2007 CAEP/ACMU Scientific Abstracts - Poster Presentations: 100-130

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

100 WHICH patients with community-acquired pneumonia are likely to have positive blood cultures? - A case control study of blood cultures in community-acquired pneumonia.

Campbell SG, McIvor RA, Joannis V, Andreou P, Urquhart DG. Department of Emergency Medicine, Dalhousie University, Halifax, NS

Introduction: Blood cultures (BC) are commonly ordered as part of the initial assessment of patients admitted with community-acquired pneumonia (CAP), yet their yield remains low. Selective use of BC would allow the opportunity to save healthcare resources and avoid patient discomfort. Objectives: To determine what demographic and clinical factors predict a greater likelihood of a positive BC result in patients admitted to hospital after being diagnosed with CAP in the emergency department (ED). Methods: A structured retrospective systematic chart audit, comparing relevant demographic and clinical details of 83 patients admitted with CAP, in whom BC results were positive, with 169 gender and date-matched controls of patients in whom blood culture results were negative. Results: On univariate analysis, eight variables were associated with a positive BC result. These were: WBC<4.5 x 10⁹/L, (LR8.96, 95% CI=2.87-27.98), anion gap>16 (LR:4.27, 95% CI=1.04-17.52), systolic BP <90 mmHg (LR:3.87, CI=1.46-10.25), creatinine >106 umol/L(LR:2.87, CI=1.65-5.00), urea>9.3 mmol/L (LR:2.49, CI=1.4-4.42), glucose <6.1 mmol/L (LR:2.18, CI=1.09-4.35), No antibiotic therapy on presentation (LR:2.01, CI=1.02-3.98) and temperature > 38°C (LR:2.00, CI=1.16-3.46). After logistic regression analysis the variables remaining significantly associated with positive BC were: WBC<4.5 x 10⁹/L, (LR):7.75, CI=2.89-30.39), creatinine >106 umol/L (LR:3.15, CI= 1.71-5.80) glucose <6.1 mmol/L (LR:2.46, CI=1.14-5.32), and temperature >38°C (LR:2.25, 95% CI=1.21-4.20). A patient with all four of these variables had a LR of having a positive BC of 135.53, (95% CI=25.28-726.68) compared to patients with none of these variables. Conclusions: Certain clinical variables in patients admitted to hospital with CAP do appear to be associated with a higher probability of a positive yield of BC, with combinations of these variables increasing this likelihood. We have identified subgroups of patients in whom blood cultures are more likely to be useful in the treatment of CAP. Key Words: Infectious disease, Blood culture, Pneumonia

101 METHICILLIN-resistant S. aureus associated skin and soft tissue infections in a non-urban Canadian emergency department

Wiebe KP. Department of Emergency Medicine, Chilliwack General Hospital, Chilliwack, BC

Introduction: Community Acquired Methicillin Resistant Staphylococcus Aureus (CA-MRSA) has recently been recognized as the predominant pathogen in acute purulent skin and soft-tissue infections (SSTI) in urban emergency departments (EDs) in the United States. More recently, CA-MRSA has received attention as a common cause of SSTI in urban Canadian EDs. Very little is known about the prevalence of CA-MRSA as a cause of SSTI in non-urban EDs. This study sought to estimate the prevalence of CA-MRSA as a cause of SSTI in one non-urban ED in southern BC. Methods: Patients presenting with purulent skin and soft-tissue infections to a community emergency department in southern BC (catchment area 100 000, ED volume 37 000 visits/yr, distance to nearest urban center approx 100km) between the months of September and December of 2006 were enrolled. For each patient, purulent material was cultured, and clinical information was obtained. Antimicrobial susceptibility testing was performed on all S. aureus isolates. Results: MRSA was isolated from 14/19 infection sites (74%). Mixed S. aureus / streptococcal species were isolated from 4/19 sites (21%). The proportion of all S. aureus isolates that were MRSA was 14/18 (78%). All MRSA isolates were susceptible to trimethoprim/sulfamethoxazole, rifampin, vancomycin, and doxycycline. Only 10/14 MRSA isolates were susceptible to clindamycin (71%). Conclusions: MRSA is the most common pathogen cultured from patients with SSTI at this community ED. Physicians practicing emergency medicine in non-urban Canadian EDs should consider culturing all purulent SSTI, and should consider modifying empirical antimicrobial therapy to cover CA-MRSA in all life or limb threatening SSTI. Further studies are needed to determine the prevalence of CA-MRSA in other emergency departments and regions across the country. Key Words: Infectious disease, Infection control, MRSA

102 Psychiatric morbidity is not increased in hospital workers one to two years after SARS

Borgundvaag B, Lancee W, Maunder R, Balderson K, Bennett J, Evans S, Fernandes C, Goldbloom D, Gupta M, Hunter J, McGillis L, Nagle L, Pain C, Peczeniuk S, Raymond G, Read N, Rourke S, Steinberg R, Stewart T, VanDeVelde S, Veldhorst G, Wasylenki D. Division of Emergency Medicine, Faculties of Medicine and Nursing, University of Toronto and McMaster University, Toronto, ON

