CAEP/ACMU 2001 Scientific Abstracts - Poster Presentations: 75-93

CAEP Abstracts

CJEM 2001;3(2):135-149

| 001-014 | 015-029 | 030-044 | 045-059 | 060-074 | 075-093 |

075 Information access by emergency medicine staff and residents.
Rowe BH, Cummings G, Sher A, Hayward R. University of Alberta, Edmonton, Alberta.

OBJECTIVES: Accessible to electronic evidence-based medical (EBM) information in emergency medicine (EM) has the potential to improve patient care and to facilitate educational activities. This study examines the reliability and validity as well as the characteristics of resource use recorded by a computer-based health information system. METHODS: CHE/CLINT is a single sign-on, password-protected, desktop application for medical information resource management. An EM-specific desktop was produced consisting of journals, on-line resources, and important links; access was provided to 70 staff and resident EM users at 2 major teaching hospitals affiliated with the University of Alberta. The data for the usage characteristics were derived from user logs and spanned the first 8 months this resource was available. Four investigators completed 50 test logins, recording log-in characteristics and comparing them to the user logs generated by the computer system in a local database. RESULTS: Times, date, location and non-internet applications were reliably recorded by the system during individual sessions (all >80% accurate). There is a spectrum of users, ranging from those who have never used this resource to "super-users" (<10%) who have readily adopted this resource. Most of the 925 logins to the CHE/CLINT program were of short duration (mean = 2 min/login) and accessed few applications (mean = 2 applications/login). The most highly used resources (n =1784) were literature searching applications such as MEDLINE (15%), and reference materials such as MD Consult (9%) or Harrison's Online (6%). ED-based resources (e.g., Micromedex [3%]) and EBM resources such as Cochrane Library (6%) were less often used. CONCLUSIONS: The data collected in CHE/CLINT are reliable and valid and can be used to determine user information access. EM information needs include literature searching and text; clinical applications and EBM resources remain underutilized. Interventions to improve use of EBM resources appear warranted.
Key words: medical informatics, evidence based medicine

076 Are homeless patients satisfied with care received in the emergency department?
Ilk L, Spence J. Department of Medicine, University of Toronto, Toronto, Ontario.

OBJECTIVES: Homelessness is an increasing threat to the health of urban communities and inner city emergency departments (EDs) are playing a major role in the care of this population. A survey, comparing satisfaction with care of homeless and underhoused (H-UH) to housed patients (H), was conducted in a university affiliated ED which services 40% of the lowest income census tracts and is situated in the highest concentration of homeless people in Canada. METHODS: A 12-question patient satisfaction survey was administered over a 4-week period in July and August 2000. Domains of interest included: participant's subjective sense of health, satisfaction with care, perceived quality of care, attention of medical staff, and discharge planning. A retrospective chart review was conducted to compare the H-UH to H patients with respect to demographics, waiting times, number of tests and procedures, and follow-up care. RESULTS: 305 surveys were completed, 113 by H-UH patients. There was no difference in the mean age of the groups. More H-UH patients were male. Urgency, waiting times, time to see an MD, and number of tests and procedures performed were not significantly different between groups. H-UH patients rated their usual health to be poorer and presenting complaints to be more serious. Homeless patients also anticipated greater difficulty in following discharge instructions. Both groups agreed that they would return to the ED, however H-UH patients did not rate their quality of care as highly as H patients. CONCLUSIONS: H-UH patients perceive their health to be poorer and presenting complaints to be more acute than H patients. Overall H-UH population tended to be less satisfied with ED care. Staff working in inner city EDs may require a greater awareness of both social and medical needs in order to better serve H-UH patients.
Key words: satisfaction, quality

077 Initiating the discussion of organ donation: a survey of physician attitudes and practices.
Hall C, Kjerulf M, Cass D. Department of Medicine, University of Toronto, Toronto, Ontario.

