Variability in trauma center outcomes for patients with moderate intracranial injury.

BACKGROUND: The variability of outcome between Trauma Centers has not been extensively studied as a possible avenue for performance improvement. Trauma Center variability in severity-adjusted survival for patients with moderate intracranial injury (MII) was studied in order to determine the association of MII-related process of care variables with outcomes. The analytic results were supplemented with peer review of MII patients with unexpected outcomes and identified potential process of care variables. METHODS: A retrospective cohort study was undertaken based on data submitted to a statewide trauma center database from July '95 through June '98. MII patients had one or more selected ICD-9-CM codes with an AIS-90 severity score of 3 or 4 but no higher. Severity adjustment was done using case matching and a logistic function based New Model that appropriately accounts for patients intubated on Emergency Department arrival. MII-related process of care variables derived from the database were identified and their relationship with outcome were evaluated individually and using multivariate methods. Trauma center personnel conducted standardized peer reviews. RESULTS: The study included data from 6765 patients treated at 26 trauma centers. Two centers (2PZW) had significantly more survivors than expected by both severity adjustment methods. Three had significantly fewer survivors than expected (3NZW). By several measures, patients treated in the 2PZW centers were more seriously injured and older than those in the 3NZW centers. CT of the head performed in the treating hospital was the only process of care variable associated with outcome in multivariate evaluations. Peer review also found little association between process of care variables and patient outcomes. However, peer review reported that 23.7% of unexpected deaths identified by case matching or the New Model were preventable or potentially preventable. Peer review also identified as medically unnecessary significant percentages of patients with unexpectedly long stays in hospital (26.4%) or in ICU (17.3%) identified by case matching. Nearly 45% of unexpected complications were judged preventable or potentially preventable. CONCLUSIONS: Two severity adjustment methods identified significant variability in trauma center outcomes for patients with MII. The difference in outcomes between the centers with better than expected (2PZW) and poorer than expected outcomes (3NZW) was substantial. Peer review identified significant opportunities for reducing unexpected deaths, stays in hospital and in ICU, and the occurrence of complications. Trauma registry data and peer reviews found little relationship between available process of care variables and patient outcomes. This study should stimulate discussions to understand reasons for outcome variability and ways to reduce it.

A new "TRISS-like" probability of survival model for intubated trauma patients.

BACKGROUND: Prehospital trauma patient field intubations and paralyzations, using neuromuscular blocking agents before emergency department respiratory and neurologic assessments are made, bias assessments and outcome evaluations using probability-of-survival models, such as TRISS and A Severity Characterization of Trauma (ASCOT). We present a newly developed "TRISS-like" probability-of-survival model for intubated blunt- and penetrating-injured patient assessment. METHODS: From a population of 51397 consecutively admitted trauma patients, this study used all 5740 (11.2% of the total injured population) intubated patients with complete data from a statewide trauma registry from October 1, 1993, to September 30, 1996. Model performance was evaluated using standard calibration and discrimination measures and z and W statistics of significance. RESULTS: The new model accurately predicted survival for blunt- and penetrating-injured intubated patients and is applicable to 11 etiologic patient populations. CONCLUSION: Study findings suggest that the new TRISS-like model should be used to assess both blunt- and penetrating-injured intubated patients. Use of this new model provides an analytical method for addressing a significant limitation of both the standard TRISS and ASCOT models, which are not applicable to intubated injured patient assessment. In addition, use of this model will complement TRISS/ASCOT assessments of nonintubated trauma patients and thus permit appropriate assessments for both intubated and nonintubated injured patient study populations.