Attitudes towards artificial intelligence in emergency medicine

OBJECTIVE

To assess Australian and New Zealand emergency clinicians' attitudes towards the use of artificial intelligence (AI) in emergency medicine.

METHODS

We undertook a qualitative interview-based study based on grounded theory. Participants were recruited through ED internal mailing lists, the Australasian College for Emergency Medicine Bulletin, and the research teams' personal networks. Interviews were transcribed, coded and themes presented.

RESULTS

Twenty-five interviews were conducted between July 2021 and May 2022. Thematic saturation was achieved after 22 interviews. Most participants were from either Western Australia (52%) or Victoria (16%) and were consultants (96%). More participants reported feeling optimistic (10/25) than neutral (6/25), pessimistic (2/25) or mixed (7/25) towards the use of AI in the ED. A minority expressed scepticism regarding the feasibility or value of implementing AI into the ED. Multiple potential risks and ethical issues were discussed by participants including skill loss from overreliance on AI, algorithmic bias, patient privacy and concerns over liability. Participants also discussed perceived inadequacies in existing information technology systems. Participants felt that AI technologies would be used as decision support tools and not replace the roles of emergency clinicians. Participants were not concerned about the impact of AI on their job security. Most (17/25) participants thought that AI would impact emergency medicine within the next 10 years.

CONCLUSIONS

Emergency clinicians interviewed were generally optimistic about the use of AI in emergency medicine, so long as it is used as a decision support tool and they maintain the ability to override its recommendations.

Sex- and Age-Specific Differences in Risk Profiles and Early Outcomes in Adults With Acute Coronary Syndromes

BACKGROUND

Adults <55 years of age comprise a quarter of all acute coronary syndromes (ACS) hospitalisations. There is a paucity of data characterising this group, particularly sex differences. This study aimed to compare the clinical and risk profile of patients with ACS aged <55 years with older counterparts, and measure short-term outcomes by age and sex.

METHOD

The study population comprised patients with ACS enrolled in the AUS-Global Registry of Acute Coronary Events (GRACE), Cooperative National Registry of Acute Coronary Syndrome Care (CONCORDANCE) and SNAPSHOT ACS registries. We compared clinical features and combinations of major modifiable risk factors (hypertension, smoking, dyslipidaemia, and diabetes) by sex and age group (20-54, 55-74, 75-94 years). All-cause mortality and major adverse events were identified in-hospital and at 6-months.

RESULTS

There were 16,658 patients included (22.3% aged 20-54 years). Among them, 20-54 year olds had the highest proportion of ST-elevation myocardial infarction compared with sex-matched older age groups. Half of 20-54 year olds were current smokers, compared with a quarter of 55-74 year olds, and had the highest prevalence of no major modifiable risk factors (14.2% women, 12.7% men) and of single risk factors (27.6% women, 29.0% men), driven by smoking. Conversely, this age group had the highest proportion of all four modifiable risk factors (6.6% women, 4.7% men). Mortality at 6 months in 20-54 year olds was similar between men (2.3%) and women (1.7%), although lower than in older age groups.

CONCLUSIONS

Younger adults with ACS are more likely to have either no risk factor, a single risk factor, or all four modifiable risk factors, than older patients. Targeted risk factor prevention and management is warranted in this age group.

Development of a list of Australian potentially inappropriate medicines using the Delphi technique

BACKGROUND

Older people are at high risk of medicines-related harms. otentially inappropriate medicines (PIMs) list has been developed to assist clinicians and researchers to identify medicines with risks that may potentially outweigh their benefits in order to improve medication management and safety.

AIM

To develop a list of PIMs for older people specific to Australia.

METHODS

The study obtained expert consensus through the utilisation of the Delphi technique in Australia. A total of 33 experts partook in the initial round, while 32 experts engaged in the subsequent round. The primary outcomes encompass medicines assessed as potentially inappropriate, the specific contexts in which their inappropriateness arises and potentially safer alternatives.

RESULTS

A total of 16 medicines or medicine classes had one or more medicines deemed as potentially inappropriate in older people. Up to 19 medicines or medicine classes had specific conditions that make them more potentially inappropriate, while alternatives were suggested for 16 medicines or classes.

CONCLUSION

An explicit PIMs list for older people living in Australia has been developed containing 19 drugs/drug classes. The PIMs list is intended to be used as a guide for clinicians when assessing medication appropriateness in older people in Australian clinical settings and does not substitute individualised treatment advice from clinicians.

Dispensing of antineoplastic medications and their impact on the dispensing of anti-dementia drugs for adults aged ≥60 years: A cohort study

History of cancer has been associated with decreased risk of dementia, but it is unclear if this is due to the use of antineoplastic medications. Participants were 442,795 adults aged ≥60 years, of whom 235,841 (53.26 %) were women. Those dispensed antineoplastic medications during 2012-2013 had lower odds of being dispensed an anti-dementia drug between 2015 and 2021 (age/sex-adjusted OR = 0.60, 95%CI = 0.55-0.66). The dispensing of antineoplastic medications was associated with an adjusted hazard ratio of 0.72 (95%CI = 0.65-0.80) of subsequent dispensing of an anti-dementia drug. Understanding the mechanisms that support this association may contribute to the introduction of novel approaches to dementia prevention.

