key: cord-0078504-u1g556ij
authors: nan
title: CAEP/ACMU 2022 Scientific Abstracts: May 29th – June 1st, 2022
date: 2022-05-23
journal: CJEM
DOI: 10.1007/s43678-022-00309-w
sha: e44a4da2e98ea2c9cf5d807b76b12c0a7e8abb14
doc_id: 78504
cord_uid: u1g556ij

nan

Introduction: CAEP atrial fibrillation (AF) guidelines recommend that emergency physicians apply the CHADS65 rule to identify patients with AF who should initiate anticoagulation. Safe direct oral anticoagulant (DOAC) prescription also requires assessment of contraindications. Little is known about the accuracy of emergency physicians in this process. The objective of this study was to report the reliability of physician identified CHADS65 patient characteristics and contraindications to the prescription of a DOAC. Methods: This was a secondary analysis of the SAFE implementation study which evaluated the safety of an emergency physician pathway to initiate anticoagulation in CHADS65 positive patients with AF. The study included adult patients who presented to one of four emergency departments with AF, between 2018 and 2020. Physicians used a hard copy form to document CHADS65 items and the presence of a predefined list of contraindications to DOAC prescription. Trained data abstractors reviewed the hospital medical records to record the presence and absence of the same variables, blinded to the physician documentation. The primary analysis was the agreement between physician and medical record review on each indication and contraindication, as calculated by Cohen's kappa. The secondary analysis was the sensitivity and specificity of physicians in identifying each variable using the medical record as the reference standard. Results:

The study included 311 patients, 148 (48%) being female, with a median patient age of 69 years. The median CHADS2 score was 1 (IQR: 0-2). Agreement between physician-recorded data and medical record data varied between 0.01 (heart failure) and 1.0 (thrombocytopenia). Kappa scores were \ 0.7 for cirrhosis and creatinine clearance \ 30 ml/min. Kappa scores were \ 0.4 for stroke, diabetes, hypertension, heart failure and interacting medications. Specificities ranged from 89 to 100%. Sensitivities ranged from 0 to 100%, with low sensitivities for stroke (65%: 39, 88), diabetes (81%: 66, 90), hypertension (75%: 68, 81), heart failure (25%: 10, 49), interacting medications (25%: 1, 78), cirrhosis (50%: 14, 86) and creatinine clearance \ 30 ml/min (50%: 17, 83). Conclusion: There was poor agreement between physician-identified CHADS65 items and medical record data. There was also poor agreement for some DOAC contraindications. This raises safety concerns for emergency physicians initiating anticoagulation in patients with AF. Keywords: anticoagulation, atrial fibrillation, safety Introduction: Mechanical chest compression devices (MCCD) are increasingly used to assist cardiopulmonary resuscitation (CPR). However, the pattern of MCCD-induced injuries and their importance are unknown. The main objective of this study was to determine the incidence of MCCD-induced injuries during CPR. Secondary objectives were to determine the incidence of MCCD-induced injuries deemed fatal or life-threatening and to compare the MCCD-induced injuries' incidence between load-distributing band and piston-driven devices. Methods: A systematic review of the literature was performed using five databases (PROSPERO: CRD42016049719). Studies reporting on the incidence of MCCD-induced injuries in adults who had a cardiac arrest during which a MCCD was used were considered eligible for inclusion. Case-reports and studies reporting on traumatic cardiac arrest or limited to a paediatric population were excluded. Injuries needed to be assessed during an autopsy (external or internal) or with an imaging modality such as a computed tomography. Injuries are presented according to their anatomic region and their type. A list of potentially life-threatening injuries was developed prior to the beginning of the study by consensus. Risk of bias assessment was performed using the Cochrane for RCT or the ROBINS-I tools. Results: Nine studies were included with a combined total of 4545 patients. A piston-driven device was used in 7 studies while a load-distributing band device in 2. Autopsies were performed to identify the injuries in 6 studies while 4 studies used computed tomography. Rib and sternum fractures were the most frequently reported injuries (500/621 and 442/1187 respectively). 5/ 84 reported a flail chest. 30/583 myocardial ruptures, 37/862 hemopericardia, 65/1073 pneumothoraces, 44/1035 hemothoraces and 55/832 lung contusions and lacerations were reported. The most frequently injured intra-abdominal organ was the liver with 52/946 injuries followed by the kidney 10/472 Life-threatening injuries tend to be more frequent with the use of MCCD. Five observational studies were considered at a serious risk of bias. Conclusion: Ribs and sternum fractures were frequent, while other significant injuries such as hemopericardium and myocardial rupture were frequently injured by MCCD Chest imaging such as computed tomography should be considered following the use of MCCD aiming to identify the induced injuries. Keywords: chest compressions, cardiopulmonary resuscitation, systematic review Introduction: Defibrillation, the restoration of normal cardiac rhythm by administering a controlled shock, is the standard resuscitative measure for patients experiencing in-hospital or out-of-hospital cardiac arrest due to ventricular fibrillation and pulseless ventricular tachycardia. However, there is still a high rate of unsuccessful shocks. As a result, there has been an increased interest in using machine learning (ML) software to improve the rate of success of defibrillation. The purpose of this systematic review is to evaluate the existing knowledge on the effectiveness of ML algorithms in predicting defibrillation success during cardiac arrest. Methods: MEDLINE, Embase, CINAHL, and Scopus were searched according to PRISMA guidelines for primary literature involving adults, children, or animals with in-or-out of hospital, stimulated, or experimental cardiac arrest, where ML algorithms were used to predict defibrillation success with restoration of spontaneous circulation (ROSC). Primary outcomes of interest include the optimal ML algorithm used, ROSC, survival to hospital or discharge, and neurological status at discharge. Independent reviewers completed study selection, data extraction, and risk of bias assessments. Results: 2369 studies were identified of which 105 full text articles were reviewed and 14 observational studies (n = 5568 patients) were included for final analysis. Cumulatively, 29 ECG waveform features were fed into 15 different ML combinations. The optimal ML algorithms across studies were neural networks (n = 4) , support vector machines (n = 3), decision trees (n = 2), logistic regression model (n = 1), random forest (n = 1), and multiple domain integrative model (n = 1) with 2 studies not finding improvement with their ML models. Amongst these, neural networks had the highest sensitivity and specificity values ([ 90%Sen, [ 95% Spec) . Support vector machine had the best performance with area under receiver operating curve of 0.938. A meta-analysis was not performed due to significant heterogeneity in study design and characteristics. Conclusion: Machine learning algorithms, specifically support vector machines and neural networks, have been shown to have potential in improving the prediction of defibrillation success for in-and out-of-hospital cardiac arrest with high sensitivity and specificity. However, this has not been conclusively established in a controlled manner and future studies are needed to further elucidate clinical applicability. Keywords: cardiac arrest, defibrillation, machine learning Introduction: A shockable initial rhythm has been proposed, with inconsistent results, as a tool to identify patients with a short no-flow time (NFT) (delay between collapse and initiation of cardiopulmonary resuscitation) for patients with an unwitnessed out-of-hospital cardiac arrest (OHCA). Since eligibility for extracorporeal resuscitation is conditional on a short NFT, patients with an unwitnessed OHCA could be candidates if an initial shockable rhythm reliably identify patients with a short NFT. The objective of the present study is to determine the sensitivity, specificity, and likelihood ratios of an initial shockable rhythm to identify patients with an NFT of five minutes or less (NFT \ 5). Methods: A multicentric North American registry was used to include adult patients who experienced a witnessed non-traumatic OHCA. Patients with obviously dead criteria or who received bystander cardiopulmonary resuscitation were excluded, as well as those for whom the initial rhythm, or the time delay before the first assessment of the electrical rhythm was unknown. For the present study, the elapsed time from dispatch call to the first rhythm after the OHCA was considered the proxy variable for the NFT. The sensitivity, specificity, and likelihood ratios of an initial shockable rhythm to identify patients with an NFT \ 5 was calculated. Results: A total of 229,632 patients were logged in the registry, of whom 31,773 patients (13.8%) were included. A total of 8580 (27.0%) had an initial shockable rhythm (Male:78%; Mean age: 64.7 years [95% confidence interval {CI} 64.3-65.0]) and 23,193 (73.0%) had an initial non-shockable rhythm (Male:63%; Mean age: 68.6 years [95% CI 68.4-68.8]). Among patients with bystander witnessed OHCA, the initial rhythm was shockable for 345 patients (45%) of the 776 who had an NFT \ 5 and for 5472 (30%) of the 18,293 who had an NFT of more than five minutes. The sensitivity of an initial shockable rhythm to identify patients with an NFT \ 5 was of 45% (95% CI 41-48), the specificity of 70% (95% CI 69-71), the positive likelihood ratio of 1.49 (95% CI 1.37-1.61) and the negative likelihood ratio of 0.79 (95% CI 0.74-0.84). Conclusion: The pres-

Characterization of the resource consumption of frequent emergency department users at an urban, tertiary-care, academic hospital K. Shih, BSc, MD, J. Zhou, Y. Ma, BSc, I. Cheng, MD, MSc, PhD University of Toronto, Toronto, ON Introduction: High frequency emergency department users (HEDU) utilise a disproportionately large amount of resources compared to their non-high frequency counterparts (nHEDU) and may benefit from interventions to reduce emergency department (ED) use and cost. As hospital catchment areas are heterogenous, this study aims to identify the demographic and cost differential of HEDU at Sunnybrook Health Sciences Centre (SHSC). Methods: This was a retrospective, observational case costing study done at SHSC, spanning all patient visits from 2017 to 2019 inclusive. Costs were ED-related only, not including admission or physician billings. HEDU were defined as patients who visited the ED more than 3 times or were admitted to hospital from the ED more than 2 times over any 6-month period. Patient-level demographics, visit-level details (chief complaint, length of stay, acuity, etc.), and costs were compared between HEDU and nHEDU. Results: A total of 187,480 patient visits occurred over the 3-year period. These visits were made by 119,030 unique patients, and 3,749 were HEDU. HEDU were more likely to be older (62.0 [61.3, 62 .7] vs 51.0 [50.9, 51.1]), have higher-acuity CTAS level (CTAS 1-3 visits 91.2% vs 86.7%, p \ 0.001), consulted to a specialty service (39% vs 23.3%, p \ 0.001), and be admitted to a medical service (23.9% vs 10.8%, p \ 0.001), though less likely to be admitted to a surgical service (6.4% vs 7.2%, p \ 0.001). Cumulatively, HEDU were more costly ($3195.15 [$3098,29, $3292 .01] than nHEDU $629.69 [$624.22, $635.16 ] and required more diagnostic imaging overall. On a per-visit basis, HEDU costs were higher ($462.48 [$455.25, $469.71] vs $449.26 [445.69, $452 .83]) with fewer diagnostic images/visit compared to their non-HEDU counterparts. There were significant differences in mode of transport to hospital, timing of visits, and use of allied health resources to facilitate disposition. Conclusion: HEDU place a disproportionate resource burden on the health system. At SHSC, it appears that this burden is predominantly secondary to decompensated medical disease. Per visit, resource consumption appears to be similar though statistically higher than non-HEDU; however, due to increased frequency of visits, HEDU have significantly greater cost and resource-consumption overall. Further identification of specific underlying diagnoses driving the HEDU at this site are needed to direct outpatient efforts to reduce ED use. Keywords: cost, frequent user, resource utilization Introduction: Frequent use of the emergency department (ED) has been a topic of study due to its complex nature. However, pediatric patients who frequently present to the ED are a distinct population that has remained understudied in literature. The objective of this study was to characterize pediatric patients who display high frequency use of the ED among residents of the Hamilton Niagara Haldimand Brant Local Health Integration Network in Ontario, Canada. Methods: A descriptive analysis of high frequency use of the ED (5? per year) during an eight-year period was undertaken among patients, 0-17 years old. The individuals were divided into eight cohorts to identify how many years they had remained a high frequency user of the ED over this timeframe. Information on ED visits, hospitalizations and patient characteristics between 2012/13-2019/20 was analyzed using data obtained from Integrated Decision Support (IDS): National Ambulatory Care Reporting System and Discharge Abstract Database. Results: 13,809 children (average age: 7.4 years) were found to have high frequency use of ED services with the majority (78%) having one year of high ED use. Less than 4% of children had high use for five to eight years. There were a total of 198,356 ED visits with an average of 14.4 visits per patient over the eight years. 125,941 visits were high acuity (Canadian Triage and Acuity Scale: CTAS 1-3) of which 2,206 (2%) were CTAS 1. The most common ICD-10 discharge diagnoses from the ED were: acute upper respiratory infections of multiple and unspecified sites (8%), abdominal and pelvic pain (7%) , and fever of other and unknown origin (5%). Of the total, 6,249 children (45%) had 15,806 hospitalizations (1-48 per person, average 2.5 hospitalizations) and spent 73,942 days in hospital (range: 1-580, average 11.8 days) over 8 years. The most common co-morbid conditions among the children were brain injury (n = 4,261, 31%) , other perinatal conditions (n = 3,165, 23%) , and asthma (n = 2,101, 15%) . Conclusion: Pediatric patients experiencing high use of the health care system are a small population requiring significant attention as the majority of these individuals seem to have one year of high ED use. Consecutive high use of the ED is a rarer phenomenon. Data tracking persistent high frequency use of the ED in the pediatric population is currently quite limited and as such, findings from this study may inform interventions at the community level. Keywords: emergency department, high use, pediatric Introduction: According to Statistics Canada, cannabis use has increased since its legalization in 2018. Whether this correlates with increased ED utilization is unclear. We compared the number and characteristics of cannabis-related ED visits pre-and post-legalization in an urban tertiary care setting. Methods: This multi-centre health records review included ED visits to the Civic and General campuses of The Ottawa Hospital over three 6-month periods: seasonal control (Oct. 17, 2017 -Apr. 16, 2018 ), pre-legalization (Apr. 17-Oct. 16, 2018 , and post-legalization (Oct. 17, 2018 -Apr. 17, 2019 . We piloted a strategy using ICD-10-CA codes to identify all cannabisrelated visits. Trained reviewers assessed visits for eligibility and used a standardized data collection tool. Our primary outcome was number of cannabis-related ED visits and secondary outcomes included patient characteristics, method of ingestion, and disposition. We performed interrupted time series (ITS) analyses and a logistic regression adjusting for sex, age, triage time, CTAS, and method of ingestion. Results: Characteristics of the 629 included visits (183 seasonal control, 231 pre, and 215 post) were: Median age 26, female 43.9%, inhaled cannabis 69.2%. There were 22 to 44 cannabis-related visits per month with a median CTAS of 3 [range 1 to 5]. Disposition included discharge from ED 85.1%, discharge after consultation 3.7%, admission 6.5%, or left without being seen 4.6%. We observed a significant increase in the number of cannabis-related ED visits/month pre-legalization (slope = 1.34; p = 0.001) and a significant decrease post-legalization (slope = -2.23; p = 0.04). Difference between the predicted number of visits post-legalization and the secular trend was -16.32, 95% CI -30.69 to -1.95; p = 0.03. During the same period, there were no significant difference between the predicted and the secular trend of total number of ED visits (-850.55, 95% CI -2730 (-850.55, 95% CI - .37 to 1029 p = 0.38) . The proportion of visits leading to admission did not significantly change (7.5% vs 4. 7%, p = 0.17) , and the proportion of those leaving without having been seen increased post-legalization (2.9% vs 7.9%, p = 0.005). There were no differences by sex, age, triage time, or CTAS. Conclusion: Our study suggests that legalization led to a decrease in cannabis-related ED visits. Future inquiry should aim to characterize acute adverse outcomes of cannabis use to inform harm reductive strategies as other jurisdictions consider its decriminalization or legalization. Keywords: cannabis, public health, substance use LO15 Demographic characteristics of patients using virtual urgent care services in Ontario As part of the funding agreement, each participating site was required to collect and report a minimum of six months of standardized patient level data. Patients who scheduled an appointment but did not complete a virtual encounter with a healthcare provider were excluded. Results: There were 22,278 VUC visits (8,075 from paediatric and 14,203 from adult sites), with a mean daily volume of 112 visits. 16,263 (73.0%) visits occurred between 12 and 8 pm, with nearly 90% happening during weekdays. Median (IQR) time from initial patient contact to first contact with the health care provider was 95 (38, 232) minutes. Median (IQR) age was 26 (4, 43) years, 13,236 (59.4%) were female and 17,554 (78.8%) patients had a primary care provider. When patients were asked why they contacted VUC, 5,442 (31.02%) patients who had a primary care provider said they could not make an appointment with their family physician. 3,932 (17.6%) visits had a triage nurse screen patients and 3 sites reported CTAS scores. 50.1% were CTAS 5, 21.6% were CTAS 4, 22.2% were CTAS 3 and 6.1% were CTAS 2. Rash, fever, abdominal pain, and COVID-19 vaccine inquiry represented 30% of the presenting complaints. Adult patients using VUC seemed to be calling in more for expedited advice as opposed to emergency care. 12,910 (65.8%) patients were discharged home, 3,179 (16.2%) were referred to the ED, 1,234 (6.3%) were referred to their primary care provider, and 1,027 (5.2%) left without being seen. Conclusion: Many patients using VUC services had a primary care provider, but access to their provider was not available in a timely fashion. Considering most presenting complaints are low acuity, future studies should examine if VUC may be safely and more economically provided by non-emergency physicians, especially for adult patients seeking VUC. Keywords: COVID-19, health policy, virtual care Introduction: There currently exists no standard productivity measure for emergency physicians (EPs). The objectives of this systematic review were i) to identify definitions and measurements of EP productivity, and ii) to evaluate factors associated with productivity. Methods: This systematic review and meta-analysis was reported according to PRISMA guidelines and registered on PROS-PERO. We searched MEDLINE, Embase, CINAHL, and ABI/ INFORM from inception to November 1, 2020. Our search strategy was peer reviewed according to PRESS guidelines. Two reviewers independently identified observational and interventional studies involving EPs working at adult or pediatric EDs in which physician productivity was a defined outcome. We excluded: non-ED settings, non-EPs, resident physicians, and studies that only reported ED productivity. We extracted demographic data, productivity metrics, and factors associated with productivity. Definitions and measurements of productivity were reported qualitatively. A meta-analysis was conducted for comparable outcomes using a random effects model with effect size presented as rate ratios. Risk of bias was assessed using the Newcastle-Ottawa scale. Results: After screening 4924 studies, a total of 44 studies (N = 2088 attendings, N = 236 EDs) were included for qualitative synthesis, of which 6 were selected for meta-analysis. Most studies were conducted in single academic centres with mixed patient acuity and population. Components of definition for EP productivity included: number of patients managed (n = 28, 63 .6% studies), revenue generated (n = 24, 54 .5%), and patient processing time (n = 10, 22.7%) . Patients per hour (n = 26 studies) , relative value units generated per hour (n = 24) , and provider-disposition time (n = 10) were primary measures of productivity. Studies evaluating scribes (n = 9) , residents (n = 3) , and mid-level providers (n = 3) reported an increase in EP productivity. Physician factors including years of experience, age, and board certification (n = 6) as well as faculty teaching scores (n = 5) did not impact productivity. EMR implementation (n = 5) was associated with decreased productivity. Scribes improved EP productivity with a pooled rate ratio of 1.16 (n = 6 studies, 95% CI 1.03-1.32). Conclusion: EP productivity is heterogeneously defined but common elements exist. The findings of this study are an important step towards the development of nationally accepted productivity metrics that can improve patient care efficiency, guide QI initiatives, and optimize physician staffing. Keywords: emergency physician productivity Introduction: Waiting in a queue is an inconvenience associated with many service-based industries including healthcare. While short wait times are typically well received by service users, the absolute wait time is less significant to patients visiting a hospital with regards to their satisfaction with care. Rather, the perceived wait time is a key measure in patient satisfaction. To enhance the wait time experience in emergency departments (ED), adult and pediatric EDs publish their wait times using real-time wait time displays (rtWTDs) in their hospital and online. The primary objective of this study is to use service industry evaluators to assess the user experience of rtWTDs for parents who would potentially access pediatric emergency departments for their children. Methods: This was a cross-sectional survey study of potential pediatric patients' parents/legal guardians on the user experience of rtWTDs. The online survey has four components that include an evaluation of socioeconomic demographics (SES), a subjective literacy survey (SLS), subjective numeracy scale (SNS), and a heuristic evaluation of real-time wait time displays. SES, SLS, and SNS were assessed to determine their impact on what information in rtWTDs is important to subgroups of the population. Four rtWTDs (numeric, graphical, multi-data, and local institution) were shown to participants in a randomized order to assess comprehension, effectiveness, and overall satisfaction of each rtWTDs. Participants were then asked to rank the four rtWTDs. Results: 157 participants completed the online survey. The mean age was 34 years. 136 (86.6%) participants were female. Mean SNS score (out of 5) was 3.88 (SD ± 0.87), and mean SLS score (out of 5) was 4.33 (SD ± 0.79). Satisfaction for the numeric and multi-data webpages were significantly greater than that of the graph webpage (p \ 0.05). Overall ratings for the numeric, multi-data, and local institution (CHEO) webpages are significantly greater than that of the graph webpage (p \ 0.05), but no statistical difference between them. SES, SNS, and SLS did not have a significant effect on survey responder's selection of preferred rtWTD webpage. Conclusion: Currently there is a limited, but emerging role for rtWTDs in Canada. Regardless of socioeconomic status, health literacy and numeracy literacy, numeric based wait time information is most preferred by parents/guardians who may access rtWTDs prior to accessing healthcare care for their children.

Keywords: communication tools, patient experience, wait times LO18 Evaluating referral patterns from Health Link to the emergency departments in Alberta I. Cooper, MSc, A. Schmaus, MSc, T. Witten, PhD, J. Bakal, PhD, J. Huang, PhD, D. Watt, MD, E. Lang, MD University of Calgary, Calgary, AB Introduction: In Alberta, Health Link (811) provides a 24-h, nurse staffed, phone resource to the public for health care advice. Health Link nurses use decision support guidelines to guide patients to seek further care in the emergency department (ED), follow up with a primary care provider (PCP) or provide self-care (SC) at home. The appropriateness of some of these referrals has been challenged, with some healthcare providers working in the ED believing that too many Health Link callers are referred inappropriately. Therefore, the aim of the present work is to provide a descriptive analysis of Health Link referral patterns to the ED. Methods: Using administrative health data from January 1, 2018-December 31, 2019, we categorized Health Link calls as likely appropriate referrals (ED referral with a Canadian Triage and Acuity Scale (CTAS) of \= 3, presenting to the ED within 24 h), less-likely appropriate referrals (ED CTAS of [ 3) or a patient over ride referral (referred to PCP/SC but then then presented to the ED within 24 h with a CTAS of \ = 3). The primary outcome was the percentage of likely appropriate referrals among all calls received by Health Link that went on to present to an ED; the secondary outcome was the percentage of patient over rides among all calls referred to PCP or SC. Results: In our two-year sample, there were 900,196 calls to Health Link with 241,103 (26.7%) referred on to the ED. However, only 44.4% of these (107, 034) followed through on the advice received and actually presented to an ED. These presentations account for 3.4% of all ED visits during the study period. Of the 107,034 patients who followed through with the advice to present, 77.3% were determined to be likely appropriate referrals, while 22.7% were determined to be less-likely appropriate referrals. Of the patients sent to the ED, the admission rate was similar to that of the general ED population (8.01% and 9.16% respectively). 86,783 patients presented as patient over rides, representing 13.2% of all calls referred to PCP or SC. Conclusion: Only a small percentage of ED visits stem from Health Link referral. In the majority of cases, Health Link referrals are likely appropriate, with a similar admission rate between the patients referred by Health Link and the general ED population. However, the present work demonstrates that there remain opportunities to improve evaluation and advice procedures through Health Link. Keywords: emergency department referral, emergency department utilization, telehealth Introduction: Rising awareness of opioid-related risks and opioid reduction initiatives may have altered emergency department (ED) approaches to analgesia. We aimed to characterize temporal shifts in ED opioid prescribing over the past decade. Methods: We studied a 10% random sample of Alberta ED visits from 2010 to 2019, extracting data for all non-pregnant adults age [ 18 excluding those with active cancer or palliative care. We linked with the AHS pharmacy database to identify opioid prescriptions and grouped them as: codeine, Tramadol, or potent (morphine, hydromorphone, oxycodone, fentanyl). We stratified patients as: naïve (no opioid prescriptions in the prior year, no OUD), exposed ([= 1 prescription, no OUD) or probable OUD (long-term opioid use, an opioid-related ICD code, or opioid agonist therapy). Univariate linear regression tested associations between prescribing year and outcomes (prescribing rate, average opioid days and oral morphine equivalent (OME) per prescription). Results: We analyzed 644,281 eligible visits, 51% female, mean age 45 years. Opioids were prescribed for 9.4% of all patients, 27% of OUD, 19% of opioid-exposed and 3% of opioid-naïve patients. Opioid prescribing decreased during the study period (b = -0.005, p \ 0.01) reflecting a 0.5% reduction per year. Time trends differed by agent and subgroup. Codeine and oxycodone use decreased (b = -0.04 and -0.03 per year, respectively) while Tramadol and potent opioid use increased (b = 0.09 and 0.06 per year). Prescriptions decreased over time in OUD and opioid-exposed subgroups (b = -0.04 and -0.005) but increased for opioid-naïve patients (b = 0.013, p \ 0.01). In this group, prescriptions increased most dramatically for Tramadol (b = 0.12, p \ 0.01) and potent opioids (b = 0.13, p \ 0.01). Opioid days per prescription decreased for OUD and opioid-exposed (b = -0.15 and -0.03, respectively) but remained constant for naïve patients. Intensity (OME) decreased over time, most dramatically for oxycodone and potent opioids in OUD patients (b = -54.8 and -128.6, respectively). Conclusion: ED opioid prescribing has steadily decreased in Alberta EDs. Lower doses and shorter durations are being used, especially for OUD and opioid-exposed patients. Rising use of potent opioids in naive patients may be cause for concern and the rising popularity of Tramadol justifies further safety investigation. Reluctance to prescribe potent agents in patients with OUD could increase subsequent risk of overdose from illicit use. Keywords: opioids, prescribing, trends LO20 Who gets the best drugs? Predictors of potent opioid prescriptions in the emergency department J. Hayward, MD, MPH, K. Lonergan, BSc, G. Innes, MD, MSc University of Alberta, Edmonton, AB Introduction: Increasing awareness of opioid-related risks may influence pain management in the emergency department (ED). Opioid prescription potency may be a risk factor for future OUD. We aimed to characterize ED prescribing for potent opioids. Methods: We studied a 10% random sample of Alberta Health Services (AHS) ED visits from 2010 to 2019, extracting data for all non-pregnant adults age [ 18. We excluded patients with active cancer or in palliative care. We linked with the AHS pharmacy database to identify opioid prescriptions and grouped these as potent (ex. oxycodone, morphine, hydromorphone, fentanyl) or non-potent (codeine, Tramadol). Patients were stratified as: naïve (no opioid prescriptions in the previous year and no OUD), exposed ([= 1 opioid prescription but no OUD) and probable OUD (long-term opioid use [LTU] , an opioid-related ICD code, or opioid agonist therapy). We used univariate linear regression to test associations between sex, age and potent opioids. We used multivariable models to adjust for other significant covariates. Results: Of 644,281 eligible patients 51% were female and mean age was 45 years. Overall, 60,270 ED visits (9.4%) were linked to an opioid prescription. In univariate models, male sex (b = -0.14, p \ 0.01) and older age (b = 0.009, p \ 0.01) predicted potent opioid prescription. After multivariable adjustment, the strongest predictors of a potent opioid were: prior OUD ); prior opioid exposure (aOR = 6.8 [6.4-7.2]); a musculoskeletal (aOR = 4.6 [4.1-5.2]), genitourinary (aOR = 2.6 [2.3-3.0]), gastrointestinal (aOR = 2.5 [2.2-2.9]) or traumatic complaint (aOR = 2.2 [2.0-2.5]). Significant but weaker predictors of potent opioid prescription included age group 31-65 years (aOR 1.50 [1.41-1.59]), tertiary or urban hospital (aOR = 1.4 [1.3-1.5] and 1.3 [1.2-1.5] respectively), psychoactive medication use (aOR = 1.26 [1.20-1.32]), non-opioid analgesic use (aOR = 1.21 [1. 16-1.26] ) and male sex (aOR = 1.11 [1.07-1.16]). Concurrent alcohol ) and stimulant use disorders (aOR = 0.66 [0.59-0.73]) were protective. Conclusion: Prior opioid prescriptions, even without evidence of OUD, greatly increase the chance of receiving a potent opioid, suggesting cumulative risk that warrants further investigation. Males aged 31-65 years are slightly more likely to receive potent opioids. Physicians avoid potent opioids in patients with alcohol and other SUDs, perhaps attempting to mitigate risk. Keywords: opioids, predictors, prescribing LO21 Do geography and emergency department type influence opioid prescribing behaviour and patient outcomes?

G. Innes, MD, MSc, K. Lonergan, BSc, J. Hayward, MD, MPH University of Calgary, Calgary, AB Introduction: Our objective was to describe patients, opioid prescribing patterns and related outcomes in tertiary, regional, rural and urban emergency departments (EDs). Methods: We studied a 10% random sample of visits from all 115 Alberta EDs from 2010 through 2019. We collated demographic and diagnosis data for nonpregnant adults age [ 18, excluding those with cancer or palliative diagnoses. We linked with the Alberta prescription information network to identify opioid prescriptions filled within 3 days of ED visit. Opioid types included codeine, Tramadol or potent opioids (morphine, hydromorphone, oxycodone, fentanyl). Outcomes included patient descriptors, opioid prescriptions and mortality or probable OUD, defined as an ED or hospital diagnosis of opioid use, 10? opioid prescriptions, 120? opioid days, or new opioid agonist therapy within 1-year of the ED visit. We used multivariable regression to adjust for age, sex, EMS arrival, complaint category, prior opioid exposure, alcohol or substance use disorder (AUD, SUD) and mental health (MH) conditions. Results: We studied 644,283 ED visits: 66,412 tertiary, 87,535 regional, 342,212 rural, 148,124 urban. Age, sex and chief complaints were similar across ED types. Tertiary sites had more EMS arrivals (23%) than regional, rural or urban sites (9%, 4%, 14%) and more high-acuity CTAS 1-2 patients (25% vs. 8%, 4%, 20%). Tertiary sites saw more AUD (11% vs 5%, 4%, 5%), SUD (9%, 4%, 3%, 4%) and MH patients (14%, 10%, 9%, 9%). Opioid prescription rates were similar across ED types (12%%, 13%, 12%, 12%), and most received codeine (60%, 57%, 62%, 63%). Tertiary and urban sites prescribed Tramadol less often (11%, 22%, 19%, 11%) and potent opioids more often (30%, 21%, 20%, 26%) . In patients without pre-existing OUD, 1-year OUD outcomes were similar across ED types (3%, 3%, 3%, 2%) and all-cause mortality was higher for tertiary and rural sites (7%, 5%, 7%, 5%) . Adjusted analyses showed that new OUD outcomes were more frequent in tertiary than regional, rural or urban sites (adjusted OR = 1.0 vs 0.85, 0.80,0.79) and all-cause mortality was also higher for tertiary sites (aOR = 1.0 vs 0.92, 0.84,0.73). Conclusions: Tertiary EDs deal with more high-risk OUD and SUD patients and prescribe more potent opioids. Outcomes after opioid prescription appear worse for patients attending tertiary sites, but this may reflect residual confounding by intention and the fact that very few deaths in this population were opioid-related. Keywords: opioid, pain, substance use disorder LO22 Does sex influence opioid prescribing and opioid-related outcomes in emergency department patients?

G. Innes, MD, MSc, J. Hayward, MD, MPH, K. Lonergan, BSc University of Calgary, Calgary, AB Introduction: Our objective was to describe the association between sex, opioid prescribing and opioid-related outcomes in emergency department (ED) patients. Methods: We studied a 10% random sample of visits from all 115 Alberta EDs from 2010 through 2019. We extracted data for nonpregnant adults age [ 18, excluding those with palliative or cancer diagnoses. We linked with the Alberta prescription information network to identify medication profile at the time of ED visit and new opioid prescriptions filled within 3 days of the visit. Our primary outcome was probable opioid use disorder (ED or hospital diagnosis of opioid use, 10? opioid prescriptions or 120? opioid treatment days or opioid agonist therapy) within 1-year of ED visit. Secondary outcomes included opioid type and dose prescribed and mortality by sex. Opioid types included codeine, Tramadol or potent opioids (morphine, hydromorphone, oxycodone, fentanyl). We used multivariable regression to adjust for age, prior opioid exposure, substance use, mental health disorders and opioid type prescribed. Results: We analyzed 644,283 ED visits, mean age 46 years, 51% female. At baseline, females were more likely (11 vs 9%) to meet OUD criteria but less likely to have alcohol use (4% vs 7%) or other substance use disorders (SUD, 3% vs 5%). 11.6% of males and 12.1% of females received (were exposed to) an opioid prescription. Females more often received Tramadol (17% v 16%) or a potent opioid (24% v 21%); males more often received codeine (63% v 59%). Both sexes received a median 5-day opioid supply (IQR, 3-10) and both received a median total morphine milligram equivalent dispense of 135 mg (IQR, . At one-year, the proportion of females/males with OUD was 11% vs 9% (same as index), and mortality was 6% vs 7.8%. Among 579,597 patients without OUD at their index visit, 9% vs 3% of exposed/unexposed females, and 7% vs 2% of exposed/unexposed males met OUD criteria during the follow-up year. Mortality rate was 5% for exposed and unexposed females, and 6% vs 7% for exposed/unexposed males. Adjusted analysis showed that males have a higher risk of new OUD (adjusted OR = 1.06; CI 1.02-1.09) and mortality (aOR = 1.6; CI 1.6-1.6) although deaths in this study were rarely opioid related. Conclusion: Opioid prescribing differed slightly by sex. ED opioid prescriptions were associated with substantial increases in new OUD for both sexes (NNT * 20). Males had higher mortality, but few deaths were opioid-related. Keywords: analgesia, opioids, substance use disorder LO23 Do different opioids carry different risk for emergency department patients?

