2007 CAEP/ACMU Scientific Abstracts - Poster Presentations: 131-159

2007 Scientific Abstracts

CJEM 2007;9(3):183-226

Abstracts: 1-30, 31-44, 46-48, 100-130, 131-159, 161-189, 191-216

Poster Presentations

131 PROSPECTIVE multicenter study of treatment and relapse following emergency department discharge for acute COPD

Rowe BH, Willis V, Mackey D, Lang E, Walker A, Ross S, Sivilotti M, Borgundvaag B, Akhmetshin E. Department of Emergency Medicine, University of Alberta, Edmonton, AB

Introduction: Risk of relapse after ED treatment of COPD exacerbations is uncertain, and previous North American research has included limited data from Canada. Our objective was to determine the treatment and relapse rate after ED treatment for COPD. Methods: 19 Canadian EDs enrolled patients (pts) over the study period. Enrolled pts underwent a structured ED interview and telephone interview 2 weeks later. Inclusion criteria were MD diagnosis of COPD, age > 35, and discharge to home. Relapse was defined as an urgent visit to any ED or clinic within 2 weeks of ED discharge; pts lost to follow-up were counted as non-relapses. Data were analyzed using Chi-2, t-test, Mann-Whitney test, and logistic regression. Results: Of 501 pts, 259 (51.7%) were discharged from the ED; 445 pts (88.8%) completed the follow-up. Most patients were discharged on oral corticosteroids (83.3%) and antibiotics (74.8%); self-reported compliance rates were 87.9% and 80.8%, respectively. Relapse occurred in 6% (95% CI: 3%-9%) by 1 week, and 13% (95% CI: 9%-17%) by 2 weeks. There was no difference in relapse based on sex (13.5% vs. 12.1%, p=0.74). Pts receiving oxygen at the initial presentation to ED were more likely to relapse (17.5% vs. 8.1%, p=0.03). More pts who relapsed had at least one ED visits for acute COPD during the past 2 years (17.0% vs. 6.6%, p= 0.021). Relapse was associated with higher respiratory rate (p=.005), lower earliest peak flow (p=.002) and oxygen saturation not on room air (28.6% vs. 7.5%, p=.000). Relapse was not associated with discharge medications. Controlling for age and sex, respiratory rate (OR= 1.2; 95% CI: 0.98-1.48) and earliest peak flow (OR= 0.23 per 100 L/min; 95% CI: 0.06-0.92) were associated with relapse. Conclusions: Overall, past COPD control (ED visits in past 2 years), ED treatments (oxygen) and initial vital signs (respiratory rate, earliest peak flow and oxygen saturation) but not treatment were associated with COPD relapse. Future research is required to target this high-risk group. Key Words: COPD, Outpatient management, Relapse

132 DEVELOPMENT of performance measures for emergency department care of patients with chronic obstructive pulmonary disease

Harris DR, Holmes A, McCarney J, Innes G on behalf of the British Columbia Emergency Department Protocol Working Group. Department of Emergency Medicine, St. Paul's Hospital, Vancouver, BC

Introduction: Despite the publication of practice guidelines for the management of patients with chronic obstructive pulmonary disease (COPD), there are no published performance indicators for the measurement of emergency department (ED) COPD care. Methods: The B.C. Emergency Department Protocol Working Group (EDPWG) is a multi-disciplinary collaborative with a mandate to develop province-wide ED protocol implementation strategies that will facilitate the introduction of ED care protocols in diverse emergency departments. These strategies are expected to incorporate the capture of performance indicators into everyday clinical practice. As part of its mandate, the EDPWG developed a series of COPD performance measures by systematically searching for existing practice guidelines, appraising their quality, selecting recommendations for inclusion in the final document, and obtaining panel feedback and consensus on the proposed measures. Results: A total of 23 measures were developed. These include patient-specific indicators (vital signs at triage), process measures (length of stay in the ED, time to medications), and outcome measures (percent with a repeat ED visit, percent with admission). Measures also covered topics such as vaccination, smoking cessation counselling and referral for education. Conclusions: The BC EDPWG COPD performance indicators provide a means to standardize measurement of care for patients with COPD in the emergency department. Key Words: COPD, ED management, Clinical practice guidelines

133 EFFECTIVENESS of a training program for emergency medicine residents in ultrasound-guided central venous catheterization

Woo MY, Frank JR, Lee AC, Thompson C, Cardinal P. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: Central venous catheterization (CVC) is a procedure commonly performed in the Emergency Department and is an essential competency for Emergency physicians. Ultrasound-guided CVC insertion is an emerging method that promises greater effectiveness and fewer complications. We developed and evaluated a novel educational training program in CVC using the ultrasound guided (USG) technique. Methods: The study was approved by the research ethics board for a pre-post evaluation design. Sixteen Emergency Medicine (EM) residents agreed to volunteer for the study. After determining their prior experience and baseline knowledge, each participant was videotaped inserting CVC in the right internal jugular vein (RIJ) on models using the USG technique. Participants then reviewed a web-based instructional module and had a practical session. Participants were again videotaped inserting CVC in the RIJ of models. The primary outcome was the change in score pre vs. post on an expert-validated performance evaluation tool used to review the video tapes in a blinded fashion. Participants also completed a questionnaire to measure any change in knowledge and perceived competence following the educational session. Results: Participants were EM residents ranging from year 1 to 5. 81% (13/16) had never attempted one USG CVC. After taking the course, participants reported that the models were realistic. Comparing pre- and post- assessments, both performance scores (12.0 vs 13.5) and global rating scores (3.5 vs 5.5) improved significantly (p<0.01, Cohen's d = 1.12 and 1.28 respectively). There was good inter-rater reliability between evaluators of the videotapes regarding performance scores (r = 0.68) and global rating scores (r = 0.75). All participants felt their confidence and technical skills were improved (p<0.01) and all felt satisfied with the course. Conclusions: This brief innovative multimodal training program was effective in enhancing EM resident competence in USG CVC insertions. Key Words: Medical education, Residency training, EM curriculum, Central line placement

