Adopting the pre-hospital index for interfacility helicopter transport: a proposal

Transcription

1 Injury, Int. J. Care Injured 34 (2003) 3 11 Adopting the pre-hospital index for interfacility helicopter transport: a proposal Leanelle Goldstein, Christopher J. Doig, Sarah Bates, Sandra Rink, John B. Kortbeek Division of Critical Care, Department of Surgery, Trauma Services, Foothills Medical Centre, University of Calgary, Street NW, Calgary, Alta., Canada T2N 2T9 Accepted 2 April 2002 Abstract Background: Interfacility helicopter transport is expensive without proven outcome benefit in trauma patients. Our objectives were to determine the fastest method of rural to urban interfacility transport, and develop a triage tool to identify patients most in need of rapid transport. Methods: Retrospective cohort study. Adults ISS 12 transported from January 1996 to December Transport time variables were compared between geographical zones. A pre-transport index (PTI) identified two patient cohorts in which outcome was assessed. Results: Air ambulance was faster than ground transport, with helicopter overall superior to fixed-wing (<225 km range). Seventy-two percent of patients with PTI < 4(n = 196) had no outcome indicating severe injury versus 29% of the PTI 4 cohort (n = 151). Mortality for PTI < 4 was 1.4% versus 22% for PTI 4. Conclusion: Interfacility helicopter transport of severely injured rural trauma patients was the overall fastest method within a 225 km range. PTI > 4 identifies patients most in need of this fast but expensive method of transport Elsevier Science Ltd. All rights reserved. 1. Introduction The need to provide rapid access to tertiary care for critically injured patients from both urban and rural locations is an essential element of a regionalized trauma care system [1 3]. Approximately 70% of trauma deaths occur in rural areas with an average death rate for rural trauma patients twice as high as for patients in urban centers [4 6]. The American College of Surgeons Committee on Trauma has stated that, in any trauma system, the goal is to transport the injured patient to definitive care in the shortest possible time [1]. Implicit in this position, and given the relatively higher mortality of rural trauma patients, it is an imperative for regional trauma centres receiving patients from rural centres to identify the best method of transfer [7,8]. Studies supporting the use of helicopters in air ambulance programs have identified two major factors contributing to improved patient outcome: level of skill provided by the transport team and shorter transport times to definitive care [9 11]. However, helicopters are an expensive resource, limiting their use in many jurisdictions. Fixed-wing and ground transport with similar equipment and staff may offer equivalent care and outcome but with less cost. Although studies have explored the association of transport team composition, none have analyzed the time elements involved in a Corresponding author. rural to tertiary care transportation system [12]. An analysis of these time elements may be useful in determining the most effective method of transport and for quality assurance within existing systems. Effective triage and transport of a trauma patient is a complex process of appropriately allocating resources for each patient situation. For rural trauma patients, these decisions include which patient, what method of transport (ground versus fixed-wing versus helicopter), and whether the patient should be triaged to a secondary (local) or tertiary (level 1) trauma centre. Unfortunately, standardized criteria to assist in these triage decisions do not exist. Attempts have been made to facilitate decision making using patient-based survey systems [13]. The most widely used pre-hospital scoring systems in the past have been the trauma score and revised trauma scores [14,15]. A more recent revision of the trauma score, the pre-hospital index (PHI), is currently used by several trauma systems in Canada. Its accuracy in conjunction with mechanism of injury criteria has recently been published [16,31]. As indicated by name, the PHI is intended for use by pre-hospital personnel at the scene of an accident. The situation for rural trauma patients is unique in that a qualified physician makes triage decisions for transport to tertiary care in the setting of a rural hospital emergency department, often after initial resuscitation measures. A more representative assessment of the severity of injury of a trauma patient in a /02/$ see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S (02)

2 4 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) 3 11 rural hospital would be a pre-transport evaluation, rather than a pre-hospital evaluation. The objectives of this study were: (1) to explore time related elements involved in the interfacility transfer of critically injured patients from rural hospitals to regional tertiary care (level 1) trauma centers, and to determine the fastest method of transport for any trauma patient injured at a distance from the trauma center; and (2) to develop and test a triage tool, the pre-transport index (PTI), which would allow us to identify those patients with severe injury who require the most rapid transport to a level 1 trauma center. 2. Materials and methods The southern Alberta trauma system was analyzed as part of a regional patient care and outcome process program (a comprehensive program of quality assurance). The Foothills Medical Center is a tertiary (level 1) trauma center and serves as the referral center for southern Alberta and southeastern British Columbia, a population of approximately 1.4 million. The geographic area is approximately 200,000 km 2, running from Golden, British Columbia to the Saskatchewan border, and from Red Deer to the Montana border. The provincial air ambulance program includes fixed-wing King Air 100 bases in Calgary and Medicine Hat and the non profit Shock Trauma Air Rescue Service (STARS) BK 117 helicopter in Calgary. Data for this study was extracted from two data sources: the Southern Alberta Trauma Registry and the Trauma Ambulance Audit Database. The Southern Alberta Trauma Registry records all trauma admissions with ISS 12. Data is collected from pre-hospital patient care records, referring hospital records and trauma center records. The Southern Alberta Trauma Registry is a collector database Fig. 1. Map of Alberta with distance from level 1 trauma center in Calgary.

