T. Husty1, M. Crandall1, D. Chesire1, D. Ebler1 1University Of Florida,Surgery,Jacksonville, FL, USA
Introduction: The State of Florida Adult Trauma Triage Criteria defines specific parameters that prompt paramedics to initiate a trauma alert, including injury mechanism and vital signs. In addition to these predefined criteria, paramedics may initiate an alert based on their own judgment. This activation is known as an alert based on Paramedic Discretion (PD). Our aim was to identify predictors of PD activations and to compare the outcomes of trauma patients who met objective alert criteria versus PD.
Methods: This is a retrospective observational study which included all trauma patients 18 years and older evaluated in our trauma center from January 1, 2007 to December 31, 2014. Demographic and injury severity variables were obtained from our trauma registry and outcomes were compared between patients who met state alert criteria and those who were brought by PD. We performed bivariate and multivariate statistics using SPSS and STATA.
Results: There were 13,305 patients who met state alert criteria and 1,188 alerted due to PD during the study period. Patients who were activated by state criteria had lower mean Glasgow Coma Scores [12 vs 14, p<0.001] and systolic blood pressures [126 vs 133, p<0.001]. On initial evaluation, 822 (6.4%) who met alert criteria and 14 (1.2%) activated via PD died [OR 3.7, 95% CI 2.5-5.5, p<0.001]. Admission rates were similar for both groups, as were lengths of stay. Regression modeling was unable to find independent or combinations of vital signs or demographics that would predict PD alerts [OR 1.4, 95% CI 0.74-2.62, p=0.31]. On multivariate modeling, PD was not an independent predictor of mortality or length of stay after controlling for injury severity and demographics.
Conclusion: As expected in an established trauma triage protocol, the mortality was higher for patients who met TA criteria but crude admission and discharge rates were similar. Though regression analysis could not elucidate predictors of PD, these results suggest PD may identify a subset of patients that benefit from trauma center evaluation. However, PD itself does not appear to be an independent predictor of mortality or length of stay after controlling for injury severity. In summary, PD remains a reasonable adjunct to state activation criteria, but further research into PD would help refine and codify these criteria.