The Ideal Trauma Transport Modality: Helicopter or Ground Ambulance? A Systematic Review and Meta-analysis

Abstract

Introduction:

Trauma-related mortality remains a global concern, causing ~ 5 million deaths annually. Helicopter emergency medical services (HEMS) enable rapid transport to difficult-to-reach areas, potentially benefiting critical trauma patients. However, the survival advantage of HEMS over ground emergency medical services (GEMS) remains unclear. This systematic review evaluates the impact of HEMS versus GEMS on in-hospital trauma mortality.

Methods:

A literature search (2014–2024) identified studies reporting trauma patient survival and emergency medical service (EMS) transport modality. Inclusion/exclusion screening used Rayyan and multiple reviewers. Secondary inter-facility and nontrauma transports were excluded. The primary outcome was in-hospital mortality, with additional analyses of injury severity score (ISS), Glasgow Coma Scale (GCS), and regional differences. Analysis was done using an inverse variance weighting method.

Results:

Of 628 articles, 29 were included, covering 3,101,673 patients (602,068 HEMS, 3,158,152 GEMS). HEMS was linked to higher in-hospital mortality (odds ratio [OR] = 1.27; 95% confidence interval, 1.05–1.53; I² = 98.99%). HEMS patients had a higher mean ISS (14.42 vs. 10.21, P = 0.02) but similar GCS (P = 0.34). Day 1 and 30-day mortality showed no significant differences. For pediatrics (7 studies, 281,906 patients), HEMS was also associated with increased mortality (OR = 2.47; P = 0.001; I² = 97.2%) and higher ISS (11.67 vs. 6.33, P = 0.11). Regional analysis showed higher HEMS mortality odds in North America but not in Asia, Europe, or the Middle East.

Conclusion:

Despite its advantages, HEMS was associated with higher trauma mortality, suggesting selective use based on patient and regional factors. Further research should refine transport criteria.

INTRODUCTION

Trauma-related deaths continue to be a significant global health issue, resulting in approximately 5 million fatalities each year.[1] A timely emergency medical services (EMS) response to trauma is critical for patient survival, with the ‘Golden Hour’—the first hour during which prompt treatment is most effective—being particularly importance.[2] As a result, the introduction of helicopter transport for civilian use seems intuitively helpful as it provides rapid transport times over long distances, access to difficult terrain, and larger medical interventions with an experienced crew.[3,4,5] However, helicopter EMS (HEMS) also carries drawbacks, including longer start-up times, higher costs, and limited cabin space.[4,5] Ultimately, the literature has been mixed regarding the benefits of both HEMS and ground EMSs (GEMS) for survival odds.[3,4,5] The purpose of our systematic review is to assess the impact of transport modality on in-hospital mortality.

METHODS

The literature database PubMed was used to identify articles that mentioned trauma patients, survival or mortality data, and an EMS transport modality (helicopter or ground transport). The search strategy consisted of keywords and synonyms: “helicopter transport,” “ground transport,” “HEMS,” “GEMS,” “trauma,” “injury,” and “mortality.” The abstract screening was done through Rayyan software and corroborated blindly by multiple reviewers, where a minimum of 2 reviewers were assigned per paper. This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines.

The inclusion criteria were limited to English-language articles – cohort studies, case–control studies, or clinical trials, published between 2014 and 2024. To calculate odds ratios (ORs), articles had to include the following data: total population transported for both ground and air and the number of deaths for both ground and air. Articles with only one transport modality were excluded due to the inability to calculate ORs. We excluded articles on GEMS and HEMS transport between inter-facility, nontrauma cases, transport via private vehicles, police transport, and medevac combat settings. Articles that did not define EMS modality as air or ground and instead used mixed transport method modalities were excluded. Similarly, articles that did not assign the number of patient deaths to a specific transport modality in data tables were also excluded [Figure 1]. All ages were included; however, separate search and in-hospital mortality analysis were conducted for articles that only addressed “children” or “pediatric” populations [Figure 1]. The majority of included studies were retrospective in design and primarily assessed the odds of mortality associated with different transport modalities. A few studies utilized cohort designs, though these were less common. Most articles focused on mortality outcomes, specifically comparing the survival rates between air and ground transport in trauma patients [Table 1].

