E M S

A New Approach to Teaching Prehospital

Trauma Care to Paramedic Students

David R Johnson, MDDarryl Macias, MDAnn Dunlap, PhD, NREMT-PMark Hauswald, MDDavid Doezema, MD

Study objective: A modification of the standard Departmentof Transportation student paramedic curriculum encouraging indi-vidualized patient assessment decreases inappropriate on-sceneprocedures (OSPs) and scene time, measured on simulated patients.

Methods: Scenario-based testing from 1991 through 1993 wasvideotaped for all students. A new trauma curriculum was intro-duced in 1992, individualizing patient assessment and prioritiza-tion of OSPs. Recorded OSPs included spinal immobilization,application of military antishock trousers, endotracheal intubation,cricothyrotomy, intravenous catheter insertion, and needle thora-costomy. Twenty videotaped random student performances of the1991 class was compared with a similar sample of 20 from the1993 class; scene times and the OSP numbers were measured.Two board-certified independent emergency physicians unfamil-iar with the students or the new curriculum reviewed all 40 testson a master videotape. Patient assessment appropriateness,scene time, OSPs, scenario difficulty, and number of inappropri-ate OSPs were evaluated using a linear analog scale. Data arepresented as means with confidence intervals (CIs), analyzed byStudent’s t test and the Mann-Whitney 2-sample test.

Results: Scene time from 1991 to 1993 decreased overall witha mean of 4.3 minutes (95% CI 2.8 to 5.8 minutes), as did thenumber of OSPs: 3.1 versus 1.7 (mean difference, 1.45 OSPs perscenario; 95% CI .91 to 1.99). Physician reviewers noted improve-ments in the appropriateness of patient assessment, scene time,and OSPs from 1991 to 1993. There was no significant differencein scenario difficulty for 1991 compared with 1993. InappropriateOSPs done on scene declined. Physician 1 indicated a mean ofinappropriate procedures of 1.6 in 1991 versus .5 in 1993.Physician 2 indicated a mean of 1.4 in 1991 versus .3 in 1993.

Conclusion: This new paramedic curriculum decreased on-scene time and inappropriate use of procedures in stabilizingthe condition of patients with simulated critical trauma.

From the Department of EmergencyMedicine, New Mexico EmergencyMedical Services Academy,University of New Mexico, School of Medicine, Albuquerque, NM.

Received for publication September 23, 1996. Revisionsreceived January 6 and July 28,1998. Accepted for publication August 12, 1998.

Presented at the Annual Meeting ofthe National Association of EMSPhysicians 1994, Portland, OR.

Address for reprints: Darryl JMacias, MD, University of NewMexico Hospital, Department ofEmergency Medicine, 4th Floor ACC,Albuquerque, NM 87131; 505-272-5062, fax 505-272-6503,E-mail dmacias@salud.unm.edu.

Copyright © 1999 by the AmericanCollege of Emergency Physicians.

0196-0644/99/$8.00 + 047/1/94995

J A N U A R Y 1 9 9 9 3 3 : 1 A N N A L S O F E M E R G E N C Y M E D I C I N E 5 1

T E A C H I N G P R E H O S P I T A L T R A U M A C A R EJohnson et al

5 2 A N N A L S O F E M E R G E N C Y M E D I C I N E 3 3 : 1 J A N U A R Y 1 9 9 9

[Johnson DR, Macias D, Dunlap A, Hauswald M, Doezema D: Anew approach to teaching prehospital trauma care to paramedicstudents. Ann Emerg Med January 1998;33:51-55.]

I N T R O D U C T I O N

Despite limited research into the efficacy of many prehospi-tal procedures, little has changed with regard to paramedictrauma care curriculum, with minimal guidance given foron-scene prioritization of resource utilization and on-sceneprocedures (OSPs).1 Paramedics are generally taught tomanage trauma patients in a strictly protocol-driven fash-ion,1 rather than accounting for individual circumstances.Although most interventions can be beneficial to selectedtrauma patients, delayed transport of others to a hospitalwhere definitive care can be delivered may result in anadverse outcome.2,3

Specifically, much of the controversy has centered on theprehospital use of intravenous fluids and military antishocktrousers (MAST).4 Other largely unproven techniquessuch as spinal immobilization and chest decompressionmay not be appropriate in the management of individualcases. Similarly, different approaches are required for vic-tims of blunt versus penetrating trauma. A single thoracicstab wound usually requires minimal field assessment com-pared with multiple injuries sustained in a motor vehiclecrash. Furthermore, the “setting” in which the traumaticevent occurs may dictate the degree of field stabilizationneeded in a given situation. The treatment of a patient in arural setting with long transport times and a paucity of

available personnel differs from treatment in an urbanlocation. Other factors, such as mechanism of injury, per-sonnel, patient access, and transport time influences deci-sionmaking, yet no published paramedic training curricu-lum exists that explicitly considers these aspects.

