Australian Critical Care最新文献

{"title":"Implementation of reusable linen in the intensive care unit: Impact on pressure injury, staff satisfaction, and environmental sustainability","authors":"Kylie Feely RN, GCCN ,&nbsp;Stacey Matthews RN, MPH ,&nbsp;Edward Quilas RN, GCCN ,&nbsp;Forbes McGain MBBS, PhD ,&nbsp;Eugene Kwek PGCert Data Science ,&nbsp;Rochelle Wynne RN, PhD","doi":"10.1016/j.aucc.2025.101311","DOIUrl":"10.1016/j.aucc.2025.101311","url":null,"abstract":"<div><h3>Background</h3><div>Pressure injuries (PIs) remain a major concern in intensive care units (ICUs), leading to increased morbidity, healthcare costs, and extended hospital stays. While various prevention strategies exist, the impact of reusable linen on PI incidence and environmental sustainability remains underexplored.</div></div><div><h3>Objectives</h3><div>The effect of transitioning from disposable to reusable linen on PI incidence among ICU patients was evaluated while assessing environmental impact and staff acceptance.</div></div><div><h3>Methods</h3><div>A before–after study was conducted in a single major metropolitan ICU comparing patient outcomes before (April 2022–March 2023) and after (April 2023–March 2024) reusable linen implementation. Data from electronic medical records and an internal risk monitoring system (RiskMan®) were analysed to identify factors associated with PI development. A nursing staff satisfaction survey was conducted, and landfill waste reduction was quantified.</div></div><div><h3>Results</h3><div>In 2114 patients, the incidence of PIs was significantly reduced after reusable linen was implemented compared to that observed while using the disposable linen (<em>p</em> =&lt; 0.05). Although a weak significant association was found between linen type and PI occurrence (χ² (1) = 4.23, <em>p</em> = 0.040, 95% confidence interval [CI]: 0.001–0.042), with a small effect size (Cramér's V = 0.0447), once adjusting for other factors, linen type was no longer a significant predictor. Acute Physiology and Chronic Health Evaluation III score (odds ratio [OR] = 1.011, 95% CI: 1.003–1.019), age (OR = 0.976, 95% CI: 0.963–0.989), sex (OR = 0.53, 95% CI: 0.340–0.823), diabetes status (OR = 1.29, 95% CI: 1.109–1.499), and ICU length of stay (OR = 1.23, 95% CI: 1.188–1.276) were significant predictors of PI development. Staff satisfaction with reusable linen was high (n = 22, 87.2%), with benefits including usability and sustainability. The transition to reusable linens eliminated 496 kg of landfill waste annually.</div></div><div><h3>Conclusions</h3><div>Reusable linen was not inferior to disposable linen for the prevention of PI and offered substantial environmental benefits. Strong staff support and reduced waste indicate reusable linen is a viable alternative in ICU settings. Further research is needed to explore long-term impacts.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101311"},"PeriodicalIF":2.7,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144858065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Extracorporeal membrane oxygenation cannula dressing and securement practices: A point prevalence study","authors":"Amanda Corley RN, PhD ,&nbsp;India Pearse RN, MCritCare ,&nbsp;Jayshree D. Lavana MBBS, FCICM ,&nbsp;Abhilasha Ahuja MBBS, FCICM ,&nbsp;Chris M. Anstey MBBS, PhD FANZCA, FCICM ,&nbsp;Emma Haisz RN, BN ,&nbsp;Rachael Parke RN, PhD ,&nbsp;Mandy Tallott RN, BN ,&nbsp;Vincent A. Pellegrino MBBS, FRACP, FCICM ,&nbsp;Hergen Buscher MD, FCICM ,&nbsp;John F. Fraser MB, ChB, PhD, FCICM","doi":"10.1016/j.aucc.2025.101298","DOIUrl":"10.1016/j.aucc.2025.101298","url":null,"abstract":"<div><h3>Background</h3><div>Effective securement of extracorporeal membrane oxygenation (ECMO) cannulae, both at the insertion site and along the length of circuit tubing, may reduce the risk of cannula migration, dislodgement, and infection, all of which can lead to adverse patient outcomes. Despite this, there are no evidence-based clinical practice guidelines to inform cannula dressing and securement practices. However, before recommendations for best practice can be made, current practice must be understood.</div></div><div><h3>Aim/Objective</h3><div>The aim of this study was to describe current ECMO cannulae and circuit tubing dressing and securement practices across Australia and New Zealand.</div></div><div><h3>Methods</h3><div>A prospective, observational point prevalence study was conducted in 11 centres across Australia and New Zealand over a 12-month period. Data were collected for every patient receiving ECMO who met inclusion criteria during 12 prespecified data collection periods, each separated by 3 to 4 weeks.