G. Innes, MD, MSc, K. Lonergan, BSc, J. Hayward, MD, MPH University of Calgary, Calgary, AB Introduction: Our objective was to determine whether emergency department (ED) patients exposed to opioid prescriptions have differing risk of adverse outcomes based on opioid type. Methods: We studied a 10% random sample of all Alberta ED visits from 2010 through 2019. We extracted demographic and diagnosis data for adults age [ 18, excluding those with active cancer, palliative care, opioid use disorder (OUD) or long-term opioid use (LTU) . We linked to the AHS prescription network to identify patient medications at the time of ED visit and opioid prescriptions filled within 3 days after. Opioid types included codeine, Tramadol or potent (morphine, hydromorphone, oxycodone, fentanyl) . Outcomes included all-cause mortality or probable OUD, defined as ED or hospital diagnosis of opioid use, 10? opioid prescriptions, 120? opioid treatment days, or opioid agonist therapy within 1 year. We adjusted for age, sex, EMS arrival, complaint category, prior alcohol use, substance use or mental health condition. Results: We studied 546,647 eligible patients, 51% female, mean age 45. Of these, 497,290 (91.0%) received no opioid prescription, 31,708 (5.8%) received codeine, 10,941 (2%) received Tramadol and 6708 (1.2%) received potent opioids. Death occurred in 5.9%, 5.1%, 4.4% and 9.0% of these patients, although few deaths were opioid-related. During follow-up, probable OUD criteria were met in 2%, 7%, 6% and 16% of patients receiving no opioid, codeine, Tramadol or potent opioid. NNT to cause an OUD outcome was 20, 25 and 7 for codeine, Tramadol and potent opioids. After multivariable adjustment, the strongest predictors of an OUD outcome (adjusted odds ratio) were prescription of a potent opioid (aOR = 3.4; 95% CI 3.1-3.7), concurrent alcohol or substance use disorder (aOR = 1.7; CI 1.6-1.7), concurrent psychoactive medication (aOR = 1.5; CI 1.4-1.5) and musculoskeletal pain complaint (e.g. back pain; aOR = 1.2; CI 1.1-1.3). Opioid naïve patients (no opioid prescription within a year preceding their ED visit) were at substantially lower risk (aOR = 0.2; CI 0.2-0.2). Potent opioid recipients had the highest rates of subsequent LTU, ED visits and hospitalizations, while Tramadol had the most favorable safety profile. Conclusions: Different opioids have different risk profiles. In this large populationbased dataset, potent opioids were associated with substantially higher rates of death, OUD and opioid-related hospitalizations while Tramadol had the most favorable risk profile. Keywords: analgesia, opioids, substance use disorder Introduction: Analgesia with fentanyl can be associated with hyperalgesia (higher sensitivity to pain) and can contribute to escalating opioid use. Our objective was to assess the relationship between emergency department (ED) acute pain management with fentanyl compared to other opioids, and the quantity of opioids consumed twoweek after discharge. We hypothesized that the quantity of opioids consumed would be higher for patients treated with fentanyl compared to those treated with other opioids. Methods: Patients were selected from two prospective cohorts assessing opioids consumed after ED discharge. Patients C 18 years treated with an opioid in the ED for an acute pain condition (B 2 weeks) and discharged with an opioid prescription were included. Patients completed a 14-day paper or electronic diary of pain medication use. Quantity of 5 mg morphine equivalent tablets consumed during a 14-day follow-up by patients treated with fentanyl compared to those treated with other opioids during their ED stay were analyzed using a multiple linear regression and propensity scores. Results: We included 707 patients (mean age ± SD: 50 ± 15 years, 47% women) in this study. During followup, patients treated with fentanyl (N = 91) during their ED stay consumed a median (IQR) of 5.8 (14) 5 mg morphine equivalent pills compared to 7.0 (14) for those treated with other opioids (p = 0.05). Results were similar using propensity score sensitivity analysis. However, after adjusting for confounding variables, ED fentanyl treatment showed a trend, but not a statistically significant association with a decreased opioid consumption during the 14-day follow-up (B = -2.4; 95% CI -5.3 to 0.4; p = 0.09). Conclusion: Patients treated with fentanyl during ED stay did not consume more opioids after ED discharge, compared to those treated with other opioids. If fentanyl does cause more hyperalgesia compared to other opioids, it does not seem to have a significant impact on opioid consumption after ED discharge. Keywords: opioid, pain Introduction: It remains unclear whether unresponsive patients with witnessed opioid overdose require rescue breathing or chest compressions. We investigated whether ventilation and naloxone without chest compressions provide effective resuscitation of apneic, unresponsive patients with witnessed opioid overdose. Methods: At a single community supervised consumption site (Insite), we conducted a retrospective chart review and administrative database linkage of consecutive patients who were unresponsive and apneic following witnessed opioid overdose between January 1, 2012 and December 31, 2017. We recorded demographics, resuscitative interventions including chest compressions, ventilation, oxygen and naloxone administration, and outcomes including neurological status and mortality. We linked Insite visits with hospital records to define the entire care episode, which concluded when the patient was discharged from Insite, emergency department, or hospital admission; or died. The primary outcome was successful resuscitation: alive and neurologically intact (ambulatory and speaking coherently, or alert and oriented, or Glasgow Coma Scale 15) at the conclusion of the care episode. Secondary outcomes included mortality and predefined complications of resuscitation. Results: We collected data on 767 patients, with a median age of 43 and 81.6% male, and obtained complete follow up on 763 patients (99.5%). No patients underwent chest compressions. All patients with complete follow up were alive and neurologically intact at the end of their care episode. Seventeen patients (2.2%) developed complications, with one resultant hospital admission. Conclusion: At a community supervised consumption site, all unresponsive, apneic patients with witnessed opioid overdose were successfully resuscitated with oxygen and/or naloxone. No patients required chest compressions. Keywords: harm reduction, opiate overdose, ventilation LO26 A quality improvement project to improve emergency department hip fracture pain management using the fascia iliaca compartment block Background: Hip fractures are painful injuries that disproportionately affect elderly patients and are associated with significant morbidity and mortality. Regional anesthesia (nerve blockade) is an effective opioid-sparing pain management strategy for emergency department (ED) hip fracture patients, with few contraindications. Of the regional anesthetic techniques described, the fascia iliaca compartment block (FICB) is favoured for its efficacy and safety profile. Despite this knowledge, use of FICB remains low.To evaluate the impact of an educational package on rates of FICB for hip fracture pain management in the ED, with a goal of achieving FICB use in greater than 50% of cases within 2 years. Aim Statement: A preintervention cohort of hip fracture patients managed over a 12-month period (Sept 30, 2019-Sept 30, 2020) was used to identify practice patterns. A survey of physicians (Oct 2020) assessed hip fracture pain management practices as well as training and comfort in performing regional anesthesia. Using a PDSA framework, serial educational interventions to address FICB knowledge gaps were developed and delivered (Sept 2020-Apr 2021) including online articles and videos, lectures, and simulations. Additionally, to reduce logistical and time management barriers, a FICB procedure kit was created and made available for use in a procedural cart. Standardized post-intervention data from hip fracture patients assessed for change in practice patterns (May 1, 2021 -Nov 1, 2021 . Additionally, a follow up survey (Aug 2021) assessed for change in physician attitudes towards use of FICB. Measures and design: Pre-intervention, 57 hip fracture patients were managed compared to 21 cases post-intervention. FICB rates increased from 3.5 to 33%. No complications from FICB were recorded. Survey results demonstrated a shift in physician attitudes towards viewing the FICB as a more important part of hip fracture pain management, increased satisfaction with level of training, and increased confidence in performing FICB. Evaluation/results: We describe a QI project that increased use of the FICB in ED hip fracture patients. Further PDSA cycles will include on-shift training with physician experts and improvements to our FICB procedure documentation form. The impact of these changes will be monitored via ongoing chart review. The educational interventions used in this QI study can be applied to other EDs to improve the quality of care hip fracture patients receive. Keywords: fascia iliaca compartment block, hip fracture, quality improvement and patient safety Introduction: Early and adequate administration of analgesia is an essential part of treating vaso-occlusive episodes (VOEs) in patients with sickle cell disease (SCD). Unfortunately, patients with SCD frequently report significant delays to analgesia in emergency departments (ED); they also describe discrimination and lack of empathy from health care providers who perceive them as ''drug seeking''. In this study, we compare the time to initial analgesia for patients with VOEs to that of patients with acute renal colic (RC) pain. Methods: This retrospective cohort study included ED visits from January 2019 to January 2020 for patients aged C 18 years with a diagnosis of sickle cell pain or renal colic with a recorded pain score C 1 and received analgesia. Outcome measures included times from triage to: (1) the first analgesic of any kind, and (2) opioid administration. Linear regression was used to estimate differences in outcome measures between disease cohorts adjusting for sex, age, ED site, ED arrival mode, Canadian Triage Acuity Scale level, arrival pain score, order set use, initial analgesia including IV route, opioid administration prior to MD assessment, and daily ED volume. Results: In total, 233 SCD (n = 100 patients) and 452 RC (n = 400 patients) ED encounters were analyzed. Patients with SCD were on average younger (SCD vs RC; 31.2 vs 49.5 years, p \ 0.0001), more likely to be female (52.0 vs 36.8%, p = 0.0053), and had a higher average number of ED visits in the last year (3.1 vs 0.5 visits, p \ 0.0001) compared to RC patients. After adjusting for confounders, patients with SCD were found to wait on average 35 min longer to receive their first analgesic compared to patients with RC (R2 = 0.92, F(1, 569) = 13.5, p = 0.0003). However, when this comparison was limited to ED visits in which opioids were administered, there was no difference between SCD and RC encounters (R2 = 0.40, F(1, 370) = 2.64, p = 0.105). Interestingly, non-opioids were less likely to be administered in SCD patient encounters (SCD vs RC; 54 vs 85% of visits, p \ 0.0001), and less frequently given as the first analgesic dose (18 vs 72% of visits, p \ 0.0001). Conclusion: Patients with SCD received analgesia later than those with RC. Patients with RC were more likely to receive non-opioids as their first analgesic than patients with SCD. Our findings suggest that there are barriers to receiving timely access to opioids in the ED which can lead to significant delays in care for patients with SCD. Keywords: sickle cell disease, time to analgesia LO28 Optimal dose of intranasal dexmedetomidine for laceration repair in children: A phase II dose-ranging study using the continual reassessment method H. Stevens, BA, A. Chapamn, BSc, N. Evans, BSc, C. Mace, BSc, C. Creene, BSc, S. Meulendyks, BN, N. Poonai, MD, MSc London Health Sciences Centre, London, ON Introduction: Lacerations are the most common paediatric injury presenting to an emergency department (ED). Suture repair can be distressing, especially for young children who commonly resist positioning and often require restraint and no effective anxiolytic currently exists. Recent evidence suggests intranasal dexmedetomidine (IND) may be effective for reducing procedural anxiety in children. Our objective was to determine the optimal dose of IND for children undergoing laceration repair. Methods: We report the interim results of a phase II, dose-ranging study using the Bayesian continual reassessment method. Children 0-17 years with a single, isolated laceration (\ 5 cm), requiring single layer closure with sutures were administered IND in increasing doses ranging from 1 to 4 mcg/kg (maximum 200 mcg), using a mucosal atomizer device. The primary outcome was sedation using the Paediatric Sedation State Scale (PSSS). The PSSS was scored every 30 s by two independent outcome assessors using video. PSSS scores of 0 or 1, 2 or 3, and 4 or 5 indicate ''over '', ''adequate'', and ''under'' sedation, respectively. Using complete case analysis, we reported the proportion of participants that achieved ''adequate'' sedation for [ 90% of the time from sterile preparation to tying of the last suture. Secondary outcomes included adverse events (AEs) and post procedure length of stay (ppLOS). Results: Forty-three of 55 planned participants have been enrolled. One participant was withdrawn prior to scoring because the laceration required operative repair. The median (IQR) age and number of sutures was 4 (3,6) and 3 (2,4), respectively. All participants received lidocaine-epinephrine-tetracaine topical anesthetic. The proportion of participants that were adequately sedated using 1, 2, 3, and 4 mcg/kg IND were 1/3 (33%), 2/9 (22%), 5/9 (56%), and 12/21 (57%), respectively. There was one AE, a self-resolving decrease in oxygen saturation to \ 90%, lasting 30 s. The participant received 4 mcg/kg and required no resuscitative interventions. The median (IQR) ppLOS for 1, 2, 3, and 4 mcg/kg IND were 67 (60,78), 76 (60,100), 106 (76,146) , and 113 (76,150) minutes, respectively. All participants complied with intranasal administration. Conclusion: Our interim results suggest that IND is safe, well tolerated, and 3 mcg/kg balances optimal sedation and ppLOS. Additional participants will receive 3 mcg/kg and although results are pending, future studies of IND in children should consider using this administration regimen. Keywords: dexmedetomidine, pediatric, sedation Introduction: Fever among infants in the first months of life remains one of the most common problems in pediatric healthcare, and * 10% harbour serious bacterial infections. Management must balance risks of under-diagnosis and harms of over-investigation, and there exists significant variation in care. Management decisions should incorporate parental preferences through a process of shared decisionmaking. Very little is known regarding parental preferences for the management of febrile young infants. We sought to describe parental attitudes regarding shared decision-making for febrile infants and to identify aspects of care most important from the family perspective. Methods: We undertook a prospective study of parents presenting with well-appearing febrile infants aged B 60 days to an urban tertiary pediatric ED between May/2020-October/2021. Following hospital discharge, parents of all eligible infants were invited to participate in structured telephone questionnaires exploring their expectations, goals of care, and desired level of decisional involvement. Questionnaires were piloted with parent partners for clarity and used multiple-choice and 5-point Likert-scales. Results: During the 18-month study, 221/301 (73%) eligible families completed postdischarge questionnaires. Among responding families, 37% of infants were aged 0-28 days, 100% underwent blood and urine testing, 48% had a lumbar puncture, 54% were hospitalized and 53% received parenteral antibiotics. At ED presentation, parents anticipated that their infant would have a physical examination (93%), blood (85%) and urine testing (86%); whereas fewer anticipated the need for hospitalization (34%), parenteral antibiotics (32%) and lumbar puncture (15%). A majority felt they received enough information regarding testing/treatment (86%), risks (79%) and benefits (86%); 90% wanted to be involved in treatment decisions. Having their expectations met was the single most important aspect the ED experience (33%), whereas involvement in decision-making was rarely selected (3%). Diagnostic testing was the most frequently selected stressor (32%). Conclusion: Management of febrile young infants is not always aligned with parental expectations. While parents wish to be involved in decision-making, setting appropriate expectations appears more important to parental satisfaction. Findings are an essential first step towards identifying barriers and facilitators and establishing family-centered outcomes for a shared-decision model of febrile infant care. Keywords: fever, pediatric, serious bacterial infections University of Alberta, Edmonton, AB Introduction: Caregiver hesitancy for their children to receive the COVID-19 vaccine remains due to concerns regarding safety and efficacy, but also due to fear of vaccine administration-related pain and distress. Study objectives were to determine caregivers' perceptions regarding both their personal and child's COVID-19 vaccine administration-related stress and fear and relate this to their likelihood to allow their child to receive COVID-19 vaccinations. Methods: This study was a secondary data analysis of a multicenter, crosssectional survey of caregivers presenting to four Canadian pediatric emergency departments. Caregivers were surveyed between December 2020 and March 2021 and completed a digital survey on their own smartphones. Introduction: Pulmonary embolism (PE) investigation in the emergency department (ED) is especially challenging for patients with suspected SARS-CoV-2 infection due to symptom overlap, increased pro-thrombotic risk, and uncertain D-dimer test interpretation in this population. The primary objective of this study was to assess the diagnostic accuracy of both age-adjusted and standard D-dimer test thresholds for predicting 30-day PE diagnosis in patients with suspected SARS-CoV-2 infection. Methods: This retrospective observational study utilized data from the Canadian SARS-CoV-2 Emergency Department Rapid Response Network (CCEDRRN) for ED visits between Mar 1, 2020 and July 2, 2021. CCEDRRN data came from 50 ED sites across eight Canadian provinces. Consecutive adult ([ 18 years) ED patients with SARS-CoV-2 testing performed at index visit and with any of the following prespecified symptoms were included: chest pain, shortness of breath, hypoxia, syncope/ presyncope, and hemoptysis. Patients were excluded if they had duplicate records or no valid provincial healthcare number. 30-day PE diagnosis was defined as positive computed tomography for pulmonary embolism (CTPE) or ED/hospital diagnostic code for PE within 30-days of index visit. Diagnostic accuracy for prespecified age-adjusted and standard D-dimer thresholds (500-5000 ng/mL) was determined for subgroups stratified by SARS-CoV-2 test result (positive vs negative). Results: A total of 52,038 patients were included. No patients were excluded. Overall, age-adjusted D-dimer (SN 96%, 95% CI 93-98%; SP 48%, 95% CI 48-49%) performed comparably to the most sensitive D-dimer threshold of 500 ng/mL (SN 98%, 95% CI 96-99%; SP 41%, 95% CI 40-42%). Other Ddimer thresholds did not achieve acceptable diagnostic performance (SN \ 90%). D-dimer performed better in SARS-CoV-2 negative patients as compared to SARS-CoV-2 positive patients for predicting 30-day PE diagnosis (c-statistic 0.88 vs 0.80). Conclusion: In this large Canadian cohort of ED patients with suspected SARS-CoV-2 infection, an age-adjusted D-dimer strategy performed comparably to a conservative 500 ng/mL D-dimer test threshold for ruling out 30day PE diagnosis, but with better ability to avoid unnecessary CTPE studies due to higher specificity. Among ED patients with suspected SARS-CoV-2 infection who require PE investigation, an age-adjusted D-Dimer strategy appears to be superior irrespective of final SARS-CoV-2 status. Keywords: diagnostic accuracy, pulmonary embolism, SARS-COV-2 LO32 Treatments, resource utilization and outcomes of COVID-19 patients presenting to emergency departments across pandemic waves evolved between pandemic waves. Our objective was to compare treatments, acute care resource utilization, and outcomes of COVID-19 patients presenting to Emergency Departments (EDs) across pandemic waves. We hypothesized that steroid use would increase and invasive mechanical ventilation would decrease in the second wave compared to the first. Methods: This observational cohort study enrolled consecutive eligible COVID-19 patients presenting to 46 EDs participating in the Canadian COVID-19 ED Rapid Response Network (CCEDRRN) between March 1 and December 31, 2020. We collected data by retrospective chart review. We previously validated retrospectively collected data elements against prospective data collection. A central coordinator monitored consecutive enrolment to minimize selection bias and verified data for quality. Our primary outcome was in-hospital mortality. Secondary outcomes were treatments, hospital and ICU admissions, ED revisits and readmissions. We used logistic regression modelling to assess the impact of pandemic wave on outcomes. Results: We enrolled 9,967 patients in 8 provinces, 3,336 from the first and 6,631 from the second wave. Patients in the second wave were younger and fewer met criteria for severe COVID-19 compared to the first wave. In the second wave, a greater proportion of patients were discharged from EDs (61.3% versus 47.2%, p \ 0.0001), but a higher proportion revisited the ED within 7 days (6.9% versus 5.8%, p = 0.025). Adjusted for patient characteristics and disease severity, steroid use increased (odds ratio [OR] 7.4; 95% confidence interval [CI] 6.2-8.9), while invasive mechanical ventilation decreased (OR 0.5; 95% CI 0.4-0.7) in the second wave. In the second wave, patients were intubated later in their hospital course (3.2 versus 2.0 days, p \ 0.0001) and for a shorter duration (12.8 versus 16.4 days, p = 0.018). After adjusting for differences in patient characteristics and disease severity, the odds of hospitalization (OR 0.7; 95% CI 0.6-0.8) and critical care admission (OR 0.7; 95% CI 0.6-0.9) decreased, while mortality remained unchanged (OR 0.7; 95% CI 0.5-1.1). Conclusion: Our study provides real-world evidence of rapid uptake of steroids in hospitals. Clinicians were able to discharge more ED patients with a slight increase in revisits. Clinicians were able to manage patients safely using less invasive mechanical ventilation, and fewer hospital and critical care resources in the second wave. Keywords: coronavirus disease 2019, outcomes, resource utilization. Introduction: Non-pharmacological interventions (NPIs) such as masking and physical distancing have effectively stemmed the spread of COVID19. However, their efficacy in preventing other virallytriggered respiratory illnesses (VRIs) is less well understood. The aim of the study was to evaluate the impact of NPI implementation at the onset of the COVID19 pandemic on VRI-related ED visits and hospitalizations. Methods: This study encompassed 459,664 ED visits and 154,185 hospitalizations across all acute care facilities in Alberta. ICD-10 codes were used to identify patients with six common VRIs expected to be sensitive to NPIs measures: influenza, asthma, community-acquired pneumonia, COPD, emphysema, and bronchitis. Three conditions not expected to respond to NPI measures were included as controls: heart failure, bowel obstruction, and acute appendicitis. Difference-in-differences analysis was performed, with NPI implementation at the onset of the pandemic serving as the interruption timepoint. The primary outcome of interest was healthcare utilization (ED visits and hospitalizations) for VRI vs. control conditions between the pre-pandemic (Feb 2015-2020) and pandemic (Mar 2020-2021) periods. Multivariable regression models were constructed to estimate the cost-avoidance associated with NPI implementation. Results: Following NPI implementation, weekly ED visits and hospitalizations for VRIs declined by 58% (P \ 0.001) and 48% (P \ 0.001), respectively. No concomitant change was observed in triage acuity or comorbidity index scores. In contrast to VRIs, healthcare utilization for controls remained consistent between the two periods, with a 1% and 3% decrease in ED visits and hospitalizations. The decrease in VRI-related healthcare utilization resulted in a $112 million reduction in resource use. After accounting for COVID19-related visits, there was still a net reduction in annual ED visits (n = 33,254) and hospitalizations (n = 1,602) for total respiratory conditions during the pandemic period. Conclusion: NPI implementation was followed by a significant decrease in ED visits and hospitalizations for VRIs but not control conditions. The observed decrease in resource utilization with no increase in acuity indicates that reductions were primarily driven by NPI implementation rather than an avoidance of healthcare settings by low-acuity patients. These findings suggest that NPI use may be an effective preventative measure for VRIs that can reduce health resource utilization and improve patient outcomes.

Keywords: COVID-19 pandemic, non-pharmacologic interventions, respiratory illnesses

Temporal association between COVID-19 lockdowns and changes in trends of mental health visits to emergency departments G. Matskiv, BSc, MPH, L. Rivera, MD, MPH, C. Lee, PhD, J. Nicol, MD, R. Grimminck, MD, S. George, BHSc, MSc, E. Lang, MD University of Calgary Cumming School of Medicine, Calgary, AB Introduction: The COVID-19 pandemic and lockdowns had a unique effect of creating access barriers while increasing mental health strains. Emergency departments (EDs) serve as a safety net for patients with mental health and substance use (MHSU) concerns. Changes in MHSU visits may serve as indicators of mental health strains and may inform future planning. This study examined the temporal association of lockdowns and trends of MHSU visits. Methods: Data for MHSU visits of all ages from Alberta's 192 EDs was obtained for a period of 24 months starting on January 1, 2019 with MHSU diagnoses spanning 14 different ICD-10 categories. We also obtained weekly counts for all ED visits to calculate the proportions of MHSU presentations. The primary objective was to examine the MHSU visit trends for changes around the lockdown dates. Secondary objective was to assess the burden of MHSU visits on EDs by examining the changes in proportions of MHSU visits. We explored the change of trend level and shift by detecting breakpoints and comparing them to the lockdown dates using ARIMA modelling. Results: The first significant change in weekly MHSU visits was around March 8th, 2020 (week 62, 95% CI 56-63), with a 16.5% drop, from 2012 to 1681 visits. The second change was on June 14th (week 72, 95% CI 72-96) with an increase from 1681 to 1867 visits. The trend of proportion of weekly MHSU visits in relation to all ED visits changed significantly on March 15th (week 63, 95% CI 62-65) with an increase from 4.6% to 5.8%. Second change was on July 5th (week 80, 95% CI 72-84) with a decrease to 5.4%. There was no other significant change in visits or proportions during the included timeline. The actual dates for the first lockdown initiation (March 15) and easing (May 15) fell within the 95% confidence intervals for both analyses. There was no trend change during the second lockdown initiation (November 13). Conclusion: While the number of weekly MHSU ED visits showed an immediate decrease following the first lockdown initiation, the proportion of MHSU visits increased modestly around the same time. Surprisingly, no significant change was detected during the second lockdown initiation, perhaps due to better access or patients' improved resilience amid the COVID-19 pandemic. The common notion that lockdowns have severely exacerbated mental health concerns does not appear to be supported by data from the EDs, however this could be related to the ED access barriers created by the fears surrounding COVID-19. Keywords: COVID-19, mental health Introduction: Previous literature suggests that the provision of analgesia and sedation is often delayed or inadequate following endotracheal intubation in the ED. Specifically, current best practices support an analgesia-first approach to minimize negative outcomes associated with poor pain control. The objective of this study was to review practice patterns in the provision of post-intubation analgesia and sedation in a multi-site community hospital emergency department (ED). Methods: A retrospective chart review was conducted to identify all ED intubations from April 2015 until December 2019. Inclusion criteria consisted of all patients intubated by means of rapid sequence intubation (RSI) in the ED, patients remaining in the ED for at least 30 min post intubation and surviving to hospital admission. Data analysis was performed by frequency to determine trends in the type of analgesic and sedative agent used post-intubation, and timing of administration. Results: 750 charts were reviewed with 487 included in the analysis after exclusion criteria and charts with insufficient data were removed. Wide variability was seen in the choice of sedative used for induction, the most common being propofol in 23% of intubations. A post-intubation sedative was given in 85% of patients at a median time of 12 min after induction with 52% of total patients receiving propofol. Only 40% of patients received a type of analgesic post-intubation administered at a median time of 26 min following induction, the most common agent being fentanyl. Rocuronium was administered as a paralytic agent in 55% of cases, succinylcholine in 27%, a combination in 1% and no paralytic in 16% of cases. 47% of those who received post-intubation sedation had hypotension documented at some point. ED length of stay postintubation was a median time of 153 min. Conclusion: Current literature and best practice point to the benefits of an analgesia-first model of post-intubation care. The current review was a first step in understanding current practice and adherence to these standards, with a future goal to facilitate a quality improvement approach that enhances care for ED intubated patients. This study signals that postintubation analgesia is insufficient and delayed in the majority of cases which warrants further review and discussion to improve patient care. Keywords: analgesia, emergency department, sedation Introduction: Patients with acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are frequently discharged from the emergency department (ED) and treated with antibiotics. The role of antibiotics in the outpatient management of AECOPD is controversial and has never been studied in the ED setting. The objective of this study was to determine whether treatment with antibiotics is associated with lower rates of return to hospital with admission in patients with AECOPD who are discharged from the ED. Methods: We conducted a secondary analysis of prospectively collected data from the validation study of the Ottawa COPD Risk Scale. We included adult patients with AECOPD who were discharged from six tertiary care emergency departments in Canada over a two-year period. The primary outcome was rate of return to hospital with admission for respiratory symptoms within 14 days of discharge from the ED. To examine the association between antibiotic treatment and rehospitalization, we performed multivariable logistic regression. Results: A total of 774 patients were included in the analysis. The mean age was 69.4 years, 50.1% of patients were female, 44.7% of patients arrived by ambulance, and 58.3% of patients were discharged with antibiotics. Twenty-nine (6.4%) patients discharged with antibiotics and 36 (11.1%) patients discharged without antibiotics returned to hospital with admission within 14 days of ED discharge. In the multivariable logistic regression analysis, treatment with antibiotics was significantly associated with lower return to hospital with admission (OR 0.56; 95% CI 0.31-1.00; p = 0.049). Male sex, arrival by ambulance, home oxygen use, chronic kidney disease, and higher COPD risk score were all independently associated with higher rates of return with admission. Conclusion: For patients with AECOPD who are discharged from the ED, we found that treatment with antibiotics was associated with lower rates of rehospitalization. The results of this study support the use of antibiotics in the outpatient management of AECOPD. Keywords: antibiotics, chronic obstructive pulmonary disease, rehospitalization LO37 Paramedic assessment of low-risk trauma patients using the Canadian C-Spine Rule and selective transport without immobilization: Are there inequities in health services? Introduction: Health inequities are defined as unfair systematic differences in health outcomes. We sought to determine the differential impact of a strategy authorizing paramedics to assess low-risk trauma patients using the Canadian C-Spine Rule (CCR) and selectively transport them without immobilization across various patient factors. Methods: We completed pre-specified subgroup analyses within a stepped wedge cluster randomized trial to evaluate the use of the CCR by paramedics. The trial took place in 11 Ontario communities (3 clusters) where paramedics evaluated all consecutive eligible stable adult patients at risk of c-spine injury using a standardized form. Trial co-primary outcomes were immobilization rate, pain and discomfort scores (10-point ordinal scale completed when transferring care to emergency staff) and (secondary) use of diagnostic imaging. We explored differential treatment effects by age (dichotomized \ 65 and C 65), sex, and any language barrier noted by paramedics (yes/ no). We used random effects logistic regression accounting for intracluster and intra-period correlation, reporting adjusted odds ratios (aOR) with 95% Confidence Intervals (95% CI). Results: We evaluated 3,625 patients: age \ 65 78.5%, female 51.6%, and language barrier 6.5%. The CCR safely reduced the proportion of patients transported with immobilization in those \ 65 (aOR 0.29; 95% CI 0.16-0.52) but not among those C 65 (aOR 0.84; 95% CI 0.43-1.64). Pain, discomfort, and diagnostic imaging was similar in both age groups. Females received less diagnostic imaging (aOR 0.54; 95% CI 0.39-0.75) vs. males (aOR 0.80; 95% CI 0.58-1.10). Immobilization, pain, and discomfort were similar in both sex groups. Only patients without a language barrier had lower immobilization rates (aOR 0.32; 95% CI 0.17-0.59), pain (aOR 0.66; 95% CI 0.50-0.88), discomfort (aOR 0.57; 95% CI 0.44-0.74), and diagnostic imaging (aOR 0.62; 95% CI 0.47-0.82). Conclusion: While older patients were more likely to be immobilized, age is a non-modifiable and justifiable element of the CCR. Differential use of imaging across sex groups could either be from inequitable/biased risk assessment in the ED or from unequal distribution of other factors such as pattern of injury. Language barrier did result in health inequities for all outcomes, and efforts should be made to mitigate its impact through language training or use of translation tools or services.

Keywords: Canadian C-Spine Rule, prehospital, randomized controlled trials

Descriptive analysis of patients with high use of ambulance services in Southern Ontario emergency departments I. Bielska, PhD, K. Cimek, MSc, K. Bath, BSc, MPH, L. Zhang, MD, MSc, G. Agarwal, PhD, B. McLeod, MSc, MHM, R. Ferron, BEd, MHM, J. Tarride, PhD McMaster University, Kitchener, ON Introduction: Research shows that individuals who have frequent use of the emergency department (ED) tend to have greater utilization of health care resources, including of emergency medical services. However, cohort studies investigating this phenomenon are not available. The purpose of this study was to determine the characteristics and health care utilization patterns of patients who frequently arrive in the ED via ambulance in Southern Ontario over eight years. Methods: Using data from the National Ambulatory Care Reporting System and the Discharge Abstract Database obtained from Integrated Decision Support, a descriptive analysis was performed on a population of individuals residing in the Hamilton Niagara Haldimand Brant Local Health Integration Network. To be included, individuals had to have C 5 ambulance arrivals to the ED during one or more fiscal years between 2012/13 and 2019/20. Information on patient profiles and ED utilization was analyzed and presented descriptively. Results: There were 12,556 individuals who used ambulance services C 5 times per year, representing 215,353 arrivals (range: 5-1,073). In total, these patients had 391,222 ED visits (55% via ambulance) of which 31% had a CTAS score of 1-2. 23% of the ED visits resulted in admission. The average age of the patients was 65 years (SD: 22) and 51% were female. Among the cohort, 9,380 patients (75%) had 1 year of high use of ambulance services, 1,841 patients (15%) had 2 years of high use, while 1,335 individuals had 3 or more years of high use. 9% of the patients resided in rural areas and 15% had a history of homelessness. The most frequent ICD-10-CA discharge diagnosis categories were: symptoms, signs, and abnormal clinical findings (27%), mental and behavioural disorders (13%), and diseases of the circulatory system (10%). A majority of the patients (72%) had C 4 chronic conditions, including arthritis (47%), renal failure (40%), diabetes (37%), and brain injury (35%). A large proportion of the patients received home care services (76%) and 17% were enrolled on coordinated care plans. Conclusion: Even though a relatively small number of individuals have high use of emergency medical services, these patients tend to use these services frequently for over half of their visits to the ED. The findings of this study contribute to the body of knowledge on ambulance service use, providing a Canadian perspective from a geographic region that includes both rural and urban areas. Keywords: emergency department, emergency medical services, high use of health care services.