Introduction: We investigated if working in SARS-affected hospitals (Toronto) was associated with increased incidence of psychiatric diagnoses including posttraumatic stress disorder (PTSD), in the one to two years following the Toronto outbreak compared to working in similar local hospitals not treating SARS (Hamilton). Methods: Health care workers (HCW's, RN's in medical/surgical inpatient units, and all staff of ICU's, ED's and SARS isolation units) at 13 academic and community hospitals in Toronto (nine) and Hamilton (four), were surveyed between Oct 23, 2004 and Sept 30, 2005. Participants completed a survey and two structured interviews. Axis 1 diagnoses were determined using the Structured Clinical Interview for DSM-IV (SCID), and categorized as symptoms preceding or following SARS, and/or still present within the past month. PTSD was diagnosed with the Clinician-Administered PTSD Scale (CAPS) for: (1) the most severe past trauma of any kind (including SARS), and (2) for the most recent trauma if different from 1. Symptoms were placed in two time frames (a) ever since the trauma and (b), in the past month. Statistical analysis included T-test's and Chi-squared tests as appropriate. Results: A total of 179 HCW's were interviewed between 13 and 25 months (median 19) after the last SARS patient was discharged from hospital. Participants in Toronto experienced more contact with SARS patients (75%) and quarantine (65%) than HCW's Hamilton (0%, 2.5%, p<0.001). The incidence of new onset major depressive or panic disorder and substance abuse/dependence were similar in both cities, and to low for statistical comparison. Only one patient met criteria for current PTSD with SARS as the traumatic event (Toronto). The lifetime prevalence of psychiatric diagnoses among HCW's who participated in the study was similar to that in the Canadian community. Conclusions: The results of this study suggest that working in a SARS affected hospital did not increase the risk of psychiatric illness. Key Words: SARS, Infectious disease, Psychiatric impact of disease

103 CLINICAL cure rates associated with appropriate and inappropriate antimicrobial therapy for methicillin resistant Staphylococcus aureus skin and soft tissue infections in the emergency department

Stenstrom R, Grafstein E, Harris DR, Innes G, Hunte G, Romney M. Department of Emergency Medicine, St. Paul's Hospital, and University of British Columbia, Vancouver, BC

Introduction: The incidence of methicillin resistant staphylococcus aureus (MRSA) skin and soft tissue infection (SSTI) is increasing rapidly in emergency departments. MRSA is resistant to commonly used first line antimicrobials used to treat SSTI. Objective: To compare the proportion of patients with MRSA SSTI treated successfully with appropriate or inappropriate antimicrobials. Methods: This was a retrospective emergency department (ED) administrative database study done in a urban tertiary care ED. Between January 2003 and December 2004, ED patients with a diagnosis of SSTI (ICD-9 682.9) and a wound culture positive for MRSA, and treated initially with outpatient intravenous (IV) therapy were included. Patients were defined as treatment "clinical cures" (did not return to the ED within 4 weeks with a new SSTI; AND were not admitted to hospital, AND did not have the IV antimicrobial therapy changed) or "clinical failure" (any of the preceding conditions met). Antimicrobials used for IV therapy prescribed were dichotomized as "appropriate" or "inappropriate" based on microbiologic testing. Results: Over the 24 month period 982 patients with MRSA SSTI were treated with outpatient IV therapy. 456/984 (46.3%) of MRSA SSTI patients were treated with "appropriate" antimicrobials (vancomycin or clindamycin), and 526/982 (53.7%) were treated with "inappropriate" antibiotics (cephazolin, ciprofloxacin, cloxacillin, or ceftriaxone). Overall clinical cure rate was 730/984 (74.1%). Clinical cure rate was 350/458 (76.4%) for patients receiving "appropriate" antimicrobials versus 370/526 (70.3%) for patients receiving "inappropriate" antimicrobials (chi-square = 2.4; P = 0.11). Conclusions: Patients receiving inappropriate antimicrobials for MRSA SSTI have similar clinical cure rate as patients receiving appropriate antimicrobials. Key Words: MRSA, Infectious disease, Antibiotic treatment

104 ERRORS, near misses, and adverse events in the emergency department: what can patients tell us?

Moore S, Provan D, Friedman SM, Hanneman K. Division of Emergency Medicine, University of Toronto, Toronto, ON

Introduction: To determine whether patients or their families can identify adverse events in the ED, to characterize patient reports of errors, and to compare patient reports to events recorded by health care providers. Methods: Prospective cohort study in downtown teaching hospital. ED patients were recruited for participation in a standardized interview within 24 hours of ED discharge, and a follow-up interview at 3-7 days following discharge. Responses regarding events were characterized and compared with physician and nurse notations in the medical record and institutional error-reporting database. Results: Of 292 eligible patients, 201 (69%) were interviewed within 24 hours of ED discharge, and 143 (71% of interviewed) underwent a follow-up interview at 3-7 days post discharge. Interviewees did not differ from the base ED population in terms of age, gender, and language. Patients identified 10 adverse events (5% incident rate), 8 near misses (4%) and 0 medical errors. 6/10 (60%) of adverse events were characterized as preventable (Two raters, kappa +/- SE is 0.7826 +/- 0.2013 (95% CI: 0.3881, 1.0000), p=0.01). Adverse events were primarily related to delayed or inadequate analgesia, and 70% were primarily related to nursing care. Only 4 /8 (50%) near misses were intercepted by hospital personnel. The secondary interview elicited 2/10 adverse events and 3/8 near misses that were not identified in the primary interview. No designation (0/10) of an adverse event was recorded in the ED medical record or in the confidential hospital event reporting system. Conclusions: ED patients can identify adverse events affecting their care. Many adverse events are not recorded in the medical record. Engaging patients and their family members in identification of errors may enhance patient safety. Key Words: Medical error, Near misses

