CAEP/ACMU 2002 Scientific Abstracts: 1-14

2003 Scientific Abstracts

CJEM 2002;4(2):124-154

001 Patients Who Refuse Transport: A Canadian Perspective.
Lehnhardt KR, Lewell MP. University of Western Ontario. London, ON

INTRODUCTION: One area of potential medicolegal risk in prehospital care comes from patients who refuse transport to hospital after emergency medical services (EMS) have established contact. This study was conducted to determine the frequency of these so-called no service (NS) calls, to identify patient population demographics, and to evaluate documentation on these calls. METHODS: A retrospective chart review was performed on all NS calls in January and July 2001. The setting was an urban environment (population 340 000) that uses a single provider EMS system, consisting of both advanced and basic life support paramedics (ALS and BLS). RESULTS: Over the 2 months, 299/4811 patients refused transport (6.2%). The male:female ratio was approximately 1.05:1 and the average age was 47.0 years (range 1-94). The most common chief complaints were trauma (36.8%) and general medical conditions (19.7%). Documentation was found to be incomplete on 91% of these calls. A complete set of vital signs was not recorded for 59/299 calls (19.7%), and specifically, blood pressure was not measured in 47.8% of patients. An initial Glasgow Coma Scale (GCS) rating was not calculated 14.4% of the time while the number of patients with a GCS <15 at presentation was 41/299 (13.7%). The average time on scene for a NS call was 17 min ± 9 min. CONCLUSIONS: A significant number of patients refuse transport to hospital. Many of these patients exhibit concerning features upon presentation, such as an initial GCS score of less than 15. The true number of patients at risk of adverse outcome is unknown, due to the lack of complete documentation in this population. Further study is planned to develop strategies that will reduce the number of NS calls.

002 The Positive Impact of Implementing a Fast Track in an Urban Emergency Department.
Hall C, Wang D, Young B. Calgary Health Region, University of Calgary. Calgary, AB.

INTRODUCTION: Urban Emergency Department (ED) overcrowding negatively impacts patient care. Dedicating limited resources to less acute patients (CTAS Level IV & V) may be perceived to be at the expense of the care of more seriously ill (CTAS Level III). This study determined the impact of triaging CTAS Level V and some Level IV patients through a Fast Track system, on both Length of Stay (LOS) and percentage of patients who leave without being seen (LWBS) in an urban ED. METHODS: The ED was redesigned to allow CTAS Level IV & V patients to be diverted 16 hrs/day to a separate area of the ED (FAST TRACK) with waiting room, dedicated physician, nursing and clerical staff. Patient data was collected prospectively on ED census, CTAS categories, LOS in hours, number of LWBS. Data were analyzed for the 4 months pre and 2 months post redesign. Statistical analysis utilizing ANOVA was carried out to assess differences in LOS, LWBS and LWBS by CTAS category. RESULTS: ED census during the study interval was consistently over 5000 patient visits per month. Mean overall LOS decreased from 4.24 hrs (SE 0.065) to 3.74 hrs (SE 0.075), p = 0.011. Overall LWBS also decreased from 9.20% (SE 0.46) to 4.64% (SE 0.53), p = 0.006. Significant reductions in LWBS were not restricted to CTAS IV and V but also occurred in CTAS III. CTAS III LWBS decreased from 8.90% (SE 0.39) to 5.99% (SE 0.25), p = 0.008. CTAS IV LWBS decreased from 11.10% (SE 1.02) to 4.55% (SE 0.73) with a p value of 0.015. Finally CTAS V LWBS decreased from 11.50% (SE 0.89) to 5.21% (SE 2.56), p = 0.038. CONCLUSION: Fast Tracking ambulatory patients in a busy urban ED positively impacts the LOS and numbers of LWBS for patients in all of CTAS category Levels III, IV and V.

003 Changing Patterns of Investigation and Treatment of Deep Vein Thrombosis in Two Emergency Departments.
Rowe BH, Holroyd BR, Willis G, Meurer D, Sukhrani N, Spooner CH, Bullard M, Kelly KD. Division of Emergency Medicine, University of Alberta, Edmonton, AB.