M A T E R I A L S A N D M E T H O D S

The standard Department of Transportation (DOT) traumacurriculum was taught to all of our paramedic classes upto 1992; our modified curriculum taught at the NewMexico EMS Academy additionally focused on diseaseprocesses and prioritization of procedures appropriatefor a given prehospital setting and patientpresentation. Issuessuch as mechanism and severity of injury, availability of per-sonnel, transport times, and patient access were included asadditional considerations individualizing patient assess-ment. Treatment algorithms were developedbased on cur-rent prehospital emergency medicine research or existingliterature5 to aid in decisionmaking.

Before passing the trauma course, each student wasrequired to assess and manage a simulated trauma patient,and to demonstrate proficiency in 2 of 3 randomly chosenperformances at the end of the entire paramedic curriculum.The simulated patient scenarios were not the same, but didnot significantly differ between classes, although studentsfrom the 1991 class were allowed unlimited numbers ofassistants. The 1993 class, however, were given varyingnumbers of assistants. This forced the student to consideralternative resources when treating the simulated patient,comparable with a true field situation. Allotted time for

Table. Student performance ratings by 2 EMS physicians.

Performance Measure Rater 1991 Class 1993 Class κ (95% CI)

Appropriateness of patient assessment Physician 1 32 mm 62 mm.70 (.54 to .86)

Physician 2 30 mm 68 mmAppropriateness of scene time Physician 1 22 mm 59 mm

.86 (.77 to .93)Physician 2 21 mm 66 mm

Difficulty of scenario Physician 1 54 mm 63 mm.26 (.01 to .51)

Physician 2 48 mm 50 mm1991: .76 (.66 to .87)1993: .88 (.82 to .93)

Inappropriate procedures Physician 1 1.6 .5.50 (.29 to .71)

Physician 2 1.4 .3

T E A C H I N G P R E H O S P I T A L T R A U M A C A R EJohnson et al

J A N U A R Y 1 9 9 9 3 3 : 1 A N N A L S O F E M E R G E N C Y M E D I C I N E 5 3

regard to the appropriateness of patient assessment, scenetime, and the difficulty of the scenario.

All data are presented as means with 95% confidenceintervals (CIs). Continuous variables were analyzed byStudent’s t test. The Mann-Whitney 2-sample test wasused where appropriate. Exact weighted κ values with95% CI were also calculated.

R E S U L T S

Twenty trauma simulations were reviewed in each of the2 paramedic classes. In both groups 13 blunt trauma and7 penetrating trauma scenarios were reviewed. The newcurriculum resulted in a decrease in overall mean scenetime of 4.3 minutes (95% CI 2.8 to 5.8 minutes). A sub-group analysis of the 2 classes demonstrated that for blunttrauma, the class of 1993 spent 2.3 minutes less on scene(95% CI .5 to 4.2). For penetrating trauma, the latter classspent an average of 7.9 minutes less on scene (95% CI6.7 to 9.0). Although there was a trend toward shorterscene times for blunt trauma in the 1993 class, the overalldecrease in scene times occurred primarily in the penetrat-ing trauma scenarios.

The Table shows results of the review by the 2 board-certified emergency physicians. Ratings by the physicianson linear analog scales showed improvements in patientassessment and appropriateness of scene time. Inter-observer agreement for appropriateness of scene time wasexcellent (κ=.86, 95% CI .77 to .93),6 whereas agreementfor patient assessment was moderate (κ=.70, 95% CI .54to .86).6 Although averages of scenario difficulty appearedsomewhat equivalent, interobserver agreement for all 20scenarios with respect to difficulty was poor (κ=.26, 95%CI .01 to .51).6 The number of inappropriate proceduresdone on scene declined. Physician 1 indicated a mean ofinappropriate procedures of 1.6 in 1991 versus .5 in 1993.Physician 2 indicated a mean of 1.4 in 1991 versus .3 in1993. The number of OSPs performed also decreased from3.1 per scenario in 1991 to 1.7 in 1993 (mean difference,1.45 OSPs per scenario; 95% CI .91 to 1.99). Interestingly,up to 20% of those deleted procedures between the classesmay have been deemed “appropriate,” but were not specif-ically measured in the study. Serious omissions (not usingspinal immobilization when indicated, for instance)would have resulted in failure; no student failed. For all 4variables, agreement between the physicians was good in1991 (κ=.76, 95% CI .66 to .87). Agreement between thephysicians was excellent in 1993 (κ=.88, 95% CI .82 to .93).The observed difference in κ value by year was not statisti-cally significant (P>.05).6

didactic material and laboratory practice in both classes wasequivalent. The same instructors were used for the sameclasses and laboratory sessions for both years. Proficiency inperforming individual procedures had been demonstratedearlier in the trauma course.

