Introduction
Severe trauma is a leading cause of death worldwide [
1]. It requires a national, systematic approach due to its significant social and economic impact, particularly among younger patients, resulting in substantial national losses and social disruption [
2]. In the Republic of Korea, severe trauma is among the 3 major causes of death, along with cancer and cardiovascular disease. As of 2010, the preventable trauma death rate was approximately 30%, notably higher than in high-income countries, which led to the creation of a national trauma treatment system [
3–
5].
In response, the Republic of Korea launched the nationwide regional trauma center (RTC) initiative in 2012 to improve the identification and prioritization of patients with severe trauma. RTCs have been operational since 2014. According to Kwon et al. [
6], the Republic of Korea’s preventable trauma death rate declined to 15.7% by 2019, with RTCs playing a key role in improving survival rates. However, most recent studies on severe trauma include patients who were transferred between facilities. While these studies highlight overall survival improvements, they offer limited insight into outcomes for patients assessed and transported directly from the prehospital setting.
Given its impact on patient prognosis, the American College of Surgeons recommends that patients with severe trauma be classified and triaged at the prehospital stage, with transport to an appropriate facility, such as a level 1 trauma center [
7]. Hospital location, regional characteristics, trauma center classification, and injury mechanisms all play important roles in trauma mortality research. The concordance rate between prehospital and hospital classifications for severe trauma is typically defined using an injury severity score (ISS) threshold of 15, though international differences complicate direct comparisons [
8]. Furthermore, few studies have examined outcomes for patients with severe trauma who are directly transported from the prehospital stage in the Republic of Korea.
Therefore, this study analyzed the impact of direct prehospital transport on outcomes and mortality in patients with ISS >15 using national data.
Materials and Methods
Study Design and Setting
This community-based observational study utilized data from a nationwide trauma registry in the Republic of Korea. The country has a population of 51.7 million and covers an area of 100,210 km2, comprising 17 administrative provinces. The emergency medical service (EMS), operated by the Korea National Fire Agency, provides care throughout all regions, managing a range of conditions from minor injuries to severe trauma and cardiac arrest.
In the Republic of Korea, EMS personnel are authorized to perform advanced airway management and fluid resuscitation under physician supervision. However, all patients must be transported to a hospital, as EMS providers do not have the authority to terminate resuscitation efforts. EMS personnel complete ambulance run sheets for every transport. The Ministry of Health and Welfare has developed a network of RTCs, establishing 17 centers, with 9 officially operational by 2019. According to EMS transport protocols, patients identified as having severe trauma are recommended for transfer to the nearest regional trauma or emergency medical center [
9].
Data Source
Community-based Severe Trauma Surveillance was first implemented in 2 regions in 2013 and expanded nationwide in 2016 by the Korea Disease Control and Prevention Agency. This study analyzed data from surveys conducted between 2016 and 2020. The surveillance included all cases classified as severe trauma by the national EMS, focusing on hospital transfers and treatment outcomes. A total of 700 medical institutions participated, and patient severity was objectively classified using the ISS system, with national expert investigators overseeing the classification process.
Patients with Severe Trauma in the Prehospital Stage Based on the Registry
In the Republic of Korea, since 2013, detailed records of EMS treatment for patients with severe trauma have been maintained to improve the quality of EMS activities. Patients included in the “severe trauma detailed treatment” registry are those classified as having severe trauma. This study analyzed all patients recorded in this registry. EMS categorizes severe trauma according to 4 criteria: (1) physiological findings—systolic blood pressure <90 mmHg, respiratory rate <10 or >29, or altered mental status based on verbal response; (2) physical examination findings—penetrating or stab wounds, chest wall deformity, suspected long bone fractures (>2), crush injuries, pulseless wounds, amputations, suspected pelvic fractures, open skull fractures, traumatic paralysis, and similar conditions; (3) mechanism of injury—including falls, high-energy vector traffic accidents, passenger vehicle accidents, motorcycle accidents, bicycle accidents, and others; and (4) additional EMS-based considerations.
Exclusion Criteria
Patients were excluded if there was a collection error in time data (such as invalid timestamps like “99:99” or a 14-month date), errors in emergency room arrival time (for example, if the emergency department [ED] arrival time was recorded as earlier than the EMS dispatch request time), cases without traumatic mechanisms or those exposed to non-trauma-related events (including thermal injuries, respiratory threats, exposure to chemicals or other substances, physical overexertion, or unspecified categories), ISS score <16, or unknown ISS. Patients with prehospital cardiac arrest due to severe trauma were also excluded, as these cases are associated with a high mortality rate and different mortality patterns [
10].