134 FACTORS affecting family medicine residents' choices in choosing further training in emergency medicine

Wang N, Frank JR, Lee AC. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: Little is known about what influences family medicine (FM) residents to pursue 3rd year training in emergency medicine (EM). We studied the factors affecting FM residents when choosing to pursue EM or other 3rd year residency options. Methods: This cross-sectional paper survey was distributed to all Ontario FM residents (years 1 & 2). The instrument was piloted then administered via FM academic sessions at 5 medical schools. Participants were excluded if they were absent the day of the survey, international graduates, already in 3rd year of residency, or funded by the Department of Defense. Participants were asked about their interest in further training and factors known to influence career choice. Results: A total of 249 of the 308 residents responded (80.8%). 67.9% were female, and 55.2% were PGY1. Interest in pursuing 3rd year training was 73.0% during medical school and 68.0% during FM residency. Among those interested in further training, 43.6% decided on EM while in medical school. However, only 20.0% actually applied for a PGY3 position. Out of 55 responses, EM was the most popular 3rd year discipline (24 responses), followed by OB (7), Palliative Care (6), and Anesthesia (4). Comparing those who did and did not pursue subspecialty training, only debt load was significantly different (p=0.03). Other factors such as desired work hours, satisfaction with clinical teaching, satisfaction with related rotations, exposure during residency, availability of 3rd year positions, future practice setting, or number of days of FM practice per week did not differentiate those that did and did not pursue subspecialty training (p>0.05). Conclusions: A majority of Ontario FM residents expressed an interest in further training. Debt load was a significant factor in the decision to apply. EM was the most popular 3rd year program among those applied for. These trends have important implications for educators and policy makers concerned about the current crisis in MD supply. Key Words: Medical education, Residency training, EM curriculum

135 CAN efficiency be learned? a novel workshop to improve physician productivity and emergency department flow

Venugopal R, Lang E, Doyle K, Unger B, Sinclair D, Afilalo M. McGill University Emergency Medicine Program, Montréal, QC

Introduction: Increasingly, the ED requires physicians to focus on productivity and manage ED through-put. We previously reported on a workshop designed to improve these skills and that had been positively evaluated among EM residents. We sought to measure perceptions related to this same workshop among practicing emergency physicians. Methods: Four hands-on workshop stations were designed to simulate key components of ED throughput. These included management of acute and minor care resources, charting/communication skills, and effective/succinct patient sign-overs. Anonymous surveys were completed after the workshop using 5-item Likert scales and qualitative responses. Data is presented with the use of descriptive statistics. Results: Fifteen practicing physicians from across Canada participated. Evaluations were completed by 93% (14/15) of participants. Physician experience averaged 13 years (range 3-27 years). 93% (13/14) rated ED productivity skills as an important part of training or professional development whereas only 28% (4/14) felt it had been somewhat taught or well taught during their training. Ratings of "definitely helpful" or "helpful" evaluations were provided by 85% (11/13) for the sign-over and communication station, by 92% (12/13) for the minor care management station, by 85% (11/13) for the acute care management station, and by 66% (8/12) for the effective charting station. Among all participants 86% (12/14) felt the overall workshop experience to be helpful or definitely helpful. Qualitative feedback suggested that more than 30 minutes per station would be preferable and that practice environment should be considered during workshop design. Conclusions: ED flow management skills are valued yet under-taught to practicing physicians. A flow workshop designed to improve efficiency skills among practicing physicians yielded positive self-assessed evaluations. Teaching this competency appears feasible and might be considered a valuable compliment to CME and professional development activities. Key Words: Administration, Professional development, Training

136 DEVELOPMENT of a training program in pediatric emergency medicine (PEM) in Lao People's Democratic Republic (PDR): a learning needs assessment

Smart KL, Millar KR. Division of Pediatric Emergency Medicine, Alberta Children's Hospital, Calgary, AB

Introduction: Lao PDR ranks in the lowest 25% of countries in socioeconomic health status and has only 41 pediatricians for a country of 6 million people. Pediatric Emergency Medicine (PEM) is a priority area for further training. Our objective was to assess the knowledge, confidence and skill of Lao pediatric residents and graduates in caring for acutely ill children with a learning needs assessment tool. Methods: A three part tool was developed and included:

Multiple Choice Questionnaire (MCQ) to assess knowledge.
Survey instrument (SI) to assess confidence using a Likert scale.
Standardised Assessment Exercise (SAE) to assess clinical skills of residents with a modified OSCE format (2 cases- shock and coma).

The MCQ and SI were given to 19 graduates and 17 residents; residents participated in the SAE. Results: Residents and graduates agreed (scores >3.5/5) with 11/14 statements asserting confidence in diagnosing, managing and performing common procedures in children. Both groups were less confident in managing trauma patients and placement of chest tubes, intraosseus (IO) needles and decompression of tension pneumothorax (scores ≤3.5/5). Less than 50% of graduates and residents correctly answered questions on advanced airway, toxicology, respiratory, metabolic, neurological and surgical emergencies. In the SAE 17/17 residents treated shock with IV crystalloid. 3/17 correctly gave a fluid bolus and continued IV hydration.15/17 would place an IO if unable to obtain an IV in a child with shock; 9/17 could demonstrate placement. 2/17 residents demonstrated a jaw thrust and 12/17 selected the correct mask size for the mannequin.17/17 used benzodiazepine as first line treatment for seizure; 1/17 correctly used benzodiazepines for ongoing seizures. 9/17 correctly gave IV glucose. Conclusions: Lao physicians may overestimate their knowledge and skill in PEM as many performed poorly in areas where they reported confidence. Information from the needs assessment will be used to develop a PEM training curriculum. Key Words: Pediatrics, Training, Skills, Assessment

137 ELECTRONIC procedural logging using handheld computers in an emergency medicine clerkship

Penciner R, Siddiqui S, Lee S. Division of Emergency Medicine, University of Toronto, Toronto, ON

Introduction: The use of handheld computers to track medical students' clinical experience is a relatively new technology in medical education. There are many benefits of electronic logs versus traditional paper logs. The purpose of this study was to assess the technical feasibility and student satisfaction of a novel electronic logging and feedback program using handheld computers during emergency medicine (EM) clerkship. Methods: Fourth year medical students (n=199) at University of Toronto were expected to participate in the electronic logging project during their 4 week EM clerkship between September 2005 to April 2006. Novel software was designed for handheld computers. Following each clinical encounter, students entered data on their handheld computer utilizing check boxes and drop down menus. Students were encouraged to frequently synchronize their handheld computer to transmit their data to a central server. Students were able to view their cumulative log in comparison to peer averages on a feedback website. They were also encouraged to review their logs with their preceptors. At the completion of their EM rotation, students were surveyed for their satisfaction with an 11 item questionnaire using a 5 point Likert scale. Results: Forty-six students (23.1%) participated in the electronic logging project. Twenty-nine students (63%) responded to the survey. Students generally found it easy to complete each encounter (69%) and easy to synchronize their handheld computer with the central server (83%). However, half the students (49%) never viewed the feedback website and most (79%) never reviewed their logs with their preceptors. Overall, only 17% found the logging program beneficial as a learning tool. Conclusions: Electronic logging using handheld computers is a feasible and effective way of tracking clinical encounters and procedures performed by medical students. However significant barriers to widespread implementation include user acceptance and technical problems. Key Words: Handhelds, Education, Logging