Adaptation of a patient-reported experience measure instrument to assess and compare the experience of care in emergency departments, walk-in and primary care clinics Université Laval/CHU de Québec, Quebec, QC Introduction: To reduce the number of emergency department (ED) visits by ambulatory patients with acute health concerns, many health jurisdictions have invested in primary care and walk-in clinics, assuming an equivalent or better patient experience. We aimed (1) to adapt and validate a patient-reported experience measure (PREM), in French and English, usable in EDs, and (2) to compare patient experience in these settings. Methods: The PREM was developed by merging two English language questionnaires used in EDs and primary care clinics in Ontario. A panel of experts and patient partners selected the questions relevant to all settings and removed the duplicates. A forward translation/backtranslation process was used to correct any remaining ambiguities. The PREM was pretested on potential users with an online survey. The respondents were asked to assess the PREM's clarity, usefulness, redundancy, content and face validities, and discrimination on a scale of 1 to 9 (1 = strongly disagree to 9 = strongly agree). A prospective cohort of ambulatory patients with acute respiratory conditions was then monitored in an ED and two walk-in clinics in Quebec City. Generalized linear models with propensity scores were used to compare settings and adjust PREM responses for age, sex, gender, ethnicity, comorbidities, patient's perception of illness severity and length of stay (a = 0.05). Results: The final PREM contains 60 questions using 3-to 5-point Likert scales, where appropriate. In the pretest (n = 22) , median responses were all above 7.5. In the pilot study, 32 participants were enrolled in the ED and 51 in the walk-in clinics. The mean age was 38.5 years (SD: 15.9) and 61.4% of the participants were women. Length of stay in minutes (mean [SD] ) was longer in the ED than in the clinics (369.0 [50.0] vs. 68.6 [22.5]; P \ 0.001). A non-significant trend was observed for a poorer experience of being respected and listened to by the physician in the ED (mean adjusted score on a 5-point scale [SD]; 4.33 [0.20] vs. 4.81 [0.13] ; P = 0.07). No statistically significant differences were found between settings in overall experience of care, propensity to recommend the setting where care was provided, and relationship with nurses. Conclusion: We developed and validated a PREM to assess and compare experience of care in EDs, walk-in and primary care clinics. We will now use it in a multicenter study that will start at the end of 2022. Keywords: emergency overcrowding, patient experience LO40 Adaptation and validation of a patient-reported cost measure questionnaire to assess and compare the costs of care in emergency departments, walk-in and primary care clinics Introduction: The Cost for Patients Questionnaire (CoPaQ) is a selfreported measure that estimates the direct (e.g., medication) and indirect (e.g., loss of income) costs of a disease to patients and their families. The aim of this study was 1) to adapt and validate the CoPaQ for use on ambulatory patients with acute health concerns; and 2) to compare the costs borne by these patients when they visit an emergency department (ED), a walk-in clinic or a primary care practice. Methods: Designed for use in a variety of outpatient settings, the CoPaQ was adapted before being used in acute ambulatory care settings such as the ED. The questionnaire was reviewed and modified by a panel of experts which included clinicians, patient partners, researchers and a linguist. Questions referring exclusively to one setting or to chronic diseases were removed or modified to apply to all care settings. Then, the CoPaQ was pretested on potential users via an online survey. The respondents were invited to evaluate the CoPaQ's clarity, utility, redundancy and validity (1 = strongly disagree to 9 = strongly agree). The questionnaire was modified according to the users' comments. The CoPaQ was then piloted on a prospective cohort of ambulatory patients with acute respiratory conditions consulting in 2 walk-in clinics and one ED in Quebec City, Canada. Mann-Whitney U tests were performed to compare the costs borne by patients in ED and walk-in clinics (alpha = 0.05). Results: The final CoPaQ contains 29 questions. All dimensions assessed in the pretest (n = 22) had a median of 8. In the pilot study, 23 patients in the ED and 28 patients in the walk-in clinics completed the CoPaQ. The mean age (SD) of participants was 38.8 (16.5) years and 64.7% were females. The median (IQR) cost per patient (CAD) in the ED was significantly higher than in the walk-in clinics ($532.27 [287.23;1556.95] vs. $111.11 [59.47;266 .02]; P \ 0.0001). Indirect costs (e.g., loss of income) constituted the majority of costs borne by patients and were significantly higher in the ED ($506.31 [234.90;1330.95] vs $96.33 [46.15;198.73] ; P \ 0.0001). Patient education and income were similar between groups. Conclusion: Using an adapted version of a patient-reported cost measure (CoPaQ), this study showed that a visit in the ED is significantly more expensive than a visit in a walk-in clinic. This tool will be used in a multicenter study beginning in October 2022 to evaluate if results remain the same across 17 participating sites in Québec and Ontario. Keywords: costs, patient-reported, questionnaire Introduction: Aortic injuries are among the most frequent fatal injuries following blunt trauma, but literature on that subject is scarce. This study aims to describe the patients' characteristics and the trauma mechanisms associated with an aortic injury in patients who suffered a traumatic cardiac arrest in the prehospital setting or early during their in-hospital care. Secondary objectives were to explore the characteristics of the aortic injuries and to explore the clinical presentation of patients with a fatal aortic injury. Methods: This is a retrospective cohort study of prehospital and early in-hospital (\ 24 h) trauma deaths using the data from the Victorian Ambulance Cardiac Arrest Registry (VACAR) cross-linked with the National Coronial Information System (NCIS) in Melbourne, Australia. Patients who sustained at least one aortic injury (from the autopsy report) were included. Results: Between 2008 and 2014, 2 754 prehospital or early in-hospital deaths related to trauma were recorded. Of those, 245 cases involved a thoracic or abdominal aortic injury (8.9% of all trauma). Patients were mostly males (79.6%) with a mean age of 45.6 years. The most common mechanism was blunt trauma (n = 220, 89.8%) with motor vehicle collisions (n = 149, 60.8%) being the most frequent trauma mechanism. In the out-of-hospital setting, 68 (27.8%) resuscitations were attempted of which only 14 patients (22.1%) had a pulse on arrival. Interventions performed on the field were manual ventilation (n = 50, 73 .5%), endotracheal intubation (n = 31, 45 .6%), thoracostomy (n = 39, 45 .6%), adrenaline administration (n = 35, 52 .9%) and defibrillation (n = 5, 7 .4%). Only 5 patients (2.0%) were ultimately transported to the hospital of which none survived. Isolated aortic disruption was the most common injury (n = 185, 84 .0%) and the descending thoracic area (n = 88, 40 Introduction: Direct transport to a trauma centre for severely injured patients is associated with decreased mortality. Ontario has provincial trauma bypass criteria, which identify patients who should be brought directly to a trauma centre and when to request an air ambulance scene call. Additionally, an air ambulance can be dispatched to the scene before the arrival of land paramedics. Land paramedics can cancel this response if they deem it unnecessary. A small study of Ontario's air ambulance (Ornge) showed 35% of their trauma scene requests were cancelled, yet the reason was unclear over 50% of the time. The primary objective of this study was to determine the reasons for canceling an air ambulance trauma scene response in Ontario. Secondary objectives included estimating how many cancelled scene calls were still transferred to a trauma centre and the associated delay to care. Methods: This was a prospective cohort study using a health administrative database at Ornge from August 2020-August 2021. A new process was established within Ornge where all cancelled scene call requests had to have a reason for cancellation documented. The receiving hospital was also collected. We then performed a probabilistic match using the Patient Transfer Authorization Centre (PTAC) database to look for a patient transported from that receiving hospital to a trauma centre within 12 h of the scene call request. Frequencies were counted and the estimated delay to arrival at a trauma centre was calculated. Results: There were a total of 3147 trauma scene requests during the study period, of which 1258 were cancelled. Reasons for scene call cancellations were: trauma bypass criteria not met (n = 910, 72 .3%), patient brought directly to a trauma centre (n = 187, 14 .9%), patient refused transfer (n = 81, 6.4%) and patient pronounced (n = 80, 6.4%). Of the 910 patients that air ambulance was cancelled from because they did not meet trauma bypass criteria, 184 (20.2%) were later transferred to a lead trauma hospital, with a mean delay of 6.13 h. Conclusion: Most scene call cancellations were due to patients not meeting trauma bypass criteria, yet over 20% of these cancelled calls still resulted in a patient transported to a trauma centre. Future interventions are needed on education, adherence, and modification of our trauma bypass criteria to ensure timely trauma care for injured patients in Ontario. Keywords: prehospital care, trauma, triage Background: Pulmonary embolism (PE) is often diagnosed in the Emergency Department (ED) by computerized tomography pulmonary angiography (CTPA). Excessive scanning contributes to unnecessary radiation exposure, ED crowding, and high costs. The diagnostic yield of CTPA (positive scans/total scans) is lower in North America than in Europe (13% vs 29%). Non-invasive testing, such as D-dimer in the context of a clinical algorithm like the YEARS criteria, to rule out PE may help decrease unnecessary CTPA scans. This has not been widely studied or implemented in Canadian EDs.To improve the diagnostic yield of CTPA to rule out PE by 5% from hospital baseline within 12 months by using the YEARS clinical algorithm. Aim statement: This single centre, multidisciplinary study included all ED patients [ age 18 with a clinical presentation suggestive of PE and investigated with a D-dimer or CTPA between Aug 2020-Jul 2021. The primary outcome was the diagnostic yield of CTPA compared to baseline. Process measures included use of the YEARS criteria as indicated by the percentage of D-dimers ordered and CTPA ordered with a D-dimer to rule out PE. The balancing measure was the number of missed PEs (PE identified on CTPA if Ddimer \ 500ug/L or ED revisit in 30 days with PE on CTPA). Multidisciplinary stakeholders were consulted to develop a YEARS-based diagnostic algorithm. PDSA cycles included a review and departmental discussion of current clinician practice patterns, multimodal clinician education of the YEARS criteria, and algorithm implementation. Run charts tracked outcome measures. Measures and Design: After 3 PDSA cycles over 6 months (Feb-Jul 2021), 1337 patients were investigated for PE vs 1076 in the 6 months prior. A lower proportion of patients had a CTPA performed [483(36%) vs 484(45%)] and the diagnostic yield was higher (15% vs 11%, p = 0.085). D-dimer usage increased (82% vs 72%, p \ 0.001) and a higher proportion of patients were screened with a D-dimer before CTPA (51% vs 38%, p \ 0.001). There were no PEs identified on CTPA in patients with a D-dimer \ 500 or in 30-day ED re-visit. Evaluation/results: Implementing the YEARS criteria safely improved the diagnostic yield of CTPAs, reduced the number of CTPAs completed, and was not associated with missed PEs. Future efforts will include implementation of computer order entry guidance based on YEARS and continued feedback to ensure long-term sustainability of the enhanced diagnostic process. Keywords: computerized tomography pulmonary angiography, quality improvement and patient safety, venous thromboembolism LO44 Implementing a regional emergency department strategy for opioid use disorder in rural and remote hospitals Background: It has been over five years since B.C. declared a public health emergency of illicit drug toxicity deaths. A recent chart review found that 66% of Interior Health (IH) patients who accessed services in the year prior to fatal overdose had their most recent health system encounter in an emergency department (ED). National guidelines recommend initiation of buprenorphine-naloxone (BupNx) as the first line treatment for opioid use disorder (OUD), and the ED represents a strategic opportunity to begin patients on a journey to recovery.The Emergency Services Network has partnered with Mental Health and Substance Use to implement a standardized strategy for the initiation of BupNx and urgent referral to community treatment in 30 EDs across a vast rural and remote region. This strategy will be fully implemented between June 2021-Sept 2022 resulting in 85% of induced patients entering community treatment and 60% of patients retained in treatment at 90 days post-encounter. Aim Statement: We engaged clinicians, researchers, operational leaders and patient advisors in the design of the program. A standardized order set was created with embedded clinical decision support for the identification of OUD, eligibility screening, and induction and titration of BupNx. EDs have been provided BupNx To-Go Packs to bridge patients to community treatment and enable home inductions. A centralized referral service was created to connect patients to community treatment within 72-h of ED encounter, and is supported by a virtual nurse prescriber program. Measures and Design: In the first 6 months we have implemented in 4 urban and 5 rural and remote EDs. We delivered 40 educational sessions delivered to over 420 physicians, nurses, and substance use specialists. 38 patients have begun BupNx treatment through their ED encounter, with 35 patients (92%) successfully transitioned to community treatment in less than 72-h. Patients will be tracked in a pharmacy dispensing database to measure retention in opioid agonist treatment at 30-180 days. Evaluation/ results: We have demonstrated the feasibility of implementing a standardized strategy for the initiation of BupNx and urgent referral to community treatment across a large rural and remote health authority. Dispensing BupNx from the ED, establishing a centralized referral service, and supporting patients with virtual nurse prescribers has removed many barriers to treatment and achieved continuity of care from the acute to community setting. Implementation and evaluation are ongoing. Keywords: opioid agonist treatment, opioid use disorder, quality improvement and patient safety Background: Paramedics call emergency physicians (EPs) for online medical consultation, yet the oversight process is rarely peer-reviewed. From March 2015 to April 2020 there were 9012 patches in Eastern Ontario, yet only 0.5% were reviewed by another EP. In Eastern Ontario in April 2020, paramedics began calling a group of 13 specialty trained Prehospital Transport Medicine physicians on call 24-7. This presented an opportunity to improve physician-paramedic patches and the quality of their review.To construct, implement, and sustain a physician Peer Patch Review (PPR) quality program with review of 100% of patches by November 2021. Reviews involve listening to call recordings and reviewing the physician/paramedic documentation. Aim statement: A patch review form was created with headings: communication, medical order, documentation, system issues, and further review required. Directions were to review 5 calls/month. Outcome measures were the number of peer reviews, cases sent for further review, and identified system issues. Process measures were orders outside of scope/inconsistent with a medical directive, order readback by paramedic, and the concordance between paramedic request and order given. Balancing measures were patch physician satisfaction. Data was presented on a dashboard visible to all patch physicians and run charts. Measures and Design: There were 1038 calls from April 21 to November 19, 2021. Outcome measures showed 589 (56.7%) had a review completed and 5.1% were sent for further review. System issues were identified in 4.6%, mostly due to technology. Process measures showed 1.9% involved orders outside of scope/inconsistent with a medical directive. There was a complete read-back of order in 78.3% of calls. Physicians accepted a paramedic proposed request 49.9%, modified it 24.4%, gave alternate orders 3.7%, and declined the request in 2.2% of calls. The response rate of the satisfaction survey was 77% with 70% of physicians indicating they were very satisfied and 30% satisfied with the new program. Evaluation/results: Our study showed a clear uptake in peer-reviews with high satisfaction. We did not yet achieve 100% review due to a backlog of calls. A peer review provides learning for the reviewer who gains perspective from listening to a colleague's patch, and the reviewee who gets feedback from a review of their call. A PPR program fits with the evolution of prehospital care and a more accountable physician interaction. Keywords: paramedic online medical control, quality improvement and patient safety LO46 Evaluation of a novel, patient-centred, peer support model of care for young adults presenting to the emergency department with mental health and substance use complaints Introduction: Responding to the nearly 90% increase in young adults accessing the emergency department (ED) for mental health and substance use, we implemented the RBC Pathway to Peers (P2P) program. This novel program provides a patient-centred, peer support model of care for young adults presenting to the ED with mental health and substance use complaints. The objective of this study was to describe the patient population using the P2P program and ED staff perception of the program. Methods: We conducted a retrospective chart review of all young adults (16-29 years) who used P2P peersupport services in an urban academic ED (annual census 65,000) between May 2020-Dec 2021. We also invited ED staff (physicians, nurses, physician assistants and nurse practitioners) to complete an online survey containing 19 questions regarding their perceptions and experience working with the peer supporters in the ED. Results: From May 2020 to Dec 2021, 2,383 patients used P2P peer support services. Mean (SD) age was 24.1 (3.9) years and 1,615 (67.8%) identified as female. The median (IQR) length of time a peer support worker spent with patients in the ED was 10 (5, 25) minutes, and 10 (5, 17) minutes for those who had a telephone follow up. 2,047 (85.9%) patients presented to the ED with a stressful physical health concern, including abdominal pain (20.1%), chest pain/palpitations (10.8%) and nausea/vomiting (4.8%). Of the 336 (14.1%) who presented with mental health or substance use concerns, 53.6% were for an acute mental health crisis, 36.9% were for substance use such as overdose, withdrawal, or intoxication. Of the 80 ED staff who completed the survey, 76 (95.0%) knew about the P2P program and how to access the peer support workers, and 78 (97.5%) said the program helped to reduce stigma facing young adults presenting with mental health and substance use. When asked to state some of the benefits of the P2P program, ED staff said the presence of the peer supporters reduced the need for behavioural intervention (sedation, restraints, security), positively affected interactions with patients, decreased patient length of stay, and that peer supporters complement existing services provided in the ED. Conclusion: This novel peer support model of care for young adults presenting to the ED with mental health and substance use has been successfully implemented. Future work will determine if the P2P program improves patient health outcomes for this at-risk population. Keywords: mental health, peer support, substance use Introduction: The Ottawa Emergency Department Shift Observation Tool (O-EDShOT) was designed to assess a resident's ability to safely manage an emergency department shift. Locally, the O-EDShOT has demonstrated strong psychometric characteristics leading to highquality workplace-based assessment data. However, trainee performance ratings may be influenced by both clinical context and assessment culture, which vary by institution. Therefore, the objective of this multisite study was to implement and gather validity evidence for the O-EDShOT across different emergency medicine (EM) training programs in Canada. Methods: The O-EDShOT was implemented at two Royal College EM training programs that are geographically distinct from the program at which it was originally developed. Faculty from each program were recruited to complete the O-EDShOT after each shift for EM residents between January and June 2021. Mean O-EDShOT scores were calculated for each form by averaging scores for rated items. A generalizability analysis was conducted to determine the reliability of the scores. This model used mean O-EDShOT score as the dependent measure, and consisted of a factorial ANOVA with training level (junior = PGY1-2, intermediate = PGY3, senior = PGY4-5), residents, and forms as factors. A dependability analysis was conducted to determine the number of forms per resident required for a reliability of 0.80. Results: A total of 298 forms were completed for 37 residents. There was a main effect of training level with statistically significant increases in mean (SD) scores between junior (4.02(0.52)), intermediate (4.60(0.36)) and senior (4.86(0.24)) residents (p \ 0.001). There was no significant interaction between training level and program site. Residents judged by supervisors as able to safely run the shift had significantly higher mean scores (4.83(0.26)) than those judged as not able (4.12(0.53), p \ 0.001). There was no significant interaction between ability to run the shift and program site. Mean scores were significantly lower for Program A (4.31(0.58)) compared to Program B (4.67(0.43), p \ 0.001). To achieve a reliability of 0.80, 17 forms per resident would be required. Conclusion: Multiple sources of validity evidence demonstrated across implementation sites support the O-EDShOT as a tool to assess a resident's ability to safely run an ED shift. Program directors seeking to include more global assessments of resident performance may consider integrating the O-EDShOT into their program of assessment. Keywords: assessment, competence, post-graduate education Introduction: It is postulated that implementation of Competency by Design (CBD) in Royal College of Physicians and Surgeons of Canada (RCPSC) programs has helped enhanced skills programs in emergency medicine (CCFP(EM)) move towards a more competencybased residency. The objectives of the study were to identify major competency-based medical educational (CBME) components of CCFP(EM) programs across the country; and determine how programs are delivering these components. Methods: A selfadministered electronic survey was sent in French and English to all 17 CCFP(EM) program directors (PDs) using a modified Dillman method. The framework proposed by Van Melle was used to organize questions within five core components of CBME. The survey was piloted on the Assistant PDs of CCFP(EM) programs at two different sites for readability and comprehensibility of questions, and to assess for content and face validity. The final survey consisted of 44 questions under six sections. Results: There was a 100% response rate. Only 65% of programs currently map their program's curriculum to an explicit outcomes-based framework. All but one program plan to map their program's curriculum to Core Professional Activities that are to be released by the College of Family Physicians of Canada (CFPC). In 35% of programs, a curriculum is organized around developmental competencies that support resident progression and all these programs use progressive stages of training with no specific assessment needed for completion of each stage. Thirty-five percent of programs have flexibility to adjust rotations based on each learner's needs. A transition to practice curriculum exists in 47% of programs. In 94% of programs, the program meets with a resident at regular, pre-defined intervals to discuss their progression. In terms of assessment: data from direct observations are incorporated 94% of the time; 35% of programs require a minimum number of direct; residents keep procedure logs in 41% of programs; 29% of programs have specific criteria for advancement; and all programs have a Competence Committee. With the introduction of CBD in RCPSC programs, 71% of program directors felt there was less understanding of the unique needs of CCFP(EM) residents. Conclusion: Programs have moved to CBME to varying degrees, but challenges in implementation include the short 1-year training timeframe and less prescriptive direction from the CFPC. Further development and innovation are needed to fully adapt CBME to a 1-year training program. Keywords: competency, medical education Introduction: As part of a broader quality improvement initiative to decrease emergency department wait times, our team is designing an evidence-based flow training curriculum for emergency department (ED) staff. We aimed to identify the skills that ED staff require to improve efficiency in their departments and the topics that should be included in a curriculum for ED physicians, nurses, nurse practitioners, and medical learners. Methods: We conducted an integrative review to identify appropriate topics to include in an ED flow training curriculum. A medical librarian conducted a systematic search of online databases, such as Ovid MEDLINE, Embase, CINAHL, and PsycINFO. Some keywords include ''emergency department,'' ''training,'' and ''wait time(s).'' Of the 7721 articles retrieved, we included 40 in the review. We extracted, coded, and analyzed the data to identify a comprehensive list of efficiency skills in the literature. We conducted a modified Delphi via an online survey to obtain expert feedback on the findings from the integrative review. Participants included physicians, nurses, and nurse practitioners practicing in Canada. They were asked to select the topics that should be included in an ED flow curriculum and list any additional topics not found through our review. We conducted the data collection and analysis until no new concepts were identified. Results: We identified 19 efficiency skills present in the literature through the integrative review. For the modified Delphi, we received responses from 39 participants in round one and 28 in round two (response rates of 57% and 41%, respectively). Respondents to the first round included physicians (56%), nurses (21%), and nurse practitioners (15%). Similar ratios were present in the second round (physicians: 68%; nurses: 14%; nurse practitioners: 11%). The top five most frequently selected topics for a flow training curriculum were: (1) flow decisions, (2) teamwork, (3) backlog and surge management, (4) leadership, and (5) situational awareness. Conclusion: The results from the review and modified Delphi provide a foundation from which to develop an ED flow-training curriculum. Notably, the experts we surveyed prioritized skills related to interpersonal conduct among the emergency department staff (e.g. teamwork and leadership). These findings add to a significant gap in the academic literature regarding the skills that emergency department staff require to improve emergency department efficiency. Keywords: emergency department staff, flow training Le T-CCT a déjà été démontré efficace pour améliorer la qualité des CT immédiatement après l'enseignement, mais la durée de rétention des acquis demeure inconnue. L'objectif de la présente étude est d'évaluer l'effet du T-CCT sur la qualité des CT 10 mois après l'intervention. Methods: Une étude de cohorte rétrospective a été réalisée au Centre hospitalier de l'Université de Montréal (CHUM). Tous les PAB aptes à effectuer des CT et ayant consenti à participer à l'étude ont été inclus. La qualité des CT a été mesurée avec des mannequins équipés d'accéléromètres qui ont enregistré une minute complète de CT. D'après ces mesures, un score de 0 à 100%, basé sur un modèle de CT idéales (Laerdal QCPRÒ) a été calculé. Le groupe de PAB ayant participé au T-CCT 10 mois plus tôt (exposés) a été comparé à celui des PAB n'ayant jamais reçu la formation (nonexposés). La variable d'intérêt principale est la proportion de PAB ayant eu un « excellent» score (c.-à-d., plus grand ou égal à 90%). Results: La performance de 412 PAB a été mesurée, dont 229 exposés à l'intervention et 175 non-exposés. 46% (106/229) des PAB exposés ont atteint la cible de performance prédéterminée, comparé à 30% (53/175) des non-exposés. En régression log-binomiale, avoir participé au T-CCT 10 mois plus tôt augmente de 1,53 fois (RR 1,53, IC 95% [1,17-1,99]) le risque des PAB d'atteindre un « excellent» score de CT. Après ajustement pour des covariables pertinentes (c-.àd., l'âge, l'expérience en RCR, la formation préalable au BLS, le sexe et l'affectation de travail sur une unité de soins critiques), l'effet de l'intervention demeure statistiquement significatif (RR 1.57, IC 95% [1.19-2.07]). Conclusion: Dans cette étude de cohorte, on observe des effets positifs sur la rétention des compétences en CT chez les PAB exposés au T-CCT 10 mois plus tôt. Des études supplémentaires sont nécessaires pour déterminer à quelle fréquence le T-CCT devrait être répété, et pour déterminer son effet dans des situations réelles de réanimation cardiorespiratoire. Keywords: cardiopulmonary resuscitation, education Introduction: Healthcare is becoming increasingly digitized, yet remote and automated machine learning (ML) triage prediction systems for Emergency Department (ED) use are lacking. The Canadian Triage and Acuity Scale (CTAS) is the 'gold standard' comparative measure for in-person care. The objective of the current study is to describe the development of a CTAS-based ML virtual triage solution that can be piloted in Canadian emergency departments. Methods: 2,675,473 patient-level records from three large Ontario healthcare organizations were included in the development and testing of the ML model (75% derivation, 25% accuracy testing). Five different pretrained models including decision tree, k-nearest neighbors, random forest, XGBoost, and neural net were tested for preliminary accuracy. The two models with the greatest predictive accuracy were further tested by adjusting hyperparameters and subsequently comparing train accuracy, test accuracy, precision, recall, F1-scores, and result plots including confusion matrices, importance plots, and validation curves. Based on these refinements, a single ML model demonstrating the greatest predictive accuracy was selected. Specific de-identified patient predictors used to evaluate predictive accuracy included chief presenting complaint(s), clinical modifiers, age, sex, self-reported pain, and vital sign metrics compared to nursing assigned triage scores, pre-and post-implementation of eCTAS. Results: After the initial screening of the pre-trained models, the random forest and XGBoost models showed comparable predictive results and were further refined using gradient boosting. XGBoost gradient boosting regressor model was selected as the final model due to its enhanced ability to analyze data in real time. For the final model, the mean absolute error (MAE) is 0.453 and the mean square error (MSE) is 0.346. Accuracy testing using the 25% reserved testing dataset compared nursing assigned CTAS triage scores to algorithm predicted CTAS scores. The algorithm predicted the same triage score in 85.3% of cases and the same or more acute triage score in 98.1% of cases. Conclusion: The ML algorithm developed in this study shows a high predictive accuracy for predicting and assigning CTAS scores and was developed using the largest dataset of its kind to date. Future work will involve conducting a pilot study of the ML algorithm and assessing its effectiveness to accurately assign triage scores remotely. Keywords: machine learning, remote triage, virtual care LO52 RECONNECT study-impRoving carE experienCes Of equity-DeserviNg groups iN thE emergenCy department: A crosssectional study from the Kingston Health Sciences Centre Introduction: Emergency departments (EDs) serve an integral role in healthcare, particularly for vulnerable populations. However, patients from marginalized groups often report negative ED experiences, including stigmatizing attitudes and behaviors which can result in care avoidance and worsening health outcomes. We engaged with historically marginalized patients to better understand their ED care experiences. Methods: We conducted a mixed-methods study at the Kingston Health Sciences Centre. Controls and 6 equity-deserving groups (Indigenous; having a disability; experiencing mental health issues; people who use licit/illicit substances; LGBTQ2S ? ; and/or facing homelessness) were included. Over a 3-month period, ED patients registering during study hours or visiting a community partner organization were invited to complete an anonymous survey about an ED experience. Our primary outcome was perceptions of ED care within the previous 24 months at the patient/provider/system level. Results: A total of 2114 participants completed the survey, 949 controls and 994 who identified as equity-deserving. Members of equity-deserving groups were more likely to attribute negative feelings to their ED experience (p \ 0.001), to indicate that their identity impacted the care received (p \ 0.001), and that they felt disrespected/judged while in the ED (p \ 0.001). Those who identified as equity-deserving were statistically more likely to report that their ED experiences were affected by staff behaviour and that better understanding of personal situation/identity/culture was needed to improve care. Members of equity-deserving groups were more likely to indicate that staff paid too little attention to their needs (p \ 0.001), that they had little control in making decisions about their care (p \ 0.001), and that it was more important to be treated with kindness/respect than to receive the best possible care (p \ 0.001). Conclusion: Members of equity-deserving groups were more likely to report negative ED care experiences than controls. Quantitative results highlight that equity-deserving individuals felt judged and disrespected by ED staff, believed that there was insufficient attention to their needs, and felt disempowered to make decisions about their care. Qualitative thematic analysis will help contextualize findings and identify actionable change to improve ED care experiences among equity-deserving groups, making care more inclusive and better able to meet healthcare needs. Keywords: care experiences, emergency department, equitydeserving Introduction: With the adoption of competency based medical education there have been repeated calls for greater use of direct observation for workplace feedback and assessment. However, a considerable amount of assessment and feedback throughout medical training occurs through means other than directly watching a trainee perform a task. The nuances regarding the content and value afforded by these indirect observations have been scarcely examined in the literature. Therefore, this study examined how residents and faculty understand and operationalize direct and indirect observation. Methods: A constructivist grounded theory approach was used to interview 10 attending and 8 resident physicians across specialties at The Ottawa Hospital about their experiences with direct and indirect observation for assessment and feedback at work. The role of each type of observation and its perceived impact on learning was explored. Data collection and analysis were conducted iteratively. Themes were identified using constant comparative analysis. Results: Direct observation was preferred for higher acuity situations, and assessment of communication and leadership skills. Indirect observation was consistently utilized across specialties and operationalized in many forms. Attending physicians felt they could provide an accurate assessment of the learner's ability to synthesize the clinical case, generate a differential diagnosis and propose management using indirect observation. Resident physicians valued indirect observation as a means to further their independent practice. However, resident physicians did not consistently find feedback based on indirect observation to be credible. This tension seemed to stem from residents not being aware of how indirect observation informed attending physicians' judgements. Resident physicians with more insight on the use and methods for indirect observation perceived this feedback as credible. Conclusion: Consistent with prior literature, this study found that direct observation was preferred in certain situations. However, in other situations, participants identified that indirect observation is more useful and appropriate, particularly with regards to its ability to assess clinical reasoning. Residents with a deeper understanding of how indirect methods of observation informed attending physicians' assessments found this feedback more credible. For indirect observation to be effective, learners should be made aware of how they are being indirectly observed and assessed. Introduction: In Ontario, individuals who use the top 5% of hospital and home care services account for 61% of hospital and home care costs. In 2012, Ontario announced the creation of Health Links in its Local Health Integration Networks (LHINs) to provide coordinated care planning for individuals with high health care service use. The objective of this study was to explore patterns of emergency department (ED) utilization among individuals receiving coordinated care planning in the Hamilton Niagara Haldimand Brant LHIN. Methods: This analysis focused on Health Links model of coordinated care planning enrollees who experienced frequent use of the ED (5? visits/year) during the 2012/13-2019/20 fiscal years. Information was gathered from the following databases accessed through Integrated Decision Support to determine patient characteristics and health care use: Client Health and Related Information System, National Ambulatory Care Reporting System, and Discharge Abstract Database. Results: 4,294 adult residents met the inclusion criteria. The average age was 69 years (range: 18-101). Participants had a mean of 10 chronic conditions (range: 0-25) with arthritis (56%), renal failure (48%), diabetes (46%), anxiety disorders (27%), and substance-related disorders (26%). Over the course of eight years, there were 148,310 ED visits, averaging 35 per individual (range: 5-1,151). For patients who had one year before and after data on ED visits, there was an 18% decrease in the number of ED visits. The most common ED discharge diagnoses (ICD-10-CA) were chronic obstructive pulmonary disease (9%), heart failure (7%), and abdominal and pelvic pain (6%). 4,025 individuals (94%) were hospitalized 34,063 times for a total of 302,627 days, averaging 75 days per stay (range 1-1,117). Of all hospitalized individuals, 2,348 had an ICU stay for a total of 31,854 days, averaging 14 days (range 1-548). The most common inpatient discharge diagnoses were chronic obstructive pulmonary disease (10%), heart failure (9%), and disorders of urinary system (3%). 1,501 of the participants (35%) were deceased during the period of exploration. Conclusion: Individuals on coordinated care plans represent a vulnerable patient population with high ED use. The results show that following care coordination enrolment, the number of ED visits decreased. Monitoring of health care utilization over time is warranted to better support these individuals in their health care navigation pathways. Introduction: Computed tomography (CT) findings of acute and chronic ischemia are associated with subsequent stroke risk in patients with transient ischemic attack (TIA). We sought to validate these associations in a large prospective cohort of patients with TIA or minor stroke. Methods: We included prospectively enrolled emergency department patients from 13 Canadian hospitals with TIA or minor stroke who had CT imaging within 24 h. Primary outcome was stroke within 90 days. Secondary outcomes were stroke within 2 or 7 days. CT findings were abstracted from radiology reports and classified for the presence of acute ischemia, chronic ischemia or microangiopathy. Multivariable logistic regression was used to test associations with primary and secondary endpoints. Results: From 8,670 prospectively enrolled patients between May 2010 to May 2017, 8,382 had a CT within 24 h. 4,547 (54%) patients had evidence of acute ischemia, chronic Ischemia, or microangiopathy on CT, of whom 175 had a subsequent stroke within 90 days (3.8% subsequent stroke rate; OR, 2.33; 95% CI, 1.62-3.36; P \ 0.001). Findings associated with an increased risk of stroke at 90 days were isolated acute ischemia (6.0%; OR, 2.42; 95% CI, 1.03-5.66; P = 0.04), acute ischemia with microangiopathy (10.7%; OR, 3.34; 95% CI, 1.57-7.14; P = 0.002), chronic ischemia with microangiopathy (5.2%; OR, 1.83; 95% CI, 1.34-2.50; P \ 0.001), and acute ischemia with chronic ischemia and microangiopathy (10.9%; OR, 3.49; 95% CI, 1.54-7.91; P = 0.003). Acute ischemia with chronic ischemia and microangiopathy were most strongly associated with subsequent stroke within 2 days (OR, 4.36; 95% CI 1.31-14.54), and 7 days (OR, 4.50; 95% CI, 1.73-11.69). Conclusion: In patients with TIA or minor stroke, CT evidence of acute ischemia with chronic ischemia and/or microangiopathy significantly increases the risk of subsequent stroke within 90 days of index visit. The combination of all 3 findings results in the greatest early risk. Keywords: computed tomography, stroke, transient ischemic attack Introduction: CT pulmonary angiograms (CTPAs) are increasingly being used for the diagnosis of PE while V/Q scans are still used in some clinical contexts. The use of CTPAs however has been previously reported to lead to a larger number of incidental findings requiring follow-up imaging. We sought to determine the frequency and type of incidental findings and alternative diagnoses found on the imaging tests of patients presenting to the ED with symptoms concerning for PE and to provide a cost estimate using a public payer perspective for the follow-up imaging recommended in the Canadian context. Methods: Adult patients who presented to the ED at St. Michael's hospital in Toronto, Ontario between the 1st of January 2016 and 31st of December 2017, and who were investigated for a PE with a CTPA or V/Q scan were enrolled. Two research team members collected data from imaging reports regarding PE, incidental findings requiring imaging, additional imaging required and alternative/additional diagnoses. Decision tree analysis was used to determine the additional costs of imaging taking into account incidental findings and inconclusive scans. Costs in Canadians dollars were estimated using data from the Canadian Agency for Drugs and Technologies in Health (CADTH). Results: 649 patients met our inclusion criteria. Most of the patients were imaged using CTPA (98.15%). Regardless of the scan type, 19.56% of scans were positive for a PE, 39.91% of scans described incidental findings requiring follow-up imaging and 47.71% of scans listed an alternative/additional diagnosis. Thoracic nodules were the most common incidental finding (40.35% of all incidentals) requiring follow-up imaging. The most common alternative/additional diagnoses were lung related (84.97%) and were predominantly pneumonias/other lung infections. Revised costs including additional scans for inconclusive initial imaging tests and additional CT Thorax without IV contrast for incidental findings would bring up the cost of a CTPA scan by 16.1% and the cost of a V/Q scan by 24.34%. Conclusion: Although CTPA is a useful diagnostic modality in the investigation of patients with suspected PE with the added benefit of providing alternative/additional diagnoses, the incidental findings that are commonly found on this imaging modality are associated with additional investigations, radiation and system costs. Ordering of CTPAs should therefore be carefully considered as part of a validated PE investigation pathway. Keywords: diagnostic imaging, incidental findings, pulmonary embolism Introduction: Nonmedical problems have an impact on the quality of life of older adults, but they are inconsistently assessed in the emergency department (ED). The main objective of this study was to determine the prevalence of nonmedical needs among older adults attending the ED for low acuity conditions. A secondary objective was to explore their knowledge of potentially available resources. Methods: This was a prospective cohort study conducted in 8 EDs between March and August 2021. Well-oriented older adults (C 65 years old) were considered eligible if the initial Canadian Triage and Acuity Scale was 3, 4 or 5. Patients who were unable to speak French or English, considered hemodynamically unstable, actively receiving palliative care in the ED and/or living in a longterm care facility (CHSLD) were excluded. Recruitment was performed during weekdays by four research assistants who rotated between the different EDs. A consensus-based questionnaire on nonmedical needs was administered. Data are presented using central tendency metrics and proportions. Results: A total of 1060 patients were included. The mean age was 77.1 (standard deviation 7.6), 55.7% were female and 41.6% reported living alone. The main medical comorbidities were HTN (n = 645, 60.8%), anxiety (n = 306, 28 .9%) and COPD (n = 211, 19.9%) . 646 patients (60.9%) reported taking C 5 medications daily and 522 (49.2%) suffering from chronic pain. The most frequent nonmedical problem was loneliness (n = 313, 29 .5%) for which 197 patients (62.9%) were aware of potentially available community resources. Other reported nonmedical problems were food insecurity (n = 91, 8 .6%), financial difficulties (n = 101, 9 .5%), challenges in accessing medication (n = 48, 4 .5%) or transport (n = 43, 4.1%) . Finally, 36 (3.4%) and 35 (3.3%) participants identified themselves as victims of psychological and/or physical abuse, respectively. Only 12 (33.0%) of those who reported psychological abuse were aware of a community-based resource on abuse. Conclusion: Approximately one third of older adults visiting the ED reported feeling lonely while other non-medical problems such as food insecurity and financial difficulties remains frequent. Designing ED processes that facilitate the assessment of nonmedical problems and linkage with community resources is required to improve care of older adults. Keywords: emergency department, nonmedical needs, older adults Introduction: Chest ultrasonography (U/S) has been reported as an accurate imaging modality, and potentially superior compared to chest x-rays in diagnosing traumatic rib fractures. However, few studies have compared U/S to the reference standard of computed tomography (CT), with no prior systematic review. Our objective was to conduct a systematic review and meta-analysis to summarize the evidence comparing the test characteristics of chest U/S to CT in diagnosing rib fractures. Methods: We completed a systematic review in adherence to PRISMA guidelines and registered with PROSPERO (CRD42021252889). A registered librarian with expertise in systematic reviews searched Cochrane Library, EMBASE, Medline, Web of Science, and SCOPUS from database inception to October 2021. The search was peer reviewed by a second librarian. Two independent reviewers further searched gray literature, completed study selection, extracted data, and assessed QUADAS-2 risk of bias. We included all English language studies that enrolled adult patients with clinically suspected rib fractures following blunt trauma, who presented to an emergency department (ED) and received both a chest U/S and CT chest. Summary measures were obtained from the Hierarchical Summary Receiver Operating Characteristic model. Results: From 1660 citations, we identified 7 studies for inclusion, of which 6 had available 2 9 2 data for meta-analysis (n = 663) . Of the 6 studies, 3 involved ED performed U/S and 3 radiology performed U/S. Chest U/S had a pooled sensitivity of 89.3% (95% CI 81.1-94.3) and specificity of 98.4% (95% CI 90.2-99.8) compared to CT for the diagnosis of any rib fracture. Identification of a fracture on U/S, defined as an underlying cortical irregularity, had a ? LR of 55.7 (95% CI 8.5-363.4) for CT diagnosed rib fracture. The absence of U/ S fracture held a -LR of 0.11 (95% CI 0.06-0.20). We were unable to detect a difference in test characteristics between ED and radiology performed U/S (p = 0.11). Overall risk of bias of included studies was high, with patient selection identified as the highest risk domain. Conclusion: Chest U/S is both sensitive and highly specific in the diagnosis of rib fractures following blunt trauma. Our review supports the use of U/S for the diagnosis of rib fractures. Keywords: rib fractures, ultrasound We recently showed that delirium recognition by ED nurses and physicians remains at \ 50%. We previously developed and tested an algorithm known as ''PREDICT'' that uses performance on a ''serious game'' to identify people at increased risk of delirium in a single center sample. Objectives: (1) To validate the predictive ability of the PREDICT algorithm to identify older people accessing the ED at increased risk for delirium in a national multi-center cohort;

(2) to compare the algorithm ability to identify both prevalent and incident delirium to clinical recognition. Methods: A prospective observational study of people C 65 years enrolled in the ED of academic 7 centers in four provinces who played a tablet-based game similar to ''Whack-a-Mole''. We assessed delirium using the Confusion Assessment Method, then asked ED nurses and physicians if the patient had delirium, and to predict risk of incident delirium on a 10 point numeric rating scale. We report the delirium rates with 95% CI, and sensitivity/specificity of the game identification compared to clinical delirium recognition. Results: We enrolled 1493 participants, all of whom completed game play; their mean age was 77.9 years; 49.2% were female, and 79 (5.3%, 95% CI 4.2-6.5%) had prevalent delirium on initial ED assessment-an additional 21 developed incident delirium. ED Nurses missed delirium in 43/78 (55.1%) prevalent delirium cases. They rated 14/21 (67%) cases that developed incident delirium as high risk ([= 6/10). Physicians missed prevalent delirium in 10/20 cases they assessed, but only rated 2/21 (9.5%) cases as high risk. Sensitivity of our algorithm was 80% for both prevalent and incident delirium, with specificity of 56% and 55% respectively. Nurses assessment of delirium was 45% and 67% sensitive for prevalent and incident delirium, and 98% and 76% specific. Physicians were 50% and 9.5% sensitive for prevalent and incident delirium, and 95% and 96% specific. Conclusion: We validated the predictive ability of the PREDICT algorithm to identify those at high risk for delirium in an independent cohort. Sensitivity for prevalent delirium recognition by nurses and physicians was sub-optimal at * 50% compared to 80% for the algorithm, although clinicians were highly specific. Nurses appeared to be better at predicting incident delirium than physicians. Use of the PrEDICT algorithm to prompt clinical re-assessment may improve sensitivity of delirium recognition. Keywords: delirium, emergency department, geriatrics LO60 Is the addition of computed tomography with angiography superior to a non-contrast neuroimaging only strategy for patients with suspected stroke or transient ischemic attack presenting to the emergency department?

