This is part of a recurring series examining landmark articles in Emergency Medicine, in the style of ALiEM’s 52 Articles
After adjusting for differences in case mix the overall risk of mortality is 25 percent lower at a trauma center than a non-trauma center. The differences in risk of death is greatest for younger patients with higher Abbreviated Injury Scale scores (≥4).
Since 1976 the American College of Surgeons have advocated categorizing hospitals for care of traumatic injuries based on available resources. This has created a regionalized approach that we see today in management of trauma patients based off trauma level designations for each hospital and is affected substantially by geographic variability between states. The National Study on the Costs and Outcomes of Trauma (NSCOT) was established to address the differences in patient outcomes and costs between various hospitals, primarily trauma versus non-trauma centers. The authors concluded that though the effects of care at a trauma center varied based on severity of injury there was a mortality benefit in the more ill cohort. This study reviewed the outcomes for over 5,000 patients, however, many limitations still exist given the retrospective nature of case review. Management of trauma patients will likely continue to follow a regionalized model for many years to come given the intensive needs of some patients compared to others. As emergency physicians our role will continue requiring us to provide exceptional care from the moment the patient arrives in our resuscitation bay and acknowledging the elements that require emergent transfer to a hospital that can provide further specialized care.
This was a large case registry trial that examined patient data from across 14 states. 18 level-1 trauma centers and 51 non-trauma centers were included in the data collected. Data was gathered on 1104 patients who died in the hospital and 4087 who were discharged. The study was based on trauma centers in urban and suburban America and conclusions therefore cannot be extrapolated to more rural areas. On average non trauma centers were smaller and less likely to be members of the Council of Teaching Hospitals. The included non-trauma centers, however, were required to see a minimum of 25 patients with major trauma annually to be enrolled. Out of 131 non trauma centers asked to participate 51 agreed, but many cited a lack of administrate support to facilitate the study and therefore refused and 7 were refused by their IRB.
Patients aged 18-84 were eligible if they were treated for moderate to severe injuries as defined by the AIS>3. Interesting exclusions included those patients aged 65 or greater whose primary diagnosis included hip fracture. Patients with major burns were also excluded. Enrollment proceeded through a flow pathway that narrowed the pool from over 18,000 to 4,000 based on missing data in the chart review. The primary outcome was mortality in the hospital as well as death at 30, 90, and 365 days after injury. Patients were evaluated using the Charlson comorbidity index to balance the study arms and factors such as obesity and coagulopathy were added to the data collection models since these variables have a large impact on outcomes in traumatic injuries.
Patients treated in non-trauma centers tended to be older and unsurprisingly had more coexisting conditions. They also represented a larger portion of female, non-Hispanic white, and insured patients with overall less severe injuries. The unadjusted case fatality rate was higher at trauma centers, 8 v. 5.9%, however when adjusting for case mix, the risk of death at one year was significantly lower when care was provided at a trauma center compared to non-trauma center, 10.4 v 13.8%. The relative reduction in risk was similar for in-hospital, 30 and 90-day mortality. The data seemed to show a critical point at an AIS≥4 and age of 55.
One of the major limitations the authors cite in reviewing their data is that there is no perfect method to adjust for referral bias—largely “…the reality [is] that trauma centers treat a higher proportion of young, severely injured patients, whereas non-trauma centers treat a higher proportion of elderly patients with coexisting conditions.” Ultimately the authors arrived at a conclusion of benefit for the patients more severely injured and <55 years old when their care was performed at a trauma center. No trials will ever definitively prove this benefit through randomization and we will most likely see little change in our regional model of trauma care. Going forward, however, we may utilize points gathered in this large review to help aid us with clinical decisions about when to transfer patients to higher levels of care.
Level of Evidence:
Based on ACEP grading of level of evidence for prognostic questions this case registry is graded a level III.
Mackenzie, E. Rivara, F. Jurkovich, G. et al. “A National Evaluation of the Effect of Trauma-Center Care on Mortality.” NEJM, 2006:354(3). 366-78.
Faculty Reviewer: Dr. Matt Siket