluate the temporal change in outcomes every 4 years from 2019 until 2030.

ETHICS AND DISSEMINATION

Human Research Ethics Committees of the respective Australian states and territories provided ethical approval to conduct this study. A waiver of informed consent was granted for the use of de-identified patient data. Study findings will be disseminated via presentations at conferences and publications in peer-reviewed journals.

Medication use and cognitive impairment among residents of aged care facilities

BACKGROUND

Potentially inappropriate polypharmacy is common in residential aged care facilities (RACF). This is of particular concern among people with cognitive impairment who, compared with cognitively intact residents, are potentially more sensitive to the adverse effects of medications.

AIM

To compare the patterns of medication prescribing of RACF residents based on cognitive status.

METHODS

De-identified data collected during telehealth-mediated geriatric consultations with 720 permanent RACF residents were analysed. Residents were categorised into cognitively intact, mild to moderate impairment and severe impairment groups using the interRAI Cognitive Performance Scale. The number of all regular and when-required medications used in the past 3 days, the level of exposure to anti-cholinergic/sedative medications and potentially inappropriate medications and the use of preventive and symptom control medications were compared across the groups.

RESULTS

The median number of medications was 10 (interquartile range (IQR) 8-14). Cognitively intact residents were receiving significantly more medications (median (IQR) 13 (10-16)) than those with mild to moderate (10 (7-13)) or severe (9 (7-12)) cognitive impairment (P < 0.001). Overall, 82% of residents received at least one anti-cholinergic/sedative medication and 26.9% were exposed to one or more potentially inappropriate medications, although the proportions of those receiving such medications were not significantly different across the groups. Of 7658 medications residents were taking daily, 21.3% and 11.7% were classified as symptom control and preventive medications respectively with no significant difference among the groups in their use.

CONCLUSION

Our findings highlight the need for optimising prescribing in RACF residents, with particular attention to medications with anti-cholinergic effects.

A rare form of dermatomyositis associated with muscle weakness and normal creatine kinase level

We present a case study of a 61-year-old Vietnamese woman who presents with features of dermatomyositis (DM), including Gottron’s papules, heliotrope rash, cutaneous ulcers, generalised weakness and pain, and weight loss with normal levels of creatine kinase (CK). She demonstrated features of interstitial lung disease and subsequently tested positive for anti-melanoma differentiation-associated gene 5 and anti-small ubiquitin-like modifier 1 activating enzyme antibodies, which belong to a DM subtype known as clinically amyopathic dermatomyositis and do not present with raised CK. She received standard treatment for DM, including oral prednisolone, hydroxychloroquine, mycopheonlate and topical betamethasone. The treatment successfully reversed skin changes; however, the patient remained generally weak and unable to carry out her activities of daily living.

Preoperative and postoperative recommendations to surgical wound care interventions: A systematic meta-review of Cochrane reviews

BACKGROUND

The increasing numbers of surgeries involving high risk, multi-morbid patients, coupled with inconsistencies in the practice of perioperative surgical wound care, increases patients' risk of surgical site infection and other wound complications.

OBJECTIVES

To synthesise and evaluate the recommendations for nursing practice and research from published systematic reviews in the Cochrane Library on nurse-led preoperative prophylaxis and postoperative surgical wound care interventions used or initiated by nurses.

DESIGN

Meta-review, guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.

DATA SOURCES

The Cochrane Library database.

REVIEW METHODS

All Cochrane Systematic Reviews were eligible. Two reviewers independently selected the reviews and extracted data. One reviewer appraised the methodological quality of the included reviews using A MeaSurement Tool to Assess Systematic Reviews 2 checklist. A second reviewer independently verified these appraisals. The review protocol was registered with the Prospective Register of Systematic Reviews.

RESULTS

Twenty-two Cochrane reviews met the inclusion criteria. Of these, 11 reviews focused on preoperative interventions to prevent infection, while 12 focused on postoperative interventions (one review assessed both pre-postoperative interventions). Across all reviews, 14 (63.6%) made at least one recommendation to undertake a specific practice, while two reviews (9.1%) made at least one specific recommendation not to undertake a practice. In relation to recommendations for further research, insufficient sample size was the most predominant methodological issue (12/22) identified across reviews.

CONCLUSIONS

The limited number of recommendations for pre-and-postoperative interventions reflects the paucity of high-quality evidence, suggesting a need for rigorous trials to address these evidence gaps in fundamentals of nursing care.