105 Do ED patients need n-acetylcysteine before receiving contrast media?

Gray SH. Yale School of Public Health, New Haven, CT

Introduction: Contrast-induced nephropathy (CIN) is a complication of contrast media administration, which is associated with significant morbidity. The role of N-acetylcysteine (NAC) as prophylaxis for CIN has been investigated in many controlled trials, however uncertainty persists over its effectiveness. Furthermore, the role of this medication in the emergency department is unclear. This systematic review examines the benefits or harms of NAC prophylaxis for contrast-induced nephropathy among patients with chronic renal failure. Methods: The Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 4, 2006), MEDLINE (January 1966 to December 2006), EMBASE (1980 to October 2006) and CINAHL (1982 to Oct Week 1 2006) were searched for randomized trials which examine the effect of NAC versus placebo or hydration, on the development of CIN in adults with chronic renal failure. Clinical outcomes and quality data were extracted from the trials, using pre-specified criteria. Results: Twenty-three trials involving 2955 participants were included. The NAC group had significantly lower rates of contrast-induced nephropathy, compared to the controls (fixed RR 0.61, 95% CI 0.49, 0.77). 17 patients must be treated with NAC to avoid one case of CIN (RD -0.06, 95%CI -0.08, -0.03, NNT 17). This outcome included significant statistical heterogeneity (X2 = 35.50, df = 19 (p=0.01), I2 = 46.5%) which may be partly explained by differences among the studies with respect to the severity of baseline renal dysfunction, the dose of NAC administered, and blinding. Conclusions: This review demonstrated a beneficial effect of NAC in reducing contrast-induced nephropathy, although there was significant heterogeneity among the included studies. These results must be interpreted within the context of the broader literature, which remains controversial, particularly in the context of the emergency department. Considering the overall risk-benefit ratio, this analysis cautiously supports the use of NAC in appropriate patients. Key Words: Systematic review, N-acetlcystine, Contrast induced renal failure

106 INTIMATE partner violence: development of an online education course for emergency department workers

Snider C, Schwartz B, Mason R. Division of Emergency Medicine, University of Toronto, Toronto, ON

Introduction: Intimate Partner Violence (IPV) is a common and continued problem in Canada. According to Statistics Canada, 7% of women reported abuse in the last five years. In Canada, 13% of women injured by their partners require medical attention. They commonly present to emergency (ED), yet those who work in emergency departments continue to express a lack of comfort in how to approach the issue of IPV in the ED. Methods: In September 2005, the Ontario Women's Directorate (OWD) commissioned an expert panel to develop a training program for ED personnel. This panel is comprised of ED physicians, nurses, social workers, assault/crisis-workers and primary researchers. Results: Through group meetings and an extensive literature review, core competencies were developed and presented to stakeholder groups. Our expert panel then worked with educators and software developers to build an innovative, scenario-based, self-directed, e-learning curriculum to educate and train ED personnel on the issues and clinical care of women injured due to IPV. Twelve sessions, each comprising three sections; an upfront interactive text based learning section, an animated scenario where learning is applied, and a final quiz, were developed. Conclusions: This presentation will describe the process of the curriculum development, implementation plans and demonstrate one of the completed modules. Key Words: Spousal abuse, Intimate partner violence, Medical education

107 DANGER in the shopping malls: epidemiology of escalator related injuries in children in Singapore

Tyebally A, Ang A, Ng KC, Anantharam V, Goh SH, Tan NC, Tan C, Heng D, Wee KP. Department of Paediatric Emergency Medicine, KK Women's and Children's Hospital, Singapore

Introduction: To study the epidemiology of escalator related injuries in children in Singapore to determine how we can educate the public on escalator safety and whether enhancements in the safety features of escalators are necessary to prevent such injuries Methods: 45 children, aged 0-16 years, who attended the Emergency Departments in the Singapore Health Serivces network for escalator related injuries from Feb 2002 to Jan 2004 were surveyed as part of the Childhood Injury Surveillance Project. Data on demographics, place of injury, host factors, environmental factors and injury particulars were collected via the use of questionnaire forms, review of in-patient records and coroner's reports. Data was recorded using the International Classification of External Causes of Injuries Codes. Results: There were 45 escalator related injuries in Singapore during the study period with no deaths. 55.6% of the patients were female and the mean age of the victims was 6.5 years. 60% of the time at least one parent was looking after the child during the incident. 13.3% of the children required hospital admission. Most of the injuries occurred to the upper limb (31.3%), lower limb (29.7%) and head (20.3%). 56.9% of the injuries were caused by falls on or off the escalator and 33.3% were crush injuries that resulted from body parts getting caught in between grooves on the escalator. 34.5% of the children suffered from open wounds and 12.7% had fractures or dislocations. Conclusions: Using escalators has become part of the daily lives of most children and the injuries caused by escalators are small but significant. Escalator injuries are preventable and we need to embark on a public education programme to teach parents basic escalator safety. Engineers and technicians also need to ensure that escalators installed are well maintained and meet basic safety requirements. Key Words: Injury prevention, Escalator related injuries, Pediatric trauma

