Australasian College for Emergency Medicine 30th Annual Scientific Meeting, 24–28 November 2013, Adelaide, Australia


Oral Program Abstracts

“CARE ED” a better way for emergency departments to care for patients

Thomas Chan MBBS FACEM MHSM AFRACMA,1,2 Suzanne Coles RN MEducation Cert Crit Care,1 Gary Campain MBBS FACEM Grad Cert HSM,1 George Braitberg MBBS FACEM,2,3 Rose Chapman RN PhD MSc,2 Sandra McGuire RN MHA5

1Monash Health, Berwick, Victoria, Australia

1Emergency Department, Casey Hospital, Monash Health, Berwick, Victoria, Australia; 2Emergency Program, Monash Health, Dandenong, Victoria, Australia; 3Monash University, Clayton, Victoria, Australia; 4Australian Catholic University, Fitzroy, Victoria, Australia; 5Casey Hospital, Monash Health, Berwick, Victoria, Australia

This paper will present the three month evaluation of a new model of care implemented within one urban district emergency department in Victoria. The new model involves early, senior-level decision-making as well as incorporating the split flow of presentations between two streams: a vertical (ambulant or chair-based patient) stream, and a horizontal (non-ambulant or trolley-based patient) stream. This innovative model will commence in September 2013. A series of quantitative and qualitative metrics will be used to evaluate the effectiveness of the introduced changes. Through the introduction of a new model of care the department aims to improve access, performance and patient experience, particularly during periods of peak activity. The model may be seen as a pilot for changes that may be introduced in emergency departments both nationally and internationally. Introduction of this new model of care is expected to contribute to achieving the National Emergency Access Targets including all patients being seen and receiving disposition within four hours, meeting all triage category wait times and reducing the percentage of patients who do not wait for treatment.

We're all part of the problem! Using ED data to engage with inpatient units

Simon Craig MBBS (Hons) FACEM MHPE,1,2 Anthony Kambourakis MBBS GradDipHlthServMt MPH FACEM AFRACMA,1 George Braitberg MBBS FACEM FACMT Dip Epi Biostats1,2

1Emergency Department, Monash Medical Centre, Melbourne, Victoria, Australia; 2Southern Clinical School, Faculty of Medicine, Monash University, Clayton, Victoria, Australia

Unnecessary variation in practice can lead to patient harm and inefficiency. The introduction of National Emergency Access Targets, and increased scrutiny on ambulance ramping, has led to considerable focus on efficiency – not only within the emergency department, but in the wider hospital system.

Standardisation of care – where appropriate – leads to improved quality and safety, and increased efficiency. However, defining “best practice” is problematic, and adherence to published guidelines is inconsistent.

This paper describes the use of ED data to engage with specialty units within the hospital. The data was used to highlight common clinical conditions cared for by the ED and each specialty unit, to quantify the relative frequency of these conditions, and to identify priority areas for standardisation of care.

The use of ED data in this way has potential to improve the quality of patient care (both within and beyond the ED), and to improve ED efficiency.

Emergency services – a service systems approach

Moira Scerri MBA,1 Jai Waters EMBA GradCertChange Management MClin Psych BPsycho (Hons),2 Marian Lee MBBS DipCH FACEM MHA3

1Management Discipline Group, University of Technology Sydney, New South Wales; Australia; 2TAFE NSW, Sydney, New South Wales, Australia; 3Emergency Department, Prince of Wales Hospital, Randwick New South Wales, Australia

Background: Australia's population is growing, ageing and living longer. Health care is critical and central to the continued well-being of citizens. Health expenditure as a percentage of GDP continues to rise and the cost of health care operations is under scrutiny. The delivery of healthcare requires operations/processes. Parallels of the operative aspects of emergency services and health care delivery have been drawn from other industries such as manufacturing. Lean manufacturing principles of resource allocation, standardisation of work-practices, minimisation of work duplication, waste and time are being used to analyse the quality and efficiency of the delivery of healthcare.

Objective: The objective of this research was to view healthcare using a service system lens. The premise is that value results from the interactions between people, technology, shared knowledge and connections between internal and external healthcare systems.

Methodology: The methodology adopted was a content-analysis of a television documentary depicting a trauma patient's journey from the pre-hospital phase to discharge from the emergency department.

Findings: The study found that efficiency was achieved through the convergent and divergent mobilisation of resources rather than a linear input – transformation – output model. Aspects of work practices were found to be important but most patient-centred outputs resulted from multi-disciplinary, distributed and collaborative decision making.

Conclusion: These findings have important implications for how healthcare work is organised and delivered. Reviewing healthcare using a service systems approach may prove beneficial in determining how health care services can be reorganised to meet the anticipated increase in demand and should be the focus of future research.

Decrease in access block associated with decreased subsequent inpatient LOS

Drew Richardson BMedSc MBBS (Hons) FACEM GradCertHE MD1,2

1Australian National University Medical School, Canberra, Australian Capital Territory, Australia; 2Emergency Department, The Canberra Hospital, Canberra, Australian Capital Territory, Australia

Background: Previous research has shown an association between access block and subsequent inpatient length of stay (LOS). Demonstration of reversibility would suggest this is causative rather than incidental.

Objective: To compare mean inpatient LOS between periods of different access block.

Methods: Retrospective descriptive study of ED and inpatient data over 6 June-December periods 2007–2012 in a tertiary hospital with a mixed ED. Presentations not transported by interstate ambulances who were admitted from ED first to an inpatient ward (not observation unit) were studied. Access block (AB) was defined as more than eight hours total ED time and LOS was defined as midnights on the inpatient ward with a minimum of one and truncated at ten days. The primary hypothesis was that there would be a correlation between AB and mean LOS.

Results: Presentations and admissions increased over time but AB varied from 43.0% (2011) to 54.6% (2007). In every period mean LOS was longer in AB patients (P < 10E-6). There was a significant correlation between the proportion of AB and overall LOS (r = 0.73, P = 0.05). Subgroup analysis strongly suggested a discontinuity in the series with r = 0.99 if 2012 data were excluded.

Conclusion: This study provides further evidence of a causative relationship between AB and LOS. The effect outside the ED is modest, amounting to 4.5 bed days saved for each 1% reduction in AB. Further research is required on effects of case mix and on possible changes in admission practice in 2012 when NEAT was introduced.

Access block point prevalence study 2013: four-hour target appears out of reach

Drew Richardson BMedSc MBBS (Hons) FACEM GradCertHE MD

Australian National University Medical School, Canberra, Australian Capital Territory, Australia

Background: The ACEM funded Access block point prevalence study included new questions on 4-hour performance in 2012 and 2013.

Objective: To describe the prevalence of access block and breaches of four hour targets in ACEM accredited EDs in 2013 and compare changes over time.

Methods: Email, telephone and fax survey of all accredited Australian EDs at 10:00 on 3 June 2013.

Results: 100 (96%) of 104 eligible EDs provided usable data, revealing 1970 patients under treatment at 10:00 and a further 501 waiting to be seen. Of those being treated, 698 (35%) were waiting for an inpatient bed of whom 450 (65%) were experiencing access block (total ED time > 8 hours). However, 5% of those waiting and 39% of those being treated had already exceeded four hours since arrival. Seventy-three patients in 25 hospitals had a dangerously long ED time of over 24 hours and seven hospitals reported a patient waiting for a bed who had been in ED more than 48 hours. Comparison of the 62 hospitals which reported in all the last five June surveys showed mean occupancy and access block were lower than in 2011, but not significantly lower than 2012. Twenty-four hour stays were at the highest level ever, with a particular increase in WA, Tasmania and the territories.

Conclusion: The steady decrease in Access Block since 2008 appears to have stalled. It is unlikely that any jurisdictions (or even many hospitals) will achieve the intended NEAT 4-hour targets in 2013.

Early experience with an overcapacity protocol

Drew Richardson BMedSc MBBS (Hons) FACEM GradCertHE MD,1,2 Michael Hall MBBS FACEM2

1Australian National University Medical School, Canberra, Australian Capital Territory, Australia; 2Emergency Department, The Canberra Hospital, Canberra, Australian Capital Territory, Australia

Background: Overcapacity protocols (OCP) mandate movement of inpatients from overcrowded EDs to ward spaces and have been shown to reduce overcrowding.

Objective: To describe the effects of introducing a new OCP.

Methods: Prospective descriptive study in a mixed tertiary 70000/year ED with ongoing process changes. Three consecutive five week periods corresponding to control, use of a new chest pain evaluation unit, and intervention were compared to each other and historical controls from the previous four years. The intervention was an OCP triggered 1700–0800 weekdays when ED was overcrowded and had more than 10 patients admitted. On trigger, three-four patients were immediately transferred to registered inpatient beds opened in areas normally closed after hours, using on-call staff. Aggregates for each five weeks were expressed as daily averages of standard ED demand and performance indicators plus calculated occupancy with inpatients.

Results: Presentations averaged 184.0, 183.8 and 187. Nine daily in the three periods, ward admissions 41.7, 44.7, and 44.6, a similar pattern but higher than historical controls. Occupancy with inpatients increased from 5.4 to 8.0 then fell to 6.7 (P < 0.001), did-not-wait changed from 8.6% to 7.7% then 7.8% (P = NS), whereas historical controls all showed significant increase in both. Total time when occupancy with inpatients exceeded 13 fell from 68:23 in the second five weeks to 6:43 and subgroup analysis suggested that this 90% reduction had practical effects on waiting time and flow.

Conclusion: This OCP shows promise as a method for mitigating the effect of peaks of overcrowding with inpatients on the ED.

Permission to pause the pathology procession? How does ED ordering of pathology tests compare with ACEM/ RACP guidelines?

Simon Craig MBBS (Hons) FACEM MHPE,1,2 George Braitberg MBBS FACEM FACMT Dip Epi Biostats,1,2 Rose Chapman RN PhD MSc1,3

1Monash Medical Centre Emergency Department, Monash Health, Clayton, Victoria, Australia; 2Southern Clinical School, Faculty of Medicine, Monash University, Clayton, Victoria, Australia; 3Australian Catholic University, Fitzroy, Victoria, Australia

Several studies have estimated that 25–40% of all tests sent to the pathology laboratory are unnecessary. This may be due to inexperienced staff, poor understanding of test characteristics, or a perceived need for “completeness”. Recently, ACEM and the RCPA (Royal College of Pathologists of Australasia) released a guideline on pathology testing in the emergency department. To date, there are no papers comparing current practice to these guidelines.

This paper will present the findings of a three-month retrospective audit of all ED electronic medical records of pathology testing for a number of commonly encountered conditions from three Victorian hospital EDs in south east Melbourne (one tertiary ED, and two urban district EDs).

The conditions reviewed included Chest pain (? STEMI and ? Pulmonary embolism); Chest infection/ pneumonia; COPD exacerbation; gastrointestinal bleeding; bleeding in early pregnancy; stroke/TIA; syncope; fractured neck of femur; renal colic; and deliberate self-poisoning. The results of the study will be used to develop a quality improvement programme to address unnecessary/inappropriate practice variance and lead to improved quality, safety and efficiency.