Health morbidities associated with the dispensing of lithium to males and females: Cross-sectional analysis of the 10 % Pharmaceutical Benefits Scheme sample for 2022

OBJECTIVES

This study examined the association of gender on the physical morbidity of individuals likely living with bipolar disorder (BD) using a comprehensive health-related database. It investigated the association between lithium dispensing (a surrogate marker for BD) and other health morbidities, considering age and sex.

METHODS

The cross-sectional study design used the 10 % Schedule of Pharmaceutical Benefits Scheme (PBS) database in Australia for 2022. Medication dispensing, age, and sex were available. A validated algorithm inferred 45 health morbidities from dispensed medicines. Statistical analyses, including logistic regression, assessed the relationship between lithium dispensing, sex, and age with inferred health morbidities.

RESULTS

The sample consisted of 1,594,112 individuals aged 10 to over 95 years. A higher proportion of women than men were dispensed lithium (0.33 % vs 0.30 %). Lithium dispensing and age were associated with higher prevalence of inferred morbidities. Women dispensed lithium had a greater physical health burden compared to men, with higher odds of chronic airways diseases, diabetes, ischaemic heart disease/hypertension, inflammation, pain, psychosis, and steroid-responsive diseases. Conversely, women dispensed lithium had lower odds of cardiac arrhythmias and hypothyroidism compared to men.

CONCLUSIONS

This study provides evidence that individuals with BD, indicated by the dispensing of lithium, experience a relatively higher frequency of physical health morbidities, with women being disproportionally affected compared to men. The findings highlight the need for comprehensive care for people living with BD, particularly women.

Applications of natural language processing at emergency department triage: A narrative review

INTRODUCTION

Natural language processing (NLP) uses various computational methods to analyse and understand human language, and has been applied to data acquired at Emergency Department (ED) triage to predict various outcomes. The objective of this scoping review is to evaluate how NLP has been applied to data acquired at ED triage, assess if NLP based models outperform humans or current risk stratification techniques when predicting outcomes, and assess if incorporating free-text improve predictive performance of models when compared to predictive models that use only structured data.

METHODS

All English language peer-reviewed research that applied an NLP technique to free-text obtained at ED triage was eligible for inclusion. We excluded studies focusing solely on disease surveillance, and studies that used information obtained after triage. We searched the electronic databases MEDLINE, Embase, Cochrane Database of Systematic Reviews, Web of Science, and Scopus for medical subject headings and text keywords related to NLP and triage. Databases were last searched on 01/01/2022. Risk of bias in studies was assessed using the Prediction model Risk of Bias Assessment Tool (PROBAST). Due to the high level of heterogeneity between studies and high risk of bias, a metanalysis was not conducted. Instead, a narrative synthesis is provided.

RESULTS

In total, 3730 studies were screened, and 20 studies were included. The population size varied greatly between studies ranging from 1.8 million patients to 598 triage notes. The most common outcomes assessed were prediction of triage score, prediction of admission, and prediction of critical illness. NLP models achieved high accuracy in predicting need for admission, triage score, critical illness, and mapping free-text chief complaints to structured fields. Incorporating both structured data and free-text data improved results when compared to models that used only structured data. However, the majority of studies (80%) were assessed to have a high risk of bias, and only one study reported the deployment of an NLP model into clinical practice.

CONCLUSION

Unstructured free-text triage notes have been used by NLP models to predict clinically relevant outcomes. However, the majority of studies have a high risk of bias, most research is retrospective, and there are few examples of implementation into clinical practice. Future work is needed to prospectively assess if applying NLP to data acquired at ED triage improves ED outcomes when compared to usual clinical practice.

Deep learning model to predict exercise stress test results: Optimizing the diagnostic test selection strategy and reduce wastage in suspected coronary artery disease patients

BACKGROUND

Cardiac exercise stress testing (EST) offers a non-invasive way in the management of patients with suspected coronary artery disease (CAD). However, up to 30% EST results are either inconclusive or non-diagnostic, which results in significant resource wastage. Our aim was to build machine learning (ML) based models, using patients demographic (age, sex) and pre-test clinical information (reason for performing test, medications, blood pressure, heart rate, and resting electrocardiogram), capable of predicting EST results beforehand including those with inconclusive or non-diagnostic results.

METHODS

A total of 30,710 patients (mean age 54.0 years, 69% male) were included in the study with 25% randomly sampled in the test set, and the remaining samples were split into a train and validation set with a ratio of 9:1. We constructed different ML models from pre-test variables and compared their discriminant power using the area under the receiver operating characteristic curve (AUC).

RESULTS

A network of Oblivious Decision Trees provided the best discriminant power (AUC=0.83, sensitivity=69%, specificity=0.78%) for predicting inconclusive EST results. A total of 2010 inconclusive ESTs were correctly identified in the testing set.

CONCLUSIONS

Our ML model, developed using demographic and pre-test clinical information, can accurately predict EST results and could be used to identify patients with inconclusive or non-diagnostic results beforehand. Our system could thus be used as a personalised decision support tool by clinicians for optimizing the diagnostic test selection strategy for CAD patients and to reduce healthcare expenditure by reducing nondiagnostic or inconclusive ESTs.