OBJECTIVES: One factor that may play a role in improving the rate of organ donation is prior communication between patients and their next-of-kin concerning their attitudes towards organ donation. One potential vehicle for the promotion of these discussions is the family physician, who could encourage patients to discuss the issue with their families. METHODS: A 35-question survey designed to assess attitudes, current practices, and barriers and opportunities regarding organ donation was mailed to each family physician and general practitioner in Toronto. Surveys could be returned either by fax, or by postage pre-paid envelopes. RESULTS: A total of 497 of 2385 surveys were completed, for a return rate of 21%. Sixty percent of respondents indicated that they felt it was appropriate for family physicians to initiate the discussion about organ donation in their office practice. However, only 40% of family physicians had raised the issue of organ or tissue donation with their patients and/or their families in the past year, and 74% indicated that they "seldom or never" introduce the topic during office visits. The 3 greatest barriers to introducing these discussions were identified as: fear of offending patients; personal lack of knowledge about the topic, and; lack of time in the patient encounter. The greatest opportunities to improve the initiation of the discussion of organ donation were identified as being the provision of pamphlets and other materials for the family physician's office and further education for family physicians about organ donation. CONCLUSIONS: While many family physicians support the concept of initiating the discussion about organ donation with patients and their families, the vast majority seldom or never do so. The provision of further educational materials, for both patients and physicians, could increase the likelihood of discussion and, therefore, of donation.
Key words: organ donation

078 Identifying potential organ donors in a university teaching hospital: How good are we?
Hall C, Kjerulf M, Cass D. Department of Medicine, University of Toronto, Toronto, Ontario.

OBJECTIVES: St. Michael's Hospital has identified as a priority the improvement of both quality and quantity in the organ donation process. To aid in quantifying the gap between potential and actual donors, a retrospective chart review was designed to identify all potential organ donors at St. Michael's Hospital in 1999. METHODS: All deaths in critical care settings over a one-year period were screened using "Donor Action" software to identify patients who were potential organ donors, using a set of clinical criteria. Once potential donors were identified, a manual chart review was performed in order to confirm eligibility of the patient, and to identify the reason(s) why donation did not occur. RESULTS: One hundred and fifty-nine charts were reviewed; 15 charts were unable to be reviewed. Fifty-one patients were found to meet the established criteria for potential organ donors. The families of 30 (59%) of these potential organ donors were approached about the possibility of donation; 17 (57%) of these consented to organ donation, of which 15 patients went on to donate organs, 1 to donate tissues only, and 1 to donate their body for research. Nineteen families (37%) were not approached, while 2 families (4%) were unavailable to either give or decline consent. Of the 34 patients who did not go on to become organ donors, over half of the families were not offered the opportunity to donate. CONCLUSIONS: When offered the opportunity to donate, over half of the families of potential donors gave consent for organ and tissue donation. A significant number of families were denied the opportunity to consider donation as part of their end-of-life decisions for their loved ones. Providing all families this opportunity could lead to an improvement in both the rate of organ donation, and in the quality of the end-of-life experience for the families.
Key words: organ donation

079 Why do patients experiencing acute chest pain bypass the nearest hospital to seek care in a tertiary care centre?
Yeung B, Friedman SM, Dubinsky I. University of Toronto, Toronto, Ontario.

OBJECTIVES: To characterize the subpopulation of patients with acute chest pain or palpitations who select a tertiary referral center over closer emergency departments (EDs) to identify patient rationale for hospital selection. METHODS: Four-week prospective cohort pilot study. Patients presenting to the ED of a downtown teaching centre (Toronto General Hospital [TGH]) with a chief complaint of chest pain or palpitations were administered a standardized survey within 72 hours of registering in the ED. The 20% who traveled the longest distance (LD) were compared to the base population (BP). Categorical variables were analyzed using the binomial inference for proportions, and continuous variables using the t-test. RESULTS: Of 106 eligible patients, 84 (79.2 %) completed the survey. The LD group, as well as study nonparticipants, were similar to BP in age and gender. Long distance travelers were more likely than the base population to be aware of a hospital closer than TGH to the point of departure ( 76.5 % vs. 46.4 %, p < 0.01), to have previously attended the closer hospital (p < 0.10), and to have attended with a complaint similar to the current problem (p < 0.08). LD patients were more likely to select TGH for subjective reasons (i.e., perceptions regarding hospital reputation or courtesy), (p < 0.10 ), and were less likely to select TGH for medical reasons (i.e., past records, physician, or procedure at TGH) (p < 0.08). Long distance travelers were less likely than BP to agree that in the event of acute chest pain or palpitations, the best choice of hospital is usually the closest one (p < 0.03) CONCLUSIONS: Patient rationale for hospital selection may be related to medical misconception and nonmedical factors, impacts upon distance traveled to seek medical attention, and has potential to adversely affect medical outcome.
Key words: chest pain

080 The incidence of vomiting in a pediatric emergency department and its influence on clinical mnagement.
Kozer E, Razavi H, Goldman R, Shi K, Keays T, Wu D, Scolnik D. Hospital for Sick Children, University of Toronto, Toronto, Ontario.