Use of an accelerated diagnostic protocol in the assessment of emergency department patients with possible acute coronary syndrome

Terry George MBBS Dip RACOG FACEM,1 Sarah Ashover BScApp (HMS) BHlthSc (Pod)(Hons),2 Louise Cullen MBBS (Hons) FACEM,3,4 Peter Larsen MBBS (Hons) FRACP FCSANZ,5,6 Jason Gibson BSc DMU,7 Jennifer Bilesky BScApp (HMS),2 Steven Coverdale MBChB FRACP FCSANZ,8 William Parsonage BMedSci (Hons) BM BS DM MRCP FRACP9,10

1Department of Emergency Medicine, Nambour General Hospital, Nambour, Queensland, Australia; 2ACRE Project, Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia; 3Department of Emergency Medicine, Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia; 4Faculty of Public Health, Queensland University of Technology, Brisbane, Queensland, Australia; 5Sunshine Coast Hospital and Health Service, Nambour General Hospital, Nambour, Queensland, Australia; 6School of Medicine, University of Queensland, Brisbane, Queensland, Australia; 7Department of Cardiology, Nambour General Hospital, Nambour, Queensland, Australia; 8Sunshine Coast Clinical School, School of Medicine, University of Queensland, Brisbane, Queensland, Australia; 9Department of Cardiology, Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia; 10Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia

Background: Patients with suspected acute coronary syndrome make up 10% of emergency department (ED) presentations. Eighty-five percent are diagnosed as non-cardiac following extended ED stay (>6–8 hours) or admission to hospital. Standard clinical assessment is problematic for achieving National Emergency Access Targets (NEAT).

Aim: To introduce an Accelerated Diagnostic Protocol (ADP) for a significant proportion of patients with undifferentiated chest pain and achieve NEAT targets in a manner acceptable to patients and staff.

Methods: A Short Low Intermediate Chest pain (SLIC) risk pathway using the ADP incorporating the Thrombolysis in Myocardial Infarction (TIMI) risk score and ECG to identify patients suitable for early discharge following negative serial sensitive cardiac troponins by two hours after presentation with planned return for outpatient exercise stress test (EST).

Results: 1762 ED presentations for chest pain between 1 July 2012 and 31 January 2013 were included. 12.4% (219) were identified for SLIC. From August, SLIC patients exceeded the NEAT benchmark. 90.9% (169/186) of SLIC patients to 31 December 2012 were followed up at 30 days. 8.3% (14/169) had positive/equivocal EST and underwent further investigation. 17.8% (30/169) did not return for EST. There were no major adverse cardiac events. Only one patient was diagnosed with coronary artery disease. 6.5% (11/169) represented to a hospital ED with chest pain within 30 days. Patient satisfaction rating was very high.

Conclusion: The introduction of this ADP for low risk chest pain has been successful in this pilot ED site.

Role of routine chest X-ray in suspected acute coronary syndrome in the emergency department

Pam Rosengarten MBBS FACEM,1,2 Eva Hagop, BSci (Hons) PhD1

1Emergency Department, Peninsula Health, Victoria, Australia; 2Monash University, Clayton, Victoria, Australia

Background: Suspected acute coronary syndrome (ACS) is a common presentation in the emergency department (ED). Assessment and diagnostic workup is undertaken for risk stratification. Chest radiography is frequently performed despite not being included in risk stratification guidelines.

Objective: The primary aim of this study was to determine the prevalence and nature of chest x-ray (CXR) abnormalities in Frankston Hospital ED patients with suspected ACS, and to determine whether CXR abnormalities resulted in alterations in patient management.

Methods: A retrospective study of patient's ≥18 years with suspected ACS (as defined by ACS workup of 12-lead ECG, cardiac monitoring and troponin assessment) from April to June 2012. Patients were excluded if they had symptoms consistent with other cardio/pulmonary pathology, trauma or vital signs outside of normal ranges on examination.

Results: 1063 patients were identified presenting to the ED with suspected ACS, of whom 265 met study criteria. Of these, 246 (93.1%) had a CXR performed in which 152 (61.5%) radiographs were interpreted as “normal” and 94 (38.1%) as “minor abnormality not requiring further action”. Only one (0.4%) patient had a radiographic abnormality requiring further investigation, which was unrelated to the patient's presentation.

Conclusion: In this patient group routine CXR did not reveal any abnormalities that resulted in a change of management. This represents a potential 23% reduction in CXR in this patient population, with opportunity to reduce the impact on cost, workload and patient investigations.

Some patients can be safely discharged after a single troponin on arrival

Yusuf Nagree MBBS MBA(E) FACEM,1,2,3,4,5 Stephen MacDonald BSc MBChB MRCP DCH FACEM1,3,5,6

1Fremantle Hospital, Fremantle, Western Australia, Australia; 2Armadale Hospital, Armadale, Western Australia, Australia; 3Centre for Clinical Research in Emergency Medicine, Harry Perkins Institute of Medical Research, Perth, Western Australia, Australia; 4Emergency Medicine, Fremantle Hospital, Western Australia, Australia; 5Emergency Medicine, University of Western Australia, Perth, Western Australia, Australia; 6Emergency Medicine, Armadale Health Service, Armadale, Western Australia, Australia

Background: Chest pain is a significant burden on emergency departments (ED). Acute coronary syndromes (ACS) remain a diagnostic challenge with many pathways recommending serial biomarkers and ECGS over hours.

Aim: We aimed to determine if a single troponin drawn on patient arrival could safely identify a cohort of patients suitable for immediate discharge.

Methods: We used data from a multi-centre prospective observational study in five Western Australian hospitals to validate a rapid discharge algorithm. Patients were recruited if they were undergoing serial troponin evaluation for suspected ACS. Exclusions were inability to consent, pregnancy, age < 18 years or inability to follow-up at 30 days. Bloods were drawn on arrival and again 8–12 hours later. The thrombolysis in myocardial infarction (TIMI) score was calculated. A patient was regarded as dischargeable immediately if they had a TIMI score of 0, normal ECG and normal initial troponin. The primary outcome was the composite of mortality, acute myocardial infarction or emergency revascularisation within 30 days.

Results: A total of 663 patients were recruited with four lost to follow-up and 76 patients (11.5%, 95% CI 9.1%-14.0%) meeting an outcome. Using our method, 127 patients could have been discharged immediately with one patient meeting a 30-day outcome (subsequently had a normal coronary angiogram), leading to a sensitivity of 98.7% (95%CI 92.9%-100%) and a negative likelihood ratio of 0.06 [95%CI 0.01–0.43].

Conclusion: A TIMI score of 0, negative initial troponin and normal ECG identifies a cohort of patients who can be safely discharged from the ED rapidly.

Post abortion care in South Australian emergency departments

Su-Min Wong MBBS, BMedSc,1 Ea Mulligan FRACMA FRACGP FACHSE MBBS BMedSc (Hons) MHealth Admin,2 Steve Quinn B Math M Math PhD3

1Department of Obstetrics and Gynaecology, Women's and Children's Hospital, Adelaide, South Australia; 2Department of Obstetrics and Gynaecology, Flinders University, Adelaide, South Australia; 3Flinders Clinical Effectiveness, Flinders University, Adelaide, South Australia

Objective: To describe presentations to emergency departments following pregnancy termination in South Australian public hospitals

Methods: 1173 first trimester medical or surgical pregnancy terminations conducted in three public hospital clinics were analysed. The primary outcome measure was presentation to an ED within 30 days of the procedure. Secondary outcomes were ED admissions and subsequent D&C surgery. Presenting complaints, ED discharge diagnosis, admission diagnosis and subsequent surgical procedures performed were collected by interrogation of OACIS clinical data repository using the “single patient lookup” function.

Results: 64 (5.5%) women presented to public hospital emergency departments following 1173 pregnancy terminations. Complaints consisted of: bleeding and/or pain (44) nausea/vomiting (5), offensive discharge (4), fever (3) and anxiety (1) referral by a health provider (3) and no complaint recorded (3). No pathology other than the presenting symptoms was recorded at discharge following 37/64 (57.8%) of the encounters. In 27/64 (42.2%) underlying pathology was identified; sepsis/endometritis (11), retained products of conception or inevitable abortion (10), threatened miscarriage(4), medication side effect(1) and one continuing pregnancy. There were 6/64 (25.0%) presentations for treatment of; retained products of conception (8), sepsis (4) or continuing pregnancy (1), pain management (3).

Conclusion: 64/1173 (5.5%) women attended EDs following abortions with only 16/1173 (1.4%) requiring admission. Most commonly women presented requiring reassurance that no sinister pathology had been identified. The most common cause of admission was for surgical management of retained products of conception (8), but this could potentially be managed with outpatient misoprostol in the future.

A governance structure to implement clinician sonology

Dimitri Giannios MBBS FACEM Grad Dip Hlth Econo & Policy CCPU,1,2 Carolynne Cormack BAppSc (Med Rad) GDipUlt AMS,1,3 Ronnie Ptasznik MBBS FRANZCR,3,4 George Braitberg MBBS Dip Clin Epi Biostats FACEM FACMT Diplomate of the American Boards of Emergency Medicine and Medical Toxicology,1,5 Robert Meek MBBS FACEM MClin Epi1,5

1Monash Emergency, Monash Health, Dandenong, Victoria, Australia; 2Grattan Institute, Carlton, Victoria, Australia; 3Diagnostic Imaging Department, Monash Health, Clayton, Victoria, Australia; 4Department of Medical Imaging and Radiation Sciences, Faculty of Medicine, Nursing and Health Sciences, Monash University, Clayton, Victoria, Australia; 5Southern Clinical School, Monash University, Clayton, Victoria, Australia

Objective/Background: For over 20 years ED clinician trauma and AAA sonography has been accepted practice, however processes for credentialing and support of emergency physician ultrasound aren't standardised in many health services. In 2013 Monash Health implemented a governance structure in order to increase emergency physician uptake of sonography and ensure that the practice met high and reproducible sonographic standards.

Methods: A hospital executive sponsored oversight committee with representatives from Diagnostic Imaging and Emergency Medicine was formed to govern the education and quality activities of the programme. The committee was tasked with ensuring compliance in programme induction, training in image acquisition and interpretation, result archiving and recording, competency evaluation, and clinical quality outcomes. Instrumental to the success of the programme was a dedicated sonographer for ongoing “hands-on” training, image review and result correlation with other modalities.

Results: The established governance framework for emergency physician ultrasound is recognised and supported by the Health Service and incorporated within the emergency physician's scope of practice. Currently 93 clinicians have performed over 1300 studies meeting verifiable quality framework standards.

Conclusion: A formal governance structure with interdisciplinary stakeholder membership is a useful institution to provide a clear pathway to a new skill implementation programme. Future directions with other disciplinary integration (e.g. Intensive Care Medicine), development in echocardiography/IVC and RUQ scanning, research using the large data base created and development of a sabbatical programme with the continued governance structure are seen as the next step.

E. coli sepsis in the ED; high mortality, high morbidity and increasing resistance. A contemporary clinical case series and exploration of an important emerging Australian pathogen

Nicole Ghedina MBBS MPHTM,1 Daniel Fatovich MBBS FACEM PhD,1,2 Lucy Fitton MBBS2

1Emergency Medicine Department, Royal Perth Hospital, Perth, Western Australia, Australia; 2University of Western Australia, Perth, Western Australia, Australia

Background: E. Coli sepsis incidence is increasing in Australia and worldwide and is associated with a very high case fatality and morbidity load. The recognition and management of sepsis in the ED is well established as an area of practice in which early recognition, appropriate treatment and correct antibiotic selection has real impact on patient outcomes.