Figure 1.

Flowchart: Identification of articles for systematic review

Table 1.

Characteristics of included articles

Author	Year	Country	Type	Age Group	Mortality Time Frame	Secondary Outcomes
Kushida Y[34]	2020	Japan	Retrospective cohort study	adult	In-hospital	NA
Aiolfi A[7]	2017	USA	Retrospective cohort study	adult	In-shotpial,1-day, 30-day	NA
Abe T[30]	2014	Japan	Retrospective cohort study	adult	In-hospital	NA
Hosomi S[28]	2022	Japan	Retrospective cohort study	adult	In-hospital	NA
Lee KH[35]	2022	S. Korea	Retrospective cohort study	adult	In-hospital, 1-day	NA
Nabeta M[15]	2021	Japan	Retrospective cohort study	adult	In-hospital	NA
Weinlich M[10]	2019	Germany	Retrospective cohort study	adult	In-hospital	NA
Buchanan IM[36]	2016	Canada	Retrospective cohort study	adult	In-hospital	NA
Kim J[31]	2015	Korea	Retrospective cohort study	adult	In-hospital	ISS
Elkbuli A[37]	2022	USA	Retrospective cohort study	adult	In-hospital	NA
Beaumont O[8]	2020	England	Retrospective cohort study	adult	In-hospital	GCS, ISS
Dominguez OH[17]	2020	USA	Retrospective cohort study	adult	In-hospital	ISS
Brown JB[18]	2016	USA	Retrospective cohort study	adult	In-hospital	ISS
Schneider AM[9]	2021	USA	Retrospective cohort study	adult	In-hospital	ISS
Michaels D[4]	2019	USA	Retrospective cohort study	adult	In-hospital	NA
Al-Thani H[16]	2019	Qatar	Retrospective cohort study	adult	In-hospital	NA
Sun H[38]	2017	USA	Retrospective cohort study	adult	In-hospital	GCS, ISS
Zhu TH[39]	2018	USA	Retrospective cohort study	adult	In-hospital	NA
Tscuchiya A[40]	2016	Japan	Retrospective cohort study	adult	In-hospital	NA
Vercruysse GA[19]	2015	USA	Retrospective cohort study	adult	In-hospital	GCS, ISS
Roman J[13]	2020	USA	Retrospective cohort study	adult	In-hospital	NA
Taylor BN[20]	2018	USA	Retrospective cohort study	adult	In-Hospital	NA
Den Hartog D[41]	2015	The Netherlands	Retrospective cohort study	adult	In-hospital	NA
Jitsuiki K[42]	2022	Japan	Retrospective cohort study	adult	In-hospital	NA
Sutherland M[14]	2021	USA	Retrospective cohort study	adult	In-hospital	NA
Guinzburg A[43]	2022	Isreal	Retrospective cohort study	adult	In-hospital	NA
Andruszkow H[3]	2016	Germany	Retrospective cohort study	adult	In-hospital, 1-day	ISS
Zadorozny EV[5]	2023	USA	Retrospective cohort study	adult	In-hospital, 1-day	ISS
Alstrup K[11]	2020	Denmark	Retrospective cohort study	adult	30-day	NA
Lapidus O[33]	2023	Sweden	Prospective observational study	adult	30-day	NA
Enomoto Y[44]	2020	Japan	Retrospective cohort study	pediatric	In-hospital	NA
Englum BR[24]	2017	USA	Retrospective cohort study	pediatric	In-hospital	NA
Blasius FM[25]	2021	Germany	Retrospective cohort study	pediatric	In-hospital	NA
Brown J[45]	2016	USA	Retrospective cohort study	pediatric	In-hospital	NA
Stewart CL[46]	2014	USA	Retrospective cohort study	pediatric	In-hospital	NA
Polites SF[22]	2017	USA	Retrospective cohort study	pediatric	In-hospital	NA
Missios S[47]	2014	USA	Retrospective cohort study	pediatric	In-hospital	NA
Brown JB (2)[21]	2016	NA	Retrospective cohort study	NA	NA	GCS, ISS
Greer JW[48]	2019	NA	Retrospective cohort study	NA	NA	GCS
Malekpour M[49]	2017	USA	Retrospective cohort study	NA	NA	GCS, ISS
Duke M[2]	2018	USA	Retrospective cohort study	NA	NA	GCS, ISS
Wandling MW[50]	2017	USA	Retrospective cohort study	NA	NA	ISS
Al-Shahrani MS[51]	2017	Saudi Arabia	Retrospective cohort study	NA	NA	ISS
Maher Z[52]	2021	USA	Retrospective cohort study	NA	NA	ISS