3 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) (Tri-Analytics Inc., Bel Air, MD) compiled by a trained study nurse (SR) who reviews pre-hospital records and patient charts for demographic data and outcomes. The Trauma Ambulance Audit Database was developed by a consensus conference of local trauma, emergency, and pre-hospital personnel in concert with local and regional health officials. The data is collected in a manner similar to the Southern Alberta Trauma Registry, however, the database was developed locally using Access (Microsoft Corporation, Seattle, WA). The first phase of the study was to examine time related elements involved in the transfer of critically injured patients from rural areas outlying Calgary to trauma care in Calgary s level 1 trauma center. An air ground ambulance trauma and critical care audit committee was formed, and included local department heads for trauma, critical care, and emergency medicine, ground and air ambulance personnel and government health officials. Potential time factors involved in rural to tertiary care patient transfers by ground, fixed-wing or rotary wing service were identified by consensus of the 15 members of the committee. Data on all time factors was collected prospectively from 1 January 1996 to 1 December 1996 for all trauma registry patients over age 15 with ISS 12 transported from outside the Calgary city limits. Each record was classified by geographical distance from the level 1 trauma center in Calgary: zone 1, 125 km; zone 2, km; zone 3, 225 km (Fig. 1). A priori, three items were considered as the essential and critical time elements to be assessed: (1) injury to trauma center; (2) dispatch to trauma center; and (3) out of hospital at risk time. (1) Time of injury to trauma center is defined as the initial recorded time of EMS dispatch to the scene of the accident, to arrival at the trauma center inclusive of all time spent at the scene, in the peripheral hospital and during transport. (2) Time of dispatch to trauma care is a calculated sum of all time elements from the initiation of the request for air transfer to arrival at the trauma center. This does not include time spent in sending hospitals prior to dispatch of an air ambulance. Dispatch to sending hospital times for helicopter versus fixed-wing are a reflection of preparation for take-off, including boarding of all crew members, actual flight times, and for fixed-wing, ground travel between airport and the rural hospital, as well as airport to the tertiary care center. For ground transport, time of dispatch to trauma care is from the time of departure of the local ground ambulance to arrival at the trauma centre (the rural ground ambulance services are based within or immediately adjacent to the hospitals). (3) Out of hospital at risk time reflects the travel time between hospitals when the only resources available for the patient are those provided by the transport team within the ground or air ambulance. Data on these three time factors is reported. The second objective of the study was the development of a triage tool to identify patients in most need of rapid transport. The PTI was developed as a modification of the Table 1 The PHI and PTI Variable PHI PTI Score Blood pressure >100 > Heart rate =120 = <50 <50 5 Respiratory status Normal Normal 0 Labored/shallow Labored/shallow/ 3 pneumothorax/rate >20 Rate <10 Needs intubation/rate 10 5 Level of consciousness Normal Normal 0 Confused/combative Confused/combative/amnestic 3 No intelligible words Needs intubation/ no intelligible words 5 Penetrating chest/abdominal trauma No No 0 Yes Yes 4 PTI or PHI score of 4 is considered major trauma. original PHI in order to account for assessment and interventions at the rural hospital (Table 1). Changes to the PHI for development of the PTI were made by expert consensus prior to the onset of the study. These included additional scores for intubation/pneumothorax. Data for calculation of the PTI was acquired by detailed review of the transferring records conducted by two of the authors (L.G., S.B.) and reviewed by a third author (C.D.). Objective criteria for data selection were developed a priori. First priority was given to variables documented directly by the rural physician during trauma resuscitation. If multiple values were present, the value consistent with more severe injury was utilized. This was done to represent more accurately the situation in the rural emergency department in which the final triage decision most likely reflects patient status during the overall resuscitation phase and assessment in that hospital. For example, a patient who was hypotensive (SBP 80)atany time during the rural hospital assessment was assigned a PTI score of 2 for blood pressure, even if hypotension resolved with fluid resuscitation. In the event that the physician documented terms (hypotensive, tachycardia, etc.) rather than specific values, the PTI score assigned was the lowest value that would indicate that the vital sign had been abnormal. Vitals noted prior to assessment and initiation of resuscitation by the rural MD were not used. In the event that documentation from the rural physician was not present in the chart, the PTI was calculated based on transport team or trauma center team records that referenced patient data from the rural hospital. For example, the records from the