Methods: We present a case series of 89 adult patients with E. coli bacteraemia on blood cultures drawn in the ED, and compare these with the E. coli bacteraemic patient's hospital wide and with previously published international case series.

Results: In our cohort 47% were female and mean age at admission was 70.5 years. Mean length of stay 11 days (range 0 to 107, median six). 11% of patients died within 30 days of admission. Forty-eight percent of patients were tachycardic at ED presentation. Shock index of >0.7 was present in 53% whilst hypotension was only present in eight percent at presentation. Sixty-six percent of patients met SIRS criteria at presentation to the ED. The most common source of infection was urinary 52%, with primary septicaemia 26%, biliary 17%, non-biliary GIT 3.4%, and device related infections 2%. We present further data regarding comorbities, demographics, initial antibiotic selection and resistance rates and clinical outcomes including need for ICU care (septic shock, dialysis, mechanical ventilation) and at discharge (increased care requirements representation rates).

There is minimal ED based data about E. coli sepsis in the current literature and this study presents important clinically relevant data for the ED clinician.

We review the emerging use of real time PCR on blood to detect bacterial pathogens. E. Coli assays have shown poor sensitivity and specificity in other studies. We present the result of a 16S PCR DNA analysis on a subset of our patients that failed to demonstrate correlation with clinical severity of illness.

Risk factors for patients who represent to Emergency and subsequently require admission: can they be identified?

Allison M Moore BMedSc (Hons) MBBS FACEM, Elaine Harvey Quality Support Officer RN BNurs Grad Cert Critical Care

Emergency Department, Sutherland Hospital, Caringbah, New South Wales, Australia

Aim: To determine which factors identify patients at risk of representation and subsequent admission.

Methods: A retrospective chart review was carried out on all patients who represented within a month to our urban district referral ED and required admission, over a 13 month period. The electronic record was examined for demographic information and triage presenting condition. “Red flags” at initial presentation were noted.

Results: There were 52508 presentations and 1016 representations to the ED in this period. 197 patients subsequently required admission. Admitted representers had demographics not significantly different to all presenters, with the exception of a significant decrease in admitted representations in the 2–16 y age group (16.3% vs 10.7%, p < 0.05).

The top 10 triage presenting conditions for admission on representation were: Shortness of breath, pain (any), chest pain, abdominal pain, flank pain, diarrhoea and/or vomiting, fever, headaches, collapse or syncope and unwell. These accounted for 79.0% of admitted representations, but less than half of all presentations and all representations (49.4% and 50.9%, p < 0.05). Flank pain was eight times more frequent in admitted representations than in initial presentations. Shortness of breath, diarrhoea or vomiting, headache and fever were between two and three times more frequent in admitted representations. “Red flags”, including vital signs or blood test abnormalities, could be identified on initial presentation for 51% of representers requiring admission.

Conclusion: These findings suggest that representers requiring admission present with conditions and clinical “red flags” which could be used to identify them as more likely to require admission at their first presentation.

Determining patient capacity to consent in the emergency department

Susan Hertzberg MBBS MHL FACEM

Emergency Department, Prince of Wales Hospital, Sydney, New South Wales, Australia

It is well recognised in both common and statutory law that adult patients have the right to determine what treatment they receive provided they are deemed to have capacity. Emergency physicians are frequently required to assess patient capacity to consent to, or refuse medical treatment. Despite there being various tools and guidelines available to aid health professionals to determine whether patients are competent to decide for themselves, there are no standardised tests that inform the assessment of capacity.

Capacity or competence is defined both in common law and statute. In Australia, the laws which govern capacity and consent differ between states and emergency physicians should be acquainted with the relevant legislation in the state in which they work. The common law has upheld the principle that all adult patients are presumed to have capacity unless that presumption can be rebutted. Minors are deemed to lack capacity but the law recognises that minors may have capacity to consent to treatment depending on the nature of the treatment and their ability to understand. In contrast to adult patients, however, the law does not recognise the right of a minor to refuse life saving or life sustaining treatment.

This talk will outline the common law test for capacity as well as the statutory definition of capacity as which is found in the Guardianship Act of the states and territories. I will describe recent Supreme Court judgments which have upheld the right of adult patients to refuse life sustaining treatments and I will address the vexed issue of the right of a minor to consent to, or refuse medical treatment.

Management of children with possible appendicitis: a survey of emergency physicians

Kieran McCabe MB BCh MBAO MRCPI,1 Franz E Babl MD MPH FRACP,2 Sarah Dalton BMed MApptMgt (Hlth) FRACP1 on behalf of the Paediatric Research in Emergency Departments International Collaborative (PREDICT)

1Emergency Department, The Children's Hospital at Westmead, Sydney, New South Wales, Australia; 2Emergency Department, The Royal Children's Hospital, Murdoch Children's Research Institute and University of Melbourne, Victoria, Australia

Objective: To describe current practice in managing children with possible acute appendicitis in paediatric emergency departments in Australia and New Zealand.

Methods: An anonymous survey was distributed to consultant emergency physicians working at PREDICT (Paediatric Research in Emergency Departments International Collaborative) sites in Australia and New Zealand in April 2013. A second, site based survey, was distributed to the PREDICT representatives at each of the 13 sites.

Results: The response rate was 100% (13/13) for the site based survey and 82% (145/176) for the physician survey. Most respondents agreed that right lower quadrant (RLQ) tenderness (94%), anorexia (92%), migration of abdominal pain to the RLQ (86%), rebound tenderness in the RLQ (60%) and RLQ tenderness produced on coughing, hopping or percussion (63%) were valuable symptoms and signs in diagnosing appendicitis. The responses regarding nausea and vomiting, measured body temperature and blood investigations were less consistent. Most physicians (61%) agreed there was a role in their department for a validated clinical practice guideline for children with possible appendicitis although only 3/13 sites reported use of a guideline for such patients.

Conclusion: This survey of experienced emergency physicians across Australia and New Zealand has demonstrated some inconsistencies in the approach to diagnosing children with possible appendicitis. The survey has also identified a desire among emergency physicians for a validated clinical practice guideline for the management of these patients. Such a guideline may improve the efficiency with which available resources are used in the investigation and management of children with possible appendicitis.

The use of guided imagery to alleviate pain in paediatric patients (4–12 year olds) undergoing treatment in the emergency department

Kristine Gilbert MBBS,1 Robyn Parker MBBS FACEM,1 Joseph Ciorciari,2 Michelle Beswick,2 Colin Carbis3

1The Northern Hospital, Epping, Victoria, Australia; 2Brain and Psychological Research Centre, Hawthorn, Victoria, Australia; 3Virtual Medicine, Preston, Victoria, Australia

Background and Aim: Pain is a significant problem for children attending the emergency department (ED). For many children anxiety forms a large component of their perceived pain. In a randomised, controlled trial we investigated whether the use of a guided imagery programme – a virtual reality, age-appropriate relaxation programme – plus standard care (analgesics and distraction techniques) would decrease pain compared to standard care alone.

Methods: Forty six children (4–12 years) who presented to the ED with an isolated limb injury were randomly assigned to either the treatment group (guided imagery plus standard care) or control group (standard care). At three points in their presentation – t = 0 mins, t = 30 mins, t = 90 mins – the subjects rated their pain on a Wong-Baker Chart. Our clinical end point was to decrease the mean pain score of >10 mm when scores were converted to a 100 mm visual analogue scale. A statistical end point was calculated using ANOVA.

Results: There was a reduction in mean pain scores of 16.3 mm at t = 30 mins and 11.4 mm at t = 90 mins in the treatment group compared to the control group. Using a three by two mixed design analysis of variance there was a statistically significant reduction in the treatment group's mean pain score over time F(2, 88) = 14.02, p < .001, partial ή2 = 0.24.

Conclusion: By targeting and treating the anxiety component of a child's pain we can decrease their overall pain. Therefore the use of a virtual reality, guided imagery programme may become a further tool in the management of paediatric pain in the ED.

Admitted patients with bronchiolitis at seven Australian and New Zealand centres: retrospective analysis of medication use

Trusha Brys MBBS FRACP,1,2 Ed Oakley MBBS FACEM,1,2,3 Franz E Babl MD MPH FRACP,1,2,3 David Krieser MBBS FRACP,5 Mary Boyle MBBS (trainee),1 Amanda Fry RN,1,2 Kim Jachno MSc,2,4 Vimuthi Chong BA MB BCh BAO MRCPI MRCPCH DTM,1,2 Jocelyn Neutze MBChB FRACP FACEM,6,7 Stuart R Dalziel MBChB FRACP PhD,8 Meredith Borland MBBS FACEM,9,10 Jason Acworth MBBS (Hons) FRACP11,12,13

1Emergency Department, Royal Children's Hospital, Melbourne, Victoria, Australia; 2Murdoch Children's Research Institute, Royal Children's Hospital, Parkville, Victoria, Australia; 3The University of Melbourne, Melbourne, Victoria, Australia; 4Clinical Epidemiology and Biostatistics Unit, Royal Children's Hospital, Melbourne, Victoria, Australia; 5Sunshine Hospital, Melbourne, Victoria, Australia; 6Emergency Department, KidzFirst Middlemore Hospital, Auckland, New Zealand; 7University of Auckland, Auckland, New Zealand; 8Starship Children's Hospital, Auckland, New Zealand; 9Emergency Department Princess Margaret Hospital, Perth, WA, Australia; 10Joint Schools of Paediatric and Child Health & Primary, Rural and Aboriginal Health Care, University of Western Australia, Perth, WA, Australia; 11Emergency Department, Royal Children's Hospital, Brisbane, Queensland, Australia; 12Queensland Children's Medical Research Institute, Royal Children's Hospital, Herston, Queensland, Australia; 13University of Queensland, St Lucia, Queensland, Australia

Background: There are no effective medications to improve the outcome of infants with bronchiolitis. Studies have shown the management of bronchiolitis to be varied even within the same region.

Aim: To describe medication use at the seven study hospitals for a recent multi-centre randomised controlled trial on hydration in bronchiolitis (Comparative Rehydration in Bronchiolitis (CRIB).

Methods: A retrospective analysis of extant data of infants two months of age (corrected for prematurity) and 12 months of age admitted with bronchiolitis at seven hospitals in Australia and New Zealand identified through the CRIB trial. CRIB study records, pathology and radiology databases were used to collect data using a standardised form and entered on a database at a single site. Medications investigated included salbutamol, adrenaline, steroids, ipratropium bromide, normal saline and hypertonic saline, steroids and antibiotics.

Results: There were 3456 patients admitted with bronchiolitis available for analysis, of which 42% received medication. Medication use in general varied by site between 27% and 48.7%. The most frequently used medications were salbutamol (25.5%). Medication use and salbutamol use increased by 8.2% and 9.3% respectively per month of age after four months. In patients admitted to ICU 81.6% received medication and 39.5% not admitted to ICU received medication.

Conclusion: Medication is being used frequently and variably in this region. Medication use and salbutamol use increased with age. Patients who were likely to be more unwell with bronchiolitis, that is those admitted to ICU or had investigations, also received medication and antibiotics more frequently.