NA: Not available, ISS: Injury severity score, GCS: Glasgow Coma Scale

The primary outcome was unadjusted in-hospital mortality. Propensity-matched groups were not included in our study. A separate analysis was done for day-1 and 30-day crude mortality. Mortality en route to hospitals, at the scene, and at other measured time points were excluded. Data for 1-year mortality were part of our inclusion criteria, but statistical analysis could not be done as there were not enough articles that used this outcome and that included both transport modality groups. Statistical analysis was done using an inverse variance weighting method. A geographical subgroup analysis was also done for our mixed-aged population which was divided into the following geographic regions: Asia, North America, South America, Africa, Oceania, Europe, and the Middle East.

The secondary outcome was calculating the mean of the median injury severity score (ISS) and Glasgow Coma Scale (GCS) between air and ground-transported groups. An independent t-test was used for statistical analysis. This systematic review is registered on PROSPERO (ID: CRD42024629274).

RESULTS

Our PubMed search yielded 628 articles, whose abstracts were reviewed. After removing 10 duplicates and excluding irrelevant articles, a total of 29 studies provided data for both air and ground transportation modalities in mixed-aged populations, making them the only ones eligible for in-hospital mortality statistical analysis. A total of 15 articles were analyzed for ISS and 8 for GCS. A total of 3,703,741 patients were included in the study on in-hospital mortality, with 602,068 transported by HEMS and 3,101,673 by GEMS. Our study included 2,331,097 (75.2%) males and 1,345,280 females (24.8%). Some articles lacked sex/gender information for specific trauma transport modalities or had incomplete EMR data. For all articles, the use of HEMS was associated with increased odds of in-hospital mortality compared to GEMS (OR = 1.46; 95% confidence interval [CI], 1.19–1.80, I² = 98.9%) [Figures 2 and 3]. For mixed-aged populations, the results are as follows: OR = 1.27; 95% CI, 1.05–1.54, I² = 99.1% [Figures 2 and 3].

Figure 2.

Forest plot: Trauma patient in-hospital mortality odds (ground emergency medical services vs. helicopter emergency medical services): (a) All articles: odds ratio (OR) =1.46; 95% confidence interval (CI), 1.19–1.80, I² = 98.9%, (b) Mixed population: OR = 1.27; 95% CI, 1.05–1.54, I² = 99%

Figure 3.

Forest plot: Sensitivity analysis of mixed population trauma patients (ground emergency medical services vs. helicopter emergency medical services): odds ratio = 1.26; 95% confidence interval, 1.05–1.52, I² = 99%

Finally, we looked at articles exclusively looking at pediatric populations. Out of 625 articles, 7 were included for analyzing in-hospital mortality. Of these 7 articles, 3 were compared for ISS. The study included a total of 281,906 patients: 185,986 men, 95,914 women, and 6 individuals with unreported gender. Of these, 52,214 were transported by helicopter, while 229,692 were transported by ground ambulance. Two articles reported age using means, three reported age ranges, and two used median ages. The use of HEMS is associated with higher odds of in-hospital mortality compared to GEMS (OR = 2.47; 95% CI: 1.24–4.90, P = 0.001, I² = 97.2%) [Figure 4]. The mean ISS was 15.50 for the HEMS group and 7.37 for the GT group (P = 0.10; 95% CI: 2.42–18.6). A confounding variable to note is that sicker patients may have been transported by helicopter.

Figure 4.

