a b s t r a c t

Purpose: The beneficial effect of the presence of an emergency physician in prehospital major trauma care is con- troversial. The aim in this study is to assess whether an emergency physician on scene can improve survival out- come of critical trauma patients.

Methods: This retrospective cohort study was conducted by using Nationwide trauma registry data between 2004 and 2013 in Japan. Severe trauma patients (injury severity score (ISS) >= 16) who were transported directly to the hospital from the injury site were included in our analysis. Patients who were predicted to be untreatable (abbre- viated injury score (AIS) = 6 and/or cardiopulmonary arrest at least one time before hospital arrival) were ex- cluded. Participants were divided into either a physician or paramedics group based on the prehospital practitioner. The primary outcome was survival rate at discharge. Multivariable logistic regression analysis was performed to compare the outcome with adjustment for age, gender, ISS, cause of injury, and pre-hospital vital signs.

Results: A total of 30,283 patients were eligible for the selection criteria (physician: 1222, paramedics: 29,061). Overall, 172 patients (14.1%) died in the physician group compared to 3508 patients (12.1%) in the paramedics group. Patients in the physician group had higher ISSs than those in the paramedics group. In multivariable logis- tic regression, the physician group had an odds ratio (OR) of 1.16 (95% confidence interval (CI) = 0.97 to 1.40, p

= 0.11) for in-hospital survival.

Conclusions: Our results failed to show a difference in survival at discharge between non-physician-staffed ambu- lances and physician-staffed ambulances.

(C) 2018

Introduction

Major trauma comprising serious and multiple injuries is often life threating. As some deaths after trauma were commonly considered pre- ventable only if injuries were detected and treated rapidly and accu- rately [1], the concept that a physician deployment to the scene of trauma for the prehospital rapid advanced intervention might improve the outcome of major trauma patients has been raised. In fact, studies have demonstrated that prompt initiation of advanced techniques such as intubation and chest decompression before hospital arrival re- duce mortality in patients with severe trauma [2,3]. Nevertheless, the efficacy of the presence of an emergency physician in the pre-hospital setting is still controversial and debated. The main argument against the involvement of an emergency physician in prehospital trauma

* Corresponding author at: Department of Emergency and Critical Care Medicine, Juntendo University Urayasu Hospital, 2-1-1 Tomioka, Urayasu, Chiba 279-0021, Japan.

E-mail addresses: [email protected] (Y. Hirano), [email protected] (H. Tanaka).

care is that it could delay Transfer time due to unnecessary Prehospital interventions [4]. Several reports have highlighted the importance of rapid patient transfer to the trauma center and of definitive trauma care, including surgery, which strongly influence outcome [5-7]. More- over, involvement of an emergency physician requires considerable fi- nancial and human resources. Therefore, it is important to demonstrate the benefits of direct physician involvement in the field, such as improved survival outcome.

Emergency medical services (EMS) and laws vary around the globe. In Japan, EMS generally adopt the Anglo-American model in which paramedics are the only prehospital scene responders. For prehospital trauma patients alive, Japanese paramedics are permitted to perform limited prehospital interventions such as intravenous administration (since 2014), a supraglottic airway device use (since 1991) and intuba- tion (since 2014) if they finished special training program. Thus, Japanese paramedics have performed medical procedures on the scene less actively than paramedics in the United State (U.S.) and Europe histrorically [8]. On the other hand, some districts have their own physician-staffed EMS system in Japan. They can perform several

https://doi.org/10.1016/j.ajem.2018.11.014

0735-6757/(C) 2018

advanced procedures including ultrasound examination, intravenous administration, intubation, thoracotomy, Resuscitative endovascular balloon occlusion of the aorta (REBOA) when the equipment was avail- able. Based on these regional differences, trauma patients in Japan rep- resent an interesting population to compare the outcome of prehospital care provided by a physician versus paramedics. In this study, we aimed to investigate whether the presence of an emergency physician on scene in the pre-hospital setting improves outcome among patients with major trauma.