Retrospective multicentre study of bronchiolitis in the intensive care unit (ICU) setting

Vimuth Chong BCh BAO MRCPI MRCPCH DTM,1 Ed Oakley MBBS FACEM,1 David Krieser MBBS FRACP,2 Franz Babl MD MPH FRACP,1 Amanda Fry RN,3 Kim Jackno MSc,3 Mary Boyle MBBS,1 Meredith Borland FACEM,4 Stuart Dalziel MBChB FRACP PhD,5 Jocelyn Neutze MBChB FRACP FACEM,6 Natalie Phillips MBBS FRACP7

1Royal Children's Hospital, Parkville, Victoria, Australia; 2Sunshine Hospital, Sunshine, Victoria, Australia; 3Murdoch Children's Research Institute, Royal Children's Hospital, Parkville, Australia; 4Princess Margaret Hospital, Perth, Western Australia; 5Starship Children's Hospital, Auckland, New Zealand; 6Kidz First Hospital, Auckland, New Zealand; 7Royal Children's Hospital, Brisbane, Queensland, Australia

Objective: To describe the rate of Intensive Care Unit (ICU) admission, type of ventilation support provided and risk factors for ICU admission in infants admitted to hospital with bronchiolitis.

Methods: Retrospective review of hospital records and Australia & New Zealand Paediatric Intensive Care (ANZPIC) registry data collected from infants two to 12 months old admitted with bronchiolitis in seven Australian and New Zealand hospitals, prospectively identified through the comparative rehydration in bronchiolitis (CRIB) study between 2009–2011.

Results: Of 3884 infants identified, there were 3589 charts available for analysis. Of 204 (5.7%) infants with bronchiolitis admitted to ICU, 162 (79.4%) received ventilation support. One hundred and thirty-three (82.1%) received non-invasive ventilation (High Flow Nasal Canulae [HFNC] or Continuous Positive Airway Pressure [CPAP]), seven (4.3%) received invasive ventilation (endotracheal intubation) and 21(13.0%) received a combination of ventilation modes. Infants with co morbidities such as chronic lung disease [OR 1.6 (95% CI 1.0–2.6)], congenital heart disease [OR 2.3 (95% CI 1.5–3.5)], neurological disease [OR 2.2 (95%CI 1.2–4.1)] or prematurity [OR 1.5 (95% CI 1.0–2.1] were more likely to be admitted to ICU. Respiratory Syncitivial Virus (RSV) positivity did not increase the likelihood of being admitted to ICU [OR 1.1 (95% CI 0.8 – 1.4)]. HFNC use increased from 13 patient episodes in 1005 patients in 2009 to 39 patient episodes in 1172 patients in 2011.

Conclusion: Admission to ICU is an uncommon occurrence in bronchiolitis, but more common in infants with co morbidities and prematurity. The majority require non-invasive ventilation, with increasing use of HFNC.

Use of sedation in children requiring a CT scan after head injury

Tammy Goldwasser MBBS,1 Silvia Bressan MD,1,3 Ed Oakley MBBS FACEM,1,2,3 Franz Babl MD MPH FRACP1,2,3

1Emergency Department, Royal Children's Hospital Melbourne, Victoria, Australia; 2Department of Paediatrics, University of Melbourne, Victoria, Australia; 3Emergency Research Group, Murdoch Children's Research Institute, Melbourne, Victoria, Australia

Background: Using head CT scans (CTB) to detect intracranial injuries in children carries risks related to radiation and sedation in uncooperative patients. This study explores sedation requirements and determinants in children having a CTB.

Objective: To determine the sedation rate and correlation with age and GCS for trauma related CTBs in children.

Methods: Retrospective review of charts of children undergoing a CTB for head injury at a tertiary paediatric ED from April 2011 to April 2013 identified through the database of an ongoing prospective study on validation of clinical prediction rules for children with head injuries.

Results: Four hundred and seventy-six patients underwent CTB after head injury. Thirty-three were intubated for medical management and two received midazolam for seizure management. Of the remaining 441, 67.3% were male and mean age was 8.4 years. 30.8% had abnormal CTs and 2.0% required neurosurgery. Twenty-eight (6.3%) were sedated for CTB (including 10 who initially failed CTB without sedation). Two patients were intubated for CTB. Sedation rate was 3.7% in patients with initial GCS ≤ 12, 6.4% in children with GCS > 12 (OR 0.56 95% CI 0.13 – 3.70; p = 0.5). Sedation rate was 17.0% in patients ≤ 4 years, 1.6% in children > 4 years of age (OR 12.21 95% CI 4.37 – 41.86; p < 0.001).

Conclusion: Children with head injury who require CTB rarely require sedation. Younger children, irrespective of GCS are more likely to require sedation. These data will be useful in assessing CTB associated risks and the allocation of resources for CTB in children.

Computed tomography for head Injuries in children: change in usage rates and radiation dose over time

Rachel May MBBS (Hons) BMedSc,1,2,3 David Krieser MBBS FRACP,3,4 Franz E Babl MD MPH FRACP,1,2,3 Ed Oakley MBBS FACEM1,2,4

1Emergency Department, Royal Children's Hospital, Melbourne, Victoria, Australia; 2Murdoch Children's Research Institute, Royal Children's Hospital, Parkville, Victoria, Australia; 3Emergency Department, Sunshine Hospital, St Albans, Victoria, Australia; 4The University of Melbourne, Melbourne, Victoria, Australia

Introduction: Head injury is a common presentation to the emergency department (ED). North American studies demonstrate increasing use of cranial computed tomography (CTB) to investigate head injury. No such data exists for Australian EDs.

Aim: To describe CTB use and radiation dose in CTB over time in two Australian EDs.

Methods: Retrospective medical record and radiology database audit by ICD 10 code for head injury in children < 16 years at a suburban general hospital (SG) and a tertiary referral children's hospital (TR), in Melbourne, Victoria. Data were obtained from SG from 2008–2012, and TR from 2004–2012 including diagnosis, CTB use, CTB result and radiation dose in millisievert (mSv). Descriptive and comparative analysis of CTB rates and radiation dose delivered.

Results: 8965 children with head injury were identified (7613 [85%] at TR and 1335 [15%] at SG). At TR and SG, 15.3% and 7.3% of children received a CTB, of which 2.9% and 1.4% had significant injury defined by PECARN criteria respectively. CTB use did not increase over time. Mean age of patients with CTB was 7.6 years. CTB was obtained more frequently in older children: <2 yo (7.1%), 2–5 yo (10.8%), 6–10 yo (14.6%), >10 yo (19.1%). Mean radiation dose at TR decreased from 55.0 mSV to 25.0 mSV, and remained steady at SG at 13.3 mSv.

Conclusion: Compared with North America, CTB use in head injuries was less and did not increase during the study period. Radiation doses decreased over time.

The utility of ultrasound in the assessment of children with suspected appendicitis

Gemma Milne MBChB (trainee), Simon Binks BMedSci BM BS MRCPCH (UK) FCEM (UK) FACEM

Wollongong Emergency Department, Wollongong, New South Wales, Australia

Introduction: Acute appendicitis is the most common indication for emergency abdominal surgery in children. The key to a successful outcome has always been early diagnosis and urgent appendisectomy before gangrene or perforation develops. Nevertheless, confirming the diagnosis of appendicitis in children and avoiding the risks of unnecessary surgery or excessive exposure to ionising radiation with abdominal CT continue to be a challenge. In published research, ultrasound has been shown to improve diagnostic accuracy in selected patients. Quoted sensitivities have been variable, ranging from 74 to 100 percent with specificities ranging from 88 to 99 percent [1]. This is not our experience in a busy regional emergency department. We believe that ultrasound is over used without careful pre-test selection of patients, is often inconclusive, needlessly delaying admission and more importantly surgical intervention.

Methods: We shall present a retrospective observational study (audit) of the use of ultrasound in the assessment of acute abdominal pain in children presenting to our department; highlighting time frames, diagnostic accuracy and change in clinical disposition. Furthermore, we shall present a derived protocol for the use of ultrasound in the assessment of suspected appendicitis in children.

Reference

1. Taylor GA, Wesson DE. Acute appendicitis in children: Diagnostic imaging. UpToDate® [updated May 23 2013, cited 25 July 2013]. Available from: www.uptodate.com

Reported ingested dose of paracetamol as a predictor of risk following paracetamol overdose

Yit Hung Leang MBBS,1,2 David Taylor MBBS MD MPH DRCOG FACEM,1 Paul Dargan MBBS FRCPE FACMT FRCP FAACT,3,4 David Wood MD FRCP FACMT FBPharacoIS,3,4 Shaun Greene MBCAHb msC (Medical Toxicology) FACEM1,2,3

1Austin Health, Austin Hospital, Heidelberg, Victoria, Australia; 2Victorian Poisons Information Centre, Melbourne, Australia; 3Guys and St Thomas' NHS Foundation Trust and King's Health Partners, London, United Kingdom; 4King's College London, London, United Kingdom

Objective: To evaluate the reported ingested dose of paracetamol as a risk assessment tool in paracetamol overdose.

Methods: Data was retrospectively obtained from a clinical toxicology electronic database linked to two United Kingdom and one Australian hospital. Plasma paracetamol concentrations (PPCs) of adult patients presenting with non-staggered paracetamol overdose between 2006 and 2012 were back extrapolated to a four-hour post ingestion PPC. Reported paracetamol dose ingested (>10 g was considered a potential hepatotoxic dose), was plotted against four-hour PPC (C4) (>150 mg/L (1000 μmol/L) was considered an indication for antidotal treatment) to determine the accuracy of reported dose ingested as an indicator for the need for antidotal treatment.

Results: Of 1246 patients, 88% were from the UK. 65.7% were female. 39% of those reporting ingestion >10 g had an extrapolated four-hour PPC (C4) above 150 mg/L and 95% of those reporting ingestion <10 g had a PPC (C4) less than 150 mg/L. (NPV 95%, PPV 39%, sensitivity 89%, specificity 60%). The data were similar when analysed singularly between populations. There was no improvement in both negative and positive predictive values despite redefining the potentially hepato-toxic dose to 8 g and 16 g. There was no statistically significant difference in test performance when early and late presenters (<8 hours and >8 hours presenters) were compared. Overall, there was a moderate correlation (r = 0.58) between reported paracetamol ingested and extrapolated four-hour paracetamol concentration (C4).

Conclusion: Reported ingested dose of paracetamol is a poor risk assessment tool in paracetamol overdose within these patient populations.

Accuracy of bedside ultrasound in confirming endotracheal placement: an interim report of a validation study

Justin Bowra MBBS FACEM CCPU,1 Maggie Chung MBBCh trainee CCPU,2 Mark Gillett MBBS FACEM Director of Emergency Medicine Research at RNSH,1 Toby Fogg MBBS FACEM1

1Emergency Department, Royal North Shore Hospital, St Leonards, New South Wales, Australia; 2Emergency Department, Prince of Wales Hospital, Randwick, New South Wales, Australia

Objective: To determine the accuracy of bedside lung ultrasound (LUS) in assessing endotracheal tube (ETT) placement using capnography and chest x-ray (CXR) as gold standards.

Methods: A prospective observational study of intubated patients in the ED, using LUS assessment of ‘lung sliding’ as the primary assessment tool for tracheal intubation and CXR/capnography as a combined gold standard. Target patient number is 120.

