The checking of X-ray reports to ensure limb fractures are not missed and patients are suitably followed-up upon Emergency Department (ED) discharge is an essential but labour intensive task. Phase I of our research is the development of computer software to reliably identify limb fractures in radiology reports. Phase II of the project is to link fractures identified in the reports with patients' disposition recorded in hospital electronic databases in order to automate the stated checking process.
To (i) assess the inter-rater reliability of clinicians in identifying limb fractures in radiology reports; and (ii) describe the early stages in the development of software to model how clinicians read and interpret the reports.
Electronic radiology reports were sampled from an adult, a paediatric and a mixed hospital ED. Software was developed to assists clinicians in the recording of their interpretation and the highlighting of text in the report which lead to his/her interpretation. Reports were interpreted as normal or abnormal (fracture, dislocation and others). Agreement was measured and reasons for disagreement sought.
Ninety-three reports were reviewed by two clinicians. Forty-seven and 30 reports were interpreted as normal and abnormal respectively by both clinicians. Sixteen reports were interpreted as normal by one and as abnormal by the other. Kappa 0.67 (95% CI 0.51–0.82). All 16 reports with disagreement were of patients with known fractures representing for scheduled or unscheduled review.
Clinicians' processes for classifying a limb radiology report need to be defined prior to their modelling by computer software.
Factors associated with adverse events during sedation of highly agitated patients in the emergency department
Jonathan Knott PhD FACEM,1,2 David Taylor MD MPH FACEM,2,3 Esther Chan BPharm PhD,4 David Kong PhD,5 Georgina Phillips MBBS FACEM,6 David Castle MD FRCP FANZCP6
1Emergency Department, Royal Melbourne Hospital, Melbourne, Australia; 2The University of Melbourne, Melbourne, Australia; 3Emergency Department, Austin Hospital, Melbourne, Australia; 4Department of Pharmacology and Pharmacy, Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong; 5Centre for Medicine Use and Safety, Monash University, Melbourne, Australia; 6Emergency Department, St Vincent's Hospital, Melbourne, Australia
Patients requiring sedation for acute agitation are a highly vulnerable population. The need for rapid sedation using benzodiazepines or neuroleptics may result in significant adverse events with a rate reported between 13% and 31%.
To identify factors associated with adverse events (AE) during sedation for acute agitation.
This study reviewed 336 patients sedated with midazolam alone or in combination with either droperidol or olanzapine for AE occurences. AEs were identified prospectively during management of acute agitation. All case histories were also reviewed and data extracted for any unreported pre-defined events. Variables including patient demographics, clinical urgency, sedation, co-morbidities and past history were tested for significant association. Factors significant to a p-value of 0.1 were entered into a regression model to isolate those with independent prediction of adverse events.
There were 39 (11.6%) patients who had 50 AEs. The factors increasing risk were: being male (Odds Ratio 1.88, P value 0.08), history of alcohol abuse (1.97, 0.06), or alcohol intoxication on arrival (3.45, 0.002). The protective factors were: regular anti-psychotic use (0.31, 0.007), history of drug abuse (0.49, 0.03), drug intoxication on arrival (0.54, 0.08) and self-presentation (0.6, 0.06). Following logistic regression only regular anti-psychotic use remained independently significant (0.37, 0.03).
Whilst AEs during acute sedation remain highly problematic for ED staff and patients, there are no factors likely to be readily available to help predict high risk populations. All patients with acute agitation requiring chemical restraint should be monitored closely until the risk of serious AEs has dissipated.
Evaluating a pre-Fellowship exam course: the Auckland experience
Kim M Yates MBChB MMedSc (Hons) PGCertClinEd FACEM,1 Louise A Finnel MBChB PGCertClinEd FACEM2
1Emergency Departments, North Shore & Waitakere Hospitals, Waitemata District Health Board, Auckland, New Zealand; 2Emergency Care, Middlemore Hospital, Auckland, New Zealand
There has been an annual Pre-fellowship Exam Course in Auckland for more than 10 years and it remains popular and fully booked with trainees. Feedback about the course is generally very positive, but we wanted to further evaluate the course, to determine whether attending the course improved exam performance. The Visual Aid Question (VAQ) component was chosen for this pilot study.
Course attendees were invited to participate and participants were allocated random codes, then randomised to complete one of two VAQs in exam conditions before the course started, writing their code on the exam booklet. The second VAQ was completed at the end of the course. Each paper was marked by 2 college examiners, and an agreed mark given for each paper. Papers were then decoded to allow before and after comparisons. Participants also filled out questionnaires before and after the course regarding their confidence of having a good strategy to pass components of the exam.
12 participants completed both VAQs. Improvements in overall VAQ grades before and after the course were minimal (median 4.5 vs 5, mean 5.3 vs 5.4) however standard deviation of grades appeared lower after the course (1.8 vs 1.0). Mean and median scores for confidence about strategies to pass all components of the exam were higher after the course than before.
The small sample size in this pilot study limits interpretation, although after the course, study participants appeared more confident about strategies to pass components of the fellowship exam.
Does the addition of the lateral film improve the diagnostic accuracy of the chest radiograph in adult patients presenting to the emergency department with suspected pneumonia?
Catherine Twomey MBChB (trainee),1 James Kwan MBBS FACEM,2 Amit Shetty MBBS FACEM,3 James Metri MBBS FRANZCR4
1Emergency Department, Westmead Hospital, Sydney, Australia; 2Discipline of Emergency Medicine, Sydney Medical School, University of Sydney; Sydney, Australia; 3Emergency Department Research Unit, Westmead Hospital, Sydney, Australia; 4Radiology Department, Westmead Hospital, Sydney, Australia
Pneumonia is a common, potentially serious and life threatening condition. Chest radiography is the investigation most commonly used to diagnose pneumonia. The lateral film delivers approximately 13 times the radiation to the breast as the PA view. The null hypothesis is that the lateral chest radiograph does not improve the diagnosis of pneumonia within the emergency department and therefore patients are potentially being subjected to unnecessary radiation.
We conducted a prospective cohort study involving 450 radiographs performed for suspected pneumonia. 15 Emergency Physicians and 15 Emergency Advanced Trainees read the PA radiographs alone and then with lateral views whilst the researcher recorded their interpretation. On both occasions, each doctor ‘reported’ 30 sets of radiographs. 15 Radiologists then performed the same exercise blinded to the Emergency Physicians results.
The sensitivity was 86% (95% CI 82–89%) with the PA alone and 89% (95% CI 84–91%) with the addition of the lateral film (p > 0.05). The specificity was 73% (95% CI 70–76%) with PA alone and 71% (95% CI 67–74%) with the addition of the lateral (P > 0.05). The data shows that when the addition of the lateral radiograph does lead to a change in diagnosis, there is no statistically significant difference in which direction the change will be.
This study demonstrates that the addition of the lateral view to a standard PA chest radiograph does not improve the diagnosis of pneumonia in the emergency department. Emergency departments should consider developing protocols to rationalize the use of the lateral chest film.
Experiments with teamwork training: is a teamwork rating scale helpful?
Kim M Yates MBChB MMedSc(Hons) PGCertClinEd FACEM,1 Willem Landman MBBS PGCertClinEd FACEM,1 Jennifer Weller MD MClinEd MBBS FANZCA FRCA,2 Boaz Shulruf PhD MPH BSc DipTch3
1Emergency Departments, North Shore & Waitakere Hospitals, Waitemata District Health Board, Auckland, New Zealand; 2Centre for Medical and Health Sciences Education, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; 3Medicine Education & Student Office, Faculty of Medicine, University of New South Wales, Sydney, Australia
Patient safety literature emphasises the importance of teamwork skills and human factors in preventing medical errors, and formal ED teamwork training improves team behaviours, decreases medical errors and improves attitudes toward teamwork. The ED continuing medical education programme at Waitemata District Health Board in Auckland includes monthly simulation centre sessions, where feedback on teamwork is given. In this study, we aimed to evaluate the usefulness and reliability of a teamwork rating scale when used in simulations.