Scenario testing for all students was recorded on video-tape. Students were informed of this policy at the beginningof the course. Videotaping at that time was done in a randomfashion. No paramedic did more than 1 scenario. In 1994, allvideotapes were collected and reviewed for recordingquality.A person unfamiliar with the study or any participants didthis review. In a few cases, scenarios were necessarily excludedfrom use because of poor videotape quality (inability to hearorders, or to clearly see what procedures the student was per-forming). Scenarios were not, however, excluded on the basisof whether the individual student received a passing evalu-ation in the scenario. From the viewable scenarios, 20 ran-domly selected student performances from each class werecopied onto a master tape. Performances were selected byrewinding the tape and selecting the first trauma assessmentscenario encountered. The scenarios of the 2 classes werematched only with respect to mechanism of injury (bluntversus penetrating trauma). All 40 performances from the2 classes were randomly copied to 1 master tape with sce-narios preceded by a brief synopsis.

Scene time was measured in seconds and was defined asbeginning when the student first made physical contact withthe patient and ending when the patient was moved to anambulance gurney and the student verbalized initiation oftransport. The number of OSPs was counted and includedendotracheal intubation, cricothyrotomy and needle thora-costomy, insertion of intravenous catheters, spinal immobi-lization,and application of MAST. “Inappropriate” OSPs weredefined as OSPs that were not indicated for a specific sce-nario (eg, cervical spine immobilization for a penetratingchest injury). Procedures that were relatively contraindicatedfor that scenario (eg, prolonged intravenous insertionattempts/MAST application on scene for a hemodynami-cally compromised patient who needs immediate transportto a trauma facility) were also deemed “inappropriate.”

Two board-certified emergency physicians with extensiveexperience in EMS and EMS medical direction reviewedthe master tape. Both assessors were blinded without pre-vious contact with either class, or with the development ofthe new curriculum. For each of the student performances,the physicians recorded the number of inappropriate OSPsbased on their impressions of reasonable medical care. Thephysicians did not use the algorithms, since the algorithmswere developed for the class of 1993. The physicians ratedeach performance on 100-mm linear analog scales with

This study had some significant limitations. Althoughscene time may be a logical surrogate of outcome2 (“thefaster to the hospital, the better”), it is not definitive.Further outcome studies are necessary to correlate scenetimes to outcome. Actual scenarios given to the studentswere not controlled for; although patient injuries weresimilar, environmental factors differed. Different scenariosgiven to the 2 groups might produce different results. The2 physician reviewers did not find any significant differencein the difficulty of the scenarios between the 2 classes.Nonetheless, groups were not specifically assessed formotor skill competency—although the same instructorstaught both classes, we did not account for how well 1student in a given class could “outperform” another. It isalso intuitive that students perform in the manner in whichthey are taught; although the reviewers had no previousknowledge of the curriculum, they may have favorablygraded faster scenarios if they themselves approve of thisphilosophy. It is impossible to conclude that the measure-ments indicate improved quality; outcome studies in thefield would be necessary to validate this. Nevertheless,previous EMS literature has not challenged paramedicteaching standards to this degree; evolution of existingstandards therefore necessitate studies such as this to beimproved on with validation from the field.

We used linear analog scales to rate the student perfor-mances in a subjective basis. Although this methodologyhas been used in rating pain, less experience exists with thismethod to measure other subjective variables.10 Linearanalog scales are powerful tools to demonstrate small sta-tistical differences between interventions; the clinical sig-nificance of these differences, however, can easily be ques-tioned.10 Although a well-defined measurable endpoint forscenario difficulty and appropriateness of OSPs was lack-ing, we used equivalent scenarios for each class. We did notspecifically capture omissions, although necessary omis-sions would have resulted in failure, which did not occur.Physician reviewers did not receive specific instructionsor protocols to compare assessments for appropriateness;they evaluated scenarios on the basis of their view of“reasonable treatment” expected of any paramedic. Morereviewers, and the availability of a well-researched, standardnationwide EMS treatment protocol would have minimizedthese problems.

