Twitter  Linkedin

Back to 2nd Annual Meeting Program

Evaluating Variation in Process of Care Metrics for Critically Injured Pediatric Trauma Patients
Katherine Flynn-O'Brien, MD, MPH and Frederick P. Rivara, MD, MPH, University of Washington, Seattle WA; Mary E. Fallat, MD, University of Louisville, Louisville Kentucky; Tom B. Rice, MD, Medical College of Wisconsin, Milwaukee WI; Christine Gall, DrPH, MS, RN, SCL Health System, Denver CO

Abstract: Introduction: Efforts to improve outcomes after pediatric trauma depend on availability of detailed data to evaluate processes of care. We developed the Pediatric Trauma Assessment and Management database by merging two independent data systems at 5 Pediatric Trauma Centers to assess variation in key care processes during different phases of the care continuum. Methods: Trauma registry and intensive care unit data were merged for all children <18 years discharged from the ICU after a traumatic injury during calendar year 2013. Additional variables abstracted targeted imaging, resuscitation, ICU and floor management practices. Chi2 test for independence was completed to compare site-specific proportions. Univariate and multivariable regression (controlling for age, mechanism and severity of injury) were used to quantify process of care variation across sites. Results: The merged database of 692 children showed large variation in use of the FAST exam (site-specific variation 0-37%, p-value <0.001), and both mechanical (0% vs. 19%, <0.001) and pharmacologic VTE prophylaxis (0-11.1%, <0.001). Cervical spine collar (C-collar) placement after head injury varied significantly by site in children transferred from another facility (50-85% with C-collar placed, <0.001), but not in children who presented directly (69-81%, 0.720). Conclusions: These data suggest substantial variation in management practices; specifically, there appears to be differential practices related to C-collar placement prior to transfer to higher levels of care. Combining two existing datasets has the potential to provide a unique platform for evaluation of processes of care across the care continuum.

Objective: Identify processes of care that vary by site based on a unique multi-institutional database.

Objective Content: There are many variations in care processes across the nation. The PTAM database is unique in that it has the potential to explore process of care throughout the care continuum. There is statistically significant variation in FAST exam, pharmacologic and mechanical DVT prophylaxis, and in C-collar placement by hospital and/or location of care.

Back to 2nd Annual Meeting Program