4 6 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) 3 11 transport team or trauma center admission may refer to the fact that the patient had been hypotensive and confused at the rural hospital. If no specific rural hospital vitals or reference to these vitals was documented, the first vital signs documented by the transport team were used. The first vital signs recorded by the transport team are when the patient is first encountered in the rural hospital prior to transport team intervention or transport. A PHI of 4 has previously been validated as an accurate pre-hospital triage tool [16]. A PTI of 4 or greater was used as a cut-off to maintain consistency with this previous literature. The PTI was applied to all trauma registry patients over age 15 with ISS 12 transported to Calgary between 1 January 1997 and 31 December Outcomes selected as indicators of severe injury were: (1) ICU stay >48 h; (2) ICU ventilator requirement >24 h; and (3) mortality after admission to the trauma center. The presence of one or more of these outcomes was considered evidence of severe or major injury. Data was entered into an Access database (Microsoft Corporation, Seattle, WA) and analyzed using SPSS software (SPSS Inc., Chicago, IL). PTI scores were calculated and related to the following three categories of outcome variables: (1) mortality; (2) one or more of the outcomes indicating severe injury (major injury); or (3) none of the outcomes indicating severe injury (minor injury). Dichotomous variables were analyzed by Fishers Exact test and the Mann Whitney U test for continuous variables. The local institutional review board responsible for human research approved the study. 3. Results One hundred and seventy-seven trauma transports met inclusion criteria for analysis of transport data (Table 2). The mean age of all trauma patients was just over 40 years with patients from zone 3 being slightly older than patients from zones 1 or 2. The predominance of males is consistent with the expected gender distribution for trauma patients. The injury severity score for patients in all zones is consistent with major trauma, with patients from zone 3 having a clinically insignificant lower score than patients from zones 1 and 2. There was a slightly higher mortality rate in zone 2 patients. Over half of the transports were within zone 1, a third from zone 2 and the remainder from zone 3 (Fig. 2). In zone 1, helicopter was the fastest method of transport (P <0.01) from time of injury to tertiary care. In zone 2, helicopter was faster than both ground and fixed-wing (P <0.01). In zone 3, fixed-wing was fastest (P <0.01 (Fig. 3). Dispatch to tertiary care time is displayed in Fig. 4, and out of hospital at risk times are displayed in Fig. 5. Ground and helicopter times appear similar for zone 1, however, the data do not account for the absence of a central dispatch call for ground transports, and therefore the dispatch time for ground transport may be underestimated. Air ambulance clearly reduces out of hospital at risk times for all zones with helicopter superior for distances <225 km (zones 1 and 2) (P <0.01). Fixed-wing was superior in zone 3 (>225 km) (P <0.01). For the second phase of the study, 1 January December 1998, 347 patients met inclusion criteria and had Table 2 Demographic data on trauma patients by zone of referral Zone 1 (0 125 km; n = 91) Zone 2 ( km; n = 53) Zone 3 (>225 km; n = 33) Age a 43.7 (19.4) 40.8 (18.9) 47.1 (21.3) Male gender (%) ISS a 25.5 (11.3) 25.7 (9.7) 21.3 (6.5) TRISS a (0.283) (0.268) (0.188) ICU length of stay (days) b 2 (0, 4.5) 1 (0, 4.0) 1 (0, 5.5) Hospital length of stay (days) b 12.0 (8.0, 26.0) 11.0 (5.3, 20.0) 7.0 (5.5, 18.0) Outcome (all cause hospital mortality) (%) The TRISS and ISS scores are calculated based on published guidelines [19,33]. The TRISS score is based on the average of patients in whom the data could be collected: intubated patients were as per the TRISS criteria excluded. a Reported as mean (S.D.). b Reported as median (25th, 75th percentiles). Fig. 2. Number of transports by zone.