INTRODUCTION: Vomiting commonly occurs in children, however, the incidence of vomiting in the pediatric emergency department (ED) and its influence on clinical management has not been described. OBJECTIVES: to establish the incidence of vomiting among children presenting to pediatric ED, and to find whether or not vomiting is associated with increased length of stay, more laboratory tests and increased risk of admission to the hospital. DESIGN: A retrospective chart review of patients presenting to the ED of tertiary care pediatric hospital. PATIENTS: all the patients 0-18 years old presented to the ED during 4 randomly selected days. METHODS: Using the hospital electronic database, the charts of all patients who presented to the ED on 4 randomly selected days from the years of 1999-2000 were systematically reviewed by research worker. For each of the charts demographic parameters were recorded along with data regarding the history of the current illness and management in the ED. RESULTS: Of the 439 patients presenting to the ED, 97 (22.1%) reported vomiting. The average age of those who vomited was 3.50 ± 0.87 years, compared to 5.09 ± 0.47 years (p = 0.0016) in those without vomiting. Within the subgroup of 329 cases without trauma, 41 reported vomiting twice or more. In this group, 1.37 tests were ordered, compared to 0.825 tests for those vomiting less than twice (p = 0.056). In the 197 patients 2 years or younger, 1.02 tests were ordered for vomiting patients, whereas 0.50 tests were ordered for patients without vomiting (p = 0.026). There were no trends seen in time spent in ED, triage class or admission rate. CONCLUSIONS: 22% of patients presented with vomiting as a symptom. In patients 2 years of age or younger, significantly more tests were ordered for vomiting patients. A similar trend was seen in non-traumatic patients vomiting multiple times.
Key words: vomiting, pediatric

081 Can we predict which children with clinically suspected pneumonia will have the presence of focal infiltrates on chest radiographs?
Lynch T, Gouin S, Larson C, Patenaude Y. Children's Hospital of Western Ontario, London, Ontario.

OBJECTIVES: To determine predictive factors for the presence of focal infiltrates in children with clinically suspected pneumonia in a pediatric emergency department (ED). METHODS: Prospective cohort study conducted between May 1998 and December 1999 in a pediatric ED. Children (1-16 years) with clinically suspected pneumonia were enrolled. Exclusion criteriae were: chronic cardiac and respiratory disease, concurrent asthma exacerbation, pneumonia confirmed by chest radiograph (CXR) in the previous 8 weeks or antibiotic use in the previous 2 weeks, gastroesophageal reflux, spastic quadriplegia and unstable condition. Data on demographic variables, presenting symptoms and signs were collected upon study enrollment. Frontal and lateral views of the chest were obtained for each patient. A consensus agreement of 3 radiologists on the presence of focal infiltrates was taken as the gold standard. RESULTS: There were 604 eligible children: 33 families declined to participate. Of the remaining 571 patients, 204 had the presence of focal infiltrates on CXR. No differences were noted between the children with and without focal infiltrates for mean age, sex, mean weight, presence of cough, coryza, bronchial sounds, irritability, poor feeding, mean oxygen saturation and time of the year. Risk factors for the presence of focal infiltrates included: history of fever (p = 0.001, odds ratio [OR] = 3.1; 95% confidence interval [CI] = 1.7, 5.3), retractions (p = 0.047, OR = 2.8; 95% CI = 1.0, 7.6), grunting (p = 0.038, OR = 7.3; 95% CI = 1.1, 48.1), crackles (0.001, OR = 2.0; 95% CI = 1.4, 2.9), and decreased breath sounds (p = 0.034, OR = 1.4; 95% CI = 1.0, 2.0). The patients with focal infiltrates were more likely to have an increased temperature (38.2°C vs. 37.8°C, p = 0.001, DM = 0.36; 95% CI = 0.15, 0.58), an increased respiratory rate (32/min vs. 29/min, p = 0.001, DM = 3.3, 95% CI = 1.4, 5.2) and an increased heart rate (130/min vs.124/min, p = 0.003, DM = 6.8, 95% CI = 2.6, 11.1). CONCLUSIONS: Chest radiographs for the detection of focal infiltrates should be limited to children who have an history of fever, increased temperature, increased respiratory rate, increased heart rate, retractions, grunting, crackles or decreased breath sounds.
Key words: pneumonia, pediatric, radiography

082 Communicating clinical information between the emergency department and family physicians: a qualitative analysis of current practices and suggestions for improvement.
Afilalo M, Lang E, Boivin JF, Colacone A, Robitaille C, Rosenthal S, et al. McGill University, Montreal, Quebec.

