Effectiveness of Physician-Led Prehospital Management in Severe Trauma Patients: A Retrospective Analysis of the Japanese Nationwide Trauma Registry

Background: The comparative effectiveness of physician-led over paramedic-led prehospital trauma management has been inconclusive. Regarding this topic, in some previous studies, the impacts of physician-led prehospital management were affected by the advantage of shortened prehospital time by helicopter transportation. This study aimed to evaluate the effect of physician-led prehospital management independent of prehospital time. Methods: This retrospective cohort study analyzed the data of severe trauma patients who were transported directly to a hospital during 2009–2018 using Japan’s nationwide trauma registry. In-hospital mortality was compared between patients who received physician-led prehospital management and those who received paramedic-led management, using 1:4 propensity score-matched analysis. The propensity score was calculated using information on patient demographics, mechanism of injury, and vital signs at the scene of injury, as well as prehospital transport time. Subgroup analysis was performed to identify patients who were most likely to benet from physician-led prehospital management. Results: A total of 30,968 patients (physician-led: 3,032, paramedic-led: 27,936) were eligible for analysis, of whom 2,766 propensity score-matched pairs (i.e., physician-led: 2766, paramedic-led: 11,064) were generated and compared. Physician-led pre-hospital trauma management showed signicant superiority over paramedic-led prehospital trauma management (in-hospital mortality: 395 [14.3%] and 1785 [16.1%], respectively; odds ratio [95% condence interval] = 0.87 [0.77–0.97], p = 0.017). In subgroup analysis, cases characterized by patient age <65 years, Injury Severity Score ≥ 25, Abbreviated Injury Scale in pelvis and lower extremities ≥ 3, and prehospital transport time <60 min likely benetted from physician-led prehospital management. Conclusions: The result of a largescale registry-based cohort study showed that physician-led prehospital trauma management was signicantly associated with survival benet independent of prehospital transport time.


Background
As the scope of medical interventions that paramedics can perform are generally limited compared to that of physicians, physician-led prehospital management is expected to improve survival outcomes in severe trauma patients. However, the effect of physician-led over paramedic-led prehospital management in severe trauma patients has been debatable. Although some randomized controlled trials [1,2] and cohort studies [3][4][5] suggested bene cial effects of physician-led prehospital trauma management, a recent systematic review concluded that evidence of the superiority of physician-led over paramedic-led prehospital management in severe trauma patients is insu cient because of the limited number of studies with high methodological quality. [6] Notably, some studies evaluating this topic included helicopter emergency medical service (HEMS) with physicians, which had two major advantages: prehospital physician-led management itself and the shortened prehospital transport time. Another systematic review [7] reported that, although physician-led prehospital trauma management was associated with lower mortality, the bene t disappeared after excluding helicopter transport as a confounder.
As longer prehospital transport time has been reported to be associated with higher mortality rates, [8] it would be important to evaluate the bene t of physician-led trauma management independent of prehospital transport time for establishing optimal dispatch criteria for prehospital physician teams.
However, studies evaluating the independent effect of physician-led trauma management, in which prehospital transport time was considered, have been scarce.
The aim of the present study was to evaluate the effect of physician-led prehospital trauma management on patient mortality considering the effect of time requiring patient transport. In addition, in subgroup analyses, we examined the characteristics of patients who were most likely to bene t from physician-led prehospital management itself.