OBJECTIVES: For patients presenting to the Emergency Department (ED) with suspected deep vein thrombosis (R/O DVT), diagnostic and treatment approaches vary widely. One way to reduce variation involves the use of clinical practice guidelines (CPG) and this study examined the effect of incorporating a CPG for R/O DVT. METHODS: The CPG was developed using Wells' clinical model (CM), D-dimer testing and venous Dopplers and was approved by a majority physician vote. In addition, justification of venography ordering was required prior to completion of the testing. Control hospital MDs received passive CME interventions, but had access to Dopplers and D-dimer testing. Patients presenting to the 2 EDs from 1999-2001 were eligible for chart review if they were >16 years of age with a R/O DVT. Primary investigations (Doppler ultrasounds, venograms, and D-dimers) were recorded; the main outcome was the percentage of venograms ordered as the first test at each hospital. RESULTS: Overall, during the study period 89 control and 338 intervention hospital charts were reviewed. Patients with R/O DVT were more likely to be female (59%) and older (mean age: 55) at both sites (p > 0.05) but control patients had fewer co-morbid conditions. Laboratory evaluation of hemoglobin, plts, and INR were similar. However, CM use (59% vs 0%; p < 0.001), D-dimer (59% vs 35%; p < 0.01) were more common at the CPG hospital, and venograms were less frequently ordered as the first test (6% vs 33%; p < 0.001) at the CPG hospital. Overall, both hospitals had low admission rates for confirmed DVTs. CONCLUSIONS: Clinical practice guidelines in the ED are potentially valuable tools for improving quality of care. This study demonstrated the successful application of a CPG for R/O DVT, specifically by reducing the percentage of patients receiving venography. Further research is required to determine how to enhance uptake of CPGs by emergency physicians.

004 Increased Emergency Department Volumes but Decreased Overall Utilization: Ontario's Hospital Restructuring Paradox.
Schull M, Chan B, Schultz S. Institute for Clinical Evaluative Sciences, Sunnybrook & Women's College Health Sciences Centre, University of Toronto. Toronto, ON.

OBJECTIVE: The hospital system in Ontario underwent substantial restructuring in the 1990s. We sought to compare the number of emergency departments (ED) in the province with the total number of ED visits during this period. METHODS: Over the study period (fiscal years 1993 to 2000), most ED physician services were paid for on a fee-for-service (FFS) basis. We obtained data on all FFS physician billing records for ED patient visits in Ontario. EDs where physicians' services were not continuously remunerated by FFS were excluded. To confirm ED status, billing data was supplemented by surveys of District Health Councils (DHC), and telephone calls to individual hospitals. Limited service EDs were defined as those limiting patient visits or ambulance arrivals to <24 hours/day. Statistical tests were not conducted since we included virtually the entire ED patient population. RESULTS: In 1993, there were 201 EDs in Ontario. By 2000, 20 (9.5%) had closed, 7 (3.5%) had reduced services and 0 had opened, leaving 181 full or limited service EDs. Across the 17 regional DHCs, the proportion of EDs that closed ranged from 0% (0 of 19) to 36% (4 of 11). Over the same period, the population of Ontario increased by 8.9%, but the overall per capita ED visit rate declined by 10.3%; these trends resulted in a decrease in the total number of ED visits of 2.2% (from 3.34 to 3.27 million visits). As a result, the average number of visits per ED rose by 10%, from 19,111 visits per ED in 1993 to 21,096 visits per ED in 2000. CONCLUSION: As a result of ED closures, the average ED in Ontario had a substantially higher visit volume in 2000 than in 1993, despite reduced overall utilization. Planners should consider these trends when trying to predict future demand for ED services.

005 Successful Implementation of a Combined Pneumococcal and Influenza Vaccination Program in a Canadian Emergency Department.
Pearson E, Lang E, Colacone A, Goulet M, Rahmani S, Virgona M, Trudel N, Afilalo M. Sir Mortimer B. Davis Jewish General Hospital. McGill University. Montreal, QC.