Data Collection and Variables
Collected data included general characteristics (age, sex, region, and hospital transfer level), accident details (time of accident, rescue request, EMS arrival, and ED arrival), intentionality, toxic exposure, and trauma mechanisms (traffic accidents, falls/slips, violence, blunt trauma, penetrating trauma, mechanical injury, thermal injury, and chemical injury). Trauma locations were categorized as home, medical institutions, schools, pathways, industrial sites, farms, public spaces, commercial facilities, and outdoor areas. ISS scores and patient disposition outcomes (mortality) were also recorded. Time-related variables included EMS activation time, EMS scene arrival time, and ED arrival time. “Time to ED arrival” was defined as the interval from EMS activation to ED arrival. To accurately assess prehospital transport time, patients transferred from other healthcare facilities were excluded during initial data curation. The study period was divided into 2016–2017 and 2018–2020 to reflect major policy changes implemented in 2018, including revisions to the EMSs Act and the Ministry of Health and Welfare’s initiative to strengthen severe trauma care.
Institution types were categorized according to the National Medical Center Statistical Yearbook: (1) level 1, RTCs; level 2, regional emergency centers; level 3, local emergency centers; and level 4, local emergency facilities.
“High-level medical institutions” were defined as level 1 and level 2 facilities, while “lower-level medical institutions” referred to level 3 and level 4 facilities. High-level institutions serve as definitive treatment centers within a regional system, equipped with specialized personnel, facilities, and resources required for advanced care in severe emergency cases. By contrast, lower-level institutions function as frontline or intermediary centers within local communities, primarily managing mild to moderate emergency cases and ensuring the timely transfer of critically ill patients to higher-level centers.
Regional categorization included metropolitan areas such as Seoul, Gyeonggi Province, Incheon, and other regions [
11].
Outcomes
The primary outcome was to assess changes in mortality among patients with severe trauma (ISS >15) who were directly transported from the prehospital stage in the Republic of Korea. The secondary outcome was to analyze factors influencing mortality. Mortality was determined at hospital discharge through review of medical records.
Statistical Analysis
To examine the association between patient characteristics and mortality, a cross-analysis was performed. Additionally, binary logistic regression analysis was conducted to assess the effect of specific variables on mortality and determine the relative importance of each factor. A stepwise selection method was used to identify the key variables in the model. Decision tree analysis was used to visualize decision-making rules based on the independent variables most strongly associated with mortality in patients with severe trauma. This technique, widely employed in data mining, enables the integration of expert knowledge into domain-specific analyses [
12]. It was applied for group classification and outcome prediction.
All analyses were conducted using SAS ver. 9.4 (SAS Institute Inc.). for data cleaning and cross-analysis, and SAS Enterprise Miner 9.4 for binary logistic regression and interactive decision tree analysis.
Ethics Statement
This study protocol was reviewed and approved by National Approval Statistics (approval number: 117105), and the requirement for informed consent was waived.
Results
Patient Flow
Figure 1 illustrates the patient selection process from the community-based severe trauma surveillance dataset in the Republic of Korea between 2016 and 2020. Initially, 237,616 patients were identified. After excluding 2,321 patients due to collection errors or missing time data—including 2,311 cases with implausible time entries (e.g., “99 hours 99 minutes” or “14th month”) and 10 cases where emergency room arrival times were recorded as earlier than EMS dispatch requests—an additional 71,490 patients without traumatic mechanisms or those exposed to non-trauma-related events (such as chemical exposure, physical exertion, or unspecified causes) were excluded. This resulted in 163,805 patients with suspected severe trauma at the prehospital stage. Subsequently, 139,238 patients were excluded based on the following criteria: 2,902 patients with unknown ISS, 118,739 patients with ISS less than 16, and 17,597 patients who had out-of-hospital cardiac arrest or prehospital death. Ultimately, 24,567 patients met the inclusion criteria. The annual distribution was as follows: 4,454 patients in 2016, 4,558 in 2017, 5,097 in 2018, 5,493 in 2019, and 4,965 in 2020.
General Characteristics
Baseline characteristics for the survival and mortality groups are shown in
Table 1. Among female patients, 4,449 (71.7%) survived, while 4,509 male patients (24.6%) died. Among patients aged ≥55 years, 4,285 (31.6%) died, and in higher-level medical institutions (levels 1 and 2), 3,352 patients (23.6%) died. During 2018–2020, 11,884 patients (76.4%) survived, and among those with an ISS of 25–75, 3,768 (39.5%) died (
Table 1,
Table S1).