Methods

Study design and data collection

This retrospective cohort study investigated the effectiveness of the involvement of an emergency physician versus paramedics in pre- hospital trauma care. The data were obtained from the Japan Trauma Data Bank (JTDB), which was established in 2003 by the Japanese Asso- ciation for the Surgery of Trauma (Trauma Registry Committee) and the Japanese Association for Acute Medicine (Committee for Clinical Care

Evaluation) to assure the quality of trauma care in Japan. During the study period, a total of 244 hospitals including over 90% of tertiary emergency medical centers in Japan participated in the JTDB [9]. The tertiary emergency medical centers are major hospitals that usually have a full complement of services including major operations, consul- tations with sub-specialists. There are no specific trauma centers in Japan, but tertiary emergency medical centers are considered to be equivalent to level I or II trauma centers in the U.S. Japan Prehospital trauma evaluation and Care (JPTEC) recommends that major trauma patients such as high-energy trauma patients should be transferred rap- idly to the nearest tertiary emergency medical center. In most cases, the physicians who attended the AIS coding course registered the data. Dur- ing the study period, AIS90 Update 98 version was used for AIS coding. The JTDB variables include patient characteristics, injury type, cause of injury, transportation type (ambulance with or without a physician, he- licopter with physician), Prehospital vital signs and treatments (oxy- genation, cervical collars, backboard, ventilation, intubation, nasal airway tube, and intravenous fluid), Abbreviated Injury Scale score, injury severity score (ISS), survival at discharge and in the emer- gency department (ED), disposition at ED, and information regarding

Fig. 1. Flow diagram of participants selection procedure. JTDB: Japan trauma data bank, EMS: emergency medical service, ISS: injury severity score, AIS: abbreviated injury scale.

the timing of events such as pre-hospital contact, hospital arrival, CT scans, blood transfusion, and surgery. The Shock Index, the Revised Trauma Score , and the Probability of survival based on the Trauma and Injury Severity Score were calculated using these

Table 1

Baseline characteristics of the study population

Characteristics Physician (n = 1222)

Paramedics (n = 29,061)

p

value

data.

Age (years) 52.6 +- 22.1 54.9 +- 22.7 b0.001

Gender (male) 873/1222 (71.4%)

20,482/29,051 (70.5%)

0.646

Ethical approval and consent to participate

The ethics committee at our institution approved JTDB data analysis (29-061). The requirement for patient consent was waived as this was an observational study using anonymous data. Permission for the use of the data was granted by the JTDB.

Injury type (blunt vs penetrating)

Cause of injury

Motor vehicle crash	804/1222	15,267/29,061	b0.001
	(65.8%)	(52.5%)
Falls	322/1222	12,917/29,061	b0.001
	(26.4%)	(44.4%)
Other reasons	96/1222 (7.9%)	877/29,061 (3.0%)	b0.001

Prehospital vital signs

1190/1222 (97.4%)

28,706/29,061 (98.8%)

b0.001

Selection of participants

SBP (mm Hg) 130 (110-151) 133 (112-157) b0.001

DBP (mm Hg) 78 (63-92) 78 (64-90) 0.576

A total of 236,698 patients were enrolled in the JTDB between 2004

MAP (mm Hg) 94.0

(79.3-110.6)

96.7 (81.7-112.0) 0.052

HR (beats/min) 86 (72-102) 84 (72-99) 0.002

and 2015. We identified 163,195 patients with blunt or penetrating trauma who were transported by the emergency medical services di- rectly from the injury site. Of these, patients with ISS >= 16 were selected for this study because of the requirement for specialized trauma care	Shock index (HR/SBP) RR (breaths/min) JCS Grade 0	0.67 (0.53-0.86) 24 (20-28) 270/1199 (22.5%)	0.63 (0.51-0.79) 22 (18-24) 8459/28,973 (29.2%)	b0.001 b0.001 b0.001
[10]. Patients with cardiopulmonary arrest before/at hospital arrival or	Grade 1	413/1199	11,178/28,973	0.004
with an AIS score of 6 were excluded from this study because their con-		(34.4%)	(38.6%)
ditions were considered non-survivable. Cases with incomplete data re-	Grade 2	183/1199	3570/28,973 (12.3%)	0.003
garding survival, pre-hospital and in-hospital vital signs [systolic blood		(15.3%)

pressure (SBP), diastolic blood pressure , respiratory rate (RR), and heart rate (HR)] were also excluded. Finally, we excluded patients who were transported by helicopter to reduce the influence of vehicle type of emergency services. Thus, a total of 30,283 patients with trauma met our study criteria (Fig. 1).

