MENTAL HEALTH INPATIENT USE OVER TIME: IDENTIFYING CHARACTERISTICS ASSOCIATED WITH INCREASED USE FOLLOWING INDEX ADMISSION

Abstract

Purpose: The purpose of this study was to describe patterns of inpatient mental health service following index admission and to identify individual and socio-environmental factors associated with high use following index admission.

Methods: Secondary data analysis of the Ontario Mental Health Reporting System (OMHRS) and the Ontario Marginalization Index (On-Marg) was performed. OMHRS—which is derived from the Resident Assessment Instrument - Mental Health (RAI-MH), and contains all inpatient mental health service use for the province of Ontario—was linked with the On-Marg—geographic socio-environmental data drawn from the Canadian Census—via a common geographic descriptor: the forward sortation area (FSA).

A retrospective cohort containing data from 2006-2014 with an intake period from 2006-2009 and a 5-year follow-up period was established. Two outcome variables were examined: high intensity episodic use, and high intensity days in hospital. High-intensity use was defined as use in the 90th percentile for either variable following index admission. Descriptive, bivariate, and multivariate statistics were used to identify variables to be included in the parsimonious iterative modelling process. Multivariate logistic regression using generalized estimating equations (GEE) was used to determine factors associated with high intensity use following index admission.

Results: Among those with an index admission to inpatient mental health services, 70% did not have further use over the 5-year follow-up. High intensity use (48+ days, 3+ episodes) following index fell into 3 categories: high use days only (2.8%), high use episodes only (5%), high use of both episodes and days (7.7%). The modelling process identified several variables associated with high intensity use. Schizophrenia and its related symptomology were shown to increase the odds of being high intensity users, while dementia, substance use, and adjustment disorders were shown to decrease odds. Two interactions were shown to predict the high intensity days outcome: marital status of never having been married with positive symptoms scores greater than 6 on the positive symptoms scale, and being male with cognitive impairment scores greater than 4 on the cognitive performance scale. Finally, middle quintile area dependency scores were shown to predict high intensity episodic use.

Conclusions: Schizophrenia and psychotic symptoms represents the primary driver of high intensity inpatient use (both days and episodes) following index admission, though the observed interactions may suggest that issues with social support may be driving higher lengths of stay in inpatient settings following the index admission. Socio-environmental factors appear to play a smaller role, after adjusting for individual risk factors, in high intensity inpatient use following index admission.