The overall survival rate after TCA remains low, as demonstrated by our study, which reported a rate of 1.8%. However, the cross-region and TC groups had a higher overall survival rate of 4.2%. The cross-region TC and TC groups exhibited a higher probability of any ROSC, improved survival to admission, and increased 24-h and 30-day survival rates. TC distance and transport time showed no significant association with survival prognosis. However, blunt injury–induced TCA was associated with poor prognosis.

The overall survival rate in our study closely aligns with the reported rate of 1.5% from a single trauma center in a study conducted in southern Taiwan from 2014 to 2016 [9]. However, when patients were transported or bypassed to TC, the overall survival rate was 4.2%, more than eight times higher than that of the non-TC group. Over the past decade, the survival rate for patients sent to TC has increased significantly. In contrast, the Pan-Asian Resuscitation Outcomes Study, conducted across 13 Asian countries, reported a survival to discharge rate of only 3.4%, regardless of hospital level [15].

Regarding survival to admission, previous studies have reported rates ranging from 16.5% to 18% [16, 17]. Our study presents similar findings, with any ROSC rate and survival to admission at 18.3% and 14%, respectively. However, in the TC group, the rates of any ROSC and survival to admission were significantly higher, at approximately 30.6% and 29.2%, respectively. In the cross-region TC group, the ROSC rate and survival to admission were also higher, at 30.5% and 18.6%. After adjusting for covariates, we found that both direct and cross-region transport to a TC were associated with improved outcomes in TCA patients compared to those in the non-TC group. A study conducted in Toronto, which aligns with our findings, indicated that the 24-h and 48-h mortality rates were lower in patients transported directly to a trauma center compared to those transported to a non-trauma center [18]. Another study indicated that the rate of mortality (including in-hospital mortality) after severe trauma was significantly lower for patients transported to a TC than for those transported to a non-TC [19]. In a Swedish study, the adjusted rate of 30-day mortality was 41% lower in patients transported to a TC than in those transported to a non-TC, especially for critically ill patients [20]. However, groups of these three papers were for all trauma patients, not just TCA patients, which is different from ours.

In contrast, some studies have suggested transporting TCA patients to lower-level hospitals. U.S. studies have shown that, after adjusting for confounders such as patient demographics, injury patterns, and Injury Severity Score, the survival rate of TCA patients was higher at Level II hospitals compared to Level I hospitals [21, 22]. However, the authors of these studies noted that performance improvement programs and regionalized trauma care could be significant predictors of survival. Neither study mentioned factors influencing survival outcomes, such as the distance to the destination or proximity to nearby medical institutions.

We found that the probability of overall survival was the highest for patients directly transported to a TC, followed by those with cross-region transportation to a TC and transportation to a non-TC. Similar trends were observed for survival to admission and 30-day survival rates. A study in southern Taiwan reported a significant association between transportation to a TC and the achievement of sustained ROSC in patients with TCA [11], which aligns with our findings. However, that study did not account for distance or proximity to nearby medical facilities. To the best of our knowledge, the present study is the first to report that cross-region transportation to a TC, despite requiring more transport time (a median of 7 min in our study), improves the probability of survival.

In Taiwan, the current protocol for transporting patients with TCA is to the nearest hospital, regardless of its level. Our findings challenge this practice. Once ROSC is achieved, TCA patients require care from a multidisciplinary team, which is only available in Level I TCs. While transport to a TC increases the probability of survival, the time required for transport plays a critical role in patient outcomes. A Taiwanese study reported a negative association between prolonged transport time and survival in patients with TCA [17]. Additionally, several studies have shown that a response time of over 8 min has no significant effect on the survival rate following trauma [7, 23]. In a relevant study, the probability of transportation to a TC decreased by 5% with a 1 km increase in the distance of the nearest TC from the TCA location; this finding suggests that the hospital distance, rather than the injury pattern, is a crucial factor influencing transportation decisions [24]. In the present study, the median transport time was 12 min (interquartile range: 9–17 min) in the cross-region TC group, while it was 5 min in the non-TC group. The longest transport time recorded in our study was 42 min. Despite the extended transport time, our findings indicate that directly transfer to a TC over a longer distance result in better survival probabilities, and the transfer distance had no significant impact on survival outcomes. Thus, considering the convenience and intensity of EMSs in Taiwan, we recommend transporting patients with TCA to a TC with a relatively short transport time. However, the optimal approach may be different for countries with limited EMS or a vast territory.

Many factors might influence the outcomes of TCA; these factors include injury pattern, cardiac arrest recognition, bystander CPR, prehospital intervention, transportation strategy, and hospital care. Some studies show that trauma mechanisms are not associated with ROSC and survival rate [25, 26], but some think that blunt injury is associated with poor outcomes [5, 6]. However, our study findings indicate that blunt injury leads to poor outcomes after TCA. Notably, no association has been observed between in-hospital mortality and pre-hospital advanced airway management, age, or bystander CPR [27].

The present study has some limitations. First, this was an observational, retrospective study. EMT–paramedics made transportation decisions considering real-time factors. These decisions might have been influenced by factors such as age and trauma mechanism. This might have introduced a selection bias in this study. Nevertheless, we attempted to correct this through multivariable analyses. Future studies should adopt a prospective design, as the current study is based on retrospective data, where patient transport decisions by EMTs may introduce bias. By implementing a prospective study design with a standardized transport protocol for EMTs to follow, the risk of bias can be minimized.

Second, Taoyuan City has a high population density and hosts numerous hospitals. In this study, we did not consider the effects of traffic congestion, day–night variations, and holidays, particularly in rural and low-density areas. Thus, our results should be interpreted with caution.

Finally, we lacked data on the patients’ anamneses, which might have influenced our results pertaining to prognosis. Nonetheless, most of the patients were young adults with relatively good health conditions.

Despite the aforementioned limitations, we recommend cross-region transportation to a TC for patients with TCA in high-density population area.