Inflating 2D convolution weights for efficient generation of 3D medical images

BACKGROUND AND OBJECTIVE

The generation of three-dimensional (3D) medical images has great application potential since it takes into account the 3D anatomical structure. Two problems prevent effective training of a 3D medical generative model: (1) 3D medical images are expensive to acquire and annotate, resulting in an insufficient number of training images, and (2) a large number of parameters are involved in 3D convolution.

METHODS

We propose a novel GAN model called 3D Split&Shuffle-GAN. To address the 3D data scarcity issue, we first pre-train a two-dimensional (2D) GAN model using abundant image slices and inflate the 2D convolution weights to improve the initialization of the 3D GAN. Novel 3D network architectures are proposed for both the generator and discriminator of the GAN model to significantly reduce the number of parameters while maintaining the quality of image generation. Several weight inflation strategies and parameter-efficient 3D architectures are investigated.

RESULTS

Experiments on both heart (Stanford AIMI Coronary Calcium) and brain (Alzheimer's Disease Neuroimaging Initiative) datasets show that our method leads to improved 3D image generation quality (14.7 improvements on Frchet inception distance) with significantly fewer parameters (only 48.5% of the baseline method).

CONCLUSIONS

We built a parameter-efficient 3D medical image generation model. Due to the efficiency and effectiveness, it has the potential to generate high-quality 3D brain and heart images for real use cases.

Lithium Dispensed for Adults Aged ≥ 50 Years Between 2012 and 2021: Analyses of a 10% Sample of the Australian Pharmaceutical Benefits Scheme

BACKGROUND

Lithium use seems to be declining in clinical practice. We examined the proportion of adults aged ≥ 50 years dispensed lithium between 2012 and 2021, and investigated the proportion of lithium users dispensed other medications.

METHODS

We used a 10% random sample data of the Australian Pharmaceutical Benefits Scheme from 2012 to 2021, and limited our analyses to adults aged ≥ 50 years. We retrieved data on lithium, other mood stabilisers, antipsychotics, antidepressants, anxiolytics and hypnotics, and medications for the treatment of other health systems.

RESULTS

We received 7081939 person-years records (53.2% women). The proportion of participants dispensed lithium decreased with age: 0.4% for those aged 50-59 years to < 0.1% for people aged ≥ 90 years. The dispensing of lithium increased over 10 years for those aged 50-69 and decreased in those older than 80 years. Among people dispensed lithium, nearly 1 in 5 were dispensed another mood stabiliser. Antipsychotics and antidepressants were dispensed to about 60% of participants dispensed lithium, with antidepressants dispensed more frequently to women than men. About 20% of people dispensed lithium were dispensed anxiolytics/hypnotics, more frequently for women than men. Medications to treat diseases of the alimentary, cardiovascular, endocrine and nervous systems were commonly dispensed to those dispensed lithium, as were antibiotics.

CONCLUSIONS

While the dispensing of lithium increased among young older adults since 2015 when guidelines for the management of mood disorders were published, our findings suggest that lithium may be under-utilised for the management of bipolar disorder in later life.

Priorities for building Australian workforce capacity to leverage population-based, routinely collected data: views from pharmacoepidemiology

AIM

To explore perspectives of leaders in pharmacoepidemiology on building workforce capacity in the routinely collected data arena to enable researchers to generate evidence to support clinical and policy decision-making.

METHODS

Semi-structured interviews were conducted between May and August 2018 with 13 leaders in pharmacoepidemiology in Australia. Discussion topics included training needs, workforce enablers, barriers and priorities for building capacity. The data was analysed using a content analysis approach.

RESULTS

Leaders identified a range of knowledge and skills that are needed to work with routinely collected data and generate evidence to support clinical and policy decision making. Enablers identified included collaborations and promoting awareness to attract new people to work with this data type. Barriers included difficulty accessing data, lack of critical mass of human capital to build skill levels and funding issues.

CONCLUSIONS

Building workforce capacity involves addressing identified enablers and barriers. Central to building workforce capacity is the harmonisation of Australia's data infrastructure, which can improve the way people work, learn, collaborate, share ideas and expand their professional network.

Performance of multilabel machine learning models and risk stratification schemas for predicting stroke and bleeding risk in patients with non-valvular atrial fibrillation

BACKGROUND

Appropriate anticoagulant therapy for patients with atrial fibrillation (AF) requires assessment of stroke and bleeding risks. However, risk stratification schemas such as CHA2DS2-VASc and HAS-BLED have modest predictive capacity for patients with AF. Multilabel machine learning (ML) techniques may improve predictive performance and support decision-making for anticoagulant therapy. We compared the performance of multilabel ML models with the currently used risk scores for predicting outcomes in AF patients.

METHODS

This was a retrospective cohort study of 9670 patients, mean age 76.9 years, 46% women, who were hospitalized with non-valvular AF, and had 1-year follow-up. The outcomes were ischemic stroke (167), major bleeding (430) admissions, all-cause death (1912) and event-free survival (7387). Discrimination and calibration of ML models were compared with clinical risk scores by area under the curve (AUC). Risk stratification was assessed using net reclassification index (NRI).