</div></div><div><h3>Results</h3><div>A total of 127 patients (adult, n = 100; paediatric, n = 27) and 256 cannulae (venous = 179, arterial = 77) were included in the analysis. Peripherally inserted cannulae were most commonly dressed with a transparent semipermeable dressing (arterial: n = 50/59 [85%]; venous: n = 127/165 [77%]), while centrally inserted cannulae were less uniformly dressed. Sutures were used to secure cannulae at the insertion site in neonatal and paediatric patients (n = 48/51, 94%) more often than in adults (n = 88/205, 43%). Circuit tubing was most frequently secured using sutureless securement devices (arterial: n = 50/77 (65%); venous: n = 93/179 [52%]). Most centres (82%) had a dressing and securement guideline; however, only 12% of insertion sites (n = 13) and 6% of circuit tubings (n = 6) were dressed and secured according to the guideline.</div></div><div><h3>Conclusions</h3><div>Variation exists in ECMO cannula dressing and securement practices across Australia and New Zealand intensive care units, and adherence to local guidelines is low. Further evidence on optimal cannula dressing and securement techniques is urgently required to inform the development of clinical practice guidelines and improve patient care and outcomes.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101298"},"PeriodicalIF":2.7,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144842470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A roadmap for applying theoretical frameworks to implementation of hospital rapid response systems: A qualitative study using focus group interviews","authors":"Gerrie Prins MD ,&nbsp;Soulaiman Lagzimi MD, MsC ,&nbsp;Erwin Ista PhD ,&nbsp;Zoran Trogrlic PhD ,&nbsp;Mart W. Groot MD ,&nbsp;Monique van Dijk PhD ,&nbsp;Diederik A.M.P. Gommers MD ,&nbsp;Jasper van Bommel PhD, MD","doi":"10.1016/j.aucc.2025.101306","DOIUrl":"10.1016/j.aucc.2025.101306","url":null,"abstract":"<div><h3>Introduction</h3><div>Rapid response systems (RRSs) are generally associated with a reduction in adverse events in hospital settings. However, limited attention has been given to the use of effective implementation strategies for successfully embedding the RSS in clinical practice.</div></div><div><h3>Objective</h3><div>The objective of this study was to identify the barriers and facilitators affecting the reimplementation of the RRS in hospital wards and to develop implementation strategies to address any barriers<strong>.</strong></div></div><div><h3>Methods</h3><div>We conducted a qualitative study using semistructured focus group interviews with nurses and physicians from nine general wards in a university hospital. Identified barriers and facilitators were categorised using the Tailored Implementation for Chronic Diseases checklist. Based on this categorisation, the Effective Practice and Organisation of Care taxonomy was used to develop ward-specific toolkits.</div></div><div><h3>Results</h3><div>A total of 112 determinants influencing the reimplementation of RRS—both facilitators and barriers—were identified and categorised across all seven domains of the Tailored Implementation for Chronic Diseases checklist. The most frequently cited barriers related to the beliefs and behaviours of individual healthcare professionals. Across all wards, half of the proposed implementation strategies fell into the categories of educational materials and meetings or audit and feedback. The remaining strategies spanned a wider range of categories, tailored to the specific needs of each ward.</div></div><div><h3>Conclusions</h3><div>This study highlights that effective implementation of an RRS requires consideration of both the behaviour and beliefs of individual healthcare providers, as well as broader organisational issues such as incentives and resources. Successful hospital-wide implementation of an RRS will require a multifaceted approach, combining tailored and educational strategies with components from guideline development and audit and feedback. Further research is needed to determine whether such an approach can lead to successful, sustainable implementation and improved clinical outcomes.</div></div><div><h3>Registration</h3><div>Not applicable.