The Waitemata ED Teamwork Rating Scale was adapted from an existing scale. Attendees at ED simulation days were invited to participate using codes to preserve confidentiality. Before the usual debrief occurred for each simulation, study participants were asked to fill out the Waitemata Teamwork Rating Scale. At the end of each simulation day, participants were asked to fill out a questionnaire gauging the usefulness of the scale for teaching teamwork. Analysis will include descriptive statistics with confidence intervals. Reliability of the instrument will be assessed by multivariate factor analysis.
Each ED simulation day had 2 attendee groups and 4 scenarios, so 16 scenarios in total, with 40 participants studied. The rating scale was felt to be useful in identifying teamwork successes and opportunities by 86% of respondents, and felt to be useful in the subsequent debriefing by 97%. Results of the factor analysis will be presented.
Using a rating scale in simulations as part of teamwork training was felt to be useful by most participants. Reliability issues will be discussed.
Effect of continuous positive airway pressure (CPAP) on mortality in the treatment of acute cardiogenic pulmonary oedema (ACPO) in the pre-hospital setting: randomised controlled trial
Michael A Austin,1,2 Karen E Wills,1 David Kilpatrick,3 Michael Gibson,4 Eugene H Walters1,3
1Menzies Research Institute of Tasmania, Tasmania, Australia; 2Department of Emergency Medicine, University of Ottawa, Ottawa, Ontario, Canada; 3University of Tasmania, Tasmania, Australia; 4Ambulance Tasmania, Tasmania, Australia
The pre-hospital use of continuous positive airway pressure (CPAP) ventilation is a relatively new management for acute cardiogenic pulmonary oedema (ACPO) and there is little high quality evidence on the benefits or potential dangers in this setting. The aim of this study was to determine whether patients in severe respiratory distress treated with CPAP in the pre-hospital setting have a lower mortality than those treated with usual care.
Randomised, controlled trial comparing usual care versus CPAP (Whisperflow®) in a pre-hospital setting, for adults experiencing severe respiratory distress, with falling respiratory efforts, due to a presumed ACPO. Patients were randomised to receive either usual care, including conventional medications (Nitrates and Furosemide) plus bag-valve-mask ventilation (bagging) and high flow oxygen, versus conventional medications plus this CPAP regimen. The primary outcome was pre-hospital or in-hospital mortality. Secondary outcomes were the need for tracheal intubation, length of hospital stay, change in vital signs and blood gas results.
Fifty patients were enrolled with mean age 79.8 (SD 11.9), male 56.0%, mortality 20.0%. The risk of death was significantly reduced in the CPAP arm with mortality 34.6% (9 deaths) in the usual care arm compared to 4.2% (1 death) in the CPAP arm (RR, 0.12; 95% CI 0.02 to 0.88; p = 0.04). Patients who received CPAP were significantly less likely to have respiratory acidosis (mean difference in pH −0.10; 95% CI 0.04 to 0.16; p = 0.001; n = 44, and elevated pCO2 (mean difference 10.0 mmHg; 95% CI 1.76 to 18.2; p = 0.014; n = 43) than patients receiving usual care. The length of hospital stay was significantly less in the patients who received CPAP (mean difference 4.3 days; 95% CI 0.26 to 0.63, p = 0.001).
We found that CPAP significantly reduced mortality, respiratory acidosis and length of hospital stay for patients in severe respiratory distress caused by ACPO who required ventilatory assistance compared with conventional care. Results from this study also support the caution in the use of hyperoxia for this patient population. This study shows the use of CPAP in the pre-hospital setting for ACPO is practicable, and improves patient outcomes.
NHMRC Centre of Research Excellence (CRE) for Chronic Respiratory Disease, and Fisher and Paykal (suppliers of the Whisperflow® CPAP device).
Are workplace-based assessments (WBAs) the weapons of mass destruction (WMDs) of assessment?
Victor Lee MB BS FACEM GCHPE
Emergency Department, Austin Health, Melbourne, Australia
Workplace-based assessments (WBAs) have been used to assess how doctors perform in the clinical workplace for almost fifteen years. They represent a new assessment modality that is being implemented in postgraduate specialty training programs in Australasia and overseas. Likewise, the ACEM Curriculum Revision Project (CRP) is undertaking to implement WBAs in Advanced Training. If you are wondering what WBAs are, why they are being utilized or whether they could possibly be coming to your own ED, then this session is for you.
To describe the different WBA tools, their rationale and the current evidence for their role in the ACEM advanced training program.
An overview of the different WBA tools and the rationale for their use will be presented. The utility of WBAs as an assessment method will be discussed, including a comprehensive review of their strengths and weaknesses. The current educational evidence base for their role in programmatic assessment and the importance of asking the right questions, in the right way, about the right things, of the right people about workplace performance will be described. Also, the formative feedback benefits of WBAs as an assessment for learning will be outlined. Finally, an update on the progress of implementation of WBAs in the ACEM advanced training program will be provided. Then you can make up your own mind about whether WBAs are WMDs of assessment or not.
Implementation and impact of clinical network managed evidence-based-practice implementation projects in Victorian emergency departments
Anne-Maree Kelly,1,2 Jan Pannifex,1 Elif Cetiner1 on behalf of the Emergency Care Improvement and Innovation Clinical Network
1Emergency Care Improvement and Innovation Clinical Network (ECIICN), Commission for Hospital Improvement, Department of Health (Victoria), Melbourne, Australia; 2Joseph Epstein Centre for Emergency Medicine Research, Western Health, Melbourne, Australia
To describe the implementation and impact of evidence-based-practice implementation projects managed by a clinical network across a range of health services in Victoria.
During 2009–2011, ECIICN (a group within the Department of Health Victoria) led three rounds of evidence-based-practice implementation projects in EDs across Victoria. Through consultation and research, a methodology based on knowledge transfer theory was developed and implemented. Quantitative analysis of impact using pre and post data was performed. We also obtained qualitative feedback about success factors and barriers.
17 hospitals participated in the first project (chest pain assessment), 19 in the second (TIA assessment and management) and 23 in the third (various projects). 29 out of 40 network member ED participated (72%), approximately equally distributed across metropolitan and rural areas. Projects were completed by 90% of ED. Projects reported statistically and clinically significant improvements in patient care including reduction of average length of stay for patients discharged after ACS rule out process (effect size 32 minutes, 95% CI 460 min, P = 0.025), increased proportion of patients with TIA discharged on anti-platelet drugs (90% vs 80%, P < 0.001), increased proportion of patients with fractured neck of femur receiving nerve block as adjuvant analgesia (64% vs 6%, P < 0.001) and increased proportion of patients with COAD treated with controlled oxygen therapy (68% vs 31%, P < 0.001).
There has been good uptake of projects with a high completion rate and clinical significant improvements. With careful planning, cross-organisation evidence-based practice implementation projects can be successful.
Does 0.9% saline therapy reduce alcohol intoxication in the emergency department: a randomised controlled trial
Siegfried R Perez MD (trainee),1,2 Gerben Keijzers MBBS FACEM,1,2,3 Michael Steele PhD,4 Joshua Byrnes PhD,3,5 Paul A Scuffham PhD3,5
1Department of Emergency Medicine, Gold Coast Hospital, Queensland Health, Brisbane, Australia; 2Faculty of Health Sciences and Medicine, Bond University, Brisbane, Australia; 3Griffith Health Institute, Griffith University, Brisbane, Australia; 4Faculty of Business, Bond University, Brisbane, Australia; 5Centre for Applied Health Economics, School of Medicine, Griffith University, Brisbane, Australia
Intravenous saline is frequently used in emergency department (ED) patients with acute alcohol intoxication despite the lack of evidence for its efficacy.
To compare combined treatment with intravenous 0.9% saline and observation against observation alone in ED patients with acute alcohol intoxication.
Single blind randomized controlled trial.
One tertiary and one urban ED in Queensland, Australia.
144 ED patients with uncomplicated acute alcohol intoxication who required observation.
Eligible patients were randomized to either 20ml/kg intravenous 0.9% saline bolus with observation or observation alone.
