A 5-Year Review on the Trauma Population and TraumaRelated Mortality, in Stockholm, Sweden

Introduction

Trauma represents a major global public health burden and remains the leading cause of death among adults under 45 years of age, while its incidence and associated mortality are steadily increasing in older populations [1-3]. Worldwide, approximately 4.4 million injury-related deaths occur annually, with marked regional variation reflecting demographic, socioeconomic, and healthcare system differences [3,4]. High‑income countries have demonstrated reductions in trauma-related mortality through the implementation of structured trauma systems, standardized protocols [5], and continuous quality improvement initiatives [1,6]. In Sweden, the Swedish Trauma Registry (SweTrau) provides a national platform for monitoring trauma epidemiology, outcomes, and system performance [5]. Karolinska University Hospital (KUH) is Sweden’s largest Level I trauma centre, serving a catchment area of approximately 2.5 million inhabitants, and the local trauma registry is a part of the national SweTrau database [79]. Previous studies from KUH reported a predominance of elderly male patients, blunt injury mechanisms—particularly falls—and a gradual reduction in preventable trauma deaths between 2007 and 2016 [7,8]. Since then, Stockholm’s population has continued to grow, and trauma activation criteria and care pathways have evolved [9,10]. However, contemporary data describing trauma epidemiology, mortality patterns, and preventability in Stockholm are lacking. The assessment of preventable trauma deaths remains a cornerstone of trauma quality improvement. Systematic peer review of trauma‑related mortality enables identification of modifiable errors and informs targeted interventions [1,8]. The present study aimed to (1) describe current trauma demographics at KUH, (2) compare characteristics of survivors and non-survivors, (3) identify causes of trauma-related death, and (4) evaluate the proportion and nature of preventable and potentially preventable deaths over a recent five‑year period.

Methods Study Design and Setting

This retrospective observational study analysed data from the Trauma Registry at the Karolinska University Hospital, Solna, Sweden, covering the period from January 1, 2018, to December 31, 2022.

Study Population

Eligible patients were aged ≥15 years and met at least one of the following criteria: trauma team activation, New Injury Severity Score (NISS) >15, or inter-hospital transfer within seven days of injury. Patients with burns, drowning, or hypothermia as primary diagnoses were excluded.

Data Collection

Extracted variables included age, sex, injury mechanism and intent, comorbidities (ASA physical status), injury severity scores (ISS and NISS), prehospital cardiac arrest, Glasgow Coma Scale (GCS) at admission, hospital length of stay, and 30‑day mortality.

Mortality Review and Classification

All deaths occurring within 30 days of injury were reviewed in multidisciplinary morbidity and mortality conferences in accordance with World Health Organization trauma quality improvement guidelines [1]. Deaths were classified as preventable, potentially preventable, or non-preventable. Causes of death were determined using ICD-10 codes and clinical chart review.

Statistical Analysis

Data normality was assessed using the Shapiro-Wilk test. Continuous variables were analysed using independent t-tests or MannWhitney U tests as appropriate, and categorical variables using χ² tests. Binary logistic regression was performed to identify factors associated with 30-day mortality. In addition to hypothesis testing, crude (unadjusted) associations between selected binary variables and 30-day mortality were summarized using odds ratios (OR) with 95% confidence intervals derived from 2×2 tables. Annual proportions of potentially preventable deaths and traumatic brain injury (TBI)‑related deaths were summarised with 95% confidence intervals, and linear trend tests were applied to evaluate temporal change. Statistical significance was defined as p<0.05. Analyses were conducted using IBM SPSS Statistics version 29.0.1.0.

Ethical Considerations

The study was approved by the Swedish Ethical Review Board in Stockholm (Dnr 2024-03805-01). The requirement for informed consent was waived due to the retrospective registry-based design.

Results Patient Characteristics

During the study period, 6,483 trauma patients were admitted, of whom 396 (6.1%) died within 30 days. Non-survivors were significantly older than survivors (mean age 60 vs. 44 years, p<0.001). Male patients predominated in both groups. Blunt trauma accounted for the majority of injuries in survivors and non-survivors. Falls were the most common injury mechanism overall, with low-impact falls more frequent among non-survivors. Intentional self‑inflicted injuries were more prevalent among non-survivors than survivors (12.9% vs. 6.9%, p<0.001). Crude effect size estimates demonstrated markedly increased odds of 30-day mortality among patients with prehospital cardiac arrest (crude OR 143.15, 95% CI 91.96-222.84) and among those presenting with GCS <9 (crude OR 6.00, 95% CI 4.39‑8.20). Increased crude odds were also observed for ASA physical status 3‑4, low‑impact falls, and self‑inflicted injuries, whereas stabbing injuries were associated with lower crude odds of death.