Comparison of Survival-Mortality by ISS Score Group, Hospital Type, and Year
Table 2 presents survival and mortality rates among patients with severe trauma, categorized by hospital type and ISS, across 2 time periods: 2016–2017 and 2018–2020. In level 1 and 2 hospitals, survival rates were 73.7% (3,108 patients) in 2016–2017 and 77.5% (7,727 patients) in 2018–2020. In level 3 and 4 hospitals, survival rates were 69.0% (3,313 patients) in 2016–2017 and 74.5% (4,157 patients) in 2018–2020. During 2016–2017, 2,107 (46.7%) patients visited level 1 and 2 hospitals annually, while 2,399 (53.3%) visited level 3 and 4 hospitals. In 2018–2020, 3,324 (64.1%) patients visited level 1 and 2 hospitals annually, compared to 1,860 (35.9%) who visited level 3 and 4 hospitals.
By ISS grouping, among patients with an ISS of 25–75, the mortality rate was 43.5% (1,550 patients) in 2016–2017 and 37.1% (2,218 patients) in 2018–2020 (
Table 2,
Table S2).
Factors Associated with Mortality in Patients with Severe Trauma
Compared to female patients (reference group), male patients had a slightly lower risk of mortality (Exp(B)=0.921; 95% confidence interval [CI], 0.857–0.989; p=0.02). Age was a strong predictor of survival: patients aged 15–54 years (Exp(B)=0.440; 95% CI, 0.411–0.471; p<0.001) and those aged <15 years (Exp(B)=0.497; 95% CI, 0.385–0.642; p<0.001) had significantly lower mortality risk compared to those aged ≥55 years. Patients residing in metropolitan areas (Seoul, Gyeonggi Province, Incheon) had a reduced risk of mortality compared to those in other regions (Exp(B)=0.920; 95% CI, 0.862–0.982; p=0.01). Treatment at higher-level institutions (level 1 or 2) was associated with a significantly lower risk of death than treatment at lower-level facilities (level 3 or 4) (Exp(B)=0.810; 95% CI, 0.760–0.864; p<0.001). Patients treated during 2016–2017 had a higher risk of mortality than those treated during 2018–2020 (reference group) (Exp(B)=1.249; 95% CI, 1.171–1.333; p<0.001), indicating improved outcomes in the later period. ISS scores were strongly associated with mortality. Patients with an ISS of 16–24 had a substantially lower risk of death compared to those with scores of 25–75 (Exp(B)=0.295; 95% CI, 0.277–0.314; p<0.001).
Regarding injury mechanism, falls/slips (Exp(B)=0.906; 95% CI, 0.845–0.971;
p=0.005), other blunt injuries (Exp(B)=0.774; 95% CI, 0.636–0.943;
p=0.01), and contact with machinery (Exp(B)=0.662; 95% CI, 0.466–0.941;
p=0.02) were associated with lower mortality compared to transport-related accidents (reference group). Penetrating injuries did not show a significant difference (
p=0.27). Patients with intentional injuries had a significantly higher risk of mortality compared to those with accidental injuries (Exp(B)=1.380; 95% CI, 1.150–1.657;
p<0.001), while violent/assault-related injuries did not show a significant association (
p=0.36). Finally, patients who arrived at the ED ≥30 minutes after EMS activation had a significantly lower risk of mortality compared to those who arrived within 30 minutes (Exp(B)=0.671; 95% CI, 0.630–0.716;
p<0.001) (
Table 3,
Table S3).
Survival Outcomes in Patients with Severe Trauma Based on Time to Arrival at ED, ISS Score, and Region
Among patients who arrived at the ED in less than 30 minutes (n=12,425), the survival rate was 70.5%, while the mortality rate was 29.5%. In this group, patients with an ISS of 16–24 had a survival rate of 79.9% and a mortality rate of 20.0%. For those in metropolitan areas, the survival rate was 81.0%, compared to 79.1% in urban areas. Among patients with an ISS of 25–75, the survival rate was 56.4% and the mortality rate was 43.6%. In metropolitan areas, the survival and mortality rates were 58.7% and 41.3%, respectively, compared to 54.7% and 45.3% in urban areas.
Patients with arrival times of ≥30 minutes (
n=12,142) had a survival rate of 78.6% and a mortality rate of 21.4%. For patients with an ISS of 16–24, the survival rate was 86.9% and the mortality rate was 13.2%. Patients with an ISS of 25–75 had a survival rate of 69.9% and a mortality rate of 35.1%. Among patients arriving at ≥30 minutes with an ISS of 16–24, survival rates were 86.9% in metropolitan areas and 86.8% in urban areas. For patients with an ISS of 25–75, survival was higher in metropolitan areas (67.8%) than in urban areas (63.6%) (
Table 4;
Figure 2).