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101306"},"PeriodicalIF":2.7,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intermittent pneumatic compression can reduce the incidence of upper extremity venous thrombosis after peripherally inserted central catheter placement in traumatic brain injury patients: A randomised controlled trial","authors":"Hongmei Liang RD, MSN ,&nbsp;Xiaoqin Wang RD, MSN ,&nbsp;Xingmei Zhou RD, MSN ,&nbsp;Ji Wang RD, MSN ,&nbsp;Chuanfeng Pei RD, MSN ,&nbsp;Long Liu MD, PhD","doi":"10.1016/j.aucc.2025.101308","DOIUrl":"10.1016/j.aucc.2025.101308","url":null,"abstract":"<div><h3>Background</h3><div>Peripherally inserted central catheters (PICCs) are commonly used in patients with traumatic brain injury (TBI) in neurosurgical intensive care units. A frequent complication of this procedure is upper-extremity venous thrombosis (UEVT), which can lead to adverse outcomes.</div></div><div><h3>Objectives</h3><div>The objective of this study was to evaluate the effectiveness of intermittent pneumatic compression (IPC) in reducing the incidence of UEVT in patients with TBI undergoing PICC placement.</div></div><div><h3>Methods</h3><div>Patients with TBI admitted to our neurosurgical intensive care unit between 2021 and 2023 were included in the study. All patients underwent PICC placement and were randomly assigned to a control or intervention group. The intervention group received IPC on the upper limb. Doppler ultrasound was used to detect venous thrombosis and measure blood flow in the upper extremity. Differences in blood flow velocities between the groups at specified time points were analysed using the Mann–Whitney U and Wilcoxon signed-rank tests.</div></div><div><h3>Results</h3><div>The intervention group showed significantly lower rates of UEVT (4.1% vs. 18.6%, <em>P</em> = 0.001), deep vein thrombosis (1.0% vs. 8.2%, <em>P</em> = 0.018), and superficial vein thrombosis (3.1% vs. 10.3%, <em>P</em> = 0.042) than the control group. Venous flow velocities on days 14 and 28 after catheterisation were significantly higher in the intervention group (all <em>P</em> &lt; 0.001), while no significant differences were observed on days 0 and 7.</div></div><div><h3>Conclusion</h3><div>IPC can reduce the incidence of UEVT and improve blood flow in the catheterised upper extremity in patients with TBI after PICC placement.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101308"},"PeriodicalIF":2.7,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144828410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"icuRESOLVE-D (Intensive Care Unit REcovery Solutions cO-Led through surVivor Engagement-Digital): Protocol for a multicentre randomised controlled trial of digital peer support for adult survivors of critical illness","authors":"Kimberley J. Haines PhD, B.Health Science Physiotherapy ,&nbsp;Nina Leggett DPT, BBiomed ,&nbsp;Elizabeth Hibbert B. Physiotherapy ,&nbsp;Yasmine Ali Abdelhamid MBBS, PhD, FCICM, FRACP ,&nbsp;Samantha Bates RN, Grad Dip Crit Care ,&nbsp;Sue Berney B.Physiotherapy, PhD ,&nbsp;Erin Bicknell BPhysio (Hons), M.Rehab Neuro Physio ,&nbsp;Sarah Booth B. Social Work ,&nbsp;Jacki Carmody BBSc, MPsych Hlth, MPsych Clinical, MAPS ,&nbsp;Christopher Cox MD, MHA, MPH ,&nbsp;Tegan Cruwys PhD, MClin Psy, PhB Sci (Hons) MAPS FCCLP ,&nbsp;Kate Emery BExSci, DPT ,&nbsp;K.J. Farley MBBS, FCICM ,&nbsp;Lauren Ferrante MD, MHS ,&nbsp;Craig French MBBS, FCICM, PhD ,&nbsp;Michael O. Harhay PhD, MPH ,&nbsp;Anne Holland BAppSc Physiotherapy, PhD ,&nbsp;Emily Karahalios PhD, BSci (Hons), MPH ,&nbsp;George Kiossoglou ,&nbsp;Marlena Klaic PhD, BOT ,&nbsp;Adam M. Deane MBBS, PhD, FCICM, FRACP","doi":"10.1016/j.aucc.2025.101303","DOIUrl":"10.1016/j.aucc.2025.101303","url":null,"abstract":"<div><h3>Background</h3><div>Peer support is a candidate intervention to improve health-related quality of life in survivors of critical illness. This trial will evaluate the effect of a codesigned digital peer support intervention (expert-facilitated, web-based peer support) compared to usual care.</div></div><div><h3>Design</h3><div>This study is planned as a randomised, multicentre, two-arm, parallel-group (1:1) hybrid effectiveness implementation trial of a digital peer support model versus usual care. This trial will include an embedded process evaluation and health economic analysis.</div></div><div><h3>Participants</h3><div>A total of 212 adult intensive care unit (ICU) survivors, recruited by telephone from 28 days after hospital discharge.</div></div><div><h3>Study setting</h3><div>Six health services in Victoria and the Australian Capital Territory, Australia—including five metropolitan and one regional hospital.