Main outcome measures
ED length of stay (EDLOS), breath alcohol levels and intoxication symptom scores using the Observed Acute Alcohol Intoxication (OAAI) tool.
Groups were comparable at baseline; blood alcohol levels were similar between groups (0.185 vs 0.195, P = 0.44) as were intoxication scores (22.3 vs 22, P = 0.90). In the intention to treat analysis, patients receiving 0.9% saline therapy had a similar EDLOS (274 min vs 287 min, P = 0.61), intoxication score decreases (3.7 vs 4.9 points per hour P = 0.16) in the first 2 hours as well as time of discharge (4.5 vs 3.7, P = 0.16).
Intravenous fluids do not decrease length of stay and intoxication symptom scores compared to observation alone. This study suggests that either approach is reasonable, and observation alone may less resource intensive.
Australian and New Zealand Clinical Trials Register: 12611000938909.
Missed chances for prevention of shaken baby syndrome
Amanda Stephens BA MBBS PhD,1 Naren Gunja, MBBS MSc FACEM FACMT,1,2 Otilie Tork, BSc(Med) MBBS FRACP,3 R Kim Oates MD DSc MHP FRACP FRCP4
1Emergency Department, Westmead Hospital, Sydney, Australia; 2Discipline of Emergency Medicine, Sydney Medical School, University of Sydney, Sydney, Australia; 3Child Advocacy Service, Royal Children's Hospital, Brisbane, Australia; 4Discipline of Paediatrics & Child Health, Sydney Medical School, University of Sydney, Sydney, Australia
High morbidity and mortality from non-accidental head injury (NAHI) in children makes prevention via early detection of at-risk populations essential. However, NAHI is a particularly contentious area of emergency medicine with significant controversies in the literature. Many cases are missed and even when abuse is suspected, studies show that over 50% of doctors have, at some time, chosen not to report.
Examine the frequency of prior medical contact in children with suspected NAHI, assess the nature of such contact and analyse the reasons for missed diagnoses and/or non-reporting.
A retrospective cohort study of consecutive cases of suspected NAHI presenting to a metropolitan children's hospital. Hospital records, community services data, and information from legal proceedings were reviewed to obtain details of contact with medical professionals prior to the index presentation with suspected NAHI.
68 cases were reviewed. 18% were seen by a health practitioner for signs or symptoms associated with maltreatment prior to diagnosis with NAHI. Seven had presented to emergency departments. Six had facial/scalp bruising. One pre-mobile infant had leg bruising. Another had a spiral fracture. Four children presented with non-specific signs such as lethargy and vomiting in the context of parental coping difficulties.
A significant number of children had relevant prior medical contact, but were not reported to child protection services. As children are frequently abused on multiple occasions with worsening brain damage resulting from recurrent insults, it is essential that ED staff be alert to the possibility of NAHI.
Evaluation of a rapid adenoviral detection test for confirmation of adenoviral conjunctivitis
Anita Ng MBBS,1,3 Pat Usher RN BAppSci (Nursing) BNursing Informatics,1 Lucy Busija BA (Hon), GradDipHealthStat MSc (AppStat) PhD,1,4,5,6 Carmel Crock MBBS FACEM BLitt,1 Atul B Shah MRCSEd PhD1,2
1Royal Victorian Eye and Ear Hospital (RVEEH), Melbourne, Australia; 2West of England Eye Unit, RD&E Hospital, Exeter, UK; 3Emergency Department, Royal Melbourne Hospital, Melbourne, Australia; 4Centre for Eye Research Australia (CERA), Melbourne, Australia; 5Centre for Research Excellence in Translational Neuroscience at Melbourne Brain Centre, The University of Melbourne, Melbourne, Australia; 6Melbourne Centre for Clinical Epidemiology, Biostatistics and Health Services Research, The University of Melbourne, Melbourne, Australia
Evaluation of efficacy of a Rapid Pathogen Screening (RPS) Adeno Detector test, an immuno-chromatographic test that detects adenoviral antigens from eye fluids (RPS, Inc., South Williamsport, PA, USA) for confirmation of adenoviral conjunctivitis, using adenoviral PCR as the gold standard.
A total of 116 patients with presumptive diagnoses of viral conjunctivitis were recruited from the RVEEH Emergency Department. Participants were first swabbed for the RPS Adeno Detector test then again for the adenoviral PCR test. The RPS Adeno Detector test result was read and recorded on the provided proforma by the trained recruiter. The PCR test result was followed up by the investigator.
Results show that the RPS Adeno Detector test is 54.5% sensitive (95% CI 41.1–69.6%) and 82.8% (95% CI 68.7–88.9%) specific. The test has a positive predictive value of 70% and a negative predictive value of 72%. These results are substantially lower than previously published values of a sensitivity of 89% and specificity of 94% for the RPS Adeno Detector test (Ophthalmology 2006; 113 (10): 1758–1764).
The results suggest that the RPS Adeno Detector test has moderately high specificity but low sensitivity for use as a sole test to confirm adenoviral conjunctivitis in the emergency department.
Impact of climate on prevalence and severity of bronchiolitis in Australia and New Zealand: a retrospective observational international multicentre study
Tobias Hoeppner,1 Meredith Borland2
1Emergency Department, Princess Margaret Hospital, Perth, Australia; 2Schools of Paediatrics and Child Health and Primary, Aboriginal and Rural Health Care, University of Western Australia, Perth, Australia
This abstract presents data for the first of eight hospitals in four climatically and geographically different metropolitan regions of Australia and New Zealand that participated in two multicentre studies of bronchiolitis. This will be extended to include data from the remaining sites which will become available by mid-2012.
Bronchiolitis is the leading cause of hospital admission during the first year of life and contributes substantially to the workload of Emergency Departments during winter months. While seasonality of bronchiolitis is well-documented, there is little published evidence examining the impact of local climate and weather on prevalence and severity of illness.
Data for all emergency presentations and inpatient admissions of infants aged 2 to 12 months to a tertiary paediatric hospital in Western Australia from 2009 until 2011 were collected. Length of stay, supplemental oxygen requirement and ICU admission served as markers for severity of illness. Official weather records were used to examine a correlation of weather with prevalence and severity of bronchiolitis.
Data was extracted for 462 hospital admissions. 322 (69.8%) of admitted infants required supplemental oxygen, 20 (4.3%) were admitted to ICU. Median length of stay was 1.81 days. The onset of peak season was preceded by a decrease in mean weekly temperatures of at least 3 degrees Celsius in all three years. Compared with the other years, mean temperatures were lower during the winter of 2010. During this season, there were more presentations to ED, a higher admission rate and all assessed outcome measures indicated more severe illness in affected individuals.
Data analysed thus far represents a useful pilot sample. The results of this study suggest an inverse correlation between temperature during winter and prevalence as well as severity of bronchiolitis. This hypothesis will be tested against seven other samples from Australia and New Zealand.
Why older patients of lower clinical urgency choose to attend the Emergency Department
Judy Lowthian PhD MPH BAppSc (SpPath),1 Cathie Smith BSW Cert Bereavement Counselling,2 De Villiers Smit MBChB FACEM,2 Johannes Stoelwinder MD FRACMA FACHSE,1 John McNeil AO MBBS PhD FRACP FAFPHM,1 Peter Cameron MBBS MD FACEM1,2
1School of Public Health & Preventive Medicine, Monash University, Melbourne, Australia; 2Alfred Hospital Emergency & Trauma Centre, Melbourne, Australia
To determine non-clinical factors associated with Emergency Department (ED) attendance by lower urgency older patients.
Design, Participants and Setting
A descriptive study comprising structured interviews with lower urgency community-dwelling patients aged ≥70 years presenting to a tertiary metropolitan Melbourne public hospital ED.
Demographic and clinical characteristics, self-reported feelings of social connectedness, perceived accessibility to primary care, reason for attending ED.
100 patients were interviewed: mean age 82 years, 56% female, 57% lived alone; 73% presented during business hours, 58% arrived by ambulance, 80% presented for illness, and 65% were discharged home within 48 hours.
Overall, 56% of patients reported feelings of social disconnectedness. Of the patients not living alone, 65% reported feeling disconnected; compared with 49% of patients who lived alone.
