147 GLASS Rule: Automotive Glass Left IntactAssociated With Safe Cervical Spine

Sochor MR, Deflorio P, Singal B/University of Michigan, Ann Arbor, MI

Background: Evaluation of the motor-vehicle collision (MVC) victim’s cervicalspine is of critical importance. Despite the implementation of validated clinicaldecisionmaking rules (such as the NEXUS criteria), many MVC victims undergounnecessary c-spine evaluations, which involve significant cost, discomfort, andradiation exposure.

Study Objectives: The ability of a single field test to exclude c-spine injury wasassessed. It is hypothesized that if the forces involved in an MVC are not sufficient todamage any of the automobile’s windows (aka greenhouse) then these forces wouldnot cause significant c-spine injury, defined as an Abbreviated Injury Score (AIS) of 2or greater.

Methods: Using a nested retrospective cohort format, data was studied from theNational Highway Traffic Safety Administration’s National Automotive SamplingSystem’s Crashworthiness Data System from 1993-2003. Inclusion Criteria were asfollows:

- occupant age 16-60 years- front seat passengers only with seat belts worn- no evidence of airbag deployment- all windows in raised positionThe incidence of serious c-spine injuries (AIS � 2) in subjects with intact and

non-intact greenhouses was then compared.Results: A total of 95,365 front seat occupants were examined, with 646 excluded

due to unknown air bag status. Of that total, 23,246 met enrollment criteria and werestudied. Of those, 12,231 had non-intact greenhouses, and 215 of these occupantshad significant c-spine injuries. Of the remaining 11,015 occupants in MVC’s withintact greenhouses, 21 occupants had serious c-spine injury. Sensitivity was 0.91(95% CI 0.87-0.94), specificity was 0.48 (95% CI 0.47-0.48), and negativepredictive value (NPV) was 99.8% (95% CI 99.7-99.9). It is assumed the sensitivityand NPV would actually be higher as independent review of the exceptions to therule revealed that 50% of the exceptions had evidence of glass breakage and wereimproperly coded in terms of the vehicle glass breakage. However, this fact is notreflected in the results as the 10,994 cases which follow the rule were not reviewed.

Conclusions: Using an intact greenhouse to predict the absence of significantc-spine injury may prove useful; this study’s NPV equals that of the NEXUS criteria.This rule may provide significant patient comfort (lack of c-collar application in pre-hospital arena) and a time and cost savings in the emergency department (no need forradiologic evaluation of these patients). Based on these findings a prospectivevalidation of the GLASS rule is needed.

148 Can We Predict What Objectives Off-ServiceResidents Have for Their Emergency MedicineRotations?

Tenn-Lyn NA, LeBlanc VR, Bandiera GW/University Health Network, Toronto,Ontario, Canada; University of Toronto, Toronto, Ontario, Canada

Study Objectives: To compare objectives identified by off-service residents withaccepted core objectives and to determine the degree to which objectives can bepredicted for a) off-service residents as a group, and b) individual off-service residentswhen their home specialty is known.

Methods: At the start of their emergency medicine (EM) rotation in an academicemergency department (ED), residents were surveyed, identifying their specialty and10 objectives for their 4-9 week rotation. Objectives were tabulated using publishedEM objectives. Internal consistency (IC) and inter-class correlation coefficient (ICC)were calculated as measures of ability to predict resident objectives as a group and asindividuals, respectively.

Results: Sixty-three forms were collected over 2 years. A total of 515 objectives(100 different objectives) were identified. Seventy-one (71%) of their objectives werein common with Taylor’s 155 objectives. Residents identified 29 additionalobjectives. Eight (26%) involved the approach to undifferentiated patient problems.The group IC was 0.913; the ICCC was 0.143. No substantial difference was foundwhen only objectives identified by residents were analyzed. Analysis by specialtyrevealed an ICC value of 0.207.

