Injury, Int. J. Care Injured (2005) 36, 579—587

www.elsevier.com/locate/injury

REVIEW

Development of pre-hospital trauma-caresystem–—an overview

*

B.R. Sharma

Department of Forensic Medicine and Toxicology, Goverment Medical College and Hospital,#1156-B, Sector 32-B, Chandigarh 160030, India

Accepted 20 September 2004

Contents

Introduction . . . . . . . . . . . . . .

Triage using multiple components

Defining a major trauma victim .

Evaluation of triage components .

KEYWORDSTrauma;Triage;Triage tools;Pre-hospital Triage;Pre-hospital trauma life

support;In-hospital triage;Trauma centres;Trauma documentation

* Tel.: +91 172 2622 145.E-mail address: drbrsharma@yaho

0020–1383/$ — see front matter # 20doi:10.1016/j.injury.2004.09.009

Summary Trauma is recognized as a serious health care problem world-wide andthere is a general agreement that trauma is best addressed in a systematic manner.Accordingly, trauma systems must be designed to use efficiently the limited healthcare resources available. The majority of injuries are of only minor, or moderate,severity, and can be managed well at local community hospitals. On the other hand, asignificant minority of injured patients will require extensive and expensive care inorder to survive, or to minimize the morbidity of injury. For a trauma system to beoptimally effective, given limited available resources and the need for urgent care ofthe seriously injured, it is critical to utilize a method of differentiating those injuryvictims, who need the specialized expertise and resources available in traumacentres, from those who can be cared for adequately locally. Rapid field assessmentof injured patients is, therefore, an essential element of any effectively functioningtrauma system. This review examines the evolution of pre-hospital trauma-caresystems, during the triage of a trauma victim.# 2004 Elsevier Ltd. All rights reserved.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 580

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582

o.com.

04 Elsevier Ltd. All rights reserved.

580 B.R. Sharma

Introduction of pre-hospital trauma life support (PHTLS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583

Secondary emergency department triage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583

Multiple and mass-casualty triage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584

Military triage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585

Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585

References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 586

Introduction

The term triage derives from the French verb ‘trier’(meaning to pick), originally describing how earlyFrench wool traders sorted wool into various cate-gories, according to its quality.1 Historically, begin-ning with Napoleon’s surgeon, Dominique-JeanLarrey, the primary goal of military triage was toprioritize care according to the severity of injuryand the chances of survival. This system providedurgent surgical care to the less critically wounded,with a view to their speedy return to battle. Withthe advent of regional trauma-care systems in the1970s, a new dimension was added to the concept ofcivilian triage, the goal being rapidly and accuratelyto identify those trauma victims whose severity ofinjury warranted the resources of a trauma centre.With the evolution of trauma-care systems in the1980s and 1990s, it became apparent that suchsystems must be ‘inclusive’. This concept was firstdescribed in the Centre for Disease Control’s (CDC)position paper on trauma-care systems.2 The modelof an inclusive trauma-care system was subse-quently incorporated into the model trauma-careplan developed by the Health Resources and Ser-vices Administration of the US Public Health Service3

and the Resources for Optimal Care of the InjuredPatient: 1993, published by the Committee onTrauma of the American College of Surgeons (ACS).4

Based on estimates that only 5—10% of all traumapatients required trauma centre facility5 and itsbeing established that the chances of survivalimprove when victims of major trauma are treatedin trauma centres,6—9 earlier system designs hadtended to exclude non-trauma centre hospitals.However, despite the reports that only the mostseverely injured 10% of patients require theexpanded resources of a designated trauma centre,it was critical that the remaining 90% should alsoreceive optimal care. The trauma-care system was

then expanded to include all facilities providingcare to injured person in a given geographical area.The goal of the inclusive trauma-care system was tomatch each hospital’s trauma-care resources withthe needs of injured persons.10 Although the traumacentre was a key component of acute care for theseverely injured, the inclusive trauma systemencompassed all phases of care, from pre-hospitalthrough acute care and rehabilitation. Inclusivesystem guaranteed that all injured patients wouldreceive optimal care, to the extent of the resourcesavailable for this, even if they do not require theresources of a trauma centre.

Information on which pre-hospital triage was andis based includes:

(A) P

hysiological data based on the circulation,respiration, abdominal, motor, and speech(CRAMS) scale; Glasgow Coma Scale (GCS);trauma score; revised trauma score and traumatriage rules etc., have the advantages of beingreadily assessable in a traditional physicalexamination and easily rendered into a numer-ical format.