patients presenting with suspected acute stroke or transient ischemic attack (TIA). Various forms of computed tomography (CT) are currently available for initial investigation, including non-contrast CT (NCCT), CT angiography head and neck (CTA), and CT perfusion (CTP). However, there is uncertainty around optimal imaging choice for cost-effectiveness, particularly for minor or resolved neurological symptoms. In addition to the cost of CTA and CTP testing, there is also concern for increased incidental findings which may contribute to the burden of overdiagnosis. Methods: In this cross-sectional observational study, analysis was conducted on 586 anonymized triage and diagnostic imaging (DI) reports for neuroimaging orders completed on patients presenting to adult emergency departments (EDs) with a suspected stroke or TIA from January-December 2019. The primary outcome of interest is the diagnostic yield of NCCT ? CTA compared to NCCT alone for patients presenting to urban academic EDs with Canadian Emergency Department Information System (CEDIS) complaints of ''symptoms of stroke'' (specifically acute stroke and TIA indications). DI reports were coded into 4 prespecified categories (endorsed by a panel of stroke experts): no abnormalities, clinically significant findings (requiring immediate or follow-up clinical action), incidental findings (not meeting prespecified criteria for clinical significance), and both significant and incidental findings. Standard descriptive statistics were performed. A two-sided p-value \ 0.05 was considered significant. Results: 75% of patients received NCCT ? CTA imaging, 21% received NCCT alone, and 4% received NCCT ? CTA ? CTP. The diagnostic yield of NCCT ? CTA imaging for prespecified clinically significant findings was 24%, compared to only 9% in those who received NCCT alone. The proportion of incidental findings was 30% in the NCCT only group and 32% in the NCCT ? CTA group. CTP did not significantly increase yield of significant or incidental findings. Conclusion: In this cohort of patients presenting with suspected stroke or TIA, an NCCT ? CTA neuroimaging strategy had a higher diagnostic yield for clinically significant findings than NCCT alone without significantly increasing the number of incidental findings identified. Keywords: diagnostic yield, neuroimaging, stroke Introduction: Abdominal ultrasound (US) or computed tomography (CT) is used by emergency department (ED) physicians to diagnose appendicitis or an alternative etiology for patients presenting with right lower quadrant (RLQ) abdominal pain. Up to 84% of US are indeterminate for acute appendicitis and require a CT. Our study aims to determine predictors for appendicitis when the US is indeterminate. Methods: We performed a health records review of adult ED patients presenting with RLQ pain and indeterminate US to two Canadian academic EDs between June 2019 and July 2020. Patient demographics, comorbidities, results of investigations, interventions and disposition were recorded. The study outcome was diagnosis of appendicitis within 30 days of the index ED visit ascertained by chart review. We used multivariable logistic regression and report odds ratios (OR). We used cubic spline functions to develop cut-off threshold for the continuous predictors and report diagnostic performance. 95% confidence intervals (CI) were calculated when appropriate. Results: 463 patients (mean age 30.3 years standard deviation 10.5, 74.9% female) with RLQ pain and indeterminate US were included, of whom overall, 45 (9.7%, 95% CI 7.3%, 12.8%) patients were diagnosed with appendicitis. A subsequent CT was performed in 227 (54.6%) patients and 39 patients (17.2%, 95% CI 12.8%, 22.6%) were diagnosed with appendicitis. Among the 236 patients who did not have a subsequent CT, six (2.5%, 95% CI 1.2%, 5.4%) patients had appendicitis, of whom four were diagnosed by the surgical team during the index ED visit and two were confirmed after the index visit discharge within 30 days. Neutrophil count (OR = 1.2, 95% CI 1.1, 1.3) and secondary signs of inflammation on US (OR = 2.2, 95% CI 1.1, 4.4) were associated with a higher likelihood of appendicitis (C-statistic 0.77, 95% CI 0.70, 0.84). Neutrophil count [ 5.5 9 109 or presence of secondary signs had a sensitivity of 89% (95% CI 76%, 96%) and specificity of 48% (95% CI 41%, 51%) for appendicitis when US is indeterminate. Conclusion: For ED patients with RLQ pain and indeterminate US for appendicitis, higher neutrophil count or secondary signs on US should increase the suspicion for appendicitis. Future large-scale prospective studies are needed for developing a risk tool. Keywords: appendicitis, predictors, ultrasound Established clinical assessment tools have uncertain performance in the ED context and non-contrast computed tomography (CT) has low sensitivity for posterior circulation abnormalities. However, CT angiography (CTA) carries added cost and potential risks, including contrast complications and incidental findings. Our group previously found large variability in diagnostic imaging utilization across different ED sites in Alberta. This study aimed to determine the diagnostic yield of CTA for patients presenting to Calgary ED's undergoing workup for a potential central cause of vertigo. Methods: Imaging reports were obtained for patients C 20 years old presenting to a Calgary ED between April 1st, 2019 and March 31st, 2020 who received a CTA after a CEDIS presenting complaint of vertigo or gait disturbance/ataxia. Exclusion criteria were: imaging ordered by inpatient providers (rather than ED physicians), CT venograms, and addendum/incomplete reports. Each report was independently reviewed and coded by two blinded reviewers based on a prespecified list of significant findings (endorsed by a panel of stroke experts) and categorized into one of the following: i) at least one finding requiring urgent management, ii) other findings (non-urgent atherosclerosis, incidental findings, or anatomic variants without clinical significance), or iii) unremarkable. Discrepancies were adjudicated by a stroke expert. Results: A total of 567 CTA reports met criteria and were included in the analysis. 117 scans (20.6%) had at least one finding requiring urgent management, 347 scans (61.2%) had other findings present, and 103 scans (18.2%) were unremarkable. Of scans requiring urgent management, 68 (58%) had vertigo-specific pathology affecting the posterior territory, while the other 49 (42%) had findings that did not convincingly explain patients' presentation for vertigo yet still required urgent intervention or follow-up. Most findings requiring urgent management were only detected on the angiographic phase of imaging (101/117; 86.3%). Conclusion: Diagnostic imaging may identify urgent pathology in up to 1 in 5 patients presenting to the ED with vertigo. Significant findings were often only detected on angiography, suggesting that CTA may be more appropriate than non-contrast CT alone for investigating patients with suspected central causes of vertigo and ataxia. Keywords: computed tomography angiography, diagnostic imaging, vertigo Introduction: Delirium is unrecognized in the ED-up to 75% of cases are missed in older adults. This leads to poor outcomes, including increased mortality. Measurement of ED recognition of delirium has been variable. Prior to 2002, retrospective chart review was predominant, but recently, prospectively interviews of ED staff are increasingly common. To compare ED delirium recognition rates in studies using prospective and retrospective methods, as well as examine temporal trends in reported ED delirium recognition. Methods: We conducted a meta-analysis according to PRISMA guidelines. We developed search strategies for Medline, Embase, CINAHL, and PsycINFO. Included studies had (1) patients aged C 65 in the ED, (2) detected delirium using a validated screening tool, and (3) had a measure of ED staff delirium recognition rates. Studies identifying only delirium tremens were excluded. Abstract and title screening, full-text screening, followed by data abstraction was completed by 2 independent reviewers. Results: Our search yielded 5274 after deduplication; 14 studies met inclusion criteria. ED delirium prevalence ranged from 5.1-24.4%. 9 studies used retrospective chart review and 5 studies used prospective interviews of ED staff to establish delirium recognition rates. The 9 retrospective studies reported a lower recognition rate than the 5 studies that used prospective interviews, though not significantly different (21.9% [95% CI = 13.9-32.8] vs. 41.0% [95% CI = 29.8-53.2]). In a posthoc sensitivity analysis, we excluded one study with confounded recognition rates. Excluding this outlier, the difference in delirium recognition rates between the retrospective and prospective approaches was statistically significant (18.9% [95% CI = 13.0-26.5] vs. 41.0% [95% CI = 29.8-53.2], p \ 0.001). There is no clear temporal trend in delirium recognition rates, with widely overlapping confidence intervals (24.7% [95% CI = 14.6-38.6] vs. 30.7% [95% CI = 18.8-46.0]). Conclusion: Use of retrospective chart reviews to establish delirium recognition may under-estimate actual recognition rates. Regardless, the bulk of the evidence reviewed suggests that delirium recognition appears to be less than 50% even when restricting our analysis to prospective studies. We did not find evidence that delirium recognition by ED staff has improved in the previous decade. Keywords: delirium, elderly, emergency LO64 Intra-articular lidocaine versus intravenous sedation for closed reduction of acute anterior shoulder dislocation in the emergency department: A systematic review and meta-analysis Introduction: Anterior shoulder dislocations are the most common type of joint dislocation seen in the emergency department (ED). Two common types of analgesia for anterior shoulder dislocation reductions have been studied and compared in the ED setting: intraarticular lidocaine (IAL) and intravenous sedation (IVS). The objective of this systematic review and meta-analysis was to compare the efficacy of IAL versus IVS on successful closed reduction of acute anterior shoulder dislocation in the ED. Methods: Electronic searches of MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials were completed and reference lists were hand searched. Randomized controlled trials (RCTs) comparing IAL to IVS for reduction of acute anterior shoulder dislocations in the ED were included. Two reviewers independently screened abstracts, assessed study quality, risk of bias, and extracted data. Data were pooled using random-effects models and reported as mean differences (MD) and risk ratios (RR) with 95% confidence intervals (cIs). Results: 12 RCTs were included with a total of 630 patients (IAL = 327; IVS = 303). There was no significant difference in reduction success between IAL and IVS (RR = 0.93; 95% CI 0.86 to 1.01), significantly lower adverse events with IAL (RR = 0.16; 95% CI 0.07 to 0.33), significantly shorter ED length of stay (LOS) with IAL (MD = -1.48; 95% CI -2.48 to -0.47), no significant difference in pain scores post-analgesia (MD = -0.04; 95% CI -1.10 to 1.02), and no significant difference in ease of reduction (RR = 0.78; 95% CI 0.59 to 1.04). The itemized cost of IAL was reported as less than IVS. Conclusion: IAL had similar effectiveness as IVS in the successful reduction of anterior shoulder dislocations in the ED with fewer adverse events, shorter ED LOS, and no difference in pain scores or ease of reduction. Given the improved safety profile, cheaper healthcare costs, and decreased time in hospital, IAL may be an effective alternative to IVS when reducing anterior shoulder dislocations, particularly when IVS is contraindicated or not feasible. Keywords: lidocaine, sedation, shoulder dislocation Introduction: An increasing number of patients taking direct oral anticoagulants (DOACs) are managed in the emergency department (ED). It is unknown whether the risk of traumatic intracranial hemorrhage (ICH) after a head injury differs between DOACs. The objective of this study was to compare the risk of ICH at the index ED visit among older adults presenting to the ED with a head injury prescribed different DOAC medications. Methods: This was a retrospective cohort study using population-based data from Ontario, Canada between 2016 and 2018. We included patients age 65 years and older prescribed a DOAC who were seen in the ED with a presenting complaint of a head injury. Patients were matched on the propensity score to create three pairwise-matched cohorts based on the DOAC prescribed (dabigatran vs rivaroxaban; dabigatran vs apixaban; rivaroxaban vs apixaban). For each cohort, the relative risk (RR) and 95% confidence intervals (CI) of ICH diagnosed at the index ED visit were calculated. Results: We identified 9,230 older adult ED patients presenting with a head injury prescribed a DOAC. There were 1,274 (13.8%) patients with a prescription for dabigatran, 3,136 (34.0%) patients with a prescription for rivaroxaban, and 4,820 (52.2%) patients with a prescription for apixaban. Overall, 5.9% of patients had an ICH at the index ED visit. After propensity score matching, there were no significant differences in the risk of ICH between any matched DOAC cohorts: dabigatran vs rivaroxaban; RR = 0.97 (95% CI 0.72-1.31); dabigatran vs apixaban; RR = 1.07 (95% CI 0.79-1.44); rivaroxaban vs apixaban; RR = 1.09 (95% CI 0.89-1.34). Conclusion: In patients aged 65 years and older with a prescription for a DOAC seen in the ED for a head injury, there were no differences in the risk of ICH between DOACs. Keywords: anticoagulation, head injury, intracranial hemorrhage Anyone seen by SADV for IPV during these periods was included. Patients \ 14 years or those presenting for non-IPV forms of violence were excluded. Data was extracted by two independent reviewers and a third audited 10% of charts for quality assurance. Primary outcome measures were frequency of IPV presentations and injury severity, assessed using the Clinical Injury Extent Score (CIES), a validated tool for IPV injuries. CIES scores [0 (no injuries)-3 (severe injuries)] were calculated by two raters who were blinded to the date of presentation. Disagreements were resolved by consensus. Results: A total of 128 patients met the inclusion criteria and were included in this study. The majority of participants were female (97%), single/never-married (59%) and had a mean age of 34 years (SD = 11). Eight patients presented more than once for IPV during the study period, for a total of 139 acute visits. The frequency of IPV visits before and during COVID was similar: 72 (52%) in Pre-COVID vs. 67 (47%) in COVID (p = 0.73). Of the COVID presentations, 43% (N = 29) of visits occurred during lockdowns (11 in LD1, 12 in LD2, and 6 in LD3), which was not significantly different from the number of non-lockdown visits (p = 0.46). Mean CIES scores were significantly higher during COVID vs. Pre-COVID [(M = 1.91, SD = 1.1) vs. (M = 1.69, SD = 1.3), p \ 0.01]. There were no differences in mean CIES scores observed between lockdown periods (p = 0.71). Conclusion: While the overall frequency of IPV presentations to SADV did not change during COVID when compared to the 15 months prior, injury severity increased. This may reflect escalation of violence as the pandemic and its restrictions persist and requires further investigation. Keywords: COVID-19 pandemic, injury severity, intimate partner violence The CRASH-2 trial demonstrated that early administration of tranexamic acid (TXA) reduced the risk of death due to bleeding in trauma patients. However, despite evidence of effectiveness, previous studies have shown poor compliance with CRASH-2 TXA administration protocols. The objective of this study was to assess physician compliance with standard TXA protocols at our centre and determine if TXA protocol compliance has changed over time. Methods: This was a retrospective medical record review of adult (C 18 years) trauma patients presenting to a level-one trauma centre from January 2012 to July 2019. Patients were included if they had at least one risk factor for significant hemorrhage, defined as systolic blood pressure (sBP) less than 90 mmHg, heart rate greater than 110 bpm, or transfusion of at least one unit of packed red blood cells (pRBCs) in the emergency department. The primary outcome of interest was early administration of TXA according to our local trauma protocol. Between-group differences of compliance were assessed using chi-square, Fisher's exact or a Student's T-test. Results: Of 582 trauma patients who presented to our centre, 445 met eligibility criteria for TXA administration and were included in the study. TXA was administered to 118 of 445 (26.5%) patients. Most patients (62.2%) met only one criterion for risk of significant hemorrhage at presentation, and only 2.5% of patients met all three criteria. The compliance rate for patients meeting only one criterion for risk of significant hemorrhage ranged from 0% to 40.5%, increasing to 54.5% for patients meeting all three criteria. Patients who received TXA had a lower sBP compared to those who did not receive TXA (106.5 mmHg vs. 126.1 mmHg; D 19.6 mmHg, 95% CI 9.62 mmHg to 29.67 mmHg) and the average number of pRBCs transfused was significantly higher (10.9 units vs. 2.7 units; D 8.2 units, 95% CI 6.5 to 10.0 units). TXA administration improved from 2.8% (1/36) in 2012 to 46.2% (18/39) in 2019 (D 43.4%, 95% CI 25.9% to 58.8%). Conclusion: Compliance with TXA administration protocols at our level-one trauma centre has improved over time but remains low. Future studies should attempt to elucidate barriers and facilitators to TXA administration for trauma patients at risk of death due to bleeding. Education strategies and interventions should be developed and implemented to increase compliance with TXA protocols for eligible trauma patients. Introduction: Severe trauma is often accompanied by hemorrhage, shock, and subsequent patient deterioration. Perhaps the most widely used prehospital strategy for the management of hemorrhagic shock or trauma accompanied by hypotension is fluid resuscitation. Current guidelines by the ATLS suggest immediate fluid resuscitation, usually beginning with 1L bolus of isotonic fluid and subsequent titration based on patient response. Efficacy of prehospital fluid delivery has recently come into question, with contemporary literature suggesting a more restrictive approach. Our objective was to evaluate the effectiveness of low/ no IV fluids in comparison to standard resuscitation in reducing mortality for trauma patients in the prehospital setting. Methods: Population-any adult with any cause of blunt or penetrating trauma in the prehospital setting with severe injury (defined as the presence of hypotension (SBP \ 90 mm Hg) and/ or a Shock Index [ 1). Intervention-low/ no IV fluids. Comparisonstandard resuscitation. Outcome-Mortality. We performed a librarian-assisted search of 4 databases (Medline, Embase, Web of Science, and CINAHL) between June 24 and June 28th, 2021, in addition to website and citation searching in August, 2021. ROBINS-1 and ROB-2 tools were used to assess risk of bias in observational and randomized studies respectively. An inverse variance method and random-effects model of statistical analysis were utilized. Data were reported as risk-ratios with related 95% confidence intervals. Heterogeneity of studies was assessed through analysis of the I2. Results: A total of 7 studies (6 observational and 1 randomized trial) were included, with a total of 3050 study participants included for analysis. Four studies compared high to low dose fluids, and three compared fluids to no fluids. The mean ISS ranged from 10 to [ 50, with injury type including both blunt and penetrating trauma. We found no difference in mortality when comparing standard resuscitation to restricted or no fluid resuscitation (RR = 0.99, 95% CI [0.80-1.22]). With the removal of two low quality studies, the I2 was reduced to 0%, and results trended to favouring low/no fluid (RR = 0.84, 95% CI [0.70-1.00], p = 0.05). Conclusion: Standard fluid resuscitation had no significant mortality benefit over restricted resuscitation in the prehospital setting for severely injured trauma patients. There is weak evidence to support reducing or withholding fluids in the prehospital setting in the context of trauma. Keywords: fluid, prehospital, resuscitation Introduction: Pre-hospital trauma team activation is felt to expedite and improve care of trauma patients; however, little evidence exists evaluating the effects of such systems. Our study objective was to assess the impact of trauma team activation prior to patient arrival on patient survival at 24 h. Our secondary objectives were to assess time to first procedure, time to computed tomography (CT), time to first transfusion, time to critical administration threshold (CAT)-defined as transfusing at least 3 units of blood within one hour, and emergency department length of stay (ED LOS). Methods: We conducted a medical records review of all trauma team activations from September 29, 2016 to March 5, 2020 at The Ottawa Hospital. We compared patient outcomes between in-field and in-hospital trauma team activation. Relevant charts were identified from the regional trauma registry, and a single reviewer performed the chart review and data entry. Our analysis included descriptive statistics. We assessed the effects of in-field activation on the study outcomes using regression models to control for possible confounders, such as injury severity score (ISS). Results: Of the 1015 trauma team activations that occurred during our study period, 762 met our inclusion criteria. Children, patients direct to the trauma service or transferred from outside hospitals, and patients that received emergent procedures prior to arrival were excluded. For in-field trauma activations, the median age was 38.0 years with 79.4% male, compared to 41.0 years and 77.5% male for in-hospital activations. In-field activations had a median ISS of 16.0 with a 38.3% incidence of penetrating trauma, relative to 12.0 and 23.9% for in-hospital activations. When controlling for ISS, in-field activations had an increase in 24-h mortality (14.4% vs. 4.5%; P = 0.0108, odds ratio 2.240). We found a decrease in time (minutes) to emergency procedure (18.0 vs. 27.0; P = 0.0005), CT (37.0 vs 42.0; P = 0.0086), first transfusion (14.0 vs. 28.0; P \ 0.0001), and ED LOS (101.0 vs. 171.0; P \ 0.0001) with in-field activation. Time to CAT (25.0 vs. 31.0; P = 0.1057) was also shorter, but did not reach statistical significance. Conclusion: Our study showed that, when controlling for ISS, in-field trauma team activation had a significant increase on 24-h mortality, with a significant decrease on time to emergency procedure, CT, first transfusion, and ED LOS. Given these results, further assessment of in-field trauma team activation is warranted. Keywords: prehospital, trauma Introduction: Trauma is the leading cause of death before the age of 40. The trauma team leader (TTL) is a ''model'' of a specific dedicated care team leader in the emergency department (ED). The presence of a trauma team led by an experienced physician could potentially facilitate and improve resuscitation, diagnosis and treatment for trauma patients, but TTL benefit is uncertain. The primary objective was to assess its impact on 72 h mortality. Secondary objectives included 24 h mortality and admission delays from the ED. Similar results were found for the proportions of patients admitted within 8 h after ED arrival (0.36 [-1.47; 2.18]). Sensitivity analyses using the two other centres as controls yielded similar results. Conclusion: TTL implementation was not associated with changes in mortality nor on admissions delays from the ED which may not be the appropriate indicators of TTL plus-value. Future studies are needed to assess the potential impact of TTL programs on other patient-centered outcomes using different quality of care indicators. Keywords: mortality, quality improvement, trauma team leader Introduction: Death from massive hemorrhage is often preventable and is managed by damage control resuscitation, which includes activation of hospital massive hemorrhage protocols (MHP). The Ontario Regional Blood Coordinating Network (ORBCoN) developed and released the first provincial evidence-based adult and paediatric MHP in May 2021, including 8 quality metrics. This study reports the results of the testing of a Beta prototype of the MHP quality metric tool across 3 Level One Trauma Centres in Ontario.

Can J Emerg Med (2022) 24 (Suppl 1):S1-S100

Vol.: (0123456789) Methods: A retrospective review (Jan 2019 -Oct 2021) was conducted of 288 MHP activations (151, Kingston; 112, Sunnybrook; 12, Thunder Bay; 13 from Children's Hospital of Eastern Ontario and several non-academic centres). Data was entered into a REDCap data collection tool (patient demographics, outcomes, and quality metrics). Differences between trauma vs. non-trauma MHP quality metrics were assessed using Fisher's exact tests. Similarly, differences on the metrics between the three centres were assessed using Chi Square tests. 288 activations were reviewed (169 non-trauma and 119 trauma). 63.3% and 54.1% of trauma and non-trauma patients were alive at discharge, respectively. Results: Trauma vs. non-trauma 8 quality metrics were as follows: (1) Received tranexamic acid \ 60 min (74% vs. 82%, p \ 0.001); (2) RBC transfusion \ 15 min (89% vs. 89%, p = 0.80); (3) Transition to group-specific components (92% vs. 94%, p = 0.70); (4) Temperature [ 35°C (66% vs. 76%, p = 0.08); (5) Hemoglobin [ 60 g/L in first 24 h (92% vs 91%, p = 0.89), below 110 g/L at 24 h (69% vs 75%, p = 0.50); (6) MHP activation appropriate (69% vs. 63%, p = 0.50); (7) Blood components wasted (18% vs 27%, p = 0.08); and (8) Transfer \ 60 min (78% vs. 94%, p \ 0.001). Significant differences between the 3 centres were as follows: (1) Conclusion: High performance was observed across all metrics, except tranexamic acid administration within 1 h of MHP activation and temperature over 35°C at termination. There was substantial variability between sites for several MHPrelated quality metrics suggesting the tool will provide valuable benchmark data to inform site-specific quality improvement initiatives. Keywords: massive hemorrhage protocol, trauma activation Introduction: Thoracic injuries are frequent complaints in the emergency department (ED) and serious complications may arise from this type of trauma. Current clinical decision tools are primarily based on the number of rib fractures-which are frequently missed on plain chest x-ray. We sought to evaluate the benefits of rib series on rib fracture identification. Methods: A prospective cohort study was conducted in four university-affiliated Canadian EDs over a 6-year period. Patients aged C 16 years presenting to the ED with a minor thoracic injury were recruited and underwent both a chest x-ray and rib series. Blinded radiologists' reports were reviewed for concordance of rib fracture identification. Univariate and weighted kappa analysis were performed. Results: 1332 patients were included. Chest x-ray identified no rib fracture for 1246 patients (93.5%), 1 rib fracture for 36 patients (2.7%), 2 for 31 patients (2.5%) and C 3 for 19 patients (1.4%). Rib series identified no rib fracture for 882 patients (66.2%), 1 rib fracture for 213 patients (15.9%), 2 for 125 patients (9.3%) and C 3 for 112 patients (8,4%). Weighted kappa (95% CI) was 0.25 (0.20-0.30) between x-ray techniques. For patients with a negative chest x-ray for rib fracture, rib series identified at least one fracture 27.3% of the time. Conclusion: Nearly 1 patient out of 3 presented an undetected rib fracture on the routine chest x-ray when rib series were added. The agreement between chest x-ray and rib series is weak for the overall number of rib fracture. Keywords: radiology, rib fractures, thoracic injury (811), is an accessible telehealth platform at no cost to the user that provides 24/7 service for individuals seeking health advice. Trained nurses assess the situation and direct callers along care pathways: to present immediately to the emergency department (ED), to see their primary care provider (PCP), or to self-care at home (SC). This descriptive study seeks to explore urban and rural differences in Health Link utilization and subsequent ED presentation. Methods: Data on Health Link callers from January 1, 2018-December 31, 2019 was extracted from administrative health data, including postal code, location of ED attended, and Canadian Triage Acuity Scale (CTAS) assigned at ED. Alberta Health provided population estimates for each postal code. This data was used to construct region-specific density of usage maps to compare the Health Link ED and the general ED populations in each region, as well as urban and rural populations as composite groups by usage density and CTAS. The primary outcome used to compare these populations in this study is users/100 population/year. Results: In this period, 900,196 individuals called Health Link, 241,103 were referred to the ED, and 58.3% (140,614) presented to the ED within 24 h of the call. Health Link callers constituted 3.4% of the total ED visit population (4, 194, 735) . While there is a greater density of ED usage in rural than urban locations (101.6 vs. 41.2 visits/100/year), a smaller proportion of individuals who call Health Link live in rural than in urban areas (2.1 vs. 2.5 callers/100/year). More urban Health Link callers that were advised to go to the ED presented with a CTAS \ /= 3 than CTAS [ 3 (1.2 vs. 0.4 callers/ 100/year). Rural citizens who did not call Health Link and went directly to the ED presented more often with a CTAS [ 3 than a CTAS \ /= 3 (60.0 vs. 35.0 visits/100/year). However, when rural citizens called Health Link and were then advised to go to the ED, they more often presented with a CTAS \ /= 3 than CTAS [ 3 (0.9 vs. 0.6 callers/100/year). Conclusion: While Health Link provides greater access of care advice to individuals from rural and remote communities, a greater density of callers originated from urban centres. When rural citizens do utilize Health Link, a greater proportion of those that go to the ED present with a higher acuity CTAS. These findings indicate that Health Link has the potential to correctly identify callers in need of ED care and should be promoted for use in rural communities. Keywords: emergency department referral, geography, telehealth Introduction: Bystander cardiopulmonary resuscitation (CPR) has been shown to restore some blow flow and decrease the ischemia experienced by patients suffering from an out-of-hospital cardiac arrest (OHCA). The period between the occurrence of the OHCA and the initiation of CPR is known as the no-flow time (NFT), while the period after the initiation of CPR until return of spontaneous circulation is known as the low-flow time (LFT). The objective of this study was to describe the association between the NFT, bystander LFT and the clinical outcomes of patients with an OHCA. Methods: A multicentric North American registry was used to include adult patients who experienced a bystander witnessed non-traumatic OHCA. Patients not treated were excluded, as well as those for whom the bystander resuscitation status, or the time delay before the emergency medical services (EMS) provided compression was unknown. For the present study, the elapsed time from dispatch call to the first EMS-provided compression after the OHCA was considered the LFT for patients with bystander CPR and the NFT for patients without bystander CPR. The association between the NFT, LFT and the clinical outcomes (survival with a good neurologic outcome [Modified Rankin Scale 0-2], survival to hospital discharge) was evaluated using a multivariable logistic regression. Introduction: Point-of-Care Ultrasound-Guided Regional Anesthesia (POCUS-GRA) in the Emergency Department (ED) provides safe and rapid analgesia for older people with hip fractures. Objectives: To assess the impact of a knowledge to practice (KTP) intervention on (1) the attempt and success rate of POCUS-GRA by ED physicians and (2) the safety and efficacy of blocks performed. Methods: This is a planned nested observational study contained within a multicenter, stepped-wedge cluster randomized clinical trial at 8 academic emergency departments in four provinces. (NCT02892968) Participants: ED physicians working at least one shift per week, excluding those already performing POCUS-GRA more than 4 times per year. Hip fracture patients [ = 65 years of age who were not delirious on arrival and had operative treatment were eligible. Analysis: We report proportions and 95% Confidence Intervals for primary outcomes. We used the Hanley Rule of Three for confidence intervals with a ''zero'' numerator. Intervention: A 2-h structured KTP training sessions, using simulated and live models. We also sent e-mail reminders and provided a procedure bundle. Those patients have heterogeneous presentations and need seniorfriendly triaging tools. Systolic blood pressure (SBP) is commonly used to assess severity and some authors advocated adjusting SBP threshold. However, the literature is heterogenous and there is no consensus on which threshold is the most appropriate for older patients. We aimed to describe and compare the relationship between mortality and prehospital SBP in older patients (65-74 years and C 75 years) and their younger counterparts (16-64 years). Methods: Data from the Quebec Trauma Registry including patients admitted in three level I trauma centres (03/2003-12/2017) were analysed. We performed logistic regressions with age and SBP to obtain mortality curves. Three time-periods (2003-2007, 2008-2012 and 2013-2017) were defined to assess a potential cohort effect. We calculated sensitivity, specificity, positive and negative predictive values for each SBP cutoffs from 90 to 130 mmHg in 10-mmHg increments. Subgroup analyses were performed for major trauma admissions (ISS C 12) and severe traumatic brain injury (TBI) admissions (GCS B 8). Primary outcome was in-hospital mortality. Results: We included 47,661 trauma admissions. Mean ISS and mortality rates were 14 and 3.9%, 14 and 8.1%, 12 and 11.7% in the 16-64, 65-74, C 75 years age groups, respectively. The relationship between prehospital SBP and mortality was nonlinear. Mortality proportionally increased with each 10-mmHg SBP increments across all age groups, and this increase was greater in older patients. The maximum S34 Can J Emerg Med (2022) 24 (Suppl 1):S1-S100

Vol.: (0123456789) combination of sensitivity and specificity was found at a SBP of 130 mmHg for both older groups (37.5% and 69.5% in the 65-74years group; 31.7% and 76.0% in the C 75 years group). Conclusion: A SBP threshold of 130 mmHg may be more accurate to triage older patients with an increased risk of mortality compared to the standard 90 mmHg. However, relationship between SBP and mortality is nonlinear. Hence, binary cut-off should be abandoned, and risk prediction model studies are required to develop senior-friendly triaging tools where SBP should be integrated as a continuous variable and compiled with other predictors because of its low sensitivity to predict mortality.

Keywords: hypotension, older adults, trauma

Utility of repeat troponin testing in emergency department patients with syncope Introduction: For emergency department (ED) patients with syncope, cardiac troponin will help identify underlying myocardial infarction (MI) and is an independent predictor of 30-day serious adverse event (SAE). However, it is unclear if serial troponin testing will improve diagnostic yield and prognostication. Methods: This was a sub-study of two prospective multicenter studies conducted to derive and validate the Canadian Syncope Risk Score (CSRS). Adult patients (age C 16 years) presenting with syncope to one of 11 Canadian EDs between 2010 and 2019 were enrolled. Patients with an obvious serious condition (e.g. hemorrhage, arrhythmia), without troponin testing and those lost to follow-up were excluded. The primary outcome was detection of a serious condition (death, arrhythmia, MI, pulmonary embolism, hemorrhage, or others) in the ED or change in CSRS category leading to variation in 30-day SAE risk estimate based on repeat troponin testing. We conducted 30-day telephone follow-up to ascertain SAEs after the index ED visit and compared ED length of stay (LOS) between patients with a single and serial troponin measurements. Descriptive statistics and two-tailed ttest were used for analyses with p \ 0.05 considered significant. Results: Of the 9540 patients enrolled in the parent study, 4996 patients [mean age 64.5 years (SD 18.8 years), 52.2% male] were included in this sub-study. 4397 (89.8%) had a single troponin, of whom 232 (5.3%) patients had a SAE identified in the ED and 203 (4.9%) patients after ED disposition. Of the 499 (10.2%) patients with [ 1 troponin test, 39 (7.8%) suffered SAE in ED and 60 (13.2%) after ED discharge. 10 patients (2.0%) had a rise from below to above the 99th %ile, of which 2 patients (0.4%), with on-going chest pain/ shortness of breath, were diagnosed with a pulmonary embolism and early respiratory failure. Both were hospitalized, and the rest did not suffer SAE. Nine patients (1.8%) had CSRS risk reclassification based on repeat troponin (7 decrease and 2 increase) and none suffered any 30-day SAE. Mean ED LOS was significantly longer for patients with serial compared to single troponin (6.9 vs. 5.3 h, p \ 0.001). Conclusion: For ED syncope patients, repeat troponin testing did not improve the diagnostic yield or outpatient prognostication. ED syncope risk stratification based on first troponin is sufficient and repeat troponin testing should be reserved for patients with on-going symptoms or abnormal ECG. Keywords: serious outcome, syncope, troponin Introduction: For patients with atrial fibrillation seen in the Emergency Department (ED) following a transient ischemic attack (TIA) or minor stroke, the impact of initiating oral anticoagulation immediately rather than deferring the decision to outpatient follow-up is unknown. We sought to determine the association of starting anticoagulation in the ED and the outcome of stroke, TIA or death at 90 days. Methods: We conducted a planned secondary data analysis of a large prospective cohort of 11,434 adults in 13 Canadian EDs between 2006 and 2018. Patients were eligible for enrolment if they were aged 18 years or older, with a final diagnosis of TIA or minor stroke with previously documented or newly diagnosed atrial fibrillation. The primary outcome was subsequent stroke, recurrent TIA, or all-cause mortality within 90 days of the index TIA diagnosis. Secondary outcomes included stroke, recurrent TIA, or death, rates of major bleeding at days 2, 7, 30 and 90. A multivariable logistic regression model was utilized to evaluate the association between stroke, TIA and all-cause mortality within 90 days of index visit Results: Of 11,434 subjects with TIA/minor stroke, atrial fibrillation was identified in 11.2% (1,286 mean age 77.3 (SD 11.1) years, 52.4% male). Over half (699; 54.4%) of these were already taking anticoagulation, 89 (6.9%) were newly prescribed anticoagulation in the ED. By 90 days, 4% of the atrial fibrillation cohort had experienced a subsequent stroke, 6.5% subsequent TIA and 2.6% died. On multivariable logistic regression, there was no association between being prescribed anticoagulation in the ED and these 90-day outcomes (composite OR 1.37, CI 95% 0.74-2.52). Major bleeding was found in 5 patients (all intracerebral hemorrhages), none in the ED initiated anticoagulation group, 2 in those already taking anticoagulation and 3 in the group not on anticoagulation at the time of ED discharge. Conclusion: Initiating oral anticoagulation in the ED following new TIA was not associated with lower recurrence rates of neurovascular events or all-cause mortality in patients with atrial fibrillation. ED initiated anticoagulation was not associated with increased rates of major bleeding. Keywords: anticoagulation, emergency medicine, transient ischemic attack Introduction: EMS alternative destination pathways provide an opportunity for redirection of low-acuity 9-1-1 patients away from overcrowded emergency departments (EDs). This study analyses the accuracy of suspected diagnoses made by paramedics and their ability to independently identify appropriate destinations for their patients.

Methods: In this cross-sectional survey involving two tertiary EDs serving a population of 400,000, paramedics recorded their suspected diagnosis at patient handover and if an alternative destination would have been suitable (urgent care, family physician, release from care with no follow up). Two ED/EMS physicians determined the accuracy of the paramedic's suspected diagnosis by reviewing the ED diagnosis and decided if the alternative destination was suitable based on ED investigations, treatments, and diagnosis. A third ED/EMS physician was used when agreement was absent. Results: 493 patient transfers were included. 329 (66.7%) diagnoses were deemed a 'reasonable match' (rater agreement = 95%), 126 (25.6) were a 'miss' (rater agreement = 98%), and 38 (7.7%) were a 'critical miss' (highmorbidity or mortality pathology was present but not identified, rater agreement = 100%). Of cases (n = 161) Keywords: initial heart rate, out-of-hospital cardiac arrest, pulseless electrical activity

Adaptation of a patient-reported outcome measure tool to assess and compare health outcomes of ambulatory patients visiting emergency departments, walk-in or primary care clinics Introduction: The objective of this study was to adapt and validate the Patient-reported outcome measure for emergency department patients (PROM-ED) to compare the quality of care in walk-in and primary care clinics to that in emergency departments (ED). Methods: The PROM-ED was adapted following the guidelines for the cross-cultural adaptation of self-report measures described by Beaton et al. 2000 . The PROM-ED contains 20 questions on 4 domains of health outcomes and is administered to patients up to 10 days after their visit. Each domain is assessed independently with a score reported as a percentage: higher indicates a better health outcome. In this study, the PROM-ED was modified by a panel of experts including patient partners, clinicians, researchers and a linguist to ensure questions were adapted and relevant to all settings under study. It was pretested with an online survey of potential users to assess whether it is intelligible, clear, useful, devoid of redundancy and if it has good apparent and content validities (1 = strongly disagree to 9 = strongly agree). A prospective cohort of ambulatory adults with acute respiratory conditions was recruited from an ED and 2 clinics in Québec city to pilot the PROM-ED in a real-life situation. Innovation Concept: Emergency physicians (EP) are expected to be competent in a variety of uncommon but life-saving procedures, including the bougie assisted cricothyrotomy (BAC). Mental practice (MP), defined as the ''cognitive rehearsal of a skill in the absence of overt physical movement'', has been shown to be as effective as physical practice in several areas of medicine. MP scripts incorporate cues from different sensory modalities to supplement instructions of how to complete the skill. Previous work has described the sensory experience of EPs completing the procedure. We describe the integration of sensory cues with a delphi-derived list of procedural steps into a cohesive script to facilitate MP. Methods: Data collection occurred through in-depth semi-structured qualitative interviews with EPs at a single tertiary care centre. Interviews were recorded and transcribed verbatim. Transcripts were coded using qualitative content analysis on a coding framework based on a previously published list of procedural steps. At each procedural step, cues that enhance MP were extracted and counted. Two investigators identified common themes among cues at each step. These cues were integrated with procedural steps in an iterative fashion to create a coherent narrative script. Curriculum, tool, or material: Eight interview transcripts were analyzed. All interviewees were EPs with experience completing a BAC in clinical practice. A total of 328 cues were identified. On average, each EP identified 13.7 cues for each procedure. All identified visual, kinesthetic and cognitive cues at key procedural steps. The final script was written from a first person perspective and incorporated a colour coding scheme to highlight important cues. The design and layout of the script was refined to produce a visually interesting format to facilitate readability and hold users attention.