All patients attended a regular General Practitioner, 31% reporting regular review appointments. Thirty-five percent reported waiting-times >2-3 days for urgent problems; 59% stated accessing care ‘after hours’ without attending ED as difficult, with 20% having attended ED 3–6 times in the previous 12 months.
Reasons for attending ED were referral by a third party, difficulty with accessibility to primary care, patient preferences for timely care, and fast track access to specialist care.
Most older patients of lower clinical urgency presented to ED due to perceived access block to primary or specialist services, alongside an expectation of more timely and specialised care. This suggests that EDs should be re-designed and/or integrated community-based models of care developed, to meet the specific needs of this age group who have growing demand for acute care.
The hows and whys of reducing residential aged care to ED transfer: a personal perspective
Glenn Arendts MBBS MMed FACEM1,2
1School of Primary, Aboriginal and Rural Health Care, University of Western Australia, Perth, Australia; 2Centre for Clinical Research in Emergency Medicine, Western Australian Institute for Medical Research, Perth, Australia
There is now good evidence that a number of individual strategies can significantly reduce transfers from residential aged care facilities to ED. Besides a reduction in the number of transfers, other relevant outcomes of these strategies (clinical; economic; qualitative) are mostly inferred or poorly described. Because of this, programs to reduce ED transfer from aged care have been criticised as a form of ageist rationing of hospital care. If improved uptake of these programs is desirable, evaluation of meaningful outcomes is required unless a strong ethical justification for rationing can be provided.
Aims and Content
This presentation will:
- Describe HOW to reduce ED transfer by summarising my own research and the work of others;
- Describe WHY outcome measures besides transfer reduction are important and need greater evaluation, especially from the perspective of transfer decision makers;
- Provide a proposed framework that identifies evidence gaps, future research to address these gaps and an ethical justification for transfer reduction where research cannot feasibly be undertaken.
Demography is destiny: ED research for the ageing population
Glenn Arendts MBBS MMed FACEM,1,2 Judy Lowthian BAppSc MPH PhD3
1School of Primary, Aboriginal and Rural Health Care, University of Western Australia, Perth, Australia; 2Centre for Clinical Research in Emergency Medicine, Western Australian Institute for Medical Research, Perth, Australia; 3School of Public Health & Preventive Medicine, Monash University, Melbourne, Australia
Emergency Department (ED) presentations are increasing disproportionately in the older population, many of whom carry a high burden of chronic disease and co-morbidity. Nearly half of all hospital bed days in Australia are devoted to people aged 65 and over, and the proportion of the population in this age bracket will double by 2050. Older people have an increased risk of ED attendance, re-attendance and admission to hospital, which impacts on all aspects of emergency/acute hospital care and patient flow. If ‘demography is destiny’, EDs will inevitably be overwhelmed by the acute care needs of older people in our community in coming decades.
Emergency medicine research has logically been focussed on care within the confines of the ED, but this constraint cannot sufficiently address the most important issues associated with the persisting rise in demand by older people for acute medical services.
Aims and Content
This presentation will draw on our current research programs in several states to describe:
- Past and projected trends in emergency health service usage by older people
- The key drivers of increasing ED presentations in this population
- The consequences of increasing hospitalisation rates in the elderly
- A rational research agenda for geriatric emergency medicine that addresses potential solutions, highlighting research priorities both within and beyond the ED. This will include a summary of our research thus far, as well as ongoing and planned research in this field.
Effect of hospital in nursing home program on the ED presentations from aged care faculties among three Queensland hospitals: a pilot study
William Lukin,1 Jingzhou Zhao,2 Rhonda Purtill RN,3 Sam Tapp RN,4 Kerry Robinson,5 Shuang Zhong,2 Xiang-Yu Hou1,2
1Dept of Emergency Medicine, Royal Brisbane and Women's Hospital, Brisbane, Australia; 2School of Public Health and Social Work, Queensland University of Technology, Brisbane, Australia; 3Clinical Nurse Consultant, Hospital in the Nursing Home Royal Brisbane and Women's Hospital, Brisbane, Australia; 4Clinical Nurse, Hospital in the Nursing Home, Royal Brisbane and Women's Hospital, Brisbane, Australia; 5Clinical Nurse Consultant, Aged Care Early Intervention and Management, Gold Coast Hospital, Brisbane, Australia
The burden placed on Emergency Departments (EDs) by residents of Residential Aged Care Facilities (RACFs) is not well known. The current pilot study aims to help answer this question. In addition we examine whether having an ED-based hospital avoidance service for these patients can influence this burden.
Four months data (March–June 2011) was collected at three hospitals in Southeast Queensland (Hospital R, Hospital G, and Hospital L; where Hospital L does not have an ED-based hospital avoidance program). The number of presentations and admissions from RACFs per 1000 aged care beds in the catchment areas were compared among the three hospitals. Pearson chi-square test for the admission rate and log-linear models for the number of presentations were used for data analysis.
For the four-month period, the number of presentations from RACFs per 1000 beds in Hospital L (76.24) was significantly higher than that in Hospital R (44.76) and Hospital G (47.13); and the admission rate per 1000 RACF beds in Hospital L (43.5) was significant higher than that in Hospital R (23.81) and Hospital G (31.35).
ED-based hospital avoidance programs could be effective in reducing the ED presentations from NHs and hospital admissions among the NH patients.
Does a screen saver reminder help improve hand hygiene practices in the ED setting?
Kim M Yates MBChB MMedSc(Hons) PGCertClinEd FACEM,1 Aashish Raj MBChB,1 Stuart Dalziel FRACP PhD2,3
1Emergency Departments, North Shore & Waitakere Hospitals, Waitemata District Health Board, Auckland, New Zealand; 2Children's Emergency Department, Starship Hospital, Auckland, New Zealand; 3Liggins Institute, University of Auckland, Auckland, New Zealand
Hand hygiene is a vital element of infection control in the health care setting. The World Health Organisation (WHO) recommends that health care workers clean their hands at 5 moments associated with patient care. Audits at our hospitals showed that compliance with these WHO recommendations is not high and that doctors had the lowest hand hygiene rates of all health care workers. In this study we aimed to evaluate the effectiveness of a visual screensaver reminder intervention for improving compliance with WHO recommendations.
In September-October 2010, the hand hygiene practices of ED doctors working at the Starship Children's ED were studied. Hand hygiene observations were completed by an observer trained to audit the WHO 5 moments of hand hygiene. The screensaver intervention consisted of photos of emergency physicians and the ED nurse manager cleaning their hands with speech bubbles reminding staff of the moments when they should be cleaning their hands. Hand hygiene audits were performed at baseline and 4 weeks after the addition of the screensaver to all computers in the department.
529 hand hygiene moments were audited at baseline and 520 post-intervention. Compliance with WHO recommendations significantly improved from 40.3% to 50.8% (P = 0.0003). The moments showing greatest improvement were cleaning hands after a procedure or body fluid exposure (36.4% v 69.2% [P = 0.0009]) and after patient contact (23.2% v 46.9% P < 0.0001).
The screensaver intervention led to a modest 10% overall improvement in compliance with WHO hand hygiene recommendations, with bigger improvements for specific moments.
Patients undergoing appendectomies: The effects of practice change
Alexa E Farrell MBChB (trainee),1 Drew Richardson MBBS FACEM GradCertHE MD1,2
1Emergency Department, The Canberra Hospital, Canberra, Australia; 2Australian National University Medical School, Canberra, Australia
There have been major changes in the diagnosis and management of suspected appendicitis.
To describe changes in the ED management of patients who have an appendectomy.
Retrospective chart review of all patients admitted through a mixed adult/paediatric tertiary ED who had appendectomy in 2001 and 2010.