138 A web-based lecture series for external resident rotators in the emergency department

Bellazzini MA. Section of Emergency Medicine, University of Wisconsin, Madison, WI

Introduction: Interns from other specialties commonly rotate in academic emergency departments. It is difficult to schedule lectures at our hospital since residents may rotate for a brief time, have other clinical commitments and rotate at various times and dates. We created a web-based curriculum of video recorded lectures including; airway management, toxicology, trauma, wound management and evaluation of chest pain to improve education. Residents took an exam before and after completing the curriculum to assess the effectiveness of the lectures. Methods: Thirty interns from specialties including family medicine, internal medicine, orthopedics and anesthesiology rotated through our emergency department over 6 months. They were asked to take a pretest, view web-based, streaming media lectures during their rotation and complete a posttest after finishing their rotation. Exam content was based on the material in the lectures. Scores before and after completing the curriculum were compared using the paired t-test. Results: Twenty-six interns completed the pretest. Their average score was 66.4%. Eighteen interns completed testing and were used for data analysis. Pretest scores of the eighteen interns who completed testing was 66.9% (SD 11.8). After completing the curriculum exam scores increased to 81.9% (SD 10.4) with a p-value of 0.001. Conclusions: A web-based, steaming media curriculum is effective in increasing knowledge of non-emergency medicine residents rotating through the emergency department. It provides a convenient way to deliver lectures when scheduling formal lectures during intern rotations is not possible or difficult. Key Words: Web-based education, Streaming media, Internship.

140 GEOGRAHIC patterns of emergency room utilization in a rural Ontario community

Furtado N, Lacroix T, Page RJ, Teeple LE. Schulich School of Medicine & Dentistry, University of Western Ontario, London, ON

Introduction: Leamington District Memorial Hospital (LDMH) is not unlike many other Ontario rural community hospitals in that it is not too geographically removed from a larger urban or academic medical centre. Emergency Room (ER) staff members have speculated that patients from outside the geographic service area travel to LDMH seeking shorter wait times. Our primary objective was to ascertain patients' perspectives on why they choose to visit the ER and to determine whether patients are "shopping around' for an ER. Methods: Patients presenting to the LDMH ER were offered a paper-based questionnaire, labeled with a unique identification number. A copy of the Triage Record (TR) with identifying information removed was made and coded with the same ID number. Results: A total of 171 surveys were collected (response rate 90.5%). Only 6% of respondents were from outside the LDMH service area. The most commonly reported reason for coming to LDMH, chosen by 78.9% of respondents, was because it was the closest ER. Mean distances from reported home communities were shorter to LDMH than to alternative ERs (p=0.0001). Patients do not rate their illness severity accurately as there is poor correlation between their self-reported severity and their assigned Canadian Triage and Acuity Scale (CTAS) level (r=0.141). Conclusions: The initial hypothesis that patients in rural areas who have access to two or more Emergency Departments will "shop around' seeking shorter wait times cannot be substantiated by the pilot data collected in the test area of Leamington, Ontario. This finding is supported by both patient reports, which favour proximity as a primary reason for choosing a particular ER, and by distance measures which confirm that LDMH was statistically significantly closer than alternative ERs in Windsor for respondents to travel to. No previous attempt has been made to correlate patient-rated illness severity with CTAS score. Key Words: ED utilization, Geographic patterns, Rural

141 THE consent and prescription compliance (COPRECO) study - does obtaining consent in the ED affect study results in a telephone follow-up study of medication compliance?

McCarvill EM, Campbell SG, Magee KD, Cajee I, Crawford M. Department of Emergency Medicine, Dalhousie University, Halifax, NS

Introduction: Patient compliance with a prescription made in the ED would be expected to result in a better patient outcome than non- compliance. The level of compliance is therefore an important area for research. Research ethics boards commonly prohibit follow up calls to patients without consent being obtained at the time of treatment. The act of consent taken at the time of prescription might, however, be expected to affect the results of a follow survey up for a number of reasons. Primary objective: To determine whether patient-reported compliance with ED prescriptions is affected by the acquisition of consent in the ED for the follow up call. Secondary objectives: To ascertain the level of compliance with ED prescriptions and to find out the degree of displeasure expressed by patients called without prior consent. Methods: Patients given prescriptions in the ED were randomized to having consent obtained during their ED visit, or at the time of follow-up call. Patients were called 7-10 days after their ED visit to determine their compliance with the prescription. Compliance rates between the two groups were compared, as was the level of displeasure expressed by patients called without prior consent. Results: Of 430 patients enrolled, 221 were randomized to consented in the ED (group 1), and 209 were called without prior consent (group 2). Of 169 (76.5%) evaluable patients in group 1, 39.64% were considered "non-compliant' compared to 49.66% "non-compliance' in the 149 (71.3%) evaluable patients in group 2 (p= 0.07). Overall, 44.34% of patients did not fill the prescription or took it incorrectly. Of patients called without prior consent only 0.67% (1/149) expressed displeasure at the call. Conclusions: Non-compliance is a significant issue for patients discharged from the ED. Although there was a definite trend toward greater compliance in patients that consented to the follow-up call, this did not reach statistical significance. Most patients do not appear to object to being called without prior consent. Key Words: Consent, Ethics, Compliance

142 ATTITUDES of medical trainees towards homeless persons presenting for care in the emergency department

Spence JM, Bandiera G, Hwang SW. St. Michael's Hospital, University of Toronto, Toronto, ON

