A Novel Application of Statewide Emergency Department Data to Classify Behavioral Health Holds.

Abstract

Context: Emergency department (ED) "boarding" for behavioral health occurs when patients remain in emergency rooms overnight due to a lack of available psychiatric beds. However, quantification of the problem and demographic characterization of affected populations have been limited and manual. When a patient stays in the ED awaiting either an inpatient psychiatric bed or community care, this is known as a behavioral health hold (BHH). Understanding of BHH requires statewide, 'real-time' data collection that does not widely exist.

Program: Collaborators leveraged an existing syndromic surveillance system to compile ED data from across North Carolina into an executive BHH dashboard by developing a custom data extract, implementing a BHH case definition, and curating the data in a secure environment. Data come from 131 hospitals submitting ED visit data to NC DETECT, and the resulting extract includes patients with ED encounters lasting at least 24 hours for a behavioral health reason from November 2022 to November 2024.

Implementation: The team implemented and evaluated multiple case definitions of BHH based on ICD-10 codes or keywords listed within the free-text encounter chief complaint. One case definition and four metrics were selected for use within a dashboard based on subject matter expert feedback. US Census Bureau data were used to create population denominators and estimate population rates, which are presented in an equity context.

Evaluation: This dashboard enabled data-driven funding and policy decisions for additional crisis stabilization resources, including Facility Based Crisis (FBC) facilities, Behavioral Health Urgent Care (BHUC) facilities, and Mobile Response and Stabilization Services.

Discussion: A cross-departmental approach provided the necessary expertise for state departments of health to capitalize on an existing syndromic surveillance system to address a behavioral health policy need. This novel approach, encapsulated in a dashboard, places intelligence about the burden of BHH directly in the hands of policy makers.