108 EM training for off-service residents: a systematic needs assessment to guide curriculum design

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

Introduction: Non-EM residents may find the EM rotation stressful and variably useful. We evaluated the perceived effectiveness of the EM rotation for off-service residents and conducted a needs assessment for a better EM curriculum. Methods: This cross-sectional web survey used Dillman methods and was conducted at 2 university affiliated hospitals with a complete spectrum of 18 residency training programs. Surveyed were all off-service residents, who rotated through the EDs over an 11-month period, and their program directors (PD). We asked about perceptions of shiftwork, aspects of the EM rotation, and priority curricular content, using Likert scale responses. Gap scores were calculated for priority content. Results: All 18 PDs (100%) and 86 residents (77.5%) responded. Residents were PGY1 (84.8%), female (56.5%), aged 24-29 (67.1%), family medicine (46.7%), surgical (16.7%), and medical (16.7%). PDs and residents rated the EM rotation similar or better than other rotations (89.5% vs 88.3%). On a series of 5-point scales, both PDs and residents were satisfied with the EM rotation including educational relevance (3.5 vs 3.6), number of shifts (3.7 vs 3.5), meeting academic commitments (3.2 vs 3.2), case variety (3.7 vs 3.9), ACLS learning (3.8 vs 3.7), and tolerance of shiftwork (3.5 vs 3.2) (all p>0.05). PDs and residents differed over ideal shift length (3.9 vs 2.3, p<0.05), bedside teaching (3.6 vs 3.1, p<0.05), exposure to trauma (3.2 vs 2.6, p>0.05) and resuscitation (3.4 vs 2.6, p<0.05). Cardiac resuscitation, central line insertion, cardioversion, and intubation were of greatest interest for enhancement. Case variety, teaching quality, hands-on role, and procedures were strengths; shorter shifts, more procedures and resuscitation exposure were recommended. Conclusions: Off-service residents and their PDs are satisfied with the EM rotation but have suggestions to enhance training. These results can be used to guide the creation of a systematic EM curriculum for these residents. Key Words: Medical education, Residency training, EM curriculum

109 A needs assessment to define the essential administrative competencies for EM training

Alrajhi A, Frank JR, Lee AC, Alamry A, Pitters C. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: Administrative aspects of Emergency Medicine are recognized as an increasing priority for training by educators, department chiefs, and certification bodies such as the Royal College. We set out to conduct the first needs assessment of essential administrative competencies for EM training in Canada. Methods: We conducted a cross-sectional mail survey sent to 18 Ontario academic ED chiefs, 5 FRCPC program directors, and 42 PGY4 and 5 FRCPC EM residents, using Dillman's methods. 44 content domains were derived from the curriculum in Salluzzo's text (Emergency Department Management 1996). Participants (n=65) were asked to rate the relevance of each content domain for typical emergency physicians as compared to the relevance for ED chiefs. Results were analyzed using paired t-tests comparing the perceived differences in the importance of each of these domains to physicians and to ED chiefs using two 5 point Likert scales. Results: The overall response rate was 42/65 (64.6%). Responses revealed 20 domains that were not identified to be of significantly different in terms of importance for both physicians and chiefs (p>0.05). Of the other domains where p<0.05, 18 domains favoured incorporation of the administrative content domains into general EM residency training, of which the highest scores were for ED flow (residency mean = 3.8/4, administrator mean = 2.4/4), disposition (3.7, 2.4), death in the ED(3.7, 2.3), and EMS(3.4,2.3). 6 content areas were deemed essential only for chiefs or those interested in senior administrative roles: business models and budgeting(2.2,3.2), meetings and committees(2.3,3.2), strategy and planning(2.3,3.1), ED facility design(2.4,3.1), material management(2.3,3.0), and facility accreditation(2.4,3.0). Conclusions: This study is the first to conduct a systematic needs assessment in administrative competencies for Canadian EM training. The results can provide an initial framework to guide the enhancement of both resident and administrator education. Key Words: Medical education, Residency training, EM curriculum

110 CREATING a reliable and valid blueprint for emergency medicine clerkship curriculum design

Langhan TS, Donnon T. Department of Emergency Medicine, University of Calgary, Calgary, AB