5 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) Fig. 3. Time of injury to tertiary care by zone by method of transport. sufficient information recorded to calculate the PTI. One hundred and ninety-six patients had PTI scores <4, and 151 patients had PTI score 4. The age and gender distribution were not different between study groups, and again are consistent with what is representative of a trauma population (Table 3). The ISS was slightly higher in the PTI 4 group. The ICU and hospital lengths of stay were different between study groups. All outcome measures indicative of a severe injury were significantly different when the PTI < 4 cohort was compared to the PTI 4 cohort (Table 4). Only three patients (1.5%) in the PTI < 4 group died compared to 33 (22%) in the PTI 4 group (P <0.001). Fifty-one (26%) of patients in the PTI < 4 group required mechanical ventilation for more than 24 h, and only 21 (18%) had an ICU length of stay >48 h. This compares to 90 patients (60%) who required ventilation >24 h and 75 patients (50%) who required ICU stay >48 h in the PTI 4 group. Obviously, length of ICU stay and mechanical ventilation may be effected by physician practice patterns, and hospital Fig. 4. Time from dispatch to level 1 trauma care by zone by method of transport.

6 8 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) 3 11 Fig. 5. Time at risk by zone and method of transport. Table 3 Demographic data on trauma patients by PTI PTI < 4 (n = 196) PTI 4 (n = 151) Age a 39.6 (20.2) 40.9 (20.1) Male gender (%) ISS a 20.0 (6.6) 23.2 (9.6) TRISS a (0.070) (0.136) ICU length of stay (days) b 0 (0, 0) 1.0 (0, 5.0) Hospital length of stay (days) b 10.0 (6.0, 14.0) 14.5 (8.5, 21.0) The TRISS and ISS scores are calculated based on published guidelines [19,33]. The TRISS score is based on the average of patients in whom the data could be collected: intubated patients were as per the TRISS criteria excluded. a Reported as mean (S.D.). b Reported as median (25th, 75th percentiles). occupancy factors. However, there was no evidence to suggest that these factors were systematically different between study groups so as to bias interpretation of results. PTI < 4, as a predictor of survival, had a sensitivity of 92%, a specificity of 62%, a positive predictive value of 98%, and a negative predictive value of 22%: a PTI < 4is Table 4 Presence of severe injury by PTI PTI < 4 (n = 196) PTI 4 (n = 151) In-hospital mortality (yes) 3 (1.5%) 33 (22%) <0.001 ICU length of stay >48 h 21 (18%) 75 (50%) <0.001 ICU ventilator days >24 h 51 (26%) 90 (60%) <0.001 One or more of above outcomes indicating severe injury 54 (28%) 106 (71%) <0.001 P associated with a 98% probability of survival and only a 2% chance of death. 4. Discussion Fixed-wing aircraft were first used to transport the wounded in World War 1 [9]. The first dedicated helicopter ambulance system was established in Denver in Retrospective studies supporting air ambulance systems include work by Moylan et al. [17] who demonstrated a significant survival advantage for patients with trauma scores between 10 and 5, transported from rural areas to tertiary care via helicopter [14]. Similarly, Boyd observed a 25.4% reduction in predicted mortality of rural trauma patients transferred by helicopter [7]. However, there have also been numerous studies that failed to attribute improved outcome to rapid transport alone, but rather demonstrated a reduction in mortality associated with the presence of a physician on either ground or helicopter transport [10,12,18,19]. Unfortunately, the widespread belief that helicopter transport itself improves outcome for critically ill and injured patients has made it impossible to obtain supporting level 1 evidence [20]. Recently, Sampalis et al. published a study evaluating the impact of implementing a regionalized trauma system [21]. Among the many variables studied, reductions in both pre-hospital time and time to definitive (level 1) care were significantly associated with improved survival. However, the controversy persists surrounding the effectiveness of the more expensive modes of transport and the selection of those patients most likely to benefit from the most rapid transport system [22 24]. The challenge still remains to develop a

7 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) regional access, referral and transportation system which ensures priority access in the minimum time for that minority of patients who are most severely injured and physiologically unstable, while maintaining a safe and cost-effective system for the majority of other patients [25,26]. Therefore, in this era of health care funding restraints, it is ideal if the patients are triaged so that the patients most likely to benefit from the most rapid transport are sent by the fastest method, while allowing patients without major injury to be transported by an alternative method. The current study is a first step in the process of ultimately resolving controversy regarding the most effective method of transferring patients from rural hospitals to trauma centers for definitive care. This was done by first defining the anatomy of a rural patient transport and all of the time variables involved, then by determining the fastest method of transport for each of three geographical zones. Specific variables analyzed were the out of hospital at risk