Grade 3	333/1199 (27.8%)	5766/28,973 (19.9%)	b0.001
AIS (>=3)
Head	698/1222	16,951/29,061	0.407
	(57.1%)	(58.3%)
Face	34/1222 (2.8%)	351/29,061 (1.2%)	b0.001
Neck	8/1222 (0.7%)	91/29,061 (0.3%)	0.064

	Thorax	647/1222	11,597/29,061	b0.001
		(52.9%)	(39.9%)
2.4. Outcome measures	Abdomen and pelvis	157/1222	2402/29,061 (8.3%)	b0.001
		(12.8%)

The primary outcome of this study was survival to hospital dis-

Spine 179/1222

(14.6%)

4918/29,061 (16.9%) 0.039

charge. The secondary outcomes were survival to ED discharge and

Upper extremity 70/1222 (5.7%) 1157/29,061 (4.0%) 0.004

time from arrival of EMS to the following events: hospital arrival, CT scan, blood transfusion, and surgery.

Lower extremity 328/1222

5089/29,061 (17.5%) b0.001

2.5. Statistical analysis

To display the patient data, the mean +- standard deviation (SD) or median with interquartile range (IQR) were used for numerical vari- ables, according to data distribution and previous reporting style. Cate- gorical variables were displayed as raw values followed by percentages in parentheses. The t-test was used to compare the means of the two samples. The Mann-Whitney U test was used to compare the medians of the two samples. The chi-square test was used to compare frequen- cies. Time from EMS site arrival to hospital arrival, CT scan, blood trans- fusion, and surgery were compared by drawing the cumulative event rate curve and log rank tests. For this analysis, we extracted only data for events achieved so that the cumulative event rate eventually reached 100% in both groups. To assess the independent effect of the presence of a physician during pre-hospital trauma care on study end- points, multivariable logistic regression for survival and cox Proportional-hazards regression analysis for time to events were per- formed. Following previous reports [11,12], covariates were carefully selected based on the assumption that none were affected directly by the intervention. These variables included age, sex, cause of injury, ISS, and pre-hospital vital signs (SBP, DBP, RR, HR, and level of conscious- ness measured by the Japan Coma Scale ). The two-sided signifi- cance level for all tests was set at 5% (p b 0.05). All analyses were performed using EZR software, version 3.3.2 (Easy R, Saitama Medical Center, Jichi Medical University; Saitama, Japan) [13].

All categorical variables are shown as n (%). Continuous variables are shown as mean +- standard deviation or median (interquartile range). Missing data are age (n = 36), gender (n = 10), JCS (n = 111), RTS (n = 1051), TRISS (n = 439).

	(26.8%)
ISS	25 (19-34)	21 (17-26)	b0.001
RTS	7.55 (5.97-7.84)	7.84 (6.90-7.84)	b0.001
TRISS (probability of survival)	0.91 (0.71-0.97)	0.93 (0.84-0.97)	b0.001

SBP: systolic blood pressure, DBP: diastolic blood pressure, MAP: mean arterial pressure, HR: heart rate, RR: respiratory rate, JCS: Japan coma scale (Grade 0: alert, Grade 1: possible eye-opening, not lucid, Grade 2: possible eye-opening upon stimulation, Grade 3: no eye- opening and coma), AIS: Abbreviated Injury Score, ISS: injury severity score, RTS: revised trauma score, TRISS: trauma and injury severity score.

Results

Characteristics of study subjects

Among a total of 30,283 patients with trauma who met our study criteria, 1222 patients were transported to hospital accompanied by a physician (physician group) and 29,061 patients were accompanied by paramedics (paramedics group). The main characteristics of the pa- tients are shown in Table 1. The mean age was 52.6 +- 22.1 years in the physician group and 54.9 +- 22.7 years in the paramedics group (p b 0.001). Blunt trauma was the most frequently occurring injury type in both groups (97.4% vs 98.8%, p b 0.001). Although there were statisti- cally significant differences in prehospital vital signs such as SBP, RR, and HR between two groups, almost all of these differences are not clin- ically significant. Severe disturbance of consciousness (JCS Grade 3) was observed more frequently in the physician group than in the

Table 2

Comparison of prehospital intervention.

there was a significant difference in the frequency of prehospital intra- venous fluid administrated (31.7% vs 0.6%, p b 0.001).