Introduction: The University of Calgary currently does not include a rotation in Emergency Medicine among the mandatory rotations for undergraduate medical education. As part of a larger curricular development initiative involving the development of a mandatory Emergency Medicine clerkship, we sought to design an examination blueprint that is congruent with learning objectives and reflecting the perceived importance of clinical presentations from the perspective of expert clinicians. The University of Calgary undergraduate medical education curriculum focuses on a clinical presentation model with 120 +/- 5 cardinal presentations. Methods: In this study, 24 specialists in Emergency Medicine (EM) in the Calgary Health Region were asked to score each of the 120 cardinal clinical presentations for "impact' and "frequency'. Multi-attribute utility theory was applied to assess the best way of combining the variables of "impact' and "frequency'. Statistical tests used were the Pearson's correlation coefficient and the Cronbach's alpha inter-rater reliability statistic. Results: 21 of 24 (87.5%) survey instruments were returned during the study period. Combining impact and frequency as a multiplicative function produced a distribution that was positively skewed towards common, high impact presentations such as chest pain. The correlation coefficients amongst high impact, high frequency clinical presentations were high, as were the correlation coefficients between low impact, low frequency presentations. Low impact, low frequency presentations demonstrated divergent validity (poor correlation) to high impact, high frequency clinical presentations. Cronbach's alpha (reliability coefficient) was 0.934 demonstrating very high internal reliability. Conclusions: Using previously established examination blueprint design techniques pioneered at our centre, we have created an Emergency Medicine examination blueprint that provides a realistic and objective measure of the relative importance of clinical presentations. This information can now be used in formal EM curriculum design and implementation, including objective writing and evaluation methods. Key Words: Medical education, Undergraduate training, EM curriculum

120 COMPUTER modeling of patient flow in a pediatric emergency department using discrete event simulation

Hung GR, O'Neill C, Gray AP, Whitehouse SR, Kissoon N. Division of Emergency Medicine, Department of Pediatrics; Office of Pediatric Surgery Excellence and Innovation; BC Children's Hospital, Vancouver, BC

Introduction: Pressures in pediatric emergency medicine (PEM), such as increasing census and overcrowding, require new strategies to address these concerns. Accurate predictions of patient flow and resource utilization in the pediatric emergency department (PED) are important in determining how PED activity could be modified to improve flow, reduce waiting times, increase efficiency and morale, and effectively direct change. Discrete event simulation (DES) was used to develop a Patient Flow Model (PFM) to test simulation scenarios designed to alleviate PED pressures. A Physician Scheduling Analysis Tool (PSAT) was designed to assist in physician scheduling. Methods: Arena DES software was used to develop a model of PED patient flow following interviews with staff and observation of 517 patients. Historical patient arrival information and observed patient flow data provided simulated patient arrival rates for the PFM and the PSAT. Validation of the PFM was performed by comparing simulated patient flow data to actual patient flow data. Previously determined staffing scenarios were applied to the simulation and the resulting patient flow indicator outputs were examined. Results: The PFM was validated on model-wide and process specific levels, with excellent validation observed on high acuity patient LOS (length of stay) and for detailed processes such as triage and registration. The simulated addition of a hospital volunteer and a second triage nurse reduced the pre-triage waiting time and the proportion of patients waiting >30-60 minutes pre-triage. Simulation of an extra physician shift to the staff schedule reduced the LOS for all patients. Conclusions: The PFM accurately represents patient flow in our PED and provides simulated patient flow data for a variety of scenarios. Based on the findings of the simulator scenarios, the authors' PED decided to make strategic decisions in reallocation of personnel resources. Less well validated aspects of the PFM would be improved by additional observational data. Key Words: Medical informatics, ED efficiency, Pediatrics

122 MEDICATIONS in the emergency department: lists, interactions, and survey of treatment

Michalski W, Leveck D, Sedran RJ. Division of Emergency Medicine, University of Western Ontario, London, ON

Introduction: Access to a complete patient medication list can be a valuable resource to the treating physician. This study was undertaken to determine what patients remembered about their current medication list, why they were on the medications listed, and how they remembered their medications. Methods: A questionnaire was administered between March and September 2006. Patients were asked demographic information and to list their medications, dose, frequency of administration, and reason for taking it. They were also asked how this information was remembered. This information was then confirmed by the patient's pharmacist. Results: Fifty-one questionnaires were successfully completed by adult emergency patients and consent was obtained (59% female). Ages ranged from 18 to 93 years old. Overall, 90% of patients could recall all of their medications, 67% knew the correct doses, and 82% could determine why they were on the medications listed. 40% respondents had their medication list memorized, 12% relied on family members to generate the list, and 48% relied on a written list or pill bottles. Males and females were similar in their abilities to remember their medications. 95% of the respondents had a primary care physician, and having a primary care physician did not correlate positively with an accurate medication list. Conclusions: Overall, 90% of respondents could accurately list their medications, and almost half could produce a current written list of medications. This did not seem to be affected by the age or sex of the patient. Multiple methods were utilized to remember their medication list. The presence of a listed primary care physician did not seem to allow for easier recollection of medications taken. Patients are most often accurate in their ability to remember their medication lists - often a trait of great value to a treating physician. In this small study, no adverse drug reactions were identified when the treating physician wrote a prescription without having an accurate patient medication list. Key Words: Medical informatics, Prescription medication