INTRODUCTION: Although many emergency departments (EDs) see a high concentration of patients that are unprotected by either pneumococcal or influenza vaccines, few centres have developed an ED vaccination program. This study assessed the extent to which a pneumococcal and influenza vaccination program could be successfully implemented in a Canadian ED. METHODS: Design: Prospective cohort study. Setting: Tertiary-care academic centre. Participants: All patients eligible to receive either influenza or pneumococcal vaccine and presenting to the ED on weekdays from 8 AM to 4 PM from November 1-30 were approached. Interventions: A questionnaire was administered to all consenting patients. Unvaccinated patients who did not plan on being vaccinated elsewhere were offered vaccination in the ED. If willing, the patient was vaccinated by a dedicated study nurse and completed a satisfaction questionnaire. RESULTS: During the study period, 753 patients (36%) presenting to the ED were eligible for vaccination with either vaccine; 86.8% (95% CI: 84-89%) on the basis of age, and 13% (95% CI: 11-15%) on the basis of chronic disease. 169 patients (22%) were excluded due to predefined exclusion criteria, 20 (3%) were missed, 35 (5%) refused consent, and 529 (70%) consented to participate in the study and completed a questionnaire. Of the study patients, 282 (53%; 95% CI: 49-57%) were unvaccinated against influenza that year and did not plan on being vaccinated elsewhere, and 279 (53%; 95% CI: 49-57%) were unvaccinated against pneumococcus and did not plan on being vaccinated elsewhere. Influenza vaccine was administered to 187 study patients (36%; 95% CI: 32-40%) while 165 (32%; 95% CI: 28-36%) received pneumococcal vaccine. CONCLUSIONS: In this setting, an ED-based vaccination program can reach a significant proportion of the clientele at risk who would otherwise go unprotected. Similar programs merit implementation on an annual basis.

006 Influenza in the Elderly and Emergency Department Overcrowding.
Schull M, Mamdani M, Redelmeier D. Institute for Clinical Evaluative Sciences, Sunnybrook & Women's College Health Sciences Centre, University of Toronto. Toronto, ON.

OBJECTIVE: Influenza outbreaks may contribute to Emergency Department (ED) overcrowding. We sought to determine the impact of influenza on ED utilization. METHODS: We obtained weekly totals of laboratory-confirmed cases of influenza (A and B) and other respiratory viruses in Toronto from January 1996 to April 1999 (number of weeks = 170). Weekly proportions of total visits to Toronto EDs due to upper respiratory (URT) and lower respiratory tract (LRT) conditions were determined, along with cardiac and psychiatric conditions as 2 control groups. Time series modeling tested the association of ED utilization and influenza cases. Covariates were other respiratory virus cases, average ED patient age and sex distribution. RESULTS: A total of 1,882,702 ED visits over the study period, with a weekly mean of 11,075. Patient age averaged 40 years with 51% female. Weekly cases of influenza ranged from 0-236 (mean = 20); cases of other viruses ranged from 0-91 (mean = 24). Among patients aged <65 years, the mean proportion of total ED visits per week due to LRT conditions was 4% and for URT conditions 6%. Among patients aged >65 years, the mean proportions were LRT 1.8% and URT 0.4%. In time series models, influenza was a significant predictor of increased ED utilization for both LRT (p < 0.001) and URT diagnoses (p < 0.001), but only among patients >65 years. For every 100 cases of influenza, there was an absolute increase of 2% and 1% in the proportion of total visits due to patients >65 years with LRT and URT diagnoses respectively. This was equivalent to relative increases of 111% for LRT and 250% for URT over their weekly means. Influenza was associated with a small decrease in cardiac utilization among the elderly (-0.2%/100 cases; p = 0.04), and was not associated with psychiatric utilization. CONCLUSIONS: Influenza virus is a major predictor of increased ED utilization for respiratory conditions among the elderly.