Clemmer et al.,11 reported that the CRAMS Scalewas most amenable to use by paramedics and that itcould be successfully implemented to direct trans-port to a level trauma centre. In Clemmer’s originalstudy, after adding a respiratory rate to classify r-espiration as being abnormal, the CRAMS score wasdetermined from the pre-hospital report and re-d-etermined on arrival at the emergency department.Patients with a CRAMS < 6 were found to have ahigher mortality and this became the cut-off for thedefinition of major trauma with a mortality rate of62% and surgery or ICU admission rate of 56%.9

The trauma score, a modification of the triageindex, was introduced as the ‘perfect predictor’.Vital signs were added to the physiological para-

Development of pre-hospital trauma-care system 581

meters, scoring was simplified and interval weight-ing was determined by a consensus conference.Field testing of the trauma score was later per-formed, using a trained nurse observer, in additionto pre-hospital personnel.12 This, however, revealeda high false negative rate and it was recommendedthat other variables, such as transport time,mechanism of injury and site be considered.13,14

In 1989, Champion et al.15 revised the trauma score,in response to the finding that it tended to under-estimate the severity of head-injured patients. Theyadopted the commonly used and accepted intervalsof the Glasgow Coma Score while dropping theobserver-dependent capillary refill and respiratoryexpansion scores, and matching the intervals forsystolic blood pressure and respiratory rate to theGCS intervals.

Under the trauma triage rule, Baxt et al.15

defined amajor injury as one that was characterizedby:

(1) p

rocedure with positive findings (life-threaten-ing if not treated) within the first 48 h of admis-sion, including thoracotomies, pericardialwindows, craniotomies and burr hole place-ment;

(2) in

-hospital fluid resuscitation of greater than1000 ml, or a transfusion required to maintainsystolic blood pressure above 89 mm Hg;

(3) in

vasive CNS monitoring with a head CT demon-strating a significant epidural, subdural, orintraparenchymal haemorrhage, or with a docu-mented elevated intracranial pressure and

(4) d

eath.

They further suggested testing the performanceof triage rules, not on the patients with obvious s-evere injury, or obvious absence of severe injury,but on the patients with equivocal diagnoses ofmajor injury.

An analysis of the different pre-hospital scoringsystems, based on physiological data demonstratedthat they could predict, which patients would die,but that none was able to identify accurately thosepatients who had sustained major life-threateninginjuries and would survive.16 This was partlybecause many patients with life-threateninginjuries have stable physiological parameters inthe field, owing to compensatory mechanisms inresponse to injury, particularly when there is onlya short interval between injury and the triage deci-sion. For this reason, it became necessary to estab-lish criteria that were not time dependent and thatcould be of particular advantage in situations wherethe time for making decisions about hospitalizationwas short.

(B) A

natomical criteria including a penetratinginjury (head to mid thigh); any traumatic ampu-tation proximal to ankle or wrist; burns invol-ving >20% of body surface area, or burns withsuspected respiratory involvement, stronglysupport referral to a trauma centre/burns cen-tre. Although a trauma system may choose todefine anatomical criteria more precisely, torsopenetration serves as a useful guideline.

(C) M

echanism of injury criteria include:(1) falls with attention to distance and surface;(2) motor vehicle crashes with fatality in same

car, passenger ejection, complicated extri-cations, violation of passenger space and

(3) pedestrian struck by motor vehicle withattention to speed, distance thrown andlanding surface.

Important information about the probability ofserious injury can be gained by evaluating the phy-sical forces involved in a traumatic event. The ob-servations of pre-hospital personnel become criticaland must be well documented if mechanisms of i-njury are to be used in the triage decision. In thecase of a motor vehicle collisions, for example, pre-hospital personnel should, if possible, documentvehicle speed, deceleration distance, violation ofpassenger compartment space, passenger ejection,and deformed contact points.17 Jones and Cham-pion18 have reported that impact speed under 20 -mph generally produce minor injuries, whereasthose greater than 25 mph without seat belts and35 mph with seat belts can produce severe, or le-thal, injuries. They also reported that the criterionof vehicle speed of 20 mph or more in direct frontalcollisions, 22 mph or more in the offset collisionsand 15 mph or more in side-impact collisions cap-tured more than 90% of patients with an injury s-everity score (ISS) greater than 16.