RESULTS

Multilabel gradient boosting classifier chain provided the best AUCs for stroke (0.685 95% CI 0.676, 0.694), major bleeding (0.709 95% CI 0.703, 0.716) and death (0.765 95% CI 0.763, 0.768) compared to multi-layer neural networks and classifier chain using support vector machine. It provided modest performance improvement for stroke compared to AUC of CHA2DS2-VASc (0.652, NRI = 3.2%, p-value = 0.1), but significantly improved major bleeding prediction compared to AUC of HAS-BLED (0.522, NRI = 22.8%, p-value < 0.05). It also achieved greater discriminant power for death compared with AUC of CHA2DS2-VASc (0.606, p-value < 0.05). ML models identified additional risk features such as hemoglobin level, renal function.

CONCLUSIONS

Multilabel ML models can outperform clinical risk stratification scores for predicting the risk of major bleeding and death in non-valvular AF patients.

A reinforcement learning-based approach for imputing missing data

Missing data is a major problem in real-world datasets, which hinders the performance of data analytics. Conventional data imputation schemes such as univariate single imputation replace missing values in each column with the same approximated value. These univariate single imputation techniques underestimate the variance of the imputed values. On the other hand, multivariate imputation explores the relationships between different columns of data, to impute the missing values. Reinforcement Learning (RL) is a machine learning paradigm where the agent learns by taking actions and receiving rewards in response, to achieve its goal. In this work, we propose an RL-based approach to impute missing data by learning a policy to impute data through an action-reward-based experience. Our approach imputes missing values in a column by working only on the same column (similar to univariate single imputation) but imputes the missing values in the column with different values thus keeping the variance in the imputed values. We report superior performance of our approach, compared with other imputation techniques, on a number of datasets.

Prevalence and Perinatal Outcomes Following In Utero Exposure to Prehospital Emergency Methoxyflurane: A 17-Year Retrospective Cohort Study

OBJECTIVE

To examine the prevalence of exposure and perinatal outcomes associated with in utero exposure to methoxyflurane.

DESIGN, SETTING AND POPULATION

Whole-population ambulance data in Western Australia (WA) were linked to the statutory perinatal data collection to identify pregnant women transferred by ambulance between 2000 and 2016. The proportion of neonates in WA exposed to methoxyflurane, fentanyl or no analgesia during an ambulance transfer was calculated. Perinatal outcomes of pregnancies exposed to methoxyflurane (n=1579) were compared to those exposed to fentanyl (n=203) or no analgesia (n=10524) using multivariable logistic regression modelling. Perinatal outcomes were considered overall and by trimester of exposure.

MAIN OUTCOME MEASURES

Primary outcomes were the prevalence of in utero methoxyflurane exposure and Apgar score on the day of delivery.

RESULTS

In the study period, 0.4% of all neonates born in WA were exposed to methoxyflurane in utero. Methoxyflurane exposure on the day of delivery (n=657) was not associated with an increased likelihood of a low Apgar score at five minutes compared with no analgesia (n=2667) (OR 1.23, 95% CI 0.91-1.67). Whereas fentanyl exposure (n=22) was associated with an increased likelihood of low Apgar score compared with methoxyflurane (OR 3.67, 95% CI 1.18-11.48).

CONCLUSIONS

Methoxyflurane is commonly used by ambulance services to treat pain in pregnant women in WA. While not recommended for use in pregnancy, pregnancies exposed to methoxyflurane did not have an increased risk of adverse perinatal outcomes in this study.

Ten answers to key questions for fluid management in intensive care

This review focuses on fluid management of critically ill patients. The topic is addressed based on 10 single questions with simplified answers that provide clinicians with the basic information needed at the point of care in treating patients in the Intensive Care Unit. The review has didactic purposes and may serve both as an update on fluid management and as an introduction to the subject for novices in critical care. There is an urgent need to increase awareness regarding the potential risks associated with fluid overload. Clinicians should be mindful not only of the indications for administering fluid loads and of the type of fluids administered, but also of the importance to set safety limits. Lastly, it is important to implement proactive strategies seeking to establish negative fluid balance as soon as the clinical conditions are considered to be stable and the risk of deterioration is low.

Long-term outcomes following rheumatic heart disease diagnosis in Australia

Aims 

Rheumatic heart disease (RHD) is a major contributor to cardiac morbidity and mortality globally. This study aims to estimate the probability and predictors of progressing to non-fatal cardiovascular complications and death in young Australians after their first RHD diagnosis.

Methods and results 

This retrospective cohort study used linked RHD register, hospital, and death data from five Australian states and territories (covering 70% of the whole population and 86% of the Indigenous population). Progression from uncomplicated RHD to all-cause death and non-fatal cardiovascular complications (surgical intervention, heart failure, atrial fibrillation, infective endocarditis, and stroke) was estimated for people aged <35 years with first-ever RHD diagnosis between 2010 and 2018, identified from register and hospital data. The study cohort comprised 1718 initially uncomplicated RHD cases (84.6% Indigenous; 10.9% migrant; 63.2% women; 40.3% aged 5–14 years; 76.4% non-metropolitan). The composite outcome of death/cardiovascular complication was experienced by 23.3% (95% confidence interval: 19.5–26.9) within 8 years. Older age and metropolitan residence were independent positive predictors of the composite outcome; history of acute rheumatic fever was a negative predictor. Population group (Indigenous/migrant/other Australian) and sex were not predictive of outcome after multivariable adjustment.