Conclusion: EPs were able to describe sensory cues at each step, including cues not present in traditional descriptions of the procedure. To our knowledge, this work represents the first attempt to combine cues identified by experienced practitioners along with procedural steps with the aim of supporting rich mental representations of a rare procedure. We expect that this script will be useful to EPs and others seeking to improve their skills in this rare procedure. Future work will evaluate the effects of the script on EP confidence and competence with the BAC. The methodology presented here may also be useful to those seeking to develop MP scripts for other skills. Keywords: innovations in emergency medicine education, mental practice Innovation concept: The majority of suspected or confirmed pregnancy losses occur in the first trimester, and these patients frequently seek care in the emergency department (ED). Our antecedent work demonstrated ED care often does not meet patient and family expectations and needs, and patients repeatedly identified a lack of educational resources for this prevalent, painful, and psychologically devastating condition. The objective was to create an innovative, freely accessible, patient and provider co-designed, web-based education platform for ED patients experiencing symptoms of early pregnancy loss. Methods: To develop our web-based patient education resource, we implemented a multi-modal approach. We first conducted a literature review to ensure up-to-date and relevant content for the website. We then conducted six semi-structured interviews of patients with lived experience, ED nurses, a gynaecologist, and an ED physician to seek their input for the website's organization and validate the content informed by our literature review. Finally, we used the thematic analysis of the qualitative interviews to structure the framework for the website's design. The initial version underwent an assessment using the Patient Education Materials Assessment Tool for Audiovisual Materials (PEMAT-A/V) and iterative feedback was received from patients and content experts, and the website was revised accordingly. Curriculum, tool, or material: four themes emerged from the interviews: the importance of validating the experience of guilt; the need for reassurance and support; need for clear and transparent medical information; and ensuring the website was both inclusive and accessible to ED patients. From this analysis and literature review, the website pregnancyemerg.ca was developed in consultation with a medical education web-designer, a gender studies expert and web-based patient education consultant. The website will be publicly available, and patients in two local EDs will receive a pamphlet with the website's QR code during ED visits for symptoms of early pregnancy complications. Conclusion: We identified an urgent need for a patient-centred, expert and patient-informed resource that can be freely accessed online in real time by patients experiencing complications in early pregnancy. This website seeks to inform patients about the physical and psychological manifestations of early pregnancy loss. This patient-informed, web-based education platform is a critical first step in advocating for improved maternal outcomes in Canada. (3) simulation. This approach suits various learning styles and strengthens autonomy by ensuring flexibility and accessibility. Methods: Students are to complete readings from the Ottawa EM Handbook. This is followed by asynchronous lectures and RLOs, which review of clinical skills and procedures and include automated grading of formative assessments. They supplement the interactive, virtually delivered small-group tutorials. While these sessions are instructor-facilitated, they are student-centred, contextually based, and foster accountability. Simulation allows learners to consolidate skills and for tutors to assess learners. Instructors perform debriefing and complete direct observation and formative assessments. Summative exam is converted online. Curriculum, tool, or material: personnel required: central IT staff for server, network monitoring, storage, and VLE maintenance; undergraduate medical education (UGME) and EM administrators to provide learner access and to assist with moderating, troubleshooting, and recording. Software: Teams, VLE, online assessment platform, simulation technology and automated captioning system/transcription to ensure accessibility. Evaluation using RUFDATA and Kirkpatrick frameworks are ongoing. In 2020, the EM clerkship program was voted to be the best rotation per Graduating Questionnaire (Kirkpatrick 1). Additional data will be obtained from multiple stakeholders (UGME/EM leaders, faculty, students, administrators) through focus group. Conclusion:

The drivers for this curriculum, such as fiscal reality and COVID-19, are here to stay. It will be important to have TEL ready to prepare for future crises. Economy of scale provided by blended learning will be supported by the reusability of materials in our curriculum. Ongoing program evaluation and maintenance cost will be assessed to ensure that our program continues to provide the best learning experience. Keywords: clerkship, innovations in emergency medicine education, technology-enhanced learning evidence to support it, this belief may be a barrier to performing the FNB in this patient population. The objective of this study is to assess if pain reduction from the FNB varies depending on the anatomical level of hip fracture. Methods: A secondary analysis of data collected previously in a prospective cohort study. Study emergency physicians completed a two-hour training session on US-guided FNB using simulators and live models. Patients were enrolled in the study if they were 65 years or older and presented to the ED with a hip fracture. Pain scores on a 10-point numeric scale were recorded pre and post the block performance. We categorized patients according to the level of the fracture into two subgroups (intracapsular and extracapsular). The mean reduction in pain score in 30 min interval was compared using students t-test with statistical significance set at p \ 0.05. Results: A total of 162 patients underwent FNB in the ED. Pain scores before the block and 30 min after were recorded in only 64 older people all of whom were included in our analysis. Among these 64, 28 (44%) had intracapsular and 36 (56%) had an extracapsular hip fracture. The mean age in each group was 77.9 and 82.3 years respectively. The mean reduction in pain was (-3.61/10 SD 2.57) and (-3.39/10 SD 3.33) respectively. The T-value (pooled variance) was 0.29, with 62 degrees of freedom and a P-value of 0.7756. Our results show that the pain reduction in the intracapsular group was at least equivalent to the extracapsular group. In fact, a non-statistically significant difference was in favour of the intracapsular group (-3.61 vs -3.39). Conclusion: The ultrasound-guided femoral nerve block is equally effective in intracapsular and extracapsular hip fractures. Anatomical level of the fracture should not be a barrier to providing fast, effective analgesia in older people suffering from a hip fracture in the ED.

Keywords: emergency analgesia, emergency ultrasound, femoral nerve block Introduction: Delays in identifying and providing definitive treatment to severe thoracic and abdominal injuries (TAI) are among the most frequent cause of preventable mortality following a trauma. However, the identification of life-threatening TAI can be challenging, particularly in the out-of-hospital setting which is characterized by its limited resources and chaotic environment. We performed a systematic review aiming to identify the predictors of internal TAI potentially applicable in the prehospital setting. Methods: This is a systematic review. We searched Medline, Embase and Cochrane library. Studies were considered eligible for inclusion if they reported original data on the association between a potential prehospital predictor and internal TAI. Exclusion criteria were: case reports, simulation studies, studies in which more than 10% of the patients were not transported by EMS and studies with no metrics on prehospital predictors of TAI. Three authors assessed the titles, reviewed the selected manuscript and extracted the data. The risk of bias was assessed using the ROBINS-I tool. Results: Our initial search strategy identified 4266 citations of which 190 manuscripts were fully reviewed. Finally, a total of 13 articles were included for a total of 41,762 patients. Four studies were performed in Netherlands and two studies in the USA. Potential predictors were the use of ultrasound 236 trauma field triage protocols (n = 7) and physiological variables such as lowest systolic blood pressure and heart rate. The sensitivity of the ultrasound to predict 43 thoracic injuries ranged between 38 and 100% and 95.6% and 100% specificity while a sensitivity of 31.3% and a specificity of 96.7% to predict intraabdominal free fluid (n = 92) . The sensitivity of field triage protocols was 72% to 94% and specificity of 40% to 78%. The outcomes assessed were presence of a pneumothorax, presence of free fluid, urgence surgery etc. More than half (n = 7) studies were able to clearly identify thoracoabdominal injuries, eight could predict an internal injury and six could predict a surgery in the first 24 h.The included studies were considered at low (n = 6), moderate (n = 5) or high (n = 2) risk of bias. Conclusion: The use of prehospital ultrasound by trained EMS providers seems to be the most efficient and accurate way to identify TAI. In the BLS system, provider should rely on a good physical exam with signs of trauma to the abdomen or thorax and vital signs that might indicate a state of shock. Keywords: predictor, prehospital, thoracoabdominal

The impact of out-of-hospital time on return of spontaneous circulation following resuscitative thoracotomy in traumatic cardiac arrest Introduction: Trauma is a lead contributor to disease burden worldwide. As trauma systems advance, an increasing number of studies are evaluating factors impacting survival in traumatic cardiac arrests (TCA) that require resuscitative thoracotomies (RT). While this informs management guidelines, data has primarily focused on injury mechanism, and patient and in-hospital factors. Prehospital variables have not been extensively explored. Therefore, this study evaluates the impact of out-of-hospital time (OOHT) on return of spontaneous circulation (ROSC) in TCA. Methods: This was a retrospective cohort study evaluating tier 1 and 2 trauma activations at two level-1 trauma centres in Toronto, Canada (January 1, 2010-December 31, 2020). Trauma patients who experienced a TCA in the prehospital setting or in the trauma bay, and who required a RT within 60 min of hospital arrival in the emergency department or the operating room, were included. The primary and secondary outcomes were ROSC and survival to hospital discharge, respectively. Results: A total of 176 patients met inclusion criteria. The mean age of our study population was 34 years (SD 16) and 15% were female, while 85% were male. Ninety-seven percent of RTs occurred in the emergency department setting. Thirty-two percent of patients achieved ROSC following RT, with 5% of all patients surviving to hospital discharge. The median time to RT from trauma bay arrival was 5 min (IQR 2-11) and the median OOHT of all patients was 26 min Introduction: Aiming to optimize emergency healthcare resource use, prehospital and emergency department (ED)-initiated redirection programs for patients with low acuity conditions have been implemented in many jurisdictions. This systematic review aims to determine the impacts of dispatch, prehospital or ED-initiated strategies to redirect patients with low-acuity conditions to non-ED settings on system-and patient-level outcomes. Methods: This is a systematic review (Prospero: CRD42021230545). Five electronic databases were searched. Controlled trials and cohort studies were considered eligible for inclusion if they reported on a dispatch, prehospital or an ED-initiated diversion strategy implemented to redirect patients with a low-acuity condition to a non-ED setting. The main outcome was the need for a subsequent emergency medical services (EMS) request or ED visit \ 30 days. Secondary outcomes included subsequent EMS requests or ED visits \ 7 days, safety metrics and patient's perspective on redirection. Results: Twenty-nine studies were included which reported on dispatch (n = 5), EMS (n = 13) or ED-initiated (n = 11) redirection. Eight studies were randomized controlled trial. Seven studies limited the inclusion to older adults ([ 60 years old) and one to pediatric patients (\ 18 years old). Within 30 days, the proportion of ED returns varied between 1.2% to 62.6% among patients redirected (n = 9). Overall, \ 7% of redirected patients revisited the ED \ 7 days following their index consultation (n = 4) . Regarding safety, \ 2% of patients were considered inappropriately redirected to a non-ED setting (n = 3). All studies reporting on patients' satisfaction suggested similar or improved satisfaction for those redirected compared to patients who had received standard ED care (n = 13) . ED-initiated strategies seemed associated with more reported barriers such as the requirement for another transport and the perception that delays are shorter, and quality of care is superior in the ED compared to non-ED settings.

The risk of bias of the included cohort studies was serious (n = 6), moderate (n = 13) or low (n = 2). Conclusion: Various programs of redirection strategies have been implemented. Their impacts varied greatly with low to very high returns to ED within 30 days following the index visit and those redirected seems to be satisfied. Methodologically robust studies are required to assess the impacts of redirection programs at a system-level. Keywords: emergency medical services, low acuity, redirection Introduction: Previous literature has identified that decision fatigue is a factor which might influence prescribing 1,2. However little evidence exists on how this might affect diagnoses and whether later decisions are incorrect (''fatiguing down'') or initial decisions are incorrect (''tuning up''). The purpose of this study is to determine how timing and order of patients over emergency shifts are associated with receiving diagnostic imaging in the emergency department, in order to further understand the effect of decision fatigue in this setting. Methods: We examine how timing and order of patients is associated with receiving diagnostic imaging in Niagara Health emergency department (ED) using 1.1 million visits between April 2013 and March 2019. Outcomes are whether a patient receives an x-ray, a CT scan, or an ultrasound with seven day admission and re-presentation to ED rates. Variables of interest are the minute in the physician's shift that a patient is seen and in how many patients the physician has previously seen. Results: Relative to the first patient, the probability of receiving an x-ray, CT scan, or ultrasound decreases by 12%, 15%, and 4% respectively if a patient is the 15th patient seen during a shift (p \ 0.05, CI 95%). Relative to the first 15 min, these probabilities increase by 4%, 4%, and 0.3% respectively if a patient is seen in the 180th minute of a physician's shift (p \ 0.05, CI 95%). Seven day admission or re-presentation rates did not show any significant association with patient order or timing in a shift and imaging ordered (p \ 0.05, CI 95%). Conclusion: Diagnostic imaging in the ED is associated with patient order and shift length; more so with patient order. This suggests that physicians make different imaging decisions over the course of a shift. The initial higher probability of receiving imaging does not translate into reductions in subsequent admissions to hospital supporting a theory of ''tuning up'' of diagnostic ability on earlier patients. This study holds implications for remodelling ED shift structures to optimize resource utilization and patient outcomes. Adult patients were included if they presented with acute ambulatory respiratory conditions potentially treatable in any of these settings. Participants were asked to answer the motivation questionnaire during a follow-up phone call one to three days after their visit. Mann-Whitney U tests were performed to compare answers in both groups (a = 0.05). Results: 28 ED and 32 WIC patients were recruited. The mean age (SD) of participants was 38 (15) years old and 66% of them were women. Patients were more likely to consult at the COVID WIC because they knew the time of their appointment (P \ 0.001) or someone recommended them to go there (P = 0.01). Alternatively, patients preferred to go to the ED because that is where they usually seek care (P \ 0.001), their medical history is known there (P \ 0.001) or out-of-pocket costs to see a healthcare professional are lower (P = 0.0498). Accessibility and perceived illness severity did not differ between groups. Income and level of education were similar between groups. Conclusion: Starting in October 2022, this validated questionnaire will be used in a multicenter prospective cohort study in Québec and Ontario that will compare the value of care in EDs, WICs and PCPs for ambulatory patients with acute health concerns. Keywords: emergency department, patients' motivation, walk-inclinic Can J Emerg Med (2022) 24 (Suppl 1):S1-S100 S41

Vol.:(0123456789) Introduction: Workplace-based assessments (WBA) are an important tool for trainee feedback and as a means of reporting expert judgments of trainee competence in the workplace. However, the literature has shown that gender bias can exist within these assessments. We aimed to determine if there is a gender-related quality gap in workplace-based assessments in our residency training program. Methods: This study was conducted at the University of Ottawa in the Department of Emergency Medicine. Four end-of-shift WBAs completed by male faculty and 4 completed by female faculty were randomly selected for each residents during the 2018-2019 academic year. Two blinded raters scored each WBA using the Completed Clinical Evaluation Report Rating (CCERR), a previously published nine-item quantitative measure of WBA quality. A 2 9 2 mixed measures analysis of variance (ANOVA) of resident gender and faculty gender was conducted, with mean WBA rating as the dependent variable. The ANOVA was repeated with mean CCERR score as the dependent variable to examine the impact of gender on the quality of WBAs. Results: A total of 363 WBAs were collected for 46 residents. There was no difference in mean WBA ratings between female and male residents (p = 0.92), and no interaction between resident and faculty gender (p = 0.62). Mean CCERR score was 25.78, SD = 4.15, indicating average quality assessments. There was no significant effects off faculty or resident gender on the quality of WBAs. Conclusion: We did not find faculty or resident gender differences in the quality of workplace-based assessment completed in our training program. While the literature has previously demonstrated gender bias in trainee assessments, our results are not surprising as assessment culture varies by institution and program.

Our study highlights the importance of not generalizing gender bias across contexts and offers a simple method that educators can use to conduct their own evaluation to determine if gender bias in trainee assessments exists within their program.

Keywords: assessment quality, gender bias, postgraduate training

The learning impact of a virtual CPR webinar for seniors Although CBME aims to be learnercentered, the current literature largely focuses on faculty or administrator perspectives. However, understanding the resident perspective is essential, particularly since assessments perceived to be inauthentic or inaccurate may interfere with learning and professional development. Our study explores residents' real-world perspectives of EPAs, including their perceived impact on learning. Methods: Using a constructivist grounded theory approach, we conducted 18 semistructured interviews with residents from disciplines that have implemented CBME at one tertiary care academic center in Canada. Data collection and analysis occurred iteratively, and themes were identified using constant comparative analysis. Results: Residents' perspectives were mixed, seemingly influenced by the attitudes and behaviors of their respective programs and faculty. In programs where EPAs were perceived as valuable, residents described a culture where EPAs were seamlessly integrated into daily routine. Faculty were both capable and willing to do these assessments, and residents intentionally sought EPAs for skillsets in which they were less confident, with the aim of eliciting specific, actionable feedback. Conversely, other programs were described as having less faculty buy-in, and residents perceived EPAs as a strictly quantitative requirement that ultimately interfered with learning and caused undue stress. In turn, these residents completed EPAs solely to fulfill perceived program requirements and advance to the next stage. Conclusion: Authenticity matters. For EPAs to be perceived as truly reflective of residents' lived experiences and as a feasible means of formative feedback and accurate assessments, the CBME learner-centered approach necessitates a supportive educational environment with palpable faculty buyin. Residency programs must build a culture that incorporates frequent low-stakes EPA assessments into daily routine and communicate EPAs as authentic facilitators of formative learning rather than arbitrary checklists for generating quantitative metrics. Introduction: Security services are an essential part of emergency department (ED) care, but their role remains poorly described. While not always directly related to violence or patient agitation itself, describing the security service's role in care may help obtain a more complete picture of the exposure to violence and agitation that staff face in clinical care. We sought to (1) identify how often security services are involved in ED care and (2) identify trends related to call frequencies.

Methods: This is a retrospective observational study at an urban academic tertiary care hospital from January 2017 to June 2021 using a database that security services maintain. We described how often security services were involved in care provision and categorized the calls based on their descriptions. Calls were adjusted for patient volumes for each year. We used linear regression to look for increases in events security services were involved in. We used ANOVA to analyze for correlation in the number of events related to shift time, day of week, and month of the year. Results: The database identified 20,033 cases of security service involvement over 4.5 years related to care in the ED. On average, security services had 6,506 involvements per 100,000 patient visits. Records were sorted into 6 categories [Code White, Patient Discharge, Medications, Restraint, Patient Transfer/Escort, and Assist-Standby]. Our analysis showed significant increases in total events over the study period (4,356 in 2017 to 9,897 per 100,00 patient events in 2021) and across all categories of events security services were involved in. Call volumes were not significantly different based on day of week or month, but were significantly different based on shift time. Conclusion: This shows how closely tied security services are in the daily operations of an ED. While limitations exist with this type of database, it provides an indirect measurement of the agitation that clinical staff are exposed to. It also shows a growing reliance on security services and highlights the importance of developing alternative strategies to prevent and de-escalate violent behaviour, especially in the evening and overnight periods when fewer allied healthcare staff, such as social workers are present. The database shows the need to better understand the patients that require security services and associated underlying factors to improve care experiences for both staff and patients. Keywords: patient agitation, workplace violence Introduction: The airway registry is a common way to report success rates and complications in intubation attempts. In 2019, our academic centre established an emergency department airway registry. We monitor four Key Performance Indicators (KPIs): post-intubation hypoxia, hypotension, first pass success (FPS), and \ 90 s intubation time, as well as intubation preparation, attempts, and complications. Data is collected through a form completed at the bedside by health care practitioners post-intubation and has proven useful as a quality assurance tool but has not been validated. No gold standard is defined for documentation of emergency intubation; therefore, we compare our paper data extraction tool to a formal chart review as it may represent a more rigorous methodology. Methods: To validate our data, two stages were devised. Firstly, we reviewed previously entered registry records with the corresponding paper form to identify rates of correctly assigned KPIs. This led to implementation of a Standard Operating Procedure (SOP) to minimize entry variability. In the second stage, a retrospective chart review was completed analysing a random subset of intubations performed at two academic hospitals. Based on previously published data comparing chart reviews to bedside data collection tools, we calculated a necessary sample size of 174, 87 charts and 87 paper forms, for a significance level of 0.05 with a power of 80%. The primary outcome was agreement on our four KPIs between the two data sets and was assessed using Cohen's Kappa. Results: 40 intubations were examined to validate the inter-rater reliability of our airway registry and 25% of cases had KPIs incorrectly assigned. Following implementation of the SOP and re-entry of all airway registry forms, the retrospective chart review was performed. Background: In response to an organizational survey revealing low safety culture scores, we implemented a ''zero harm'' approach to eliminate preventable harm across a wide variety of clinical areas. We aimed to achieve this objective within three years. Our aim was to eliminate preventable harm across 5 clinical areas within 3 years: achieving zero falls with injury; eliminating hospital-acquired pressure injuries (HAPIs); minimizing central line-associated blood stream infections (CLABSIs); avoiding irretrievable specimens; and maximizing medication reconciliation rate. Our goal was to make 33% improvements in these areas over our first year. Aim Statement:

We developed a five-part strategy for cultural and process redesign that included (1) engaging leadership, (2) developing an organizationspecific patient safety framework, (3) monitoring specific quality aims based on high-risk, high-volume, high-cost and problem-prone areas, (4) standardizing a three-part review process that includes a root cause analysis for moderate and critical patient safety incidents, (5) communicating progress to staff in real time via unit-specific electronic dashboards. Measures and design: In less than one year, we increased patient safety incident reporting by 37% while simultaneously decreasing falls with injury by 39%, pressure injury rates by 37% and central line-associated blood stream infections (CLABSIs) by 34%. We also improved medication reconciliation rate by 3.3% and decreased our irretrievable specimen rate to 0. Finally, we noted increased awareness around patient safety within clinical teams, with open discussions about patient safety becoming a routine part of patient care. Evaluation/results: This study describes an initiative that sought to introduce system-wide changes to practice and patient safety culture in a rapid time frame. Results suggest that our five-step approach to transformation may confer substantial gains in patient safety for peer institutions. Next steps include continuing to expand and monitor quality aims as we progress through our journey to eliminating preventable patient harm in our healthcare system. Keywords: quality improvement and patient safety, safety culture, zero harm

Reducing opioid prescriptions from the emergency department: a two-pronged quality improvement initiative Background: Pain is one of the most common reasons for visiting the emergency department (ED). While adequately managing a patient's pain is important, the liberal prescription of opioid medications can be problematic. Studies show that more than 60% of opioids prescribed by emergency physicians (EP) are unused. These opioids pose a risk for abuse, diversion, and unintentional ingestion. To decrease the quantity and frequency of ED opioid prescriptions by 15% in 1 year ending September 2021. Aim statement: We conducted a twopronged intervention to decrease opioid prescribing practices for discharged ED patients. We implemented a default quantity on all electronic opioid prescriptions at 10 tablets. EP could modify the default quantity up to 30 tablets. A quarterly audit on physicians' opioid prescribing practices was conducted. Anonymized data was distributed to physicians with comparison to other EP, the group mean and highlighting those more than 1 standard deviation above the mean. PDSA cycles in design of default electronic opioid prescriptions and graphic display of audit data were conducted. Three reviewers completed titles and abstracts screening, full text screening and data extraction, with conflicts resolved by a fourth party. The NIH quality assessment tool for before-after (Pre-Post) study was used to assess studies. Narrative synthesis of data was performed as meta-analysis was not conducted due to heterogeneity between studies. Results: We identified 2196 potential studies and 21

were included. All studies included interdisciplinary healthcare workers who worked in the ED, totalling 1306 participants across the studies. All but one of the studies were prospective. Most of the studies' objectives were to solely detect LST in the ED (n = 10), with the remainder being in the era of Covid-19 (n = 5), while opening a new facility (n = 4) and during a mass casualty simulation (n = 2). The methods for assessing LST ranged from debrief, post-simulation surveys, video and audio review. All studies reported identifying LST through either a single or multiple ISS sessions. The number of LST detected ranged from 32 to 843 in the included studies. The LST detected across studies can be grouped into 3 categories: System errors (i.e. problems with guidelines, procedures and staffing), individual errors (i.e. task specific challenges, knowledge and communication), and environmental errors (i.e. organization of rooms, location of equipment and medications). Most of the studies were of good quality (n = 14) , with 4 being rated as ''fair'' and 3 as ''poor'', mostly due to lack of reporting specific information on participants. None compared ISS to other methods for LST detection.

Conclusion: This systematic review demonstrates that ISS is consistently effective in detecting LST in various ED contexts. These results can be used to create and implement targeted interventions to mitigate system errors in the ED and improve quality of patient care. NPT provides a framework to understand how individuals and teams navigate the process of embedding new models of care as part of normal practice. All physicians who had worked in the virtual ED were invited to participate. Thematic analysis, using inductive coding and deductive coding to NPT domains, was performed to describe the manifest and latent content. Results: A total of 14 physicians were interviewed. The data illustrated two groups: one group moved to normalize the virtual ED in practice, while the other was unable to fully adopt it. These groups differed in their perception of the patient benefits as well as the criticality of their role in the virtual ED. The first group saw value for patients (coherence) and were motivated by patient satisfaction and relief witnessed (reflexive monitoring) at the end of the virtual appointment. In contrast, the other group did not find virtual ED work reflective of urgent care (cognitive participation) and felt their skills as ED physicians to be underutilized. For them the virtual ED more closely resembled primary care. The limited ability to examine patients and a sense that patient issues were not fully resolved at the end of the appointment caused frustration. These physicians signed up for fewer shifts and did not experience the continuously evolving model and enhanced capabilities of the virtual ED platform. Introduction: At the start of the COVID-19 pandemic, emergency department (ED) volumes decreased. Unanticipated effects of delaying or not presenting to the ED for emergent conditions may result in adverse events. The objective of this study was to examine ED volumes for two emergent surgical diagnoses during the pandemic and compare outcomes during the pandemic to a historical control period. Methods: This was a retrospective population-based study using Ontario administrative data. We selected two surgical diagnoses not expected to change during the pandemic (appendicitis and ectopic pregnancy). We examined frequency of monthly ED visits for each diagnosis during the pandemic. Secondary outcomes (ambulance arrival, hospital admission, surgery and 30-day mortality) were compared among patients with each diagnosis during the pandemic (Mar 15 -Dec 31, 2020) compared to a control period (Mar 15-Dec 31, 2018 and 2019). We conducted multivariable regression to examine secondary outcomes during the pandemic compared to the control period. A sensitivity analysis of the first 1.5 months of the pandemic (Mar 15-May 31, 2020) examined differences in outcomes limited to the start of the pandemic. Results: There was an initial reduction in monthly ED visits for appendicitis (791 visits Feb 2020 vs 650 visits Apr 2020; D17.8%) and ectopic pregnancy (297 visits Feb 2020 vs 221 visits Apr 2020; D25.6%) during the pandemic, which then returned to baseline. Patient baseline characteristics did not vary between the control period and pandemic. Patients seen during the pandemic had increased odds of ambulance arrival compared to the control period (appendicitis OR: 1.47, 95% CI 1.35-1.61; ectopic OR: 1.41, 95% CI 1.16-1.71). In the sensitivity analysis, patients with appendicitis at the beginning of the pandemic were less likely to receive surgery compared to the control period (OR: 0.72, 95% CI 0.60-0.85), while there was increased odds of surgery among patients with an ectopic throughout the pandemic (OR: 1.28, 95% CI 1.04-1.55). There were no differences in other outcomes. Conclusion: Ontario ED volumes for appendicitis and ectopic pregnancy during the pandemic initially declined but quickly returned to baseline. Patients with appendicitis were less likely to receive surgery at the start of the pandemic compared to control periods, however, this difference dissipated as the pandemic progressed. Patients with an ectopic were more likely to undergo surgery during the pandemic. Keywords: COVID-19, patient outcomes, volumes

Paramedic-led fall assessment protocol to identify older adults who could safely be treated on-site and referred to an outpatient resource: a feasibility study Introduction: The number of emergency medical services (EMS) interventions for older adults following a fall from standing is increasing. Approximately two thirds of these patients are discharged directly from the emergency department (ED). This study aims to examine the feasibility of implementing a fall assessment protocol administered by emergency medical technicians (EMT) on-site to determine which patients can be safely referred to an outpatient specialized resource without being transported to an ED. Methods: This is a pilot study. A standardized fall assessment protocol was created and adapted to the prehospital setting. A selected group of EMTs in the city of Sherbrooke (Quebec, Canada) were thereafter trained. The on-site fall assessment was administered to older adults ([ 64 years old) who requested an ambulance after having fallen from standing (or lower) between October 2019 and March 2020. To ascertain safety, patients were transported to the ED regardless of the findings. Data were collected prospectively. The primary outcome was the presence of a condition deemed to require ED care. Secondary outcomes included the potential reduction of non-essential transport to the ED. Results: A total of 125 patients were included of which 36.0% were female and the mean age was 84.3 (standard deviation (SD) 7.9) years old. The initial fall assessment protocol would have recommended transport to the ED for 108 patients (86.4%). Following the EMS intervention, 111 patients were transported to the ED as 14 patients refused to be transported. The duration of the EMS assessment, including the fall assessment, was 31 (SD 11.4) minutes. Most patients (57.7%) were discharged from the ED. Infections and fractures were the most common admission diagnoses. Four patients (23.5%) for whom the fall assessment protocol suggested that transport to the ED was not required were admitted of which two (11.8%) died during their hospitalization. Conclusion: This study showed that the implementation of an EMT-led fall assessment protocol is feasible. However, before widespread implementation, a fall assessment protocol needs to be robustly developed. The upcoming steps are to appropriately derive a protocol that could have a stronger impact on resource utilization and be used safely in our context. Keywords: emergency medical services, fall, geriatric

A cohort study of high use of the emergency department among adult patients in Ontario

Can J Emerg Med (2022) 24 (Suppl 1):S1-S100 WIC. Inclusion criteria were: i) patients over 18 years old, ii) ambulatory during the entire visit, iii) discharged home with a diagnosis of URTI, pneumonia, acute asthma AE-COPD. The cost of care processes administered by fiscal year was estimated with time driven activity-based costing. Primary outcome was the proportion of patients returning in any ED or WIC at 72 h and 7 days after the index visit. Secondary outcomes were the mean cost of care for the index visit, the incidence of antibiotic prescription (patients with URTI) and the compliance to guidelines on use of antibiotics (patients with pneumonia, AE-COPD). Data were extracted from electronic medical records and provincial physician billing database. Multilevel generalized linear models were used and an overlap weight approach to adjust for confounding ( (COPD) is associated with exacerbations and high risk of serious outcomes. Our goal was to determine the appropriateness of the ED management of COPD exacerbations (COPDe) and short-term serious outcomes.

Methods: This was an observational cohort study by health records review of COPDe cases seen from June 30th, 2017, to July 1st, 2020 at 2 large academic EDs. We included all patients with the primary diagnosis of COPDe. The electronic medical record (EMR) for each case was reviewed by two learners and issues were reviewed by the faculty supervisor. Demographic and clinical data were abstracted into an electronic database and the Ottawa COPD Risk Score (OCRS) was calculated for each. Short-term serious outcomes (SSO) included ICU admission, endotracheal intubation, myocardial infarction and NIPV. We also noted death at 30 days. Cases were judged for appropriateness of treatment according to explicit indications and standards developed by a respirology and an ED physician for inhaled beta agonists, inhaled anticholinergics, IV/PO corticosteroids, IV/PO antibiotics, oxygen therapy, NIPV, and endotracheal intubation. Results: We enrolled 500 cases with mean age 71.9, female gender 51.2%, admitted 50.2%, and death 4.4%. The calculated OCRS score was [ = 2 for 70.8% of patients. The treatments provided were inhaled beta agonist (82.6%), inhaled anticholinergic (76.6%), corticosteroids (75.2%), antibiotics (71.0%), oxygen (63.8%), NIPV (8.8%) and endotracheal intubation (0.6%). Overall, 50.0% of cases were judged to have had inadequate management due to missing treatments. Specifically, the proportion of missing treatments were inhaled beta agonist (17.0%), inhaled anticholinergic (22.6%), corticosteroids (24.4%), antibiotics (12.8%), oxygen (0%), NIPV (2.0%) and endotracheal intubation (0% Introduction: Many patients with alcohol use disorder (AUD) present to the emergency department (ED), providing an important opportunity for early intervention. There are several effective pharmacological interventions for the treatment of AUD which effectively decrease alcohol consumption and promote abstinence, and access to Rapid Access Addiction Medicine (RAAM) clinics has made these treatments more widely available. The objective of this study was to determine the frequency of ED prescribing for anti-craving medications for AUD and determine the number of ED patients referred to a RAAM clinic. Methods: This was a single center retrospective chart review of adult (C 18 years) patients presenting to an urban academic ED (annual census 65,000) between Jan 2020-June 2021 with any alcohol-related complaint. Patients were excluded from the analysis if they left against medical advice (LAMA) or if they already had an addiction medicine physician as part of their healthcare team.

Results: During the study period, there were 2,138 ED visits related to an alcohol-related presenting complaint. Of the 200 (9.4%) randomly sampled charts reviewed, 17 (8.5%) patients LAMA and 23 (11.5%) patients were already followed by addiction services. Of the remaining 160 patients included in the analysis, 112 (70.0%) were male, 32 (20.0%) were homeless, mean (range) age was 42 (16-82) years, and 24 (15.0%) had known polysubstance use. 16 (10.0%) had an ED diagnosis of alcohol withdrawal. 34 (21.2%) patients were offered counselling, 25 (15.6%) were referred to a RAAM clinic, and 13 (8.1%) patients were referred to their primary care provider for ongoing management. No patient was administered anti-craving medication in the ED. 141 (88.1%) patients were discharged home, 15 (9.3%) were admitted for an alcohol related illness or medical problem exacerbated by alcohol use, and 4 (2.5%) were transferred to another healthcare facility. Of those discharged home, 1 (0.7%) patient was prescribed naltrexone. Conclusion: ED patients with presenting complaints related to the use of alcohol frequently display symptoms associated with AUD including alcohol withdrawal and have a high rate of admission. Despite the availability of effective treatment strategies for AUD including anti-craving medications and specialized addiction care, their use remains low. Additional research is required to identify barriers for providers offering these therapies in the ED setting. Keywords: addiction medicine, alcohol use disorder, anti-craving medication

Hepatitis B immune status of healthcare workers presenting to the emergency department for assessment after blood or body fluid exposure secondary objective was to compare the HB non-immune status rates between Saskatchewan Health Authority (SHA) HCWs and non-SHA HCWs with BBFEs. Methods: Following ethics approval, a retrospective chart review was conducted on a convenience sample of patients presenting between January 1st, 2020 and December 31st, 2020 to any of the three emergency departments (EDs) in Saskatoon, Saskatchewan with chief complaints or discharge diagnoses related to BBFEs, needle pokes, or bites. Collected data included age, sex, chief complaint, occupation, type of exposure, and HB surface antibody (anti-HBs) titres. Patients were considered HB non-immune if anti-HBs titres were \ 10 IU/L. To examine the HB non-immune status rates between HCWs and non-HCWs with BBFEs or between SHA HCWs and non-SHA HCWs with BBFEs, first Chi-square tests were used, and then multivariate logistic regression models were performed controlling for some demographic, geographical and clinical covariates. Results: Of the 1034 patient charts reviewed for eligibility, 345 were included. HB non-immune status rates were 34/256 (13.3%) among HCWs, and 20/89 (22.5%) among non-HCWs (p \ 0.001).