Conclusion: Using standard guidelines alone would miss over 25% of objectivesidentified by EM off-service residents. Identification of process-oriented objectivesnot captured by standard guidelines suggests resident insight into the undifferentiatednature of ED patient problems. Calculation of association measures reveals that

residents have varying objectives for their EM rotations, objectives which arepredictable for residents as a group, but not for individuals. Additionally, there is nosubstantial improvement in the ability to predict if the resident’s home specialty isknown. These findings emphasize the need for learner-centered objective setting.Implications of these findings for practice include the need to avoid drawingconclusions about an individual resident’s objectives based on knowledge of theirhome specialty alone, and to develop educational tools to encourage and facilitatelearner-centered, self-directed objective setting for off-service residents in emergencymedicine.

149 Effect of Advanced Cardiac Life Support Courseson General Practitioners Working in EDs in Iran

Khalaj S, Nejati A, Moharari RS, Narenjbon SA, Abbasian A/University of Tehran,Tehran, Iran (Islamic Republic of)

Study Objective: In order to determine the effectiveness of Advanced CardiacLife Support (ACLS) course, 82 general practitioners who worked in emergencydepartments of Tehran hospitals were evaluated before and 3 months aftercompleting the course.

Methods: Eighty-two general practitioners who worked in emergencydepartments of hospitals owned by Tamin Ejtemaei Company (a big insurancecompany in Iran which owns many big hospitals) were evaluated by an ObjectiveStructured Clinical Exam (OSCE) before starting an ACLS course. The examincluded the standard ACLS checklist. Three months after completing a 3-day ACLScourse administered by 4 emergency physicians and an anesthesiologist (all coursetutors were faculty members of Tehran University of Medical Sciences), anotherOSCE exam was arranged and the checklist was tested again. The courses and examswere sponsored by the educational center of Tamin Ejtemaei Company.

Results: Less than half of the trainees were men (41%) with mean age of 35.1 �5.1. In the first exam, the mean score was 10.1 � 3.1, conversely related to the meanage of physicians. The highest score was 22. In the second exam (administered threemonths after completing the course), the mean score was 26.3 � 4.3 and the highestscore rose to 29.

Conclusion: The ACLS course was effective in improving the performance ofthose who completed it and all trainees were satisfied. In our country, emergencymedicine is a new specialty and we have less than 30 academically trained emergencyphysicians. Therefore, almost all of our emergency departments are run by generalpractitioners. So it seems that more ACLS courses are needed to update theinformation and skills of those general practitioners who work in emergencydepartments.

150 Standardized Patients or Patient Simulators?Reality Perception In Emergency MedicalEducation

Brendel R, de Best J, de With C, de Vries G/Regional Ambulance ServicesUtrecht, Bilthoven, The Netherlands; Hogeschool Leiden, Leiden, TheNetherlands; University of Utrecht, Utrecht, The Netherlands

Study Objectives: During simulation training and assessments of Dutchparamedics standardized patients and/or patient simulators are used. A standardizedpatient is an actor representing syndromes. A patient simulator is an advancedcomputer-controlled dummy permitting parameters to vary, depending on giventreatment. To demonstrate whether there is a difference in reality perception duringtraining or assessments between using just a patient simulator and using astandardized patient in combination with a patient simulator.

Methods: During a 5-week period participants, performing a yearly assessment(n�25), were placed at random in group 1 (with patient simulator) or group 2(standardized patient with patient simulator). All participants got the same case. Afterthe assessment every participant filled in a questionnaire concerning reality perceptionof the simulation. The assessors filled in a questionnaire concerning the noticeableaspects of reality perception besides the standard score list.

Results: The total score of group 1 (patient simulator) and group 2 (standardizedpatient with patient simulator) were compared. With an independent t-test asignificant difference was found between the total score of participants in group 1(M� 3.10; SD � .691) and group 2 (M�3.96; SD � .804); t (23) � �2.852, p �.009. The total score of assessors showed a significant difference as well, which wascalculated with a Mann-Whitney-U-test. (Z � � 2,484; p � .013).