OBJECTIVES: Accurate and timely bi-directional transmission of clinical information between the emergency department (ED) and family physicians (FPs) is important for rational and informed patient care. The objective of this study was to evaluate current information exchange practices and to query both emergency physicians (EPs) and FPs on the components of an ideal communication tool. METHODS: As part of a larger study aimed at designing, implementing and measuring the impact of a secured electronic communication tool, focus groups were conducted at 3 hospitals where a total of 16 EPs and 13 FPs participated. Three hypothetical cases were presented as a starting point for a series of questions and discussions. All comments were audio-taped/transcribed and physicians were asked to complete questionnaires following each session. RESULTS: EPs rarely sought information from FPs despite the perception that this would be of value. FPs at all 3 sites reported that there was currently little or no clinical information being sent to them regarding their patients' ED visits. EPs ranked the most recent EKG, a list of current medical problems and the past medical history as the most important information to be obtained from FPs. FPs ranked new or changed medications, treatment plan and blood test results as the most valuable information required from the ED. All physicians agreed that enhanced communication would reduce duplication, specialty consultations, and inappropriate ED use. FPs were however concerned about the potential for unsafe discharge from the ED and the need to take on new patients and responsibilities as a result of receiving such data. CONCLUSIONS: The lack of communication of clinical information is a major deficiency in the continuity of care provided for patients who frequent the ED. FPs and ED physicians believe that the provision of specific clinical and lab data will enhance both the quality and efficiency of patient care.
Key words: communication, quality

083 Do pediatric emergency physicians change the management of patients with suspected pneumonia after interpretation of chest radiographs by radiologists?
Lynch T, Gouin S, Larson C, Patenaude Y. Children's Hospital of Western Ontario, University of Western Ontario, London, Ontario.

OBJECTIVES: To measure the levels of confidence of pediatric emergency physicians ( PEPs) in their clinical and radiological diagnoses of pneumonia and in the radiologists' chest radiograph (CXR) interpretation. METHODS: Prospective cohort study conducted between May 1998 and December 1999. Children (1-16 years) with clinically suspected pneumonia were enrolled while 1 of the 11 recruiting PEP was attending in the emergency department. Exclusion criteriae were: chronic cardiac or respiratory disease, asthma exacerbation, immunodeficiency gastro-esophageal reflux and unstable patients. Frontal and lateral CXRs were obtained for each patient. RESULTS: 604 children were eligible; 33 families declined to participate. Of the remaining 571 patients, a 100% response rate was achieved amongst the PEPs. Prior to interpreting the CXRs, the PEPs thought that there was clinical evidence of pneumonia in 30.6% of the patients. PEPs would prescribe antibiotics regardless of their CXR interpretation in 30.1% of the patients (18.2% for suspected pneumonia and 11.9% for a different infectious focus). After reviewing the CXRs, 59.0% of patients were suspected by the PEPs to have pneumonia. If the PEPs believed that the CXRs were consistent with pneumonia, they would use antibiotics even if the radiologist's interpretation was negative 86.9% of the time. If the PEPs believed that the CXRs were negative for pneumonia, they would not order antibiotics even if the radiologist's interpretation was positive 22.6% of the time. CONCLUSIONS: The PEPs diagnosis of pneumonia nearly doubled in frequency with the addition of a CXR to the clinical exam. This study demonstrates interesting trends in the level of confidence that the PEPs have in the radiologist`s CXR interpretation.
Key words: pneumonia, pediatric, radiography

084 Can pediatric emergency physicians correctly localize pulmonary infiltrates on chest radiographs?
Lynch T, Gouin S, Larson C, Patenaude Y. Children's Hospital of Western Ontario, University of Western Ontario, London, Ontario.

