2003 CAEP/ACMU Scientific Abstracts: 1-20
CAEP Abstracts
CJEM 2003;5(3):179-209
Disclaimer: The large number of abstracts submitted and the short time interval between submission and publication did not permit communication with authors, abstract revision of CJEM editorial review. The following abstracts are presented, unedited, as they were submitted to the CAEP Research Committee. Abstract authors are from the department or division of emergency medicine of their respective universities unless otherwise specified.
abstracts : 001-020 : 021-040 : 041-060 : 061-080 : 081-100
001 Multicenter Prospective Validation of the New Orleans Criteria for CT in Minor Head Injury.
Stiell IG, Clement C, Rowe B, Brison R, Schull M, Wells GA, Greenberg G, Cass D, Holroyd B, Worthington JR, Reardon M, Eisenhauer M, for the CCC Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: The New Orleans Criteria (NOC) for CT in minor head injury were previously derived and validated in a single site cohort of 1,429 patients. This study prospectively and explicitly evaluated the accuracy, reliability, and acceptability of the NOC in multiple sites. METHODS: This prospective cohort study was conducted in 9 tertiary care EDs and enrolled adult minor head injury patients with witnessed loss of consciousness, amnesia, or confusion and a GCS score of 15. More than 350 physicians completed data forms and interpreted the NOC status for patients prior to diagnostic imaging. In some cases 2nd physicians performed interobserver assessments. The outcome standards were 'need for neurological intervention' and 'clinically important brain injury'. Analyses included kappa coefficient, sensitivity, and specificity with 95% CIs. RESULTS: The 1,733 patients enrolled over 30 months had these characteristics: mean age 37.7 (range 16-99), male 68.2%, ambulance arrival 72.1%, clinically important brain injury on CT 5.0%, unimportant injury 2.7%, neurological intervention 0.5%, death 0%. The NOC classified patients for neurological intervention (N = 8) with sensitivity 100% (95% CI 63-100), specificity 12.3% (11-14), and would have required CT for 87.8%. The NOC also classified patients for 87 important brain injuries with sensitivity 100% (95% CI 96-100) and for 48 unimportant brain injuries with sensitivity 95.8% (95% CI 85-98). The kappa value for MD interpretation of the NOC was 0.47 (-0.13 - 1.0). MDs under-estimated the risk in 5.4% and were uncomfortable applying the rule in 11.5%. CONCLUSIONS: The NOC have proven to be very sensitive for identifying important brain injury. Interobserver agreement is fair but specificity is very low. The NOC may be appropriate for minor head injury patients in the U.S. where current CT rates are high but widespread implementation elsewhere would dramatically increase use of CT outside of the U.S. Key words: clinical prediction rule, diagnostic imaging, Computed tomography
002 Multicenter Comparison of GCS and RTS Scores at Scene versus at Trauma Hospital.
Al-Salamah M, McDowell I, Stiell IG, Wells G, Nesbitt L. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
OBJECTIVE: Prehospital triage of trauma patients is challenging. This study evaluated the predictive validity and reliability of the Revised Trauma Score (RTS) and Glasgow Coma Scale (GCS) by comparing scores at scene by paramedics to those made at the ED by physicians. METHODS: This multicenter prospective cohort study was conducted in 20 communities as part of the Ontario Prehospital Advanced Life Support (OPALS) Study, which evaluates the impact of ALS programs. EMS care was provided at both the BLS-D and ALS level. Included were adult trauma patients with ISS >12 and who were treated at 12 regional Level 1 trauma hospitals. We linked ambulance call report data to that from the Ontario Trauma Registry to compare scene and ED evaluations. Analysis included Cronbach's Alpha; Intraclass Correlation Coefficients (ICC); area under the Receiver Operating Characteristic (ROC) curves and Kendal's Tau C Correlation Coefficient (Tc) for predicting survival. Mann-Whitney U-test was performed to test association with ICU admission for survivors. RESULTS: For the 922 patients, scene and ED RTS scores had similar internal consistency with similar patterns of inter-item correlation (Alpha = 0.82, 0.85). Scene and ED GCS scores also had identical inter-item correlations (Alpha = 0.93, 0.93). Interobserver agreement between scene and ED scores was very good both for cases requiring 20-minute transfers (ICC for RTS 0.92, ICC for GCS 0.85) and for those with 15-minute transfers (ICC for RTS 0.96, ICC for GCS 0.88).
|
|||||||||||||||||||||||||
CONCLUSIONS: Paramedic scoring at scene showed good internal consistency and interobserver reliability and are as valid as ED physician scores in predicting survival. Although paramedic scores did not predict ICU admission, timing of the score seemed to be the main factor, rather than the performance of paramedics. Key words: emergency medical services, trauma
003 The Value of History in the Diagnosis of Subarachnoid Hemorrhage for Emergency Department Patients with Acute Headache.