Survival Outcomes in Patients with Severe Trauma Based on Time to Arrival at ED, Year, and Institution Level
Patients who arrived at the ED within 30 minutes had a lower survival rate (70.5%) compared to those with arrival times of ≥30 minutes (78.6%). High-level institutions showed higher survival rates (70.1% in 2016–2017 and 72.8% in 2018–2020) compared to lower-level institutions (66.2% and 72.0%, respectively). In contrast, patients arriving at the ED after ≥30 minutes demonstrated consistently higher survival rates, with improvements from 75.8% in 2016–2017 to 79.9% in 2018–2020. Within this group, high-level institutions had superior survival outcomes (77.3% in 2016–2017 and 80.7% in 2018–2020) compared to lower-level institutions (74.0% and 77.8%, respectively) (
Table 5;
Figure 3).
Discussion
Our study observed improvements in the Republic of Korea’s trauma system, as indicated by a decreased mortality rate among patients with severe trauma (ISS >15). This reduction included cases involving direct prehospital transportation, which were analyzed by institution, year, and ISS score. Although longer transport times may sometimes be required, selecting an appropriate hospital and transporting patients directly to high-level institutions can enhance survival rates.
This nationwide study in the Republic of Korea analyzed patients with severe trauma (ISS >15) who were directly transported following EMS assessment. Between 2016–2017 and 2018–2020, the overall in-hospital mortality rate decreased by 5.2%, with consistent improvements observed across hospital levels and ISS subgroups (
Table 2). Mortality in urban areas was 1.1% lower than in metropolitan areas. However, transport times exceeding 30 minutes were more frequently observed in urban areas (68%), likely due to the relative distance to high-level trauma centers. Subgroup analysis revealed that among patients with higher ISS scores (25–75), survival was better in metropolitan areas, despite longer transport times being less frequent. This suggests that factors related to hospital level—such as available resources and trauma care capacity—may have a greater influence on outcomes than transport time alone. These findings highlight the potential importance of destination hospital capability in prehospital triage decisions and underscore the need for further investigation into regional differences in trauma care accessibility.
Consistent with previous studies [
13,
14], higher ISS scores were associated with increased mortality, while high-level institutions had an 11.6% lower mortality rate than lower-level institutions (Exp(B)=0.884,
p<0.0001). Between 2016–2017 and 2018–2020, survival rates among severe trauma patients improved significantly, coinciding with a substantial increase in annual visits to high-level institutions—from 2,107 cases (46.7%) to 3,324 cases (64.1%). Additionally, a higher proportion of patients transported to high-level institutions had transport times exceeding 30 minutes (8,003 cases, 65.9%) compared to those transported in under 30 minutes (6,174 cases, 49.7%). Consequently, among directly transported patients, those with transport times longer than 30 minutes exhibited higher survival rates. These results suggest that, for patients with severe trauma, direct transport to high-level institutions—despite longer transport times—may be more beneficial than rapid transport to nearby hospitals.
Kim et al. [
15], in a single-center study in the Republic of Korea, demonstrated that direct transport to a high-level trauma center was associated with improved survival rates, even when transport time exceeded 60 minutes. Another study reported that transport times of up to 90 minutes did not significantly impact mortality [
16]. Similarly, a pediatric trauma study in Sweden found no significant association between prolonged transport time and 30-day mortality or functional outcomes [
17]. However, other studies have shown that each 10-minute delay in total transport time increases the risk of poor functional outcomes by 6%, that transport times exceeding 50 minutes are predictive of poor functional outcomes in trauma patients [
18], and that transport times longer than 30 minutes are associated with worse outcomes in patients with penetrating injuries [
19]. While findings are mixed, this ongoing debate is influenced by differences in regional and national healthcare systems, inclusion of transferred patients, and interfacility transfer times. Well-designed studies are needed to establish an optimal transport time cutoff in the Republic of Korea, taking into account the healthcare system and transport modalities, including air transport.
Limitations
This study had several limitations. First, final diagnoses and detailed disease categories were unavailable, restricting subgroup analyses and detailed statistical classifications. However, as these data have not been disclosed, our study provides an overall assessment of patients with severe trauma and ISS scores who were directly transported from prehospital settings. Second, we included only EMS-assessed patients with trauma. Patients with severe trauma who were not transported by EMS may have influenced the distribution of ED visits. In the Republic of Korea, approximately 15% to 20% of patients with severe trauma visit emergency institutions directly without EMS involvement [
6,
20]. Although EMS transport remains predominant, direct-visit patients require different considerations. In another National Emergency Department Information System-based study that analyzed severe trauma using international classification of disease-based injury severity score rather than ISS, 83,646 patients (22.8%) visited ambulatory hospitals, a smaller number than the 217,482 patients (59.2%) transported by public EMS. However, this difference should be considered when interpreting the results, given the differences in underlying data and the definitions of severe trauma patients [
21]. Future studies should include these patients for a more comprehensive analysis.