Conclusion 

This study provides the most definitive and contemporary estimates of progression to major cardiovascular complication or death in young Australians with RHD. Despite access to the publically funded universal Australian healthcare system, one-fifth of initially uncomplicated RHD cases will experience one of the major complications of RHD within 8 years supporting the need for programmes to eradicate RHD.

Maintaining quality of life in patients with chronic obstructive pulmonary disease (COPD) by extending the maintenance phase of community-based pulmonary rehabilitation: protocol for a randomised controlled trial (ComEx3 Study)

Introduction

Pulmonary rehabilitation is a core component of the treatment of people with chronic obstructive pulmonary disease (COPD); however, the benefits gained diminish in the ensuing months. The optimal strategy for maintaining the benefits is unclear with weekly supervised maintenance exercise programmes proposed as one strategy. However, the long-term future of maintenance programs is dependent on quality evidence.

Methods and analysis

The ComEx3 randomised controlled trial will investigate the efficacy of extending a weekly supervised maintenance programme for an additional 6 months following an initial 10-week maintenance programme (intervention) by comparing with a control group who receive the same 10-week maintenance programme followed by 6 months of usual care. 120 participants with COPD will be recruited. Primary objective is to determine health-related quality of life over 12 months. Secondary objectives are to determine functional exercise capacity trajectory and to perform an economic evaluation of the intervention to the health system. Outcomes will be analysed for superiority according to intention-to-treat and per-protocol approaches.

Ethics and dissemination

Approval has been received from the relevant ethics committees. Findings will be disseminated in peer-reviewed journals and conferences, targeting those involved in managing people with COPD as well as those who develop policies and guidelines.

Clinical trial registration

ANZCTR 12618000933257

Ten answers to key questions for fluid management in intensive care

This review focuses on fluid management of critically ill patients. The topic is addressed based on 10 single questions with simplified answers that provide clinicians with the basic information needed at the point of care in treating patients in the Intensive Care Unit. The review has didactic purposes and may serve both as an update on fluid management and as an introduction to the subject for novices in critical care. There is an urgent need to increase awareness regarding the potential risks associated with fluid overload. Clinicians should be mindful not only of the indications for administering fluid loads and of the type of fluids administered, but also of the importance to set safety limits. Lastly, it is important to implement proactive strategies seeking to establish negative fluid balance as soon as the clinical conditions are considered to be stable and the risk of deterioration is low.

Evaluating Machine-Learning Models for Predicting Hospital Transfers in Administrative Data: A Study of Admissions for Myocardial Infarction

IntroductionHospital administrative data is a valuable source to measure myocardial infarction (MI) rates. However, admission counts are susceptible to over-inflation if the patient is transferred multiple times during a single episode of care, and variables denoting transfers may not be reliable. To obtain an accurate number of events, hospital transfers need to be correctly identified. Objectives and ApproachWe assessed multivariable logistic regression and various machine-learning models to predict transfers in hospital administrative data. Using Western Australian linked hospital data, we identified records from 2000-2016 with a principal discharge diagnosis of MI. Our standard method to compare against was a 24-hour look-back to identify a transfer using just admission and separation dates from the current and previous records for the same patient. Multivariable logistic regression and decision trees with various boosting algorithms were used to predict if a single record was a transfer, using variables recorded in the admission (e.g. age, sex, type of hospital, admitted from, emergency/elective admission). The performance of each model was calculated using metrics including area under the curve (AUC). ResultsRecords in the training, validation and testing samples had similar characteristics: mean age=68.9 years, 66% were male and 58% admitted to tertiary hospitals. Gradient Boosting Decision Tree (AUC=0.887, 95%CI: 0.886-0.887) outperformed multivariable logistic regression (AUC=0.875; 95% CI: 0.869-0.881) and random forest models (AUC=0.859; 95% CI: 0.853-0.865). Conclusion / ImplicationsMultivariable logistic regression and machine-learning models are able to identify transfers in a single record from existing variables. They can be used in unlinked hospital administrative data where records belonging to the same patient cannot be identified.