Perry JJ, Stiell IG, Wells GA, Mortensen M, Lesiuk H, Wallace G, Sivilotti M, Kapur A. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: It is unclear which ED patients with an acute headache require investigation for SAH. This study determined the value of history and resolution of headache for predicting SAH. METHODS: This 2-year prospective cohort study was conducted at 3 university tertiary care EDs. Adults with an acute headache peaking within 1 hour and without neurological deficit were enrolled. Excluded were recurrent headaches of similar type, trauma, or previous SAH/brain neoplasm. Emergency physicians completed data forms prior to investigation. The outcome, SAH, was defined as SAH on CT, xanthochromia in the CSF, or the presence of red blood cells in the last tube of CSF with positive cerebral angiography. Patients were asked historical questions and to rate their pain on a scale of 0-10; 0 represented no pain and 10 the worst. Analysis included univariate and multivariate logistic regression. RESULTS: The 589 enrolled patients had a mean age of 42.9 years, were mostly women (60.6%), and had mean peak pain of 8.5 (SD=2.1). Anti-migraine treatment was given to most patients (83.9%); total pain relief was associated with benign headaches versus SAH (p=0.031); however 6.1% of patients with total pain relief had SAH. We compared patients with and without SAH and determined their unadjusted and adjusted odds ratios (** P<0.05):
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CONCLUSIONS: Age >50, neck pain and vomiting were related to the outcome of SAH and should be sought in all patients presenting with an acute headache. Although relief of headache was statistically more common in the benign headaches, it does not rule out SAH. Key words: Subarachnoid hemorrhage
004 The Epidemiology of Acute Myocardial Infarction Emergency Department Presentations in Alberta.
Holroyd BR, Bullard M, Welsh R, Yiannakoulias N, Spooner CH, Rosychuk R, Svenson L, Schopflocher D, and Rowe BH. Division of Emergency Medicine, University of Alberta, Edmonton, AB.
OBJECTIVES: Many patients with acute myocardial infarction (AMI) present to the emergency department (ED) and despite potentially serious consequences, limited information exists about this diagnosis in a defined population. This study examines the epidemiology of AMI presentations to the ED using a unique provincial data set. METHODS: All cases of AMI from all ages presenting to Alberta EDs were eligible for inclusion. Data were derived from the cohort of patients treated in Alberta EDs in all 17 health regions over 1 year (fiscal 00/01). Data were extracted from the Ambulatory Care Classification System (ACCS) database, which contains computerized MD-diagnosed abstracts coded using ICD-9 coding, by medical record nosologists in each hospital. Descriptive statistics and crude presentation rates are presented. RESULTS: AMI (ICD-9 410.x) accounted for 3543 (0.21%) of the 1.7 million annual visits to EDs in Alberta during 1 year; the overall population of the province is 2.8 million. Males presented more commonly (2,471; 70%) than females. Patients > 49 years old comprised 3,028 (85%) of the cases while patients > 69 years old comprised 1505 (42%) of the cases. Presentation rates demonstrated little variation based on weekday presentation or month of the year. The ED presentation rate for AMI was 1.12 / 1,000 persons in rural settings and 0.97 / 1,000 persons in urban settings. In the patients with AMI there were a total of 421 ED visits in the preceding 14 days prior to their AMI presentation. The most common ICD-9 diagnosis for these visits was "Chest Pain" 786.5 (40 visits). CONCLUSION: Myocardial infarction is a serious ED presentation which demonstrates marked gender differences, but fails to demonstrate significant variation across weekday, month or region. However, ED visits immediately preceding an AMI presentation are not infrequent and need further evaluation. Key words: myocardial infarction
005 Characteristics of Patients with Chest Pain Arriving by Ambulance vs Private Transport.
Christenson J, Innes G, Grafstein E, Boychuk B, Wong H, Singer J, Wanger K, Berringer R. Department of Emergency Medicine, St. Paul's Hospital and University of BC, Vancouver, BC.
OBJECTIVES: Approximately 50% of patients with a final diagnosis of AMI arrive to emergency departments by ambulance. Regional policies for the care of ACS often focus on patients transported by ambulance. We hypothesized that patients with chest pain arriving by ambulance are demographically distinct and more likely to have acute coronary syndrome (ACS). METHODS: We compared patients enrolled in a prospective cohort study of 1819 patients presenting with chest pain who arrived by ambulance (AMB) or not (nonAMB). Subjects were consecutive consenting patients > 25 years old, living in British Columbia with complete follow-up at 30-days. Explicitly defined diagnoses were applied for categories of ACS: AMI, definite unstable angina (UA), possible UA, no ACS with an adverse event (AE) and no ACS/AE. We compared presenting demographics, risk factors, diagnostic outcomes and mortality. Differences with 95% confidence intervals were calculated. RESULTS: 564 (31%) of patients arrived by ambulance. The AMB and nonAMB groups were 64.6 vs 55.4 years old (95% CI of difference; 7.7,10.8), were 52% vs 61% male (95% CI of difference; -14,-4%), took a median time of 100 min vs 135 min from symptom onset to ED arrival (95% CI of difference; 42,-6), had an incidence of 54% vs 35% of past AMI or angina (95% CI of difference; 14,23),had a 19% (106/564) vs 11% (135/1255) final outcome of AMI (95% CI of difference; 4,12) and a 8% (43/564) vs 9% (114/1255) incidence of no unstable angina (95% CI of difference;-4,1). CONCLUSIONS: Patients with chest pain who arrive by ambulance are different from those not using ambulance services. They are older, less likely to be male, have shorter duration of symptoms, more likely to have previous coronary disease and more likely to have ACS. However, as many patients with ACS arrive by private means, highlighting the need for regional carepaths to include strategies for patients who do not arrive by ambulance. Key words: acute coronary syndromes, myocardial infarction