time, the time from dispatch to tertiary care, and the time of injury to tertiary care. Examination of each of these variables for any rural location within the local trauma system allowed us to develop an accurate means of defining the fastest method of transport within each of the three specified zones. Our study demonstrated that air transport was faster than ground for the three a priori defined geographical zones for interfacility transfers. This study did not address direct ground transfers versus scene flights. The out of hospital at risk time is the most direct reflection of travel time over distance for each of the three ambulance systems. The rapid air speed of fixed-wing aircraft is somewhat negated by the time required for airport take-off, travel from rural hospital to airport, and transport from receiving airport to trauma center. When considering this variable, it is important to recognize that in addition to the time involved, the true risk to the patient is dependent on the level of expertise of the transport team and the facilities within the transport vehicle. It has been observed that with ICU level specialty specific care, long transport time itself may not be associated with poor outcome [27]. However, given that the level of care provided en route is variable and a subset of trauma patients require surgery to manage truncal/head injuries, it remains prudent to provide the most rapid transport possible to critically injured patients. Identification of these most severely injured patients prior to transport was the objective of the second phase of this study. The second objective of the present study was to develop a triage tool specifically designed for use in trauma system transport dispatch in the setting of the rural hospital trauma resuscitation. The widely used PHI provided the basis of this new triage tool [16]. The PHI is a triage oriented trauma severity scoring system consisting of five elements: systolic blood pressure, pulse, respiratory status, level of consciousness and presence of penetrating abdominal or chest trauma [16]. The calculated sum is used to classify patients as minor trauma (PHI 0 3) or major trauma (PHI 4). In the original development and validation PHI study, patients defined as major trauma were those who required emergency surgery within 24 h, or those who died within 72 h of admission. A PHI 4 was found to be a sensitive predictor of major trauma based on these criteria (P < 0.001). Prospective validation of the PHI was performed by Koehler et al. [28] who applied the score in a multi-center trial and found it to be accurate in predicting need for emergency surgery within 4 h (P < ) and mortality within 72 h (P < ). A PHI of 4 is a triage tool with high specificity in identifying major trauma defined by ISS 16. In an urban system, it has been validated as an accurate triage tool when combined with mechanism of injury criteria [29 31]. Whether a tool, applicable in an urban and regional trauma system, could be applied to a rural population with prolonged pre-hospital times and times to definitive care had not been demonstrated. In development of the PTI, modifications of the PHI were made to allow for clinical evaluation and procedures performed by the rural physician during the resuscitation. These modifications were agreed upon by consensus of available expert opinion based on a pilot study. The major modifications included scores for the physician s assessment that airway protection or head injury protocol necessitated intubation, as well as scores for pneumothorax requiring chest tube. The PTI score was divided into <4 versus 4 in keeping with the original PHI validated outcomes. In general, a score of at least 4 points implies either significant alteration of a single variable, minor alteration in two or more variables, or the presence of penetrating chest or abdominal trauma. A PTI score <4 signifies minor injury, as compared to PTI score of 4, which indicates major injury. The underlying hypothesis of this study is that patients with higher PTI scores in the rural hospital would experience outcomes consistent with severe or major injury during admission to the trauma center. Conversely, patients with PTI scores <4 would be unlikely to experience these outcomes. The outcomes selected were: (1) ICU stay >48 h; (2) ICU ventilator requirement >48 h; and (3) mortality. These outcomes were chosen based upon existing research including the original PHI study [16], the major trauma outcome study [32] and a variety of other citations from trauma and critical care literature. Overall, this study demonstrated that PTI score of <4 versus 4 was able to identify patients who had outcomes consistent with major injury. Seventy-two percent of the patients with PTI < 4 had none of the indicators of severe injury. Only 29% of the patients with PTI 4 could be considered to have minor injury (P < 0.001). Conversely, only 28% of the patients with PTI < 4 had major injury, versus 71% of the patients with PTI 4 had at least one outcome indicating severe injury (P < 0.001). Therefore, our study has demonstrated that PTI can accurately identify patients with major trauma as determined by clinical outcome, and therefore could be used to prioritize transport and allocate appropriate use of limited and expensive air ambulance and particularly helicopter services.