Study design and setting
We conducted a nationwide registry-based retrospective cohort study, wherein we analyzed data from the Japan Trauma Data Bank (JTDB) between scal years 2009 and 2018. The JTDB was established in 2003, and it is overseen and maintained by the Japanese Association for the Surgery of Trauma and Japanese Association for Acute Medicine. The details of all trauma patients who suffered a severe injury at any region of the body, with an abbreviated injury scale (AIS) score of ≥ 3, were registered in the JTDB. During the study period, the JTDB received records from approximately 280 hospitals, including almost all government-approved tertiary emergency hospitals in the country. The database includes information on injury mechanisms, prehospital time (including the paramedic dispatch time, physician contact time, and hospital arrival time), patient baseline characteristics (including vital signs at the scene of injury and upon arriving at an emergency department [ED]), procedures performed, and survival status at hospital discharge.
In Japan, the operation of prehospital physician teams, such as dispatch criteria and operating time), varies according to the medical control area. The coverage area also varies largely depending on whether it is an urban or rural area. The physicians are delivered in a car or a helicopter according to the system of the medical control area. They are not always trauma surgeons but are those usually working at an ED and trained to provide basic prehospital trauma management such as assessment with sonography, tracheal intubation, chest drainage, intraosseous infusion, and temporal hemostatic maneuver using a tourniquet. Regarding uid resuscitation, prehospital blood transfusion is not common in Japan, and only the administration of the crystalloid solution is provided in many cases.
This study complied with the principles of the 1964 Helsinki Declaration and its later amendments. The Ethics Committee of Tokyo Medical and Dental University approved this study (#2192). The requirement for informed consent from each patient was waived because of the study's retrospective design and the use of anonymized patient data.

Study population
Patients who met all of the following criteria were included: (1) patients who suffered injuries of Injury Severity Score (ISS) ≥ 16, (2) patients who were transferred directly from the scene of injury, and (3) patients whose speci c information regarding times of injury, physician contact, and hospital arrival were available. We excluded patients from the analysis if they met at least one of the following criteria: (1) cardiac arrest at the scene of injury, (2) unsalvageable injury de ned as AIS = 6, (3) missing data required for analyses, and (4) unrealistic or outlier values on prehospital time course. Outlier values on prehospital time course were determined using robust regression analysis in which time of injury, physician contact time, and hospital arrival time were accounted for.

Data collection
We collected information on the following items from the JTDB: age; sex; mechanism of injury; type of injury (i.e., blunt or penetrating); year, season, and time of injury; time of physician contact; time of hospital arrival; systolic blood pressure, heart rate, and respiratory rate at the scene of injury, consciousness level at the scene of injury (recorded using the Japan Coma Scale [9]); The highest score of AIS values for each region of the body; ISS; and patient survival status at hospital discharge.

De nition and outcome
We identi ed patients with physician-led prehospital management by comparing time of physician contact and time of hospital arrival. Eligible patients were divided into the physician-led prehospital management group (physician-led group) and the paramedic-led prehospital management group (paramedic-led group). Season of injury was divided into four categories by quarter, beginning in January. Time of injury was divided into four zones every 6 hours, beginning at 0:00. The study outcome was inhospital mortality.

Statistical analysis
Considering the heterogeneity of trauma patients between the physician-led group and the paramedic-led group, we used a propensity score matching analysis [10] to compare the outcome between the two groups. In this analysis, a logistic regression model was applied to estimate the propensity score for each patient, predicting physician-led prehospital management based on age, sex, mechanism of injury, type of injury, year of injury, systolic blood pressure and respiratory rate at the scene of injury, consciousness level at the scene of injury, and ISS, in addition to prehospital transport time (from injury onset to hospital arrival). Considering the differences in sta ng according to working hours and in road conditions according to season, both time and season categories of injury were incorporated into the model. These variables were chosen based on the clinical perspective, and the predictive accuracy of these variables for in-hospital mortality was assessed using C-statistics. Propensity score matching extracted 1:4 matched pairs from the physician-led and paramedic-led groups; this ratio was determined based on the feasibility of match balance and maximum use of patient data. Match balance between the groups was assessed by the absolute standardized mean difference (ASMD); values < 0.1 were considered acceptable. The caliper width was set as the standard deviation of the logit-transformed propensity score multiplied by 0.1 to achieve well-matched balance between the two groups. The chi-square test was used for intergroup comparison in the propensity score-matched cohort. The effectiveness of physician-led prehospital management was also evaluated in a sensitivity analysis in an overall study cohort (i.e., not the propensity score-matched cohort) using a multivariate logistic regression model. In this model, the aforementioned variables used in the propensity score calculation were used as the covariates.
Multicollinearity was assessed by the variance in ation factor, with the tolerance value set at < 2.
Subgroup analysis was performed in the propensity score-matched cohort to assess potential candidates who would bene t from physician-led prehospital management. We evaluated the p values for the interaction of the following dichotomized categories for the study outcome: age (< 65 vs. ≥65); sex (male vs. female); type of injury (blunt vs. penetrating); blood pressure at the scene of injury (< 90 mmHg vs. ≥90 mmHg); shock index de ned by the heart rate/systolic blood pressure ratio (< 1 vs. ≥1); presence or absence of coma (de ned by Japan Coma Scale > 30 at the scene of injury); the highest AIS scores on the head, chest, abdomen, and pelvis and lower extremities (< 3 vs. ≥3); ISS (< 25 vs. ≥25); and the time lapse between the time of injury and the time of hospital arrival (< 60 min vs. ≥60 min).
Descriptive statistics were reported as counts and percentages for categorical variables and medians and the 25th − 75th percentiles for numeric or ordered variables. Predictive statistics were reported as odds ratios (ORs) and 95% con dence intervals (CIs). The level of signi cance was de ned as two-sided p < 0.05 for all statistical analyses. All analyses were performed using R 3.