</div></div><div><h3>Main outcome measures</h3><div>The primary outcome is the self-reported health status at 90-days post hospital discharge using the EuroQol Visual Analogue Scale. Secondary outcomes will be measured at baseline 28 days after hospital discharge and at 90 days and 180 days after hospital discharge. Secondary outcomes include EuroQol five-dimensional five-level (EQ-5D-5L), Assessment of Quality of Life 6D, Impact of Events Scale-Revised, University of California Los Angeles three-item Loneliness Scale, Patient Health Questionnaire for Depression and Anxiety (PHQ-4), Connor–Davidson Resilience Scale, Patient Activation Measure, and healthcare costs.</div></div><div><h3>Intervention</h3><div>The codesigned digital peer support intervention is a rolling program consisting of four 1-h peer support sessions, occurring fortnightly, delivered via Zoom and facilitated by an expert clinical social worker or psychologist. The sessions include a formal (standardised education focused on ICU recovery) and an informal component (peer-to-peer discussion and peer support).</div></div><div><h3>Discussion</h3><div>This is a trial of a novel, codesigned, digital intervention aimed at improving recovery following critical illness. Through this randomised, parallel-group, two-arm, hybrid type 1 effectiveness implementation trial, we will examine the effectiveness and cost-effectiveness of a digital health peer support model and identify implementation factors necessary for scaling (if effective). Due to the digital health design, this trial has the potential to enhance equity of access to post-ICU services.</div></div><div><h3>Trial registration</h3><div>This study was registered to the Australian New Zealand Clinical Trials Registry (ACTRN12624000267550). This trial will be conducted in compliance with all stipulation of this protocol, the conditions of the ethics committee approval, the National Health and Medical Research Council, National Statement on ethical Conduct in Human Research (2007 and updates), and the Integrated Adde","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101303"},"PeriodicalIF":2.7,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144842349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Medical emergency team stand-down decision-making: Characteristics, documented decisions, and outcomes documented between single and repeat medical emergency team patients—A retrospective review","authors":"Natalie A. Kondos RN, BNurs, BBiomedSci(Hons) ,&nbsp;Jonathan Barrett MBBs, MPH, FRACP, FCICM ,&nbsp;Jo McDonall RN, PhD ,&nbsp;Tracey Bucknall RN, FAAN, GAICD, PhD","doi":"10.1016/j.aucc.2025.101310","DOIUrl":"10.1016/j.aucc.2025.101310","url":null,"abstract":"<div><h3>Introduction</h3><div>Decisions to end a medical emergency team (MET) call have been infrequently studied. Premature ending of MET calls may compromise patient outcomes. The aim of the study was to describe clinicians' documentation practices upon ending MET calls and to compare patients with single and repeat MET call activation on the initial call.</div></div><div><h3>Methods</h3><div>A retrospective review was conducted at a metropolitan hospital in Melbourne, Victoria, from Oct 1st, 2018, to September 30th, 2019. From a total of 8648 initial MET calls, 500 were included in the sample, 250 single and 250 repeated (≥2) MET calls. Data from patients’ index MET call were analysed using univariate analyses and descriptive statistics. Variables included documentation of the MET call stand-down decision and associated decision-making elements, demographic and admission characteristics, and patient outcomes. We compared documentation of MET call stand-down decision-making with expert consensus on essential MET call stand-down decision-making elements.</div></div><div><h3>Results</h3><div>Key differences in the documentation of the essential MET call stand-down decision-making elements were that repeat MET patients had a higher proportion of care outcomes (post-MET call) documented (72%) than single MET patients (48.8%). Treatment decisions were documented over 75% of the time and an escalation plan was documented less than 50% of the time for both MET call patient groups. Repeat MET call patients were twice as likely to die in hospital (15.2% versus 7.6%, <em>p</em> = 0.01), had double the hospital length of stay (21 versus 10 days, <em>p</em> = 0.031), and were three times more likely to be discharged to rehabilitation services rather than home (28% versus 9.6%, <em>p</em> = 0.001).</div></div><div><h3>Conclusion</h3><div>There were differences at the index MET call in documentation and outcomes between patients who required a single MET call and those who required repeat MET calls. Prospective observational research is recommended to better understand the MET call stand-down decision-making process at the patient bedside, environmental influences, and the impact on further patient deterioration.