OBJECTIVES: To compare the accuracy of pediatric emergency physicians (PEPs) with a radiologist in the localization of pulmonary infiltrates on chest radiographs (CXRs). METHODS: Cohort of CXRs ordered for the clinical suspicion of pneumonia in children (1-16 years) who presented to the emergency department between May 1998 and December 1999, while 1 of the 11 recruiting board eligible/certified PEP was present. Exclusion criteriae were: chronic cardiac or respiratory disease, asthma exacerbation, immunodeficiency gastro-esophageal reflux and unstable patients. Following randomization, the PEP had only access to the frontal CXR view for interpretation in half of the cases; otherwise, the frontal and lateral views were available. All the CXRs were reviewed by a single, blinded radiologist who had access to both views. RESULTS: There were 604 eligible patients; 33 families declined to participate. Of the 571 recruited patients, 332 (58.1%) had the presence of pulmonary infiltrates according to the PEPs versus 204 (35.7%) according to the radiologist (p < 0.05). The rate of agreement between the radiologist and the PEPs for each location was: right upper lobe = 52.3%, right middle lobe = 29%, right lower lobe = 18%, left upper lobe = 21%, lingula = 23% and left lower lobe = 33%. There was only a statistical difference in the comparison of the rates of agreement for 1 vs. 2 views for the lingular location (12% vs. 50%, p = 0.016). CONCLUSIONS: PEPs had a significant rate of overcall of pulmonary infiltrates compared to a radiologist. Accurate localization of pulmonary infiltrates by PEPs was sub-optimal. Lingular infiltrates appeared to be significantly overcalled if access to only the frontal view was provided.
Key words: pneumonia, pediatric, radiography

085 Regional variations in cycling helmet use in Alberta communities.
Nykolyshyn K, Belton K, Petruk J, Cheung M, Wiebe N, Rowe BH. KIDSAFE Connection, University of Alberta, Edmonton, Alberta.

OBJECTIVES: This study examined the use of helmets in adults and children in 2 major Canadian urban regions and examines regional variation of interest for injury prevention planners. METHODS: A prospective survey of cyclists was performed between 05-09/2000 in 2 major urban areas in Alberta, Canada. Trained research assistants recorded cyclist demographics and helmet wearing patterns in 2 urban and 8 non-urban areas. A random selection of roadways, commuter paths, bicycle paths, parks, schools and campuses were sampled. Schools were over-sampled in order to increase the target population (children) data. Rates are reported with 95% confidence intervals (95% confidence interval [CI]). All variables are compared using univariate and adjusted analyses. RESULTS: From 4,157 cyclist encounters, 4,141 valid helmet observations were made. Overall, 2,259 (55%; 53-56) cyclists were wearing a helmet; however, only 75% (95% CI: 73-77) were wearing their helmet correctly. Patterns of use varied according to age: 74% (95% CI: 72-77) of children, 29% (95% CI: 25-32) of adolescents, and 52% (95% CI: 50-54) of adults were wearing helmets. In addition, location (Calgary 63% vs. Edmonton 45%; p < 0.001) and gender (girls vs. boys: 64% vs. 50%; p < 0.001) proportions differed. Location of observation (urban vs. non-urban) was not associated with helmet wearing, although non-urban rates were similar to those exhibited by the closest major center. Overall, models predicting helmet use varied based on age grouping. CONCLUSIONS: These results identify large within and between region variation in the use of cycling helmets. Without the successful implementation of mandatory cycling helmet legislation, injury prevention planners need to use these data to implement interventions that are focussed on age groupings, gender and place of residence. Further work is required to understand these variations and design implementation strategies to correct these disparities.
Key words: brain injury, injury preventation

086 Frequent fliers in the emergency department.
Ovens HJ, Chan BTB. Mount Sinai Hospital, Institute for Clinical Evaluative Sciences, Toronto, Ontario.

OBJECTIVES: Heavy users of emergency services (often called "frequent fliers") have been described for at least 25 years. However previous studies were confined to a single institution. Access to a population database allowed us to identify the characteristics of heavy users of emergency department (ED) services, describe their migration patterns and use of services outside the ED across a large number of institutions. METHODS: This was a population-based, observational, cross-sectional study, set in the province of Ontario, Canada, in 1997/98, using the provincial billing database. Frequent fliers, defined as individuals with at least 12 ED physician assessments per year, were compared to those with 1 to 11 per year. The main outcome measures were number of ED assessments and institutions visited; diagnosis at ED visit; and visits to physicians outside the ED. RESULTS: Frequent fliers constituted 0.3% of the patients visiting EDs but accounted for 3.5% of total ED visit volume. They were present in academic, urban and rural EDs. Consistent with past research, they tended to be adults aged 25-64 living in low socioecomic neighbourhoods with a high prevalence of psychosocial conditions. Migraine headache was their most prominent organic condition (9.6% of ED visits vs. 1.0%; p < 0.001). Most frequent fliers made the majority of their visits (83%) to the same institution. Frequent fliers had more visits to office-based GPs (19.2 vs. 5.5) and more specialist referrals (4.0 vs. 1.0). CONCLUSIONS: Frequent fliers are ubiquitous, and all EDs are faced with the challenges they present. Their identification can be readily accomplished by individual institutions, and data-sharing between EDs would have limited benefit. Because these patients are heavy users of office-based services, a strategy to manage their care better will require coordination between EDs, GPs and other health providers. Migraine is a previously unrecognized cause of heavy ED use which deserves attention. Research supported by the physicians of Ontario through a Physician Services Incorporated Foundation Grant.
Key words: emergency department, utilization

