A model for an integrated emergency medicine/trauma service.

Abstract

To describe a model for an integrated multidisciplinary trauma service and to compare survival outcomes for patients resuscitated by either emergency medicine (EM) or surgical housestaff assigned to the trauma service. A prospective observational study was performed using injured patients evaluated in the trauma room at Hartford Hospital from July 1 through December 31, 1995. Inclusion criteria included an ICD-9-CM code of 800 through 959.9 and any of the following: transfer from another hospital, admission to the intensive care unit, hospitalization for > or = 23 hours, survival probability of < or = 90%, or Abbreviated Injury Score of > or = 3. Patients were excluded for burns necessitating transfer to a burn unit for definitive care, and for missing data elements that prevented a patient from being analyzed by the TRISS method. Data elements included mechanism of injury, Injury Severity Score, Revised Trauma Score, probability of survival, age, gender, and whether an EM resident was team leader. Patients in the EM cohort (group 1) were compared with patients for whom a surgical resident was team leader (group 2) for all data elements and for hospital survival. TRISS analysis was performed to evaluate outcomes in comparison with national norms. After exclusions, 609 patients were left for analysis. There were 141 (30%) resuscitated with an EM resident as team leader. No significant difference was found for matched variables between the groups. Both groups had good comparability with the Major Trauma Outcome Study (MTOS) database baseline, with M scores of 0.949 and 0.942, respectively. Outcomes for both groups also compared favorably with the MTOS norm for survival, with Z scores of 2.38 and 2.35 for groups 1 and 2. These results suggest that in this model of integrated EM/trauma service, equivalent survival outcomes occur whether EM or surgery housestaff act as team leaders.