Results: This is an interim report, current patients enrolled is 61. No oesophageal or endobronchial intubations have been performed to date. Sensitivity of LUS for airway placement of the ETT (rather than in the oesophagus) was 100% (95% CI: 94.07 % to 100.00 %). Sensitivity of LUS in confirming tracheal placement of the ETT (rather than in the bronchus) was 96.7% (95% CI: 88.63 % to 99.51 %) using CXR as gold standard. On two occasions, the operator incorrectly diagnosed endobronchial intubation with LUS.

Conclusion: In clinical settings where CXR and capnography are unavailable or delayed (e.g. pre-hospital, rural/remote and retrieval) LUS may prove an accurate alternative in the determination of ETT placement.

Does medical student simulation training improve time to CPR and defibrillation? A randomised trial

Nathan Dryburgh MBBS BBiot,1 C Jackson,1 Gary Rogers,2 N Jones de Rooy,2 Gerben Keijzers MBBS FACEM MSC Epidemiology1,2

1Emergency Department, Gold Coast Hospital, Queensland, Australia; 2School of Medicine, Griffith University, Nathan, Queensland, Australia

Introduction/Objective: Clinical Learning though Extended Immersion in Medical Simulation (CLEIMS) methodology is a highly resource-intensive teaching approach used at the local University that involves high-fidelity simulations. Areas of interest in the CLEIMS include advanced life support (ALS). Time to defibrillation in cardiac arrest with a shockable rhythm is one of the few interventions that are known to improve survival. We aimed to assess and quantify components of resuscitation, comparing two teaching strategies for medical students.

Methods: A cohort of 80 year three post-graduate medical students were invited to participate in the randomised trial. Forty-four participants were randomised to the ‘intervention arm’ receiving full CLEIMS methodology while 36 participants were randomised to the ‘control arm’ receiving the programme of practical workshops and didactic teaching sessions without the simulation training. At the completion of the teaching week, both arms undertook an unexpected group simulation (14 CLEIMS groups and 12 control groups of 2–4 students) resuscitation scenario on an unresponsive ‘patient’ with a shockable rhythm. Investigators used recorded audio-visual material to observe and time student performance on two key resuscitation skills: time to CPR and time to defibrillation.

Results: Preliminary findings indicate that time to defibrillation was 96 seconds in the CLEIMS group compared to 129 seconds in the control group. Final analyses are pending and will be presented at the ASM.

Conclusion: Full conclusions will be presented at the ASM. The data may influence deliberations about the permanent addition of the CLEIMS methodology to university MBBS training.

Accuracy of rapid disposition decision by emergency clinicians

Andrew Backay MBBS,1 Biswadev Mitra MBBS MHSM PhD FACEM,1,2 De Villiers Smit MBChB FACEM,1 Gerard O'Reilly MBBS FACEM MPH MBiostat AStat,1 Adam Bystrzycki MBBS FACEM PGDipEcho,1 Martin Keogh Grad Dip CCRN Grad Dip Management1

1Emergency & Trauma Centre, The Alfred Hospital, Melbourne, Victoria, Australia; 2Department of Epidemiology & Preventive Medicine, Monash University, Clayton, Victoria, Australia

Objectives: Rapid disposition protocols are increasingly being considered for implementation in emergency departments (EDs). Among patients presenting to an adult tertiary referral hospital, this study aimed to compare disposition prediction accuracy of a ‘rapid decision’ at the conclusion of history and examination, compared to disposition following standard assessment.

Methods: Prospective observational data were collected for one month. Clinicians filled out a questionnaire within five minutes of obtaining a history and clinical examination for eligible patients. Predicted patient disposition was compared to final disposition.

Results: There were 301 patient episodes included in the study. Predicted disposition was correct in 249 (82.7%; 95%CI; 78–86.8) cases. Accuracy at predicting discharge to home appeared highest amongst emergency physicians at 95.8% (95% CI: 78.9–99.9). Predicted admission rates were 20.3% (95% CI: 11.0–32.8) higher with a rapid decision.

Conclusion: Rapid assessment by ED clinicians can predict disposition accurately but was associated with a potential increase in admission rates. Any model of care utilising rapid disposition decision making should involve an emergency physician, establishment of inpatient systems for further assessment and a culture of timely inpatient team transfer of patients to the most appropriate treating team for ongoing patient management.

Do emergency nurses and doctors agree in their triage assessment of dyspneic patients?

Beng Leong Lim MBBS (UNSW) MRCS (A&E) FAMS, Xing Mei LIEW BSc (Nursing) Master (Nursing), Alicia VASU BSc (Biomedical Sciences) MPH, Kim Chai CHAN MBBS (Singapore) FRCS (A&E) FAMS

Tan Tock Seng Emergency Department, Singapore

Objective: Identifying high risk patients at triage in the emergency department (ED) to receive immediate treatment is critical. The inter-observer agreement (IOA) between doctors and nurses on triaging adult ED dyspneic patients has not been investigated. The aim of our study was to investigate their IOA on this task.

Methods: We conducted a prospective observational study to investigate their IOA on assessing dyspneic patients at triage at a tertiary adult ED. We compared eight previously trained nurses with doctors on a convenience sample. Each patient was assessed by a nurse and two doctors (Dr 1 and 2); blinded to one another. They would decide on the patients' clinical status, pulmonary auscultation findings, four-point Patient Acuity Category (PAC), triage-initiated investigations and treatment. We traced each patient from ED arrival till disposition. The primary outcome was the proportion of patients escalated to the resuscitation room from consult. Secondary outcomes included proportions of various PAC status assigned by them and their IOA, their patterns of distribution and IOA on triage evaluation, investigations and treatment. We reported IOA as the kappa (k) statistic [95% confidence intervals (CIs)] and percent agreement. IOA was poor if k < 0.20; fair if k = 0.21–0.40; moderate if k = 0.41–0.60; good if k = 0.61–0.80 and very good if k = 0.81–1.00.

Results: We recruited 302 patients; median age was 61 years. There were 22 (7.3%) patients escalated to the resuscitation room from consult; only one because of clinical deterioration. The range of proportions of assigned PAC status by nurses, Drs 1 and two were: PAC 1: 20.2–24.2%; PAC 2: 71.5–72.8%; PAC 3: 4.3–7.0%; PAC 4: 0.0–0.7%. Their IOA was: Nurse versus Dr 1: k: 0.58 (0.48–0.68); Nurse versus Dr 2: k: 0.55 (0.45–0.65); Dr 1 versus 2: k: 0.65 (0.56–0.74). Their IOA was at least fair to moderate [k: 0.33 (0.22–0.43) to 1.00] in the remaining steps from evaluation to treatment except for ordering of electrocardiogram [k: 0.19 (0.10–0.27) to 0.45 (0.35–0.55)]. Their percent agreement ranged from 81 to 100% for all assessment steps except for ordering of chest X-ray (66 to 73%) and electrocardiogram (61 to 71%).

Conclusion: Trained ED nurses were safe and could manage adult dyspneic patients as well as doctors at triage.

The comparison of eGFR and serum creatinine in identification of chronic kidney disease and prevalence and diagnosis rate of CKD in a primary care sample

Brian O'Connell MICGP,1 Liam G. Glynn MD MICGP FRCGP FRCSI,2 Anna Holdgate FACEM MMed,3 Richard Tobin MRCPI M Sc,4 Donal Reddan MRCPI MHS5

1Registrar, Western General Practice Training Programme, Galway, Ireland; 2Senior Lecturer in Primary Care, Department of General Practice, NUI, Galway, Ireland; 3Associate Professor, Emergency Medicine Research Unit, Liverpool Hospital, Sydney, Australia; 4General Practice Principal, Tobin Healthcare, Castlebar, Co. Mayo, Ireland; 5Consultant Nephrologist, Merlin Park Regional Hospital, Galway, Ireland

Introduction: The imminent arrival of routine estimated Glomerular Filtration Rate (eGFR) calculations brings a new opportunity to identify previously undiagnosed patients with chronic kidney disease (CKD). This study identifies and compares patients with eGFR < 60 ml/min/1.73 m2 and normal creatinine, with patients with elevated creatinine in an Irish primary care population.

Methods: Consecutive serum creatinine requests over a four-month period from a single primary care site were examined. Using the modified MDRD equation eGFR was calculated for each patient and the charts of all 502 patients in the sample were reviewed and data collected on patient characteristics and factors that could influence CKD prevalence.

Results: 502 individual adult patients met inclusion criteria. 76 (15.1%) were shown to have eGFR < 60, of whom only 6 (8%) had a previous diagnosis of CKD in their primary care record. Of the 76 patients, 49 had an elevated creatinine while the remaining 27 were identified only using the eGFR method. Comparisons between these two groups showed significant difference in gender but no significant difference in age, recorded BP or treatment with cardiovascular medications.

Conclusions: The use of eGFR rather than simple creatinine measurement increases the diagnostic rate for chronic kidney disease, particularly amongst older women, with an additional 5.4% of our population diagnosed with CKD as defined by eGFR < 60. Despite an estimated prevalence of CKD of 15.1% in our primary care sample, the rate of diagnosis in the medical records was low and many of these patients were not receiving recommended treatments to minimise cardiovascular complications.

The appropriateness and adequacy of documented treatment limitations in a tertiary regional Australian hospital: a prospective cohort study

Laura Berkeley MBChB (Advanced Trainee Emergency Medicine)1, Julian Stella MBBS FACEM,1 Neil Orford MBBS FCICM FANZCA PG Dip Echo PhD Scholar,2 Charlie Corke MBBS BSc MRCP (UK) FCICM,2 Tania Elderkin Master Health Science GCertCCNurs Clinical Research Nurse RN,2 Sharyn Milnes MBioEth GDipNurs GCertCCNurs RN3

1Department of Emergency Medicine, The Geelong Hospital, Geelong, Victoria, Australia; 2Intensive Care Unit, The Geelong Hospital, Geelong, Victoria, Australia; 3School of Medicine, Deakin University, Waurn Ponds, Victoria, Australia

Objective: Documenting treatment limitations (TL) during hospital admission is of increasing importance with an ageing population. To be of use TL must be precise, clearly documented, and communicated. Instruments to identify patients with end-of-life needs have not been evaluated in Australia.

Aim: To examine the prevalence and adequacy of TL, and one-year outcomes of in-patients qualifying for supportive/palliative care using an objective tool.

Methods:Design: prospective cohort study. Setting: Two 24-hour periods in a tertiary regional hospital (Nov-Dec 2011), Australia. Participants: 633 adult in-patients. Measures: Need for discussion of TL – assessed by UK Gold Standards Framework (GSF) Prognostic Indicator (criteria identifying patients in their last year of life). Adequacy of documentation and one-year mortality – assessed by review of records, and national death index linkage pending. Analyses: Descriptive statistics.

Results: There were 171 (27%) patients with ≥1 GSF criterion, with discussion of TL documented in 60(36%) cases, of which less than half 25 (42%) were adequately documented. The majority of TL were documented by admitting team (57%) or covering team (22%). The one-year mortality for patients identified using the GSF was 37% (63) overall compared with 40% (24/60) in those with a TL (p = 0.664, chi squared test).

Conclusion: Utilisation of GSF identified a patient population with a one-year mortality rate of 37%, suggesting a population with high palliative care needs that would benefit from TL discussions. Despite community advanced care directives, or TL during previous admissions, end-of-life discussions are frequently overlooked and poorly documented.

