Needs Based Assessment of Trauma Systems 2, is it ready for primetime? A natural experiment testing its reliability

Dennis W Ashley; Etienne E Pracht; Melissa J Tetzlaff-Bemiller; Regina S Medeiros; Elizabeth V Atkins; and the Georgia Research Institute for Trauma Study Group

doi:10.1097/TA.0000000000003288

Needs Based Assessment of Trauma Systems 2, is it ready for primetime? A natural experiment testing its reliability

J Trauma Acute Care Surg. 2021 Sep 1;91(3):489-495. doi: 10.1097/TA.0000000000003288.

Authors

Dennis W Ashley¹, Etienne E Pracht, Melissa J Tetzlaff-Bemiller, Regina S Medeiros, Elizabeth V Atkins; and the Georgia Research Institute for Trauma Study Group

Affiliation

¹ From the Department of Surgery (D.W.A.), Atrium Health Navicent, Macon, Georgia; Department of Health Policy and Management (E.E.P.), University of South Florida, Tampa, Florida; Social Sciences Department (M.J.T.-B.), Augusta University; Department of Surgery (R.S.M.), Augusta University Health, Augusta; and Georgia Trauma Care Network Commission (E.V.A.), Atlanta, Georgia.

PMID: 34432754
DOI: 10.1097/TA.0000000000003288

Abstract

Introduction: Needs Based Assessment of Trauma Systems 2 (NBATS-2) attempts to predict the impact on patient volume and travel time for patients when a new trauma center (TC) is added to the system. The purpose of this study was to examine NBATS-2 predictive accuracy regarding expected volume and travel times of trauma patients at a newly designated TC and nearby legacy TCs when compared with actual data.

Methods: Needs Based Assessment of Trauma Systems predictive model for volume of trauma patients at the new TC was run based on 25th, 50th, and 75th percentiles of both state and National Trauma Data Bank (NTDB) patients per 100 TC beds. This was compared with the actual number of trauma patients from the State Discharge Data set before (2011-2012) and after (2016-2017) designation of the TC. Analysis was then augmented using the geographic information system (ArcGIS) spatial modeling to characterize median travel times for actual trauma patients, before and after designation of the TC.

Results: Both state and NTDB 25th, 50th, and 75th percentiles resulted in significant overestimation of volume at the new TC in 2016. After another year of TC maturation (2017), overestimation decreased but was still present. The 25th percentile from state and NTDB data sets provided the most accurate predictions. For the legacy TCs, the model switched from under to overestimation as the state and NTDB percentiles increased. The geographic information system accurately showed patients traveling <40 minutes to a TC nearly doubled.

Conclusion: Needs Based Assessment of Trauma Systems 2 provides an excellent template for state strategic planning; however, it overestimates new TC volume and under/overestimates volumes for legacy TCs depending on the state and NTDB percentiles used. This study shows that population density of the county in which the new or legacy TC is located should be considered when choosing the appropriate state or NTDB percentile. The geographic information system appropriately showed a decrease in trauma patient travel times after TC designation.

Level of evidence: Care Management, level V.

MeSH terms

Databases, Factual
Geographic Information Systems*
Georgia
Health Services Accessibility*
Humans
Needs Assessment / organization & administration*
Reproducibility of Results
Time Factors
Trauma Centers / organization & administration*
Travel
Wounds and Injuries / therapy