087 Candidate perspectives on emergency medicine residency programs: using survey feedback to respond to candidate perceptions.
Rosenblum RM, Lund AJ, Spooner CA, Rowe BH. University of Alberta, Edmonton, Alberta.

OBJECTIVES: Graduating MDs face residency decisions sooner than in previous years. This poses challenges to both candidates and resident selection committees. Candidates with less clinical exposure to various specialties must rely more heavily on information gained during the pre-Match "courtship." Our goals were to understand how candidates make their decisions and how they perceived the University of Alberta Emergency Medicine program, to make appropriate changes in our own processes, and to test the impact of those changes. METHODS: The study period was 2 years. All applicants interviewed in 01/1999 (n = 28) and 01/2000 (n = 25) were surveyed by mail, using a standardized form, after the interview but prior to the final national residency match. Based on initial feedback, between 1999 and 2000, the interview process was changed in several ways, including a series of less formal smaller panels, encouraging questions and increased contact between candidates and current residents. RESULTS: We obtained response rates of 46% (13) and 52% (13) respectively. Candidates judge programs by criteria such as provision of information, resident accessibility, and facilities. They also rely heavily on the interview itself. All programs rated well in some areas and poorly in others. Our program rated poorly in 2 aspects of the interview: the interview format and the chance to ask questions. After interview restructuring, the proportion of candidates who rated the University of Alberta interview highest rose from 7% in 1999 to 67% in 2000 (p = 0.004). CONCLUSIONS: Many candidates will share their perspectives on EM program selection processes when asked. Program changes can impact strongly on candidate experience and choices. Obtaining candidate feedback can help programs critically assess the strengths and weaknesses in their selection processes. Programs can then make appropriate formatting changes.
Key words: postgraduate training

088 Wild in the city: piloting a wilderness medicine course for medical students.
Denny CJ, Cheng IS, Schull MJ. University of Toronto, Toronto, Ontario.

OBJECTIVES: Wilderness medicine is a crucial component of emergency medicine. However, wilderness medicine topics such as prehospital care and environmental emergencies are often not included in medical school core curricula. Our objectives were to pilot a short wilderness medicine course for medical students, to measure student satisfaction, and to estimate the course's effectiveness at improving knowledge. METHODS: This prospective observational study was performed at a university medical school. Medical students in all years were invited to participate. The 2-day course was held over a weekend in October 2000. It combined didactic classroom presentations by emergency medicine residents and attending physicians with interactive outdoor simulations. Participants were asked to complete evaluations of the course. Furthermore, pre and post-course multiple-choice question testing was conducted based upon exams generated by course faculty. All evaluations and tests were optically scanned. The paired-sample t-test and 95% confidence intervals (CIs) were used to analyse results. RESULTS: Twenty medical students (years 1 through 4) volunteered to participate in the course. Of the fifteen students who completed evaluations, all fifteen would recommend the course to their colleagues. Moreover, all would have the course offered again next year. The most frequent suggestion (n = 8) was to increase the proportion of interactive learning in the curriculum.Of ten medical students who consented to pre and post-course testing, pre-course test mean scores were 51.9% (standard deviation [SD] 11.0); post-course test mean scores were 67.1% (SD 10.2). Post-course test results were significantly improved (p < 0.006), with a mean improvement of 16.7% (95% CI 6%-27%). CONCLUSIONS: This is the first study to measure outcomes in a wilderness medicine course for medical students. Our results suggest that medical students respond positively to such an intervention. More importantly, our results demonstrate that medical student knowledge improves with this short, focused intervention.
Key words: wilderness medicine, medical education

089 Enrolment in research studies in emergency medicine.
Scheibel N, Spooner CH, Sukhrani N, Cunningham R, Kelly KD, Holroyd BR, et al. Mayo Clinic, Rochester, Minnesota.