Introduction: Many homeless persons (HPs) use emergency departments (EDs) for care due to barriers to access. Negative attitudes of healthcare workers have been identified as a major theme when HPs are asked about difficult aspects of being homeless. Exposure alone to homeless patients may negatively impact attitudes of medical trainees (MTs). Objectives: To describe the attitudes of MTs towards HPs pre and post-completion of an ED rotation. Methods: The study was conducted in an urban Canadian ED (55,000 visits/yr; 15% HPs). Using a validated 11-question survey, MTs assigned to the ED, were surveyed, before and after rotations, regarding their attitudes towards homelessness. Included were 4 additional questions regarding comfort level providing care. Results: 160 MTs completed rotations. 131 (82%) pre-ED and 113 post-ED surveys were collected. 18 (16%) were lost to follow-up. Mean age was 27 years and 47% were male. 73% had 1 parent with a professional/associate professional occupation. Baseline data showed most had an affiliation with HP: 64% comfortable eating with and 67% comfortable meeting with HP. Cutbacks in housing assistance, low minimum wage and welfare were identified as social causes of homelessness by 51%, 49% and 92% respectively. Substance abuse was identified with homelessness by 66% of MTs. Most (84%) disagreed that little can be done. 77% were comfortable dealing with homeless mental health patients. However, 55% felt overwhelmed by the complexity of problems. The majority (92%) felt care should address medical and social needs. Overall scores did not change after the rotation (Mean pre-rotation score: 45.8; post-rotation: 46.1: p=0.59). Significant differences were found for 2 questions: increased comfort meeting HP (67% vs 74%; p=0.05), and the impact of low wages on homelessness (49% vs 39%; p=0.008). Conclusions: MTs have strong affiliation with homeless persons. Exposure to homeless persons alone during an ED rotation did not impact overall scores. Key Words: Homeless, Attitudes, Survey

143 ARE CTAS level V: non-urgent patients admitted to a rural hospital?

Anstett N, Knight J, Milne WK. South Huron Hospital and University of Western Ontario, London, ON

Introduction: The Canadian Triage and Acuity Scale (CTAS) was "ruralized" in 2002 to allow patients triaged as Level V: Non-Urgent to be discharged from the Emergency Department (ED) by a nurse without seeing a doctor. However, ongoing monitoring was considered essential to ensure that this was safe and effective. A recent study done in an urban ED concluding it would be measurably unsafe to triage non-urgent patients away from the ED. This was because it would lead to inappropriate refusal of care and because some non-urgent patients were admitted to the hospital. The purpose of this study was to evaluate whether any CTAS Level V: Non- Urgent patients were admitted to this small rural hospital. Methods: This study took place at South Huron Hospital (SHH), a typical low volume rural ED in Exeter, Ontario. A retrospective chart review was conducted to determine if any CTAS Level V: Non-Urgent patients were admitted to SHH from January 1, 2005 to December 31, 2005. The CTAS triage scores and hospital admission data were collected for analysis using a retrospective chart review. Results: During the one year study period, 11 546 patients were triaged in the ED and 634 (5.5%) were admitted to hospital. Of all patients triaged at the hospital, 4140 patients (35.9%) presenting to the ED were triaged CTAS Level V and of those 31 (0.75%) were admitted to hospital, totaling 4.9% of all admissions. Conclusions: A small percentage of CTAS Level V: Non-urgent patients were admitted to SHH during the one-year study period in this small rural hospital. Key Words: Rural, Triage level, Hospital admission

Table 1, Abstract 143
CTAS	# of patients	% of all ED visits	# and % of patients admitted
I	31	0.3	4 (0.6)
II	231	2.0	83 (13.1)
III	1705	14.8	193 (46.2)
IV	5439	47.1	223 (35.2)
V	4140	35.9	31 (4.9)
Total	11546	100.0	634 (5.5)

144 RURAL hospital CTAS times

Vlahaki D, Milne WK. South Huron Hospital and University of Western Ontario, London, ON

Introduction: The Canadian Triage and Acuity Triage Scale (CTAS) was adopted and implemented in 1999. The main goal of CTAS is to more accurately define patients' needs for timely care, and provide a common standard to enable emergency departments (EDs) to compare their performance against certain operating "objectives'. An Ontario study published in 2005 showed that these benchmarks were not being met. The purpose of this study was to compare CTAS times from a rural hospital to the provincial averages. Methods: All ED visits to South Huron Hospital (SHH) for 2004 were reviewed. Time to assessment by physician was quantified for each CTAS level. The average length of stay (LOS) for each CTAS level was also quantified. These were compared to the provincial average. Results: CTAS Guidelines SHH Ontario Conclusions: Time to physician assessment at this rural hospital ED exceed CTAS guidelines at every triage level. These times are also significantly better than the provincial average. In addition, the average LOS in the ED at this rural hospital are significantly better than the provincial average. Key Words: Rural, Triage level, Benchmark, Time to assessment

Table 1, Abstract 144. Time to physician assessment
CTAS	Guidelines (min)	SHH (min)	Ontario (min)
1	3	1	36
2	15	12	120
3	30	24	174
4	60	27	156
5	120	26	126

Table 2, Abstract 144. Average length of stay in ED
CTAS	SHH (min)	Ontario (min)
1	203	522
2	189	612
3	151	504
4	78	282
5	53	192

145 DEMOGRAPHICS of medical missions in western Guatemala: experiences from the University of Illinois at Chicago Department of Emergency Medicine International Program

Templeman TA, Lin JY, Schlichting AB. Department of Emergency Medicine, University of Illinois at Chicago, Chicago, IL

Introduction: Since 1999 residents and attendings from The University of Illinois at Chicago Medical Center have partnered with Pastoral de la Salud, a long-term non-governmental organization that works extensively in Guatemala. For a week each year, a group of physicians and support staff from The University of Illinois at Chicago travel to Guatemala where they disperse to remote villages and hold daily clinics for the locale in association with the village's local health promoters. Methods: All documented patient encounters in villages in rural Guatemala precepted by UIC resident and attending physicians for two, five day periods in February of 2002 and 2003 were reviewed for demographics, presenting condition, medications dispensed and specialty referrals. Presenting condition was categorized into abdominal pain, diarrhea, infections, malnutrition, eye complaints, gynecologic complaints, musculoskeletal complaints, respiratory complaints, skin complaints and other. Data were categorized and examined for descriptive and frequency variables. Results: There were 1717 patient encounters at 23 village clinic sites. The patients were 70.8% female and the mean age was 33.5 + 22.4 years. The top presenting condition was musculoskeletal complaints (30.6%) followed by other (15.6%) and abdominal pain (15.5%). The top three medications dispensed were analgesics (28.0%), vitamins (24.4%) and antibiotics (17.3%). 9.1% of patients were referred for further care. Conclusions: The vast majority of presentations were related to musculoskeletal complaints and this was supported by analgesics being the most commonly dispensed medication. The relative acuity in the clinics was low. This raises a concern that resources might be better directed towards local health promoter education instead of direct patient care. Key Words: Rural, Third world, Epidemiology

146 HEALTH status of asylum-seekers in Kibondo, Tanzania

Bartels, SA. Brigham and Women's Hospital, Harvard Humanitarian Initiative, Boston, MA