007 Serum Alpha-Glutathione S-Transferase Following Supratherapeutic Dosing of Acetaminophen in Human Volunteers.
Sivilotti MLA, Montalvo M, Brison RJ, Linden CH. Departments of Emergency Medicine, and of Pharmacology & Toxicology, Queen's University. Kingston, ON.

INTRODUCTION: Alpha-Glutathione S-transferase (alpha-GST) is a promising new biomarker of end-organ toxicity following acetaminophen (APAP) overdose. Previous work has shown alpha-GST appears in the serum shortly after overdose, but not after therapeutic doses of APAP. Understanding the temporal dose-response profile of serum alpha-GST release is necessary prior to clinical application of this biomarker, and to explore its potential as a surrogate outcome for interventional studies in humans. We sought to characterize the early response of serum alpha-GST to a single supratherapeutic dose of APAP in healthy subjects. METHODS: Prospective human volunteer study. Fasting subjects received 100 mg/kg liquid APAP, and serum assayed every 2 hours until 10 hours for APAP, AST, ALT, bilirubin and alpha-GST concentrations. Patients with known risk factors for APAP hepatotoxicity were excluded. RESULTS: 24 healthy subjects (age 18-42; 13 male) achieved serum APAP levels of (mean ± sd) 347 ± 65 micromole/L (4 h), and 93 ± 53 micromole/L (10 h). AST and ALT remained unchanged (final-initial AST -11 ± 18, ALT -2.0 ± 3.1 IU/L). On average, alpha-GST levels did not change substantially (average peak-baseline 2.6 ± 6.3 microgram/L). Two subjects, however, were clear outliers ("responders") with substantial increases in alpha-GST (final 35 and 14.5 microgram/L vs initial 7.5 and 0.5 microgram/L, respectively), despite unchanged serum transaminse concentrations (final-initial AST +2 and +9, ALT -5 and +6 IU/L). CONCLUSIONS: Serum alpha-GST is unlikely to rise appreciably following supratherapeutic but subtoxic ingestions in healthy subjects, supporting its use as an early "rule out" marker in the overdose setting. Moreover, the ability to demonstrate a response in alpha-GST in a small subset of subjects suggests a model to screen patients for vulnerability to APAP-induced hepatic injury. This biomarker might ultimately improve the specificity and dosing of N-AC therapy, and help resolve controversy regarding risk factors (e.g. alcoholism) for hepatic injury following APAP exposure.

008 Diagnostic Accuracy of Helical Computed Tomography in the Diagnosis of Renal Colic in the Emergency Department: A Meta-Analysis.
Meyers C, Lang E, Dankoff J, Moore S, Martin K, Afilalo M. Sir Mortimer B. Davis-Jewish General Hospital, McGill University. Montreal, QC.

INTRODUCTION: Non-enhanced helical computed tomography (NHCT) has replaced intravenous pyelography as the diagnostic test of choice for suspected renal colic in many institutions. Several studies have reported diagnostic parameters for this test, but many have been flawed by small numbers and methodologic problems. The objective of this study was to perform a systematic review of all available evidence and to pool data from studies meeting specific methodologic criteria in order to better define the diagnostic accuracy of NHCT in emergency department (ED) patients with suspected renal colic. METHODS: Prospective studies reporting the sensitivity and specificity of NHCT in patients with acute flank pain were identified by computerized MEDLINE and manual searching. Using validated criteria, retrieved studies were reviewed by 2 investigators who were blinded to authorship, journal of publication, and results. Our a priori criteria for study inclusion were prospective data collection, enrolment of individuals with and without disease, CT done without contrast and interpreted by a blinded investigator, and comparison to an acceptable reference standard. Patient data from studies meeting these inclusion criteria were extracted and analyzed as a summary receiver operating characteristic curve using a random effects model, from which pooled sensitivity, specificity, and likelihood ratios were derived. RESULTS: Four studies met all a priori inclusion criteria, for a total of 313 patients. For the purposes of the meta-analysis, 1patient originally classified as a true positive, and 3 patients originally classified as true negatives, were reclassified as false positives based on definitions used for the analysis. This resulted in a pooled sensitivity of 95% (95% CI: 91%-97%), a specificity of 93% (85%-97%), and positive and negative likelihood ratios of 13.57 (6.07-32.33) and 0.054 (0.031-0.106), respectively. CONCLUSIONS: NHCT is a useful test for the diagnosis of renal colic in the ED.