Forest Plot: In-hospital mortality for pediatric population articles (ground emergency medical services vs. helicopter emergency medical services). (a) Pediatric population: odds ratio (OR) =2.47; 95% confidence interval (CI), 1.24–4.90, I² = 97.1%, (b) Pediatric population, sensitivity analysis: OR = 1.93; 95% CI, 1.31–2.84, I² = 92.8%

HEMS-transported patients had an average median ISS of 14.4 and GCS of 14.4, compared to GEMS-transported patients with an ISS of 10.2 and GCS of 14.8 (P = 0.02 for ISS, P = 0.34 for GCS). Five articles provided data for day-1 mortality and 3 articles were included for 30-day mortality. No significant difference in mortality was observed in the day-1 group (OR = 0.78; 95% CI, 0.32–1.88, I² = 98.7%) [Figure 5] or 30-day group (OR = 1.13; 95% CI, 0.27–4.70, I² = 95.5%) [Figure 5].

Figure 5.

Forest plot: Mixed population articles, day-1 and 30-day mortality (ground emergency medical services vs. helicopter emergency medical services): (a) Day-1 mortality: odds ratio (OR) =0.78; 95% confidence interval (CI), 0.32–1.88, I² = 98.7%, (b) 30-day mortality: OR = 1.13; 95% CI, 0.27–4.70, I² = 95.5%

Of the 29 articles, 12 reported the average age for the HEMS/GEMS groups, eight reported the median, four gave a range, and two did not assign the age with the transport modality but rather incorporated age into trauma type or propensity-matched groups, respectively.

A separate analysis was done on different geographic regions. There were only sufficient data for North American (n = 15 articles), Asian (n = 8), European (n = 4), and Middle Eastern (n = 2) countries. For North American patient populations, HEMS use had increased mortality odds (OR = 1.44, 95% CI, 1.12–1.81) [Figure 6]. For Asian, Middle Eastern populations, and European countries, there was no significant difference in mortality odds between HEMS and GEMS groups. Their respective results are as follows: OR = 0.98; 95% CI, 0.74–1.31, PI: 0.46–2.09; OR = 1.17; 95% CI 0.33–4.18; and OR = 1.52; 95% CI, 0.00–2215.34 [Figures 6 and 7].

Figure 6.

Forest plot: In-hospital mortality for North America and Asia (ground emergency medical services vs. helicopter emergency medical services): (a) North America: odds ratio (OR) =1.44, 95% confidence interval (CI), 1.12–1.81, PI: 0.72-2.83, (b) Asia: OR = 0.98, 95% CI, 0.74–1.31, PI: 0.46–2.09

Figure 7.

Forest plot: In-hospital Middle East and Europe (ground emergency medical services vs. helicopter emergency medical services): (a) Middle East: odds ratio (OR) =1.52; 95% confidence interval (CI), 0.00–2215.34, PI: 0.00–350507.46, (b) Europe: OR = 1.17; 95% CI, 0.33–4.18, PI: 0.18–7.52

DISCUSSION

Mortality analysis

Our study reveals that both adult and pediatric HEMS-transported patients had an association with higher in-hospital mortality than GEMS-transported patients, a result previously unreported in previous systematic reviews.[6] Moreover, adult HEMS-transported patients had higher ISS scores than GEMS-transported patients while GCS between the groups was not statistically significant. For pediatric patients, no significant difference was seen in ISS scores. When broken down by geography, no difference in crude in-hospital mortality was seen except for North American populations, where HEMS use was associated with increased mortality odds.

In contrast to our results, Butler et al. showed that the use of HEMS yielded improvements in trauma mortality in their systematic review, yet acknowledged the multifactorial nature of trauma survival beyond transport modality.[6] Similarly, many articles included in our review pointed out the merits of HEMS transport. Using a stepwise logistic regression analysis, Aiolfi et al. mentioned that HEMS provides a survival benefit in patients with isolated severe head trauma when corrected for using GCS, age, and other risk factors.[7] However, they also mention that crude mortality rates were higher with HEMS transport. Michaels et al. also reported an increased crude mortality rate among HEMS-transport patients, yet after adjusting for age, ISS, and gender, HEMS-transport conferred increased survival odds using a logistic regression model.[4] In England, Beaumont et al. drew the same results.[8] Indeed, when mortality is adjusted for using parameters such as ISS, age, and injury severity, among many others, the benefits of HEMS for trauma survivability were seen.[9,10]