Understanding emergency department staff needs and perceptions in the provision of palliative care

Freya M Shearer BSc (Hons),1 Ian R Rogers MBBS FACEM,1,2 Leanne Monterosso PhD BNurs (Hons) RN RM FACN,1,2,3 Gail Ross-Adjie PhD RN MACN,1,2 Jeremy R Rogers BSc2

1St John of God Murdoch Hospital, Murdoch, Western Australia, Australia; 2The University of Notre Dame, Fremantle, Western Australia, Australia; 3Edith Cowan University, Joondalup, Western Australia, Australia

Objective: The primary aim of this study was to investigate Australian ED staff perspectives and experiences regarding palliative care provision. Secondary aims were to assess staff views about death and dying, their awareness of common causes of death in Australia and their assessment of which patients most warrant a palliative care approach.

Methods: All medical and nursing staff working in a private ED in Perth Western Australia were asked to complete a combined quantitative and qualitative survey. The survey tool, previously used in other healthcare settings, uses a combination of Likert scales and open ended questions. Descriptive statistics and intergroup comparisons were made for all quantifiable variables, while formal content analysis was used for text responses.

Results: Surveys were returned by 22 doctors and 44 nurses, with most reporting only working knowledge of palliative care but clinical proficiency in symptom control. Confidence in palliative care provision was lower amongst nursing than medical staff but educational needs were similar. Cancer diagnoses were consistently overestimated, and dementia and COPD underestimated, as the most common causes of death. Only six of 63 (9.5%) of respondents identified the correct top five causes of death.

Conclusion: Our study suggests that though ED staff expressed confidence regarding symptom management in palliative care, they lacked understanding of the patients in whom a palliative approach could be applied and sought further education in areas such as end-of-life communication and ethical issues. ED specific training and clinical interventions in palliative care provision would seem to be needed and justified.

Is the ED the right place to die? Exploring opportunities to improve care for patients with advanced cancer

Claudia Helena Marck BSc MSc,1 Jennifer Weil MBBS (Hons) FRACP FAChPM,2 Heather Lane MBChB FRACP,3 Tracey Joy Weiland BBSc (Hons) PhD MPsych(Health)/MAPS,1,4 Jennifer Philip MBBS PhD Mmed FAChPM Grad Dip Pall Med,2,3 Mark Boughey MBBS BMedSc Dip Pall Med MPH FAChPM,2,3 George Alexander Jelinek MBBS MD Dip DHM FACEM1,4

1Emergency Practice Innovation Centre, St Vincent's Hospital Melbourne, Victoria, Australia; 2Department of Palliative Care, The University of Melbourne (St Vincent's Hospital Melbourne), Victoria, Australia; 3Centre for Palliative Care, The University of Melbourne (St Vincent's Hospital Melbourne), Victoria, Australia; 4Department of Medicine, The University of Melbourne (St Vincent's Hospital Melbourne), Victoria, Australia

Many people with advanced cancer present to EDs, and increasingly end-of-life care is being recognised as a core task for ED clinicians. As part of ongoing research to provide a better understanding of these experiences for patients and health care providers we present the results of a cross-sectional survey of ED clinicians. Earlier qualitative work informed the development of survey items, and online survey completion took approximately 15 minutes. Six hundred and eighty-one responded including 444 emergency physicians (13.5% of Australian ED physicians) and 237 emergency nurses, with 51.4% from major referral centers. Most (70.9%) reported finding caring for dying patients rewarding, yet 83.8% believed that ED is not the right place to die and 73.6% felt unable to provide optimal care. The barriers and enablers to providing optimal care will be presented. In addition, 65% of ED clinicians agreed that many patients who are close to the end-of-life undergo treatment in the ED that is unlikely to benefit them, mainly due to a lack of discussion or documentation regarding limitations of care. Given the ageing population and the expecting rapidly rising incidence of cancer specifically, it is expected that ED attendances by patients with advanced cancer will increase considerably. This will place significant demands on ED resources and staff given their complex care needs and requirements. Interventions to improve care in EDs for these patients must address information provision to patients, documentation, communication between disciplines, and consideration of afterhours support by palliative care services.

Supported by the Victorian Cancer Agency.

Developing robust, peer reviewed, credible resources for emergency clinicians

Sally McCarthy MBBS FACEM MBA, John Mackenzie MBChB DipMSM FACEM, Vanessa Evans, Sophie Baugh

NSW Agency for Clinical Innovation, Emergency Care Institute, New South Wales, Australia

Emergency clinicians are required to assess, diagnose, treat and discharge undifferentiated patients within defined time based access target performance. Evidence Based Medicine dictates that clinicians implement the most current and up-to-date evidence into clinical care to ensure high quality, safe patient care is delivered at the right place and at the right time. Delivering this care requires credible and accessible resources to be readily available to clinicians on the shop floor.

The internet has allowed the formation of online communities of interest, with rapid dissemination of up to the minute clinical knowledge, research findings and practical tips and tools, when compared to modes of knowledge translation by traditional educational and professional bodies. Despite leadership and participation by subject matter experts in the on-line milieu, organically developed rapidly responsive on-line clinical material often lacks the robust credibility of traditional peer review and organisational imprimatur.

A middle ground, whereby the need for responsiveness and rapidity of development of on-line clinical resources is balanced against the need for peer review and “official” sanction of such resources is required.

This presentation will discuss the process developed by the NSW Emergency Care Institute to meet these inevitably conflicting objectives. A framework for the identification of knowledge needs, resource gaps, and a process to create, peer review, endorse and publish such resources will be described. This was developed to assist ED with credible, accessible clinical tools that are relevant to the clinical practice of time poor clinicians.

NSW ED quality framework implementation project

Sophie Baugh, Sally McCarthy MBBS FACEM MBA

Emergency Care Institute, Agency for Clinical Innovation, Chatswood, New South Wales, Australia

An integrated framework operating over the whole spectrum of emergency medicine, not just service provision, is essential. This presentation will describe the project being led by the NSW Emergency Care Institute (ECI) that involves rolling out the ACEM developed ED Quality Framework to 24 emergency departments (EDs) across NSW. The framework illustrates the requirements of EDs to provide high quality patient care services and develop quality initiatives.

The ECI has developed resources and supporting tools to allow assessment and implementation of the framework and Quality Support Officer (QSO) posts have been funded in 24 EDs throughout NSW for 12 months. These posts support ED teams to establish a quality team, implement and evaluate the ED Quality Framework, develop a workplan and undertake a range of quality projects. These projects are mapped against the framework domains of Clinical; Education and Training; Research; Administrative; Professional. All sites are undertaking two mandatory projects reviewing sensible test ordering and death reviews and M&M (Morbidity and Mortality) meetings in the ED.

The aims of the presentation will be to explain the rationale and aims of the NSW project; outline the project structure; discuss qualitative feedback from participating sites; provide the audience with project outcomes; discuss evaluation and sustainability.

The project is due to finish in December 2013 therefore this will offer an early insight into the progress and findings.

Errors, appeals and incidents: designing clinical governance for EDs

Susan Ieraci MBBS FACEM

Emergency Department, Bankstown Hospital, Balmain, New South Wales, Australia

As hospital medicine seeks to standardise quality and safety procedures, there is a risk of over-simplification and misapplication of inappropriate tools to the ED setting. ED is a multi-patient environment that requires prioritisation of scarce resources across conflicting patient needs. The tool that is commonly used to review clinical “incidents” – the Root Cause Analysis or RCA – is a blunt tool that is performs poorly at assessing prioritisation in an environment of constraint. It also involves retrospective analysis in assessing actions taken and decisions made without the benefit of hindsight. Complex decision-making in ED can be compared and contrasted with the production of judicial decisions in the law. The legal system provides an appeal process as a matter of course. In medicine, a diagnosis that is overturned on review is generally classified as an error. This contrast will be explored in greater detail in the session. A purpose-designed incident review process for EDs would include the following characteristics: ● Clinicians involved in the incident included on the review team ● Only information available at the time is known to the team ● The review assesses prioritisation as well as cognitive processes ● A process map of ED activity and conflicting demands is produced ● Any proposed recommendations are reality-tested prior to documentation. A detailed proposed structure for a more appropriate incident review methodology for EDs will be presented in the session.

Introduction of bedside workplace based assessments, in the form of mini CEX and DOPS forms on emergency medicine registrars, at a major urban trauma centre using action research methodology

Alfredo Mori MBBS FACEM FCEM DipEMHC

The Alfred Hospital, Melbourne, Victoria, Australia

Bedside workplace based assessments (WBAs) are accepted formative and summative assessment methods of trainees in many medical specialties. The ACEM will introduce bedside WBAs at least as formative assessments of advanced trainees in emergency medicine by 2016. The mini CEX and DOPS are the most reliable and validated bedside WBA tools. We investigated the feasibility of introducing bedside WBAs in emergency medicine trainees in the form of mini CEX and DOPS in a major Australian trauma centre. An action research group was formed consisting of the lead investigator, at least one DEMT and an advanced trainee who met monthly to identify real or perceived barriers to performance of bedside WBAs at work while trainees are at work. We tested the feasibility of a voluntary, suggestive, opportunistic model of bedside WBAs in progressive phases: a pilot study, staff education phase, the causative relationship of a dedicated bedside WBA assessor and number of mini CEX and DOPS completed, and the impact of volunteered non clinical time from other FACEMS in the ED on the frequency of bedside WBAs performed.

Quantitative results included the number of assessments performed, in what areas of the ED, and on what days of the week, and what times of shifts. Seniority of FACEM assessors and trainees was also recorded. This study also encouraged bedside feedback form FACEM assessors as well as self-feedback by trainees for every encounter.

Qualitative data was studied in the form or thematic tropes from audio recorded action research group meetings, a staff survey or knowledge, attitudes and beliefs to bedside WBAs, and the comments and suggestions on the mini CEX and DOPS forms themselves.

We believe this is the largest study of bedside WBAs specific to EM trainees in Australasia. The study results will inform the Board of Education at ACEM of the feasibility of bedside WBAs as formative assessments of trainees in a major urban emergency department.

New FACEMs: training pathways and perspectives

Katie Moore BAppSc (Hons) BSocSci (Hons), Holly Donaldson BAppSc MPH, Mary Lawson BAppSc (Hons), Andrew Gosbell, BAppSc PhD

Australasian College for Emergency Medicine, West Melbourne, Victoria, Australia

Objective: The purpose of this study was to understand the trainee experience and pathway to Fellowship through the Australasian College for Emergency Medicine training programme. Progression through the programme is self-determined and largely dependent on workplace influences. As such, little is known about trainee progression, trainee's perceptions of the training programme or how they advance through their early careers as Fellows.

Methods: Data was extracted from the College's member's database on a cohort of Fellows elected to Fellowship during the period January 2010 to December 2012. Data included the length of time spent in the training programme; age at completion; country of primary medical qualifications; and region(s) where training was undertaken. This was combined with data obtained from Fellows who responded to a targeted online survey and included questions on perceptions of the training and examination processes; career outcomes; and continuing professional development (CPD) intentions.

