Predictors of Alternate Level of Care within Mental Health Inpatient Units across Ontario
MetadataShow full item record
Purpose: The main purpose of this study was to identify clinical predictors of alternate level of care (ALC) status within inpatient mental health units across Ontario. In addition, the distribution of ALC days among designated patients, the proportion of patient episodes designated ALC over time, and the demographic and clinical characteristics of mental health inpatients are also described. Methods: This project was conducted via a secondary data analysis of data obtained from the Resident Assessment Instrument – Mental Health (RAI-MH). RAI-MH data was provided by the Ontario Mental Health Reporting System (OMHRS) for the years 2005 to 2014, including information on ALC days. Furthermore, an additional ALC variable was obtained from the Wait Time Information System (WTIS) for the years 2011 to 2013. All analyses were performed using admission episodes only. For the descriptive statistics, rates of ALC designations and odds ratios (OR) were examined across several characteristics, providing a basis for variable selection in the multivariate modeling stage. To determine the clinical predictors of ALC status, a multivariate logistic regression was performed in OMHRS using a 30+ ALC day cut-point. Once the model was ascertained in OMHRS, it was tested again in WTIS and adjusted accordingly, providing a final model for 30+ ALC days. Results: The proportion of ALC episodes designated as ALC over time was low (~3% for 1+ ALC days and ~1% for 30+ ALC days), but the distribution of ALC days across patients was comparatively large (Median = ~20 days). The final logistic regression model for 30+ ALC days identified several predictors of ALC status, including IADL, ADL and cognitive impairment, social isolation, aggression, older age, male sex, speaking a primary foreign language, greater lifetime admissions to a psychiatric hospital, symptoms of depression, positive psychosis and social withdrawal, disorders of childhood/adolescence, intellectual disabilities, and substance use. Conclusions: There are a multitude of risk factors implicated in ALC designations among mental health inpatients in Ontario, indicating that their needs are varied and complex. Policy makers and mental health care practitioners may utilize early predictors of ALC status to introduce treatment interventions and policies that reduce the risk of ALC designations in mental health settings.