009 Serum Alpha-Glutathione S-Transferase Becomes Elevated Shortly after Subtoxic Acetaminophen Overdose.
Sivilotti MLA, Bird DB, Montalvo M, Aaron CK, Brison RJ, Linden CH. Departments of Emergency Medicine, and of Pharmacology & Toxicology, Queen's University. Kingston, ON.

INTRODUCTION: Alpha-Glutathione S-transferase (alpha-GST) is a promising new biomarker of end-organ toxicity following acetaminophen (APAP) overdose. Unlike traditional liver function tests, alpha-GST appears in the serum shortly after overdose in patients who fall above the Rumack-Matthew nomogram threshold for initiating N-acetylcysteine (N-AC). The serum alpha-GST profile following lesser ingestions of APAP in humans is unknown. We sought to quantify the early alpha-GST response following subtoxic exposures to APAP. METHODS: Prospective, observational pilot study of patients presenting within 4 hours of a single acute APAP overdose, with a 4-hour serum APAP level of 500-1000 micromole/L (i.e. below but within 50% of the nomogram threshold). Levels were measured every 2 hours until 10 hours post ingestion. RESULTS: 8 patients (median age 18 years, range 15-59; 2 male) were studied. 4-hour APAP levels were (mean ± sd) 815 ± 212 micromole/L, and all patients were considered low-risk by conventional criteria. Despite persistently normal serum AST and ALT in all patients, 4 had serum alpha-GST levels above the 95% ile upper limit of normal (peak 13, 30, 72, and 79 microgram/L; normal <7.5 microgram/L). Three of these patients had elevated alpha-GST at presentation, and levels had normalized by 10 hours (in the absence of N-AC therapy) in 2 of these 3. CONCLUSIONS: alpha-GST appears in the serum shortly after APAP exposure in patients falling below the "possible hepatotoxicity" nomogram zone. Small elevations in alpha-GST may represent reversible, subclinical injury to vulnerable centrilobular hepatocytes. This phenomenon suggests end-organ toxicity occur even during the classically taught 8-hour window of adequate glutathione protection. This biomarker deserves further study, given its potential to help risk-stratify patients in whom the nomogram cannot be applied (uncertain time of ingestion, repetitive dosing), and to reduce unnecessary or prolonged antidotal treatment with N-acetylcysteine

010 Does Taking a History Help in the Evaluation of Potential Cervical Spine Injury?
Stiell IG, Brison R, Clement C, Bandiera G, Holroyd B, Dreyer J, McKnight RD, Morrison L, Reardon M, Worthington JR, Battram E, Wells GA, for the CCC Study Group. Division of Emergency Medicine, University of Ottawa, Ottawa, ON.

OBJECTIVES: Many physicians pay little attention to history-taking in patients with potential C-spine injury. This study measured the accuracy and reliability of specific history findings in alert trauma patients. METHODS: We conducted a prospective cohort study in 10 tertiary care EDs and enrolled alert and stable adult trauma patients at risk for neck injury. MDs took a standardized history from patients and recorded results. In some cases, 2nd physicians performed interobserver assessments. Patients underwent radiography to determine the outcome, clinically important C-spine injury. Analyses included univariate association, kappa, sensitivity, specificity, adjusted odds ratio by stepwise logistic regression. RESULTS: We enrolled 8,924 patients with mean age 36.8 years and important C-spine injury 1.7%. The following Table shows % of injury and non-injury patients with the history findings, p-value, kappa coefficient, sensitivity, specificity, adjusted odds ratio.