Results: The cohort of new Fellows consisted of 57% males and 43% females, averaged 38.9 years at attainment of Fellowship, predominantly undertook training in NSW (23%), Victoria (22%) and QLD (21%) and took a median time of 63.6 months (range, 46.9- 144.5 months) to complete advanced training. The majority of Fellows were satisfied or very satisfied with the training programme (83.3%) and supervision (77.6%). Areas for improvement included College resources and support for exam preparation. The majority of Fellows had a consultant position secured (60.9%), were working within a public hospital setting (70.8%) and had begun or were planning to undertake the College's CPD (99.4%).

Rotating medical students through private hospitals: does it work?

Kate Field BMedSC MBBS MPH&TM FACEM,1,2 Iain Field BSc (Hons) GradDip (Hons) PhD3

1Accident and Emergency, Calvary Hospital, Hobart, Tasmania, Australia; 2Emergency Department, Royal Hobart Hospital, Hobart, Tasmania, Australia; 3Macquarie University, Sydney, Australia

Increasing numbers of medical students in conjunction with decreasing numbers of in-patient beds means that new health education opportunities and clinical experiences for students need to be sought and utilised. The University of Tasmania has been collaborating with Calvary Health Care Tasmania (South) in providing educational opportunities for medical, nursing, pharmacy and paramedic students. But has it provided any benefit for the students?

Over 15 months, quantitative and qualitative data was prospectively collected at the start and conclusion of the medical students' rotation through Calvary Hospital. Calvary is a private, not-for-profit hospital located over two campuses in Hobart, with more in-patient adult beds than the only public hospital in the region. During their rotation, students spent time in the private emergency department, medical ward, surgical theatres, anaesthetics and private rooms.

We discovered that medical students gained valuable clinical experiences and enjoyed their rotation. They have a better understanding of how private healthcare functions within the Australian healthcare system in addition to the benefits and risks private patients' experience. Importantly, they also gain an understanding regarding the impact that the private healthcare system has on the public hospital system. We were able to demonstrate that there is also an increased understanding of the funding system, and the cost of healthcare, for both public and private patients.

This is a landmark study that indicates that medical student placements in private hospitals are valuable and worthwhile experiences, and that private hospital rotations for medical students are a potential educational avenue that should be further exploited.

Is the blood culture a useful test for paediatric pneumonia?

Sathyakala Vijayanand MBBS DNB Family Medicine (paediatric trainee),1 Simon Craig MBBS (Hons) FACEM MHPE,1,2 Jim P Buttery MBBS FRACP MSc MD3,4,5

1Monash Medical Centre Emergency Department, Clayton, Victoria, Australia; 2Southern Clinical School, Faulty of Medicine, Monash University, Clayton, Victoria, Australia; 3Infectious Diseases Department, Monash Children's Hospital, Clayton, Victoria, Australia; 4Department of Paediatrics, Monash University, Clayton, Victoria, Australia; 5SAEFVIC, Murdoch Children's Research Institute, Royal children's Hospital, Melbourne, Victoria, Australia

Background: Blood cultures are commonly ordered for children presenting to the emergency department with pneumonia. Little is known about the utility of blood cultures in this population in an Australian setting. This study aims to identify the prevalence of bacteraemia in children with community acquired pneumonia, and to evaluate the impact of positive blood cultures on subsequent clinical management.

Methods: A retrospective study was conducted on paediatric emergency presentations with an ED diagnosis consistent with pneumonia from January 2011 to May 2012. Patients were eligible for study if they were previously well with a working diagnosis of community-acquired pneumonia. Variables collected included demographic data, chest x-ray reports, details of antibiotic prescription during hospitalisation, clinical severity of illness and pneumonia-related complications.

Results: 903 presentations were identified. Six hundred and twenty-three patients fulfilled criteria to be evaluated for the study. Blood cultures were performed in 41% of the children with a clinical diagnosis of pneumonia, yielding 22/260 (8.4%) positive cultures. Of these, 12 were contaminants, and 10 grew pathogenic organisms. The prevalence of true bacteraemia was 3.8%. Of children with complicated pneumonia,15% were bacteraemic. None of the true positive blood cultures resulted in a change in the management of the children with pneumonia.

Conclusion: Blood cultures have low clinical utility in children with community acquired pneumonia. Routine ordering does not appear to be justified in this patient population.

ED physician adherence with departmental clinical guideline for evaluation of infants < 2 months of age with potential serious bacterial illness

Alison Kent MB ChB MSc FACEM,1 Paula Midgley MB ChB MD MRCP (FRCPCH),2 Tom Beattie MB FRCSE FRCPE3

1Lyell McEwin Hospital, North Adelaide, South Australia, Australia; 2Department of Child Life and Health, University of Edinburgh, Scotland; 3University of Edinburgh, Edinburgh, Scotland

Background: Emergency physicians adhere poorly to clinical guidelines for the management of potentially septic young infants. In addition studies report that physician ‘intuition’ may be more accurate than screening criteria in identifying infants with serious bacterial illness (SBI).

Aim: To examine both compliance with clinical guidelines regarding assessment of infants < 2 months of age with suspected SBI presenting to a general emergency department (ED) and to investigate “physician intuition” in managing SBI in this group.

Outcome measure: Adherence to ED guidelines for managing SBI; infant management with and without investigation (a proxy for “physician intuition”).

Patients and Methods: A retrospective case review of all infants < 2 months of age suspected of having SBI. Clinical, laboratory and outcome data were retrieved and analysed.

Results: Complete data was available for 249/264 (94%) of eligible infants. 153 (58%) were aged less than four weeks. One infant was managed according to guidelines, despite 12 (4.8%) having confirmed SBI. 9/12 (75%) infants with SBI were well appearing and 8/12 (66%) were afebrile at presentation. Infants with SBI and “ill-appearing” infants had significantly more investigations than those without SBI or who appeared well (p < 0.05 in both cases)

Discussion: This study confirms that SBI is rare and that there is poor adherence to departmental guidelines. Physician intuition played a significant role in managing infants with suspected SBI. This suggests current guidelines should be re-written, but further analysis of physician behaviour is needed to determine where and how physician intuition can be incorporated.

Why do children present to emergency departments? A prospective observational pilot study

John Cheek MBBS FACEM,1,2,3,4 George Braitberg MBBS FACEM FACMT Dip Epi Biostats1,4

1Emergency Department, Monash Medical Centre, Clayton, Victoria, Australia; 2Emergency Department, Royal Children's Hospital, Parkville, Victoria, Australia; 3Murdoch Children's Research Institute, Parkville, Victoria, Australia; 4Monash University, Clayton, Victoria, Australia

Background: While there has been discussion in the literature about the number of “general practice” (GP) type patients presenting to emergency departments (EDs) little has been reported in the paediatric emergency medicine literature. It has been estimated that up to 50% of paediatric presentations in Victoria fall within this patient type. Some paediatric emergency departments are currently experiencing levels of access block previously seen only in adult departments. It is important to understand the reasons why parents/guardians present their children to EDs rather than see their GP and, equally as important, to determine why GPs refer children to the ED.

Aim: The primary aim of this study is to describe the expectations of parents and doctors (ED doctors and referring GPs) regarding the type of care provided in a tertiary paediatric ED. A secondary aim is to identify factors associated with ED attendances to determine whether any processes can be implemented to improve the “appropriateness” of ED presentations in this population.

Methods: The setting is a large tertiary referral ED. Qualitative and demographic data will be collected by survey and EDIS. The survey will identify the views of parent/guardian, treating ED Doctor and referring GP (if relevant) for each paediatric presentation over a two-week period (projected 1000 attendances).

Outcomes: Descriptive statistics for survey results will be presented. A comprehensive adult and paediatric literature review on these presentation types will be undertaken.

Busted! Management of paediatric upper limb fractures: not all that it's cracked up to be …

Erin Mills MBBS (Hons) B Med Sci,1 Simon Craig MBBS (Hons) FACEM MHPE,1,2 Ed Oakley MBBS FACEM2,3

1Monash Medical Centre Emergency Department, Monash Health, Clayton, Victoria, Australia; 2Southern Clinical School, Monash University, Melbourne, Victoria, Australia; 3Royal Children's Hospital, Melbourne, Victoria, Australia

Background and aim: Optimal management of limb fractures includes splinting and adequate analgesia. In children with fractures, previous studies have demonstrated inadequate analgesia, however, the prevalence of splinting has not been formally investigated. We aimed to assess the impact of a computerised reminder on the use of splinting prior to X-ray in paediatric ED patients with deformed upper limb fractures.

Methods: Phase one was a retrospective study of 1407 paediatric emergency department patients with upper limb fractures. The records of those with fractures requiring manipulation were identified, and reviewed for data on demographics, clinician seniority, pain scoring, splint application and analgesic use. Phase two was a prospective study of 518 patients in the same population after implementation of a computerised pop-up reminder to give analgesia and apply a splint, to compare splint application and analgesia provision after the intervention.

Results: Phase 1: 212 patients were identified. Of these, 22% had a splint applied prior to X-ray and 76% of patients were prescribed analgesia in the first hour after presentation.

Phase 2: 86 patients were identified. Of these, 49% had a splint applied prior to X-ray, and 72% were prescribed analgesia in the first hour after presentation.

Conclusion: This study identified significant shortcomings in ED management of children with deformed upper limb fractures. Only a minority were splinted prior to X-ray, and a quarter did not receive any analgesia in the first hour after presentation. A computerised reminder significantly improved splint application rates in this population.

Topical lignocaine to improve oral intake in children with painful infectious mouth ulcers: a blinded, randomised placebo controlled

Sandy M Hopper MBBS FACEM FRACP,1,2 Michelle McCarthy MN,1 Chasari Tancharoen MBBS,3,4 Katherine J Lee PhD,2,3 Francesca Orsini MSc,2 Franz E Babl MD MPH FRACP1,2

1Emergency Department Royal Children's Hospital, Melbourne, Australia; 2Murdoch Children's Research Institute, Melbourne, Australia; 3Department of Paediatrics, Faculty of Medicine, Dentistry, and Health Sciences, University of Melbourne, Melbourne, Victoria, Australia; 4Junior Medical Staff Department, St Vincent's Hospital, Melbourne, Victoria, Australia

Objective: To establish the efficacy of 2% viscous lignocaine in increasing oral intake in children with painful infectious mouth conditions compared with placebo.

Methods: Randomised placebo–controlled trial of viscous lignocaine versus placebo at a single paediatric emergency department. Study staff, clinicians, nurses, care givers, and participants were blinded to the group assignment. Children with acute infectious ulcerative mouth conditions (gingivostomatitis, ulcerative pharyngitis or hand foot and mouth disease), and poor oral fluid intake were randomised to receive 0.15 ml/kg of either 2% viscous lignocaine or placebo with identical appearance and flavor. The primary outcome was the amount of fluid ingested in the 60 minutes following administration of the intervention, with a difference in fluid intake of 4 ml/kg considered clinically important. Secondary outcomes were specific ml/kg fluid targets and the incidence of adverse events. The trial was registered with the Australian and New Zealand Clinical Trials Registry.

Results: 100 participants were recruited (50 per treatment group) all of whom completed the 60 minute fluid trial period. Oral intake one hour after drug administration was similar in both groups: lignocaine – mean (standard deviation) 9.48 ml/kg (7.02 ml/kg) vs. placebo 9.32 ml/kg (7.39 ml/kg); mean difference 0.15 ml/kg (95% CI –2.7 ml/kg to 3.0 ml/kg); p = 0.9). Little evidence for difference between groups was found in secondary outcomes. There were no adverse events in either group.