8 10 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) 3 11 Though a retrospective review such as this is subject to bias, attempts were made to control for potential sources of error. All trauma registry data was entered at the time of discharge. The air transport times were collected during the study period and entered monthly. PTI data was entered by two of the investigators after collection of all transport times. There are important limitations of the study. Only patients who survived their accident to be transported to the rural hospital, and who survived the initial resuscitation and transport to the trauma center, were included. Inclusion of deaths in a prospective study will be essential to evaluate the impact of survival with long transport times. A certain amount of subjective assessment was required for collection of PTI data. Attempts were made to control for this by reviewing a pilot sample and specifying any changes required for PTI calculation (prior to data collection for the study itself). As well, the prioritization of data collection (rural physician documentation, before reference to rural hospital vitals, then initial transport team vitals) was assigned and adhered to for each chart reviewed. Finally, the study was not designed to evaluate the influence of the transport time variables or variation in care at the sending hospitals on mortality. Each of these will be examined during future prospective validation of the PTI. We believe the results of this study support a trial to triage to mode of transport based on PTI. Helicopters will be reserved for PTI 4, within the defined operating radius (225 km). Trauma patients with PTI < 4 being referred to the tertiary trauma center will be transported by ground, if <125 km, and by fixed-wing for distances >225 km. 5. Conclusion In this era of health care funding restraints, the appropriate allocation of such limited and costly resources as air ambulance transport must be examined. The challenge is thus the development of a regional access referral and transport system which ensures priority access in the minimum time for that minority of patients who are most severely injured, while maintaining a safe and cost-effective system for the majority of other patients [19,20]. This has been a multi-step process involving first the definition of the time requirements for each transport method (ground versus fixed-wing versus helicopter) for rural locations at various distances from the trauma center. Then, effectively determining the most rapid method of transport for each of three defined geographical zones. Finally, the present study then involved the development of a rural trauma triage tool, the PTI, based upon the existing PHI [16,28]. The PTI has been retrospectively applied to 2 years of rural trauma transports. Specific outcomes were defined as indicators of severe injury, and using PTI < 4 versus PTI 4, we were able to reliably identify those patients unlikely to have severe injury outcomes (minor injury), as well as those more likely to have severe or major injury. Through identification of these patient groups, the PTI may be useful in the identification of patients who should be transported by the fastest possible method, as well as those who are safe to transport by an alternative more economical method. Limitations of this study include the small sample size, the retrospective nature of the data, and the fact that mortality and morbidity outcomes were not specifically related to transport times, but rather method of transport. This data is currently being collected for future evaluation. Finally, the PTI must be validated prospectively in order to determine its true utility in the triage of rural trauma patients. Accuracy, over, and under triage rates will have to be determined prospectively. PTI demonstrates potential as a trauma triage tool developed specifically for use in determining transport mechanism for rural trauma patients. References [1] Resources for optimal care of the injured patient. American College of Surgeons Committee on Trauma, [2] American College of Surgeons Committee on Trauma, [3] American College of Surgeons Committee on Trauma. Advanced trauma life support manual, [4] Cales RH. Trauma morbidity in Orange County: the effect of implementation of a regional trauma system. Ann Emerg Med 1984; 13(1): [5] Leicht MJ, Dula DJ, Brotman S, Anderson TE, Gessner HW, Parrish GA, et al. Rural interhospital helicopter transport of motor vehicle trauma victims: causes for delays and recommendations. Ann Emerg Med 1986;15(4): [6] Baker SP. The injury factbook. 