Introduction: Along the Tanzanian Burundi border there are three way stations, which house up to 75 people each and are intended to provide shelter, food, and water while Burundian asylum-seekers wait to be granted refugee status in Tanzania. In the fall of 2006, the Burundians began to repatriate but they returned home to a devastating drought. The Burundians returned to Tanzania but were denied refugee status and could not return to the refugee camps. Instead, they continued to be housed in the way stations, which quickly became overcrowded and unsanitary. The health situation was deteriorating since appropriate health care facilities were lacking and no disease surveillance mechanisms existed. Methods: Four weeks were spent supervising the outpatient clinics at the three way stations in Kibondo, Tanzania. Daily duties included assessing ill patients who required transport to hospital, escorting ill patients to hospital, and epidemiologic monitoring such as crude mortality rates (CMR) and under-five mortality rates. Results: At the end of March, 10,200 asylum-seekers were being housed in the three way stations. The CMR was stable and below the emergency threshold, reaching 0.7 deaths / 10,000 people / day at its highest. The under-five mortality rate reached the emergency threshold of 2.5 deaths / 10,000 people / day in the last week of March. The major causes of morbidity were malaria (50%), respiratory infections (30%) and diarrhea (9%). The major causes of mortality were respiratory infections (31%), neo- natal death (23%), malaria and malnutrition (15% each) and other (15%). Conclusions: Overcrowding was a major issue in the way stations and presented a challenge to the provision of adequate water and sanitation. The under-five mortality rate reached the emergency threshold by the end of March. Respiratory infections and neo-natal deaths were the most common causes of under-five mortality, although malnutrition was a co-morbidity in almost all childhood deaths. Key Words: Refugees, Third world, Epidemiology

147 DIAGNOSTIC accuracy of emergency department targeted ultrasonography for intrauterine pregnancy: a meta-analysis of current evidence

McRae AD, Edmonds M, Murray H. Dept. of Emergency Medicine, Queen's University, Kingston, ON

Introduction: This systematic review examined the diagnostic accuracy of Emergency Department Targeted Ultrasonography (EDTU) in symptomatic first-trimester pregnancy and generated a pooled estimate of the sensitivity and specificity of EDTU for the detection of intrauterine pregnancy (IUP). Methods: The literature was systematically searched. Studies were selected using predefined inclusion criteria based on guidelines for the critical appraisal of evaluations of diagnostic studies. The sensitivity and specificity of EDTU for IUP was abstracted from selected studies. Pooled estimates of sensitivity and specificity were calculated if the studies showed no evidence of statistical heterogeneity. Three separate analyses were planned: on all included studies, on studies using "definite IUP" (gestational sac & yolk sac or fetus with cardiac activity) as the diagnostic criterion for IUP, and on studies exclusively employing transvaginal sonography. Results: Seven studies met inclusion criteria. The specificity of EDTU for IUP in most studies exceeded 98%. The sensitivity in most studies exceeded 90%. There was significant statistical heterogeneity between the included studies (p<0.01), so a pooled analysis of all included studies was not performed. There was no evidence of statistical heterogeneity between studies using "definite IUP" as the diagnostic criterion for IUP (p=0.0945). The pooled specificity from these studies was 99.2% (95% CI 95.7-99.8%). The pooled sensitivity was 78.0% (95% CI 73.9 to 81.6%). Studies exclusively employing transvaginal sonography were too clinically heterogeneous to permit a pooled analysis. Conclusions: EDTU accurately identified patients with normal IUPs. When "definite IUP" was used as the diagnostic criterion for IUP, the specificity exceeded 99%. Patients with definite IUPs identified using EDTU may be safely discharged from the ED with outpatient follow-up. Patients with no definite IUP identified using EDTU require further evaluation to exclude ectopic pregnancy. Key Words: Emergency Department Targeted Ultrasonography, Intrauterine pregnancy

148 DO emergency physicians overstate the severity of the patient's clinical picture when consulting other physicians?

Hasan AA, Delaney JS, Correa J, Beique M. Accident and Emergency Department, Bahrain Defence Force Hospital, East Riffa, Bahrain

Introduction: Consultations in the Emergency Department (ED) are an important daily act in which all Emergency Physicians (EPs) participate. Obtaining timely and professional advice, intervention or acceptance of transfer of care from another physician is a vital, yet sometimes stressful procedure for EPs. The goal of this research was to ascertain whether EPs sometimes overstate certain aspects of their patient's clinical picture in an attempt to get quicker consultation with less conflict from other physicians. Methods: A web-based survey was sent to 1038 staff Emergency Physicians practicing in Canada. The survey was sent five times over a period of one year. It included 34 questions addressing different issues pertaining to the way EPs practice the consultation process. Results: 525 EPs responded to the survey (50.6% response rate). Of these, 126 (24.0%) stated that they never overstate the severity of their patient's clinical picture to consultants, whereas 399 (76.0%) stated that they did overstate with varying frequencies. Younger physicians were more likely to overstate than older physicians (RR 1.43, p <0.05). The frequency of this practice also varied between different provinces. Conclusions: The process of obtaining consultations in the ED can be a source of stress for EPs. Most EPs surveyed answered that they did overstate aspects of a patient's condition on occasion in order to get more timely consultation with less conflict from the consulting physicians. Key Words: Consultations, Survey

149 FACTORS influencing knowledge translation of the Canadian C-spine Rule among ED nurses

Danseco E, Davies B, Clement CM, Brehaut JC, O'Connor A, Stiell IG. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: The Canadian C-Spine Rule (CCR) is now being validated for use by ED triage nurses. The objective of the study was to determine the facilitators and barriers to knowledge translation of the CCR prior to its actual implementation by nurses. Methods: We conducted a qualitative study in the EDs of 2 teaching and 4 community hospitals, where 6 nurse research coordinators gathered feedback from ED administrators, nurses, paramedics and physicians about the application of the CCR in the current validation study. The coordinators' verbal reports were recorded and transcribed. A qualitative analysis was conducted on the aspects of the rule that providers and administrators liked or not, the problems/ challenges they encountered, and suggestions for the upcoming implementation phase of the rule. Resulting themes and categories were validated by an independent rater, the site coordinators, and research team members. Results: 15 facilitators and 16 barriers were identified, and were categorized into 6 themes: 1) characteristics of the rule; 2) attitudes towards change; 3) organizational characteristics; 4) physician buy-in; 5) staff training/education; and 6) leadership/project management. Ease of use was an important facilitator in non-teaching hospitals while for those in the teaching hospitals the following were most often cited: perceptions of rule validity, value for patient care, value for nursing practice and value for improving organizational efficiency. Lack of physician buy-in was cited as a barrier in both types of hospitals, as was the impact of heavy workload. Conclusions: This knowledge transfer study assessed multiple levels (organizational, provider) and stakeholders (administrators, staff, physicians). Configurations of facilitators and barriers are different in teaching versus non-teaching hospitals. Implementation strategies will have to be tailored based on the prominent facilitators and barriers to increase knowledge translation of the CCR amongst ED nurses. Key Words: Clinical decision rule, Knowledge translation, Nurses