Despite numerous studies showcasing the advantages of HEMS in trauma survival, the conclusion remains uncertain. Using Kaplan–Meier cumulative mortality curves, Alstrup et al. demonstrate that HEMS-transported patients had higher mortality rates.[11] HEMS was also associated with higher mortality in cases of septic shock.[12] Moreover, Roman et al. reported higher mortality rates among HEMS-transported patients with large burns, however no difference between transport modalities among patients with small burns.[13] Furthermore, Sutherland et al. and Nabeta et al. reported no difference in patient outcomes between HEMS and GEMS groups.[14,15] In Qatar, Al-Thani et al. conducted a univariate analysis showing increased ISS and in-hospital mortality in patients transported by HEMS. However, a multivariate analysis revealed that transport modality had no impact on survival.[16] Several other papers also corroborate our result of increased ISS scores in HEMS groups.[9,10,16,17,18]

Given the uncertainty surrounding HEMS use, it is also important to discuss variable operational costs. Vercruysse et al. reported an average helicopter ride of $18,000 in the United States while Roman et al. reported an average cost of $29,000.[13,19] Likewise, a systematic review by Taylor et al. concluded that HEMS is generally more expensive, with unclear evidence regarding its overall benefits.[20] Thus, HEMS transport should be considered only when necessary. Moreover, the difficulty in conclusively evaluating the cost-effectiveness of HEMS transport stems from variations in health/trauma systems. It is this concept that also applies to evaluating the ideal transport modalities for reducing trauma mortality. Brown et al. emphasized the need for patient-specific use of HEMS and a call for more research to identify which patient subgroups benefit most from HEMS transport.[21] Brown et al. also stressed the importance of understanding the interplay of prehospital distance/time and HEMS provider resources, while Al-Thani underscored the significance of identifying environmental factors that may affect prehospital transport time. Ultimately, many factors should be taken into consideration when evaluating survival odds.[16,21]

This study also evaluated the effect of EMS transport modality on a subset of the pediatric population. HEMS-transported pediatric trauma patients had an increased risk of mortality and an increased mortality risk compared to adult populations (OR: 2.29 vs. 1.27). Both pediatric and adult HEMS-transported patients had higher ISS scores, yet the difference was only significant among adults (P = 0.11 vs. 0.02). Polites et al. also reported increased mortality odds in HEMS-transported children but saw equivalent mortality odds when propensity matched.[22] Similarly, Hakakian et al. saw no difference in mortality between HEMS/GEMS groups before and after ISS matching.[23] In contrast, other investigators saw increased survival odds with HEMS use in pediatric populations using regression analysis.[24,25] The heterogeneity with these results can be explained by the same factors explained in adult populations. Within our study, the increased mortality odds for crude in-hospital mortality (OR: 2.47 vs. 1.27) among pediatrics can be explained by a smaller sample size of pediatric studies. Although age may be a differentiating factor, Englum et al. demonstrated that this was not a factor in survival odds, instead mental status and trauma type were linked to increased survival odds.[24] The admission to an adult trauma center (ATC) or pediatric trauma center (PTC) may also be a factor as pediatric patients treated at PTCs as opposed to ATCs improved survival.[26,27]