(R Foundation for Statistical
Computing, Vienna, Austria).

Results
A ow diagram of the patient selection process is presented in Fig. 1. A total of 30,968 patients were eligible for analysis based on the inclusion and exclusion criteria. Of them, 3,032 patients (9.8%) received physician-led prehospital management. Physician-led prehospital management was more likely to be provided during daytime (between 6:00 and 17:59) than during nighttime. Median ISS (25th, 75th percentiles) was 25 (18,32) in the physician-led group and 21 (17,27) in the paramedic-led group, suggesting that patients in the physician-led group suffered more severe injuries than those in the paramedic-led group. The time lapse from the time of injury to physician contact was shorter in the physician-led group than in the paramedic-led group ( in the physician-led group and in 3,417 (12.2%) patients in the paramedic-led group. Baseline characteristics of the overall study cohort are summarized in Table 1. The variables used for propensity score estimation had high accuracy for predicting in-hospital mortality with C-statistics of 0.87. Via the matching process, a total of 2,766 propensity score-matched pairs (2,766 and 11,064 patients per physician-led group and paramedic-led group, respectively) were generated. All the ASMD values of the adjusted variables for the severity adjustment were < 0.1, indicating a wellmatched balance ( A signi cantly reduced in-hospital mortality rate was observed in the physician-led group in the propensity score-matched population (OR = 0.87, 95% CI, 0.77-0.97; p = 0.017). In the sensitivity analysis conducted with the overall study population using logistic regression analysis, the variance in ation factors for all the variables were less than 2, which eliminated the issue of multicollinearity in the model. The result also showed the signi cant association between physician-led prehospital treatment and reduced in-hospital mortality (adjusted OR = 0.86, 95% CI, 0.75-0.98; p = 0.021). Abbreviations: ASMD, absolute standardized mean difference; IQR, interquartile range; AIS, abbreviated injury scale; ISS, injury severity score; ED, emergency department.
The results of subgroup analysis are summarized in Fig. 2. Patient who had age < 65 years, severe injuries with ISS ≥ 25, injury with AIS ≥ 3 in the pelvis or lower extremities, and total transportation time < 60 min likely bene ted from physician-led prehospital management.