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101310"},"PeriodicalIF":2.7,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144842348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting delirium in intensive care unit patients every 8 hours with machine learning: Model development and evaluation","authors":"Kei Imai RN, MSN ,&nbsp;Takeshi Unoki RN, PhD ,&nbsp;Naoto Takahashi PhD ,&nbsp;Megumi Horikawa RN","doi":"10.1016/j.aucc.2025.101305","DOIUrl":"10.1016/j.aucc.2025.101305","url":null,"abstract":"<div><h3>Objectives</h3><div>Delirium in the intensive care unit (ICU) is associated with poor short- and long-term outcomes, and identifying patients at risk of delirium during ICU stay remains difficult. This study aims to develop a machine learning–based prediction model for identifying delirium occurrence every 8 hours during ICU stay.</div></div><div><h3>Methods</h3><div>We retrospectively collected data from the electronic medical records in a single-centre mixed ICU. Adult patients who were admitted to the ICU between January 2023 and December 2023 and spent more than 24 h in the ICU were eligible for this study. The outcome was delirium defined as an Intensive Care Delirium Screening Checklist score of 4 or more. Four machine learning algorithms (XGBoost, LightGBM, CatBoost, and random forest) were used to develop prediction models using a holdout method.</div></div><div><h3>Results</h3><div>273 patients were included in the study, and 170 patients (62.3%) experienced delirium. The dataset consists of 2321 delirium assessments of which 822 (35.4%) were delirium positive. CatBoost demonstrated the best performance; area under the curve, mean precision, and brier score on the test dataset were 0.886 (95% confidence interval [CI]: 0.857–0.909), 0.804 (95% CI: 0.749–0.852), and 0.131 (95% CI: 0.115–0.149), respectively. The model achieved an accuracy of 0.816 and specificity of 0.900. Precision was 0.775 while maintaining recall at 0.668. Routinely collected nursing observational variables, including Intensive Care Delirium Screening Checklist subscores and the Glasgow Coma Scale score, played a significant role in predicting delirium.</div></div><div><h3>Conclusions</h3><div>Our machine learning–based prediction model demonstrated potential in identifying patients at risk of delirium. However, further research with a larger sample size and greater heterogeneity in the patient population would be needed to guide nursing interventions. Our prediction model would enable nursing professionals to identify patients at high risk of delirium using only routinely collected variables. Nurses would be able to implement timely preventive care and optimise staffing and the patients’ therapeutic environment to reduce risk factors of delirium.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101305"},"PeriodicalIF":2.7,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144828411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Network meta-analysis reinforces concerns over oral antiseptic safety","authors":"Elena Conoscenti BSNurs (Hons.), MSc,&nbsp;Stijn Blot BNurs, BMid, MNurs, PhD","doi":"10.1016/j.aucc.2025.101304","DOIUrl":"10.1016/j.aucc.2025.101304","url":null,"abstract":"","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101304"},"PeriodicalIF":2.7,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring explainable machine learning techniques to aid dysphagia risk identification: A feasibility study","authors":"Melanie L. McIntyre BHSc SpPath, GradCertClinEd ,&nbsp;Yuxi Liu BEng, PhD ,&nbsp;Joanne Murray PhD, BAppSc(Speech Pathology), CPSP ,&nbsp;Shaowen Qin BEng, MEng, MS(Applied Mathematics), PhD ,&nbsp;Timothy Chimunda MBChB, FCICM, AMC, MCC, MACEM ,&nbsp;Sebastian H. Doeltgen MSLT, PhD","doi":"10.1016/j.aucc.2025.101307","DOIUrl":"10.1016/j.aucc.2025.101307","url":null,"abstract":"<div><h3>Background</h3><div>Machine learning offers opportunities to identify complex risk patterns in large data sets. We explored the methodological feasibility, and proof of concept, of applying machine learning techniques to dysphagia (swallowing difficulty) risk identification for adult patients who required endotracheal intubation within an intensive care unit (ICU).</div></div><div><h3>Aim</h3><div>The aim of this study was to explore the methodological feasibility and proof of concept of developing machine learning models for dysphagia risk identification for adult patients who required endotracheal intubation within an ICU.