Rate of abdominal X-ray fell from 51/252 (20.2%, 95CI 15.6–25.8) in 2001 to 15/343 (4.4%, 2.7–7.3) in 2010. Rates of CT rose from 6.6% (4.0–10.7) to 13.4% (10.1–17.6), and of ultrasound non-significantly from 27.7% (22.2–33.9) to 34.7% (29.7–40.0). 2 patients in 2001 and 1 in 2010 had both CT and US. In those who had CT or US the pathologically proven rate of appendicitis was 76.9% (65.8–85.4) in 2001 and 71.3% (63.7–78.0) in 2010, compared to 84.5% (78.0–89.4) and 77.1% (70.1–82.9) in those without imaging. Patient care time (PCT), from seen by doctor to departed ED, was less than 6 hours for 204 in 2001 and 205 in 2010, but because of increased presentations, 80th Centile PCT increased from 5:58 (hh:mm) to 8:21. Time from departing ED to appendectomy showed a similar pattern, with 192 (2001) and 194 (2010) within 12 hours, but 80th Centile delays of 12:33 and 21:02. 9% of patients with pathologically proven appendicitis in 2010 had a normal CRP and WCC.
There has been increased use of CT and US and increased delays to operation without a decrease in negative appendectomies. The data suggest that increasing demand has not been associated with increased treatment capacity.
Design and rollout of standardised approach to paediatric sedation in Victorian emergency departments
Anne-Maree Kelly,1,2 Jan Pannifex,1 Elif Cetiner1 on behalf of the Paediatric Procedural Sedation Reference Group, Emergency Care Improvement and Innovation Clinical Network
1Emergency Care Improvement and Innovation Clinical Network (ECIICN), Commission for Hospital Improvement, Department of Health (Victoria), Melbourne, Australia; 2Joseph Epstein Centre for Emergency Medicine Research, Western Health, Melbourne, Australia
Children often require minor procedures in emergency department (ED) for which sedation is needed. Information from EDs in Victoria indicated that processes for paediatric procedural sedation were variable both between and within health services. ECIICN was aware that the Royal Children's Hospital Melbourne and Sunshine Hospital had collaborated to develop a standardized approach, including training and credentialing. This had not been widely disseminated.
This project aims to improve safety and reduce variation in clinical practice with respect to paediatric procedural sedation in EDs. It is managed by ECIICN with support of an expert reference group; however implementation is conducted at the local ED level, with scope for local adaptation if required.
The approach is multimodal and grounded in quality and safety theory. It includes an evidence-based training manual (and associated materials e.g. for assessment of competence), a treatment record form that aids patient selection and preparation, a train-the-trainer day for local lead clinicians, before and after clinical governance self-assessment tools and a training DVD on child and family centred care. The project will be evaluated by a clinical audit, review of key clinical governance criteria, number of clinicians trained and qualitative feedback about the training program and resources.
15 ED are participating; 11 metropolitan and four regional/rural. Results of the clinical audit and clinical governance changes as a result of the project will be presented.
Age-and gender-specific changes in ED demand
Drew Richardson MBBS FACEM GradCertHE MD
Australian National University, Canberra, Australia
Demand for ED services is rising faster than population growth. There has been little study of whether demand increases are evenly distributed by age and gender.
To identify changes in age- and gender-specific ED demand in a setting of rising overall demand.
Observational comparison of population and ED data from July 2005 to June 2006 and July 2010 to June 2011 in a tertiary ED with over 80% of presentations living in a defined local population. Presentations from the two periods were classified as Primary, Secondary or non-local by stated location of residence. Presentations were age-standardised using 5-year groups to the June 2001 national population (as ABS recommends).
Presentations from the local population increased by 12.5% over 5 years from 44493 to 50042. From the Primary area, age specific presentations increased by 6.7% from 208 per 1000 [95% CI 205–211] to 222 [218–225] in females but declined by 3.7% from 228 [225–231] to 220 [217–223] in males. From the Secondary area, female presentations increased by 28.1% (47 [45–49] to 61 [59–62] ) and males by 13.0% (62 [60–64] to 70 [68–72] ). Subgroup analysis showed age-standardised ambulance arrivals rising 12.7% in females and 7.7% in males from Primary, and 22.6% and 21.3% from Secondary. The increase in Primary area ambulance arrivals was not seen in those over 75 years.
Age-specific presentations from this ED's primary population have increased in females but decreased in males over 5 years of increasing demand. This is a new finding.
Low incidence of available advance directives in elderly ED patients who do not survive 60 days
Catherine Greenshields Medical student (4th year), RN BNurs,1 Drew B Richardson MBBS FACEM GradCertHE MD1,2
1Australian National University, Canberra, Australia; 2Emergency Department, The Canberra Hospital and Australian National University Medical School, Canberra, Australia
Documented advance directives (ADs) are beneficial to staff and patients when EDs must manage complex unwell elderly patients near their end-of-life.
To identify the incidence of documented ADs in ED patients who subsequently died and any relationship with residential care living.
Retrospective chart review of all patients over 65 presenting to a mixed tertiary ED in one year who died within 60 days, identified from a previously validated register. Data abstracted were documented existence of AD at arrival, residential care status and whether a documented end-of-life decision was made during the ED visit.
9077 visits were made by 6314 patients aged over 65, of whom 506 (8.0% [95% CI 7.4–8.7] ) died within 60 days (287 as inpatients, 36 in ED, 183 elsewhere). 6 cases were excluded for inadequate notes. Residential care patients were much older (85.8 [84.6–87.0] years vs 78.6 [77.7–79.6] ) and spent longer in ED. Only 22 patients had documented AD on ED arrival, 6/353 (1.7% [0.7–3.9] ) who lived at home and 16/147 (10.9% [6.6–17.3] ) in residential care (P < 10–5, χ2). Some 24.1% [19.8–29.0] of those living at home and 42.9% [34.8–51.2] of those in residential care had a documented decision about appropriateness of further care made during their ED visit, and only 10/148 of these decisions were to continue full active treatment.
Only 4.4% of elderly ED patients who did not survive 60 days had any form of documented advance directive available at presentation, yet a further 29.6% had end-of-life decisions made in the ED.
A comparison between emergency physicians (EP) & intensivists of the considerations taken into account when withdrawing life sustaining health care
Philip Richardson MBChB FACEM LL.B
Royal Brisbane & Women's Hospital, Brisbane, Australia
Withdrawal of acute life sustaining health care, or a decision not to offer life sustaining health care or to only offer limited life sustaining health care is a core clinical practice that uniquely occurs in most Emergency Departments (EDs) every day.
A similar decision making process also occurs in the Intensive Care Unit (ICU).
Intensive Care Physicians were ask to complete a questionnaire indicating the considerations they took into account, and there importance, when limiting or withdrawing treatment in the ICU
This was compared to a questionnaire completed by EP when limiting or withdrawing treatment in the ED.
The objective was to ascertain if there were any significant differences between the 2 specialities.
What does it take to be in charge of the ED? A qualitative study
Simon Craig FACEM GCHPE,1,2 Jonathan Dowling FACEM PgCertTox (Cardiff)1,2
1Emergency Department, Monash Medical Centre, Melbourne, Australia; 2Monash University, Melbourne, Australia
This paper describes the views of senior emergency medicine trainees and emergency physicians (EPs) about the ‘in charge’ role following the introduction of a targeted educational program in 2011–2012. The program comprised ‘registrar in charge’ (RIC) shifts, which allocated a senior emergency registrar to be ‘in charge’ of the ED from morning handover with one-to-one supervision and support by an EP.
Each registrar was allocated at least one RIC shift during their 13-week ED term. Structured questionnaires were administered to registrars and EPs, and gathered qualitative data regarding perceptions of important qualities required to run an emergency department, what the EP was hoping to teach during the shift, and whether the trainee learned anything new or unexpected about the in-charge role. Data were analysed using grounded theory methodology.
During the study period, 16 senior ED registrars were rostered for 26 RIC shifts. The study questionnaires were completed by 16/16 registrars and 13/16 EPs. Major themes that were common to registrars and EPs included communication skills, knowledge and experience, delegation, professionalism and organisational skills.
Additional themes that were more prominent in EP responses included multitasking, dealing with interruptions, managing patient flow, and being aware of the whole department. Registrars emphasised the importance of looking after fellow staff members, and communicating with the nurse in charge.
EPs and registrars appear to have different perceptions about the important qualities required to run an emergency department. Targeted education to address these differences may assist registrars adjust to the in-charge role.
