A descriptive study on the use of restrictive interventions for potentially or actually violent patients in the emergency department

Abstract

BackgroundRestrictive interventions (chemical, physical, or mechanical restraints) to manage patients who are potentially or actually violent in the emergency department (ED) can be harmful and costly. Non-restrictive interventions are advocated; but this must be preceded with an understanding of patient characteristics that influence their use. A study was conducted to describe the use of restrictive interventions and ascribe it with patient characteristics in the ED. MethodsRecords from October 2020 to March 2021 in the occupational violence database were used to analyse patient characteristics and restrictive interventions. Logistic regression was used to establish influencing factors of restrictive interventions adjusting for clinically relevant confounders. ResultsOf the N = 1276 potentially or actually violent patients, 70 % received restrictive interventions. Chemical restraint was common, with 1 in 2 patients receiving either oral medication or intramuscular injection. Probability of restrictive interventions were higher in patients who were intoxicated [(adjusted odds ratio (aOR) 3.48, 95 % confidence interval (CI) 1.675–7.21)], had high triage score (aOR 2.084, 95 % CI 1.094–3.96), and were in the ED involuntarily (aOR 1.494, 95 % CI 1.105–2.020). ConclusionThe results reveal the need for multifaceted approaches that limit the presentations of, and minimise restrictive interventions among, potentially or actually violent patients.