Conclusion: There was little evidence that viscous lignocaine is better than a flavored gel placebo in improving oral intake in children with painful infectious mouth ulcers.

Non-invasive cardiac output monitoring in emergency resuscitation – the NICER trial

Ogilvie Thom MBBS FACEM,1 Helena Cooney BSc,2 David Taylor MBBS MD MPH DRCOG FACEM,3 Jessica Kasza PhD,4 John Fraser MbChB PhD FCICM5

1Nambour Hospital, Nambour, Queensland, Australia; 2Redcliffe Hospital, Redcliffe, Queensland, Australia; 3Austin Health, Austin Hospital, Heidelberg, Victoria, Australia; 4Department of Epidemiology and Preventive Medicine, Monash University, Alfred Hospital, Prahran, Victoria, Australia; 5University of Queensland Critical Care Research Group, The Prince Charles Hospital, Chermside, Queensland, Australia

Background: The USCOM monitor allows Emergency Physicians to rapidly and non-invasively measure stroke volume.

Aim: To determine if fluid resuscitation guided by non-invasively measured response in stroke volume improves lactate clearance and haemodynamic indices in critically ill patients in the emergency department

Methods: Randomised controlled pilot study of adult patients requiring urgent fluid resuscitation and having two out of three of the following: heart rate > 100 bpm, systolic blood pressure < 100 mmHg and lactate > 1.5 mmol/L. Participants randomised to standard care vs fluid resuscitation guided by stroke volume response.

Results: 110 participants were enrolled. There was a significant difference between groups at enrolment with more participants in the stroke volume receiving pre-hospital fluid, 12 (22%) vs 27 (46%), P = 0.004 and in enrolment heart rate, 118 bpm(110–127) vs 105 bpm (99–113), P = 0.01. There was no difference in the amount of fluid given as boluses between groups (1379 ml (1182–1592) vs 1618 ml (1360–1875). Primary outcome, four hour lactate, showed no difference, 2.2 mmol/L (1.7–2.7) /L vs 2.7 mmol/L (1.8–3.7), P = 0.57. Mortality was similar in both groups, 9 (17%) vs six (10%), P = 0.25. Univariate analysis demonstrated a difference in the proportion achieving normalisation of vital signs (HR < 100, SBP > 100, 30% vs 51%, P = 0.04). However multivariate analysis controlling for differences between the groups showed the significance was not sustained, P = 0.20.

Conclusion: In this pilot study, stroke volume response guided fluid resuscitation in critically ill ED patients did not improve outcomes.

REACTED: a prospective before and after validation of a two hour rule out for Acute Coronary Syndrome utilising a multi-marker panel

David Mountain MBBS FACEM,1,2 Tor Ercleve MBBS FACEM,1,2 Peter Allely, FACEM2 Jeremy Rogers BSc,1 Eric Yamen MBBS FRACP,3 Brendan McQuillan MBBS FRACP1,3

1University of Western Australia Medical School, Western Australia, Australia; 2Emergency Department, Sir Charles Gairdner Hospital, Western Australia, Australia; 3Cardiovascular Medicine, Sir Charles Gairdner Hospital, Western Australia, Australia

Background: ACS diagnosis in ED is difficult, with reported high miss rates and admission rates with ACS in only 25–30%. Paired early serial multi-markers are considered acceptable versus late troponins to rule out AMI. However they have not been widely applied or introduced into actual practice. Prolonged waits and late decision making cause severe overcrowding in EDs, particularly monitored bays. This issue compromised patient flow, cardiology management and meeting time based targets in our 600 bed tertiary academic ED.

Intervention: A new rule out strategy was implemented using myoglobin, CK-MB and troponin (mod sensitivity) at arrival and two hours associated with a new guided approach (based on recent heart foundation guidelines) to risk stratification.

Aims-a) describe current major adverse cardiac events (MACE) rates b) whether introduction significantly reduces ED length of stay (LOS) and c) whether admissions rates change.

Important outcome measures: ED LOS, discharge/admission rates to hospital/CCU, MACE at 30 and 45 days, sensitivity and specificity of troponin alone or as part of a multi-marker.

Results: 1044 consecutive chest pain patients had data collected, 429/ 615 pre/post intervention, well matched for important features. ED LOS significantly reduced (6:56 hours to 5:29 (p < 0.0001)). Admission rates (CCU and medical units) dropped by 25% (p < 0.001) and overall 20% more patients were discharged from ED. MACE occurred in 16% pre/post with missed MACE in 2% (3.1%/1.3%, before/after (NS)). Sensitivity and specificity for cardiac markers for both strategies were not different.

Conclusion: shortened serial cardiac marker sampling works in real practice, markedly reducing ED LOS and admission rates with no obvious change in missed MACE.

Ambulance ramping (delayed off-stretcher) is caused by admitted patients

Yusuf Nagree MBBS MBA€ FACEM,1,2,3 Sally Burrows BMath Grad Dip Med Stat4,5

1Centre for Clinical Research in Emergency Medicine, Harry Perkins Institute of Medical Research, Perth, Western Australia, Australia; 2Emergency Medicine, Fremantle Hospital, Fremantle, Western Australia, Australia; 3Emergency Medicine, University of Western Australia, Perth, Western Australia, Australia; 4Royal Perth Hospital, Perth, Western Australia, Australia; 5School of Medicine and Pharmacology, University of Western Australia, Western Australia, Australia

Background: Ambulance Ramping (delayed off-stretcher time) is a significant problem for hospitals in the Western World. It occurs when ambulances are unable to unload their patient in the emergency department (ED). This leads to delays in patient assessment and management. The ED is full because it is occupied by patients who are awaiting a hospital ward bed, but because there are no beds in the hospital, they must wait in the ED causing overcrowding.

Aim: We aimed to determine the main reasons why ambulances were ramped.

Methods: Data was obtained from 1st January 2008 to 30th June 2013 from the emergency department Information System (EDIS) of three major tertiary hospitals in Western Australia. Variables extracted included arrival date, triage category, mode of arrival, diagnosis and disposition. Ramping time was extracted from EDIS. An estimation was made of the number of GP patients using a diagnosis methodology previously described in the literature.

Correlations were performed between ramping time and access block, total number of attendances, total number of GP-type patients and total time admitted patients spent in the ED.

Results: There was a high correlation between ramping time and access block (r = 0.74), between ramping time and total admitted patient time (r = 0.78) with very little correlation between ramping time and total attendances (r = 0.28) and even less correlation with GP-type attendances (r = 0.16).

Conclusion: Ambulance ramping is primarily caused by admitted patients in the emergency department and is not well correlated with total number of patients attending or total number of GP-type patients attending.

Complementary and alternative medicine use among paediatric emergency department patients

David Taylor MBBS MD MPH DRCOG FACEM,1,2 Reetika Dhir MBBS FRACP,1,3 Simon Craig MBBS (Hons) FACEM MHPE,4,5 Thalia Lammers MBBS MD,6 Kaya Gardiner RN MPH,7 Kirrily Hunter RN,1 Paul Joffe MBBS (Hons) BA BSc BMedSci FRACP,3 Franz Babl MD MPH FRACP,2,3,7 David Krieser MBBS FRACP6

1Emergency Department, Austin Health, Melbourne, Victoria, Australia; 2Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia; 3Royal Children's Hospital, Melbourne, Victoria, Australia; 4Emergency Department, Monash Medical Centre, Clayton, Victoria, Australia; 5Department of Medicine, Monash University, Victoria, Australia; 6Paediatric Department, Sunshine Hospital, Victoria, Australia; 7Murdoch Children's Research Institute, Parkville, Victoria, Australia

Objective: To determine the prevalence, nature and adverse effects of complementary and alternative medicine (CAM) use among paediatric emergency department (ED) patients and the perceptions of CAMs among the CAM administrators.

Methods: We undertook a cross-sectional survey in three EDs between January and June, 2013. A convenience sample of parents accompanying paediatric patients was invited to complete a self-administered questionnaire regarding their administration of CAM (within the previous day, one and twelve months) and their perceptions of CAM.

Results: The parents of 409 patients participated. Of these patients, 180 (44.0%, 95%CI 39.2–49.0) and 17 (4.2%, 95%CI 2.5–6.7) had taken a CAM within the previous 12 months and on the day of presentation, respectively. There were no gender or ethnic differences in CAM use (p = 0.88 and 0.46, respectively). However, CAM use was significantly more common among older patients (p < 0.001) and tended to be more common among those with chronic disease (p = 0.09). Fish oil, acidophilus, garlic, chamomile and cranberry were the most commonly used CAMs. Reported adverse events were rare. Parents who had administered CAMs tended to believe that CAMs were safe, drug free, could prevent illness, were more effective than prescription medicines and were safe when taken with prescription medicines (p > 0.05). Data collection is ongoing.

Conclusion: CAM use is common among paediatric ED patients. There is the potential for interactions between CAMs taken on the day of presentation and drugs administered in the ED. Parents who administer CAM tend to have differing perceptions of CAM safety.

Porte d'entree: is your hand hygiene a liability?

Abel Wei MD,1 Joost Hopman MD DTMH2

1Emergency Department, Radboud University Medic Center, Nijmegen, the Netherlands; 2Department of Microbiology, Radboud University Medical Center, Nijmegen, the Netherlands

In the chaos of the ED we tend to forget the fundamentals of hand hygiene. Health care associated infection is one of the most important patient safety problems. Without proper handy hygiene, the personnel become the vector for pathogens responsible for nosocomial infection.

An observational study on the compliance of hand hygiene was carried out in 2012 at ED of University Medical Centre St. Radboud Nijmengen. The subjects, include both ED nurses and physicians, were unaware of the identity of the observer. The compliance is rated on five indications advised by World Health Organisation: before and after touching a patient, before an aseptic procedure, after exposed to patient's body fluid and after touching patient surroundings.

144 observations were recorded in the period of three weeks at random intervals during morning and evening shifts. The compliance of hand hygiene of the ED physicians is 25% and the ED nurses is 29%.

Several interventions have been implemented after the observational study. In collaboration with department of hygiene and infection prevention, we identified locations across our ED where the alcohol disinfectant dispensers were needed. We had doubled the amount of alcohol disinfectant dispensers at our ED department, from 25 to 55. Ten of which are mounted to intravenous drip and indwelling catheter trolleys, specifically targeting the zero compliance of hand hygiene before aseptic procedures. The choice of hand disinfectant had changed to Sterillium Med because of 50% shortening of contact time compared to the previous product, efficacy against noro virus and proven skin friendliness.

To increase the personnel's awareness on hand hygiene we are planning to implement hand hygiene as theme of the week. During the first week of each month there'll be posters hanging up in the department, special attention paid to explaining five interventions to each other during the handovers, wearing badge with “hand hygiene role model”, black-light box to check the adequacy of applying the disinfectant. Most important of all, the compliance will be measured again to document improvement.

After the first of the three-theme week the compliance of the ED physicians is 35% and the ED nurses is 41%. The second and third week theme week takes place at week 29 and 33. The result will be ready for publish for ACEM 2013 congress.

The project Porte d'entrée has been approved by the CEO of UMC St. Radboud Nijmegen and will be enforced at every patient related department. Our ED is the pilot in this hospital-wide project.

Ancillary