Mental Health Services in Canada: Building a Model of Mental Health Care Utilization
Lewis, John Dufton
MetadataShow full item record
Existing research literature shows that mental health care services are under-utilized among individuals who have mental health problems. In Canada, it is estimated that only 40% of individuals who have mental health problems are provided with mental health care. Although some past research have examined predictors of mental health care utilization, there are gaps in our knowledge of how these predictors interact with one another and how these predictors specifically affect mental health care utilization in Canada. In this study, data from the 2007/2008 Canadian Community Health Survey (CCHS) were used with the goal of developing a more accurate picture of trends in mental health issues and care in Canada (N = 131,061, weighted N = 28,030,943). Guided by Andersen’s Behavioural Model of Health Care Use (2008), associations were examined between mental health care utilization (i.e., consultation with a psychologist, accessing mental health care services, and receiving care from a mental health specialist) and contextual factors (e.g., province of residence, health region), predisposing individual characteristics (e.g., gender, age, minority status), individual enabling factors (e.g., employment, income), individual need (i.e., stress, mental health well-being), general health behaviours (e.g., number of consults with health professional), and outcomes factors (e.g., satisfaction, difficulties getting services) were explored. Associations were observed between mental health care utilization and a variety of variables across most of the categories proposed by Andersen (2008). When tested as a model, the group of variables related to need generally showed the strongest influence on utilization. However, when examining individual variables contributing to the model, four predictors (perceived mental health, physician visits, income, and age) generally made the largest contributions to the models. These predictors represented factors proposed in Andersen’s model. The models were statistically significant, however, both models were limited in the amount of variation explained for consultation with a psychologist and receiving mental health care (Nagelkerke R2 = .145 and Nagelkerke R2 = .271 respectively). Clinical and theoretical implications and future research directions are discussed.