150 NON-ADHERENCE with emergency department discharge prescriptions

Hohl CM, Abu-Laban RB, Zed PJ, Sobolev B, Brubacher JR, Tsai G, Kretz P, Nemethy K, Bjilsma JJ, Purssell RA. Department of Emergency Medicine, Vancouver General Hospital, Vancouver, BC; Faculty of Medicine, University of British Columbia, Vancouver, BC

Introduction: Non-adherence with medication is associated with increased morbidity and health care costs. Our objectives were to determine the incidence of non-adherence to Emergency Department (ED) discharge prescriptions, and to describe associated factors. Methods: This prospective study was carried out in a tertiary care center with an annual census of 65,000. We enrolled a convenience sample of patients during shifts which mirrored our ED's discharge pattern. Research assistants (RAs) recorded information on patient demographics, socioeconomic factors, illicit drug use, family physician access and herbal remedy use. Discharge prescriptions were documented and 2 weeks later a provincial prescription dispensing database was used to determine if prescriptions had been filled. RAs made up to 5 attempts to contact patients by telephone, and during the follow-up interview asked patients to perform a pill count to assess adherence. Results: Of 1965 patients screened, 301 were discharged with a prescription and agreed to participate. Follow-up was successful in 257 patients (85.4%). The most frequent diagnoses were skin and soft tissue infections, back pain and urinary tract infection. The most common prescriptions were for acetaminophen with codeine (27.2%), ciprofloxacin (8.2%) and cephalexin (6.2%). Ninety-one patients (35.3%, 95%CI 29.7%-41.3%) were non-adherent with 1 or more medications. Regression modeling indicated a trend towards increasing non-adherence with the prescription of 2 or more medications (1.75, 95% CI 0.98-3.11) but not with socioeconomic factors. Non-adherence to medication was associated with a trend towards increased ED revisits (OR 1.6, 95% CI 0.9-2.9). Conclusions: Non-adherence with ED discharge prescriptions is common and not independently associated with socioeconomic factors. This study was not powered to evaluate any association between non-adherence and increased health services utilization, but our results suggest a trend that merits further evaluation. Key Words: Prescription, Compliance, Discharge advice

151 A prospective analysis of procedural sedation practices among academic emergency physicians

Peddle M, Lehnhardt K, McLeod S, Mosdossy G. Division of Emergency Medicine, The University of Western Ontario, London, ON

Introduction: There are a wide variety of pharmacologic agents that are used for procedural sedation in the emergency department (ED). The purpose of this study was to examine current practice for procedural sedation among emergency physicians. Methods: This prospective, observational study was conducted in two academic EDs (80,000 annual patient visits). Eligible patients undergoing procedural sedation were identified by a nurse or physician. The primary outcome was recovery time. Other outcomes included patient, nurse, and physician satisfaction, as well as complications. Vital signs were recorded prior to sedation and every 5 minutes thereafter. Upon completion of the procedure, the Aldrete post-anaesthesia score was calculated at 5-minute intervals, with full recovery defined as a cumulative score of 10. Patient, nurse and physician satisfaction was assessed using a 7-point Likert scale. Results: Fifty-six patients were enrolled; two were excluded (one critically ill, one chart missing). Forty-seven patients received propofol and 7 patients received midazolam and fentanyl. Propofol was used with adjunctive agents in 29 (61.7%) patients: 23 patients received fentanyl, 3 patients received lidocaine, and 3 patients received morphine. Mean ± standard deviation recovery time for the propofol group was 17.3 ± 8.8 minutes, compared to 27.9 ± 5.7 minutes for the midazolam group (p=0.003). Complications occurred in 7 (14.9%) patients from the propofol group and in 2 (28.6%) patients from the midazolam group (p=0.44). Complete post-procedural retrograde amnesia was present in 34 (72.3%) patients who received propofol and in 2 (28.6%) patients who received midazolam (p=0.02). There were no differences in patient, nurse or physician satisfaction between the two groups. Conclusions: In this study, propofol was more commonly used than midazolam for procedural sedation in the ED. Propofol resulted in shorter recovery times, lower complication rates and no difference in patient, nurse and physician satisfaction when compared to midazolam. Key Words: Sedation, Recovery times, Propofol, Midazolam

152 A prospective study of the provision and documentation of discharge instructions at a tertiary emergency department

Bijlsma JJ, Cheyne R, Abu-Laban RB. Department of Emergency Medicine, Vancouver General Hospital, University of British Columbia, Vancouver, BC

Introduction: The final step in emergency department (ED) care involves the provision of discharge instructions. These instructions are assumed to have a positive effect on patient outcome, but research is lacking. Our primary objective was to determine the proportion of ED patients who report receiving discharge instructions. Secondarily, we examined factors associated with the provision and documentation of discharge instructions. Methods: This prospective observational study was carried out in a tertiary ED with an annual census of 65,000. A convenience sample of patients from representative times of day, ED areas, and emergency physicians participated in a structured discharge interview. Charts of these patients were subsequently reviewed for documentation of instructions using explicit criteria. Results: 275 patients cared for by 18 emergency physicians were enrolled. The mean patient age was 48.6 years (range 19-99) and 56.2% were male. In 36.7% of cases the patient's first language was not English, and in 17.8% of cases English was not the primary language spoken at home. ED translation was required in 13 cases (4.7%). Patients reported the provision of physician discharge instructions in 234 cases (85.1%, 95%CI 80.3% to 89.1%) and nurse discharge instructions in 121 cases (44.0%, 95%CI 38.0% to 50.1%). Physician and/or nurse documentation of discharge instructions was present in 77.5% and 27.3% of cases respectively. 29 patients (10.5%) reported the absence of physician discharge instructions despite documentation to the contrary on their chart. A multivariate logistic regression model indicated physician discharge instruction provision was independent of patient age, gender, need for translation, time of day, or ED location. Conclusions: A significant proportion of patients at a tertiary ED report they do not receive discharge instructions, although in over half of such cases there is medical record documentation to the contrary. Further research on the cause and implications of this is warranted. Key Words: Documentation, Discharge advice