Guideline compliance in the international response to disasters: the response to Haiti examined from the perspective of an emergency medicine trainee in the field
Nicky Dobos MBBS MPH
Emergency Department, Royal Melbourne Hospital, Melbourne, Australia
The earthquake in Haiti on 12 January 2010 left more than 1.5 million people displaced, over 300 000 people injured and an estimated 220 000 deaths. The scale of the resulting morbidity far outweighed the capabilities of an already poor health infrastructure. Following previous responses by the international health community to disasters in resource-poor settings, the World Health Organization (WHO) and the Pan American Health Organization (PAHO) had developed a report and strategy to improve the cost-effectiveness of utilising foreign field hospitals (FFHs) in the aftermath of sudden impact disasters. Despite the lessons learnt from previous disasters requiring international assistance, the global health response to the earthquake in Haiti was perceived by many state and independent organisations as uncoordinated, ill-equipped and fraught with unacceptable shortcomings.
This presentation will examine the international health response in Haiti with reference to the WHO/PAHO international guidelines, comparing the viewpoints of both the 2012 report and those of a trainee emergency specialist working in the field.
Firstly, the contents of the WHO/PAHO guidelines will be summarised. The lessons learnt will then be described against these guidelines using dual information sources: the subsequent 2012 report and the individual emergency medicine perspective of the author as both a witness and worker in the field.
From the perspective of both the 2012 report and the author's structured observations, there were many instances in which the WHO/PAHO guidelines were not adopted successfully or at all. Some examples include:
- Lack of coordination, organization and communication between FFHs
- Uneven allocation of resources and lack of basic quality standards
- Lack of surveillance systems,
- Unsatisfactory security levels
- Poor mental health and psychological support
- Poor hygiene and sanitation in camps and hospitals, and unsafe disposal of medical waste and dead bodies
Despite the existence of guidelines in the international response to disasters in resource-poor settings, there were many failures of guideline compliance during the Haiti earthquake response.
In your expert opinion: the use and abuse of medical evidence
Amanda Stephens BA MBBS PhD
Emergency Department, Westmead Hospital, Sydney, Australia
Clinical findings from emergency department presentations often form the basis for medical evidence used in legal proceedings. In medicine, and particularly emergency medicine, some uncertainty (of diagnosis, prognosis or treatment) is inevitable and does not preclude sound medical management. However, the uncertainty taken for granted in the ED has serious repercussions for legal outcomes when medical science is translated into evidence for the courts.
This presentation examines the law's treatment of expert medical opinion and scientific research. It uses data from a large study of suspected non-accidental head injury cases (‘shaken baby syndrome’) and analyses of case law to show that medicine and law have fundamentally different understandings of the term ‘evidence’. It will also be demonstrated that the law is poorly equipped to distinguish between good and bad science. This is seen, for example, in courts' acceptance of experts who propose that vaccines cause subdural and retinal haemorrhages.
The practice of law is an art. Medicine involves both science and art. The challenge in child protection is to successfully marry them to ensure just outcomes. The art of translating medical evidence into courtrooms requires the careful interpretation of the scientific uncertainties.
It is salutary to appreciate that whilst acknowledgement of uncertainty is a virtue in medicine, it is often the opposite in the law where even slight uncertainty may introduce ‘reasonable doubt’ and, at least in child protection cases, may result in the baby being thrown out with the bathwater.
Developing a consensus on measuring quality in the ED
Peter Cameron MBBS FACEM MD
Chair Emergency Medicine, Hamad Medical Corporation, Doha, Qatar
The importance of the specialty of Emergency Medicine in modern healthcare systems is now becoming clear to most health administrators and policy makers. With increased funding and associated scrutiny, the community needs to know whether the emergency service they are getting is good/bad or indifferent. Unfortunately, the most prevalent measure of ‘good performance’ is the length of time patients stay in the ED.
Although this is a useful process measure, we do a lot more in EDs than push people through quickly.
A consensus meeting was held in London in November 2011 sponsored by the International Federation for Emergency Medicine and hosted by the UK College of Emergency Medicine. This was followed by a meeting in Dublin 2012 (at the ICEM) to reach an international consensus on how best to measure quality in emergency medicine. The implications of this international consensus are potentially important as we push governments to improve the quality of emergency care internationally.
This presentation will focus on the context and outcomes of this consensus driven process.
Trauma registries: What is the experience in developing countries?
Gerard O'Reilly MBBS FACEM MPH (International Health) MBiostat,1,2 Manjul Joshipura MBBS MS,3 Peter Cameron MBBS FACEM MD,1,2 Russell Gruen MBBS PhD FRACS4
1Victorian State Trauma Registry, Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Commercial Rd, Melbourne, Australia; 2Emergency and Trauma Centre, Alfred Health, Melbourne, Australia; 3Department of Violence & Injury Prevention & Disability (VIP), World Health Organisation, Geneva, Switzerland; 4National Trauma Research Institute, Alfred Health, Commercial Rd, Melbourne, Australia
Trauma systems reduce mortality. The trauma registry is a key driver of improvements in trauma care. Developing countries have begun to develop trauma systems but there is relatively little trauma registry experience in less-resourced settings.
The aim of this study was to review the published experience of trauma registries in developing countries.
A structured review of the literature was performed. All abstracts referring to a trauma registry in a developing country as defined by the United Nations Human Development Index were included. The Trauma Registry Assessment Tool was applied to the search results to collect and analyse the experience of trauma registries in developing countries.
There were 84 articles, 76 of which were sourced from 46 registries. The remaining 8 articles were perspectives. Most were from Iran, followed by China, Jamaica and Uganda. Regarding data methodology, only 4 registries used ISS to define inclusion criteria. Most registries collected data on variables from all five variable groups. Several registries collected data for less than 20 variables. Only 3 registries measured disability using a score. The most common scores of injury severity were ISS, followed by RTS, TRISS and the Kampala Trauma Score.
Several countries are responsible for the majority of developing country trauma registry publications. ISS is unlikely to be used to define inclusion criteria, and the Kampala Trauma Score is not uncommonly used to measure injury severity. The trauma registry experience of developing countries can inform trauma registry development in similar settings.
Predictors of missing data in a trauma registry: an observational study
Gerard O'Reilly MBBS FACEM MPH (International Health) MBiostat1,2, Peter Cameron MBBS MD FACEM1,2, Damien Jolley MSc (Epidemiol) MSc AStat3
1Emergency and Trauma Centre, The Alfred, Commercial Rd, Melbourne, Australia; 2Victorian State Trauma Registry, Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia; 3Centre of Research Excellence in Patient Safety, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia
Trauma registry data are almost always incomplete. Traditional approaches to dealing with missing data lead to biased outcomes and conclusions. Multiple imputation can reduce bias in registry analyses but the ideal approach would be to improve data capture.
The aim of the study was to identify which patients were most likely to have incomplete data.
An analysis of prospectively collected regional trauma registry data over one year was performed. Variables used for trauma system benchmarking were analysed. Logistic regression analyses were performed to identify predictors of missingness. These analyses were conducted following complete data estimation using multiple imputation.
There were 2520 cases. The variables with the greatest proportion of missing observations were respiratory rate, GCS and systolic blood pressure. Data for these physiological variables were more likely to be missing when the patient died in hospital. GCS and respiratory rate were more likely to be missing when they were abnormal. A major predictor of a missing GCS or respiratory rate was an abnormal pre-hospital GCS; an additional predictor of a missing blood pressure was whether the patient was transferred to a second hospital for definitive care.
Death in hospital was a predictor of missingness for all of the primary hospital physiological variables. An abnormal GCS and respiratory rate were more likely to be missing from the dataset. In addition to reducing bias during data analysis, multiple imputation can be used to inform improvements in data capture.
Population changes do not explain rising demand for emergency care by the elderly
Judy Lowthian PhD MPH BAppSc (SpPath)
School of Public Health & Preventive Medicine, Monash University, Melbourne, Australia
The aim of this research program (5 studies) was to analyse the main elements of increasing demand for emergency services across metropolitan Melbourne (pop 4.2 million) with a special focus on clarifying the impact of an ageing population. Despite a number of initiatives by health care funders (in this case State Government) and service providers, demand continues to rise. This issue has major implications for quality of care and future planning and resourcing.
