O1022: Evaluating the Screening Tool for Early Predictors of Post-traumatic Stress Disorder in Predicting Post-traumatic Stress Outcomes in a Pediatric Trauma Cohort

Background: Millions of children are treated for traumatic injuries in the United States each year. Though most patients recover well, more than 30% of injured youth develop signs and symptoms of post-traumatic stress disorder (PTSD). Early identification of at-risk children can inform prompt intervention to prevent adverse secondary health outcomes associated with post-traumatic stress. The Screening Tool for Early Predictors for PTSD (STEPP) is a pediatric acute care screening tool previously tested using DSM-IV PTSD criteria. To determine the tool’s performance with current diagnostic criteria, we sought to evaluate STEPP in identifying pediatric trauma patients at risk for DSM-V post-traumatic stress outcomes.

Methods: We performed a secondary analysis of a prospectively enrolled cohort originally used to evaluate the PsySTART triage tool. In this study, we identified children ages 8-17 years who met criteria for a trauma alert activation at a freestanding quaternary children’s hospital from June 2019-July 2021. Children were excluded if they were: victims of intentional trauma; required immediate psychiatric intervention; presented with a GCS ≤12; wards of the state or likely to face criminal charges; deaf and/or unable to converse in English. Social workers administered STEPP as part of routine care for all trauma activations. At one-month post-trauma, patients were contacted to complete an interview regarding life stressors, mental health service utilization, and the Child PTSD Symptom Scale for DSM-V (CPSS-5). Only children with complete STEPP and study follow-up data were included in the analysis. We evaluated the sensitivity, specificity, positive and negative predictive values for positive parent STEPP, child STEPP, and overall STEPP (child or parent positive) screens; given the clinical implications of identifying children with varying levels of post-traumatic stress, these test performance measures were calculated at different thresholds of CPSS-5 symptom severity and functional impairment.

Results: We identified 167 children who met our study inclusion criteria (median [IQR] age, 12.6 [10.3-15.5] years; 59.9% male; 70.7% White and 98.1% non-Hispanic) (Table 1). The most common mechanism of injury was motor vehicle crashes (25.1%). 23.4% of patients had level 1 trauma activations. The performance characteristics of a positive parent, child, or overall STEPP score in predicting CPSS-5 symptom severity scores ≥20, as well as ≥1 functional impairments, at one month follow-up are summarized in Table 2.

Conclusion: STEPP did not accurately identify children with DSM-V post-traumatic stress symptoms or functional impairments after injury. Reliably identifying children at risk of developing different levels of traumatic stress symptoms and functional impairments is critical in informing a staged behavioral health care model to prevent and manage PTSD. These findings underscore the continued critical need to develop and validate novel screening tools to characterize post-traumatic stress and PTSD risk in the pediatric trauma population.