123 ARE there significant calendar cycles that predict emergency department use?

Kingsley SJ. Division of Emergency Medicine, University of Toronto, Toronto, ON

Introduction: Accurate predictions of daily emergency department (ED) usage can be valuable for staffing and resource allocation. While much data have been published correlating specific disease incidence with calendar cycles, only anecdotal evidence of specific weekday, lunar and yearly cycles of daily ED usage exists, and is not proven. We endeavoured to identify significant calendar trends and their contributions to daily ED usage using Fourier Transform Analysis (FTA). FTA is a powerful mathematical tool that can graphically identify trends in periodic data and their relative contribution to the overall data series, without the need for a priori hypotheses required by more traditional statistical methods. Methods: Daily ER volumes were recorded from our institution, an urban teaching hospital with yearly ED volumes > 55,000. 1,024 consecutive daily volumes underwent FTA to identify significant contributing calendar cycles. 365-day cycles in ER volume were hypothesized to be dependent on weather, with 30 or 31-day and 7-day cycles representing lunar and weekday effects respectively. Identified cycles were correlated with their hypothesized calendar cause (weather, lunar cycle or weekday) and subject to multivariate regression to create a calendar prediction model for ER volumes. Results: The strongest contributing calendar cycle occurred yearly (every 377±18.71 days), signal-to-noise ratio=6.91. The second most significant calendar effects noted were weekly (7.0±0.06 days, S:N=6.18) and twice weekly cycles (3.5±0.01 days, S:N=5.90), corresponding to Monday and Thursday weekdays. Multivariate regression of weather (daily temperature) and weekday as the only independent variables created an ED volume prediction model with a fit of r^2=0.28, F=97.6. Conclusions: FTA is a useful analytic tool to identify cyclical data trends. Using FTA in this preliminary study, we identified that weekday and outdoor temperature were substantial predictors of daily ED usage, alone accounting for 28% of daily ER volume fluctuation. Key Words: Medical informatics, ED administration, Computer modeling

124 SENSIBILITY survey of community acquired pneumonia and neutropenic fever electronic clinical decision support systems

Graham TAD, Bullard M, Rowe BH. Department of Emergency Medicine, University of Alberta, Edmonton, AB

Introduction: Emergency Department (ED) clinicians are increasingly exposed to guidelines and treatment recommendations. To help access and recall these recommendations, electronic Clinical Decision Support Systems (eCDSS) have been developed. This study examined the use and sensibility of two common eCDSS. Methods: eCDDS for community acquired pneumonia (CAP) and febrile neutropenic (NF) were developed by multi-disciplinary teams and have been accessed via an intranet-based homepage (eCPG©) for several years. Sensibility is a term coined by Feinstein that describes common sense aspects of a survey instrument. It was modified by emergency researchers to include four main headings: 1) Appropriateness; 2) Objectivity; 3) Content; and 4) Discriminative Power. Sensibility surveys were developed using an iterative approach for both the CAP and NF eCDSS and distributed to all 25 emergency physicians at one Canadian site. Results: The overall response rate was 88%. Respondents were 88% male and 83% were less than 40; all were attending ED physicians with specialty designations. An unexpected number had never used the CAP (21%) or NF (33%) eCDSS; 54% (CAP) and 21% (NF) of respondents had used the eCDSS less than 10 times. Overall, both eCDSS were rated highly by users with a mean response of 4.95 (SD 0.56) for CAP and 5.62 (SD 0.62) for NF on a 7-point Likert Scale. The majority or respondents (CAP 59%, NF 80%) felt that the NF eCDSS was more likely than the CAP eCDSS to decrease the chances of making a medical error in medication dose, antibiotic choice or patient disposition (4.61 vs. 5.81, p = 0.008). Conclusions: Despite being in place for several years, eCDSS for CAP and NF are not used by all ED clinicians. Users were generally satisfied with the eCDSS and felt that the NF was more likely than the CAP eCDSS to decrease medical errors. Additional research is required to determine the barriers to eCDSS use. Key Words: Medical informatics, Clinical decision support, Guidelines

125 THE use of usability engineering methods to define usability problems in emergency department electronic clinical decision support systems

Graham TAD,* Kushniruk AW.† *Department of Emergency Medicine, University of Alberta, Edmonton, AB, and the †School of Health Information Sciences, University of Victoria, Victoria, BC

Introduction: Emergency Department (ED) clinicians are increasingly exposed to electronic Clinical Decision Support Systems (eCDSS). For a health care CDSS to be used, it must not only be factually accurate, but also designed in a way that users will find pleasant to use. Usability refers to how quickly people can learn to use something, its efficiency in use, its memorability, its proneness to error, and its enjoyability. Methods: eCDDSs for community acquired pneumonia (CAP) and febrile neutropenia (NF) were developed by a multi-disciplinary team and can be accessed via an intranet-based homepage (eCPG©). One aspect of the usability of CAP and NF eCDSS was examined using the cognitive walkthrough methods described by Neilsen and Kushniruk. All pages of the eCDSS were scrutinized by the researchers, and the actions and system responses required to complete each goal recorded. Actions and responses were examined to specifically look for usability problems, and aspects of the systems that could cause adverse events. Results: Assuming all potential fields were filled, the CAP eCDSS had 5 Goals, 37 Sub-goals, 56 Actions to complete all Sub-goals, and 52 Problems. The NF eCDSS had 3 Goals, 44 Sub-Goals, 63 Actions to complete all Sub-goals and 60 problems. Recurring problems were included in the totals. Eight categories of usability problems emerged: 1) Ambiguous Wording; 2) Layout; 3) Defining User Input; 4) Clinical Decision Support; 5) Convenience of Use; 6) Relevance to the ED; and 7) Incomplete or Out of Date Information. Conclusions: The interaction of physicians with CDSS is complex and poorly understood. Assessment tools for computer systems have been developed based on cognitive and usability engineering principles. While the cognitive walkthrough does not quantify the seriousness of a problem, its systematic approach can identify previously unnoticed issues, including potential adverse medical events. Usability engineering methods hold promise in characterizing usability problems with CDSS in the ED and elsewhere. Key Words: Medical informatics, Clinical decision support, Guidelines