(D) A

ge and associated conditions elderly traumavictims are reported to be at greater risk thanyounger ones for errors in triage, since a givenimpact results in a greater magnitude of injuryin such persons.19,20 Similarly, young childrenwith traumatic injuries require special care andshould, therefore, be triaged to a trauma centrewith the appropriate resources and special com-mitment to the care for the injured child.4 Oncein the trauma centre, patients with pre-existingdisease and elderly patients are reported toconsume more resources than those withoutpre-existing illness, and younger patients, withsimilar injuries.21,22 Hostile environmental con-ditions, such as prolonged exposure to heat orcold, and associated diseases, such as diabetes

582 B.R. Sharma

mellitus, should also influence the triage deci-sion.23

Triage using multiple components

Ideal criteria for triage would be 100% accurate indifferentiating risk in victims with life-threateninginjuries from those not at risk of death, but suchaccuracy is not attainable. Any system of triagedecision-making will necessarily be imprecisebecause of biological variability, the short intervalbetween injury and evaluation by pre-hospital per-sonnel, and the difficulty of diagnosing thoracic andabdominal injuries in the field. Hence, it is a car-dinal principle that the triage method should err onthe side of over-estimation of risk (over-triage), noton the side of under-triage, so as not to result inpreventable deaths. The American College of Sur-geon’s Committee on Trauma4 recommended thefollowing ‘pre-hospital triage decision scheme’:

Step 1: evaluate vital signs and level of conscious-ness:

�Glasgow Coma Scale <14 or�Systolic blood pressure <90 or�Respiratory rate <10 or >29 or�Revised trauma score/paediatric trauma score <11/<9

Decision: yes, take to a trauma centre/alert trauma team; no,look for step 2.

Step 2: assess anatomy of injury:

�All penetrating injuries to head, neck, torso and extre-mities proximal to elbow and knee�Flail chest�Combination of trauma with burns�Two or more proximal long bone fractures�Pelvic fractures�Amputation proximal to wrist and ankle

Decision: yes, take to a trauma centre/alert trauma team; no,look for step 3.

Step 3: evaluate for evidence of mechanism ofinjury or high-energy impact:

�Ejection from automobile�Death in same passenger compartment�Extrication time > 20 min�Falls > 20 feet�Rollover�High-speed auto crash, initial speed > 4 mph major autodeformity, >12 in. intrusion into passenger compartment�Auto-pedestrian/auto-bicycle injury with significant(>5 mph) impact�Pedestrian thrown or run over�Motorcycle crash >20 mph or with separation of riderfrom bike

Decision: yes, contact medical control and consider transport to atrauma centre/consider trauma team alert; no, look for step 4.

Step 4: evaluate for age and associated condi-tions:

�Age <5 or >55�Cardiac disease, respiratory disease�Insulin-dependent diabetes, cirrhosis of liver, or morbidobesity�Pregnancy�Immunosuppressed patients�Patient with bleeding disorder or patient on anticoagu-lants

Decision: yes, contact medical control and consider transport to atrauma centre/consider trauma team alert; no, re-evaluate.

Defining a major trauma victim

It is necessary to define the victim of major traumabefore the component parts of the triage decisioncan be evaluated.24 A broad definition would bevictim whose magnitude of injury requires theresources of a trauma centre. However, it may takehours, or even days, to determine the true magni-tude of a patient’s injury, yet there is the need toidentify major-trauma victims at entry into thetrauma system. The injury severity score providesa useful retrospective definition of the majortrauma victim.25 A commonly accepted definitionfor major trauma is an ISS > 15. This requires asingle grade 4 injury on the abbreviated injury scale(AIS) or two AIS = 3 injuries.

Since the primary objective of a regional traumasystem is to decrease mortality and morbidity frominjury, it is essential that the criteria used in thesystem be sufficiently sensitive to identify the vastmajority of victims at risk from life-threateninginjury. This can be accomplished by combining phy-siological and anatomical criteria with the mechan-ism of injury and other criteria. However, incombining triage components, an appropriate bal-ance has to be maintained between the medicalobjective of identifying the majority of patients atrisk and the economic objectives of minimizingover-triage. Multiple studies in the literature haveanalyzed the methodology of field triage.26—30 Thegeneral consensus of these studies is that physiolo-gical status alone is not a very sensitive indicator ofinjury severity.