HCWs were more likely to be exposed to blood (36.7% vs 14.6%, p \ 0.001), while non-HCWs were more likely to be exposed to saliva (20.2% vs 8.6%, p \ 0.001 (2) How does COVID-19 impact the gap? Methods: We conduct a case study of a group of ED physicians who belong to a practice plan that receives lump-sum funding from a provincial government to provide acute care services. The physicians collectively determine pay for different shifts, including premiums for less desirable shifts. The physicians also receive 'shadow billing' for services performed. We compile and analyze detailed compensation data from 2007 to 2021 and will conduct a physician survey in early 2022. Results: While women are underrepresented in this practice plan, the gap in the proportion of women has improved over time: from 12/43 (28%) in 2007 to 35/89 (39%) in 2020. To our knowledge, no physicians in this group identify as transgender or non-binary. Assignment of shifts is equitable across genders, though physicians often trade shifts after assignment. Men do not work more of the higher-paid shifts than women, so trading does not lead to an 'hourly' gender wage gap. However, there is an observed gap in the average annual shift earnings ratio between men and women (1.14), as women work less hours on average than men. Larger gender gaps in the average annual total earnings ratio (1.18) are observed when including shadow billing. While these differences are quite large, variability in annual earnings within genders and small sample sizes mean that we cannot reject a hypothesis test of equal average earnings between genders. Conclusion: There is no average hourly wage gap between male and female physicians in this plan, however we do observe an average annual gender earnings gap. Part of the difference in earnings is due to differences in average hours worked across genders. Our findings differ from other research on the gender wage gap in medicine and may be partly explained by the physicians' collective decision-making processes in this practice plan. We will explore possible reasons for the hours differences in a follow-up survey, such as leaves, non-clinical professional responsibilities, and the relative distribution of childcare responsibilities between male and female physicians. Survey results will allow us to construct more precise estimates of the gender gap by controlling for labour supply. Keywords: equity, gender, pay Introduction: Some older persons presenting to the emergency department (ED) and living in the community are at risk of adverse outcomes such as loss of independence and return to ED. Cognition, frailty and functional status are 3 important and measurable aspects of health in older adults, but which is most predictive of adverse outcomes? OBJECTIVE To determine whether frailty, function or cognition is most predictive of adverse events defined as return to ED (RTED), calling emergency services, admission to long-term care (LTC) or death at 1 and 3 months. Methods: This is a secondary analysis of a prospective cohort study (September 2015 to April 2016). Participants [ = 65 years were enrolled at 3 different Canadian EDs. Inclusion criteria were presenting to the ED and living independently in the community. Exclusion criteria included patients unable to communicate, give consent or use a tablet. Frailty, function and cognition were assessed respectively using Clinical Frailty Scale (CFS), OARS and Montreal Cognitive Assessment (MOCA). Logistic regression models with 95% confidence intervals were explored to generate hypotheses. Results: We recruited 300 older adults, mean age was 75.7 years and 52% were male. A total of 72 patients (24.0%) had adverse events by 3 months. No domain was found to be predictive of adverse events at 1 month. Frailty was the most predictive domain at 3 months. Models that used MD and RA assessments of CFS (AUC [ 0.590) were better than models using patient selfassessment (AUC 0.499), and models using 2 assessments of CFS were the most feasible and predictive (AUC [ 0.625). In all models that included CFS assessed by an MD or RA, probability of adverse outcome increased as frailty increased (OR 1.5), but patients selfassessment showed robust reverse directionality (i.e. increased adverse event rates with lower self-assessments of frailty, OR 0.725 to 0.652). Conclusion: We found that no variable was predictive of 1 month outcomes. The CFS was the most predictive and feasible predictor of adverse events at 3 months. Any combination involving CFS done by physician or RA had good predictive value. The reverse directionality of predictive value between MD/RA rated CFS and patient self-reports should be further explored in future research. Introduction: Patient-centered care (PCC) is widely defined as a holistic approach to providing care that includes patient involvement, communication, access to services, well-trained staff, and an environment that meets patients' psychosocial, physical, and cultural needs. PCC is an emerging priority in many healthcare settings, yet it has not been incorporated into emergency department (ED) practice in a standard way. The goal of this study is to conduct a systematic review examining PCC in the ED to better understand how EDs undertake this method, and the objectives are to determine what components are included in PCC practices in the ED and what the impacts of PCC are on ED outcomes. Methods: Search terms were identified and search strategies developed by a medical librarian. The primary strategy (PubMed MEDLINE) was peer reviewed using PRESS and translated to search Embase (Elsevier), CINAHLPlus (EBSCO), PsycINFO (EBSCO), and Cochrane (Wiley). All articles identified by the database searches were examined for eligibility, assessed for quality, and underwent data extraction by two independent reviewers. The Joanna Briggs Institute convergence integrated approach was used to ''qualitize'' any quantitative data, followed by meta-ethnographic analysis to determine qualitative themes across all included articles. Results: 13 articles were included in final data extraction and analysis. The most cited components of PCC in the literature were comfort of environment (n = 8), communication (n = 7), education (n = 7), involvement of patient/family in information sharing and decision making (n = 7), respect and trust (n = 7), continuity and transition of care (n = 7), and emotional support (n = 5) . Four studies (n = 4) assessed the impacts of PCC components on various outcomes in the ED, and results demonstrated decreased length of stay (n = 3), reduced number of patients who left without being seen (n = 1), and greater patient satisfaction (n = 1) with implementation of PCC-related interventions. Conclusion: PCC can be a valuable contribution to emergency medicine practice. This study contributes to the literature on PCC in the ED. It can also be used to assist in the development of training modules for staff on implementation of PCC. By using the outlined components of PCC and implementing some of the suggested methods from the literature, it is possible to develop a comprehensive list of actionable PCC practices that create a better ED environment and desirable outcomes. Introduction: The treatment of primary headache disorders in the emergency department (ED) is variable and may be considered suboptimal with untimely pain relief and side effects. The role of peripheral nerve blocks (PNBs) in the management of primary headaches is currently unknown. This national postal survey aimed to examine Canadian ED physicians' practice patterns with respect to drug treatment and perspectives on the use of PNBs. Methods: We administered a national postal survey to a random sample of 500 ED physicians listed in the Scott's Canadian Medical Directory according to a modified Dillman's method. We piloted the survey before mailing the remaining questionnaires. The initial mailout included the questionnaire with an unconditional $5 coffee card and was followed by up to four reminders to non-responders. The final contact was delivered using Canada Post Xpresspost TM . Physicians were asked 12 questions regarding their demographics and practice setting, frequency of medication use and perspectives towards PNBs. Frequency of use was measured by collapsing responses into ''high'' and ''low''. We excluded physicians who were not treating adults in emergency medicine. Results: Of 500 mailed surveys, 468 were successfully delivered of which 179 physicians responded (response rate 38.2%). 144 were eligible to participate. Most physicians were male (63.9%); 80.6% had been in practice for C 10 years with 50.7% in a community or district general teaching hospital. Pharmacotherapies used in high frequency were dopamine antagonists (69%), dopamine antagonists with or without ketorolac (54.2%), ketorolac alone (53.1%), fluid boluses (54%) and acetaminophen (51.4%). We found opioid use to be low (0.7%). 79 (55.6%) physicians reported prior experience with PNBs and 55.1% believe they are effective and safe (85%). Occipital nerve blocks were the most commonly used PNB compared to the sphenopalatine ganglion (SPG) block and trigger point injections. 84.4% would consider PNBs as a first-line treatment option given sufficient evidence from a future trial. Conclusion: Current ED physicians comply with recommendations from the Canadian Headache society with IV ketorolac alone, as well as dopamine antagonists with or without ketorolac as commonly used pharmacotherapies; opioid use was very low. A large proportion of physicians have never used a PNB in their practice; however, most respondents would consider them as a first-line treatment option given sufficient evidence from a future trial. Keywords: headache, pain, peripheral nerve block Introduction: COVID-19 has been associated with decreases in Emergency Department (ED) utilization. EDs disproportionately serve those facing barriers to healthcare. Our objective was to assess changes to ED use by patient demographic during the first wave of COVID-19. Methods: We conducted a retrospective study using Alberta provincial administrative data. We estimated population rates and counts of ED visits and hospital admissions from ED for March 15, 2020, to June 30, 2020 and the same period in 2019. Poisson means tests assessed the significance of changes between periods. We conducted sub-analysis for children (0-19), seniors (age 65 ?), female patients, remote residents, and First Nations. We also examined outcomes for a pre-specified set of life-threatening diagnoses to better understand whether reductions in ED use may have resulted in missed needed care. We report outcomes for serious cardiac diagnoses here. Results: All subgroups of interest utilized the ED more than the overall population in both 2019 and 2020 on a per capita basis. Between 2019 and 2020 there was an estimated 34% (p \ 0.05) relative reduction (RR) in ED visits. were aged between 30-49. Female physicians made up only 22% and 33% of the workforce in rural and urban centres, respectively. Early career physicians (\ 10 years of practice) were more likely to practice in rural areas (63%) compared to their more experienced counterparts. Emergency physicians working in rural centres worked on average 28 days less in emergency medicine compared to urban practitioners. Conclusion: Most CFPC emergency physicians practice in urban areas. Those who practice in rural centres are more likely to be early career, male physicians. Rural physicians tend to work fewer days compared to their urban counterparts. Recruitment strategies should focus on keeping these early career physicians in rural communities and practicing EM. Future analysis will describe the volume and types of resuscitative and procedural codes billed to further explore practice variations between rural and urban centres. Keywords: emergency medicine, health human resource planning, rural medicine Introduction: Patients with limited English proficiency (LEP) face barriers to communication leading to inferior health outcomes when compared with English proficient patients. Professional translation services have been shown to improve healthcare outcomes for patients with LEP but are often underutilized. To report the incidence rate of LanguageLine use, a professional telephone translation service and to describe the patient demographic factors and physician perspectives associated with its' use. Methods: This was a retrospective chart review of patients who visited Kingston Health Sciences Centre's Emergency Department (ED) and Urgent Care Centre (UCC) between 2016-2021. Patients were classified as LEP if they had a non-English primary language listed on their electronic medical record. Patient records were cross-referenced with LanguageLine invoices to identify those who accessed translation services. The age, sex, date and time of arrival, primary language, official language, chief complaint, and Canadian Triage Acuity Scale (CTAS) score of patients with LEP who used the LanguageLine service while in the ED/UCC were compared to patients with LEP who did not use the service. T-tests and chi-square tests were used for the analysis of continuous and categorical variables, respectively. In addition, a survey distributed to ED physicians and residents collected perspectives on the facilitators/barriers to LanguageLine use for LEP patients.

Results: A total of 37 500 visits from LEP patients occurred, 118 (0.31%) of which used LanguageLine. LEP patients were more likely to receive translation services if they were younger (p \ 0.001), had a higher CTAS score (p \ 0.001), or spoke Arabic (p \ 0.001). All 16 staff/residents who responded to the survey (30% response rate) had at least one patient with LEP in the preceding month, with only 3/16 (19%) accessing LanguageLine for these patients. Further, 5/16 (31%) respondents reported never using the service with 4/5 (80%) unaware the service existed. Among those aware of LanguageLine, 7/12 (58%) reported the availability of a family member who could translate as a reason for not accessing the service. Conclusion: This is the first study in Canada to look at both the incidence rate of hospital translation service use and the patient factors that may be associated with its' use. Study results will help to inform interventions such as provider education on the use and availability of translation services in the ED for their LEP patients.

Keywords: health services accessibility, limited English proficiency. Introduction: Necessary COVID-19 physical distancing and selfisolation have the potential to increase emotional loneliness (EL) and social loneliness (SL), defined as the absence of an intimate relationship and the absence of a broader engaging social network, respectively. Older adults using the emergency department (ED) are particularly vulnerable. Previous research has demonstrated an association between loneliness and increased risk of mortality rivaling that of smoking. Few studies have examined the impact of loneliness associated with COVID-19 on older patients discharged from the ED. Objective: To assess the impact of COVID-19 on EL and SL in older adults discharged from the ED. This will inform the design of postdischarge interventions. Methods: This is an observational cohort study. The survey was conducted in English over the phone in patients C 70 years old who were discharged from the Mount Sinai ED in Toronto, ON. We excluded patients with low English proficiency, cognitive impairment, critical illness, or living at a nursing home. We assessed loneliness using the 6-item DeJong Gierveld Loneliness Scale. The scale includes 3 questions on EL and 3 questions on SL. Participants were asked to rate each of the 6 items of the scale using a 5-point Likert (4) Having a Plan. We report on a pilot use of PROM-ED in a single emergency department to measure the effect of improvement work on patient-reported outcomes. Methods: We included adult patients who received care over a two week period in one urban ED until target enrolment of 200 was reached. Included patients were directly discharged after ED care, and had one of a broad number of preselected reasons for ED visits which are in keeping with the scope for which PROM-ED was developed. Patients included were contacted systematically by phone and were asked to complete the four domains of the PROM-ED questionnaire. We linked PROM-ED data to demographic characteristics and the reason for the ED visit. We conducted a descriptive analysis and conducted univariate analyses for the effect of age, gender, age, reason for ED visit and time from ED care on the PROM-ED scores. Results: Phone contact was attempted with 517 patients, 270 calls were successful and 200 (38.6%) patients completed the questionnaire. Reasons for ED visit (most commonly abdominal pain, cardiovascular symptoms, extremity pain or injury and respiratory symptoms) and sex (56% women) were representative of the overall cohort during that period. The questionnaire was completed at a mean of 3.9 days (range 3-7). The group 50 to \ 60 years old reported significantly worse scores on understanding and reassurance. Patients seen for back pain reported significantly worse reassurance and symptom relief and patients with a laceration or puncture reported significantly better reassurance scores. Sex and time from ED care did not appear to have a significant effect. Conclusion: This constitutes the first use of PROM-ED for quality monitoring and description of patient scores according to their reason for seeking care. These measurements provide baseline scores to evaluate the effect of a planned initiative to improve communication with patients at discharge. Keywords: outcome measurement, patient-reported outcome, quality improvement is currently being evaluated in the Alberta context. We tested for differences in the PROM-ED scores and item functioning among people seen before and during the Covid-19 pandemic in Alberta's 16 busiest emergency departments. Methods: Since July 2019, the PROM-ED tool has been administered to patients before and during the pandemic (N = 367 and 320, respectively), 15-45 days after the ED visit. 116/92 (32%/ 29%) were 44 or younger, and 85/83 (23/26%) older than 65. The analysis focused on 16 items measuring the four PROM-ED outcome domains: symptom relief, understanding your health concern, reassurance, and having a plan. Confirmatory factor analyses were conducted to evaluate the hypothesized four-dimensional structure, internal consistency and reliability of the PROM-ED items; observed domain scores were subsequently described before and during COVID. Results: Confirmatory factor analysis provided evidence for the four PROM-ED domains. We found no significant differences in the raw domain scores and in reliability. However, differences at the item level were found in two of the domains, understanding the health concern and having a plan, which suggest the need to make item level corrections in changes of very significant changes in context. Conclusion: A widely disruptive event such as the pandemic can shift the way people understand and respond to items in outcome measures.

The results suggest a need to better understand how people may respond differently to questions about their outcomes before and during COVID, and to accommodate such differences psychometrically prior to interpreting the domain scores. Making comparisons of PROM-ED domain scores over time and across samples may require score adjustments. Interpretation is critical; there is a need for future qualitative work to examine how different people respond to PROM-ED items, as well as how knowledge users may use this information in their decision-making. Keywords: COVID-19, patient outcomes, patient questionnaire Introduction: The COVID-19 pandemic has created new challenges for critical care events in Emergency Department (ED) negative pressure rooms, and forced new adaptive solutions for limited equipment supplies and timely access. The creation and implementation of specialised ED equipment carts (eg. resuscitation, difficult airway, pediatrics) have previously been shown to improve patient care. The objective of this study was to complete a scoping literature review on the use of specialized trauma carts in the ED setting. Methods: A scoping review was completed using PRISMA-ScR methodology to assess for evidence of current use of specialised Trauma Carts in the ED. An electronic literature search was conducted using appropriate search terms (1995-current), with no language restrictions. Titles, abstracts and full-articles were screened independently by two reviewers and consensus was reached with the research team. Articles addressing the creation and implementation of specialized trauma carts (including burn injuries) in the ED setting were included. We extracted information addressing cart equipment composition, costs, implementation, staff evaluations and patient care outcomes. Results: After duplicates were removed, a total of 4341 titles, 244 abstracts and 31 full articles were reviewed. A total of 7 articles met review criteria. A heterogenous mix of outcomes were assessed for patient and provider benefit. Described carts varied widely in design and treatment goals. Key benefits include improved access to trauma and burn specific equipment, decreased time to interventions, potential cost savings, reduced equipment contamination, and potential improvement in patient outcomes. Conclusion:

The retrieved evidence supporting the use of specialized trauma carts was sparse, but reported a variety of benefits related to standardized equipment, ease of timely access, cost savings and improved patient outcomes. Further exploratory research on the utility of trauma carts is warranted. Introduction: Homelessness is a growing concern across Canada. Persons with no fixed address (NFA) have increased rates of emergency department (ED) utilization and increased healthcare spending compared to the general population. They have higher rates of acute and chronic illnesses, as well as all-cause mortality. To date, there has been no data collected on London, Ontario's homeless population, their health burden, and their utilization patterns of the ED. The primary objective of this study was to describe ED visits for adult patients with no fixed address in London, Ontario to assess for potential areas to improve care. Methods: This was a retrospective chart review. Patients 18 years or older with NFA visiting London EDs between Jan 1 and Dec 31, 2018 were included. ED visits were identified using either a diagnosis of ''homeless'', a lack of postal code, or by a postal code for a known local shelter. Demographics, health information, and utilization patterns were examined. Data were summarized as means and proportions. Results: A total of 4,186 NFA visits were identified, equating to 1,182 patients. The mean number of visits per person was 3.5, the average age was 39, and 73% were male. As for health care utilization, 46% arrived by EMS, 40% had a family doctor and 92% were registered with OHIP. The top 3 discharge diagnoses were: medical (42%), drug use (29.5%), and mental health (21%). We did not identify an obvious seasonal trend for rates of presentations, and time of day for ED registration was relatively similar across 24 h. Conclusion: This study indicates this population already has reasonable access to primary care/OHIP, suggesting that joint ED-primary care initiatives may help. A large percentage of visits were drug use related, thus an ED addictions service may also be of benefit. This study hoped to look at referrals for ED social work (SW) and outpatient addictions services, however electronic charting of these referrals was rarely found. Despite this, the data suggests that patients with NFA would likely benefit from 24-h access to SW given the consistent rate of ED visits outside of daytime hours. Because this population relies heavily on EMS, successful improvement of services may then decrease their utilization of EMS, resulting in system cost savings. This study provides a base for potential interventions and prospective studies that may help London's homeless population and in turn positively impact health care system utilization. Keywords: homelessness, quality improvement, resource utilization Introduction: Alcohol use disorder (AUD) is a significant and growing problem that affects emergency departments (EDs) across Canada. Medical management of AUD-including naltrexone, acamprosate, and gabapentin -has been shown to reduce alcohol consumption and promote abstinence. Despite the fact that treatments are available, they are infrequently utilized: less than 10% of patients with AUD ever receive anti-craving medications and fewer still have been shown to receive this treatment in the ED. One potential cause is the lack of anti-craving or addiction medicine training that emergency physicians receive as part of their medical training, instead having to rely on supplemental courses. The objective of this study was to understand the perspectives of emergency department care providers, focusing on current management strategies for AUD and perceived barriers to prescribing anti-craving medications. Methods: This was a cross-sectional survey of all ED physicians, nurse practitioners, and physician assistants at an urban, academic, tertiary care centre in August 2021. Participants were asked questions regarding demographics, previous training in the medical management of AUD, current practice strategies to manage AUD, perceived barriers to anticraving administration, and beneficial education strategies. Results: A total of 23 ED care providers responded to the survey, 19 (82.6%) of which were physicians. Three (13.0%) reported being in their first five years of practice, while eight (34.8%) reported 21 or more years of experience. While 19 (82.6%) participants reported previous training in the medical management of AUD, only five (21.7%) said they currently prescribe anti-craving medications in the ED. Overall, participants reported frequently discussing addiction medicine referrals, and have advised patients on the need for medical management but most (15, 65.2%) report a lack of confidence in prescribing anticraving medications themselves. Conclusion: Despite the availability of effective anti-craving treatments for the medical management of AUD, the majority of ED care providers surveyed report limited use and confidence in utilizing these strategies. More detailed prescribing instructions, and ED-specific education have been identified as methods to improve this competency amongst care providers, and further research is needed to explore the efficacy of intervention strategies.

Keywords: alcohol use disorder, anti-craving, education Background: The current first line treatment for opioid use disorder (OUD) is buprenorphine and naloxone. Despite availability of tools to assist with OUD management, such as the Clinical Opioid Withdrawal Scale (COWS), there remains low comfort levels with the utilization of buprenorphine and naloxone within emergency departments (ED).We aim to increase the frequency of buprenorphine and naloxone prescribing for OUD patients in our ED by 10% over 6 months. To achieve this, we aim to increase ED physicians' comfort in prescribing buprenorphine and naloxone, along with nurses' comfort in assessing OUD and monitoring for precipitated withdrawal. Aim statement: Our outcome measures were rate of buprenorphine and naloxone administration, physician comfort in prescribing, and nursing comfort with COWS score and precipitated withdrawal monitoring. Our process measures were length of stay (LOS) and total dose of buprenorphine and naloxone. Finally, our balancing measures were hospital admission and repeat ED visits within 72 h. We used an iterative design with several improvement initiatives. From physician's perspective, an 8-item needs assessment survey was distributed amongst all ED physicians at a large urban community hospital to assess comfort and barriers to prescription. , 1980 to June 1, 2021. Studies were scrutinized for quality and validity and assigned a level of evidence as per Oxford Center for Evidence-Based Medicine guidelines. Results: Amongst 422 initially screened articles 5 were eligible for inclusion; within these studies a total of 257 FICBs were conducted. The mean NVPS score post intervention (out of 10) for the FICB group was 1.53 compared with 3.6 for traditional analgesic methods. Success rates of the FICBs were 89.3% and included only 2 serious adverse events compared 7 within the standard of care arm. However, ratings of patient satisfaction post intervention were equivocal between groups with both being assigned median scores of 9 out of 10. Conclusion: FICBs are suitable for use in prehospital settings as a regional analgesic technique for proximal femur fractures. It carries a low risk of adverse events and may be performed by healthcare practitioners of various backgrounds with suitable training. Given statistically significant findings in pain reduction FICBs appear to be more successful than parenteral or oral opiates and sedative agents alone and can be used as an appropriate adjunct to pain management. Keywords: prehospital, regional anesthesia, trauma Introduction: Language barriers (LB) between patient and healthcare provider decrease quality of care in an emergency department (ED). Due to perceived time and effort associated with obtaining a professional interpreter, this service is underutilized in the ED. If the need for an interpreter is known early in the patient's visit to the ED during triage, there may be more time to access available resources. The process with which registered triage nurses (RNs) manage patients with LB has not yet been studied. The purpose of this study was to assess the current standard of care for RNs in identifying and documenting LBs of patients visiting the ED, and to illustrate their comfort doing so. Methods: RNs employed in the ED of two sites of a major academic center in a metropolitan city were asked to complete a 21-item anonymous, bilingual, online questionnaire via LimeSurvey, collected over 3 weeks. With a confidence level set at 90%, a margin of error at 5%, and a consideration for a 50% response rate, the sample size was set at 46 (total triage RNs employed at both sites was 92 Introduction: Intra-articular and hematoma blocks are considered as less resource-intensive alternatives to procedural sedation for painful procedures such as reducing fractures and dislocations. This study measured the impact of intra-articular and hematoma blocks on the emergency department (ED) length-of-stay (LOS) and the delay before readiness for safe patient discharge compared to procedural sedation for common orthopedic injuries. The need for a second reduction attempt was also evaluated. Methods: Using a case-control methodology, a retrospective study of patients who presented to a level one trauma centre between January 1st, 2018 and December 31st, 2020 was performed. Inclusion criteria were adults ([ 17 years old) presenting with a diagnosis of an acutely displaced wrist or ankle fracture or a dislocated shoulder who underwent at least one reduction attempt in the ED. ED LOS was defined as the delay between the initial registration to the patient's departure and readiness to be discharged was defined as the delay between first emergency physician contact and the reassessment during which the patient was fully-alert with adequate analgesia. The delays are presented in minutes using medians (IQR) and were compared using Wilcoxon Mann-Whitney Tests For qualitative content analysis, comments were coded by two independent reviewers in Dedoose and analyzed using inductive category development until thematic saturation was reached. Sub-codes were subsumed into major coding categories to identify important themes. Results: Three major themes and an overarching theme of satisfaction emerged from this analysis. Our findings uncovered that paediatric VC use was motivated in part by a desire to avoid the hospital environment. Inperson ED visits were reported to be challenging and stressful, particularly due to perceived infection risk. Respondents also appreciated that the paediatric VC emergency clinic provided reassurance by assisting in navigating the healthcare system and felt the virtual option allowed them to use healthcare resources responsibly. A third theme that emerged is the convenience and ease of access to virtual care allowed for improved family-centred care in vulnerable populations. The overarching theme of satisfaction was emphasized by the numerous comments for this service to be sustained and offered postpandemic. Conclusion: This study describes the user experience and perceived impacts of virtual emergency care in the paediatric population. Our study indicates that virtual care was an attractive option for parents to keep their family safe and provided guidance during an uncertain time. The value to vulnerable populations and a strong patient desire for continued availability post-pandemic will be important considerations for future research of this rapidly developing area of care Introduction: Emergency departments in rural communities are essential for providing care to patients with both acute conditions and those who are unable to access primary care. Over the last several years, the number of new family physicians providing emergency medicine (EM) services has declined. Given, that it is becoming more difficult to find family physicians to provide EM services, it is important to understand the physicians who do provide this service in rural Ontario. Our aim was to describe the demographics and practices of the rural family physicians and general practitioners providing EM services across Ontario. Methods: Data from 2017 was collected from the ICES Physician database (IPDB) and the Ontario Health Insurance Plan (OHIP) billing database. Information from the IPDB included encoded physician demographics, practice region and certification information. Sentinel billing codes (i.e., a billing code unique to a particular clinical service) was used to define the services that each physician was providing. A total of 18 unique services were identified. Demographic and practice patterns of rural physicians practicing emergency medicine were then described based on their geographic location in either rural northern or rural southern Ontario. Results: From the ICES database, a total of 1192 rural family physicians or general practitioners were identified out of a total of 14,443 physicians with similar certifications. From the rural physician population, a total of 620 physicians practised emergency medicine which accounted for 33% of their practice on average. The majority of the physicians practicing emergency medicine were between the ages of 30-49 and in their first decade of practice. The most common services in addition to emergency medicine were clinic, hospital medicine, palliative care, and mental health. There was little difference in the demographics or practices of rural physicians in southern and northern Ontario. Conclusion: Physician demographic and service provision information is useful for health human resource planning and physician allocation. Our study provides baseline information about the demographics of rural family physicians and general practitioners practicing emergency medicine and their other clinical responsibilities. With the current changes in family physician practices and the decline in the provision of emergency medicine in rural areas, it is important to understand this physician population. Keywords: administrative data, health human resource planning, rural physicians

The quality assurance process for positive urine culture follow-up in a tertiary care emergency department: A pilot study Introduction: Urinary tract infections are commonly seen and treated in the emergency department (ED). As part of management, urine cultures are often ordered with results pending at the time of discharge. Institutional quality assurance (QA) processes, that employ dedicated personnel to ensure appropriate follow-up, often have equivocal outcomes. We sought to understand the times taken to follow up clinically relevant post-discharge urine culture results with a view towards identifying opportunities for quality improvement (QI). Methods: We conducted a health records review at an academic tertiary care ED. All patients 18 years and older who were seen and discharged between July 1, 2020 and June 30, 2021 and had a positive urine culture were eligible for inclusion. One hundred patient charts were selected using a random number generator and data was abstracted manually using a structured data collection form. We performed descriptive analyses on the times to physician and nursing actions. The project received local research ethics board approval prior to initiation. Results: Of the 100 patients (mean age 66.4 years (SD 24.6), 59 female) included in the study, 65 (mean age 71 years, SD 21.5) required follow-up for their culture results. 85% (n = 55) of cultures requiring follow-up grew a single organism. The mean number of days from ED discharge to follow-up completion was 3.4 days (SD 2.1). The mean time from ED discharge to receiving a culture result was 1.3 days (SD 1.5) and the mean time for transfer of results for physician review was 1.4 days (SD 1.2). The mean time for the physician to provide further follow-up instructions after receiving the alert was 2.6 h (SD 6.6). The mean time between a nurse receiving instructions and following up with the patient was 15.6 h (SD 23.6). 98.4% (n = 64) of patient follow-ups were documented. 72% (n = 46) of these patients were successfully contacted on the first attempt at a callback. 23% (n = 15) required a second attempt and 5% (n = 3) required three attempts. Conclusion: Nearly two thirds of the urine culture results in our study required follow up. In general, more than three days was required to complete the QA process with the longest delay attributed to transferring culture results to physicians for review. Our results suggest that future QI work should explore ways of facilitating more efficient transfer of results to the physician, like automation with machine learning and natural language processing. Keywords: follow-up, quality improvement, urine culture . Results: 319 individuals experienced highly frequent use of the ED during one or more of the eight years of study, accounting for 78,265 ED visits (range: 56-2,780 visits). These patients had 6,077 hospitalizations (range 0-200) of which 38% were in an inpatient mental health unit. 86% of the sample had four or more chronic conditions of which the following were most common: anxiety disorders (83%), substance-related disorders (81%), depression (69%), brain injury (61%), personality disorders (57%), and schizophrenia (55%). 179 individuals (56%) had one year of highly frequent ED use; 69 had two years; 24 had 3 three years; 24 had four years; and 23 had five or more years. On average, the patients were 45 years old (SD: 16 years), ranging from 18-89 years old. There were 178 males (56%) and 141 females (44%). Based on the administrative data, 199 (62%) individuals had a history of homelessness or of being precariously housed. 55% of the respondents were recipients of home care services and 9% were on coordinated care plans. Conclusion: Although they represent a minority of the ED patient population, individuals with highly frequent use of the ED have complex health challenges and often use a significant amount of health care resources. Understanding more about the unique characteristics of this group of patients in Southern Ontario is essential in steering future interventions aimed at helping this population in the community. Keywords: emergency department, highly frequent user P20 Spatial analysis of acute road traffic injuries in Kenya to address regional and economic disparities through targeted injury prevention initiatives Introduction: Injury risk, is affected by the social determinants of health including social and economic inequities, which are more pronounced in low and middle-income countries (LMIC) like Kenya. These inequalities lead to a disproportionate burden of injury in LMIC, including road traffic injury (RTI). This increased risk leads to an enormous health burden for emergency departments (ED) and exacerbates an already stressed healthcare system. The study objective was to identify the geographic hotspots of RTI in Kisumu City, Kenya, to be used in the development and implementation of injury prevention initiatives to reduce the RTI incidence and the ED injury health burden. Methods: All RTI arriving to the ED of Jaramogi Oginga Odinga Teaching and Referral Hospital in Kisumu City over a 4-week pilot period of injury surveillance data collection (2019-07-15 to 2019-08-11) were mapped within a GIS heat mapping program by collision types, vulnerable road user (VRU) group, age and gender, to determine regional RTI hotspots. A road safety audit tool was adapted for LMIC that incorporated road design, infrastructure, safety elements for pedestrians and motorcyclists, and other injury factors to create a final safety score for each hotspot. These scores were compared to satellite imaging to identify injury prevention targets and create recommendations for traffic safety. Results: RTI was the common injury type occurring in nearly half (171/368; 47%) of trauma in the ED. Spatial analysis revealed clusters of crashes in high population density areas, primarily the market and slum neighbourhoods, and also along the main highway into Kisumu City. Nine hotspots were identified for three VRU groups -pedestrians, motorcyclists and other vehicle occupants. The highest RTI crash site was the Kondele round-a-bout, the main entry intersection into the urban area of Kisumu City. This intersection scored a 18/62 for road safety. Conclusion: Crash hotspots were identified by spatial analysis and GIS/heat mapping. Safety recommendations for each hotspot were created. A socio-ecological health model will be used to address underlying factors for injury risk and design injury prevention programs and policies that address regional and economic disparities. The proposed countermeasures need to be accessible to all citizens in the highest risk regions to successfully, and equitably, mitigate crash risk in Kisumu and decrease injury burden in the ED.

Keywords: global health, injury prevention, road traffic injury Introduction: This study was conducted to evaluate the first pass success rate of two supraglottic airway devices (SGA) used by paramedics in New Brunswick, Canada: the King LTS-D and the igel. Significant heterogenicity exists internationally in the airway devices used in the pre-hospital setting during cardiac arrest. While there is some limited data available comparing the two devices, there is minimal research about ease of use. In a setting where seconds count, it is important to identify the device that offers the most straight forward and reliable application. As such, this research will investigate the first pass success rate and mean number of insertion attempts of the airway devices used by paramedics in New Brunswick. Methods: We used over 2,400 patient care records compiled by Ambulance New Brunswick between February 1st 2015 and September 1st 2020. We investigated the first pass success rate of the two airway devices using a 2 9 2 Pearson chi-square test for association. Successful insertion was determined by end tidal capnography & oxygen saturation. Secondly, we determined whether there were differences in mean number of attempts required for successful insertion. Results: Our study suggest that there is a statistically significant association between airway devices and first pass success-favouring the i-gel in successful insertion X2 (1) Introduction: Transfer of nursing home residents to hospital for diagnostic imaging is common, costly, often requires an ambulance, and can expose residents to unnecessary harms. Considering these concerns and emergency department (ED) overcrowding, some jurisdictions have implemented mobile radiography services to reduce the burden on EDs and ambulance services. Data on where and how nursing home residents receive imaging in Saint John, New Brunswick (NB) is required to determine if mobile radiography might benefit residents and systems in similar communities. This data was unavailable prior to this study and will be essential for future studies to determine whether mobile radiography might improve care in communities with similar demographics while also being cost-effective. Our objective was to determine how and where nursing home residents in an Atlantic Canadian city receive their X-ray (XR) and computed tomography (CT) services. Methods: This observational study reviewed retrospective data on all XR and CT investigations conducted on all nursing home residents in Saint John, NB (population 126,00) in the 2020 calendar year. The primary outcome was the investigation frequencies and the setting of access. Data on presenting complaint, specific diagnostic imaging order, discharge diagnosis, specialties consulted, admission/discharge time, and date were also collected. All values were reported descriptively with a 95% confidence interval. Results: There were 521 visits by 311 unique nursing home residents. There was a total of 920 imaging investigations (688 XRs and 232 CTs). Most investigations were ordered in the ED (696 of 920; 75.6%; CI 72.8-78.3%) with the remaining (224 of 920; 24.4%; 21.7 to 27.2%) provided through outpatient services. The three most common investigations were chest XR (307), hip/pelvis XR (142), and head CT (114) and orthopedic surgery was the most common consult (24). About half (52.1%) of the visits to the ED that required imaging resulted in admission. Of the nursing home residents who received imaging in the ED, 23.0% received only XR imaging (no CT scan) and were discharged back to the nursing home within a mean ED stay of 5.15 h. Conclusion: The ED is the primary hospital access point through which nursing home residents receive XR and CT scans. About a quarter (23.0%) of visits to the ED by nursing home residents fit the profile of visits that could potentially be avoided with a mobile radiography service. Keywords: emergency department, nursing home residents, radiography Introduction: Rapid access to a major trauma centre improves the outcome of severely injured older adults. The objective of this systematic review was to determine the diagnostic accuracy of trauma triage tools to identify severely injured older adults and those at risk of adverse outcomes. Methods: Three databases (MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials) were searched along with preprint servers and google scholar. Cohort studies were considered eligible if they reported a measure of diagnostic accuracy (sensitivity, specificity, and/or the proportions of overtriage and undertriage) of prehospital triage protocols to predict the severity of the injuries, in-hospital critical resource use or adverse outcomes. Risks of bias were assessed using the ROBINS-I tool. The study protocol was registered in the Prospero database (CRD42019134354). Results: Twenty-eight studies were included of which 24 were from the USA. Each study included between 463 and 2 783 377 older adults. Definition of an older adult was heterogeneous ranging from C 50 to C 70 years old. Twenty studies reported on the diagnostic accuracy to predict the severity of the injuries. Sensitivity of current tools ranged between 39.9 and 93.0% (n = 11). Modifying physiological criteria (systolic blood pressure or Glasgow Coma Scale) or incorporating new elements such as comorbidities or anticoagulation modestly improved the sensitivity of current protocols (n = 8) . Theoretical sensitivity of modified tools or newly derived protocols ranged between 40.0 and 92.1% but improving the sensitivity was associated with decreasing specificity (n = 8) . Sensitivity of current triage tools (n = 6) and prehospital care professionals' compliance to the destination recommended by the triage protocols (n = 5) decrease for older patients ([ 80-85 years old) . Five studies reported on the accuracy to predict mortality with a sensitivity ranging between 15.7 and 90.0%. Risk of bias was considered low (n = 3), moderate (n = 13) or severe (n = 12) . Conclusion: Current prehospital trauma triage protocols are insufficiently sensitive to identify severely injured older adults and, even when identified, older adults were less likely to be transported to a major trauma centre. Methodologically robust research on the development of geriatric-adapted protocols is required. Keywords: older adults, trauma, triage This study aims to identify barriers and facilitators among physicians that prevent them from using the guideline-based algorithm for the diagnosis and treatment of BPPV. Methods: We conducted semi-structured interviews with eleven emergency room physicians working at three sites. We used purposive sampling, contacting physicians until we reached data saturation. Interview questions were designed to understand potential barriers and facilitators to guideline recommendations uptake and use. Responses were analyzed according to the Theoretical Domains Framework, and overarching barriers and facilitators themes were identified. Results: We identified four themes encompassing 13 theoretical domains. These included the awareness and the use of the BPPV guideline-based algorithm within their current practice; the understanding of benefits and motivations to use the pathway; patients' comfort and the emergency department limitations; the ability to get the information and its ease of use. Even though the majority of interviewees were not currently following the BPPV algorithm to diagnose and treat patients, they did agree that the recommended guidelines would improve patients' outcomes. Other facilitators included having video instructions on hand to help physicians perform maneuvers that are rarely required to treat patients with BPPV. Barriers to the implementation of the pathway included patients' comfort when they are too symptomatic and constraints within the emergency department itself. Conclusion: Clinicians supported the recommendations of the clinical practice guidelines for BPPV. We identified important considerations for the development of a structured intervention to address the barriers and leverage the facilitators in order to increase uptake of these recommendations. Introduction: Up to one third of neurological emergencies are misdiagnosed by the initial emergency medicine (EM) physician (1). Given the risk of worse outcomes secondary to delays in treatment and misdiagnosis, it is important to assess core competencies, evaluate teaching and assessment practices, and identify barriers surrounding neurological emergency training in Canadian EM Residency Programs. Methods: We conducted a self-administered crosssectional survey of EM program directors (PDs) in North America. We conducted a literature search using MEDLINE to look at training in neurological emergencies in EM Residency Programs. We then developed a questionnaire through a development stage (item generation and reduction, questionnaire formatting and pretesting) and a testing stage (pilot testing, clinical sensibility, reliability, and validity). We semi-structured conducted cognitive interviews with EM faculty (purposive sampling) to ensure clarity of the questionnaire, and a clinical sensibility survey to rate the survey's face validity, redundancy, and comprehensiveness (content validity). We distributed the survey via email to Canadian EM PDs in the Royal College and Family Medicine Enhanced Skills in EM Programs.

