Funded Projects

(SDAS) Variations In Care And Outcomes For Children With Traumatic Brain Injury

Grantee: Children's Hospital Medical Center
Principal Investigator: Tara Rhine
Project Number: R40MC26806
Project Date: 04/01/2014

Age group(s)

• Perinatal/Infancy (0-12 months)
• Toddlerhood (13-35 months)
• Early Childhood (3-5 years)
• Middle Childhood (6-11 years)
• Adolescence (12-18 years)

Abstract

Traumatic brain injury (TBI) is the most common cause of death and disability for children. There is a deficiency in the standardization of evidenced-based care for the management of pediatric TBI. Evaluating the comparative effectiveness and harms of different health care practices will highlight strategies to improve the care of and outcomes for children hospitalized with TBI. These findings will inform the development of updated evidenced- based TBI care guidelines, as well as identify practice patterns that need to be standardized for future multicenter randomized clinical trials. This will have significant impact on the health of children given the high incidence and morbidity associated with TBI. Purpose: In response to the Maternal and Child Health Bureau's Strategic Research Issue #III of evaluating "services and systems to assure quality of care for maternal and child health populations", we propose to examine variations in care practices (e.g. neurosurgical procedures, medication administration) and in-hospital outcomes (e.g. mortality, cardiac arrest, length of stay) for children hospitalized with TBI among 43 pediatric hospitals across the United States. The aims of the proposal are to characterize care practices, morbidity, and mortality for children with TBI and then assess the effects of these care practices, as well as injury characteristics and hospital characteristics, on in-hospital outcomes, while controlling for hospital and patient level covariates. Design: Retrospective cohort study Methods: Data will be retrospectively collected from the Pediatric Health Information System (PHIS) database by identifying youth hospitalized for TBI, ages 0-18 years, from January 1, 2004 - December 31, 2013. The PHIS database contains administrative data (e.g. demographics, diagnoses) and resource utilization data (e.g. pharmaceuticals, imaging) from 43 not-for-profit, tertiary care pediatric hospitals in the United States. Descriptive comparisons will be made among data collected over the past ten years from the PHIS hospitals with regards to patient population demographics (e.g. race, age), TBI characteristics (e.g. mechanism, injury pattern), care practices, and in-hospital outcomes: mortality, cardiac arrest, mechanical ventilation >96 hours, length of stay, cost of stay, and readmission and/or emergency room evaluation within 30 days of initial TBI, with the primary outcome being mortality. Mixed effects regression modeling will be used to compare care practices and outcomes between PHIS hospitals. To determine if care practices, injury characteristics, and hospital characteristics (overall size, trauma level designation) predict in-hospital outcomes, a mixed effects model using generalized estimation equations (GEE), with subjects as random effects, will be created.