Evaluation of triage components

The accuracy of triage methods is defined by its‘sensitivity’ (the fraction of injured patients cor-rectly identified by the triage method) and ‘speci-ficity’ (the fraction of non-injured patients correctlyidentified by the triage method), which are proper-

Development of pre-hospital trauma-care system 583

ties unique to a given triage method applied to agiven class of injury. Terms more commonly used,and more easily understood, in defining the accu-racy of a triage method are ‘over-triage’ (a falsepositive resulting from the identification of a majortrauma victim, who, on retrospective analysis, doesnot meet the criteria for major trauma) and ‘under-triage’ (a false negative resulting from the incorrectidentification of a patient as not having a majortrauma when in fact such an injury exists).31 Thesensitivity and specificity are interdependent andcan not be determined independently as they arepopulation-specific, in that they depend on theinjury-severity mix in the population being studied,and, therefore, cannot be transferred from onesystem to another.

Introduction of pre-hospital trauma lifesupport (PHTLS)

The pre-hospital trauma life support (PHTLS) pro-gramme started as a pilot project in 1983—1984 inthe states of Lowa, Connecticut and Louisiana.32

Since then, it has been recognized by the AdvancedTrauma life Support faculty and the American Col-lege of Surgeons Committee on trauma as the stan-dard for the training of paramedical personnel in theconcepts of pre-hospital trauma care; it has beenpromulgated internationally in the same fashion asthe advanced trauma life support (ATLS) pro-gramme.33 In 1986, ATLS was introduced in Trinidadand Tobago, and, in 1990, an outcome studyreported a positive impact of the programme ontrauma patient survival.34 A 100% decrease in mor-tality was reported among the trauma patients withinjury severity scores �16, after the introduction ofATLS programme. However, compared with NorthAmerican standards35,36 survival was still consider-ably lower and anecdotal reports from physiciansattending the ATLS course indicated that traumapatient outcome was adversely affected because ofa lack of pre-hospital trauma care. In 1992, thePHTLS programme was introduced in Trinidad andTobago, with the objective of decreasing mortalityby improving pre-hospital trauma care. Studies havedemonstrated an improvement in overall traumapatient mortality from 15.7 to 10.6% following thePHTLS course,37 although it has been argued thatthis decrease in mortality may be a function of time,rather than being attributable to the PHTLS pro-gramme.

The PHTLS programme teaches two levels of pre-hospital care: the basic and the advance pro-grammes. Basic programme includes non-invasiveskills, such as chin lift, jaw thrust, oro-pharyngeal

airway and oxygen application for airway problems;C-spine control by simple immobilization techni-ques, such as sand bags, straps and spine boards;haemorrhage control by direct pressure dressings;and fracture stabilization by simple splinting tech-niques. The advanced programme includes theteaching of such techniques as venous access andendotracheal intubation.

Since the promulgation of the PHTLS programme,pre-hospital trauma care has continued to evolve,although not without controversy. Areas of conflict-ing views include whether field personnel shouldrapidly transport them to a proper receiving facilityfor definitive care or attempt to stabilize the victimsof trauma (‘‘scoop and run, or stay and play’’).38—40

The training of paramedical personnel in pre-hospi-tal use of definitive airway techniques, such asendotracheal intubation, and the pre-hospitalestablishment of venous access with intravenousfluid administration have been questioned.41—44 Ithas been suggested that other factors, such as dis-tance from the trauma unit, duration of shock andthe nature of the injuries sustained may be impor-tant in determining the effect of pre-hospital intra-venous fluid administration on trauma patientoutcome.45 In a large urban trauma centre withrapid transport, pre-hospital intravenous fluidadministration, particularly for penetrating torsotrauma, was not shown to improve survival.46

One area of pre-hospital trauma care thatappears to improve outcome is the use by pre-hospital personnel of advanced techniques of airwaycontrol.47—49 Furthermore, cervical spine control,splinting of fractures, haemorrhage control and useof oxygen may account for the patient’s arriving atthe hospital setting in a more optimal state. Thisstatus, in turn, should be expected to improve thechances of survival, as well as decrease the morbid-ity rate after injuries. Another important factor indetermining outcome after trauma is pre-hospitaltime, which depends on the time taken for on-sceneintervention, as well as the speed of the trans-port.50,51

Secondary emergency department triage

Initially, a modified version of the criteria of the ACStriage decision scheme was used as the basis for fulltrauma team mobilization. It was an extremelysensitive, but not specific, predictor of injuriesserious enough to require Intensive Care Unitand/or operating room resources. From the mod-ified ACS scheme, a two-tiered trauma teamresponse algorithm was defined and subsequentlyrefined. There were two response teams: the full