Frailty and Initiation of Cardio-Protective Medicines Following Incident Acute Coronary Syndrome in Older People: A Linked Data Study

IntroductionBeta-blockers, renin-angiotensin system inhibitors and statins are evidence-based pharmacotherapies for preventing recurrent acute coronary syndrome (ACS), however their use in older people may depend on clinical factors such as comorbidity and frailty. Deriving comorbidity information from linked administrative data is well established. The recent development of the Hospital Frailty Risk Score (HFRS) now allows ascertainment of frailty from linked administrative data. Objectives and ApproachWe determined the ability of the HFRS to discriminate between groups of older patients initiated on these three cardio-protective medicines following hospitalisation for an incident ACS. We used a 15-year look back to identify incident ACS cases between 2005 and 2008 from Western Australian hospital data and linked them to national pharmaceutical dispensing records. The study was limited to patients aged ≥65 years who had not received these medicines in the last two years and were discharged alive from their index ACS admission. Separate competing risk regression models assessed the association between HFRS and initiation of each medicine, controlling for comorbidities and other variables available from the linked datasets. ResultsOverall, increasing levels of HFRS were associated with lower subdistribution hazards of initiation of each medicine examined, however these varied by age group and sex. For example, the subdistribution hazard ratios (SHR) for beta-blocker initiation among men in the oldest age group (≥85 years) were 0.65 (95% CI: 0.44-0.95) and 0.51 (95% CI: 0.29-0.90) in the intermediate and high HFRS groups respectively compared to the low frailty group. The corresponding SHR for women were 1.01 (95% CI: 0.76-1.35) and 0.89 (95% CI: 0.63-1.27), respectively. Conclusion / ImplicationsThe HFRS applied to linked administrative data discriminates between groups of older people who received cardio-protective medicines following an incident ACS. There is potential for it to be used in other chronic disease conditions.

Increased risk of 2-year death in patients who discontinued their use of statins

OBJECTIVE

This study examined the association between statin usage (discontinued, reduced or continued) and two-year death following a 21% increase in the Pharmaceutical Benefits Scheme (PBS) consumer co-payment in Western Australia.

METHODS

A retrospective observational study in Western Australia using linked administrative Commonwealth PBS data and State hospital inpatient and death data (n = 207,066) was undertaken. We explored the two-year all-cause and ischemic heart disease(IHD)/stroke-specific-death in individuals who discontinued, reduced or continued statin medication following the January 2005 PBS co-payment increase, overall, by beneficiary status (general population vs. social security recipients) and by a history of admission for ischemic heart disease or stroke. Non-cardiovascular (CVD)-related death was also considered.

RESULTS

In the first six months of 2005, 3.3% discontinued, 12.5% reduced and 84.2% continued statin therapy. We found those who discontinued statins were also likely to discontinue at least two other medicines compared to those who continued therapy. There were 4,607 all-cause deaths. For IHD/stroke-specific death, there were 1,317. For all non-CVD-related death, there were 2,808 deaths during the 2-year follow-up period. Cox regression models, adjusted for demographic and clinical characteristics, showed a 39%-61% increase in the risk of all-cause death for individuals who reduced or discontinued statin medication compared to those who continued their statin medication (Discontinued: Adj HR = 1.61, 95% CI 1.40-1.85; Reduced: Adj HR = 1.39, 95% CI 1.28-1.51). For IHD/stroke-specific death, there was an increased risk of death by 28-76% (Discontinued: Adj sHR = 1.76, 95% CI 1.37-2.27; Reduced: Adj sHR = 1.28, 95% CI 1.10-1.49), and for non-CVD-related death, there was an increased risk of death by 44-57% (Discontinued: Adj sHR = 1.57, 95% CI 1.31-1.88; Reduced: Adj sHR = 1.44, 95% CI 1.30-1.60), for individuals who discontinued or reduced their statin medication compared to those who continued.

CONCLUSIONS

Patients who discontinued their statin therapy had a significantly increased risk of IHD and stroke death. Health professionals should be aware that large co-payment changes may be associated with patients discontinuing or reducing medicines to their health detriment. Factors that lead to such changes in patient medication-taking behaviour need to be considered and addressed at the clinical and policy levels.

Improving patient-provider communication about chronic pain: development and feasibility testing of a shared decision-making tool

Background

Chronic pain has emerged as a disease in itself, affecting a growing number of people. Effective patient-provider communication is central to good pain management because pain can only be understood from the patient’s perspective. We aimed to develop a user-centered tool to improve patient-provider communication about chronic pain and assess its feasibility in real-world settings in preparation for further evaluation and distribution.

Methods

To identify and prioritize patient treatment goals for chronic pain, strategies to improve patient-provider communication about chronic pain, and facilitate implementation of the tool, we conducted nominal group technique meetings and card sorting with patients with chronic pain and experienced providers (n = 12). These findings informed the design of the PainAPP tool. Usability and beta-testing with patients (n = 38) and their providers refined the tool and assessed its feasibility, acceptability, and preliminary impact.

Results

Formative work revealed that patients felt neither respected nor trusted by their providers and focused on transforming providers’ negative attitudes towards them, whereas providers focused on gathering patient information. PainAPP incorporated areas prioritized by patients and providers: assessing patient treatment goals and preferences, functional abilities and pain, and providing patients tailored education and an overall summary that patients can share with providers.

Beta-testing involved 38 patients and their providers. Half of PainAPP users shared their summaries with their providers. Patients rated PainAPP highly in all areas. All users would recommend it to others with chronic pain; nearly all trusted the information and said it helped them think about my treatment goals (94%), understand my chronic pain (82%), make the most of my next doctor’s visit (82%), and not want to use opioids (73%). Beta-testing revealed challenges delivering the tool and summary report to patients and providers in a timely manner and obtaining provider feedback.

Conclusions

PainAPP appears feasible for use, but further adaptation and testing is needed to assess its impact on patients and providers.

