Cannabis use among individuals in inpatient psychiatry: Trends in use, factors associated with use, and risk of 30-day readmission

Abstract

Purpose: The Canadian Government has approved Bill C-45 to legalize and regulate the production, sale, and use of non-medical cannabis. Certain subpopulations, including those with psychiatric disorders, need to be uniquely considered during policy development and continually monitored as they may be particularly vulnerable to negative outcomes associated with cannabis use. This research was conducted to evaluate current prevalence and patterns of past 30-day cannabis use among individuals admitted to inpatient psychiatry in Ontario.

Methods: This study evaluated patterns of past 30-day cannabis use among individuals admitted to inpatient psychiatry in Ontario, Canada from 2006 to 2016 (n=160322). An individual’s first admission during the study period (index admission) was used for the analysis. Prevalence rates were established, and trends were stratified by age, sex, and psychiatric diagnosis. Variables significantly associated with past 30-day use at the bivariate level were identified. Block modelling was carried out with significant variables to construct a logistic regression model that describes characteristics associated with past 30-day use. This process was repeated for 30-day readmissions. Cannabis was added to the final model to determine whether it remained significantly associated with readmissions after controlling for other variables. Factors associated with readmissions among past 30-day cannabis users and non-users were identified.

Results: Past 30-day cannabis use increased from 15.4% to 25.3% from 2006 to 2016. Although more males reported past 30-day cannabis use, non-males had a greater increase in use across the study period. Persons ages 18 to 24 had the highest rates of cannabis use. However, older age categories also had increases in rates of use over time. Younger individuals with substance-related and addictive disorders, males with schizophrenia and other psychotic disorders, and those with mood disorders all had greater odds of reporting past 30-day cannabis use, while a neurocognitive diagnosis was associated with reduced odds. Greater use was found among those with shorter lengths of stay, a history of violence, experiences of traumatic life events and financial hardship, poorer medication adherence, fewer lifetime psychiatric hospitalizations, and more severe symptoms and clinical measures (positive symptoms, social withdrawal, cognitive performance, mania symptoms, and indicators of addiction). After controlling for other variables, cannabis was significantly associated with readmissions for those exhibiting positive symptoms. Variables associated with readmissions for recent users were highly related to psychosis and may be associated with more complex and less compliant patients.

Conclusions: A steady increase in past 30-day cannabis use from 2006 to 2016 was found. Continual monitoring of psychiatric admissions following legalization is important to determine whether an increase in cannabis use is associated with increased admissions. Several characteristics describing past 30-day cannabis users in inpatient psychiatry mimic that of users in the general population. Additional variables associated with past 30-day cannabis use were identified for the inpatient population which can be used to follow this population. Past 30-day cannabis use was significantly associated with 30-day readmissions for those exhibiting positive symptoms. Appropriate education and care planning is crucial in order to improve clinical outcomes and reduce unnecessary readmissions.