584 B.R. Sharma

response team included a trauma surgeon and/or anintensive care fellow, in addition to the paediatricsurgery residents, intensive care nurse, emergencydepartment nurse and nursing coordinator of themodified response team, all of whom mobilizedimmediately; the modified/limited response teamincluded an attending trauma surgeon, house-staffand radiology technician, without mobilization ofother level–—1 resources. The two-tier algorithmwas reported to differentiate specifically thetrauma patients whose injuries warranted fulltrauma teammobilization from those whose injuriescould be managed with normal emergency depart-ment care. Prospective evaluation and refinementof the criteria were performed to improve sensitiv-ity. The limited response became expedited care bythe emergency department staff with a mandatorytrauma consultation within 60 min. Selective use ofpersonnel and hospital resource, such as operatingroom services, laboratory studies and protectivegear, has been shown to yield cost savings. Anupgraded response was possible at any time andunder-triaged patients did not seem to suffer anyadverse effects due to the downgraded traumaresponse.

Multiple and mass-casualty triage

For optimal allocation of resources in the treatmentof trauma, it would be useful to be able to decide asearly as possible, which patients would benefit mostfrom transport to a dedicated trauma centre and,conversely, which patients could safely be excludedfrom entry into this system. Basic triage of traumavictims in the field becomes critical when availableresources are likely to become overwhelmed, as in amultiple- or mass-casualty situation.

The definition of a multiple- or mass-casualtysituation depends on both the number of victimsand the availability of medical resources for thespecific area affected.52 A multiple-casualty eventmay be defined as one in which the number ofvictims stresses local resources to the point at whichtreatment must be prioritized so that the victimswith the most urgent needs receive the earliestdefinitive care. In a multiple-casualty event, oncedisaster protocols have been initiated, the firstpriority is to identify casualties belonging to‘immediate care category’. This group includes vic-tims with obvious major anatomical injuries oraltered physiological status, such as abnormalrespiration, cardiovascular instability, an alteredlevel of consciousness, or lateralizing neurologicalsigns. Mechanism of injury and other componentparts of the triage decision are not usually used

since they are associated with unacceptable ratesof over-triage. Victims in need of medical care, butnot at immediate risk of losing their lives, or a limb,or those who are so seriously injured that they arenot expected to survive their injuries, are groupedunder ‘delayed care category’. It is assumed thatthe victims in the ‘stable (non-ambulatory) cate-gory’ can safely wait until the needs of the victimsrequiring immediate attention have been met,whereas the ‘walking wounded’ are taken awayfrom the disaster scene to a safer area for latertreatment.

In a mass-casualty event, local resources areoverwhelmed and, as a result, resources must bedirected toward victims with the greatest probabil-ity of survival: accordingly, the criteria for allocat-ing victims to the various treatment categories needto be modified. The first and most challenging mod-ification involves shifting the most critically injuredvictims from the immediate care to the ‘dead anddying category’. The threshold for this shift dependson the balance between the number of victims andthe magnitude of their injuries and the availabilityof resources over time.

A ‘category I’ event represents a mass casualtysituation for which resources from the immediatesurrounding region are readily available to meet thelocal needs. It is assumed that definitive care wouldbe provided within two hours from the time ofarrival of the first responder at the scene of thecasualty. Victims allocated priority in this event willinclude those with fixed and dilated pupils,advanced neurological findings, injuries to thetorso, a systolic blood pressure of <50 mm Hg afterinitial resuscitation, or massive burns.

A ‘category II’ situation requires extensive sup-port from surrounding and distant regions, withdefinitive care reaching the victim within 12 h.Patients allocated priority in this event wouldinclude those with torso injuries, unstable vital signsnot responding to initial resuscitation, intracranialinjuries with deteriorating levels of consciousness,lateralizing neurological signs, or second or thirddegree burns involving >50% of the body surface.

Those situations requiring extensive out of regionsupport, including military and definitive surgicaltherapy within 24 h (such as torso trauma requiringsurgical care, craniocerebral injuries requiring cra-niotomy), or prolonged support in ICU, are allocatedto ‘category III’.

A ‘category IV’ event represents an overwhelm-ing disaster situation in which the interval frominjury to definitive care is measured in days. In suchsituations, medical resources would be directedprimarily to the ‘walking wounded’, based on thephilosophy of ‘‘ the greatest good for the greatest

Development of pre-hospital trauma-care system 585

number’’. Because of limitations of time andresources, treatment at the scene after triageshould be limited to victims for whom there is areasonable probability that such treatment willdecrease morbidity and mortality.53 As such, onlythe basic elements of trauma care like airway con-trol, control of external bleeding etc., can beaddressed effectively at this level, the primary goalat the scene remaining rapid triage and evacuation.In summary, multiple- and mass-casualty triage dif-fers from single-victim triage in that the resourcesfor dealing with it are limited and patient care mustbe prioritized and pragmatic.