Injury Severity and Clinical Outcomes

Non‑survivors had significantly higher injury severity scores, with 77% presenting with ISS >15 compared with 28% of survivors. Prehospital cardiac arrest occurred in 42.6% of non-survivors. A GCS score <9 at admission was observed in 36% of non‑survivors versus 3% of survivors.

Predictors of 30-Day Mortality

Logistic regression identified age ≥60 years, ASA physical status 3‑4, GCS <9, high injury severity scores, and intentional injury as independent predictors of 30-day mortality.

Preventability and Causes of Death

Of the 396 deaths, 28 (7.0%) were classified as potentially preventable and 2 (0.5%) as preventable. The annual proportion of potentially preventable deaths declined from 12.3% in 2018 to approximately 5% in 2021-2022, although the trend did not reach statistical significance. Delayed time to surgery or definitive treatment was the most frequently identified error. Traumatic brain injury was the leading cause of death and accounted for an increasing proportion of fatalities from 2018 to 2022, rising from 32.9% to 47.4% of trauma-related deaths.

Discussion

This study provides a contemporary, five‑year overview of trauma epidemiology, mortality patterns, and preventability at a Swedish Level I trauma centre. Several important findings emerge. First, trauma-related mortality remains strongly associated with advanced age, severe injury, and impaired neurological status at admission. Second, traumatic brain injury (TBI) continues to be the dominant cause of death and appears to constitute an increasing proportion of trauma fatalities over time. Third, the proportion of preventable and potentially preventable deaths is low and has declined compared with earlier reports, although system-level delays—particularly delayed access to definitive surgical or interventional care—remain a recurring contributor to potentially preventable mortality [7,8].

Demographic shift and the growing burden of geriatric trauma

Age was one of the strongest predictors of 30-day mortality in this cohort. Patients aged ≥60 years had more than fourfold higher odds of death compared with younger patients, even after adjustment for injury severity and physiological derangement [11-18]. This finding is consistent with prior Scandinavian and international studies demonstrating that chronological age, comorbidity burden, and reduced physiological reserve substantially worsen trauma outcomes [7,13,14,16]. Older trauma patients are also more susceptible to in-hospital complications, including infections, which may further contribute to adverse outcomes in this population [15]. The predominance of low-impact falls among non-survivors further highlights the shifting epidemiology of trauma in high-income countries. Falls from standing height or similar mechanisms are often perceived as low-energy injuries, yet in older adults they frequently result in devastating intracranial injury. These findings support prioritization of fall‑prevention initiatives and targeted risk-reduction strategies in elderly populations [18].

Neurological injury as the principal driver of trauma mortality

Traumatic brain injury was the leading cause of death across all study years and accounted for an increasing proportion of fatalities from 2018 to 2022. Although the observed temporal trend did not reach statistical significance, the consistent increase aligns with broader epidemiological observations of declining haemorrhage-related deaths and a relative rise in deaths due to severe neurological injury [16]. Severe neurological impairment at admission, reflected by a GCS score <9, was among the strongest predictors of mortality.

Injury severity, physiological derangement, and prehospital cardiac arrest

Markers of injury severity and physiological collapse were strongly associated with mortality. Patients presenting with ISS or NISS >15 had markedly increased odds of death. Prehospital cardiac arrest was associated with extremely high mortality, reflecting the profound physiological insult and limited salvageability of this subgroup. These findings support continued investment in advanced prehospital care and rapid access to definitive trauma services.

Intentional injury and mechanism‑specific patterns

Intentional injuries, particularly self‑inflicted trauma, were more frequent among non-survivors and independently associated with increased mortality. This observation is consistent with prior studies demonstrating higher lethality and complexity among patients presenting after self-harm [11,12]. These patients may present with severe physiological compromise, complex injury patterns, and concomitant psychiatric illness, complicating both acute management and post-resuscitation care.

Preventability, system performance, and opportunities for improvement

The low proportion of preventable and potentially preventable deaths suggests continued improvement in trauma system performance [7,8]. Nevertheless, delays in definitive care remain a recurring issue, indicating residual vulnerability in time-critical pathways [8]. Advances in haemorrhage control and resuscitation strategies, including the use of tranexamic acid, may also influence evolving patterns of trauma mortality in mature systems [17].

