Risk factors for abusive head trauma in the pediatric population

Objective: Abusive head trauma (AHT) is the leading cause of death from physical child abuse in children younger than 5 years of age in the United States. The mortality rate among patients with AHT is 25%, and the recurrence rate of child abuse rises to 35% when there is a lack of intervention. Thus, identifying child abuse is crucial yet especially challenging for infants and toddlers as they are preverbal. Current guidelines for child abuse do not sufficiently address the specific needs of a younger population. This study aimed to evaluate clinical factors associated with abuse among such populations.

Methods: The National Trauma Data Bank was queried from 2017 to 2019 for patients younger than 3 years with acute head trauma. Patients who were suspected of having experienced child abuse (suspected child abuse [SCA] group) were propensity score matched with patients who were not suspected of having experienced child abuse (non-SCA group) based on demographics, comorbidities, and Glasgow Coma Scale (GCS) scores. Paired Student t-test and chi-square tests were used to compare differences in hospital outcomes between the two groups. Multivariable regression analysis was used to determine factors associated with SCA (p < 0.05).

Results: The authors identified 10,844 patients in the SCA group and 27,912 in the non-SCA group. Regression analysis results showed that patients in the SCA group had higher rates of prematurity (OR 2.30, p < 0.001), GCS scores < 13 (OR 1.79, p < 0.001), congenital disorders (OR 1.56, p < 0.001), and public insurance use (68.38% vs 52.88% p < 0.001). Black and Hispanic patients were more likely to be in the SCA group (OR 1.56, p < 0.001 and OR 1.35, p < 0.001, respectively). Following propensity score matching, SCA patients had a longer length of hospital stay (3.17 vs 1.34 days, p < 0.001) and higher mortality rate (4.89% vs 3.58%, p < 0.001).

Conclusions: Acute head injuries in the SCA group were associated with prematurity, congenital disorder, low GCS score, and public insurance use. As such, the current guidelines should implement clinical history and insurance type to better reflect the at-risk patient population when evaluating infants and toddlers for potential abuse. There could be overidentification of child abuse among Black and Hispanic patients, and further research is warranted.