	Physician (n = 1222)	Paramedics (n = 29,061)	p value	The physician and paramedics groups were compared for time from EMS site arrival to hospital arrival, CT scan, blood transfusion, and sur-
Prehospital intervention				gery by determining the cumulative event rate curve and log rank
Oxygenation	906/1222 (74.1%)	9174/29,061 (68.4%)	b0.001	tests (Fig. 2A-D). Compared with the paramedics group, time to hospi-
cervical collars	804/1222 (65.8%)	18,620/29,061 (64.1%)	0.23	tal arrival and time to CT scan were significantly longer in the physician
Backboards Ventilation Intubation	853/1222 (69.8%) 35/1222 (2.9%) 87/1222 (7.1%)	17,714/29,061 (61.0%) 259/29,061 (0.9%) 823/29,061 (2.8%)	b0.001 b0.001 b0.001	group (hospital arrival: 37 [25-52] vs 28 [22-37], p b 0.001. CT scan: 81 [60-102] vs 66 [52-89], p b 0.001). Conversely, time to blood transfu-
Nasal airway tube	7/1222 (0.6%)	70/29,061 (0.2%)	0.05	sion was significantly shorter in the physician group than in the para-
Intravenous fluid	387/1222 (31.7%)	171/29,061 (0.6%)	b0.001	medics group (137 [95-200] vs 172 [111-265], p b 0.001). There was

All categorical variables are shown as n (%).

paramedics group (27.8% vs 19.9%, p b 0.001). Patients in the physician group had more severe trauma scores with higher ISS (median [IQR]: 25 [19-34] vs 21 [17-26], p b 0.001) and lower RTS (7.55 [5.97-7.84] vs

7.84 [6.90-7.84], p b 0.001) than those in the paramedics group. The probability of survival was also lower in the physician group compared with that in the paramedics group (0.91 [0.71-0.97] vs 0.93 [0.84-0.97], p b 0.001).

Main results

Table 2 presents the comparison of pre-hospital treatment. There were significant differences in prehospital treatment including oxygen- ation, backboard stabilization, respiratory ventilation, and airway intu- bation between the physician and paramedics groups; specifically,

a difference in time to surgery between the two groups, although it

was not statistically significant (174 [119-276] vs 195 [130-294], p =

0.056).

Table 3 compares Survival outcomes (survival to ED discharge and survival to hospital discharge) between the two cohorts. Fewer patients in the physician group survived to ED or to hospital discharge than in the paramedics group, although absolute difference was quite low (sur- vival to ED discharge: 97.4% vs 98.8%, p b 0.001; survival to discharge: 85.9% vs 87.9%, p = 0.04).

The results of multivariable logistic regression for survival outcomes are shown in Table 4. The presence of a physician did not improve sur- vival to hospital discharge (odds ratio (OR) = 1.16. confidence interval (CI) = 0.97 to 1.40, p = 0.109) or survival to ED discharge (OR: 0.69. CI

= 0.47 to 1.01, p = 0.058). Cox proportional-hazard regression analysis revealed that time to hospital arrival and time to CT scan, adjusting for age, sex, cause of injury, ISS, and pre-hospital vital signs (SBP, DBP, RR, HR, and JCS), were longer in the physician group (hospital arrival: haz- ard ratio (HR) = 0.59. CI = 0.55 to 0.62, p b 0.001. CT scan: HR = 0.74. CI

Fig. 2. Comparison of cumulative event rate curve of (A) hospital arrival, (B) CT scan, (C) blood transfusion, and (D) surgery. (A) 883 of data were missing of required information. 318 of data were detected as outlier and removed from analysis. (B) 362 of data were detected as outlier and removed from analysis. (C) 453 data were detected as outlier and removed from analysis. (D) 2400 data were detected as outlier and removed from analysis. EMS: emergency medical service.

Table 3 Comparison of disposition at ED and hospital discharge.
	Physician (n = 1222)	Paramedics (n = 29,061)	p value
Disposition at ED
Survived	1190/1222 (97.4%)	28,679/29,018 (98.8%)	b0.001
(ICU admission)	1083/1222 (88.6%)	23,059/29,018 (79.5%)	b0.001
(Ward admission)	107/1222 (8.8%)	5620/29,018 (19.4%)	b0.001
Died	32/1222 (2.6%)	339/29,018 (1.2%)	b0.001
Disposition at hospital discharge Survived	1050/1222 (85.9%)	25,553/29,061 (87.9%)	0.04
(Home)	384/1222 (31.4%)	12,210/29,061 (42.0%)	b0.001
(Transferred)	666/1222 (54.5%)	13,343/29,061 (45.9%)	b0.001
Died	172/1222 (14.1%)	3508/29,061 (12.1%)	0.04

All categorical variables are shown as n (%). 43 of data for disposition at ED is missing. ED: emergency department, ICU: intensive care unit, CI: confidence interval.