Trial registration

This study was approved by the University of New England Independent Review Board for the Protection of Human Subjects in Research (012616–019) and was registered with ClinicalTrials.gov (protocol ID: NCT03425266) prior to enrollment. The trial was prospectively registered and was approved on February 7, 2018.

Systematic review and literature appraisal on methodology of conducting and reporting critical-care echocardiography studies: a report from the European Society of Intensive Care Medicine PRICES expert panel

Background

The echocardiography working group of the European Society of Intensive Care Medicine recognized the need to provide structured guidance for future CCE research methodology and reporting based on a systematic appraisal of the current literature. Here is reported this systematic appraisal.

Methods

We conducted a systematic review, registered on the Prospero database. A total of 43 items of common interest to all echocardiography studies were initially listed by the experts, and other “topic-specific” items were separated into five main categories of interest (left ventricular systolic function, LVSF n = 15, right ventricular function, RVF n = 18, left ventricular diastolic function, LVDF n = 15, fluid management, FM n = 7, and advanced echocardiography techniques, AET n = 17). We evaluated the percentage of items reported per study and the fraction of studies reporting a single item.

Results

From January 2000 till December 2017 a total of 209 articles were included after systematic search and screening, 97 for LVSF, 48 for RVF, 51 for LVDF, 36 for FM and 24 for AET. Shock and ARDS were relatively common among LVSF articles (both around 15%) while ARDS comprised 25% of RVF articles. Transthoracic echocardiography was the main echocardiography mode, in 87% of the articles for AET topic, followed by 81% for FM, 78% for LVDF, 70% for LVSF and 63% for RVF. The percentage of items per study as well as the fraction of study reporting an item was low or very low, except for FM. As an illustration, the left ventricular size was only reported by 56% of studies in the LVSF topic, and half studies assessing RVF reported data on pulmonary artery systolic pressure.

Conclusion

This analysis confirmed sub-optimal reporting of several items listed by an expert panel. The analysis will help the experts in the development of guidelines for CCE study design and reporting.

Predictors of ceasing or reducing statin medication following a large increase in the consumer copayment for medications: a retrospective observational study

OBJECTIVES

Previous Australian research has shown that following the 21% increase in patient copayments for medications on the Pharmaceutical Benefits Scheme (PBS) in 2005, the use of lipid-lowering therapy declined by 5%. This study aimed to determine the demographic and clinical characteristics of individuals who continued, reduced or ceased their use of statin medication in 2005.

STUDY TYPE

Retrospective observational study using routinely collected administrative data.

METHOD

We used pharmaceutical claims, hospital separations and mortality records from 2000 to 2005 for the Western Australian population. The cohort comprised stable users of statin medication in 2004. Based on changes in statin use between 2004 and 2005, we identified individuals who: 1) continued using statins; 2) reduced their use by ≥20%; or 3) ceased therapy for at least the first 6 months in 2005. Multivariate logistic regression models were used to determine whether the demographic and clinical characteristics of the three groups differed.

RESULTS

There were 205 924 statin users identified in Western Australia as of December 2004. After the January 2005 Pharmaceutical Benefits Scheme (PBS) copayment increase, 3.2% of users ceased their regular statin therapy, 12.9% reduced statin use and 83.9% continued statin use. This represented a 2.1% increase in statin users reducing or ceasing therapy compared to 2004. Predictors of cessation and reduction of statin therapy included younger age, greater socio-economic disadvantage, residing in very remote areas, having general beneficiary status, being a new statin user, having no prior history of ischaemic heart disease, having no prior history of a coronary artery revascularisation procedure, taking no other cardiovascular medication or diabetic medication, taking an increased number of medications, and having a lower level of adherence to statin medication in 2004.

CONCLUSION

Compared to 2004, an additional 2.1% of statin users reduced or discontinued medication use in 2005, which may be attributed to an increase in the medication copayment. Individuals with general beneficiary status, and younger and healthier people were at particular risk of cessation or reduction in statin use in 2005.

Correction to: A decade of progress in critical care echocardiography: a narrative review

The original version of this article unfortunately contained a mistake.

Dispensing Patterns of Blood Pressure Lowering Agents in Older Australians From 2006 to 2016

BACKGROUND

Increasing numbers of blood pressure lowering (BPL) agents are being prescribed for both primary and secondary prevention of cardiovascular disease, especially in the older population. The aim of this study is to describe the temporal trends and patterns of BPL dispensing among older Australians (aged ≥65 years).

METHODS

We utilized prescription claims data from the Australian Pharmaceutical Benefits Scheme (PBS) for a 10% random sample of people aged ≥65 years. The PBS, funded by the Federal government, provides subsidies to make medicines more affordable for Australian residents. We restricted our analysis to "long-term concession" individuals, who would use PBS for the majority of their medication needs. BPL agents were identified using the World Health Organization Anatomical Therapeutic Chemical classification codes. The annual prevalences and proportional distributions of BPL dispensing by categories were summarized from 2006 to 2016. Direct standardization was applied to indicate changes of BPL dispensing over time.