Conclusion: Training and assessment using the combination of a standardizedpatient with a patient simulator seems to be preferred above the use of just a patient

ICEM 2008 Scientific Abstract Program

516 Annals of Emergency Medicine Volume , . : April

simulator. The reality perception using the combination is higher. Because it concernsa small sample, the authors intend to continue this research.

151 Development of a Core Physiology Curriculum forPostgraduate Training in UK Emergency MedicineUsing a National Delphi Model

Kilroy DA, Driscoll PA/College of Emergency Medicine, London, United Kingdom

Study Objectives: To devise and develop a national consensus-based physiologycurriculum for postgraduate training in UK emergency medicine.

Methods: A 4-round iterated Delphi study was undertaken between May andSeptember 2007. One hundred-sixty fellows of the UK College of EmergencyMedicine were selected by random sampling from the national College database andinvited to participate. From the initial respondent pool, smaller cohorts ofparticipants undertook parallel Delphi exercises to develop subsections of thephysiology curriculum. Curricular content was developed based upon respondents’opinions of the usefulness of long-listed items in daily clinical practice in theemergency department. Feedback was based upon a 4-part Likert scale coupled withfree text analysis. Content was refined using a pre-determined cutoff value for Likertagreement as each Delphi round was completed. The final curricular document wasarranged into systems-based sections and was mapped to the College Membershipmultiple-choice question bank.

Results: One hundred-fifteen of the invited 160 fellows participated. From aninitial list of over 850 discrete items of physiology, a consensus-based curriculum of320 focused aspects of relevant physiological knowledge was developed. Thecompleted document is now used both as a Web-based learning resource for traineesand as a content guide for multiple-choice question development within the College.

Conclusion: Using a national consensus technique facilitates rationalisation ofknowledge in this and other key basic sciences. Robust curricular development toolsare of value both as learning guides and as a platform from which to develop relevantassessment materials.

152 Resident Productivity: Does Shift Length Matter?

Jeanmonod D, Jeanmonod R, Ngiam R/Albany Medical College, Albany, NY

Background: It has been established in the emergency medicine (EM) literaturethat residents are able to evaluate more patients per hour as they progress throughtraining. Shift length has been assessed with regard to quality of life issues. It isunknown if shift length has any influence on patients evaluated per hour by residents.

Study Objectives: To assess whether there is a difference in number of patientsevaluated per hour by second year residents as a function of shift length.

Methods: This is a retrospective chart review of patients evaluated in theemergency department by second year residents in a 65,000 volume tertiary carecenter. The study protocol was reviewed by the IRB and found to be exempt. Secondyear resident shifts of 12 and 9 hours in length were included. The 9 hour shiftsprovide 1 hour overlap, such that three 9 hour shifts provide 24 hours of residentcoverage. The 12 hour shifts provide no overlap and also result in 24 hours ofresident coverage. No resident working a 9 hour shift picked up a new patient after 8hours during the study period. Second year shifts being covered by off-servicerotators, first year residents, and third year residents were excluded. Shifts on weeklyconference day were excluded. Patient turnaround time (ie, time to disposition) wasnot assessed. Data were collected by review of the computerized clinical tracker whichprovides a time line of patient registration and physician assignment for all patientsseen in the emergency department (ED). Data were checked against the computerizedmedical record to verify the physician who dictated the chart for each patient as wellas against the residents’ monthly schedule. A patient was determined as having beenevaluated by a resident if the resident initiated care on the patient and dictated thehistory, physical exam, and plan. Data were assessed for normality and analyzed usingtwo-tailed T test.

Results: 193 nine hour shifts and 90 twelve hour shifts met our inclusion criteria.Residents working 12 hour shifts evaluated an average of 1.06 patients per hour, andresidents working 9 hour shifts evaluated an average of 1.15 patients per hour (95%CI 0.031 - 0.151). In an ED with 96 to 120 hours of resident coverage per day, thisresults in 9 to 11 additional patients seen by residents working 9 hour shifts. In ourdepartment with 9 ED months in the second year of residency, this results in a totalof about 180 additional patient encounters per resident during that year.