In conclusion, the adoption of a new curriculum, whichstresses the mechanism of injury, and the setting in which atraumatic event occurs, enables paramedics to individualizepatient care. Such individualized care can result in a decreasein scene times and the number of procedures performed onthe scene in simulated trauma scenarios. These effects can

T E A C H I N G P R E H O S P I T A L T R A U M A C A R EJohnson et al

5 4 A N N A L S O F E M E R G E N C Y M E D I C I N E 3 3 : 1 J A N U A R Y 1 9 9 9

D I S C U S S I O N

The efficacy of the standard DOT curriculum has not beenstudied extensively, yet most paramedic curricula arebased on this standard. The existing curriculum may insuf-ficiently tailor assessment and treatment for individualcircumstances, likely resulting in longer scene times. In ourcurriculum, students from the latter class demonstratedshorter scene times, less inappropriate procedures, and wereencouraged to use “clinical judgment,” which is not com-monly taught to paramedic students.

For instance, we stressed that unless there is a necessarydelay in transport, all intravenous lines are to be started enroute to the hospital; previous field studies validate this.7

Selective spinal immobilization was also taught. Those withminor mechanisms of injury, no neck pain or tenderness,no neurologic complaints or findings, and no distractinginjuries would not require cervical spine immobilization.Immobilization is mandated if these criteria are not met.8

Furthermore, spinal immobilization would be unnecessary inmost cases of penetrating trauma, especially in theabsence of neurologic signs or symptoms.9 Unnecessaryspinal immobilization may be of limited benefit, and maydelay transport where definitive care would better serve apatient, given that full immobilization can also take almost3 minutes to accomplish under optimal conditions.9Moreover,evidence supporting the efficacy of backboards in prevent-ing further neurologic injury is lacking, and couldactually bedeleterious to the patient.9 However, pending further out-come data, overimmobilization may be prudent in uncertaincircumstances.

Students were taught to evaluate and treat traumapatients in consideration of available personnel and train-ing level, patient access barriers, transport unit availabil-ity, and need for other resources such as extrication and firesuppression. These factors were weighted in a fashionakin to real field triage: immediate life-threateninginjuries are dealt with on scene, with subsequent trans-port; injuries that would threaten survival at a later timewould be treated en route. Students also learned resourcemanagement skills based on initial dispatch informationbefore scene arrival. When videotapedperformances from aclass before the new curriculum were compared with thosefrom after the new curriculum, there was a significantdecrease in scene time. This was likely a result of the decreasein OSPs performed. On subgroup analysis, it appears thatthe major decrease in scene time was for simulated pene-trating trauma cases. Gervin and Fischer2 suggested thatthere is a significant correlation between survival frompenetrating trauma to the heart and scene time, thus scenetime may be a surrogate marker for survival.

T E A C H I N G P R E H O S P I T A L T R A U M A C A R EJohnson et al

J A N U A R Y 1 9 9 9 3 3 : 1 A N N A L S O F E M E R G E N C Y M E D I C I N E 5 5

be shown to be appropriate when reviewed by experiencedEMS physicians. The effect on patient outcomes will haveto await carefully designed field studies.

R E F E R E N C E S1. Bledsoe BE, Porter RS, Shade BR: Paramedic Emergency Care, ed 3. Upper Saddle River, NJ:Brady Prentice Hall, 1994:553-569.

2. Gervin AS, Fischer RP: The importance of prompt transport in salvage of patients with pene-trating heart wounds. J Trauma 1982;22:443-448.

3. Spaite DW, Tse DJ, Valenzuela TD, et al: The impact of injury severity and prehospital pro-cedures on scene time in victims of major trauma. Ann Emerg Med 1991;20:1299-1305.

4. Mattox KL, Bickell W, Pepe PE, et al: Prospective MAST study in 911 patients. J Trauma1989;29:1104-1112.

5. Rosen P, Barkin RM: Emergency Medicine; Concepts and Clinical Practice, ed 3. St. Louis:Mosby–Year Book, 1992.

6. Fleiss JL: Statistical methods for rates and proportions. Wiley Series in Probability andMathematical Statistics, ed 2. New York: John Wiley & Sons, 1981:212, 236.

7. Slovis CM, Herr EW, Londorf D, et al: Success rates for initiation of intravenous therapy enroute by prehospital care providers. Am J Emerg Med 1990;8:305-307.

8. Ringenberg BJ, Fisher AK, Vrandeta LF, et al: Rational ordering of cervial spine radiographsfollowing trauma. Ann Emerg Med 1988;17:792-796.

9. Hauswald M, Ong G, Tandberg D, et al: Out of hospital spinal immobilization: Its effect onneurologic injury. Acad Emerg Med 1998;5:214-219.

10. Todd KH, Funk JP: The minimum clinically important difference in physician-assigned visualanalog pain scores. Acad Emerg Med 1996;3:142-146.