</div></div><div><h3>Methods</h3><div>In this cohort study, two large healthcare databases were linked using deterministic logic. All participants received invasive mechanical ventilation in an ICU. Several machine learning model candidates were explored. Insights into the model decision-making have been provided using SHapley Additive exPlanation values.</div></div><div><h3>Results</h3><div>A total of 59 811 patients from 42 sites were included in the study. The top five most influential factors in determining the presence or absence of dysphagia at a cohort level were duration of mechanical ventilation, age, cardiac admission, neurological admission, and Acute Physiology and Chronic Health Evaluation III score.</div></div><div><h3>Conclusion</h3><div>There is a promising prospect of machine learning in dynamic dysphagia risk screening, which we propose should be considered for clinical use in the future. The patient-specific influence of each risk factor in determining the presence or absence of dysphagia highlights the importance of determining risk based on the individual patient's unique combination of risk factors, and not on cohort means, as has been done previously.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101307"},"PeriodicalIF":2.7,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Findings of a pilot randomised controlled trial of an early psychiatric assessment, referral, and intervention study for intensive care patients","authors":"Dylan Flaws PhD, FRANZCP ,&nbsp;Stuart Baker MBBS, FCICM ,&nbsp;Adrian Barnett PhD AStat FASSA ,&nbsp;Kylie Jacobs RN, MNr (Critical Care) ,&nbsp;Olivia Metcalf BA/BSc (Hons), PhD ,&nbsp;Sue Patterson ,&nbsp;Hamish Pollock BM, FCICM ,&nbsp;Emma Proctor BSN, BBehSci (Psychol) ,&nbsp;Mahesh Ramanan PhD, FCICM ,&nbsp;Alexis Tabah MD, FCICM ,&nbsp;Tracey Varker BSc(Hons), PGradDip(Statistics), PhD","doi":"10.1016/j.aucc.2025.101302","DOIUrl":"10.1016/j.aucc.2025.101302","url":null,"abstract":"<div><h3>Introduction</h3><div>Intensive care unit (ICU) survivors can experience physical, cognitive, or psychological impairments that adversely affect long-term quality of life. The aim of this open-label, single-site, two-arm parallel, pilot randomised controlled trial was to assess the feasibility and acceptability of incorporating a psychiatric consultation into an existing post-ICU clinic.</div></div><div><h3>Methods</h3><div>Post-ICU clinic attendees were invited to participate. Consenting participants were allocated to either treatment as usual plus psychiatric review or treatment as usual using block randomisation. When attending the clinic, participants completed a standard battery of outcome measures followed by an unstructured clinical interview. The intervention arm also had a psychiatric consultation focusing on diagnostic clarification and psychoeducation within 2 weeks. All participants completed an acceptability measure after their appointment. Approximately 6 months after their appointment, participants repeated outcome measures and additional questionnaires. Data were analysed descriptively. A cost analysis assessed the time taken and cost for a psychiatrist to deliver such an intervention.</div></div><div><h3>Results</h3><div>Of 53 clinic attendees, 25 (47%) were invited to participate; 11 (44%) individuals consented, and 10 (91%) completed follow-up. Participants in both arms rated the interventions provided as acceptable. On a 10-point acceptability scale, the mean result for the control was 7.9 compared to 9.0 for the intervention. At 6-month follow-up, mean Hospital Anxiety and Depression Scale Anxiety subscales increased from 2.0 to 8.0 for controls but reduced from 8.8 to 4.9 for the intervention arm. Depression subscales increased from 2.7 to 6.1 for controls but reduced from 9.8 to 6.3 for the intervention arm. Mean Post-Traumatic Stress Disorder Checklist for Diagnostic and Statistical Manual (DSM)-5 scores reduced from 54.0 to 23.3 for controls and 24.3 to 6.1 for the intervention arm. A psychiatrist providing two half-day clinics weekly could see up to 368 patients, costing AU$63,100 per year to the local service.</div></div><div><h3>Conclusion</h3><div>Incorporating a psychiatric assessment with standard postintensive care clinics appears both acceptable and feasible. A randomised controlled trial should further evaluate the clinical efficacy of such an intervention.</div></div>","PeriodicalId":51239,"journal":{"name":"Australian Critical Care","volume":"38 6","pages":"Article 101302"},"PeriodicalIF":2.7,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}