Conclusions: Shorter shift lengths appear to result in more patients evaluated perhour by second year residents and an increase in patient encounters.

153 Effectiveness of Medical Simulation on Knowledgein Septic Shock Management During Pre-ClinicalMedical Training

Daniel-Underwood L, Van Ginkel C, Lee D, Wong M, Dizaiy S, Fry-Bowers EK,Denmark TK, Nguyen HB/Loma Linda University, Loma Linda, CA

Background: Current advances in the early management of septic shockmanagement include antibiotics, and early goal-directed therapy. However, thewidespread application of these crucial therapies is limited by clinician knowledge,skills and experience.

Study Objectives: Evaluate the use of simulation-based teaching during the pre-clinical years of medical training to increase future clinician knowledge in thetreatment of septic shock.

Methods: This is a prospective interventional study at a university-based medicalsimulation center. A 5-hour curriculum including didactic lectures, skill workshopson central line insertion and intubation, and simulated case scenario of a septic shockpatient were administered to incoming first-year and second-year medical students. Askills checklist including 21 tasks was completed during the case scenario. An 18-question pre-test, post-test and 2-week post-test were given to evaluate theeffectiveness of the curriculum on student knowledge and knowledge retention. Thestudents also completed a survey at the end of the curriculum utilizing a Likert scale.The Wilcoxon rank sum test was used to compare differences in test scores.

Results: Sixteen first-year and sixteen second-year medical students were enrolled.The pre-test scores were 51.4�/�10.4% and 51.6�/�11.0%, respectively. Thepost-test scores were 81.6�/�10.3% and 84.4�/�7.4%, respectively, andsignificantly higher than pre-test scores, p�0.01. The 2-week post-test scores were74.3�/�13.3% and 82.3�/�8.9%, respectively, and significantly higher than pre-test scores, p�0.01. The task performance during the simulated case scenario was95.2�/�3.9% and 92.9�/�8.3% completion, respectively. From a scale of 1-5,first-year students noted their pre-curriculum and post-curriculum confidence levelsin managing septic shock patients were 1.0�/�0.0 and 2.9�/�0.6, respectively,p�0.01. Second-year students noted their confidence levels as 1.3�/�1.0 and 3.3�/�0.7, respectively, p�0.01. First and second-year students agreed or strongly agreedthat the curriculum should be a requirement during medical school training, 4.5�/�0.7 and 4.7�/�0.7, respectively.

Conclusion: Medical simulation is an effective method of educating medicalstudents regarding septic shock management prior to their clinical training.

154 Integrated Clerkship

Gaufberg SV, Ogur B, Hirsh D/Cambridge Hospital, Harvard Medical School,Cambridge, MA

Study Objectives:● To assess the effectiveness of an innovative longitudinal integrated model of med-

ical education designed to remedy limitations inherent in traditional medical ed-ucation

● To describe the role of emergency medicine (EM) education within our longitu-dinal integrated model.Methods:

● Traditional third-year medical education typically consists of resident-run wardteam rotations on randomly ordered discipline-specific inpatient services. Thiseducational structure may result in the following:

● Limited patient contact before diagnosis or after discharge● Limited connection to patients● Limited contact with experienced faculty● Ethical erosion● Marginalization of core themes such as communication skills, professionalism and

empathy.

In our program, a group of 12 third-year medical students follow a panel ofpatients from all of the core disciplines longitudinally for their entire third year.Students participate in most major health events for their cohort patients:accompanying them to or meeting them in the emergency department (ED), specialtyappointments and procedures, caring for them when they are admitted to hospital,and doing home visits. Emergency medicine is a key component of the Integrated

ICEM 2008 Scientific Abstract Program

Volume , . : April Annals of Emergency Medicine 517