153 PROCEDURAL sedation and analgesia facilitator" - an expanded scope role for paramedics in the emergency department

Campbell SG, MacKinley RP, Petrie DA, Etsell G, Froese P, Warren D, Kovacs GJ, Urquhart DG, Magee KD and the Advanced Care Paramedics of the QEII. Department of Emergency Medicine, Dalhousie University and the Queen Elizabeth II Health Sciences Centre, Halifax, NS

Introduction: Procedural sedation and analgesia (PSA) are accepted as a standard of care in emergency departments (ED). PSA requires careful monitoring of a patient's cardio-respiratory status, and an ability to act immediately and appropriately in the event of any untoward event. The knowledge and skills necessary for this are a natural extension of the expertise of Advanced Care Paramedics (ACP). We report a series of PSA's conducted by ACP's over a 19 month period at a busy teaching hospital. Methods: This is a retrospective descriptive study presenting data from a registry recording details of all cases of ACP-facilitated PSA conducted in our ED between August 1, 2004 and February 28, 2006. baseline characteristics, reason for the procedure, medications used and adverse events are reported. Results: 1334 ACP-facilitated PSAs were conducted during the period. According to definitions used by this study, occurred in only 11 (0.9%) patients, and "hypotension' in 0.6% of patients. One significant adverse event was recorded, that of pulmonary aspiration. Medications used for PSA included fentanyl (94.1%of cases), propofol (65.5%), midazolam (36.7%) and ketamine (2.3%). No long term adverse events as a result of PSA recorded. Conclusions: PSA conducted in the ED by specifically trained ACPs is not associated with a significant number of adverse effects. This role should be recognized and subjected to further study. Key Words: Sedation, Analgesia, Adverse events, Propofol, Fentanyl

154 PEDIATRIC Canadian Triage and Acuity Scale as a predictor for outcome and resource utilization

Ma W, Jarvis DA, Goldman RD. Pediatric Research in Emergency Therapeutics (PRETx) Program, Division of Pediatric Emergency Medicine, The Hospital for Sick Children Department of Pediatrics, University of Toronto, Toronto, ON

Introduction: The aim of this study was to examine hospitalization rates, diagnostic resource utilization in the emergency department (ED), and ED length of stay (LOS) with respect to the Canadian Pediatric Emergency Department Triage and Acuity Scale (Ped-CTAS) in a large tertiary academic ED. Methods: All Pediatric patients (0-19 years of age) who were seen in the ED during 12 randomly selected days between May 1st 2005 and April 30th 2006 were included in the study. Information was collected to investigate the relationship between Ped-CTAS levels, hospitalization rates, ED length of stay and diagnostic test utilization. Results: 1618 (97.4%) of the 1661 patients presenting during the study period were included in the study. Hospitalization rates were 0% in the "non-urgent' category to 2.5% in "semi-urgent', 15% in "urgent' and 45% in "resuscitation-emergent'. Blood, imaging, cultures, urinalysis, and consultation rates all increased with increasing acuity. Conclusions: The Ped-CTAS level correlates well with hospitalization rates and investigational resource utilization in the ED. Hence, Ped-CTAS may serve as a tool in resource planning. Key Words: Hospitalization rates, Diagnostics, Resource utilization, Pediatric, Triage

155 IMPLEMENTATION of triage liaison physician (TLP) to alleviate ED overcrowding: evidence synthesis

Tam SL, Holroyd BR, Bullard M, Rowe BH. Department of Emergency Medicine, University of Alberta, Edmonton, AB

Introduction: Studies of the effectiveness of triage liaison physicians (TLP) and related interventions have been reported; however, there is a lack of consolidation of all available evidence. This report aims to identify and synthesize all evidence regarding the effectiveness of TLP implementation to alleviate Emergency Department (ED) overcrowding. Methods: Comprehensive searches of relevant databases such as MEDLINE, PubMed, Web of Science, and the grey literature were conducted seeking comparative clinical studies of TLP in the ED to reduce overcrowding. Search strategies were based on core keyword searches, and included searches up to year 2006. Published and unpublished studies were included and language restrictions were not applied. Results: Initial screening identified 10994 potentially relevant citations. Thirteen relevant citations were included; 3 were randomized controlled trials and 10 before- after studies, ranging from 2001 to 2006 and all in English language. These studies include one abstract, 5 studies published in peer- reviewed journals and 7 reports of Rapid Medical Evaluation available on a California emergency medicine website. Variation in reporting outcomes was impressive and prevented pooling of outcomes data. Overall, the evidence suggests that TLP implementation results in decreased patient length of stay, patients left without being seen, number of episodes of ambulance diversion, and waiting time for diagnostic procedures. Patient satisfaction was also reported to be significantly improved. There was no cost evaluation studies identified. Conclusions: Overall, the evidence regarding the role of TLPs in reducing ED overcrowding is substantial, albeit of only moderate quality and often dispersed in the grey literature. Although variations in the TLP strategies, low quality study designs and poor reporting of outcomes were identified, these results suggest hospitals should examine TLP as an option to address ED overcrowding. Key Words: Triage, Liaison physicians, Overcrowding

156 THE impact of long-term care facility bed availability on emergency department overcrowding

Fan J, Al-Darrab A, McIsaac M, Worster A, Upadhye S, Woolfrey K, Fernandes CMB. Division of Emergency Medicine, McMaster University, Hamilton, ON

Introduction: An indicator of emergency department (ED) overcrowding is the inability for admitted patients to be transferred out of the ED to an in-patient bed. A major causal factor is the lack of in-patient beds due to non-acute medical care occupancy by patients waiting for long-term care facility (LTC) placements. Methods: The objective of this study was to describe the relationship between LTC bed vacancy and ED length of stay (LOS) of admitted patients. This was a retrospective panel data analysis using a generalized least squares dummy variable model. Administrative data for a single large urban Canadian city was obtained from hospital and community care access centre (CCAC) databases from May 2003 to February 2005 for an observation period of 95 weeks. The main dependent variable was the average weekly ED LOS (hours) for admitted ED patients (an indicator of ED overcrowding). The primary explanatory variable was the total weekly number of LTC bed vacancies. All explanatory variables were considered statistically significant if p <= 0.05. Results: The regression modeling showed that the number of LTC bed vacancies was a significant predictor of ED overcrowding; a weekly increase in 1 bed was associated with a weekly decrease of 1.22 minutes (95% CI: 0.61, 1.82 minutes) in ED LOS for admitted patients. Also significant were the weekly number of medical admissions; every 1 increase in total weekly number of medical admissions, there was a weekly 7.2 minutes (95% CI: 2.2, 12.0 minutes) increase in overcrowding. The presence of overcrowding in the preceding week was also a significant covariate; each hour of overcrowding from the previous week, there was a weekly increase of 30.0 minutes (95% CI: 24.1, 36.0 minutes) in overcrowding in the subsequent week. Conclusions: There is a statistically significant relationship between community LTC bed vacancies and ED overcrowding. However, ED overcrowding in the preceding week appears to be the most important factor affecting future overcrowding. Key Words: Overcrowding, Admitted patients, Long-term care