RESULTS

Age-standardized dispensing of BPL agents increased by 8% among older Australians from 2006 to 2016 (58%-66%). BPL dispensing in males has exceeded that in females since 2009. Angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers were the dominant BPL agents dispensed, with more than 55% of all BPL users over time. Dispensing of diuretics decreased from 27% to 21%, calcium channel blockers decreased from 30% to 25%, while β-blockers remained stable (29%-31%). The use of fixed-dose combinations increased over time from 23% to 31%.

CONCLUSION

The prevalence of BPL dispensing steadily increased among older Australians from 2006 to 2016. The changes in the patterns of BPL dispensing were largely in line with contemporary changes to clinical guidelines for an aging population.

Practical approach to diastolic dysfunction in light of the new guidelines and clinical applications in the operating room and in the intensive care

There is growing evidence both in the perioperative period and in the field of intensive care (ICU) on the association between left ventricular diastolic dysfunction (LVDD) and worse outcomes in patients. The recent American Society of Echocardiography and European Association of Cardiovascular Imaging joint recommendations have tried to simplify the diagnosis and the grading of LVDD. However, both an often unknown pre-morbid LV diastolic function and the presence of several confounders-i.e., use of vasopressors, positive pressure ventilation, volume loading-make the proposed parameters difficult to interpret, especially in the ICU. Among the proposed parameters for diagnosis and grading of LVDD, the two tissue Doppler imaging-derived variables e' and E/e' seem most reliable. However, these are not devoid of limitations. In the present review, we aim at rationalizing the applicability of the recent recommendations to the perioperative and ICU areas, discussing the clinical meaning and echocardiographic findings of different grades of LVDD, describing the impact of LVDD on patients' outcomes and providing some hints on the management of patients with LVDD.

Needs and priority areas for building capacity for working with linked data in the Australian pharmacoepidemiology workforce

IntroductionLinked data are increasingly used in pharmacoepidemiology studies to enhance value beyond that which can be achieve from stand-alone pharmaceutical data. The complexity of pharmaceutical data can make any linked data analysis challenging and it is imperative that this is matched by the human capacity to perform this work. Objectives and ApproachResearch is needed to understand the state of the current pharmacoepidemiology workforce and to prioritise its capacity building needs. We aim to profile the Australian pharmacoepidemiology workforce to explore views, needs, priority areas and perspectives relevant to capacity building. Participants are the regular pharmacoepidemiology workforce (Group 1) and senior medicines stakeholders (Group 2). Following a literature review and consultation with a group of key informants, we developed survey and interview instruments for each group. We piloted the instruments in February 2018 and study data collection is planned for March 2018. We will use a mixed-methods approach to analyse the data. ResultsWe conducted a review of existing literature and identified workforce views, needs and priorities at four levels: personal, team, organisation and wider community. During the consultative process, the informants highlighted the multidisciplinary nature of the pharmacoepidemiology workforce including many with non-health related backgrounds. They also raised concerns about attracting applicants with suitable skills and experience, job satisfaction, career progression and workforce retention. We developed instruments to (i) further explore these issues, (ii) ascertain their experience with linked health data, (iii) determine their training needs, and, (iv) learn about their future intentions. We will present findings on issues pertinent to the Australian pharmacoepidemiology landscape and suggest priorities for building workforce capacity. Conclusion/ImplicationsThis study will provide empirical evidence to support and prioritise capacity building in the Australian pharmacoepidemiology workforce to improve their ability to work with linked data. The instruments that we developed and findings may be relevant to phamacoepidemiology workforce in other countries and other emerging fields that use linked data.

Identification of myocardial infarction type from electronic hospital data in England and Australia: a comparative data linkage study

OBJECTIVE

To determine the utility of International Classification of Diseases (ICD) codes in investigating trends in ST-segment elevation myocardial infarction (STEMI) and non-ST-segment elevation myocardial infarction (NSTEMI) using person-linked electronic hospitalisation data in England and Western Australia (WA).

METHODS

All hospital admissions with myocardial infarction (MI) as the principal diagnosis were identified from 2000 to 2013 from both jurisdictions. Fourth-digit ICD-10 codes were used to delineate all MI types-STEMI, NSTEMI, unspecified and subsequent MI. The annual frequency of each MI type was calculated as a proportion of all MI admissions. For all MI and each MI type, age-standardised rates were calculated and age-adjusted Poisson regression models used to estimate annual percentage changes in rates.

RESULTS

In 2000, STEMI accounted for 49% of all MI admissions in England and 59% in WA, decreasing to 35% and 25% respectively by 2013. Less than 10% of admissions were recorded as NSTEMI in England throughout the study period, whereas by 2013, 70% of admissions were NSTEMI in WA. Unspecified MI comprised 60% of all MI admissions in England by 2013, compared with <1% in WA. Trends in age-standardised rates differed for all MI (England, -2.7%/year; WA, +1.7%/year), underpinned by differing age-adjusted trends in NSTEMI (England, -6.1%/year; WA, +10.2%/year).

CONCLUSION

Differences between the proportion and trends for MI types in English and WA data were observed. These were consistent with the coding standards in each country. This has important implications for using electronic hospital data for monitoring MI and identifying MI types for outcome studies.