Background: Approximately 1.5% of all admissions result in an against medical advice (AMA) discharge in the US. These individuals are almost twice as likely to be readmitted when compared to non-AMA patients, and are more likely to have a history of drug use. Numerous studies have investigated these patients who discharge AMA; however, detailed characteristics of the subset of just intravenous (IV) drug users is very limited. This population is at risk of increased morbidity and mortality due to related organ system, infectious, or mental health complications. In this study, we aimed to describe the sociodemographic and clinical characteristics of people who inject drugs (PWID) and discharged AMA.
Methods: A retrospective cohort study was designed to investigate PWID and discharged AMA. The study population included male or female patients, age greater than 18 years, within the Trihealth Hospital System in Cincinnati, Ohio, who discharged AMA over a span of 5 years (Sept 2015-July 2020). The information was obtained from the Epic electronic medical records system, using the following search terms: patient left before being or having seen, refused treatment, procedure or treatment not conducted or done, refused treatment, elopement or eloped, left without treatment or AMA and opioid or opiate. Outcomes of interest were demographics, readmission and emergency department (ED) length of stay (LOS). Descriptive statistical analysis was performed using proportions for categorical variables and median interquartile range (IQR) for continuous nonparametric variables.
Results: Four hundred fifty-five patients were included in the study. The study population was predominantly white (91.3%), male (58.5%), and single (70.2%) with a median age of 33 (IQR 27-39). Sixty-nine percent were on Medicaid and 21.3% were self-pay or had no insurance. The prevalence of positive Hepatitis C infection status was 18.8% and previously diagnosed psychiatric conditions such as Major depressive disorder was found to be present in 5.8% of the population, Generalized anxiety disorder in 7.7%, and Bipolar I or II disorders in 2.4%. A majority left the ED (56%) and 37% left the inpatient setting. Forty-four percent were admitted for drug overdose and 30% for infectious complications. The 90-day readmission rate was 2%. Lastly, during the study period, 27 (6.3%) patients had a second AMA discharge, 5 (1.7%) had a third one, and 1 (0.4%) had a fourth one.
Conclusions: People who inject drugs and discharged AMA represent a homogeneous group of predominantly young white single males with Medicaid or no insurance who discharge from the ED or as an inpatient. Identifying vulnerable populations at high risk for AMA discharge and implementing preventative interventions early in their hospital course may be beneficial in improving AMA discharge rates, subsequent readmission rates, and overall morbidity in this patient population.