Limitations

This study is limited by its retrospective, single-centre design. Although SweTrau provides high-quality registry data, misclassification and missing data cannot be entirely excluded [5]. The classification of preventability is inherently subjective, although standardized multidisciplinary peer review mitigates this limitation [1,8].

Conclusion

Trauma-related mortality at Karolinska University Hospital is increasingly driven by advanced age, severe injury, and traumatic brain injury. The continued decline in preventable deaths reflects effective trauma system development, yet delays in definitive care persist. Focused strategies addressing geriatric trauma and system‑level inefficiencies may further improve outcomes.

Conflict of interest: The authors declare that they have no conflicts of interest

Ethical approval and consent to participate: Ethical approval for this study was obtained from the Swedish Ethical Review Authority in accordance to Swedish and European ethical guidelines (DNR 2024-03805-01). Consent for participation was deemed not required as per the ethical review guidelines as all data was coded and cannot be traced to individual persons.

Funding: There was no funding for this research.

Consent for publication: Not applicable.

Acknowledgments: We would like to thank Trauma Register, Karolinska University Hospital, for the data.

Reference

1. World Health Organization (2009) Guidelines for trauma quality improvement programmes.
2. Wilson W, Grande C, Hoyt D (2007)Trauma. New York: Informa Healthcare.
3. World Health Organization (2021)Injuries and violence. World Health Organization.
4. Amaefule KE, Dahiru IL, Ejagwulu FS, Maitama MI (2020) Trauma mortality in the emergency department of a tertiary hospital in a lowincome country: it’s time to walk the talks. West Afr J Med.37: 131-137. 
5. SweTrau. Årsrapport 2022. Swedish Trauma Registry; 2022.
6. Candefjord S, Asker L, Caragounis EC (2022) Mortality of trauma patients treated at trauma centers compared to non-trauma centers in Sweden: a retrospective study. Eur J Trauma Emerg Surg.48: 525536.
7. Ghorbani P, Falkén M, Riddez L, Sundelöf M, Oldner A, et al (2014) Clinical review is essential to evaluate 30-day mortality after trauma. Scand J Trauma Resusc Emerg Med.13: 18.
8. Ghorbani P, Strömmer L (2018) Analysis of preventable deaths and errors in trauma care in a Scandinavian trauma level-I centre. Acta Anaesthesiol Scand.62:1146-1153.
9. Stockholms stad. Statistik (2021) Befolkningsöversikt Stockholm årsrapport.
10. Linder F, Holmberg L, Björck M, Juhlin C, Thorbjornsen K, et al (2019) A prospective stepped wedge cohort evaluation of the new national trauma team activation criteria in Sweden - the TRAUMALERT study. Scand J Trauma Resusc Emerg Med.30: 52.
11. Singhal A, Ross J, Seminog O, Hawton K, Goldacre MJ (2014) Risk of self-harm and suicide in people with specific psychiatric and physical disorders. J R Soc Med.107: 194-204.
12. Varley J, Pilcher D, Butt W, Cameron P (2012) Self-harm is an independent predictor of mortality in trauma and burns patients admitted to ICU. Injury.43: 1562-1565.
13. Sammy I, Lecky F, Sutton A, Leaviss J, O’Cathain A (2016) Factors affecting mortality in older trauma patients - a systematic review and meta-analysis. Injury. 47: 1170-1183.
14. Grossman MD, Miller D, Scaff DW, Arcona S (2002)When is an elder old? Effect of preexisting conditions on mortality in geriatric trauma. J Trauma.52: 242-246.
15. Bochicchio GV, Joshi M, Knorr KM, Scalea TM (2001) Impact of nosocomial infections in trauma: does age make a difference? J Trauma.50:612-617.
16. Søreide K, Krüger AJ, Vårdal AL, Ellingsen CL, Søreide E, et al (2007) Epidemiology and contemporary patterns of trauma deaths: changing place, similar pace, older face. World J Surg. 31: 2092-2130.
17. Roberts I, Shakur H, Coats T, Hunt B, Balogun E, et al (2013) The CRASH-2 trial. Health Technol Assess.17:1-79.
18. Beserra Da Silva R, Costa-Paiva L, Siani Morais S, Mezzalira R, Oliveira Ferreira ND, et al (2010) Predictors of falls in women with and without osteoporosis. J Orthop Sports Phys Ther.40:582-588.