Geographic analysis

Our results show that HEMS use is associated with increased in-hospital mortality odds in North American countries, with no difference in mortality observed in Asian, European, or Middle Eastern countries. These differences may be attributed to EMS staffing, hours of operation, and overall trauma care infrastructure.[28] For example, HEMS staffing in Japan is exclusively physician-staffed unlike in the United States and operations are done during daylight.[28] This may explain why HEMS use was associated with decreased mortality odds in traumatic brain injury (TBI) patients in Japan and why the OR for Asian articles was the lowest among all regions analyzed (OR = 0.98; 95% CI, 0.74–1.31).[28,29] Notably, 6 of the 8 studies in this region were from Japan, with the remainder from South Korea. In South Korea, geographical characteristics such as a small land mass, high urban population density, and scattered rural areas in mountainous terrain make helicopter transport a more effective option than ground transport. However, it remains underutilized due to concerns over cost, safety, and being newer and less robust.[30] In contrast, Qatari HEMS care has grown in the last few decades. The Qatar Trauma System (QTS) has extensive training for its healthcare providers with continuous education on advanced trauma care techniques.[16] Additionally, the system operates 24-hour EMSs and has grown its HEMS fleet in recent decades.[16] This extensive HEMS system may thus explain the HEMS survival benefit seen in Al-Tahani’s analysis. However, at the unadjusted mortality level, our Middle Eastern mortality analysis (which includes Israeli HEMS and Qatari HEMS) demonstrated no difference between HEMS and GEMS as previously mentioned.

Like Japanese EMS systems, German HEMS use is activated only in serious trauma cases, such as head trauma, major blunt or penetrating injuries, critical hemorrhage, and life-threatening trauma requiring rapid transport to specialized care.[28,31] The centrality and specificity of German trauma care/HEMS is similar in Scandinavian countries.[11] Notably, the overuse of HEMS has been more established in the United States and may account for the differences in cost-effectiveness.[14,19] Yet, our results indicate that transport modality did not affect crude mortality odds in European (OR = 1.17; 95% CI 0.33–4.18, PI: 0.18-7.52) and Middle Eastern countries (OR = 1.52; 95% CI, 0.00–2215.34, PI: 0.00–350507.46). Once more, mortality odds become heterogeneous when propensity matching is done, and these differences may be explained by the previously mentioned factors.[11,32] It is important to note the variability in annual HEMS operating costs for countries. Taylor et al. contextualize factors such as fuel and HEMS personnel costs in relation to health system funding and the geographic environments in which HEMS operate.[20] Thus, the cost of HEMS may thus serve as a barrier for resource-limited countries to employ successful HEMS care. This is evident as most of our HEMS data and literature contributories come from resource-rich countries.

Strengths and limitations

Our study is the first of its kind to break down trauma transport modality odds by age and geographic variation, including a total of 3,703,741 patients across 38 articles over 10 years. This comprehensive analysis captures the variability among pediatric populations and across different HEMS modalities worldwide. By including diverse regions – North America, Asia, Europe, and the Middle East – it offers valuable insights into regional differences in EMS systems and trauma care. The use of a multi-author blinded abstract review strengthens the reliability of our findings by minimizing bias.

As for limitations, this review utilized PubMed as the primary and only literature database. The exclusion of other literature databases may have limited the breadth of literature capture, and this is acknowledged as a methodological constraint. Furthermore, the included studies conferred high heterogeneity with no subgroup analysis generating an I² less than 95%. In addition, our systematic review only analyzed crude mortality between these two transport modalities. Thus, factors such as prehospital distances, staffing of HEMS/GEMS, type of trauma, severity of trauma, the nature of trauma centers, and patients’ disposing health conditions were not accounted for. Investigating the relationship between the type of trauma and transport modality would be fruitful. What is clear is that time to receive trauma care is key.[33] Intuitively, HEMS transport may provide this advantage, but lowered mortality odds remain unclear and over-use in minimally traumatized patients, in particular children, is seen.[19,20,21,22]

CONCLUSION

Adult trauma patients transported by HEMS are associated with higher in-hospital crude mortality and higher ISS scores than those transported by GEMS, with no difference in GCS. The mortality odds between HEMS and GEMS-transported patients showed no significant difference for day-1 or 30-day survival. For pediatric populations, HEMS use was also associated with increased in-hospital crude mortality odds with no difference in ISS scores. Geographic variations show no difference in mortality odds between HEMS and GEMS modalities, except in North American countries where HEMS use is associated with increased mortality odds. Ultimately, we advocate for a holistic and patient-specific approach to trauma care as transport modality alone is not the sole factor governing the survival odds of patients.

Research quality and ethics statement

The authors followed applicable EQUATOR Network (https://www.equator-network.org/) guidelines during the conduct of this research project.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.