006 Understanding and Improving Low Bystander CPR Rates: A Systematic Review of the Literature.
Vaillancourt C, Stiell IG, Wells GA. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: Bystander CPR rates remain low in North America. CPR is a crucial yet weak link of the chain of survival for cardiac arrest. We sought to understand the determinants of bystander CPR and factors associated with successful training. METHODS: For this systematic review, we searched 11 electronic databases, 1 trial registry, and 9 scientific websites. We performed hand searches and contacted 6 international content experts. We reviewed without restriction all communications pertaining to WHO should learn CPR, WHAT should be taught, WHEN to repeat training, WHERE to give CPR instructions, and WHY people lack the motivation to learn and perform CPR. We used standardized forms to review papers for inclusion, quality and data extraction. We grouped publications by category and classified recommendations (A, enough evidence to support to E, enough evidence not to support) using a standardized classification system based on the level of evidence (I-1, meta-analysis to III, expert opinion). RESULTS: We reviewed 2254 articles and selected 370/2254 for complete evaluation. We included 221/370 papers in the systematic review. Differences in study design precluded a meta-analysis of those publications. We classified 22 recommendations. Recommendations with the highest scores (A,I-2) were: 1)Dispatch-assisted CPR instructions; 2)Teaching CPR to family members of cardiac patients; 3) Braslow's self-training video; 4)Maximizing time spent using manikins; and 5)Teaching concepts of ambiguity and diffusion of responsibility. Other examples include: Mass training events (C, II-3), pulse taking by laymen (D,I-2), and CPR using chest compressions alone (E,I-2). CONCLUSIONS: We performed the most extensive systematic review of bystander CPR determinants to date. We evaluated and classified the potential impact of interventions currently believed to improve bystander CPR rates. Our results may help international communities to design the best intervention to improve their bystander CPR rates. Key words: cardiac arrest, resuscitation
007 Do Point-of-Care Troponin-I Assays Reduce ED Length of Stay for Patients with Chest Pain?
Scheirer G, Grabher J, Wood V, Christenson J, Stagg A, Innes G. Department of Emergency Medicine, Royal Alexandra Hospital, Edmonton, AB.
INTRODUCTION: Cardiac markers are used to rule-out myocardial infarction (MI) in the ED. Lab-based testing takes 1-3 hours and may tie up ED stretchers. Point-of-care (POC) tests can be performed rapidly at the bedside, potentially reducing time-to-disposition, enhancing throughput and improving access for waiting patients. Our objective was to determine whether POC troponin testing reduces length of stay (LOS) in ED patients undergoing serial assays. METHODS: This randomized clinical trial was based at Edmonton Royal Alexandra Hospital, Nanaimo Regional General Hospital and Victoria General Hospital. Consenting patients having serial troponin assays were randomized to quantitative LAB testing or qualitative POC testing using the Cardiac Status troponin I test kit (Spectral Diagnostics). Primary outcome was mean ED LOS. Patients were followed for 30-days to document final diagnosis and key outcomes. Test parameters were determined for both modalities. RESULTS: The study sample included 88 POC patients and 71 LAB patients (n=159), with similar baseline characteristics. In the POC group, there were 3 index MIs, 2 late MIs and 1 revascularization (4 of 6 were admitted). In the LAB group there were 4 index MIs and 3 revascularizations (all admitted). The table shows that ED LOS was shorter in the POC group (p=.02). At 30-days there were no significant differences in test utilization or ED re-visit and readmission rates but LAB patients had more revascularization procedures (p=.08).
|
|||||||||||||||||||||||||||||||||||||
CONCLUSIONS: POC testing was associated with shorter LOS. POC tests may be less sensitive and more specific for ACS events occurring within 30 days. Key words: myocardial infarction, troponin, diagnosis
008 The Epidemiology of Angina and Acute Coronary Syndrome Emergency Department Presentations in Alberta.
Holroyd BR, Bullard M, Welsh R, Yiannakoulias N, Spooner CH, Rosychuk R, Svenson L, Schopflocher D, and Rowe BH. Division of Emergency Medicine, University of Alberta, Edmonton, AB.
