142 Mental Health Emergency Department Visits: 24 Hours and Counting, Characteristics Associated With Prolonged Length of Stay

Abstract

Mental health visits, including substance abuse and psychiatric illness, continue to be among the most common and expensive presentations to emergency departments (ED). Acutely ill ED patients await psychiatric evaluation and disposition for times ranging from hours to more than 10 days. Our objective is to examine the relationship of psychiatric diagnostic categories with ED disposition and length of stay. We performed a retrospective analysis of all mental health ED visits using the National Hospital Ambulatory Medical Care Survey (NHAMCS) from 2001-2011. Patients with any one of the three recorded International Classification of Disease; Ninth Revision (ICD-9) diagnostic codes indicative of substance abuse or primary psychiatric diagnosis were included. Dual diagnosis was defined as ICD-9 codes with both substance abuse and psychiatric diagnosis. Psychosis, self-harm, and bipolar disorder were categorized as severe disease according to convention. Disposition was categorized as: discharge, admission and transfer. Length of stay was defined as time of ED arrival to disposition (min) and categorized into time periods of >6, >12, >24 hours. We examined demographic, and visit-level factors associated with increased ED length of stay using chi-square tests for categorical variables, and f-tests of means for continuous variables. Of the more than 65 million ED mental health visits, substance abuse (41%), depression (23%) and anxiety (26%) comprised the majority of diagnoses. Dual diagnosis and multiple psychiatric diagnoses accounted for 9 and 10% of patients. While patients with anxiety were infrequently admitted (8.3%) or transferred (3.7%), the majority of patients with severe mental illness, required either admission or transfer. Of patients with psychosis, 34% were admitted and 24%, transferred; of patients with self-harm, 33% were admitted and 29% transferred; of patients with bipolar disorder, 29% were admitted and 40% transferred. Regardless of disposition, patients with bipolar disorder, psychosis, dual diagnosis, multiple psychiatric diagnoses and depression had increased odds of >24 hour length of stay respectively (OR 3.7, (95% CI 1.5-9.4); 2.9, (2.1-3.8); OR 2.4, (1.7-3.2); OR 1.8, (1.2-2.8) respectively. When length of stay was analyzed by disposition status, we found that even for patients discharged with these diagnoses, similarly increased lengths of stay> 24 hour were observed (data not shown.) We found that ED mental health patients with more severe disorders had increased rates of admission and transfer and longer lengths of stay overall. Our finding that the most seriously ill patients had increased odds of ED lengths of stay > 24 hours, even when ultimately discharged, demonstrates unacceptable conditions, in which patients are held and do not receive essential psychiatric care. Further investigation of the systems affecting these patients including, placement of involuntary holds, availability of ED psychiatric consultants or outpatient resources, would delineate potential intervention points for the care of these vulnerable patients.