OBJECTIVES: Enrolling patients presenting with common emergency department (ED) problems into prospective cohort studies is an important contribution to clinical research. However, compliance with enrollment is often poor, even in academic centres. This study examines enrollment in a deep vein thrombosis (DVT) study. METHODS: In a prospective fashion, we monitored the use of a DVT clinical model (CM) in patients >18 years who presented to EDs with suspected acute DVT. Physicians were asked to complete the single-page CM sheet and request consent for 3-month telephone follow-up. Consent to conduct the research was agreed to by 3 North American EDs with academic interests. Physicians were made aware of the study through advertising, peer contact and nursing input. To facilitate enrollments, nurses placed forms on the ED charts and incentives were established. This analysis examines the uptake of the clinical model. RESULTS: Overall 1,067 patients presented to the 3 EDs with a diagnosis of rule-out DVT; the CM was completed in 489 patients (46%). Enrollment did not depend on age (57 vs. 59, p > 0.01) or gender (male 41% vs. 46%, p > 0.01), time (daytime vs. after hours: 49% vs. 53%, p > 0.05) or day of presentation (weekday vs. weekend: 46% vs. 42%, p = 0.26). Factors significantly influencing enrollment were hospital location (site 1: 60%, site 2: 33%, site 3: 43%, p = < 0.001) and individual physician (range of enrollment: 0% to 100%). Follow-up calls were successful in 85% of enrolled cases. DVT was confirmed in the ED in 83 (17%) patients and in a further 13 (3%) patients at follow-up. CONCLUSIONS: Even in academically interested EDs, the enrollment of patients in prospective clinical studies is less than desirable. Enrollment appears most closely related to institutional and physician attitudes. Novel approaches to enhancing enrollments and improving efficiencies in clinical research are required in order to improve the validity of ED research.
Key words: research methodology

090 Medical coverage of Tallships 2000 Festival of Sail, Halifax, Nova Scotia.
Thompson J, Petrie D, Cain E. Dalhousie University, Halifax, Nova Scotia.

OBJECTIVES: Mass gatherings (gatherings of more than 25,000 people) present unique challenges to those responsible for providing medical coverage such as identifying who actually requires help, and access to, and extrication of sick individuals. Despite the relative frequency of, and challenges presented by mass gatherings, little analytical literature exists to guide medical preparation for these events. Instead, the literature discusses different approaches that have been used in providing medical coverage and trends that have emerged. The following is a description of a mass gathering, Tallships 2000 Festival of Sail, Halifax, Nova Scotia. METHODS: Participating Emergency Health Services Nova Scotia (EHSNS) paramedics were briefed on the use of the Canadian Emergency Department Triage and Acuity Scale (CTAS) and applied these criteria to patients encountered during the event. Data on each patient encounter (demographic, chief complaint, CTAS, and disposition) was collected prospectively with CTAS 4 and 5 patients being treated and released by paramedics or advised to transport themselves to the emergency department, and CTAS 1-3 patients being transported by ambulance to the ED for evaluation. RESULTS: Approximately 1 million people participated in the Tallships Festival over 5 days. Event medical coverage was provided by EHSNS and St. John's Ambulance 18 hours/day. St. John's Ambulance staff collected no data. EHSNS personnel included roving paramedics on bicycles, a paramedic manned tent, and several ambulances dedicated to the event during peak hours. There were 3 patient encounters and 0.36 transports by ambulance per 10,000 spectators. All but 1 of the patients transported to hospital from the event were sent home after ED evaluation. CONCLUSIONS: Paramedics encountered fewer patients at the Tallships Festival than typically reported at similar size events elsewhere (typically 5-20/10,000 spectators), although transport incidence was similar.
Key words: mass gatherings, emergency medical services

091 Cardiopulmonary resuscitation: predicting outcome in Winnipeg using the Prognosis After Resuscitation (PAR) Score.
Young J. University of Manitoba, Winnipeg, Manitoba.

