Racial Disparities in Emergency Restraint Use for Agitated Patients

ABSTRACT

Background: The COVID 19 pandemic and the murder of George Floyd have prompted healthcare organizations to reexamine racial inequities in their care, challenging us to produce lasting, fundamental change. Mental health disorders, both diagnosed and undiagnosed, have increased in volume and developed new challenges for acute care practitioners during the pandemic. Additionally previous research has suggested that there are intrinsic and extrinsic biases that affect how care is delivered to patients presenting with mental health crises. Method(s) Through nominal group technique, we identified topics for equitable-care-oriented QI in the emergency department (ED) of our Level-1 Trauma center. Initial review of triage, left-without-being-seen, and fast-track data did not demonstrate significant racial disparities in standard benchmarks. We therefore focused on behavioral codes and restraint use. We prospectively collected data on all behavioral codes over a 3-month period, including demographics, visit characteristics, and certain aspects of restraint use including type of restraint, length of restraints, medication use, and reinitiation of restraints. In addition to tracking these metrics, employee perceptions of the psychiatric mental health emergencies were polled and evaluated. Result(s) Our QI process identified varying levels of disparities in care. Over the study period, white, non-white, and black patients comprised 50.5%, 49.5%, and 28.7% of the ED patient population, respectively, and 50%, 50%, and 44% of the patients who were subject to behavioral codes. Of those patients who had behavioral codes called, restraints were used for 64.8% of white patients, 64.3% of non-white patients, and 67.2% of black patients. Of those arriving by ambulance or police, 20% arrived with pre-hospital restraints or handcuffs, and of those, 90.9% were placed in restraints on arrival to the ED. Of those patients who had restraints placed, 4-points were used for 34.1%, 26.1%, and 25.5% of white, non-white, and black patients, respectively, and the restraint chair was used for 30.7%, 38,6%, and 41.8% of those same groups. Medications were given to 80.7%, 88.7%, and 91.4% of white, non-white, and black patients who were placed in restraints, respectively, and to 77.4%, 80.6%, and 83.3% of those same groups of patients who were not placed in restraints. None of the differences were statistically significant. Of those patients who had restraints placed and then discontinued, 13% were re-restrained at some other point during their visit. Among other responses, nearly half of all ED employees thought that patients should ideally not be restrained during behavioral codes and that, if necessary, the restraint chair provides a better experience than 4-point restraints. Conclusion(s) Continuous QI around a variety of measures can identify disparities and targets for sustained anti-racist improvements in emergency department care. This study will guide further intervention and education around inequities in care in our department and has prompted further consideration of, when restraints are deemed necessary, preferentially using less invasive measures like the restraint chair over 4-point restraints. Although decision-making around chemical and physical restraints for mental health emergencies is complex and difficult to study, EDs should carefully examine their use through continuous QI in order to optimize patient-centered outcomes.