OBJECTIVES: Patients with angina or acute coronary syndromes (AACS) present to the emergency department (ED) and despite potentially serious consequences and the need for aggressive management, limited information exists about this diagnosis in a defined population. This study examines the epidemiology of AACS presentations to the ED using a unique provincial data set. METHODS: All cases of AACS from all ages presenting to Alberta EDs were eligible for inclusion. Data were derived from the cohort of ED patients in all 17 provincial health regions over 1 year (fiscal 00/01). Data were extracted from the Ambulatory Care Classification System (ACCS) database, which contains computerized MD-diagnosed abstracts coded using ICD-9 coding, by medical record nosologists in each hospital. Descriptive statistics and crude presentation rates are presented. RESULTS: AACS (ICD-9 411.1, 411.8, 413.x, 410.7) accounted for 11,023 (0.65%) of the 1.7 million annual visits to EDs in Alberta during 1 year; the overall population of the province is 2.8 million. Males presented more commonly (6455; 59%) than females. Patients > 49 years old comprised 9,545 (87%) of the cases while patients > 69 years old comprised 4,956 (45%) of the cases. Presentation rates demonstrated some variation by day of the week but not by month. The ED presentation rate for AACS was 4.20 / 1,000 persons in rural settings and 3.39 / 1,000 persons in urban settings. In the patients with AACS there were a total of 1171 ED visits in the preceding 14 days prior to their AACS presentation. The most common ICD-9 diagnosis for these visits was "Chest Pain" 786.5 (146 visits). CONCLUSION: Angina and acute coronary syndromes are potentially serious and common ED presentations. They demonstrate limited monthly variation but considerable variation by day of the week and region. ED presentations immediately preceding an ED visit for AACS are commonly undifferentiated, not infrequent and warrant further evaluation. Key words: acute coronary syndromes
009 Diagnostic Value of C-Reactive Protein for ACS in ED Patients with Chest Pain.
Christenson J, Hunte G, Boychuk B, Rosenberg F, Innes G, Grafstein E, Wong H, Singer J. Department of Emergency Medicine, St. Paul's Hospital and University of BC, Vancouver, BC..
OBJECTIVES: Inflammatory markers are predictive of future events in patients with established acute coronary syndromes (ACS) but it is unclear how they can be used as diagnostic tests in patients presenting with chest pain. We postulated that C-reactive protein on arrival to the ED in patients with chest pain would predict the diagnosis of ACS. METHODS: A convenience sample of patients prospectively enrolled to develop an early discharge decision rule were followed for 30-days to determine outcomes of AMI or definite unstable angina. High-sensitivity CRP was determined on the first blood specimen. Sensitivity and specificity were determined for different CRP diagnostic thresholds. RESULTS: Of 214 patients, 58 had a 30-day outcome of AMI or definite unstable angina. The area under the ROC curve was 0.55 indicating poor overall diagnostic accuracy. For a diagnostic threshold of 1 IU/L, the sensitivity was 41/58 = 71% (95% CI: 59%, 82%) with a specificity of 41/156 = 26% (95% CI: 19%,33%). For a diagnostic threshold of 5 IU/L, the sensitivity was 15/58 = 26% (95% CI: 15%, 37%) with a specificity of 104/156 = 67% (95% CI: 59%,74%). For a diagnostic threshold of 10 IU/L, the sensitivity was 5/58 = 9% (95% CI: 1%, 16%) with a specificity was 124/156 = 79% (95% CI: 73%, 86%). CONCLUSION: CRP does not appear to provide diagnostic value for ACS in an undifferentiated population of patients with chest pain presenting to the ED. A larger cohort is needed to confirm the diagnostic accuracy with confidence. Further work may help to clarify whether CRP provides added value to ECG and other markers in specific subsets. Key words: myocardial infarction, C reactive protein, diagnosis
010 Determination of Sample Size Parameters for Community Intervention Cluster Trials in Cardiac Arrest.
Vaillancourt C, Wells GA, Stiell IG, for the OPALS Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: Community interventions to improve survival from cardiac arrest are becoming more common. Individuals belonging to a community share characteristics that make them non-independent. We sought to determine the intra-class correlation and inflation factor for sample size calculation of community intervention cluster trials in cardiac arrest. METHODS: We analyzed data prospectively collected within the Ontario Prehospital Advanced Life Support Study. This study has the largest population-based cohort of adult out-of-hospital cardiac arrests from 20 communities and 11 base hospitals with BLS-D and ALS paramedics. We used one-way analysis of variance to obtain the mean square error among (MSC) and between (MSW) base hospitals for cardiac arrest survival. We calculated the intra-class correlation (p) using p=(MSC-MSW)/(MSC+(m-1) MSW) where m is the cluster size, the inflation factor (IF) using IF=(1+(m-1)p), and the required sample size (n) to find a 2% absolute difference in cardiac arrest survival (80% power, two-sided 5% alpha error). The number of clusters required (k) is given by k=2n/m. RESULTS: From 1995 to 2000, there were 7,707 consecutive cardiac arrest cases: mean age 68.9, 67% male, 37% VF/VT, 16% bystander CPR, and 4.0% survival to discharge. Intra-class correlation for cardiac arrest survival=0.0019 and IF=2.3. In other words, the sample size calculation for a community intervention needs to be multiplied by 2.3 to account for the lack of independence of individuals within communities. For example, we would need 8,000 cardiac arrests in 12 communities to find a 2% difference in survival from cardiac arrest. CONCLUSION: It is essential to know the intra-class correlation factor to calculate the required sample size for community intervention cluster trials. This factor is often estimated and had not been published for cardiac arrest research before. Our results can have major impact on the design and analysis of community intervention cardiac arrest research. Key words: cardiac arrest, resuscitation