Military triage

The principles of military triage form the basis ofmost civilian triage protocols and reflect the pro-blems and issues of mass-casualty events morethan those of single-victim injuries.54 The guidingprinciple of military triage in multiple-casualtysituations is to strive to do the greatest good forgreatest number of injured. The aim is to identifythose patients with minimal injuries, who can betreated quickly and returned to duty, those withpotentially life-threatening, but salvageable, inju-ries, so that they can be treated, stabilized andevacuated to receive more definitive care, andthose likely to die from their injuries, so that theycan be provided with comfort care. This imposesmaking difficult decisions about injured patientswho, in other situations, would be given heroicresuscitation.

Documentation

The measurement and documentation of injuryseverity are prerequisites for the development,evaluation and improvement of trauma-care sys-tems, as well as for the advancement of publicpolicy for control of injuries. The scarcity of injurydata on the estimated 50% of fatally injured personswho die at the scene of accident, or in transit,continues to represent a significant informationgap. Demographics and injury information on thesevictims, if it exists, is found only in autopsy reports.Lack of a uniform system of death investigationsometimes renders these data incomplete and noteasily accessible.55 As such, proper documentationshould become an integral element of any pre-hos-pital trauma-care system.

Many hospitals have found that a trauma registryis an efficient method for storing and analyzing suchdata. Basic data elements in the registry should

include demographics (patient’s age, sex, injurycause and injury type), information on pre-hospitalcare, data on the process of acute care (treatment,major surgical procedures, identification of theattending discipline, response time), clinical items(sequential measurements of RTS and blood alcohollevel), final anatomical diagnoses (from examina-tion, X-rays/CT/MRI, surgery, or autopsy) and out-come data (discharge status, hospital and ICU lengthof stay, complications and functional disability atdischarge).56 Trauma registries must be founded oncomplete, accurate data that include explicit andaccurate descriptions of physiological derangementand injuries as well as accurate injury coding basedon ICD-9-CM codes (allocated and recorded in themedical records by clinicians, not non-medical per-sonnel).57,58

Conclusions

� T

he fundamental goal of trauma system triagebeing to send the severely injured to traumacentres, while avoiding their overload withpatients who can be managed effectively else-where, further prospective studies are necessaryto refine triage criteria.

� T

he goal in setting triage criteria should be toprovide a protocol for properly categorisinginjured patients, transporting them to appropri-ate hospitals, and ensuring an appropriateresponse to each trauma patient.

� In

the current climate of health-care reform, theeconomic viability of trauma systems will dependon prudent triage. Instruments for triage mustsupport the inclusive trauma-care system, provid-ing optimal care to all injured patients, within theframework of the available resources.

� W

ith increasing cognizance of the limitedresources of a trauma system, it becomes increas-ingly important to describe the major traumavictim with an operational definition based onresource consumption (i.e., length of stay, needfor surgery, ICU admission, etc.), rather than withthe more conventional definition based on ISS.When such an operational definition is used, over-and under-triage will correlate appropriately withthe available resources of the system.

� In

the inclusive trauma systems of the future, thetriage instrument must ensure that all resourcesare utilized efficiently. The issue of transferringpatients from one facility to another, according totheir medical needs, has to be another function ofsystem triage. Finally, triage may be needed fortrauma patients after they reach a trauma centre,in order to effect a tiered response, i.e., the

586 B.R. Sharma

activation of more or less comprehensive teamsand protocols, depending on the severity ofinjury).

References

1. Oxford English Dictionary. 500th Anniversary ed. OxfordUniversity Press; 1971.

2. Third National Injury Control Conference: Setting theNational Agenda for Injury Control in the 1990s. AtlantaGA, US Department of Health and Human Services, Centrefor Disease Control; 1991.

3. Model Trauma Care System Plan. Washington DC: US Depart-ment of Health and Human Services, Public Health Service,Health Resources and Services administration; 1992.

4. American College of Surgeons Committee on Trauma.Resources for Optimal Care of the Injured Patient: 1993.American College of Surgeons, Chicago; 1993.