CONCLUSIONS: A history of 'age 65', 'dangerous mechanism', or 'paresthesias' put patients at significantly higher risk of C-spine injury. A history of 'rear-end MVC' or 'delayed onset neck pain' put patients at much lower risk. All 5 of these findings are also reliable and should be useful to clinicians evaluating the risk of C-spine injury in alert trauma patients.

011 Management and Outcomes of Out-of-Hospital Seizure Patients Attended to by EMS.
Saginur M, Stiell IG, Nesbitt L, Martin MT, Brisson D, Doherty J, Beaudoin T, for the OPALS Study Group. Division of Emergency Medicine, University of Ottawa, Ottawa, ON.

OBJECTIVES: Little is known about the outcomes and diagnoses of seizure patients seen by EMS. This study describes the management, hospital disposition, and final diagnoses of these patients. METHODS: This health records review constitutes the seizure sub-study of the Ontario Prehospital Advanced Life Support (OPALS) Study, the largest prehospital study yet conducted. The OPALS Study will assess the impact of prehospital ALS on patient outcomes. This sub-study was conducted in a city of 750,000 with a BLS-D EMS system. Included were all adult out-of-hospital seizure patients seen over a 6-month period. Data sources were ambulance call reports, centralized dispatch data, ED and in-hospital records. Seizure duration was defined as the total time of individual seizures plus interictal periods not interrupted by regained consciousness (GCS >=14). Analysis included descriptive statistics with 95% CIs. RESULTS: Of 154 suspected seizures, 129 were true seizure cases, with the following characteristics: mean age 46 (range 18-92), male 70%, seizure history 80%, generalized 90%, not transported 7%. Prehospital seizure activity ended prior to EMS arrival in 65% of cases and prior to arrival at ED in 78%. Mean prehospital seizure duration was 20 minutes (range 1-220). ED records were available in 111 of 120 cases (92.5%): 33% received IV anticonvulsants and 6% were intubated. From the ED, 35% were admitted (ICU 5%), 62% discharged, and 3% left AMA. Admitted patients' median length of stay was 5 days (range 1-610), including 0.3 days in ICU. Overall survival was 97.5%. Discharge CPC scores were: 'good' 50%, 'moderate disability' 9%, 'severe disability' 41%. Chronic epilepsy caused 78% of cases; new-onset seizure diagnoses were: tumour 4%, alcoholism 2%, trauma 2%, CVA 1.5%, other known 5%, and unknown 9%. CONCLUSIONS: This comprehensive review offers the first in-depth profile of prehospital seizure patients. These data are essential for the design of future studies of EMS seizure management.

012 Comparison of Three Immobilization Techniques in the Management of Acute Distal Radius Fractures.
Grafstein EJ, Jackson C, Innes GD, Christenson JM, Boychuk BA, Stothers K, McCormack R. Providence Health Care, St. Paul's Hospital, UBC. Vancouver, BC.

OBJECTIVE: To compare the effectiveness of cylindrical casts (CC), volar-dorsal (VD) splints and Muenster modified sugar tong splints (M) for the immobilization of distal radius fractures. METHODS: A multicentre, randomized trial at 4 Vancouver hospitals. Emergency physicians/orthopedic residents reduced all fractures primarily. Patients with undisplaced fractures or unsuccessful primary reductions were excluded. Patients >18 years of age who underwent successful closed reduction of a displaced distal radius fracture were randomized to CC, VD or M immobilization. Physicians were trained in all 3 techniques. Patients had telephone follow up at 2 days, x-ray and orthopedic follow up at 7 and 28 days, and functional assessment at 2 and 6 months. Individual x-ray views from all patients were placed in random sequence, and a blinded radiologist assessed shortening, dorsal angulation, radial inclination, and dorsal comminution. The primary radiological outcome, loss of reduction, was based on explicit criteria. The primary clinical outcome was need for secondary reduction. RESULTS: 40 CC patients, 31 VD patients and 30 M patients were analyzed (n = 101). Baseline fracture characteristics, including shortening, angulation, radial inclination, and dorsal comminution pre-reduction, were similar in all groups. By 7 days, 23% had radiographic loss of reduction, including 8 (20%) in the CC group, 5 (16%) in the VD group, and 9 (30%) in the M group (p = .17). Overall, 17% underwent secondary reduction, including 9 (23%) CC patients, 2 (6.4%) VD patients and 6 (20%) M patients (p = 0.16). Categorical pain scores and analgesic requirements at 1 week did not differ between groups. CONCLUSION: Volar-dorsal splints were associated with less surgery and better radiographical outcomes. Differences were clinically important but not statistically significant because of sample size. Ease of application suggests that volar-dorsal splints may be the best ED immobilization method for distal radius fractures.