011 OPALS Study Phase III: What is the Impact of Advanced Life Support on Out-of-Hospital Cardiac Arrest?
Stiell IG, Wells GA, Spaite DW, Nesbitt L, Cousineau D, De Maio VJ, Campeau T, Dagnone E, Nichol G, Field BJ, Beaudoin T, Brisson D, for the OPALS Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: The Ontario Prehospital Advanced Life Support (OPALS) Study is designed to systematically evaluate the effectiveness of EMS interventions for critically ill and Injured patients. OPALS Phase III tested the incremental impact on out-of-hospital cardiac arrest survival of adding an ALS program to a multicenter BLS-D EMS system that had previously optimized defibrillation (Phase II JAMA 1999). METHODS: This multicenter before-after controlled clinical trial was conducted in 17 communities (population 20,000 to 750,000) and enrolled all adult out-of-hospital cardiac arrest patients during the 12-month BLS-D rapid defibrillation phase and the subsequent 36-month ALS phase. Paramedics were fully trained to ALS standards including endotracheal intubation and administration of IV drugs. The primary outcome was survival to hospital discharge. Chi-square and logistic regression analyses were performed. RESULTS: The 5,637 patients enrolled during the BLS-D (N=1,391) and ALS (N=4,246) phases were well matched and had these characteristics: mean age 69.2 (range 16-102), male 66.7%, witnessed 51.7%, bystander CPR 14.7%, initial rhythm VF/VT 32.3%, defibrillator at scene <8 minutes 93.3%. During the ALS phase, success rates were intubation 93.7% and IV insertion 89.0%. From the BLS-D to the ALS phase, the admission rate increased (10.9% vs 14.6%; P<.001) but survival did not change (5.0% vs 5.1%; P=.82) for all rhythms combined. Logistic regression also found the odds ratio for ALS to be non-significant (0.91; 95% CI 0.6-1.3). There was no survival improvement for any subgroup including cases witnessed by bystander (7.1% vs 6.8%; P=.80) or by EMS (13.5% vs 16.6%; P=.41); with rhythm VF/VT (12.9% vs 13.2%; P=.87) or PEA (1.4% vs 2.4%; P=.27). CONCLUSIONS: The OPALS Study is the largest multicenter controlled trial of out-of-hospital cardiac arrest. The addition of prehospital ALS interventions did not improve patient survival in a previously optimized rapid defibrillation EMS system. Key words: emergency medical services, cardiac arrest, resuscitation
012 Emergency Waiting Room Care: Are Some of Our Emergency Patients Being Poorly Cared for?
Grafstein EJ, Innes GD, Stenstrom R, Christenson J, Hunte G. Department of Emergency Medicine, Providence Health Care, Vancouver, BC.
BACKGROUND: ED overcrowding creates a situation where some ED patients are triaged to and cared for in the waiting room/hallway (WR). OBJECTIVE: To compare baseline characteristics, ED utilization and adverse outcomes in patients with WR care versus those triaged to an acute care (AC) bed. Our hypothesis was that patients managed in the WR would have more adverse events. METHODS: A retrospective cohort study at St. Paul's Hospital, a Canadian, tertiary center with 45,000 annual visits. By linking physician order entry and emergency department administrative databases, we reviewed all patients who required an AC bed from November 2000 to November 2002. We used Chi-square analysis and t-tests to compare patients cared for exclusively in WR versus those whose entire care was in an AC bed. RESULTS: 34,226 patients required an AC bed. Of these, 5,355 (15.6%) had their entire care in WR, 5,341 (15.6%) were triaged to WR and later transferred to an AC bed, and 23,530 (68.7%) were triaged directly to an AC bed on arrival. WR patients were younger, lower acuity, and seen less quickly than patients in a bed. WR patients had shorter ED length of stay (LOS) and left without being seen (WBS) or against medical advice (AMA) more often. WR patients were more likely to return requiring hospitalization within 7 days (revisit-admit) than AC bed patients.
CONCLUSIONS: Triage nurses accurately identified patients who could safely be managed in the WR. High left WBS/AMA rates suggest lower patient satisfaction but we failed to show clinically important differences in revisit-admit rates. Key words: Overcrowding, quality
|
|||||||||||||||||||||||||||||||||||||||||
013 Multicenter Controlled Clinical Trial to Evaluate the Impact of Advanced Life Support on Out-of-Hospital Chest Pain Patients.