OBJECTIVES: The success of cardiopulmonary resuscitation (CPR) is determined by many factors not the least of which is the patient's baseline health. Many studies have reported different scoring systems to aid with the prognosis of CPR. The aim of this study is to assess the reliability of the Prognosis After Resuscitation (PAR) Score when attempting to predict CPR outcome in Winnipeg. METHODS: A retrospective review of all adult cardiac arrest calls was conducted for the period January 1 to December 31, 1999, at the Health Sciences Centre (HSC), and for the period January 1 to December 31, 1998, at the St. Boniface General Hospital (SBGH), both in Winnipeg, Manitoba. All resuscitation records and patients' charts were reviewed by a single observer to determine each patient's PAR Score. PAR Score variables were defined prior to the review and included age, presence or absence of sepsis, pneumonia, cancer, renal insufficiency, (creatinine >130 mmol/L), myocardial infarction as the presenting diagnosis, and homeboundness. Dichotomous data was analysed with Chi-squared analysis and the t-test was used to compare means of continuous data. RESULTS: There were 107 patients from HSC, and 47 patients from SBGH included in this review. At both sites, there was no difference between survivors and non-survivors with respect to age or gender. Baseline illness heavily influenced outcome, such that no patient at either site with a PAR Score greater than 8 survived. (Score ranged from -2 to +31.) Regardless of PAR Score, no patient with cancer survived a cardiac arrest at either hospital. CONCLUSIONS: PAR Scores to determine successful outcome, correlated closely with results of other studies. If these scores can be fine tuned with greater specificity, then attempts to resuscitate patients with poor prognostic indicators may be avoided. This information provides a useful framework when discussing resuscitation options with patients.
Key words: resuscitation, cardiac arrest

092 Barriers to effective feedback of trainees in the emergency department.
Ovens HJ, Jowatt J. Mount Sinai Hospital, Toronto, Ontario.

OBJECTIVES: "Inadequate feedback" is a common complaint from residents on emergency department (ED) rotations. Residents were surveyed on their opinions of feedback, their personal experiences and suggestions for improvement. METHODS: A survey was developed through meetings with an educator, literature review, and a focus group with residents. The study population comprised Mt. Sinai PGY1s between July 97 and October 98 and all PGY2s in the University of Toronto Family Medicine in 98/99. Two mailings were done between December 1998 and April 1999. Quantitative replies were entered and analyzed using chi-square analysis in SAS. Qualitative replies were organized by 1 author into themes and independently analyzed by 3 reviewers. RESULTS: The response rate was 44.7% (59/132). Fifteen were from Mt. Sinai residents; 44 from other sites. There were no significant differences by site, gender or year of training. Residents expressed a strong preference for immediate feedback by explicit case review and end-of-shift review as opposed to end of rotation. They felt feedback was impeded because "the ED was too busy" (43%) or "ED staff did not volunteer feedback" (37%). Only 12% felt their ED feedback was better than on other rotations; 31% felt it was worse. 54% felt the process of feedback was different than on other rotations. CONCLUSIONS: Residents wish immediate comments on their case management; they perceive they get more and better feedback on other rotations. ED staff require training in the provision of feedback and strategies to provide it in the busy ED environment.
Key words: medical education, postgraduate training

093 Improving access to acute stroke care in Southeastern Ontario.
Reed A, Jones G, Groll D, Bolton C. Queen's University, Kingston, Ontario.

OBJECTIVES: Southeastern Ontario (SEO) has a population base of 475,000, encompasses an area of 20,000 square kilometers and has 8 hospital corporations and 14 ambulance services. A Regional Acute Stroke Protocol (RASP) for SEO utilizing Kingston General Hospital (KGH) as the Regional Stroke Centre was implemented in 1999. The RASP includes: protocols for paramedics to identify patients who meet criteria for thrombolytic therapy, including hospital by-pass protocols; the formation of a "stroke team"; and arrangements for regional hospitals to emergently transfer patients directly to the stroke team. This is an evaluation of the activity during the first year of the RASP. METHODS: A retrospective chart review was undertaken for all RASP activations during the first 12 months since implementation. RESULTS: Since July 1999, the RASP was activated 191 times. 166 (87%) of patients were diagnosed with a cerebrovascular event, 93 of which were ischemic strokes. 68 (36%) patients bypassed another closer hospital, 35 (18%) were emergently transferred from a peripheral hospital, and 88 (46%) were from the direct KGH catchment area. 42 (22%) of all RASP patients received rtPA (45% of all ischemic strokes). CONCLUSIONS: Thrombolytic therapy can be provided to patients with acute stroke over a large geographic area. Further prospective evaluation is planned.
Key words: stroke, therapy

| 001-014 | 015-029 | 030-044 | 045-059 | 060-074 | 075-093 |