Results: To date, 39% (n = 12) of Canadian EM PDs completed the survey. 100% of respondents consider the management of neurological emergencies a necessary part of EM training. 67% of programs spend less than 10% of lecture time dedicated to neurological emergency topics. 42% of programs spend less than 10% of total simulation time, and 58% of programs spend 11-20% of total simulation time on neurological emergencies. Topics are not standardized across programs, regardless of training stream. Commonly covered topics include airway management, ventilation and sedation in neurologically injured patients and traumatic brain injury. Topics not commonly covered are acute spinal cord injury and meningitis. 58% of PDs believe residents would benefit from a structured course dedicated to the management of neurological emergencies to attain competencies in this domain, but no programs offer such courses. Conclusion: Management of neurological emergencies is an essential skill for EM physicians, however, there is no standardization of priority topics, methods of delivery or evaluation for training residents in these domains. Our next steps include the distribution of a similar survey to residents, and extension of both surveys to the United States to compare training in North America. Keywords: curriculum, education, neurological emergencies Introduction: Acute uncomplicated pyelonephritis is defined as pyelonephritis limited to non-pregnant, pre-menopausal women with no known relevant urological abnormalities or comorbidities. There exist several guidelines from specialty organizations on best practices with respect to investigating and managing uncomplicated pyelonephritis. While there is some equipoise, they do largely agree on a few key recommendations, including routine urinalysis and urine cultures for all cases and standard empiric antibiotic therapy, either with a quinolone or trimethoprim-sulfamethoxazole. The goal of this initiative is to ensure that patients at our large academic emergency department are receiving evidence-based care with regards to management of uncomplicated pyelonephritis. Methods: We conducted a retrospective chart review, including all female patients aged 50 or younger with an ED diagnosis of pyelonephritis (Jan 2019-Jan 2021). A total of 37 charts met the inclusion criteria. We recorded demographic data, as well as data on ED investigations, management, and outcomes. Data was analyzed for descriptive statistics using Microsoft Excel. Results: The average age for our included patients was 31.6 years old. With respect to ED investigations, urinalysis was obtained on 97% of patients, while urine culture only 69%. Point-ofcare ultrasonography was performed on 27% of patients, with another 27% receiving a formal ultrasound and 8% receiving CT imaging. 35% of patients received a dose of IV antibiotics (77% Ceftriaxone) and 57% of patients received a dose of PO antibiotics in the ED. 40% of patients were discharged on oral ciprofloxacin, 37% on oral TMP-SX. 60% of urine cultures were positive for E. coli. Of these, 20% were resistant to ciprofloxacin, 20% were resistant to trimethoprimsulfamethoxazole (TMP-SX), and 7% resistant to both. Introduction: Simulation provides exposure to high-acuity low-occurrence patient scenarios without sacrificing patient care. This needs assessment survey aimed to gather opinions from relevant stakeholders regarding the development of an Emergency Medicine (EM) simulation curriculum for local senior medical students. The primary objective of the study was to determine which concepts were most valuable to include in the curriculum. The secondary objective was to determine any barriers to simulation education. Methods: Participants were current medical students interested in EM, recent medical school graduates (2017 or later) currently enrolled in an EM residency program, and BC-based EM staff, undergraduate EM educators, and medical simulation educators. A 15-question survey was emailed to participants gathering information about the participant's role and current involvement in simulation, ideal objectives and frequency of simulation, and perceived barriers to simulation education. Free text options allowed participants to provide additional scenarios and general feedback. Results: There were 52 survey responses (17 medical students, 20 staff, 15 residents) with a 46.8% response rate.

Most participants felt that simulation should be offered monthly to medical students. In regards to learning objectives, most medical students prioritized knowledge acquisition, whereas staff and residents felt that focus should be placed on crisis resource management (CRM) skills. Participants of all categories shared concern for learner embarrassment as a primary barrier to simulation education. Staff concerns also included limited resource allocation and availability for simulation. Medical students identified ACLS, trauma, and airway management as most important. Residents and staff identified ACLS, shock, and trauma. Conclusion: Overarching simulation program objectives should include CRM, but also focus on building general EM knowledge including ACLS, trauma and shock scenarios. It is critical for simulation programs to ensure a safe learning environment. There are existing resource barriers, but current students, recent graduates and staff agree that EM simulation for senior medical students should be offered more often than currently available. Keywords: curriculum, simulation, undergraduate medical education -19) zones are determined at emergency department (ED) triage and may be influenced by clinical parameters such as presenting symptoms, recent travel, or vital signs; however, these may not be reflective of the final diagnosis. There is also evidence that assessment in dedicated zones may lead to early diagnostic closure. Delays in diagnosis could lead to further patient harm but there is limited knowledge on rates or characteristics of non-respiratory illness (NRI) diagnosed in dedicated COVID-19 zones.

Methods: This was a retrospective chart review of electronic medical chart data from a single urban community ED in Toronto, Ontario, using the RECORD framework. Charts for all patients triaged to the COVID-19 assessment zone between March 12 to April 30, 2020, were reviewed. Exclusion criteria were: diagnosis of upper respiratory infection, pneumonia, COVID-19 infection, flu or flu-like illness, or assessment in any other area of the ED. Using a standardized data abstraction form, we extracted the following variables for patients with non-respiratory illness diagnosis: age, gender, time of day for visit, chief complaint, CTAS score, presence of abnormal vital signs, sent for evaluation by an outside source, length of stay, diagnostic testing (imaging, COVID PCR test, other), and 7 and 30-day return visits. Results: A total of 4475 charts were screened. 89 patients were identified as NRI. The most common NRI diagnosis was chest pain not yet diagnosed (n = 28), followed by gastroenteritis (n = 11), and anxiety (n = 8) . Serious diagnoses such as acute coronary syndrome (n = 3), congestive heart failure (n = 2), or venous thromboembolism (n = 1) were rare. There were no acute surgical issues identified. 47.2% of these patients had respiratory symptoms, 30% of patients had gastrointestinal or cardiac symptoms, and 10.1% had a history of recent travel. Greater than 90% had a systolic blood pressure [ 120 mmHg or heart rate [ 100 BPM at triage. 10.2% of patients had return visits within 7 days, but none of these were in patients with serious diagnoses. Conclusion: Diagnosis of NRI in COVID-19 zones of EDs is rare. The most common diagnoses are non-specific such as chest pain NYD or gastroenteritis. ED staff can consider the presence of non-respiratory symptoms or abnormal triage heart rate or blood pressure as possible risk factors for NRI, although this would need to be further validated.

Keywords: coronavirus, triage

Introduction: Prolonged length of stay (LOS) in the intensive care unit (ICU) impacts patient outcomes, ICU costs, and ICU bed availability. Understanding the factors that influence ICU LOS may allow for prediction of patients at risk of prolonged LOS, and for targeting of early interventions to reduce LOS in those patients at risk of prolonged LOS. In Nova Scotia (NS), ICU LOS has not yet been described and the variables impacting the LOS in Nova Scotian ICUs have not yet been identified. 1. Describe ICU LOS of patients admitted to NS ICUs. 2. Create a predictive model to describe variables contributing to increased LOS. 3. Assess differences in mortality based on ICU LOS. Methods: We conducted a retrospective review of data collected in the NS provincial ICU database. We included all patients admitted to one of twelve adult ICUs in NS between April 1 and March 31 2021. All data were analyzed using the statistical programming language R. Descriptive statistics on ICU LOS were generated. A linear mixed effects model was fit to examine differences in mortality based on ICU LOS. A generalized linear mixed effects model was used to predict ICU LOS from the following predictor variables identified a priori: APACHE IV predicted mortality, age, diagnosis, need for ventilation in the first 24 h, ICU admission location, location prior to ICU admission, occurrence of delirium during ICU stay, and mobility level. Results: There were 12,078 ICU visits in the period under study. The average ICU LOS was 4.99 days (SD 8.79 days). Significant variation in ICU LOS was seen, with higher LOS in older patients, patients admitted with respiratory diagnoses, patients requiring mechanical ventilation, and patients with higher APACHE IV predicted mortality. Correlation was found between ICU LOS and the following categories: admission diagnosis category, admitting center, location prior to ICU, APACHE predicted mortality, and delirium. Factors correlating with extended LOS (7 ? days) included ventilation, center, APACHE-predicted mortality, and delirium. The odds of mortality was found to decrease significantly with increasing LOS. Conclusion: We have identified several variables that correlate with increased ICU LOS in NS. Further study is required to determine if interventions targeted at patients with these variables would impact ICU LOS. Any interventions targeting ICU LOS would also have to closely study impacts on mortality as it was found that longer ICU LOS was correlated with lower odds of mortality. Keywords: length of stay Introduction: Virtual patient care has seen incredible growth since the beginning of the COVID-19 pandemic. In the fall of 2020, the Ministry of Health introduced a pilot program of 14 virtual urgent care (VUC) initiatives across the province of Ontario to enable healthcare providers and patients to safely connect from a distance. The objective of this study was to determine why patients chose VUC and to describe their experience using VUC services. Methods: Patients who completed a VUC visit from December 2020 to September 2021 were invited by email to complete a standardized, 25-item online survey, which included a series of questions related to satisfaction and patient reported outcome measures (PROM). Results: Of the 2,177 anonymous responses, 73% were female, 87% had higher than post-secondary education, 93% were English speaking, and 90% had a primary care provider (PCP). Of those who had a PCP, 47% tried to call their PCP and couldn't get an appointment or the wait was too long. 69% would have gone to the ED if VUC was not available. 94% rated their overall experience with VUC as 8/10 or greater. 90% of patients were very satisfied with the ease of registration and scheduling, 88% were satisfied with the software and logging into the clinic, and 93% were satisfied with the wait time. 96% of patients felt comfortable connecting with the health care provider virtually, 95% felt their privacy was respected, 94% said the VUC provider spent sufficient time with them, and 91% reported the VUC visit was thorough. 50% of patients would have preferred to meet the health care provider in person, 35% preferred VUC and 15% were unsure. For the PROM of patient understanding, [ 80% said they had answers to all the questions they had related to their health concern and had as much information as they currently needed. For the PROM of reassurance [ 80% felt at ease and reassured about their health concern after their VUC visit. Finally, for the PROM of patient's having a plan, 80% believed they were able to manage the issue, had a plan they could follow, and knew what to do if the issue got worse or came back. Conclusion: The vast majority of patients using VUC services had a PCP, but access to their PCP was not available in a timely fashion. Patients using VUC were extremely satisfied, however, considering the demographics of those using VUC, future work should focus on health equity and making VUC more accessible to disadvantaged and vulnerable groups. Innovation Concept: Return visits and adverse events have historically been viewed through an assumption of individual failure and responsibility rather than understanding and correcting the system of care. There is increasing emphasis on analyzing these incidents, which are often presented at Morbidity and Mortality (M&M) rounds, through the lens of quality improvement (QI). Despite common quality issues across emergency departments (EDs), the lack of a safe and organized platform to share adverse events between institutions leads to reluctance in providers to share opportunities for learning and improvement. To address this gap, we created the Emergency Medicine (EM) QI Digest, a monthly newsletter shared with EDs across the Greater Toronto Area (GTA). Methods: Stakeholder engagement with ED QI leads across the GTA identified three goals: 1) share ED adverse events and clinical cases through the lens of QI, 2) learn from QI initiatives from other EDs, and 3) create a community of practice to safely disseminate this knowledge in a more targeted approach than traditional social media. Each organization contributes content for one issue per year: two de-identified clinical cases analyzed through a QI lens employing a Kirkpatrick model to guide readers from reaction to learning and change, one or two QI projects, and the promotion of citywide M&M education rounds. Curriculum, tool, or material: The EM QI Digest launched in September 2017 and is entering its fifth year. Through 26 issues, we shared 52 clinical cases and 41 QI projects on a wide variety of topics, including reducing unnecessary tests, promoting medication safety, avoiding diagnostic error, and enhancing ED flow. Site leads across eight academic and community EDs distributed the Digest to 700 providers (including doctors, nurses, nurse practitioners and physician assistants), as well as to residents. Multiple organizations reported local impacts, from adapting QI initiatives to learning from cases. Conclusion: The EM QI Digest has successfully broken down traditional institutional siloes and provided a safe and supported platform to foster sharing of knowledge and collaboration to improve patient care. This confirms the effectiveness of this knowledge translation method, leading to sustained enthusiasm and participation across a region despite the lack of dedicated financial and administrative support. This grassroots initiative has been sustained for five years, and can be replicated in different settings. Keywords: innovations in emergency medicine, patient safety, quality improvement 

We conducted a mixed methods study to better understand First Nations members' experiences with emergency medical services (EMS), and paramedic experiences when delivering care on First Nations communities. The research was undertaken in partnership with the Alberta First Nations Information Governance Centre, Alberta Health Services, the University of Alberta, and the University of Calgary. Methods: Provincial administrative data on EMS responses to First Nations municipalities from January 2018 to November 2020 were analyzed descriptively. Virtual sharing circles were hosted in July 2021 and included Elders, First Nations patients and caregivers, paramedics, other health professionals, EMS medical directors and provincial EMS leaders. Thematic analysis of the qualitative data was performed in partnership with paramedic and Indigenous health researchers. Results: Persons living within First Nations municipalities used EMS 3.6 times more than the overall Alberta population in 2020 (48.7 events per 100 persons vs. 13.4). First Nations EMS use rose in all Treaty areas from 2018 to 2020 (a 15.4% increase provincially). Forty-five individuals participated in sharing circles. Participants identified transport barriers, including distances to hospitals, as a cause of EMS use. Participants also described cases of First Nations patients declining transport to hospitals by EMS, due to lack of options to return home following care, or being stranded far from home at hospital discharge. Lack of paramedic cultural and geographic knowledge and training were discussed as barriers to providing timely and appropriate care. Stereotypes of First Nations people reported to operate within emergency care included ideas of First Nations patients as using EMS for non-urgent concerns, as substance using and as drug-seeking. Paramedics described negative experiences with receiving hospital staff dismissing patient conditions. Specific hospitals were viewed by both patients and paramedics as undesirable care destinations for First Nations patients. First Nations paramedics introduced perspectives of the advantages and difficulties of practicing in their own communities. They also spoke of a lack of collegiality they experienced with other paramedics. Conclusion: Our study offers a better understanding of issues impacting paramedic care on First Nations and a foundation for service improvement. There is a need for greater support for First Nations members to enter and remain in the paramedic profession. Introduction: Status First Nations members make up 9.4% of emergency department (ED) visits in Alberta, and have reported negative experiences in ED. As part of a mixed methods study on quality of care for First Nations members, we sought to document First Nations members' experiences before going to ED, while in the ED, leaving the ED and after the ED. This complements interviews we conducted with ED providers on First Nations members' care (reported separately).

Methods: In partnership with First Nations organizations and under the direction of a First Nations Knowledge Holder (LB), we conducted interviews with First Nations Health Directors and sharing circles with community members. Partner organizations invited participants. Data were transcribed verbatim and analyzed thematically. The project adheres to principles of First Nations Ownership, Control, Access to and Possession of data. Results: Sharing circles were attended by 24 community members belonging to 10 Nations as well as 2 non-First Nations care providers. Four First Nations Health Directors participated in interviews. Sharing circle participants' ages ranged from 20 to 70 years old, with the majority (61.5%) over 50. Participants included 7 men (27%), 18 women (69%), and 1 transgender person (4%). Seventyseven percent lived on reserve. They reported visiting ED 3.4 times on average in the previous year, with a range of 0 to 20 times. The majority (77%) had experience taking older relatives to ED and all but 1 had experience taking children to ED. Participants spoke about systemic barriers to care before attending ED, while in ED, at discharge and after leaving ED. Themes included avoidance of care, avoidance of specific hospitals, over-prescribing of narcotics, assumptions of drug use, harmful care, inadequate or inappropriate care, racism and concerns with wait times. Participants also provided explanations for quantitative findings of the wider project, and made recommendations for systems change to improve care. These included recommendations to build connections between EDs and First Nations-run health services. Conclusion: Results document First Nations members' deep and extensive knowledge about the barriers to emergency care they face. System and department level initiatives to improve care for First Nations members should be conducted though partnerships that share power between health systems, First Nations leaders and knowledge holders. Keywords: equity, First Nations, qualitative methods Introduction: Acute infectious diarrhea is usually a self-resolutive disease, but still represents an important cause of outpatient visits. Risk stratification tools could help identify patients at low risk of complications who could self-treat at home without medical care. The objective of this systematic review was to identify tools that predict the risk of complications for patients presenting to an outpatient clinic or an emergency department (ED) with acute infectious diarrhea. Methods:

We searched Medline, Embase, Cochrane Library, Web of Science and CINAHL databases from inception to July 22, 2021. The PICOS statement was as follows: population (adult or pediatric patients visiting the ED or an outpatient clinic with acute infectious diarrhea), intervention (scale, algorithm for risk evaluation), comparison (none), outcome (intravenous rehydration, hospitalization) and study design (observational or interventional studies). Two reviewers independently screened abstracts and full texts for eligibility, extracted data using the checklist CHARMS and assessed the risk of bias using PROBAST. The study protocol was registered on PROSPERO (CRD 42020166242).

Results: Five articles reporting on two different tools were identified. The EsVida scale was developed to assess the risk of hospitalization of children aged 1-13 years old with acute infectious diarrhea but has no external validation. Initially developed to assess dehydration severity in children, the Clinical Dehydration Scale (CDS) is an 8-point score that has also been studied to evaluate its performance in predicting the risk of complications of children aged 1 month to 5 years old. To predict intravenous rehydration, a CDS score C 1 (3 studies) has a sensitivity of 0.73 to 0.88 and a specificity of 0.38 to 0.69, while a CDS score C 5 (3studies) has a sensitivity of 0.06 to 0.32 and a specificity of 0.94 to 0.99. Similarly, to predict hospitalization, a CDS score C 1 (2 studies) has a sensitivity of 0.74 to 1.00 and a specificity of 0.34 to 0.38, while a CDS score C 5 (3 studies) has a sensitivity of 0.26 to 0.62 and a specificity of 0.66 to 0.96. High heterogeneity among studies and high or unclear risk of bias precluded a meta-analysis. Conclusion: As a riskstratification tool, the CDS has only been studied in preschool children with acute infectious diarrhea and has yielded inconsistent results. Further research is needed to develop and validate a risk-stratification tool suitable for adults. Keywords: acute infectious diarrhea, intravenous rehydration, risk stratification tool

What are the rates of missed pelvic inflammatory disease diagnoses in the emergency department Introduction: Pelvic inflammatory disease (PID) is a common infectious cause of lower abdominal pain in females encountered and treated in Canadian emergency departments (ED). Previous evidence suggests that approximately two thirds of cases of PID go unrecognized. The exact characterization of missed ED diagnoses of PID, and the long term sequalae of these missed diagnoses has not been studied. Methods: We conducted a multi-centre retrospective medical records review. We interrogated the ED database of our urban academic ED from July 2018-Jan 2021 to identify cases using ICD-9 codes. Data were abstracted from individual patient records by two trained research assistants using a standardized collection form. Information on ED course, vitals and investigations were extracted as well as data from clinical documents from emergency physicians, consulting specialist physicians, nursing and allied health. Continuous variables were described using counts, means, medians and IQR. The primary outcome was the missed diagnoses of PID. Results: Between July 2018 and Jan 2021, we identified 160 female patients diagnosed with PID in our urban ED. The median age was 29 (IQR 23-37). The majority of patients for whom data was available (n = 115, 92%) were sexually active at time of diagnosis. Thirty-one patients (19%) had previously been diagnosed with PID and 38 (23%) had a past medical history of a sexually transmitted infection. Twelve patients (7%) had presented to the ED with a discharge diagnosis of possible PID in the 7 days prior to diagnosis. A further 4 patients presented to the ED 30 days prior to diagnosis with discharge diagnoses consistent with possible PID. The majority of patients only had one presentation prior to diagnosis (n = 12), however, one patient had three visits prior to diagnosis. The discharge diagnoses in the 30 days prior to PID diagnosis included: abdominal pain not yet diagnosed (n = 3, 9%), urinary tract infection or pyelonephritis (n = 3, 9%), and pelvic and perineal pain not yet diagnosed (n = 2, 6%). Conclusion: Of patients ultimately diagnosed with PID in the ED, 13% presented to the ED in the 30 days prior to their diagnosis with discharge diagnoses consistent with possible PID. Our study suggests that rates of missed PID are lower than previously reported, but still represent an important area for quality improvement for emergency physicians. Keywords: diagnosis, pelvic inflammatory disease, women's health Methods: We conducted a multi-centre retrospective medical records review. We interrogated the ED database of our urban academic ED from July 2018-Jan 2021 to identify cases using ICD-9 codes. Data were abstracted from individual patient records by two trained research assistants using a standardized collection form. Information on ED course, vitals and investigations were extracted as well as data from clinical documents from emergency physicians, consulting specialist physicians, nursing and allied health. Continuous variables were described using counts, means, medians and IQR. Results: Between July 2018 and Jan 2021, we identified 160 female patients diagnosed with PID. The median age was 29 (IQR 23-37), and the majority of patients for whom data was available (n = 115, 92%) were sexually active at time of diagnosis. Thirty-one patients (19%) had previously been diagnosed with PID, and 38 (23%) had a past medical history of a sexually transmitted infection. Of the 160 patients diagnosed with PID, 23 patients did not have a pelvic exam performed: i) no reason documented (n = 14, 9%), ii) patient unable to tolerate the exam (n = 7, 4%), iii) patient declined (n = 2, 1%). Of patients having a pelvic exam, the majority (n = 92, 67%) had documented CMT. Treatment of PID in this study was variable: 7 (4%) patients diagnosed with PID in our cohort did not receive any antibiotics, and only 62 (38%) of patients received an antibiotic regime (antibiotic choice, frequency, and dosing) consistent with published Canadian or British Columbia guidelines. Patients receiving first line antibiotics were less likely to return to the ED within 30 days of discharge. Conclusion: In this study of females with PID diagnosed and treated in the ED, 14% did not receive pelvic examination, and only 38% received antibiotics consistent with published guidelines. Further research is urgently needed to improve the treatment and diagnosis of PID in the ED, and highlight emergency physician knowledge gaps around this important diagnosis.

Keywords: guideline adherence, pelvic inflammatory disease, women's health Introduction: Pelvic inflammatory disease (PID) is a common infectious cause of lower abdominal pain in females encountered and treated in Canadian emergency departments (ED). Given the high prevalence of PID and it's considerable morbidity, additional knowledge is required to describe the Canadian context. Methods:

We conducted a multi-centre retrospective medical records review. We interrogated the ED database of our urban academic ED from July 2018-Jan 2021 to identify cases using ICD-9 codes. Data were abstracted from individual patient records by two trained research assistants using a standardized collection form. Information on ED course, vitals and investigations were extracted as well as data from clinical documents from emergency physicians, consulting specialist physicians, nursing and allied health. Continuous variables were described using counts, means, medians and IQR. Results: Between July 2018 and Jan 2021, we identified 160 female patients diagnosed with PID in our urban ED. The median age was 29 (IQR 23-37). The majority of patients for whom data was available (n = 115, 92%) were sexually active at time of diagnosis. Thirty-one patients (19%) had previously been diagnosed with PID and 38 (23%) had a past medical history of a sexually transmitted infection. The most common presenting symptoms were lower abdominal pain (n = 126, 79%), abnormal vaginal discharge (n = 71, 44%), and nausea/vomiting (n = 70, 44%) . Of patients having a pelvic exam, the majority (n = 92, 67%) had documented cervical motion tenderness. The majority of patients had normal vital signs (temperature \ 38 n = 156, 98%; heart rate \ 100 n = 108, 68%) and normal leukocytes (mean white blood cells 11.2, CI 10.4-12). Positive chlamydia and gonorrhea on microbiological testing were uncommon (genital chlamydia n = 18, 11.3%, genital gonorrhea n = 10, 6.3%). A majority of patients (n = 103, 64%) had imaging performed in the ED. Forty (25%) patients were admitted, and 24 (15%) patients returned within 30 days of discharge. Of those returning within 30 days, seven (29%) were diagnosed with recurrent PID, and an additional 5(21%) were diagnosed with abdominal pain not yet diagnosed. Conclusion: PID is a complex disease with a heterogenous presentation. Our findings highlight the importance of clinical symptoms and physical exam and demonstrate that chlamydia and gonorrhea are less common than previously documented. In this study, high rates of return to the ED and admission, indicate a significant morbidity burden for patients. Keywords: pelvic inflammatory disease, women's health Introduction: The COVID-19 pandemic has imposed pressure for changes in the care provided in emergency departments (ED). We sought to explore the perceptions of different types of ED workers involved in resuscitation teams regarding these practice changes, aiming to understand the barriers and facilitators affecting their adoption. Methods: We conducted this exploratory qualitative study using a narrative analysis approach. Participants were members of a multidisciplinary resuscitation team in an urban tertiary care academic ED (nurses, orderlies, respiratory therapists, emergency physicians, and case managers). They were recruited using a purposive sampling technique to maximize variation and aim for even representation of occupation and shift type. Focus groups with workers in the same occupation were interviewed in a semi-structured manner. The data were analyzed thematically using an inductive approach, combined with a structuring framework (Theoretical Domains Framework). Two members of the research team corroborated the coding and analysis of the data. Results: Thirty-three participants took part in six focus groups in March and April 2021. The changes described were related to protective measures, site organization, team functioning and care protocols. Participants identified barriers to their adaptation such as: (1) information overload due to multiple sources of information and frequent changes, (2) Background: Intubation is a high-risk intervention in the emergency room. In recent years emergency department airway registries have been implemented around the world as a quality improvement and research tool to better understand and optimize this important procedure. In 2019, an airway registry was initiated at two tertiary emergency departments in southwestern Ontario, referred to as site A and B. Data is currently collected using paper forms completed by clinicians after intubation. Sporadic enrollment has been a challenge, especially at site B which captured only 16.5% of emergency department intubations on average in the first 6 months of 2021, compared to 54.7% at site A.The aim of this project is to increase enrollment to 50% or greater at Site B in 6 months. Aim statement: Initially a stakeholder analysis was performed involving departmental leadership at site B as well as those involved in the airway registry. Conversations with these parties were also used to help inform process mapping which was conceptualized via Ishikawa diagram. Four quality improvement interventions targeting perceived barriers derived from process mapping were planned at onemonth intervals starting in July of 2021. They were evaluated using the Plan, Do, Act, Study cycle and interventions as well as monthly rate of enrollment was tracked using a run time graph. Interventions include adding the airway registry form in the standard documentation for resuscitations, the charge nurse placing an identification sticker on the airway form and handing it to the clinician, new signage in the department reminding clinicians to complete the form and monthly updates to the physician and management groups marking progress.

Measures and design: The timeline had to be readjusted in the context of institutional crises related to COVID. Despite these setbacks, result have been promising to date with average monthly enrollment increasing to 30% after the first intervention-including the form in the resuscitation package. Analysis is pending for the second intervention and the third and fourth are planned for December and January respectively. Evaluation/results: With the increasing prevalence of airway registries, understanding how to optimize the rate of capture is of increasing importance. Given that a basic literature review demonstrated no prior publications on interventions to improve compliance this quality improvement initiative offers novel insight into the science of optimizing airway registry compliance with promising initial results. Introduction: Emergency airway management involves sedating, paralyzing, and introducing a tracheal airway device in a critically ill or injured patient who requires emergent airway management. This highrisk and complex procedure requires extensive training for the operator and careful selection of sedative and paralytic medications. Existing emergency airway registries focus on large, often international, academic medical centres, and report first-attempt success rates greater than 80%. Our objectives were to determine the first-attempt success rate of emergency intubations recorded in the registry in a large urban academically affiliated community hospital emergency department (ED) and compare success rates between paralytic medications used. To meet these multiple goals, we created the Emergency Medicine AirWAy REgistry (EM-AWARE). Methods: ED patients requiring emergency airway management were identified using a departmental form (paper or electronic) filled out by physicians prospectively from 2015 to 2020. Data collected included patient characteristics (age and sex), intubation operator characteristics, operator pre-and peri-intubation methods, post-intubation success rate and complications. Missing data was retrospectively collected by trained research assistants. We reported proportions with 95% confidence intervals and comparisons between paralytic use and type were made using Fisher exact tests. P values \ 0.05 were considered significant. Results: During the study period, 179 intubation events were recorded at our centre (mean age, 55 years; 41% female, 56% male, 3% not recorded Dalhousie University, Saint John Regional Hospital, Saint John, NB Introduction: Overnight shift work has been reported as a major contributing factor to the excessively high burnout rates seen in emergency medicine physicians. Casino-shift scheduling has been used to mitigate this issue in emergency departments (EDs). While benefits of the model have been demonstrated with regards to physician well-being, there is a lack of reported evidence assessing if this scheduling model impacts patient flow. The objective of our study was to determine the effect of a casino-shift trial on overnight patient flow variables in a tertiary centre emergency department. Methods:

We conducted an analysis of routine administrative data for overnight (10:00 pm-10:00am) patient flow variables comparing a two-month casino-shift trial (September 9th, 2019-November 4th, 2019) with control periods (the two-month period preceding and following the trial period (July-Aug 2019 and Nov-Dec 2019) and the same corresponding six-month period for the preceding year (July-Dec 2018) at the Saint John Regional Emergency Department. We included all patients presenting to the ED overnight between 10:00 pm-10:00am during the study period. Our key outcome measures were wait time to medical assessment (WT), total ED length of stay (LOS), and admission rates. Introduction: The detrimental effects of language barriers on patients in the Emergency Department (ED) are well documented. The use of trained language interpreter services (TLIS) has been shown to improve patient outcomes and decrease hospital readmission rates. This study aims to assess the current perceived level of TLIS use by Emergency physicians and residents in Saskatchewan EDs. Secondary objectives include identifying reported barriers and associated solutions to optimal TLIS use, and factors that influence the preference for or against using TLIS. Methods: An 18-question survey was designed to assess physicians and residents across ED sites in Saskatchewan, Canada. The survey was distributed to 175 ED physicians and residents by e-mail. Statistical analysis was completed, and data were summarized as frequencies or means. Thematic analysis of qualitative data of reported barriers and solutions was performed to identify common themes among responses. Results: The survey was completed by 41/175 ED physicians and residents, with a response rate of 23.4%. Participants practiced in geographically diverse areas, including urban (42.5%), regional (22.5%), and rural (35.0%) ED sites. Respondents reported significantly lower rates of TLIS use in current practice as compared to a hypothetical ideal rate of use, as determined by each respondent, across all ED scenarios. Commonly reported barriers to TLIS use included time constraints (32%), inadequate physician training for TLIS services (34%), and lack of awareness of the interpreter services available (41%). Conclusion: Trained language interpreter services are currently underutilized in Saskatchewan EDs, which is consistent with previous findings across North America. Physicians report using TLIS significantly more in an ideal scenario compared to current practice. The most common barriers to use include a lack of physician awareness on the current interpreter services available, inadequate training, and time constraints. Potential solutions include increasing awareness of TLIS, offering additional training materials and opportunities, and introducing TLIS triage protocols. Keywords: emergency department, interpretive services, language

Predictors of adverse outcomes in elders hospitalized for isolated orthopedic trauma: A multicenter cohort study Introduction: Patients 65 years of age or older now represent more than 51% of injury hospitalizations in Canada. There is a knowledge gap on the characteristics of elderly injury patients who could benefit the most from an interdisciplinary treatment approach to improve outcomes and reduce resource overuse. We aimed to identify variables that are associated with adverse outcomes in elders admitted to a trauma center for an isolated orthopedic injury. Methods: We conducted a multicenter retrospective cohort study between between April 1st, 2013, and March 31st, 2019 on elders hospitalized with a primary diagnosis of isolated orthopedic injury (n = 19,928) . Data were extracted from the provincial trauma registry (Registre des traumatismes du Québec-RTQ). We used a multi-level logistic regression to estimate the associations between potential predictors and adverse outcomes (extended length of stay, mortality, complications, unplanned readmission, and adverse discharge destination). Results: Comorbidities, type of orthopedic injuries, admission in the year before the injury, age, Glasgow Coma Scale score, severe orthopedic injury (Abbreviated Injury Scale 3), concomitant head injury, and biologic sex were all significant predictors of adverse outcomes. In addition, patients admitted for surgical care had a lower odds of mortality, unplanned readmission, and adverse discharge destination. Conclusion: We identified nine predictors of adverse outcomes in patients C 65 years of age admitted to a trauma center for orthopedic injury. These variables will eventually be used to develop a clinical decision rule to identify elders who may benefit the most from interdisciplinary care. Keywords: elders, geriatrics, orthopedic

Factors associated with mortality in patients sustaining a moderate trauma brain injury: a multicenter cohort study Introduction: Trauma brain injuries (TBI) may cause long-term consequences and are among the leading cause of mortality.Because of the ageing population, more TBI due to falls are globally reported, and elderly patients have higher rates of TBI-related deaths. The literature also reported older moderate TBI (mTBI) patients are associated with higher mortality rate compared to their younger counterparts. There are few studies which focus on moderate TBI and elderly-related factors such as comorbidities. We aimed to identify factors associated with mortality among patients admitted for moderate TBIs. Methods: We conducted a retrospective cohort study (2003) (2004) (2005) (2006) (2007) (2008) (2009) (2010) (2011) (2012) (2013) (2014) (2015) (2016) (2017) on patients C 16 years of age with a moderate TBI (Glasgow Coma Scale, GCS: 9-12) admitted to any adult-level I Trauma Centers emergency department. Data were extracted from the Quebec Trauma Registry (RTQ). We excluded patients who died on arrival. To identify factors associated with mortality, we used logistic regression to develop multivariate models. We also performed sensitivity analysis. We thus developed two separate multivariate logistic regression models including all mTBI patients with severe concomitant extra cranial injury (Abbreviated injury scale, AIS C 3) and those with minor extra-cranial injury (AIS B 2). Results: We included 1,143 patients with a mTBI. Overall, 71.3% were male, 50% had a GCS of 9 or 10; 64.9% had GCS motor response of 5 points; 41.9% were transferred from another health care facility. The in-hospital mortality was 17.3%. Increasing age, certain comorbidities such as diabetes, coagulopathy, other comorbidities (metastatic cancer, connective tissue disorders, gastro-intestinal disorders, and rheumatic disease) was associated with increased odds of mortality. Transferred patients, and GCS motor response of 5 points were associated with decreased odds of mortality. In sensitivity analysis, male sex, cardiovascular disease, diabetes, and mechanism of injury (falls) were associated with increased odds of mortality among patients admitted for mTBIs with minor and severe extra cranial injuries. Conclusion: This study showed that comorbidities, GCS motor response, transferred patients, and face injury were important factors of mortality among patients admitted with moderate TBI. The minor and severe concomitant extra-cranial injuries had no effects on mortality. These factors could be used to identify patients with moderate TBIs with higher risk of mortality. Keywords: brain, geriatrics, trauma P53 Twenty-four hour and one-week health system utilization and outcomes of urgently triaged callers to a nurse-managed provincial health information telephone service after initiation of supplemental virtual physician assessment ) had no health system encounters over the next 7 days. Eight (0.1%) callers died within 7 days of a HeiDi VP assessment, 5 of whom were advised to attend ''ED now''. 54 (2.9%) callers with a ''try home treatment'' disposition were admitted to hospital within 7 days of a VP assessment, and no callers who were advised home treatment died. Evaluation/results: This is the first study in Canada to rigorously and comprehensively examine health service utilization and outcomes arising from the addition of VPs to an RN-staffed provincial health information telephone service. Our findings suggest HeiDi VP supplementation safely reduces the overall proportion of callers advised to seek urgent in-person visits. In future work we intend to conduct an economic evaluation and investigate predictors of concordance between advice given and subsequent service utilization.

Keywords: health service utilization, quality improvement and patient safety, telehealth Introduction: Unrecognized delirium in the emergency department (ED) remains common (50-75%) despite a threefold mortality increase for those discharged home. Current models posit that delirium may be a problem with brain functional connectivity. Scalp recording of electrocerebral activity may offer a non-invasive means of detecting delirium in the ED. Objective: To review the use of oscillatory-based functional connectivity measures in assessing altered states of consciousness and to determine the feasibility of using EEG in the ED. Background: The prevention of falls with harm for patients continues to present challenges for hospitals globally. Patients admitted to hospital have an increased risk of falling during their inpatient stays. Falls are the leading cause of preventable injury during hospital admissions, and may lead to decreased mobility, functioning and participation in daily activities. Implementing falls prevention strategies-especially from the outset of patient admission-may be associated with decreased falls during inpatient stays.We implemented a zero harm approach to falls prevention. Our aim was to reduce falls with injury by 25% within one year. Aim statement: We implemented a multifaceted and multidisciplinary quality improvement falls prevention strategy that included facilitating organization-wide education, adopting the Morse Fall Risk Assessment tool, displaying real-time unitspecific fall rates, and implementing a transparent root-cause analyses process after falls. Our outcome measure was falls with injury per 1000 patient days. Measures and design: We observed a decrease in the rate of patient falls with injury from 2.03 (baseline period) to 1.12 (one year later) per 1000 patient days. We also observed increases in awareness around falls prevention, and patient safety incident reporting. Evaluation/results: Our zero harm approach to fall prevention from the ED to hospital wards reduced falls with injury while improving our patient safety culture.

