609 - Factors Associated with Emergency Holds in Acute Pediatric Mental and Behavioral Health Presentations

Publication Number: 1586.609

Christopher Gable, Children's National Health System, Washington, DC, United States; Mickey Emmanuel, Childrens National, Washington, DC, United States; Marshall Brown, Seattle Children's Research Institute, Seattle, WA, United States; Sydney Meszaros, Children's National Health System, Washington, DC, United States; Olivia Miller, Children's National Health System, Alexandria, VA, United States; Theresa Ryan. Schultz, The George Washington University, Jenkintown, PA, United States; Monika Goyal, Children's National Medical Center, Washington, DC, United States; Shilpa J.. Patel, Children's National Hospital, Bethesda, MD, United States

Background: Youth emergency department (ED) visits for mental or behavioral health (MBH) concerns are increasing. Many patients with MBH concerns arrive under an emergency hold, which allows law enforcement or a qualified clinician to involuntarily detain a person for psychiatric evaluation if they believe that person has an acute mental illness, and, without immediate care, may harm themselves and/or others. The impact of pediatric emergency holds on ED care has not been well described.

Objective: We sought to describe sociodemographic and clinical factors associated with emergency hold status.

Design/Methods: We conducted a retrospective analysis of patients ages 6-18 years presenting to a large, urban, pediatric ED between January 2019-December 2024 with an MBH concern. We summarized sociodemographic and clinical characteristics using descriptive statistics. To identify factors associated with emergency hold status, we conducted a multivariable logistic regression analysis adjusted for these respective variables and applied generalized estimating equations (GEE) for repeated measures.

Results: Of the 15,162 patients with an MBH complaint, 16% arrived under emergency hold, 62% were Black, 61% female, 64% had public insurance, 38% had acute suicide concerns, 17% had an aggression-related ICD-10 diagnosis, and 8% had mobile crisis involvement prior to arrival (Table 1). Emergency hold status was significantly associated with Black race (aOR:3.04; 95%CI 2.23-4.15), physical restraint use while in the ED (aOR:2.11; 95%CI 1.34-3.34), documented aggression-related ICD-10 diagnosis (aOR:2.82; 95%CI 2.44-3.26), mobile crisis involvement (aOR:3.97; 95%CI 3.32-4.75), one or more ED visits in the prior 12 months (1 ED visit: aOR:1.96; 95%CI 1.68-2,30; 2+ ED visits: aOR:4.09; 95%CI 3.21-5.22), and longer ED lengths of stay (aOR:1.02; 95%CI 1.01-1.02) (Figure 1). Emergency hold status was inversely associated with having private insurance (aOR:0.25; 95%CI 0.29-0.42) and one psychiatric admission in the prior 12 months (aOR:0.7; 95%CI 0.55-0.89) (Figure 1).

Conclusion(s): 1 in 6 youth presenting to the ED with an MBH concern arrived under an emergency hold. Youth under an emergency hold were more likely to identify as Black and be physically restrained in the ED, after controlling for sociodemographic and clinical factors including aggression-related ICD-10 diagnosis. Understanding factors related to initiation of emergency hold may help to inform interventions that alleviate previously reported disparities in restraint use.

Table 1: Demographic and Clinical Covariates by Emergency Hold Status and Multivariate GEE Model for Emergency Hold Use
Adjusted odds ratio is adjusted for age, race, biological sex, ethnicity, year of visit, hour of visit, insurance status, Emergency Services Index (ESI) level, CAMHD-CS classification, return to ED visits within 7 days, 30 days, and 12 months, psychiatric hospitalizations within 12 months, neurodivergent disorders, laboratory and radiology tests.


Figure 1: Forest Plot of Sociodemographic and Clinical Variables Associated With Emergency Hold Status