Discussion
The present study evaluated the independent effect of physician-led prehospital trauma care by including a larger number of patients than that included in previous studies and by considering prehospital time.
The results demonstrated the independent survival bene t of physician-led prehospital trauma care. Furthermore, subgroup analysis revealed the speci c subpopulations that might be likely to bene t from physician-led prehospital management. These ndings could provide additional insights into the topic of prehospital trauma care, because some previous studies showing the superiority of physician-led prehospital trauma management would have been largely bene tted by the advantages of shortened prehospital transport time by HEMS [11,12] in addition to the independent effect of physician-led prehospital care.
Several theoretical advantages of physician-led over paramedic-led prehospital management of severe trauma cases should be considered. First, physicians can provide a broader scope of surgical and medical interventions. Although medical interventions that paramedics are permitted vary depending on the country or area, the scope of their role tends to be limited compared to that of physicians. In Japan, paramedics responding to trauma patients without cardiac arrest are limited to performing spinal motion restriction, external xation of bone fractures, oxygen administration using a mask, and administration of Ringer's solution (only to patients with shock). Second, physicians' interventions were reported to have a higher rate of success than those performed by paramedics; for example, a previous study showed a higher rate of achieving successful advanced airway management, [13][14][15] which prevents secondary brain injury, in physicians than in paramedics. [16] Third, physicians can make precise and exible clinical decisions following the latest trauma management strategy rather than uniform simpli ed management, such as introduction of restrictive uid management based on strategic permissive hypotension. [17] Because the JTDB lacks detailed information on the treatments provided by physicians in prehospital settings, speci c interventions contributing to survival bene t could not be mentioned in the present study. However, the results of the subgroup analysis in the present study suggested the candidates who were more likely to bene t from physician-led prehospital management. Patients younger than 65 years old were likely to receive survival bene t than patients aged ≥ 65 years. A previous study assessing the characteristics of geriatric trauma patients reported a positive linear relationship between age and mortality risk, [18] suggesting that the effects of any treatment might be smaller in older patients.
Furthermore, similar to the previous study showing the effectiveness of physician-led prehospital management especially in severe trauma patients, [19,20] our data suggested that patients with ISS ≥ 25 were likely to bene t from physician-led prehospital management than patients with ISS < 25. Interestingly, patients who suffered severe injury in the pelvis or lower extremities were more likely to bene t from physician-led prehospital management than those without severe injury in these regions; however, such an association was not observed in the head, chest, and abdominal trauma. This could be partially explained by the nature of the procedures performed at the scene of injury or during transportation. Interventions that can be provided in the prehospital setting are generally limited to simple procedures; although physicians often can temporarily improve deteriorated vital signs by some effective procedures, including the use of a tourniquet or resuscitative endovascular balloon occlusion of the aorta, they cannot repair multiple and complicated organ injuries anatomically in the prehospital settings. Thus, treatments in prehospital settings might have been provided as bridging therapies until de nitive care that can be provided after ED arrival. This hypothesis also could explain the result that total prehospital time had signi cant interaction in the effectiveness of physician-led prehospital management. That is, prolonged prehospital time might have reduced the effect of physician-led prehospital treatments because of the nature of the bridging therapy. These ndings from subgroup analyses would serve as a basis for establishing optimal indication for dispatching physician-led teams to the scene of injury.
There were several limitations to this study that should be acknowledged. Because of the retrospective nature of this study, residual confounding was unavoidable. The number of patients with penetrating injury was limited. Physician dispatch criteria did not follow standardized protocols. Furthermore, detailed information on the treatments delivered by a physician was not available in the JTDB. As medical interventions that paramedics can provide vary depending on countries, the results are not always applicable in some countries. Despite these limitations, to the best of our knowledge, this was the rst largescale retrospective cohort study that showed the independent survival bene t of physician-led The Ethics Committee of Tokyo Medical and Dental University approved this study (#2192). The requirement for informed consent from each patient was waived because of the study's retrospective design and the use of anonymized patient data.

Consent for publication
The manuscript does not include individual patient's data. Therefore, consent is not applicable for this study.

Availability of data and materials
An overview of the Japan Trauma Data Bank (JTDB) is available at http://www.jtcrjatec.org/traumabank/index.htm. The detailed data in the JTDB that support the ndings of this study are available from Japan Trauma Care and Research but restrictions apply to the availability of these data, which were used under license for the current study, and hence are not publicly available.

Competing interests
The authors declare that they have no competing interests.

Funding
Not applicable Authors' contributions AE drafted and revised the manuscript, prepared the study concept and design, and performed statistical analysis and data interpretation as well as accepts responsibility for conduct of research, nal approval, and study supervision; MK revised the manuscript, prepared the study concept and design, and performed data interpretation as well as accepts responsibility for conduct of research and nal approval; AS performed data acquisition and revised the manuscript as well as accepts responsibility for conduct of research and nal approval; YO revised the manuscript and performed data interpretation as well as accepts responsibility for conduct of research and nal approval. Figure 1 Flow diagram of the process of patient selection Subgroup analysis for the effect of physician-led prehospital trauma management on in-hospital mortality