Stiell IG, Nesbitt L, Wells GA, Beaudoin T, Spaite DW, Brisson D, Nichol G, Field BJ, Lyver MB, Munkley DP, Luinstra LG, Cousineau D, for the OPALS Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: There is little published evidence regarding the optimal EMS management of chest pain. Our study tested the impact of advanced life support (ALS) EMS programs on chest pain patient outcomes. METHODS: This multicenter before-after controlled clinical trial was conducted in 17 communities (population 20,000 to 750,000) as part of the Ontario Prehospital Advanced Life Support (OPALS) Study, which evaluates the impact of EMS programs for multiple conditions. During the before phase, care was provided at the BLS-D level. During the after phase, ALS providers performed endotracheal intubation and administered IV drugs. Data were collected from ambulance reports, centralized dispatch data, ED records, and in-hospital records. Chi-square and Student's t-test analyses were performed. RESULTS: The 4,601 patients enrolled during the two 9-month BLS and ALS phases were well matched for clinical and demographic features and had these characteristics: mean age 66.6 (17-102), female 50.4%, EMS status 'severe/life threatening' 48.7%, ICD-9 final diagnoses: chest pain NYD 17.7%, MI 17.0%, other non-cardiac 15.1%, unstable angina 14.6%, stable angina 9.4%, G.I. 8.6%, respiratory 5.9%, dysrhythmias 5.8%. During the ALS phase, patients received these EMS interventions: intubation 0.1%, IV fluid bolus 5.1%, SL NTG 64.4%, SL ASA 49.8%, IV morphine 6.1%, IV furosemide 4.5%, IV adenosine 0.8%, IV lidocaine 0.3%, IV atropine 0.3%. There was a 64.7% relative reduction in the primary outcome, overall mortality, from the BLS to the ALS phase (5.1% vs 2.8%; P<.001). Other outcomes also showed improvement from BLS to ALS phase: EMS-judged improved (20.1% vs 50.9%; P<.0001); admitted (76.7% vs 48.9%; P<.001); discharged to home (65.8% vs 68.7%; P<.01). A large mortality reduction was seen in the MI subgroup (19% vs 10%; P<.001). CONCLUSIONS: This is the largest controlled trial of out-of-hospital chest pain patients and clearly shows important benefit from ALS programs for mortality and other outcomes. Key words: emergency medical services, chest pain
014 Reasons Why Patients Leave Without Being Seen from the ED.
Channan P, Bullard M, Alibhai A, Saunders D, Rowe BH. Division of Emergency Medicine, University of Alberta, Edmonton, AB.
INTRODUCTION: Recently, patients leaving without being seen (LWBS) by a physician have become an important emergency department (ED) concern; however, research into this problem has been limited. The purpose of the present study was to determine the acuity level, reasons, and outcomes of patients who LWBS at two EDs. METHODS: This prospective study, contacted all patients who LWBS at two Canadian EDs during six 7-day sampling periods in the summer of 2002. Following medical record review, a telephone questionnaire with the patient or guardian was completed up to 14 days after the original visit or 6 attempts. RESULTS: 8531 patients registered and 395(4.6%) LWBS during the sampling periods. Of these, 361 were confirmed LWBS cases; 46% were female and the mean age was 37. An urgent triage score was assigned to 26% of these LWBS. Median patient arrival in the hours before and during arrival was 11 (IQR: 9, 13) and median delay to seeing an MD for a triage-matched control was 79 minutes (IQR: 44, 135). Follow-up contact was made with 261(72%) and 245(94%) agreed to participate. The most common reason for LWBS in adults was "fed up with waiting" (49%) and in children was "feeling better"(44%). Overall, 60% of LWBS cases sought alternative medical attention for their symptoms; of the patients who did not seek any medical attention, 28 (29%) were triaged as urgent. Six patients were hospitalized and two of those required urgent surgery after seeking subsequent medical attention. One patient who did not follow up with a physician died our days after ED registration. CONCLUSIONS: The majority of LWBS results from impatience during periods of peak ED volumes. Many patients seek alternative medical care within one week of their presentation and, while rare, complications do occur. While most ED LWBS cases are minor and low-risk ailments, further research is required to determine methods of reducing LWBS, especially "high risk" patients. Key words: quality, outcomes
015 Inter-Rater Reliabilities of the Emergency Severity Index (ESI) vs. the Canadian Triage Acuity Scale (CTAS): a Randomized Controlled Trial
Worster A, Gilboy N, Fernandes C, Eitel D, Eva K, Geisler R, Tanabe P. Department of Emergency Medicine, Hamilton Health Sciences, Hamilton, ON
INTRODUCTION: The Emergency Severity Index (ESI v. 2) has been developed as a triage acuity scale for use in the emergency department (ED). This five-level tool rates patients from level 1 (most serious) to level 5 (least serious). Unlike other triage acuity scales such as the Canadian Acuity Triage Scale (CTAS), the ESI combines both the level of acuity and the patients clinical resource needs into a single measure. It has been proven to be reliable and valid in several U.S. ED settings but has yet to be evaluated in a Canadian ED setting or compared with CTAS. The objective of this study was to compare the inter-rater reliability of CTAS and ESI v. 2. METHODS: Ten triage nurses from four urban EDs were randomly assigned to one of two groups for three hours of CTAS refresher or ESI introductory triage training. Following their training, the nurses in each group independently and in a blinded fashion assigned triage scores to 200 prospectively collected, local ED cases. RESULTS: There were no significant differences found between the two groups with respect to: age (p=0.053); years in the ED (p=0.13); hours of prior CTAS training (p=0.57); or years experience in triage (p=0.61). An intraclass correlation to measure the inter-rater reliability of the CTAS group was 0.9079 (0.897, 0.987) and not significantly different from that of the ESI 0.8926 (0.882, 0.985). An inter-test G-study performed on the variance components derived from an ANOVA revealed G (5) = 0.896 (0.817, 0.992). CONCLUSIONS: After just three hours of introductory training, experienced triage nurses were able to perform triage assessments using ESI v. 2 with the same inter-rater reliability as those with experience and refresher training in CTAS. This comparison study also shows that the triage scores assigned using CTAS generalize to those assigned using ESI v. 2. Key words: triage
