ATLS practices and survival at rural level III trauma hospitals, 1995-1999.

OBJECTIVE: To determine whether Advanced Trauma Life Support (ATLS) practices characterizing initial resuscitation and interfacility transfer at rural trauma hospitals are associated with risk-adjusted survival.
METHODS: Retrospective, observational analysis of rural injured patient survival. Process-of-care variables were associated with TRISS (trauma and injury severity score)-derived Z-statistics (95% confidence intervals) for high-risk population subsets (defined below). INCLUSION CRITERIA: all patients > or = 12 years of age entered into a statewide trauma system, January 1, 1995, to December 31, 1999, and initially presenting to Level III trauma centers (N = 4,961). EXCLUSION CRITERIA: pronounced dead on arrival (n = 26), directly admitted to hospital (n = 3), and unknown disposition at first hospital (n = 2). Process variables include: intubation in emergency department (ED) given Glasgow Coma Scale (GCS) score < 9 [INTUB], administration of blood products in ED given systolic blood pressure (SBP) < 90 mm Hg [BLOOD], trauma surgeon presence within 5 minutes of patient arrival given GCS < 9 mm Hg or SBP < 90 mm Hg [UNSTABLE-TS], trauma surgeon presence within 5 minutes of patient arrival given injury severity score (ISS) > 15 [ISS-TS], transfer to higher level of care given ISS > 20 and no hypotension [TRAN], transfer to higher level of care given GCS < 9 [TRAN-GCS].
RESULTS: For the high-risk subpopulations, the following Z-scores (with and without an intervention) were found:
CONCLUSIONS: Some ATLS interventions (BLOOD, TRAN, and TRAN-GCS) are associated with improved survival for selected high-risk subgroups in these 21 rural Level III trauma hospitals.