Background: EMS responses to nonfatal opioid overdoses represent an opportunity to engage people who use opioids and facilitate their linkage to various treatment, social, and other services. However, a notable share of patients refuse to be transported to the emergency department (ED). Non-transports prevent the continuation of care in a hospital setting, potentially inhibiting the initiation of or referral to treatment and other services, and are associated with increased risks of a subsequent overdose. They also cause hospital claims data to miss a non-trivial share of overdoses.
Objective: This paper examines what patient- and event-level characteristics are associated with increased odds of patients refusing transport after an overdose and discusses possible policy responses.
Methods: The paper is a cross-sectional analysis of the latest (2022) data from the National EMS Information System, a nation-wide database of EMS activations. It uses multiple logistic regressions with a binary outcome measure of whether the patient was released against medical advice. Patient-level predictor variables include age, gender, and race/ethnicity. Event-level predictors include patient acuity, system response time, scene time, prior naloxone administration, type of EMS agency, location type, urbanicity, and census region.
Implications: Identifying factors associated with transport refusals can improve our understanding of the overdose crisis in the United States by examining events not included in hospital data. It is also critical for designing policy and practice interventions that would either reduce transport refusals or mitigate their negative impacts, e.g., via pre-hospital treatment induction.