016 Evaluation of Triage Nurse Satisfaction with Training and Use of an Electronic Triage Tool.
Bullard MJ, Meurer D, Pratt S, Colman I, Holroyd BR, Rowe BH. Division of Emergency Medicine, University of Alberta, Edmonton, AB.
OBJECTIVES: The Canadian Triage and Acuity Scale (CTAS) is a nationally recognized triage standard; however, application of CTAS in busy EDs by multiple users reduces reliability and limits validity. This study was designed to determine if an electronic, complaint-based triage tool (eTRIAGE) could be easily learned and rated satisfactory by nurses in a busy urban ED. METHODS: 9 volunteer triage nurses each received 2 hours of didactic teaching, 2 hours of mentoring during their first eTRIAGE shift, and access to a free time training module. A satisfaction questionnaire was completed at the end of each shift (using a 7-point Likert scale). All eTRIAGE shift data were captured in a database for analysis with traditional paper triage interactions randomly captured for time comparison with eTRIAGE encounters. Users were categorized based on self-reported PC experience into novice (NOV) or experienced (EXP) computer users. RESULTS: The study nurses were experienced (median age 45) with 24 (IQR: 21,25) years of nursing. Overall, 2122 eTRIAGE encounters were recorded over 67 shifts; only 112 (5.2%) encounters had a nurse override the eTRIAGE score. EXP users rated eTRIAGE more favorably than NOV users with respect their: comfort using (5.88 vs 4.76), ease (3.96 vs 3.05), helpfulness (3.76 vs 3.18), interference with patient interaction (2.60 vs 4.08), and increased triage time (4.44 vs 5.21). Measured triage times (minutes/patient) were similar for paper-based (2.45m) and eTRIAGE (EXP = 2.42m vs. NOV = 2.85m). CONCLUSIONS: eTRIAGE was easily learned even by novice users and speed was at least comparable to paper-based triaging. The tool also satisfied nursing perceptions of patient acuity, as 5% of encounters resulted in an override of the eTRIAGE score. The database provides search and reporting capabilities. If ongoing studies confirm both reliability and validity, templated electronic triage should be adopted in place of paper-based triage relying on nurse memory and experience. Key words: triage
017 The Predictive Validity of the Canadian Triage and Acuity Scale (CTAS).
Stenstrom R, Grafstein E, Innes G, Christenson J. Department of Emergency Medicine, St Paul's Hospital and University of BC, Vancouver, BC.
INTRODUCTION: The Canadian Triage and Acuity Scale (CTAS) has been shown, in various emergency department (ED) studies, to possess good reliability. No studies to date have assessed the association between CTAS level and outcomes such as utilization and patient disposition. METHODS: Objective: To establish the predictive validity of the CTAS in relation to patient disposition, ED length of stay (LOS), hospital LOS (for admitted patients) and utilization (lab tests and imaging) based on a large administrative dataset. This retrospective cohort study was conducted at St. Paul's Hospital, a Canadian tertiary care institution with over 45,000 visits yearly. On arrival to the ED, all patients are assigned a CTAS level (from 1 to 5) by the triage nurse. ED LOS, hospital LOS, and patient disposition, were assessed for over 70,000 patient visits between October 2000 and October 2002. As a proxy for utilization, the proportion of patients having any imaging (CT, ultrasound, X-ray) or a CBC was established for each CTAS level. RESULTS: Multivariate ANOVA for mean ED LOS and hospital LOS by CTAS level rendered statistically significant results for ED LOS. Chi-square and trend analyses were conducted for proportion of patients admitted, having a CBC, or any imaging were significant (Bonferonni correction for multiple comparisons used).
|
|||||||||||||||||||||||||||||||||||||||||||
CONCLUSIONS: Based on this sample of over 70,000 patient visits the CTAS has excellent predictive validity for clinical and utilization outcomes. Key words: triage
018 Does HIV Co-Morbidity Increase the Utilization of the Emergency Department?
Grafstein EJ, Stenstrom R, Buchan V, Francis M. Department of Emergency Medicine, Providence Health Care, Vancouver, BC.