= 0.70 to 0.79, p b 0.001). Conversely, time to blood transfusion was shorter in the physician group than in the paramedics group (HR = 1.26. CI = 1.10 to 1.45, p = 0.001). There was no difference in the time to surgery between the two groups (HR = 1.06. CI = 0.95 to 1.19, p = 0.29).

Discussion

To the best of our knowledge, the current study is one of the largest cohort studies investigating the effectiveness of advanced trauma care delivered by the emergency physician in the prehospital setting. In sum- mary, trauma patients transported with an emergency physician were injured more severely than those accompanied with only paramedics in Japan. However, we were not able to demonstrate an improvement in survival in physician-staffed ambulance when adjusting for severity of clinical status.

The improvement on outcome by a physician-staffed ambulance has still been controversial. Yequiayan JM et al. demonstrated that prehospital management by an emergency physician was associated with a significant reduction in 30-day mortality in their multicenter co- hort study on patients with severe blunt trauma requiring admission to university hospital intensive care units within the first 72 h [14]. How- ever, this study might have major institutional bias because they inves- tigated only patients transferred to the university hospital (equivalent to level I trauma center in the U.S.) and 62% of patients treated by non-physician were first admitted to general hospitals. On the other hand, the results of a recent matched pair analysis of the trauma registry of the German Trauma Society showed no difference in mortality during hospitalization and within the first 24 h between prehospital physician and non-physician management [12]. Similarly, the present study did not find a Survival benefit with the involvement of a physician in the prehospital trauma care despite the fact that we included only patients

Table 4

Primary and secondary outcome comparisons.

Logistic regression analysis	Survival OR	95% CI	p value
Survival to hospital discharge	1.160	0.967 to 1.400	0.109
Survival to ED	0.689	0.468 to 1.010	0.058

Cox proportional hazard regression analysis HR 95% CI p value Time to events from the site arrival of EMS

Hospital arrival 0.588 0.554 to 0.623 b0.001

CT scan 0.741 0.695 to 0.791 b0.001

Blood transfusion 1.261 1.097 to 1.449 0.001

Surgery 1.062 0.950 to 1.188 0.290

Multivariable logistic regression analysis for survival outcomes and Cox proportional-haz- ard regression analysis for other secondary outcomes were performed with adjustment for age, gender, Injury Severity Score, cause of injury, and pre-hospital vital signs (systolic blood pressure, diastolic blood pressure, respiratory rate, heart rate, and Japan Coma Scale).

OR: odds ratio, CI: confidence interval, ED: emergency department, HR: hazard ratio, EMS: emergency medical service.

with severe trauma (ISS >= 16). Actually, a considerable number of less severe patients receive prehospital care by a hospital physician in the real world. Therefore, we could reduce financial costs and human re- sources because we showed no major clinical benefit with physician in- volvement in prehospital trauma care.

The current study demonstrated that compared to paramedic manage- ment, prehospital advanced interventions, particularly delivery of intrave- nous fluid and airway management, were more frequently performed by a physician in Japanese EMS. However, the issue is that prompt prehospital treatment does not guarantee clinical benefit. For example, recent studies indicate that prehospital administration of fluid to patients with trauma could be deleterious because of impaired clotting [15,16]. Most importantly, advanced prehospital intervention is time consuming, potentially prolonging on-scene time or transportation time, which considerably af- fects outcome [5-7]. The current study found that prehospital time was sig- nificantly prolonged in the physician group compared to the non-physician group. This finding is controversial, possibly due to differences in study de- sign, emergency medical systems, number of hospitals, and the country size between previous studies [4,12,17], nevertheless delayed hospital arrival was one of the reasons for the failure of the presence of a prehospital phy- sician to improve outcome.

We evaluated not only prehospital time but also time to CT scan for di- agnosis and time to blood transfusion and surgery for treatment. Diagnosis by whole-body CT scan in early trauma care significantly increases the probability of survival for patients with polytrauma [18,19]. Our study dem- onstrated that time to CT scan was significantly longer among patients treated by a prehospital physician than that among patients treated by paramedics. In contrast, blood transfusion was achieved significantly more rapidly in patients treated by a physician. Prehospital blood transfu- sion is unusual in Japan. Therefore, this difference was attributed to the prompt order for blood transfusion in the hospital. Accurate prehospital in- formation about trauma patients provided by a physician on scene may re- sult in rapid preparation and infusion of blood. Although variations in hospital capability could remain, early blood transfusion could be beneficial aspect by a physician-staffed ambulance. Finally, time to surgery was not significantly different between the physician and paramedics group, indi- cating that unaltered speed of definitive control of bleeding by operation could be one of the reasons why prehospital early advanced intervention by a physician could not show the improvement of the survival rate for trauma care.