Keywords: falls prevention, quality improvement and patient safety, zero harm

Measuring the association between beta blockers administration and good neurological outcome in cardiac arrest patients Studies have demonstrated that older adults, despite being assigned a 'low acuity' CTAS score, are more likely than their younger counterparts to be admitted. The primary objective was to describe characteristics and outcomes of older adults presenting to the ED who are assigned a 'low acuity' triage score. The secondary objectives were to compare characteristics and outcomes: (1) to a younger control group; and (2) between age subgroups of older adults.

Methods: A retrospective cohort study was performed on ED visits of patients assigned a CTAS score of 4 or 5, considered 'low acuity', between July 1 -September 30, 2019 at two urban EDs in St. John's, NL. All patients aged 65 years and above were selected for this study, who were then divided into age subgroups of 65-74 years, 75-84 years, and C 85 years for analysis. Patients aged 40 to 55 years were selected as controls. The primary outcome was admission to the hospital at initial ED visit. The secondary outcomes included length of stay in the ED and re-visit to the ED within 14 days, among others. Chi-squared and ANOVA tests were performed where appropriate. Statistical significance was set at p \ 0.05. Results: Older adults were more likely to arrive via EMS, and present with genitourinary issues and falls/mobility issues compared to controls (p \ 0.05 for all). Older adults also received a social work consult more frequently (p \ 0.05). Admission was not more likely in older adults (p [ 0.05), but they had more frequent ED visits and hospital admissions in the previous six months (p \ 0.05). Among subgroups, those C 85 years were more likely to be triaged in a bedded unit, have social issues, receive IV fluids, IV antibiotics, and consultation for admission (p \ 0.05 for all). Those C 85 years were also more likely to be admitted than those \ 85 years (p \ 0.05). Length of stay in the ED and re-visit rates were not different (p [ 0.05). Conclusion: Admission rates were not higher in older adults compared to younger counterparts. However, older adults C 85 years required more hospital resources, had more extensive work-ups, and were more likely to be admitted compared to those 65-74 years. Introduction: Overcrowding Emergency Departments (EDs) is associated to higher morbi-mortality and suboptimal quality of care for ED patients. Management strategies have mostly focused on the early identification and redirection of low-acuity patients to primary care settings; however, impacts are non-conclusive and the generalizability of the identification strategies in other care settings is low. The purpose of this study is to assess the impacts of a redirection process using an electronic clinical decision support system on ED performance indicators. Methods: We performed a retrospective observational study in a ED of tertiary trauma center where a redirection process of low-acuity patients was implemented. The process was based on a clinical decision support system relying on an algorithm based on chief complaint, performed by nurses at triage and not involving physician assessment. All patients visiting the ED from January 2014 to December 2016 were included. We compare ED performance indicators before and after the implementation of the redirection process (June 1st 2015). We performed an interrupted time series analysis adjusted for age, gender, month, day of visit, time of day, triage category and congestion. Results: Over the 468,140 ED visits 9,546 patients have been redirected to a nearby primary health clinic (8% of post-intervention ED visits). After the implementation of the redirection process, the median length-of-triage was similar, median time-toinitial assessment decreased by 14 min ([-17; -12] , p \ 0.001), the median length-of-stay increased by 33 min ([17; 48] , p \ 0.001), the proportion of patients that left without being seen by an emergency physician decreased by 2% ([-3; -2], p \ 0.001). Conclusion: The implementation of a redirection process of low-acuity ED patients based on a clinical support system is associated with the improvement of different ED performance indicators aside from the ED length-of-stay. and integrity (the extent to which a program embodies key features) of CBD implementation, as well as capture early outcomes post-implementation. Methods: This national three-year study surveyed EM program directors annually after implementation (June 2019 (June , 2020 (June , 2021 ). An electronic survey was designed to evaluate the degree of implementation of CBD key features using an innovation configuration mapping approach (anchored scale 1 to 5), and to identify perceived benefits and challenges of CBD implementation. A subset of respondents participated in a follow-up semi-structured interview to better understand their implementation experience and perspectives. Responses to open-ended questions and interview transcripts were analyzed using a simple thematic approach. Results: Mean annual response rate from EM program directors was 6/14 (43%). In 2019, 80% of survey respondents agreed or strongly agreed with the statement that CBD was going well in their program. This percentage declined in more recent years (66. 7% in 2020, and 71.4% in 2021) . In 2021, program directors indicated that competence committees were the most implemented of the key features, whereas individualized resident learning plans were the least. For six of the eight key features, the mean level of implementation decreased from 2019 to 2021. In 2020, 33% of program directors indicated that residents described CBD as having a negative impact on their health and wellness, compared to 57.1% in 2021. The most frequently identified benefits over time included more frequent and better quality feedback for residents, more objective review of residents, and evidence informed individualized learning plans. Common challenges over time were the time, workload, and resource investment in CBD, completion of EPA assessments, and culture change. Conclusion: EM programs are working towards fully implementing key features of CBD. It is unclear if decreasing overall implementation scores represent a decrease in fidelity or a more accurate depiction of the status of implementation. The integrity of implementation is still a work in progress for programs, as many struggle with culture change and the emerging perception of negative impact on resident wellness. Keywords: Competence by Design, education, residency Introduction: In response to the COVID-19 pandemic, the Ontario Ministry of Health introduced a pilot program of 14 virtual urgent care (VUC) initiatives across the province to encourage physical distancing and provision of care by telephone and video-enabled visits. The implementation of the VUC pilot is currently being evaluated by an external academic team. The objective of this study was to understand patient and provider experiences with VUC to determine barriers and facilitators to optimal virtual care as it rapidly expands during the current pandemic and beyond. Methods: The qualitative component of the evaluation used one-on-one telephone interviews with patients, families, providers, and program administrators as the main method of data collection. Patient and family participants were invited to participate by the triage nurse after their VUC visit; all providers and administrators involved in key sites were also invited to participate. Data analysis, using thematic analysis, occurred in conjunction with data collection to monitor emerging themes and areas for further exploration. Results: We completed 14 patient and family interviews and 16 provider and administrator interviews (n = 30 total) . Key themes from the patient and family data included (a) comfort with not having to attend the ED in the height of the pandemic, (b) access to virtual care was highly valued (c) high satisfaction with care outcomes. Providers expressed guarded optimism for VUC including (a) tremendous enjoyment of the work, (b) notice of the distinction between ''expedited advice'' versus ''urgent/emergent care'' (c) the importance of the nurse triage role. Finally, administrators spoke of (a) the logistic realities of setting up the clinics, (b) significance of staffing challenges and (c) sustainability concerns. Conclusion: Virtual care options are valued by patients, families, and providers; however, the nature of care needed by those accessing VUC and who can best provide that care needs to be evaluated to position it for sustainability. Understanding how virtual care performs from both a provider and patient perspective during the current crisis has implications for the practice of ambulatory medicine beyond the COVID-19 pandemic. Keywords: patient and provider experience, qualitative research, virtual urgent care Background: Medical procedures can be a cause of pain/distress for children in the emergency department (ED). Non-pharmacological strategies such as preparation, comfort positions, distraction and relaxation techniques can help children cope with various procedures. A specific dedicated professional such as a Child Life Specialist (CLS) can facilitate these methods for pain and distress management.To evaluate the impact of introducing a CLS in the procedural care of children in the pediatric ED of a tertiary care hospital. Aim statement: This quality improvement (QI) initiative was supported by CHU Ste-Justine Foundation and the hospital-wide initiative Tout doux, which aims to alleviate procedural pain and distress across a pediatric hospital. A pilot project involving the introduction of a CLS in the ED was implanted in July 2021. To evaluate the impact of pain and distress management techniques on patients, CLS interventions were prospectively recorded to report the number and success of strategies used, including non-pharmacological methods employed by the CLS, over a 2-month period. Measures and design: Between Sept 1-Nov 1 2021, the CLS was present during 40 shifts, participating in 137 procedures for 108 patients (mean age: 5.5 yo [6 m-17 yo]). The most frequent procedures were 82 (60%) blood draws/IV-lines insertion. Parents were intramuscular injection. However, for more sustained painful procedures such as fracture reduction, there is insufficient data to recommend NO monotherapy. Keywords: meta-analysis, nitrous oxide, systematic review Introduction: Pelvic injuries can be life-threatening, mostly due to the associated vascular injuries and the risk of rapid hypovolemic shock. Early access to a major trauma centre and specialized interventions are therefore essential to mitigate morbidity and mortality associated with pelvic fractures. This study aims to describe the characteristics of patients with pelvic fracture and the care they are receiving at a major trauma centre. Methods: This is a retrospective cohort study. Prehospital and in-hospital medical records off patients with a pelvic fracture who were treated at the emergency department (ED) of the Hôpital de L'Enfant-Jesus (Quebec City, Canada), a level one trauma centre, between 01/09/2017 and 01/09/2021 were reviewed. Exclusion criteria were isolated hip fracture or pubic ramus fracture. Data are presented using proportions and means with standard deviations. Results: From the 261 patients identified, data are currently available for 155 patients of 59% are men with a mean age of 55.7 ± 21.6 years. Motor vehicles accidents was the main trauma mechanism (n = 70), followed closely by fall (n = 63). Ninety-four patients (61%) were transported directly from the place of accident to the trauma centre, despite 94% fulfilling at least one criterion of provincial prehospital trauma triage protocol suggesting direct transport to the level one trauma centre. Considering all patients, 83 (54%) were directed to the reanimation room at ED arrival and stayed there for an average of 30 ± 17 min. The trauma team leader was activated for 49 patients (31.6%). Forty-five patients (29%) needed transfusion of blood products of which 16 (36%) needed a massive transfusion protocol. Overall, 89 patients (57.4%) required a pelvic surgery with a mean delay of 12.1 ± 6.5 h. Only 7 patients (4.5%) required a pelvic embolization with a mean delay of 2.6 ± 1.5 h, but on the initial computed tomography, 13 patients (8.4%) showed an active bleeding in the pelvic region. A total of 14 patients (9.0%) died during their hospital stay. Mean in-hospital length-of-stay was 22 ± 22 days and 36% were discharged directly to their home. Conclusion: The results presented are preliminary, but we can observe that the proportion of patients with pelvic fracture directly transported to a major trauma centre was insufficient and delays before angio-interventions long. Improving prehospital identification of patients with a pelvic fracture and developing processes that expediate access to specialized trauma care are required. Keywords: fracture, pelvis, trauma

Evaluating the effectiveness of portable near-infrared spectroscopy (NIRS) devices for the detection of TBI in emergency medical settings: A systematic review M. Conrad, BHSc, MPH, C. Barrington, BSc, MSc, C. Laverty, MD Queen's University, Kingston, ON Introduction: Traumatic brain injury (TBI) is a major cause of death and disability worldwide, due to the occurrence of intracranial hematomas, edema and biochemical changes that occur in the brain following injury. The diagnosis of intracranial hematomas in emergency medical settings is currently made with clinical assessments and confirmed with Computed Tomography (CT) of the brain. In settings where CT is not available, clinical assessment alone may be insufficient to identify life threatening hematomas in TBI patients. Portable Near-Infrared Spectroscopy (NIRS) devices are a non-invasive method of measuring cerebral oxygenation, which can indicate whether a hematoma is present. This study aims to investigate whether portable NIRS devices are effective for the detection of intracranial hematomas in adult patients with closed head injury in emergency medical settings. Methods: A systematic literature search was conducted on August 18, 2021. Studies in English were included from inception to August 18, 2021. The Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE and CINAHL databases were searched. Records from unpublished studies including Clini-calTrials.gov and the WHO's ISRCTN were also searched. Screening, data extraction and quality assessment were done in duplicate by independent reviewers, with conflicts resolved by third party adjudication. Study quality and risk of bias were assessed using the QUADAS-2 tool. This study is registered on PROSPERO (CRD42021267735). Results: A total of 8 studies met the criteria for inclusion. 6 studies reported primary outcomes of test sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Reported sensitivity ranged from 75 to 100%, with specificity reported between 44.4 and 81.2%. PPV was reported between 35.50 and 85% and NPV ranged from 72.84 to 100%. One study reported successful NIRS signal collection as a primary outcome, and one study reported qualitative measures of NIRS feasibility such as ease of use of the device. The average time to complete a NIRS scan amongst the 5 studies reporting this metric was 3 min. Conclusion: The results of this study show potential for the use of portable NIRS devices as triage tools or clinical decision aides in settings such as the prehospital environment or other medical settings where CT is not available. High quality research in the form of randomised controlled trials in emergent settings is needed to further prove the clinical benefit of portable NIRS devices. Keywords: closed head injury, near-infrared spectroscopy, traumatic brain injury Background: Some emergency departments (EDs) face a defined closure time which can result in processing capacity issues to manage all patients before department closure. Cobequid Community Health Centre ED (CCHC ED) is located in Halifax, Nova Scotia (NS) with operational hours from 0700 to 2400 with an annual census of 44,000 visits. There is no inpatient or observational capacity so all patients must be discharged from the facility at close. Surges on busier days resulted in patients being triaged and waiting (often many hours) until closure time, only to be redirected to other open EDs in Halifax without emergency physician (EP) assessment. This process was recognized for potential risk to patient safety and patient dissatisfaction.Safely manage newly presenting patients with a consistent approach in overcapacity situations. Aim statement: An Over Capacity Procedure (OCP) was developed to address situations where patient volumes exceeded available emergency staffing resources. The OCP was developed and implemented by our ED Operations Committee with education of the healthcare staff. The procedure involved initiating physician triage of CTAS 2 & 3 patients once defined overcapacity criteria were met. EP decided appropriate disposition. A retrospective quality review of patient outcomes for the period of June 1, 2015 to January 31, 2021 was performed. This study used administrative data supplemented by structured chart review to study all patients managed by the OCP at CCHC ED. Data from ED information system, the Nova Scotia Health Authority electronic medical record and NS Vital Statistics were used. Our primary outcome was death within 72 h of ED discharge without repeat ED visit. Secondary outcomes included number of OCP patients assessed by EP and number of patients presenting to other EDs. Measures and design: A total of 3271 patients were managed by OCP during the study period representing 1.3% of total visits. There was 1 death identified within 72 h of discharge without repeat ED visit on day 3 post ED visit. OCP affected an average of 7.4 patients on days initiated with an average of 1 patient registering at another site before 0700 (maximum was 9). Sixty-three percent of OCP patients were assessed by EP. Evaluation/results: Our OCP was effective in safely managing patient volumes that exceeded processing capacity before closure. There was not significant impact on other facilities. These findings have potential application to other EDs facing closure. Keywords: patient safety, overcapacity protocol, quality improvement P74 Application of music pedagogy to medical education: A scoping review K. Hermann, MD, G. Comeau, PhD, K. Chen, MD University of Ottawa, Department of Emergency Medicine, Ottawa, ON Introduction: Skills obtained in musical training have been found to be transferrable to surgical skills. The skill of deliberate practice, which is nurtured in musical education, also benefits learners when learning clinical skills as well as when acquiring knowledge. Different styles of coaching might also affect the learning experience of medical students. Other concepts frequently discussed in CBME literature, such as growth mindset and mastery of learning, have been used in musical education for a few centuries. In this scoping review, we reviewed the current literature and identify themes that address parallelism in music pedagogy and medical education, and to assess the current state of evidence on the transferability of music pedagogy to competency-based medical education (CBME). Methods: We searched The Cochrane Library (CCTR and COCH), Medline, Embase, PsycInfo, ERIC, Web of Science, Scopus, and ProQuest Dissertations and Thesis Global to retrieve eligible systematic reviews, peer-reviewed studies, editorials, essays, reports, and letters to the editor. We excluded non-English articles and articles that cover medical humanities (54 Articles during title and abstract screening and 9 during full text screening) as they describe the use of music to teach listening and empathy without noting specific pedagogy approach as a foundation. Results: Of 1178 articles initially identified, 688 were included for title and abstract screening after deduplication; 143 articles underwent full-text screening and 55 articles were included for data extraction. We identified 8 themes from the content analysis: Peak performance, Warm-up, Mental practice, Feedback, Learning culture, Deliberate practice, Coaching and Improvisation. Conclusion: Although music pedagogy had been in development for centuries, research has only recently begun to address the transferability of musical training concepts in medical education. The process-oriented training in music illuminates tacit and overt teaching strategies that can be applied in medicine. Reflection, self-assessment, and emphasis on excellence in musical education contrast with the traditional ''see one, do one, teach one'' in medical education. This is an area rich with research potential and its intersectional nature brings forth possibilities for innovation in medical education. Keywords: deliberate practice, medical education, pedagogy IFTs if ongoing opioid analgesia administration is required for pain control. The effectiveness of pain management and efficiency of the EMS system during IFTs are factors impacting patient care which have not yet been examined. Our objectives were to determine whether patients' pain is adequately controlled during IFTs, and to examine the usage of ACPs during IFTs for the purpose of opioid analgesia. Methods: A retrospective chart review of EMS patient care and computer-aided dispatch system records was completed. Adult IFTs from community hospitals to a regional, non-academic hospital, between July 2018 and December 2018 were included. Exclusion criteria: bariatric transfers, diverted transfers, transfers to/from renal dialysis, unstable patients, patients on vasoactive medications. The primary outcome was the absolute value and change in patient pain scores. Descriptive statistics were obtained through analysis in Excel. Results: There were 438 IFTs during the period analyzed. The mean age was 67.1 ± 17.9 and 48% were female. Forty percent (n = 174) of IFTs were related to orthopedic or surgical problems. Of these, 32% (n = 55) had no pain score documented. The majority, 64.6% (n = 283), of IFTs were facilitated by an ACP or ICP. In \ 5 missions, primary care paramedics requested ACP assistance due to inadequate pain control. ACPs completed 14 missions (4.9% of ACP missions) where the only ACP-specific task performed was the administration of opioid analgesia. There was no significant difference between initial and final patient pain scores for missions where analgesia was administered by paramedics (6.0 ± 1.23 vs 4.8 ± 1.18; a = 0.05). There were no significant differences in changes in pain scores when stratified by transport distance or duration. Conclusion: This study suggests that pain management may be inadequate during IFTs. Further research into EMS analgesia delivery and modalities may be warranted. Limited documentation of pain severity during IFTs may limit the external validity of this study. This study also showed that a large proportion of transfer missions are being completed by ACP and ICPs despite low rates of opioid analgesic administration. Further research into the tasking of advanced EMS providers may be useful in improving resource allocation. Keywords: analgesia, pain, transfer Introduction: La réanimation par circulation extracorporelle (RCEC) consiste à utiliser un appareil externe pompant et oxygénant le sang afin de rétablir une fonction cardiopulmonaire chez des personnes qui subissent un arrêt cardiaque (AC). É tant donné les ressources nécessaires pour ce traitement, une pronostication efficace est primordiale pour une juste utilisation des ressources. Il a été proposé que l'oxymétrie par spectroscopie proche infrarouge (SPIR), une technique de monitorage non invasive, puisse être utilisée à des fins de pronostication à la suite d'une RCEC. Ainsi, l'objectif de l'étude est d'explorer l'association entre les valeurs de SPIR et la survenue d'un bon devenir chez les patients ayant bénéficié d'une RCEC. Methods: Une étude de cohorte rétrospective observationnelle a été réalisée à l'aide du registre de cas potentiels RCEC de l'Hôpital du Sacré-Coeur de Montréal (20 décembre 2011 au 19 avril 2019). Les patients n'ayant pas bénéficié d'une RCEC et ceux sans mesures de SPIR cérébrales ont été exclus. L'issue principale était la survie au congé avec un bon devenir neurologique ('Cerebral performance category' = 1-2) et l'issue secondaire la survie au congé hospitalier. L'évolution de la SPIR (valeur moyenne, minimale et maximale) au cours des 72 h a initialement été décrite. Les valeurs de SPIR (appareil INVOS TM 5100C) entre les patients avec un bon et un mauvais devenir ont par la suite été comparées à chaque étape à l'aide d'un test de t. É tant donné la nature de l'étude, tous les patients éligibles ont été inclus. Results: Parmi les 37 patients dans le registre au moment de l'extraction, seuls huit ont pu être inclus dans la présente étude (7 hommes [85,5%]; âge médian = 59 ans ). Quatre (50%) patients ont survécu au congé hospitalier, dont trois (37,5%) avec un bon devenir neurologique. Nous n'avons pu mettre en évidence de différences statistiquement significative entre la première SPIR médiane (70,2 [± 5,1] vs 64,7 [± 10,9]; p = 0,45), maximale (77, Introduction: Dizziness/vertigo is the third most common presenting complaint to primary care clinics and emergency departments. The vast majority of patients presenting with dizziness are due to a benign selflimiting process. However, up to 5% have a serious central neurological cause. Clinicians are uncomfortable with the risk stratification of patients based on history and physical exam. There is a lack of standardized risk stratification of patients resulting in economic and cost implications. Currently emergency medical services are over utilized and up to a third of patients undergo advanced imaging, the vast majority of which are negative. Our goal was to quantify the average cost of care for patients presenting to the emergency department (ED) with vertigo and identify areas of potential cost savings. Methods: We performed a review of medical records from a tertiary care ED between Sep 2014-Mar 2018, including adult patients with dizziness (vertigo, unsteadiness, lightheadedness), and excluding those with symptoms [ 14 days, recent trauma, GCS \ 15, hypotension, or syncope/ loss of consciousness. Five trained reviewers used a standardized data collection sheet to extract data. Results: A total of 3,109 records were identified and 2,309 patients included (mean age 56 years SD ± 20, Female 58.9%, Kappa 0.91). Only 63 central causes (ischemic stroke 60.3%, TIA 22.2%, multiple sclerosis 7.9%, intracerebral hemorrhage 6.3%, tumor 1.6%) of dizziness were identified. Imaging was performed in 1008 (43.7%), laboratory investigations in 2099 (90.9%), and emergency medical services (EMS) transported 522 (22.6%). Average cost per patient for ED-based evaluation was $1160.32 (SD ± 847.19).

The evaluation of patients with vertigo/dizziness currently involves the utilization of substantial ED resources. Imaging and EMS transport are particularly high-cost areas for potential intervention to optimize cost of care for vertigo patients in the emergency department. Keywords: dizziness, healthcare spending, vertigo Introduction: Trauma is a major concern in young adults but the increase in life expectancy has contributed to profound changes in the demographics of trauma patients and older adults have become a rapidly expanding age segment of this population. Exploring the specificities of this population to optimize trauma care is crucial. Weassessed the evolution of incidence, epidemiological and trauma pattern characteristics, and outcome of trauma patients in a 15-years period. Methods: We conducted a multicentre retrospective observational study, which included data extracted from the Quebec Trauma Registry. All patients aged 16 years and over admitted to level-I adult trauma centres with a primary diagnosis of injury between 2003 and 2017 were included. The population was stratified into two age groups: 16 to 64 years and 65 years and older. Descriptive analyses were performed. Results: A total of 53,324 patients were included and 24,822 (46.5%) were aged C 65 years. Median age increased from 57 [36-77] to 67 years. The proportion of older adults (C 65 years) significantly increased from 41.8% in 2003 to 54.1% in 2017; and since 2014 they represent more than 50% of admissions. Falls were constantly the main injury mechanism among older adults (minimum was 84.7% in 2010) and this is also the case among younger adults since 2009. The proportion of motor vehicle accidents slightly decreased in both groups (from 9.9 to 7.8% among older adults and from 41.6 to 37.8% among younger adults). Severe injuries (AIS C 3) to the thorax (3.2-12.1% and 9.1-19.6%) and the spine (2.4-7.4% and 8-13.1%) increased in both groups (C 65 and 16-64 years) while upper (3.6-0.8% and 6.1-2.4%) and lower extremities injuries decreased (52.9-28.6% and 23.4-17.8%). Severe head trauma significantly increased from 16.8% to 25.5% among older adults. The proportion of severely injured (ISS C 12) older patients almost doubled from 17.6% in 2003 to 32.3% in 2017. Among those, mortality varied from a minimum of 16.1% (2017) to a maximum of 21. 8% (2005) . With regards younger adults, the proportion of severely injured patients increased (41.8-53.9%) but their mortality decreased (9.4% to 5.3%). Conclusion: It is undeniable that the demographic changes deeply affected the characteristics and injury patterns of Canadian trauma patients. These findings emphasize the urgent need to optimize prevention and injury care in the older patient's population. Keywords: changes, epidemiology, geriatric trauma Violence Treatment Centre (SADVTC) in London, Ontario sees patients by self-referral or referral by healthcare providers. The number of patients experiencing DV that seek care in the Emergency Department (ED) prior to or after SADVTC visit is unknown. Methods: The primary objective of this study is to determine the number of DV survivors presenting to the ED within 7 days of their SADVTC visit in London, Ontario, and frequency of return visits to ED within 48 h of initial presentation for concerns related to DV. A retrospective review of ED records for all patients presenting to the SADVTC with primary concern of DV between April 1, 2014 and December 31, 2019 was conducted. Inclusion criteria were defined as age C 16 years, indication of experiencing DV in the chart, presenting to London Health Sciences Centre ED (University Hospital, Victoria Hospital Adult and Children's Hospitals), or St. Joseph's Urgent Care Center. Individuals \ 16 years or patients who experienced sexual assault were excluded. Data was extracted from the ED charts including demographics, CTAS, ED diagnosis and referral details. Data was summarized using percentages or median [IQR] where applicable. Results: 119 patients presented to SADVTC during study period and 111 met inclusion criteria. Median [IQR] age was 34.0 [27.0-46.0] years and 103/111 (92.8%) were female. 78/111 (70.7%) were seen in ED within 7 days of presentation to the SADVTC with DV-related concerns. Of those seen in ED, 66/78 (84.6%) of were CTAS 2/3 and 12/78 (15.4%) were CTAS 4/5. 57/78 (73.1%) were referred to the SADVTC during their ED encounter whereas 21/78 (26.9%) were not referred or a referral was not documented. Four patients (5.1%) had a return ED visit within 48 h of their first ED presentation. Conclusion: Majority of patients were seen in the ED within 7 days of SADVTC presentation with emergent or urgent concerns. Given this, healthcare workers need to be comfortable managing concerns specific to DV. Almost 75% were referred to the SADVTC during the ED encounter but there may be room for improvement by ensuring physicians are aware of the scope of care provided by the SADVTC. Limitations of this study include missing data due to incomplete charting and inability to capture return ED visits at peripheral centers. Keywords: domestic violence, intimate partner violence, spousal violence Introduction: Prompt identification and management of life-threatening injuries are essential to improve the outcomes of severely injured patients. The aim of this study was to assess the delays between emergency department (ED) arrival and life-saving interventions (LSI) during the care of traumatized patients. Methods: This is a retrospective cohort study of all injured patients who presented to the ED of a level one trauma centre during a two-year period and who had at least one LSI performed. The list of LSI was established by consensus and included: endotracheal intubation, chest decompression, administration of blood products, activation of a massive transfusion protocol (MTP), pelvic binder application, thoracotomy, trauma team leader (TTL) activation and surgical control of bleeding through angiointervention or surgical intervention. Delays are presented in minutes (min) using median (med) and range. Results: A total of 910 patients were included. The mean age was 53.2 (SD 20.8; min-max: 1-98) years old, 680 (73.1%) were men and 142 (15.6%) died. The main trauma mechanisms were falls (n = 394) followed by motor vehicle accidents (n = 294) . Among patients initially directed to a resuscitation room (n = 544), the delays before the following LSI were: endotracheal intubation (n = 83: med 41; range 1-389), administration of the first blood product (n = 121: med 31, range 2-424), MTP activation (n = 44: med 22; range 6-616), pelvic binder application (n = 22: med 26; range 2-101) and TTL activation (n = 219; med 21, range 0-185). For the whole cohort (n = 910), the median delays were 103 min (range 0-1374) before the first chest decompression (n = 145) , 148 min (range 35-1215) before spleen angiointervention (n = 32), 130 min (range 39-329) before liver angiointervention (n = 11) and 139 min (range 40-282) before kidney angiointervention (n = 5) . Conclusion: Delays before key LSI such as blood products administration, MTP and TTL activations as well as access to angiointervention could likely be improved. Subgroup analyses based on the patients' characteristics, trauma mechanisms, initial vital signs and injuries sustained will allow us to explore the factors associated increased delays and opportunities to improve the care provided to severely injured patients. Keywords: emergency department, trauma Keywords: epidemiology, inflammatory arthritis, most responsible diagnosis Introduction: Rapid sequence induction (RSI) is a method of airway control involving the administration of sedatives and paralytics, followed by endotracheal intubation. While it is considered an essential skill for emergency physicians, it is not performed frequently. Physicians often rely on simulation training to maintain their skills. To ensure the effectiveness of these simulations, they must accurately represent clinical encounters. In this study, we compared key patient and case level characteristics between simulations and clinical encounters through evaluation of data from an emergency airway registry. Methods: The EM-AWARE airway registry was created using retrospective data from real and simulated intubations at our centre from 2016 to 2020. Information collected included data points surrounding the intubation event such as patient demographics, clinical presentation, intervention metrics, and outcomes. These data points were then analyzed to compare simulation vignettes to real life cases to evaluate how well they represented actual clinical encounters. Recruitment bias was minimized using a multi-method analysis. Results: We found that our simulated scenarios were not fully representative of the patient population needing emergency airway management. Of the 179 patient charts that were reviewed, 87.15% (95% CI 81.40-91.34%) had a medical indication for RSI, while 10.06% (6.38-15.41%) had a trauma indication. Of the 36 simulations reviewed, 2.78% (\ 0.01-15.42%) were medical and 75.00% (58.74-86.44%) were trauma. Demographics of the patient population also differed significantly between simulated and clinical cases, where 41.3% of real-life patients were female and 2.78% female in the simulations. The average age of patients seen in the ED was 55, compared to 36 in the simulated scenarios. Conclusion: Our findings suggest that our simulations do not reflect typical clinical encounters faced in real life settings, and thus may fail to provide physicians with appropriate practice and training. The findings from this study will allow us to improve the approach to RSI simulated training by improving the realism of case vignettes. Development of additional case vignettes focusing more on the management of medical indications for intubation may be beneficial in simulation team training. Keywords: emergency airway, rapid sequence induction, simulation Introduction: Pain is one of the most common presentations to the pediatric emergency department (ED), and opioids are amongst the top three medications used to treat it. Understanding the reasoning behind physicians' opioid prescribing practices is vital to safer practice, particularly in the context of the ongoing Opioid Crisis. The primary objective of our study was to describe pediatric emergency physicians' decision-making process when prescribing opioids for acute pain management. Methods: This study employed qualitative methodology, using one-on-one semi-structured interviews within a grounded theory analytic framework. We used purposeful sampling to recruit pediatric emergency physicians across Canada. Interviews were conducted over telephone by a qualitative methods-trained interviewer, from December 2019 to January 2021. Transcript analysis occurred concurrently with data collection, allowing for considerations around data saturation and theory development. Results: A total of 11 interviews were completed. Participants represented the main Canadian geographic regions. Interviews revealed nine major themes: (1) treatment setting, (2) medical considerations, (3) physician confidence in the evidence, (4) pain assessment, (5) family-specific considerations, (6) safety concerns, (7) physician personal experiences, (8) physician professional context, and 9) the Opioid Crisis and media influence. All participants identified challenges managing acute pain presentations in the ED, emphasizing the need for better guidance, evidence-based data, and knowledge translation. A family-centered approach was recognized as the goal of practice. After considering all other factors, most physicians indicated that the Opioid Crisis had minimal impact on their analgesic decisionmaking final outcomes. Suggestions for the future included addressing emerging challenges such as the management of opioid dependency and withdrawal in the pediatric setting. Conclusion: Our study explored the decision-making processes for managing acute pain, isolating significant barriers, facilitators, and considerations when pediatric emergency physicians prescribe opioids. This can help inform knowledge translation strategies for safer practice and optimize acute pain management in the emergency department. Keywords: children, opioid crisis, prescribing patterns Introduction: Older adults ([ 65 years) presenting to the emergency department (ED) with delirium are at high risks of adverse outcomes including mortality, long term functional impairments and prolonged hospitalization. These outcomes may vary based on the type of delirium, or motor subtype (i.e. hypoactive, hyperactive, mixed delirium) but this has not been well understood. Identifying factors that distinguishably predict each motor subtype may provide a better understanding of delirium and patient outcomes. Methods: Secondary analysis of a prospective multi-center observational study. Participants included older adults (C 65 years), excluding those with an ED stay \ 4 h, critical illness (CTAS 1), or from a nursing home. We assessed delirium with the Confusion Assessment Method (CAM) and utilized the Richardson Agitation Sedation Scale (RASS) to assess motor subtypes. Three binary logistic regression models were used to contrast: 1) no delirium vs. hypoactive delirium; 2) no delirium vs. hyperactive delirium; and 3) hypoactive vs. hyperactive delirium. Independent measures including age, sex, education level, Mini-Mental State Examination (MMSE), and Charlson Comorbidity Index (CCI) scores were adjusted for in the models. Results: The study sample included 1577 subjects (mean age 81 years, 49% female, 93.3% no delirium, 2.0% hypoactive delirium, 4.7% hyperactive/mixed delirium). Sex, age, education, presence of family/friends, ED length of stay and comorbidities were not statistically significant in any model. CCI was a significant predictor of hypoactive and hyperactive delirium contrasted to no delirium. MMSE was a significant predictor of the motor subtypes across all models. Markedly, a 1-point score increase in MMSE was associated with a 0.91 reduction in odds of having hypoactive delirium compared to hyperactive delirium (p \ 0.05). Conclusion: MMSE and CCI scores are predictive of both hypoactive and hyperactive delirium. Our hypothesis-generating analysis suggests that those with severe cognitive impairment are more likely to have hypoactive delirium than hyperactive delirium. Keywords: delirium, geriatric emergency medicine, motor subtypes P85 Implementation of a regional prehospital-initiated redirection program for patients with low-acuity conditions: An 18-month retrospective cohort study Introduction: Increased ambulance requests and emergency departments visits were anticipated at the COVID-19 pandemic beginning. In response, provincial authorities requested regional organizations to implement prehospital-initiated redirection strategies for patients with low acuity conditions. This study aims to describe the profile of redirected patients and the barriers encountered following the implementation of an emergency medical services (EMS)-initiated redirection program. Methods: This is a retrospective cohort study. All EMS interventions during which the regional coordinating centre (Capitale-Nationale region, Quebec, Canada) in charge of the redirection process was called were reviewed. The redirection centre was available between 8 am and 4 pm from Monday to Friday. On a voluntary basis, the on-site emergency medicine technician (EMT) had the opportunity to call the coordinating redirection centre to initiate redirection of patients with non-urgent condition. Results: Between April 27th, 2020 and October 26th, 2021, 2327 calls were received at the redirection centre of which 1179 patients were finally redirected. The main reasons for not redirecting a patient (n = 1148) were the potential for deterioration or an exclusion criterion after reviewing the case (n = 829, 72.2%), patient not consenting to be redirected (n = 80, 7 .0%) and no outpatient clinic availability (n = 59, 5.1%) . The most frequent initial 911 dispatch Clawson codes of those redirected were sick person (n = 316, 18 .3%) and falls (n = 100, 8.5%) . Most redirected patients were initially considered as low priority (P4 or P7) by the 911 dispatch call centre (n = 708, 61.7%) . The main complaints of redirected patients were non-traumatic lower limb pain (n = 180, 15 .3%), back pain (n = 141, 12.0%), mental-health related conditions (n = 106, 9.0%), fall (n = 75, 6.4%), dizziness or vertigo (n = 74, 6.3%) and general functional decline (n = 70, 5.9%). Patients were frequently redirected to a general practice clinic (n = 544, 46 .1%), their own family physician (n = 144, 12.2%) or a community-based resource (n = 91, 7.7%) . Conclusion: An EMS-initiated redirection program can be implemented in our setting. Patients with non-traumatic pain were the main population redirected while the 911 dispatch ambulance priority level was not a limiting factor in the decision to redirect patients. Promoting the program to EMTs and leveraging this opportunity to develop patient-centered models of care are the upcoming steps. Keywords: emergency medical services, low acuity conditions, redirection Can J Emerg Med (2022) 24 (Suppl 1):S1-S100 

A restrained lying position was used in 61 (45%) children

the remaining, most procedures

Inhaled nitrous oxide (NO) is a dissociative anesthetic gas that produces anxiolysis, sedation, and analgesia. NO does not require IV access, has a rapid onset and offset of action, and has good safety data. Evidence-Grading of Recommendations Assessment, Development, and Evaluation (GRADE), respectively. Where meta-analysis wasn't possible, we summarized results using Tricco et al.'s classification system of ''neutral'', ''favorable'', or ''unfavorable''. The primary outcome was procedural distress. Secondary outcomes included pain, depth of sedation, procedural success, and adverse events (AEs)

NO was deemed either ''favorable'' or ''neutral'' but never

Juvenile Idiopathic Arthritis (JIA) and hospitalization. National Ambulatory Care Reporting System data (years 2008-2018) provided outcome data

Most responsible diagnosis codes for 'Factors influencing health status' and 'symptoms, signs and abnormal findings without a diagnosis' accounted for 33-36% of visits by adults with IA conditions. The next most frequent reasons for attending the ED/UCC were infection (range 11-12% by type of IA) and injury (10-12% by type of IA). For youth with JIA, 25% of all visits were for injury and 21% for infection. The proportion of visits for disease flare varied

On behalf of the CAEP Research Committee, it is our pleasure to introduce the dedicated team of volunteers who made CAEP's 2021 Grant Competition and 2021/22 Abstract Competition such a huge success. We could not have achieved this without the support of our volunteers and our generous EM Advancement Fund (EMAF) donors