013 What is the Role of the History in the Assessment of Patients with Minor Head Injury?
Stiell IG, Greenberg G, Reardon M, Brison R, Clement C, Battram E, MaPhail I, McKnight RD, Schull M, Eisenhauer M, Rowe BH, Cass D, for the CCC Study Group. Division of Emergency Medicine, University of Ottawa, Ottawa, ON.

OBJECTIVES: The history is often ignored in patients with minor head injury. This study measured the accuracy and reliability of specific history findings in minor head injury. METHODS: This prospective cohort study enrolled patients who presented with GCS 13-15 after loss of consciousness, amnesia, or confusion at 10 tertiary care EDs. MDs recorded standardized history findings and in some cases 2nd physicians performed interobserver assessments. Patients underwent CT to determine the outcome, clinically important brain injury. Analyses included univariate association, kappa, sensitivity, specificity, adjusted odds ratio by stepwise logistic regression. RESULTS: We enrolled 3,121 patients with mean age 38.7 years, important brain injury 3.0%, neurological intervention 1.4%. The Table shows % of injury and non-injury patients with the history findings, p-value, kappa coefficient, sensitivity, specificity, odds ratio. CONCLUSIONS: A history of 'age 65', 'dangerous mechanism', 'loss of consciousness >5 mins', 'amnesia >30 mins prior to injury', or 'repeated vomiting' each puts patients at higher risk of important brain injury. These findings are also reliable and should be incorporated into decision rules for the management of minor head injury patients.

014 Protective Equipment Use in In-Line Skating: an Observational Survey.
Rowe BH, Cheung M, Wiebe N, Nykolyshyn K, Belton K, Petruk J, Klassen TP. University of Alberta, Edmonton, AB.

INTRODUCTION: In-line skating is an increasingly common cause of injuries in emergency departments. Some protective equipment (helmets, wrist guards) have been shown to reduce injuries and others are recommended (elbow and knee pads); however, use of these devices varies. This study examined the variations in the use of in-line protective equipment. METHODS: A prospective survey of in-line skaters was performed between 06-08/2001 in an urban Alberta centre. Trained research assistants recorded skaters' demographics (age, gender) and protective equipment wearing patterns (helmets, wrist guards, elbow and knee pads) during a summer period. A random selection of roadways, commuter paths, valley trails, parks, schools and campuses were sampled. Rates are reported with 99% confidence intervals (99% CI). RESULTS: Overall, 615 valid observations were made. The sample's protective unadjusted equipment wearing percentages, from highest to lowest, were: 41% (99% CI: 36, 46) for wrist guards, 19% (99% CI: 15, 22) for knee pads, 15% (99% CI: 11, 18) for helmets, and 7% (99% CI: 4, 9) for elbow pads. Only 3% (99% CI: 1, 4) of observed in-line skaters wore all 4 forms of protective gear. Fifty-four per cent (99% CI: 49, 60) wore no protective gear at all. Fewer males than females used all forms of protective wear; the only difference between genders that was not significant was use of elbow pads. Children were more likely to wear helmets as compared to both adolescents (p < 0.0001) and adults (p = 0.0006). Overall, the models predicting helmet use were different for each age grouping (e.g., youth, adults). DISCUSSION: These results identify large within region variation in the use of protective devices for in-line skating in a large urban area. Injury prevention planners must use these data to adopt interventions that are focused on age and gender groupings. Further work is urgently required to understand these variations and design implementation strategies to correct these disparities.