5. Accident Facts 1984. Chicago, IL; National Safety Council;1984.

6. Shackford SR, Cooper GF, Hollingsworth-Friedlund P, EastmanAB. The effects of regionalization on quality of trauma care asassessed by concurrent audit prior to and after institution oftrauma system: a preliminary report. J Trauma 1986;26:812.

7. West JG, Gazzaniga AG. Impact of regionalization: TheOrange County experience. Arch Surg 1983;118:740.

8. West JG, Trunkey DD, Lim RC. System of trauma care: a studyof two counties. Arch Surg 1979;114:455.

9. Kilberg L, Clemmer TP, Clawson J, et al. Effectiveness ofimplementing a trauma triage system on outcome: a pro-spective evaluation. J Trauma 1988;28:1493—8.

10. Eastman AB,Walsh JC. Survey of trauma systems. In: NarayanRK, Wilberger Jr JE, Povlishock JT, editors. Neurotrauma.New York: McGraw Hill; 1995. p. 69—84.

11. Clemmer TP, Orme JF, Thomas F, et al. Prospective evaluationof the CRAMS for triaging major trauma. J Trauma 1985;25:188—91.

12. Moreau M, Gainer PS, Champion H, et al. Application of thetrauma score in the pre-hospital setting. Ann Emerg Med1985;14:1049—54.

13. Champion HR, Sacco WJ, Hannan DS, et al. Assessment ofinjury severity: the triage index. Crit Care Med 1980;8:201—8.

14. Champion HR, Sacco WJ, Copes WS, et al. Revision of thetrauma score. J Trauma 1989;29:624—9.

15. Baxt WG, Jones G, Fortlage D. The trauma triage rule: a newresource based approach to the pre-hospital identification ofmajor trauma victims. Ann Emerg Med 1990;19:1401—6.

16. Emmerman CL, Shade B, Kubincanek J. A comparison of EMTjudgement and prehospital trauma triage instruments. JTrauma 1991;31:1369—75.

17. West JG, Murdock N, Baldwin N, Whalen E. A method forevaluating field triage criteria. J Trauma 1986;26:265.

18. Jones JS, Champion HR. Trauma triage: vehicle damage as anestimate of injury severity. J Trauma 1989;29:646.

19. West JG. An autopsy method for evaluating care. J Trauma1981;21:32.

20. Baker SP, O’Neill B. The injury severity score: an update. JTrauma 1976;16:882.

21. Oreskovich MR, Howard JD, Copassis MK, Carrico CJ. Geriatrictrauma: injury patterns and outcomes. J Trauma 1984;24:565.

22. Milzman DP, Boulanger BR, Rodriquez A, et al. Pre-existingdisease in trauma patients: a predictor of fate independentof age and injury severity score. J Trauma 1992;32:236.

23. Morris JA, MacKenzie EJ, Edelstein SL. The effect of pre-existing conditions on mortality in trauma patients. JAMA1942;263:.

24. Young WW, Young JC, Smith JS, Rhodes M. Defining the majortrauma patient and trauma severity. J Trauma 1991;31:1125.

25. Baker SP, O’Neill B, Hadden W, Long WB. The injury severityscore: a method for describing patients with multiple injuriesand evaluating emergency care. J Trauma 1974;14:187.

26. Baxt WG, Berry CC, Epperson MD. The failure of prehospitaltrauma prediction rules to classify trauma patients accu-rately. Ann Emerg Med 1989;18:21.

27. Knopp R, Yanagi A, Kallsen G, et al. Mechanism of injury andanatomic injury as criteria for prehospital trauma triage. AnnEmerg Med 1988;17:895.

28. Codington EM, Young JC, Shufflebarger CM. The utility ofphysiologic status, injury site, and injury mechanism inidentifying patients with major triage. J Trauma 1988;28:305.

29. Ornato J, Milnek E. Ineffectiveness of the trauma score andthe CRAMS Scale for accurately triaging patients to traumacentres. Ann Emerg Med 1985;14:1061.

30. Kane G, Engelhardt R, Celentrano J, et al. Empirical devel-opment and evaluation of prehospital trauma triage instru-ments. J Trauma 1985;25:482.

31. Eastman AB, Lewis FR, Champion HR. Regional trauma systemdesign: critical concepts. Am J Surg 1987;154:79.

32. McSwain NE. History of PHTLS. In: McSwain NE, Butman AM,McConnell WK, Vomacka RW, editors. Basic and advancedpre-hospital trauma life support. 2nd ed. St. Louis: MosbyYear Book Inc.; 1990. p. xx—xxiii.

