Do outcome measures for trauma triage agree?

OBJECTIVE

The goal of trauma triage is to match resources to the needs of seriously injured patients. The trauma triage literature has used a variety of outcome measures to assess appropriate trauma activation. The objective of this study was to determine the agreement between procedural and nonprocedural outcome measures in a population of seriously injured patients transported to a single trauma center.

METHODS

Study authors reviewed all "level 2" trauma activations (January 2002-December 2003) at an American College of Surgeons (ACS) Level 1 trauma center. "Level 2" trauma activations were based on modified ACS Committee on Trauma (COT) triage criteria. Outcomes were classified as nonprocedural (Injury Severity Score [ISS] > 15 and intensive care unit [ICU] admission) and procedural (nonorthopedic emergent surgery, emergency chest tube placement, emergency department intubation, emergency department transfusion, or emergent interventional radiology care).

RESULTS

Of 479 patients, five were transferred out of hospital. The remaining 474 were predominantly male (62%), with a mean age of 39.7 years. Their average ISS was 13.2. There were nine deaths. For all subjects, 144 (30%) were admitted to the ICU, 172 (36%) had an ISS > 15, 80 (17%) received an emergent procedure, and 46 (10%) went for emergent surgery. Kappas comparing agreement of ISS > 15 with emergent resuscitation and emergent surgery were 0.31 and 0.15, respectively. Kappas comparing ICU admission with emergent resuscitation and emergent surgery were 0.51 and 0.26, respectively.

CONCLUSIONS

We identify moderate to poor agreement between nonprocedural and procedural outcomes of trauma triage in this population.