719 - Predictors of a Non-Accidental Trauma Evaluation for Young Children with Isolated Head Injuries: A Cross-Sectional Study

Publication Number: 3696.719

Catherine Coughlin, Boston Children's Hospital, Boston, MA, United States; Michael Monuteaux, Boston Children's Hospital, Boston, MA, United States; Melissa Deane, Boston Children's Hospital, Boston, MA, United States; Hiu-fai Fong, Boston Children's Hospital, Boston, MA, United States; Lois Lee, Boston Children's Hospital, Boston, MA, United States; Sara Schutzman, Harvard Medical School, Boston, MA, United States; Rebekah Mannix, Boston Children's Hospital, Harvard Medical School, Boston, MA, United States; Todd W. Lyons, Boston Children's Hospital, Wayland, MA, United States

Background: Clinicians must consider non-accidental trauma (NAT) in infants presenting to the emergency department (ED) with head trauma. However, significant heterogeneity exists in which head-injured patients undergo an evaluation for NAT.

Objective: To evaluate factors associated with undergoing evaluation for NAT in children with head injuries diagnosed in the ED.

Design/Methods: We performed a retrospective cross-sectional study of children < 24 months old presenting to a Pediatric Health Information System (PHIS) ED from 2016-2023 with a diagnosis of head injury. Our primary outcome was the performance of a skeletal survey in the ED, a proxy for NAT evaluation. Our secondary outcome was the diagnosis of child abuse or injuries with high-specificity for NAT (retinal hemorrhages, rib fracture, scapular injury, and sternum injury based on ICD-10 codes). Multivariable logistic regression modelling was used to examine the association between the outcomes above adjusting for patient age, sex, race and ethnicity, insurance payor, presence of complex chronic conditions (CCC), presence of a traumatic brain injury (TBI), and Childhood Opportunity Index (COI) reporting adjusted odds ratios (OR) with 95% confidence intervals (CI).

Results: We included 172,049 ED encounters for head injuries. Median age was 10.0 months [interquartile range (IQR) 6,10 months] and 78,483 (45.7%) were female (Table 1). Of these,16,787 (9.8%) children underwent evaluation for NAT and 1,255 (0.73%) had a high specificity injury or diagnosis of child abuse. Factors associated with increased odds of NAT evaluation included younger age ( < 3 months), White non-Hispanic race, male sex, public insurance, low COI scores, and CCC (Table 2). Increased odds of high-specificity injury or child abuse were found among males and children with public insurance, CCC and those diagnosed with a TBI (Table 2).

Conclusion(s): Race and COI are associated with undergoing evaluation for NAT but not identification of a high-specificity injury or child abuse. Understanding factors associated with differential utilization of skeletal surveys will be essential to implement interventions to improve equity in NAT evaluations.

Table 1: Demographic and clinical characteristics of infants presenting to the emergency department with head injury, stratified by non-accidental trauma (NAT) evaluation (as measured by skeletal survey utilization) and the presence or absence of injuries with a high specificity for NAT
1 High Specificity= injuries with high-specificity for NAT (retinal hemorrhages, rib fracture, scapular injury, and sternum injury based on ICD-10 codes)
2 Validated composite measure of resources and conditions at the census tract-level comprised of 44 indicators in three domains (education, health and environment, and social and economic) (https://www.diversitydatakids.org/child-opportunity-index)
3 Feudtner C, Feinstein JA, Zhong W, Hall M, Dai D. Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation. BMC Pediatr. 2014 Aug 8;14:199. This is a well-validated classification system used in healthcare to identify children with multiple chronic illnesses.
4 The presence of the given diagnosis in any position (i.e., primary or non-primary diagnosis)



Table 2: The association between non-accidental trauma (NAT) evaluation (as measured by skeletal survey utilization) and high specificity injury and child abuse diagnosis with demographic and clinical factors
1 n=171,429; 620 (0.4% of total sample) observations were dropped due to missing data
2 n=16,726; 61 (0.4% of total sample) observations were dropped due to missing data
3 High Specificity injury = injuries with high-specificity for NAT (retinal hemorrhages, rib fracture, scapular injury, and sternum injury based on ICD-10 codes)
4 Validated composite measure of resources and conditions at the census tract-level, comprised of 44 indicators in three domains (education, health and environment, and social and economic) (https://www.diversitydatakids.org/child-opportunity-index)
5 Feudtner C, Feinstein JA, Zhong W, Hall M, Dai D. Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation. BMC Pediatr. 2014 Aug 8;14:199. This is a well-validated classification system used in healthcare to identify children with multiple chronic illnesses.
6 The presence of a traumatic brain injury diagnosis in any diagnosis list position (i.e., primary or non-primary diagnosis)