126 A retrospective study of patient compliance with emergency department referral to a cardiovascular evaluation and risk assessment clinic

Wojtowicz JM, Dowling S, Nanji A, MacLeod DB. Departments of Family Medicine and Emergency Medicine, University of Calgary, Calgary, AB

Introduction: Cardiovascular disease is a leading cause of death among Canadians and has a substantial impact on the health care system. The C-ERA (Cardiac Evaluation and Risk Assessment) clinic is utilized by emergency physicians in Calgary to facilitate the investigation of patients presenting with symptoms of cardiovascular disease. Patient compliance with emergency department (ED) discharge instructions is often poor, resulting in sub-optimal patient care. The primary objective was to determine the proportion of patients compliant with ED referral to C-ERA. In addition, we examined the diagnoses and outcomes of patients attending C-ERA. Methods: The present study was a retrospective review of 385 patients referred to C-ERA between June 1, 2004 and April 7, 2005. Hospital charts and the database at the Medical Examiner's office were reviewed for patients who did not attend or cancel their C-ERA appointment. Results: The majority of patients (364/385, 94.5%) were compliant with referral to C-ERA. Included in the compliant group were 20 patients who called to cancel their appointment and 28 who rescheduled and later attended their appointment. Twenty-two patients did not attend or cancel their appointment. One patient was in hospital at the time of their C-ERA appointment. No deaths were reported in hospital records or at the Medical Examiner's office for non-compliant patients. Most patients (315/340, 93%) who attended their appointment completed their recommended investigations. Of these patients, 221 (70%) were diagnosed with non-cardiac or low risk cardiac disease. Ninety-four patients (30%) were diagnosed with a more serious cardiac condition. No patients were diagnosed with an acute coronary syndrome. Four patients required angioplasty and 2 underwent cardiac surgery. Conclusions: The majority of patients referred to C-ERA from Calgary EDs were compliant with referral. Other out-patient clinics may be able to improve their rate of compliance with referral by implementing the referral system used by C-ERA. Key Words: Coronary artery disease, ED follow-up, Patient compliance

127 HYPOTHERMIC modulation of anoxic brain injury: a survey of Canadian emergency physicians

Kennedy J, Green RS, Stenstrom R. Department of Emergency Medicine, Department of Medicine (Critical Care), Dalhousie University, Halifax, NS; Department of Emergency Medicine, University of British Columbia, Vancouver, BC

Introduction: Mild induced hypothermia (IH) for survivors of cardiac arrest has been demonstrated to reduce morbidity and mortality. Despite this, data from other countries has demonstrated that the incorporation of IH into clinical practice is uncommon. The objective of this study is to characterize the use of induced hypothermia (IH) by Canadian emergency physicians. Methods: An internet-based survey was distributed to members of the Canadian Association of Emergency Physicians. Participants were asked to provide data on the frequency of use of IH and on the methods employed for IH. In addition, potential barrier to the incorporation of IH into practice were explored. Results: In total, 1328 members of CAEP were contacted by email and 247 (18.6%) responded, with the majority working in an academic center (60.3%). Forty-seven percent (47%, 95% CI = 40.8-53.2) of respondents indicated that they have utilized IH in clinical practice and 34.8% (85/247) worked in a department that had a policy or protocol for the use of IH. The presence of a departmental policy or protocol for the use of IH was a strong predictor of the use of IH (chi-square 51.7, p<0.001). Barriers against the incorporation of IH included a lack of institutional policies and protocols (39%) and a deficit in resources (29%). Consultant lack of support for IH was relatively uncommon (9%) in Canadian practice. Conclusions: Although a substantial proportion of Canadian emergency physicians have incorporated IH into their practice, a large proportion of respondents in this survey have not. Emergency departments should develop policies or protocols for the use of IH in cardiac arrest survivors to optimize patient outcomes. Key Words: Cardiac arrest, Hypothermia, Hypoxic brain injury