2nd ed. Oxford University Press, New York, [7] Thomas F, Wisham J, Clemmer TP, Orme Jr JF, Larsen KG. Outcome, transport times and costs of patients evacuated by helicopter versus fixed-wing aircraft. West J Med 1990;153:40 3. [8] Urdaneta LF, Miller BK, Ringenberg BJ, Cram AE, Scott DH. Role of an emergency helicopter transport service in rural trauma. Arch Surg 1987;122: [9] Meier DR. Evolution of civil aeromedical helicopter aviation. South Med Journal 1989;82(7): [10] Baxt WG, Moody PM. The impact of a rotorcraft aeromedical emergency care service on trauma mortality. JAMA 1983;249(22): [11] Boyd CR, Corse KM, Campbell RC. Emergency interhospital transport of the major trauma patient: air versus ground. J Trauma 1989;29(6): [12] Hamman BL, Cue JI, Miller FB, O Brien DA, House T, Polk HC, et al. Helicopter transport of trauma victims: does a physician make a difference? J Trauma 1991;31(4): [13] Maslanka AM. Scoring systems and triage from the field. Emerg Med Clin North Am 1993;11(1): [14] Champion HR, Sacco WJ, Carnazzo AJ, Copes W, Fouty WJ. Trauma score. Crit Care Med 1981;9(9): [15] Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA, Flanagan ME. A revision of the trauma score. J Trauma 1989;29(5): [16] Koehler JJ, Baer LJ, Malafa SA, Meindertsma MS, Navitskas NR, Huizenga JE. Pre-hospital index: a scoring system for field triage of trauma victims. Ann Emerg Med 1986;15: [17] Moylan JA, Fitzpatrick KT, Beyer III AJ, Georgiade GS. Factors improving survival in multi-system trauma patients. Ann Surg 1988; 207(6):

9 L. Goldstein et al. / Injury, Int. J. Care Injured 34 (2003) [18] Schiller WR, Knox R, Zinnecker H, Jeevanandam M, Sayre M, Burke J, et al. Effect of helicopter transport of trauma victims on survival in an urban trauma center. J Trauma 1988;28(8): [19] Boyd CR, Tolson MA. Copes WS. Evaluating trauma care: the TRISS method. J Trauma 1986;27(4): [20] Burney RE, Fischer RP. Ground versus air transport of trauma victims: medical and logistical considerations. Ann Emerg Med 1986; 15(12): [21] Sampalis JS, Denis R, Lavoie A, Frechette P, Boukas S, Nikolis A, et al. Trauma care regionalization: a process-outcome evaluation. J Trauma 1999;46(4): [22] van Wijngaarden M, Kortbeek J, Lafreniere R, Cunningham R, Joughin E, Yim R. Air ambulance trauma transport: a quality review. J Trauma 1996;41(1): [23] Burney RE. Efficacy, cost and safety of hospital-based emergency aeromedical programs. Ann Emerg Med 1987;16(2): [24] Champion HR. Helicopters in emergency trauma care. JAMA 1983;249(22): [25] Macione AR, Wilcox DE. Utilization prediction for helicopter emergency medical services. Ann Emerg Med 1987;16(4): [26] Baxt WG, Moody P, Cleveland HC, Fischer RP, Kyes FN, Leicht MJ, et al. Hospital based rotorcraft aeromedical emergency care services and trauma mortality: a multi-center study. Ann Emerg Med 1985;14(9):61 6. [27] Jacobs LM, Bennett B. A critical care helicopter system in trauma. J Natl Med Assoc 1989;81(11): [28] Koehler JJ, Malafa SA, Hillesland J, Baer LJ, Rogers RN, Navitskas NR, et al. A multi-center validation of the pre-hospital index. Ann Emerg Med 1987;16: [29] Baxt WG, Berry CC, Epperson MD, Scalzitti V. The failure of prehospital trauma prediction rules to classify trauma patients accurately. Ann Emerg Med 1989;18:1 8. [30] Plant JR, MacLeod DB, Kortbeek JB. Limitations of the pre-hospital index in identifying patients in need of a major trauma center. Ann Emerg Med 1995;26(2): [31] Bond RJ, Kortbeek JB, Preshaw RM. Field trauma triage: combining mechanism of injury with the pre-hospital index for an improved trauma triage tool. J Trauma 1997;43(2): [32] Champion HR, Copes WS, Sacco WJ, Lawnick MM, Keast SL, Bain Jr LW, et al. The major trauma outcome study: establishing national norms for trauma care. J Trauma 1990;30(11): [33] Baker SP, O Neill B, Haddon Jr W, Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14(3):