Our study has several limitations. First, there may have been institu- tional bias. We did not control for the capability of each hospital for trauma care and this affects outcome. However, over 90% of the hospi- tals participating in the JTDB are tertiary emergency medical centers ap- proved by the Japanese Ministry of Health, Labor and Welfare as being competent in delivering trauma care. Second, there is minimal data on exactly which procedures were performed by physicians or paramedics on scene. Our data did not contain information on some prehospital ad- vanced procedures by a physician such as ultrasound, REBOA, and chest drainage, which might independently affect the outcome. Third, there were cases of missing data which could affect the results. Multiple

imputation (MI) was considered as a method to handle missing data [20]. However, MI may also generate bias. We chose to exclude patients with missing data because of a large study population. Fourth, there was a striking difference in prehospital intravenous fluid administrated be- tween two cohorts. We assume that this is mainly due to the Japanese EMS system which did not allow paramedics to perform fluid adminis- tration in prehospital trauma care till it was permitted in 2014. As there are some evidence that prehospital administration of fluid to patients with trauma could be deleterious [15,16], huge difference of this single prehospital intervention could have influence on outcome. Fifth, the current study includes only participants without prehospital cardiopul- monary arrest transferred by ambulance so that the results do not fit in all trauma cases. In fact, Abe et al. demonstrated that transport by heli- copter with a physician could be associated with improved survival at discharge compared to ground services using the same JTDB database [11]. In addition, efficacy of advanced life support by a prehospital phy- sician for trauma patients with cardiopulmonary arrest compared to basic life support by paramedics could be beneficial and still debated [21,22]. Finally, we did not evaluate quality of life of surviving trauma patients because we could not obtain detailed information regarding patient status at/after discharge. However, the quality of post-trauma life is an important clinical goal in the assessment of major trauma care.

Conclusion

In conclusion, this nationwide retrospective cohort study failed to show a difference in mortality between non-physician staffed ambu- lances and physician-staffed ambulances. The presence of a physician on scene might lead to rapid resuscitation but delay definitive care.

Abbreviations

EMS emergency medical services

REBOA resuscitative endovascular balloon occlusion of the aorta JTDB Japan trauma data bank

JPTEC Japan prehospital trauma evaluation and care AIS abbreviated injury scale

ISS injury severity score

ED emergency department

RTS revised trauma score

TRISS trauma and injury severity score SBP systolic blood pressure

DBP diastolic blood pressure RR respiratory rate

HR heart rate

SD standard deviation

IQR interquartile range

JCS Japan coma scale

OR odds ratio

CI confidence interval

HR hazard ratio

Funding

This research was supported by JSPS KAKENHI Grant Number 17K11593.

Conflict of interest

There are no conflicts of interest for all authors in this study.

Acknowledgement

We thank all investigators who coordinated or participated in the JTDB. We would like to thank Editage (www.editage.jp) for English lan- guage editing.

Author contributions

All of the authors jointly conceived of and designated this study. YH and TA conducted data cleaning and analyzed the data. All of the au- thors interpreted the data. YH drafted the manuscript. All of the authors reviewed and discussed the manuscript. All authors read and approved the final manuscript.