128 IMPACT of first-responder CPR on defibrillation opportunities following the detection of asystole

Vu EN, Innes KC, Holmes AE. Global Medical Services, Vancouver, BC

Introduction: The updated AHA guidelines for CPR and ECC were published in the fall of 2005 recommending in situations of unwitnessed cardiac arrest (CA), before the application of an automatic external defibrillator (AED), it is reasonable to attempt 5 cycles (about 2 minutes) of CPR prior to rhythm analysis and attempting defibrillation (Class IIb). It is suggested that with effective CPR, one can induce ventricular activity that is more amenable to defibrillation, i.e. ventricular fibrillation (VF) or pulseless ventricular tachycardia (pVT). Methods: Global Medical Services (GMS) provides medical oversight for >50 professional and volunteer first-responder (FR) AED programs throughout Canada. Following a CA, AED data is downloaded to a central database in Vancouver, BC. Each download is reviewed by one of two physicians and pre-determined data points, including rhythm analysis, are entered into the database. We performed a retrospective review of the prospectively gathered database for all CA from 19/04/2001-06/25/2006. The primary endpoint was detection of VF/pVT after deliver of FR CPR when the initial rhythm was asystole. The secondary endpoint was time interval between detection of asystole and return of a shockable rhythm. Results: From 19/04/2001-06/25/2006 we received AED downloads from 982 prehospital CA. Asystole was the original rhythm detected in 326 of the cases. Of these, VF/pVT was subsequently detected in 20 (6.1%) cases after FR CPR. The average time interval from detection of asystole to the detection of VF/pVT was 07:25 (min:sec) [range: 00:55 to 15:56]. Conclusions: Our data suggest that FR CPR could create opportunities for defibrillation in approximately 6 out of 100 cardiac arrests in which the original rhythm is asystole. In light of the updated AHA guidelines for CPR and ECC, this data likely under represents the potential impact of CPR on out-of-hospital CA and underscores the importance of early and effective CPR in conjunction with an AED program. Key Words: Coronary artery disease, Cardiac arrest, CPR, Defibrillation

129 IMPACT of increased availability of primary angioplasty on reperfusion times for ST-elevation MI in the ED

Stacey M, Borgundvaag B. Division of Emergency Medicine, University of Toronto, Toronto, ON

Introduction: Current guidelines for the management of uncomplicated ST-elevation MI (STEMI) suggest that primary angioplasty (PCI) is the preferred initial therapy only if it can be performed within 90 minutes of hospital presentation (or 60 minutes from when fibrinolysis could have been delivered). We examined reperfusion times at our institution over the past 8 years, using both treatments, to determine our performance as emphasis has shifted towards PCI. Methods: A preliminary medical records search (using ICD-9/10 codes for myocardial infarction) was performed on all patients registered in our ED between April 1998 and August 2006. To be included, patients had to be diagnosed with STEMI within 2 hours of presentation, and treated with reperfusion therapy. We performed a structured chart using predefined criteria for all eligible patients. Statistical analysis included simple descriptive statistics (mean, standard deviation, etc), t-tests, and linear regression analysis. Results: 995 charts were reviewed, and 169 charts met criteria for inclusion. The mean (+ SD) age of participants was 61.5 + 13.7 years, and 78% were male. Thirty-seven percent of patients presented between 08:00-17:00 hrs. The mean (+SD), median, lower and upper quartile ranges for door to needle times were 49+34, 40, 26 and 62 minutes. The similar values for door to balloon times were 160+67, 139, 110 and 184 minutes. The mean difference between door to needle and door to balloon time was 110 minutes. Thirty-five percent of thrombolysis, and 3 percent of PCI patients were treated with the recommended time window. There was no overall change in time to lysis or PCI over the study period, however there was a significant reduction in door to balloon time during daytime hours. Conclusions: Fewer patients were treated within the recommended time with PCI compared to fibrinolysis. Door to needle times have remained unchanged, while door to balloon times are improving for PCI during daytime hours. Key Words: Coronary artery disease, Myocardial infarction, Revascularization, Thrombolysis, Time to treatment

130 WHY are emergency departments holding back on holding chambers? Facilitators and barriers to change

Hurley K, Sargeant J, Duffy J, Sketris I, Sinclair D, Ducharme J. Department of Emergency Medicine; Office of CME; Faculty of Management; and College of Pharmacy & CHSRF/CIHR Chair in Health Services Research, Dalhousie University, Halifax, NS

Introduction: Best available evidence points to the therapeutic equivalence of portable inhalers and holding chambers for delivery of beta-agonist respiratory medications to children in the Emergency Department (ED). Yet only a minority of pediatric EDs in Canada have made the change. The objective of this study was to explore the perceptions surrounding use of portable inhalers and holding chambers in the ED and the facilitators and barriers to practice change. Methods: This was a qualitative study guided by principles of grounded theory. Data were collected through focus groups and individual interviews at two sites in Eastern Canada: Hospital A, where inhalers and holding chambers are used routinely; and Hospital B, where prevailing practice is use of nebulization. Participant encounters were transcribed verbatim and analyzed for emerging themes. Results: At Hospital A, 6 physicians and 7 nurses participated in separate focus groups. Four interviews were conducted with physician, nurse, respiratory therapy and pharmacy leaders. At Hospital B, 4 physicians and 3 nurses participated in focus groups while 6 leaders were interviewed. Barriers to adoption of inhalers and holding chambers included: increased workload; increased equipment costs; myths regarding the superiority of nebulization; and inter-professional conflict. Conclusions: The most prominent concern for health care professionals, particularly nurses, about administering medications with inhalers and holding chambers was the time demand. This resulted in continued administration of nebulization despite knowledge of evidence to the contrary. Professional territorialism hampered efforts to ameliorate workload issues through use of respiratory therapists in the Emergency Department. Key findings from this study can be used to inform a change program to close the gap between evidence and practice with respect to use of inhalers and holding chambers in the Emergency Department. Key Words: Inhalational therapy, Asthma, Aerochamber, Nebulizer