BACKGROUND: The prevalence of human immunodeficiency virus (HIV) in the Downtown Eastside of Vancouver is one of the highest in Canada. We hypothesize that HIV positive patients are sicker and utilize significantly greater resources within the ED. OBJECTIVE: To compare the ED length of stay (LOS), admit rate and frequency of procedures in HIV patients versus non-HIV patients. METHODS: A retrospective cohort study between June-September 2001. HIV positive and HIV negative patients were identified in a previous study using a self-reporting questionnaire. Baseline data including age, sex, and mode of arrival to the emergency department were obtained for each patient through linkage with the ED administrative database (NERD). Self-reported HIV negative patients from the same period were used as controls. Surrogate markers for emergency department utilization - ED LOS, admission rate and procedures performed were measured. Comparisons were made using Student's t-tests for continuous and Chi-square analysis for discrete variables. RESULTS: There were 501 distinct HIV patients with 821 visits and 752 controls with 1144 visits during the study period. 354/1144 of the controls had Canadian Triage and Acuity Scale (CTAS) levels 1-3 versus 359/821 of the HIV patients. 157 of the HIV patients were admitted versus 76 of the controls. The number of procedures performed was not different between the two groups. ED LOS for discharged patients was significantly shorter in the control group.
|
|||||||||||||||||||||||||||||||||
CONCLUSIONS: HIV patients tend to be sicker and spend more time in the department. The number of procedures performed does not account for this. Co-morbid illnesses such as HIV have an impact on the utilization of emergency departments. Key words: HIV, utilization
019 Emergency Department Presentations of Atrial Fibrillation (AF) In Alberta, Canada.
Diner BM, Yiannakoulias N, Holroyd BR, Bullard M, Spooner CH, Rosychuk R, Svenson L, Schopflocher D, Rowe BH. Division of Emergency Medicine, University of Alberta, Edmonton, AB.
OBJECTIVES: Atrial fibrillation (AF) is the most common sustained arrhythmia and the most common arrhythmia seen in the emergency department (ED); however, the burden of acute AF on EDs is virtually unknown. This study examines the epidemiology of AF presentations to the ED using a provincial database. METHODS: All patients presenting to Alberta EDs were eligible for inclusion. Data were derived from a population of patients treated at Alberta EDs in 17 health regions over 1 year (fiscal 00/01). Data were extracted from computerized abstracts coded similarly across all regional EDs contained within the Ambulatory Care Classification System (ACCS) database. Diagnostic categories are recorded using ICD-9coding by medical record nosologists in each hospital and represented the primary physician discharge diagnostic code. Descriptive statistics, crude and adjusted presentation rates are reported. RESULTS: Over 1 year, 1.7 million ED visits were recorded; 3276 (0.2%) patients aged > 17 made 5023 presentations to the ED during this period with a diagnosis of AF. Males (1691; 52%) and females were similarly represented. The elderly (>60 years) accounted for 76% of all cases of AF. Daily variation was high with peak presentations occurring at 5 PM. The provincial presentation rate was 1.0/1000persons with a prevalence of 1.5/1000. Patients experienced frequent relapses of AF in the first 30 days and one year (16.4% and 32.8%, respectively). The proportion experiencing TIA or AMI within the 365 days after an AF presentation was 1.5% and 1.4% respectively; the proportion of patients experiencing CHF was 9.6%. Approximately 60% of all patients were discharged and 9.7% were admitted to ICU. CONCLUSIONS: Discharge for acute AF is common, reflecting the outpatient treatment of this disease in Canada. Recurrence and complications are also frequently observed; however, TIAs are less common than reported elsewhere. Further research is needed to understand the role of outpatient treatment of AF on these outcomes. Key words: atrial fibrillation, dysrhythmia
020 Determination of Accurate Out-of-Hospital Cardiac Arrest Location in 20 Communities.
Vaillancourt C, De Maio VJ, Stiell IG, Wells GA, Nesbitt L, Martin MT, Cousineau D, for the OPALS Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON.
INTRODUCTION: Many communities are implementing CPR training and public access defibrillation programs with little information on cardiac arrest location. We sought to determine accurate out-of-hospital cardiac arrest location in 20 communities. METHODS: Prospective cohort. The OPALS study is the largest population-based cohort of adult out-of-hospital cardiac arrest cases in 20 communities with BLS-D and ALS paramedics. We merged the OPALS and provincial dispatch databases using a unique identifier to obtain addresses where cardiac arrest occurred. Addresses were then retrieved from the municipal property assessment corporation database to obtain accurate description of cardiac arrest location. Cardiac arrest location was classified within 26 predetermined categories. Missing information was submitted to base hospitals for review. Data was analyzed using descriptive statistics. RESULTS: From 1995 to 2000, there were 7,707 consecutive cardiac arrest cases. Mean age 68.9, 67% male, 37% VF/VT, 49% witnessed, 16% bystander CPR, and 4.0% survival to discharge. CONCLUSION: This is the largest review of cardiac arrest location ever conducted. Most cardiac arrests occur in private locations (84.7%) compared to public places (15.3%). Communities should review locations of their cardiac arrest when designing CPR training and public access defibrillation programs. Key words: Cardiac arrest
|
|||||||||||||||||||||||||||||||||||||||||||||
Search
Downloads
-
219.53 KB
