Leveraging FITBIR Data to Improve Clinical Practice of Severe TBI
Abstract
Traumatic brain injury (TBI) is a leading cause of death and disability in the United States. In 2013, there were approximately 2.5 million emergency department (ED) visits, 282,000 hospitalizations, and 56,000 deaths related to TBI in the U.S. Additionally, many TBI survivors live with significant disabilities, resulting in substantially reduced quality of life and increased socioeconomic burden. After someone suffers from a severe TBI, additional brain damage may occur from sudden changes in our vital signs. For many years, doctors have focused on preventing this additional brain damage by monitoring the level of pressure inside the brain, blood pressure, and other vital signs to determine whether they are becoming dangerously low or high. These vital signs that are captured in the first few days after TBI may help doctors to predict whether a patient will recover the function of his/her brain. In our past research studies, we have used statistical models and data science to learn what changes in these vital signs may predict if the patient recovers or not after brain injury. In these initial studies, we were very close in being able to accurately predict how a patient will recover from a TBI. With this new research proposal, we plan to expand on our previous studies by creating new prognostic statistical models that we expect to have much higher accuracy. This is possible by analyzing a larger amount of data from many research trials that have studied TBI within the past 10 years. By including different research trials, we can create prognostic statistical models that are more applicable to the general population. We will be able to use about three times as much data as in our previous studies, hoping to find patterns in the data that will increase our accuracy to prognosticate whether someone will or will not recover from a severe TBI. Additionally, we plan to expand the number of days’ worth of vital signs information that we use in the models. We aim to address the following two topics of interest according to FY19 JPC-6/CCCRP FITBIR Analysis Award Topics of Interest: (1) Utilization of FITBIR data to inform clinical practice guidelines for TBI (2) Exploration of the relationship between intensity of emergency, critical, and acute care to long-term disability This study has the potential to produce high-impact results in the following ways. First, high-accuracy prognostic models can guide the management of TBI because informed decisions about the likely outcomes for certain patients can play an important role in whether they should receive treatments that may not have any benefit. Second, the results could also inform clinical practice on ways to treat elevated pressure inside the brain that lead to better outcomes. Both civilian and military Service members who suffer severe TBIs are likely to be positively impacted by these results.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 10, 2021
- Source ID
- W81XWH2010770
Entities
People
- Jose-miguel Yamal
Organizations
- United States Army
- University of Texas Health Science Center at Houston