P2.158: Law enforcement presence during pediatric emergency department encounters and associated encounter level outcomes

Background: Law enforcement intersects with emergency medical care often. Qualitative data describes challenges faced by physicians, nurses, and law enforcement officers (LEOs) when medical and criminal justice priorities intersect in the emergency department (ED). Despite known challenges, the prevalence of these interactions and impact on care delivery and outcomes is unknown. This study uses a validated, electronic health record (EHR)-based algorithm to identify LEO presence during pediatric ED encounters to examine related patient demographics and encounter level outcomes including length of stay (LOS), and time to clinical care during pediatric ED encounters with and without LEO presence.

Methods: .This is a retrospective cohort study of EHR data from ED encounters at a single children’s hospital between 10/1/2021 and 10/1/2022. Encounters were included if patients were < 22 years old and excluded if transferred from another hospital. A validated multivariable probabilistic model including a keyword algorithm, chief complaint, and arrival method was used to identify LEO presence. Demographic and time variables were extracted from discrete EHR data fields. To investigate demographic variables associated with LEO presence, odds ratios were calculated from multivariable logistic regression with all other variables in the model. Adjusted means were calculated from linear regression models after log transforming the time variables. Antilogs were taken of the model means to revert to the original time measure. LOS and time to vitals, provider, triage, and room were adjusted for triage level and arrival method.

Results: Of the 147,126 encounters meeting inclusion criteria, the EHR-based algorithm identified 2,169 with LEO presence (1.5%). Encounters involving patients aged 12-17 years old and those with public insurance had the highest odds of LEO presence (p < 0.0001), while encounters with Hispanic or Latino patients had lower odds of LEO presence (p < 0.0001). There was no significant association between LEO presence and race or sex (Table 1). The adjusted mean LOS for encounters with LEO presence was more than twice as long as those without LEO presence (5.3 vs 2.4 hours, p< 0.0001). Time to first vitals, time to triage, time to room, and time to provider evaluation were also all significantly longer in encounters with LEO presence (p < 0.0001) (Table 2).

Conclusion: In this population, ED encounters with LEO presence were significantly longer and aspects of initial clinical evaluation (e.g. first vitals) are also delayed. This supports qualitative data describing perceived delays in care and prolonged encounters during ED encounters involving LEOs. Encounters with adolescents had the highest odds of LEO presence, likely reflecting LEO presence for firearm injuries, assault, and mental health-related complaints. There was no association between LEO presence and race however larger studies in more diverse populations are needed. Further studies to identify impact on clinical outcomes (e.g. time to pain medication) are needed.