33. Collincott PE. Advanced trauma life support: past, present,future–—16th Stone Lecture, American Trauma Society. JTrauma 1992;33:749.

34. Ali J, Adam R, Butler AK, et al. Trauma outcome improvesfollowing the advanced trauma life support programme in adeveloping country. J Trauma 1993;34:890.

35. Copes WS, Champion HR, Sacco WJ, et al. The injury severityscore revisited. J Trauma 1988;28:69.

36. Champion HR, Copes WS, Sacco WJ, et al. The major traumaoutcome study: establishing national norms for trauma care.J Trauma 1990;30:1356.

37. Ali J, Adam R, Gana TJ, Bedaysie H, Williams JI. Effects ofpre-hospital trauma life support programme on pre-hospitaltrauma care. J Trauma 1997;42(5):786—90.

38. Jacobs LM, Sinclair A, Beiser A, et al. Pre-hospital advancedlife support: benefits in trauma. J Trauma 1984;24:8.

39. Smith JP, Bodai BI. The urban paramedic’s scope of practice.JAMA 1985;253:544.

40. Smith JP, Bodai BI, Hill AS, et al. Pre-hospital stabilization ofcritically injured patients: a failed concept. J Trauma 1985;25:65.

41. Pepe PE, Stewart RP, Copass MK. Pre-hospital management oftrauma: a tale of three cities. Ann Emerg Med 1986;15:1484.

42. Border JR, Lewis FR, Aprahamian C, et al. Panel: pre-hospitaltrauma care–—stabilize or scoop and run? J Trauma 1983;23:708.

43. Reines HD, Bartlett RL, Chudy NE, et al. Is advanced lifesupport appropriate for victims of motor vehicle accidents:the South Carolina highway trauma project. J Trauma 1988;28:563.

44. Lewis FR. Pre-hospital intravenous fluid therapy: physiologiccomputer modeling. J Trauma 1986;26:804.

45. Hamilton SM, Breakey P. Fluid resuscitation of the traumapatient: how much is enough? CJS 1996;39:11.

46. Bickell WH, Wall MJ, Pepe PE, et al. Immediate vs. delayedfluid resuscitation for hypotensive patients with penetratingtorso injuries. N Eng J Med 1994;331:1105.

Development of pre-hospital trauma-care system 587

47. Jacobson LE, Gomez GA, Sobieray RJ, et al. Surgical cri-cothyroidotomy in trauma patients: analysis of its use byparamedics in the field. J Trauma 1996;41:15.

48. Durham III LA, Richardson RJ, Wall MJ, Pepe PE. Emergencycentre thoracotomy: impact of pre-hospital resuscitation. JTrauma 1992;32:775.

49. Rhee KJ, O’Malley RJ. Neuromuscular blockade-assisted oralintubation in the pre-hospital care of the injured patients.Ann Emerg Med 1994;23:37.

50. Baxt WG, Moody P. The impact of advanced pre-hospitalemergency care on the mortality of severely brain-injuredpatients. J Trauma 1987;27:365.

51. Margolin DA, Johann DJ, Fallon WF, Malangoni MA. Responseafter out-of-hospital cardiac arrest in the trauma patientshould determine aeromedical transport to a trauma centre.J Trauma 1996;41:721.

52. Mass casualties triage. In: Wiener SL, Barrel J, editors.Trauma management for Civilian and Military Physicians.Philadelphia: WB Saunders; 1986. p. 536.

53. Super G, Groth S, Cleary V. START: a training triage module.Newport Beach, CA: Hoag Presbyterian Memorial Hospital;1983/1984.

54. Triage and evacuation. In fieldmedical service school studentmanual camp. Pendleton CA; 1992.

55. Riddick L, Long W, Copes W, et al. Automated coding ofinjuries from autopsy reports; quoted by Champion HR inTrauma. 3rd ed. USA: Appleton and Lange; 1996, p. 66.

56. Sharma BR. Clinical forensic medicine–—management ofcrime victims from trauma to trial. J Clin Forensic Med2003;10(4):267—73.

57. Rutledge R, Fakhry S, Baker C, et al. Injury severity grading intrauma patients: a simplified technique based upon ICD-9coding. J Trauma 1993;35:497.

58. Long W, Sacco W, Copes W, et al. An evaluation of experthuman and automated abbreviated injury scale and ICD-9-CMinjury coding; quoted by Champion HR in Trauma. 3rd ed.USA: Appleton and Lange; 1996, p. 66.