Longitudinal analyses were undertaken of >10,000,000 patient episodes comprising emergency ambulance transportations (1995–2008), hospital ED presentations (2000–2009) and hospital admissions through EDs (2000–2009), using unique population-based datasets and data linkage. Numbers and rates of transportations, ED presentations and emergency admissions; and ED and hospital length of stay (LOS) were analysed. Predictive modelling forecast future demand; and regression models evaluated the effects of multiple factors on these outcomes. A descriptive study explored elderly patient viewpoints on their needs for urgent care.
The volume and rates of utilisation of all emergency healthcare services over the study periods increased beyond that expected from population changes. Emergency transportations rose by an average annual rate of 4.8% (95% CI 4.3 to 5.3%); the greatest increase in usage by the elderly. Predictive modelling suggested further increases up to 69% by 2015. Investigation of the trends in ED presentations identified similar increasing disproportionate representation of the elderly. ED attendances by all age groups rose annually by 3.6% (95% CI 3.4 to 3.8%). ED LOS increased over the study period for older, more acutely unwell patients requiring hospital admission. The number of elderly patients re-attending ED ≥3 times within a 12 month period doubled in a decade, contributing to 23% of total elderly visits in 2008/2009.
Further analysis of patients admitted to hospital from EDs identified a growth in emergency hospital admissions, driven by a 60% increase in the number of single day/overnight admissions, representing an average annual growth rate of 6.1% (95% CI 5.7 to 6.5%). The elderly were disproportionately represented amongst these short-stay admissions.
Because of the disproportionate increase in utilization by the elderly, interviews were conducted with patients of low urgency, aged ≥70 years, who could have chosen to use alternative providers. The main reasons underpinning ED attendance were expectations for timely specialist care for chronic conditions and reduced accessibility to primary care including home visits.
There is a continuing increase in demand for emergency care by the elderly that is greater than population growth. Expectations of care for the elderly, by patients and their carers are increasing, and EDs might not be best placed to manage this. The solution probably lies in a community-based system of care currently not available.
Do you reckon I should sit? Can we predict success in the ACEM Fellowship Examination?
Simon Craig FACEM GCHPE,1,2 Jennifer Brookes2,3
1Emergency Department, Monash Medical Centre, Melbourne, Australia; 2Monash University, Melbourne, Australia; 3Emergency Department, Box Hill Hospital, Eastern Health, Melbourne, Australia
The ACEM fellowship examination has a worrying low pass rate (less than 60% of candidates passing from 2009 onwards). Our objective was to identify factors that allow reliable prediction of success or failure – information which may be useful to those involved in examination preparation.
From 2010 to 2012, five consecutive cohorts of Victorian ACEM trainees sat a marked practice written examination and completed surveys regarding examination preparation. Subsequent ACEM fellowship examination outcome (pass/fail/deferred) was obtained by contacting trainees and using ACEM annual reports.
Univariate analysis was undertaken to determine which factors would most reliably predict examination success. ‘Intention to sit’ analysis was undertaken – each sitting of the practice examination was viewed separately.
59 individuals sat the practice exams, with 82 datasets available (a number sat the practice exam more than once). Pass/fail data was available for 52 datasets, and pass/‘no pass’ data available for all 82 datasets.
The strongest positive predictors for success included higher scores in the practice examination (OR 1.77, 95% CI 1.20–2.62), and belief that working on the floor (OR 1.95, 95% CI 1.005–3.813) and attending a course (OR 1.85, 95% CI 1.02–3.35) was helpful for examination preparation. Years since graduating from medical school (OR 0.65, 95% CI 0.54–0.844) and the use of medical journals (OR 0.311, 95% CI 0.109–0.883) were negatively associated with examination success.
Performance in a practice exam is a strong predictor of success or failure in the ‘real’ examination. Other identified factors warrant further study.
The physiological effects of hydrocortisone replacement in early septic shock
Matthew Maiden BSc BMBS FACEM FCICM,1,2 Marianne Chapman BMBS PhD FANZCA FCICM,1,2 Guy L Ludbrook MBBS PhD FANZCA,2 David J Torpy MBBS PhD FRACP,3 Tim R Kuchel BVSc MVS,4 Iain J Clarke PhD5
1Intensive Care Unit, Royal Adelaide Hospital, Adelaide, Australia; 2Discipline of Acute Care Medicine, University of Adelaide, Adelaide, Australia, 3Department of Medicine, Royal Adelaide Hospital, Adelaide, Australia; 4South Australian Health and Medical Research Institute (SAHMRI), Adelaide, Australia; 5Department of Physiology, Monash University, Melbourne, Australia
The role of hydrocortisone (HC) in the management of septic shock remains controversial. While there have been a number of clinical trials with conflicting mortality benefit, the physiological effects of providing HC in early septic shock are unknown.
Subjects and Setting
Prospective blinded randomised placebo controlled pre-clinical trial in a validated 24-hour ovine model of septic shock.
Animal ethics approval was obtained from all institutions. Sixteen ewes were anaesthetised for insertion of carotid arterial line, pulmonary artery catheter, tracheostomy and insertion of cannulae in the coronary sinus, renal vein, hepatic vein and femoral vein. Animals remained sedated and received protocol guided management of ventilation, fluids and noradrenaline. Sepsis was induced by intravenous E. coli. After 2 hours of sepsis, sheep received continuous infusion of hydrocortisone (0.1 mg/kg/hour) or placebo. Primary outcome was 24-hour total dose of noradrenaline (NorA). Secondary outcomes included haemodynamics, metabolic status and renal function.
Study group baseline demographics were evenly matched. All animals developed a hyperdynamic septic response. Plasma cortisol fell in placebo animals (P < 0.01) and was maintained at pre-septic levels by HC treatment (P < 0.05). Two animals in the placebo group died before 24-hours. The 24-hour NorA dose was no different between placebo group (208 ± 160 μg/kg) or HC group (167 ± 101 μg/kg; P = 0.73). There was no treatment effect on any haemodynamic variable, temperature, pH, lactate, organ-specific oxygen extraction or renal function.
In a 24-hour model of septic shock, HC infusion did not reduce NorA requirement nor alter any other measured physiological variable.
Primum non nocere: how to reduce the harm from PE investigation
Marcin A Sosnowski MBChB FRCS Ed FCEM Consultant EM,1 Sayed Hamid MBChB RMO,2 Alina Ali MBChB RMO2
1Whanganui District Health Board, Whanganui, New Zealand; 2Whanganui Hospital, Whanganui, New Zealand
Over-investigation of low-risk patients with suspected pulmonary embolism (PE) represents a growing problem. The current practice tends to involve a clinical pathway, initiated by a high level of suspicion, followed by a D-dimer testing to decide who needs the more invasive forms of diagnostic tests. The D-dimer is highly sensitive but extremely non-specific test which was supposed to reduce the invasive investigations of thromboembolic disease but its overuse has had a completely opposite effect. From the current evidence we know that some patients are more likely to have a PE than others and, maybe more importantly, the current approach to treatment of small PEs is likely to cause more harm than good to that patient population. In the PERC study Kline et al. estimate the potential harm caused by diagnosing and treating a PE as being 1.8%. This means that if the patient's likelihood of having a PE is less than 1.8%, we are more likely to causing harm by investigating the PE than the PE itself is.
A thorough literature search shows the paucity of evidence supporting treatment of small PEs and suggests a pathway which would reduce unnecessary testing in the low risk patients. Based on the PERC validation study and Wells PE criteria, we combined the 2 rules to minimise the harm to the patients with a low probability of a PE. Supported by the PIOPED II investigations, the pathway was retrospectively tested on over 1 years' worth (125 CTPAs) of patients at Wanganui Hospital in New Zealand.
