O1014: The BRACHA – Is It Able to Assess Risk of Agitation in the Pediatric ED?

Background: Rates of children presenting to the pediatric emergency department (PED) for mental and behavioral health (MBH) concerns are increasing. Although agitation among MBH patients poses a significant safety risk, early prediction of unsafe behaviors may help ameliorate this risk. The BRACHA is a validated 14-question assessment performed in the PED that is predictive of aggression in the inpatient psychiatry setting; however, no such tools exist for prediction of unsafe behaviors in the PED. The purpose of this study is to assess the discriminative ability of BRACHA to predict agitation requiring intervention (ARI) within the PED among patients presenting for MBH concerns.

Methods: This was a retrospective cohort of patients 5-18 years old, presenting with MBH concerns to two PEDs affiliated with a tertiary care children’s hospital over a 9-year period. All patients had a BRACHA completed by trained personnel. Demographic, assessment, and intervention data abstracted from the electronic medical record were used to characterize the cohort. Primary outcome was ARI in the PED defined as an episode of agitation or unsafe behavior requiring physical (i.e., physical hold or mechanical restraint) and/or pharmacologic intervention. Analysis focused on patients requiring both physical and pharmacologic intervention—those deemed at highest risk of injury. Receiver operator characteristic (ROC) curves were used to evaluate the discriminative ability of the BRACHA to predict ARI within the PED. The patient population was divided in a randomized manner into two cohorts—Derivation and Validation—in a 7:3 ratio, respectively. The Youden J Index was used to determine the optimal cutoff score for identification of patients at highest risk for agitation requiring both physical and pharmacologic intervention. Performance characteristics using the determined cutoff are described for the Validation cohort.

Results: During the study period, there were 32,905 MBH encounters with documented BRACHA scores. A total of 3,526 patients had at least one episode of ARI in the PED—2,027 required pharmacologic intervention and 2,250 required physical intervention. The area under the ROC for BRACHA as a predictor of agitation requiring both types of intervention was 0.83 and 0.82, in the Derivation and Validation cohorts, respectively. A BRACHA of ≥ 6 was identified as the optimal cutoff. This provides a sensitivity of 84.8%, specificity of 66.4%, PPV of 5.6%, and NPV of 99.5% in Validation cohort. Patients with a BRACHA ≥ 6 had odds ratio 12.0 (95% CI: 9.7, 14.7) for requiring both pharmacologic and physical intervention in the PED, compared to those with lower BRACHA scores.

Conclusion: BRACHA score is significantly associated with ARI within the PED and is most predictive of agitation requiring both physical and pharmacologic interventions. Patients with a BRACHA score ≥ 6 are at significantly increased odds of having ARI in the PED.