Definitive care in level 3 trauma centres after severe injury: A comparison of clinical characteristics and outcomes.

Background: The role of level 3 trauma centres (TC) in inclusive trauma systems has not been well defined. The absence of nationally recognised inter-facility transfer criteria for inclusive systems has often left individual level 3 TCs to decide upon their own what their spectrum of care is and particularly which severely injured patients to admit for definitive care.

Methods: Retrospective cohort study in which the principal objective was to compare the characteristics and outcomes of severely injured (injury severity score>15) adult patients (≥18 years) who received definitive care at level 3 centres with severely injured adult patients who were transferred to level 1-2 TCs during the same time period. Data were derived from the National Trauma Data Bank (2010-2011). First, we utilised hierarchical logistic regression models to evaluate the risk-adjusted mortality of patients admitted at level 3 TCs compared to those who were transferred to level 1-2 TCs. Subgroup analysis was carried out for patients with isolated traumatic brain injury (iTBI). Finally, we explored the extent of variation in risk-adjusted mortality across level 3 TCs.

Results: We identified 6433 severely injured patients who received definitive care across 150 level 3 TCs and 41,165 severely injured patients transferred to level 1-2 centres. Patients who received definitive care at level 3 TCs had a lower comorbidity burden and different injury profiles compared to those transferred to level 1-2 centres. There was no difference in crude mortality (10% vs. 11%, standardised difference 0.04); however, after risk-adjustment, the odds of death for patients who received definitive care at level 3 TCs were 1.24-fold higher (95%CI 1.08-1.43) when compared to those transferred to level 1-2 centres. A trend towards a higher likelihood of death at level 3 centres was observed when analysis was limited to patients with iTBI. Risk-adjusted mortality across level 3 TCs was with few exceptions, homogeneous (<10% of level 3 TCs were outliers with higher/lower mortality).

Conclusions: Level 3 trauma centres are providing definitive care for a subset of severely injured patients. Our findings suggest that the outcomes of severely injured patients admitted at level 3 centres might be worse compared to those transferred to level 1-2 centres; a finding independent of performance outliers. Further work is required to elucidate the determinants of admission after severe injury at level 3 trauma centres.