References

1. Cales RH, Trunkey DD. Preventable Trauma deaths. A review of trauma care systems development. JAMA 1985;254:1059-63.
2. Bushby N, Fitzgerald M, Cameron P, Marasco S, Bystrzycki A, Rosenfeld JV, et al. prehospital intubation and chest decompression is associated with unexpected sur- vival in major thoracic blunt trauma. Emerg Med Australas 2005;17:443-9.
3. Meizoso JP, Valle EJ, Allen CJ, Ray JJ, Jouria JM, Teisch LF, et al. Decreased mortality after prehospital interventions in severely injured trauma patients. J Trauma Acute Care Surg 2015;79:227-31.
4. Spaite DW, Tse DJ, Valenzuela TD, Criss EA, Meislin HW, Mahoney M, et al. The im- pact of injury severity and prehospital procedures on scene time in victims of major trauma. Ann Emerg Med 1991;20:1299-305.
5. Zafar SN, Haider AH, Stevens KA, Ray-Mazumder N, Kisat MT, Schneider EB, et al. In- creased mortality associated with EMS transport of gunshot wound victims when compared to private vehicle transport. Injury 2014;45:1320-6.
6. Funder KS, Petersen JA, Steinmetz J. On-scene time and outcome after penetrating trauma: an observational study. Emerg Med J 2011;28:797-801.
7. McMonagle MP, Flabouris A, Parr MJ, Sugrue M. Reducing time to Urgent surgery by transporting resources to the trauma patient. ANZ J Surg 2007;77:241-6.
8. Suzuki T, Nishida M, Suzuki Y, Kobayashi K, Mahadevan SV. Issues and solutions in introducing Western systems to the pre-hospital care system in Japan. West J Emerg Med 2008;9:166-70.
9. Abe T, Uchida M, Nagata I, Saitoh D, Tamiya N. Resuscitative endovascular balloon occlusion of the aorta versus aortic cross clamping among patients with critical trauma: a nationwide cohort study in Japan. Crit Care 2016;20:400.
10. Galvagno Jr SM, Haut ER, Zafar SN, Millin MG, Efron DT, Koenig Jr GJ, et al. Associa- tion between helicopter vs ground emergency medical services and survival for adults with major trauma. JAMA 2012;307:1602-10.
11. Abe T, Takahashi O, Saitoh D, Tokuda Y. Association between helicopter with physi- cian versus ground emergency medical services and survival of adults with major trauma in Japan. Crit Care 2014;18:R146.
12. Bieler D, Franke A, Lefering R, Hentsch S, Willms A, Kulla M, et al. Does the presence of an emergency physician influence pre-hospital time, pre-hospital interventions and the mortality of severely injured patients? A matched-pair analysis based on the trauma registry of the German Trauma Society (TraumaRegister DGU(R)). Injury 2017;48:32-40.
13. Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 2013;48:452-8.
14. Yeguiayan JM, Garrigue D, Binquet C, Jacquot C, Duranteau J, Martin C, et al, French Intensive Care Recorded In Severe Trauma Study Group. Medical pre-hospital man- agement reduces mortality in severe blunt trauma: a prospective epidemiological study. Crit Care 2011;15(1):R34.
15. Bogert JN, Harvin JA, Cotton BA. Damage control resuscitation. J Intensive Care Med 2016;31:177-86.
16. Hussmann B, Heuer M, Lefering R, Touma A, Schoeneberg C, Keitel J, et al. Prehospital volume therapy as an independent risk factor after trauma. Biomed Res Int 2015; 2015:354367.
17. Kulla M, Helm M, Lefering R, Walcher F. prehospital endotracheal intubation and chest tubing does not prolong the overall resuscitation time of severely injured pa- tients: a retrospective, multicentre study of the Trauma Registry of the German So- ciety of Trauma Surgery. Emerg Med J 2012;29:497-501.
18. Huber-Wagner S, Lefering R, Qvick LM, Korner M, Kay MV, Pfeifer KJ, et al. Effect of whole-body CT during trauma resuscitation on survival: a retrospective, multicentre study. Lancet 2009;373:1455-61.
19. Huber-Wagner S, Biberthaler P, Haberle S, Wierer M, Dobritz M, Rummeny E, et al. Whole-body CT in haemodynamically unstable severely injured patients–a retro- spective, multicentre study. PLoS One 2013;8:e68880.
20. Moore L, Hanley JA, Turgeon AF, Lavoie A, Emond M. A multiple imputation model for imputing missing physiologic data in the National Trauma Data Bank. J Am Coll Surg 2009;209:572-9.
21. Fukuda T, Ohashi-Fukuda N, Kondo Y, Hayashida K, Kukita I. Association of Prehospital advanced life support by physician with survival after out-of-hospital cardiac arrest with blunt trauma following traffic collisions: Japanese registry- based study. JAMA Surg 2018;153(6):e180674.
22. Kondo Y, Fukuda T, Uchimido R, Hifumi T, Hayashida K. Effects of advanced life sup- port versus basic life support on the mortality rates of patients with trauma in prehospital settings: a study protocol for a systematic review and meta-analysis. BMJ Open 2017;7:e016912.