157 A population-based study of the association between socio-economic status and emergency department utilization in Ontario, Canada

Khan Y, Schull MJ, Moineddin R, Glazier RH. Division of Emergency Medicine and Department of Public Health Sciences, University of Toronto, Toronto, ON

Introduction: The objective of this study was to investigate the association between emergency department (ED) utilization and socio-economic status (SES) in Ontario, Canada. Methods: For Ontario respondents age 20-74 years, Canadian Community Health Survey 2000-01 responses were linked to Ontario Health Insurance Plan physician utilization data for 1999-2001 and the National Ambulatory Care Reporting System for ED utilization in 2002. SES was defined according to high school completion. The primary outcome was less-urgent ED visit, corresponding to Canadian Triage and Acuity Scale 4 and 5 and not admitted to hospital. Results: Overall, 31.4% of the sample used an Ontario ED in 2002. The majority of visits (59.1%) were classified as less urgent. Fair or poor self- perceived health was the largest predictor of ED use for less-urgent and more urgent visits. Respondents with low SES were more likely to have less urgent visits (OR 1.65, 95% CI 1.35-1.94) and more urgent visits (OR 1.39, 95% CI 1.09-1.68) after controlling for age, sex, income, self-perceived health, urban or rural location, regular doctor, and non-ED physician visits. SES was not associated with having less urgent versus more urgent visits (OR 0.92, 95% CI 0.68-1.14). Conclusions: In a setting with universal health insurance, worse health status is the largest predictor of ED utilization but low SES is independently associated with increased use of the ED, regardless of triage category. Our study lends support to findings in other health systems that those using EDs are more ill and more disadvantaged, which has implications for strategic planning to address ED overcrowding. Key Words: Emergency department (ED) utilization, Socio-economic status (SES), Overcrowding

158 THE role of overcrowding on Alberta EDs: a survey of nursing directors

Rowe BH, Tam SL, Latoszek K, Holroyd BR, Bullard M, Yoon P. Department of Emergency Medicine, University of Alberta, Edmonton, AB

Introduction: A recent report documented the prevalence and severity of emergency department (ED) overcrowding at large Canadian hospitals; however, no study has examined the issue at a provincial and nursing level. This study was designed to examine the perceptions of ED Nursing Directors with respect to ED overcrowding. Methods: A 38-item, paper-based questionnaire was distributed to ED Nursing Directors at 102 hospitals with EDs or Health Care Centre with acute care services in the province of Alberta, Canada in the Fall of 2006. ED administrative data were obtained from Alberta Health and Wellness registries. Data are compared using chi-square and T-tests, where appropriate. Results: Overall, 100 Directors (98%) responded; 80.4% of whom reported that their EDs were in rural areas staffed mainly by Family Physicians (54.5%) on duty. These EDs saw a median annual volume of 13701 patients (IQR: 7000, 22000), had a median of 21 (IQR: 12, 32) acute care beds and 6 (IQR: 4, 11) standard treatment spaces. Most nurses reported that their triage area was the responsibility of non-specialized RNs (77.1%); most used CTAS (90.7%), but without decision support (86%). Most reported that their EDs were without fast track areas (80%) and did not have designated observation units (61%). Some directors (21%) reported that overcrowding was a major or severe problem during the past year, which negatively affected the quality of patient care (55%), as well as nursing recruitment (56%). Some directors commented that a lack of medical and nursing staff were major contributing factor to ED overcrowding, especially in rural areas. Conclusions: Overcrowding is a serious nursing problem in EDs across Alberta. It affects nurse recruitment, resulting in many EDs in non-urban areas poorly staffed to deal with the overcrowding problem, and leads to poor patient care. The perspective of ED nursing directors reinforces the impression of ED medical directors that ED Overcrowding is a medical crisis. Key Words: Emergency department (ED) utilization, Nurses, Overcrowding

159 CTAS reliability across provincial health regions

Grafstein E, Stenstrom R, Westman J, McLaughlin M, Worster A. Department of Emergency Medicine, Providence Health Care & St. Paul's Hospital, Vancouver, BC

Introduction: CTAS is increasingly being used to determine payment models and departmental efficiency based on time to physician. There have been many studies that have demonstrated the reliability of CTAS within an institution. There are very few studies that compare CTAS within and across health regions composed of different hospitals with different nursing education and administration. Our objective was to assess the inter-rater reliability of CTAS using scenarios and considering agreement on exact matching of triage level in nurses working in these different hospitals. Methods: Nurses enrolled in a CTAS training program were solicited to participate in this study using 50 scenarios. These scenarios were derived from a random selection of prospectively collected triage cases. These cases also reflected the anticipated ratios of CTAS 1 - 5 cases that an average emergency department would see. Results: 127 nurses from various sites within 3 provincial health authorities participated. There were 82 nurses from the Vancouver Coastal Health Authority. Baseline characteristics are included in the table below. Fourteen (11%) of nurses had received formal triage training. An unweighted kappa statistic was developed for each scenario. Exact match of the triage level occurred in 63.5% of cases and within one level in 92% of cases. The mean unweighted kappa value of all scenarios was 0.544 (95% CI: .0.457-0.631). Conclusions: The inter-rater reliability of CTAS in a broad unrelated cohort of ED triage nurses is moderate. This suggests that in order to improve triage reliability so that meaningful comparisons can occur at a regional and provincial level, training programs and/or systems that link triage to presenting complaint should be considered. Key Words: Triage, Inter-rater reliability

Table 1, Abstract 159
Baseline Characteristics	Mean (SD)
Age (years)	40.4 (8.9)
Years of nursing experience	17.1 (9.4)
Years of triage experience	7.2 (6.8)
Previous triage training	14 (11%)
Previous work at another hospital	55 (43%)