125 CTPAs (Computer Tomography Pulmonary Angiogram) performed (122 patients) with a total 13.5% incidence rate of PE (slightly lower than the 16.9% in Middlemore). However 71 patients (56%) who had a CTPA fell into a low risk group. Radiologists reported 2 probable and one possible PE. This gave an incidence rate of positive CTPAs of 4.6%. All 3 PEs reported in this group were small and likely to be of no clinical significance. Based on the PIOPED II study, 2 of them were likely to be false positive. This gives a rate of true PEs as 1.4%. This was below the PERC treatment threshold of 1.8% and is likely to have resulted in a net harm to this patient population.
49 of the 71 patients had a D-dimer test performed. Of those 46 were positive, giving this test a sensitivity of 100% (all the +ve CTPAs had a high D-dimer) but specificity of only 6.5% (all 3 low D-dimers did not have a PE).
In our institution, combining the PERC and Well's PE pathways with the appropriate use of D-dimer blood test would result in a 56% reduction of CTPAs and was likely to improve long term patient outcome in our institution. Further studies are strongly indicated to ascertain whether patients with small PEs are likely to benefit from a long term treatment. In the meantime clinically effective pathways can be used and locally validated to minimise the risks caused by the unnecessary investigations of low risk patients. D-dimer results should only be applied in patients with a moderate risk of a PE, based on the predictive pathways as it is only likely to increase the rate of unnecessary investigations. If a D-dimer had been performed it can be safely ignored in low-risk patients.
Long-term effect of a walk-in-centre on ED demand
Drew Richardson MBBS FACEM GradCertHE MD
Australian National University, Canberra, Australia
In May 2010 the ACT opened a nurse-led Walk-in-Centre (WIC) on the campus of the tertiary hospital. Research identified an immediate spike in low acuity ED workload and a medium term referral rate amounting to 1–3% of ED presentations.
This study aimed to identify long term trends in ED workload associated with the WIC.
Prospective descriptive study of ED presentations and ward admissions in the 728 days before and after the WIC opened. Each was analysed as 26 × 28 day periods and regression was used to establish a linear and an exponential model for changes over time.
There was a 12.4% increase in daily presentations from 151.7 [95% CI 149.4–153.9] before the WIC to 170.5 [167.8–173.3] after, with an 18.7% increase in daily admissions from 32.0 [31.4–32.7] to 38.0 [37.2–38.9]. The two models provided a reasonable degree of fit (r > 0.7) and indistinguishable results, giving an annual rise in presentations of 5.0% (linear) and 5.1% (exponential) before the WIC and 5.7% and 5.7% after. The calculated difference between best-fit before and after models at the time the WIC opened was an increase of 2.0 (linear) or 1.8 (exponential) presentations per day (just over 1%).
This model shows opening of the WIC was associated with a small increase in ED workload, consistent with previous studies, and was not associated with any decrease in the underlying growth rate. Rising numbers and acuity present a threat to long term viability of the ED which the WIC has not addressed.
A clinical intervention trial of nurse-initiated analgesia for paediatric patients in the emergency department
David Taylor,1,2 Simone Taylor,1 Kathy Jao,2 Shyan Goh3
1Austin Health, Melbourne, Australia; 2University of Melbourne, Melbourne, Australia; 3Monash University, Melbourne, Australia
We aimed to evaluate the impact and safety of a nurse-initiated analgesia (NIA) intervention for paediatric patients in the emergency department (ED).
We undertook a pre- and post-intervention trial in a large, tertiary referral, mixed ED. The intervention comprised development and implementation of a comprehensive NIA Standing Order. In addition to paracetamol, which nurses could initiate pre-intervention, they could administer non-steroidal anti-inflammatory drugs, opioids and topical local anaesthetics prior to a doctor assessing the patient. All nurses were trained and credentialed prior to the intervention. Patients aged 5–17 years with a triage pain score of ≥4 (Wong-Baker or numerical rating scale) were eligible for enrolment. Parental satisfaction with overall ED pain management was measured using a 1–6 point ordinal scale (1 = very unsatisfied, 6 = very satisfied).
Fifty-one children were enrolled in both the pre- and post-intervention periods. Patient gender and mean age, weight and triage pain score did not differ between the groups (P > 0.05). In the post-intervention period, significantly more patients received NIA (3.0% vs 43.9%, P < 0.001) and the median time to analgesia was significantly reduced (57.6 min vs 23.0 min, P < 0.01). At follow up 48 hours post-discharge, the proportion of parents who were very satisfied with their child's pain management overall increased in the post-intervention period (47.1% vs 66.7%, P < 0.05). No adverse events were observed during either period.
Nurse-initiated analgesia is safe, significantly reduces time to analgesia and is associated with high levels of parental satisfaction.
The optimal technique for removal of upper airway foreign bodies: a repeated-measures, cross-over trial in a porcine model
David Taylor,1,2 John Lippmann,3 Christine F McDonald,1,2 Gary Nolan,1 Ron Slocombe,2 Tony Walker4
1Austin Health, Melbourne, Australia; 2University of Melbourne, Melbourne, Australia; 3Diver Alert Network, Melbourne, Australia; 4Victoria University, Melbourne, Australia
Anterior chest thrusts (with the victim sitting, thrusts applied to the lower sternum towards a supported spine) are recommended by the Australian Resuscitation Council for clearing upper airway obstruction by a foreign body. However, due to a lack of evidence, lateral chest thrusts (with the victim lying on their side) are no longer recommended. We compared anterior and lateral chest thrusts and the Heimlich manoeuvre in the generation of airway pressures.
This was a repeated-measures, cross-over, clinical trial of eight anaesthetised, intubated, adult pigs. For each animal, ten trials of each technique were undertaken with the upper airway obstructed. A chest/abdominal pressure transducer, a pneumotachograph and an intra-oesophageal balloon catheter recorded peak chest/abdominal thrust, peak expiratory airway and pleural pressures, respectively.
The mean (SD) thrust pressures generated for the anterior, lateral and Heimlich techniques were 120.9 (11.0), 135.2 (20.0) and 142.4 (27.3) cmH2O, respectively (P < 0.0001). The mean (SD) peak expiratory airway pressures were 6.5 (3.0), 18.0 (5.5) and 13.8 (6.7) cmH2O, respectively (P < 0.0001). The mean (SD) peak expiratory pleural pressures were 5.4 (2.7), 13.5 (6.2) and 10.3 (8.5) cmH2O, respectively (P < 0.0001). At autopsy, no rib, intra-abdominal or intra-thoracic injury was observed.
The lateral chest thrust and Heimlich techniques generated significantly greater airway and pleural pressures than the currently recommended anterior thrust technique. The findings support the re-introduction of the lateral chest thrust technique for clearing upper airway obstruction by a foreign body, especially as the Heimlich technique has been associated with organ damage.
Prevalence of access block 2012: has NEAT made a difference?
Drew Richardson MBBS FACEM GradCertHE MD
Australian National University, Canberra, Australia
The ACEM funded access block point prevalence study included new questions on a 4-hour performance in 2012.
To describe the prevalence of access block and breaches of 4-hour targets in ACEM accredited EDs in 2012 and compare changes over time.
Email, telephone and fax survey of all accredited Australian EDs at 10:00 on 4 June 2012.
89 of 101 eligible EDs provided usable data, revealing 1826 patients under treatment at 10:00 and a further 577 waiting to be seen. Of those being treated, 607 (33.2%) were waiting for an inpatient bed of whom 419 (69%) were experiencing access block (total ED time >8 hours). However, 10% of those waiting and 47% of those being treated had already exceeded 4 hours since arrival. Comparison of the 69 hospitals which reported in all the last 3 June surveys showed significant improvement in both occupancy with patients (Departmental mean 23.7 patients in 2011, 20.2 in 2012, P = 0.003, paired t-test) and access block (Departmental mean 6.5 patients in 2011, 4.7 in 2012, P = 0.02, paired t-test). This improvement was seen in all States except NSW.
Access block remains a significant issue but has improved steadily since 2008. Weather effects may have reduced workload on the survey day. Efforts to reach the NEAT appear to be contributing to the improvement but it is unlikely that many EDs will achieve the intended threshold of this 4-hour target.