Use of Quebec medico-administrative databases for mental health studies: Opportunities, methodological challenges and limitations – The case of depression among diabetic subjects

The use of medico-administrative databases for studies on mental health issues is very common because of the large number of people in these databases and the possibility to carry out long-term studies. Several challenges, such as identifying people with a disease of interest or being exposed to a risk factor, have to be overcome through validation studies to ensure an optimal use of these resources. Moreover, limits (lack of certain relevant information) and the coverage of about 40% of Quebec’s population by the public drug plan are to be considered in the interpretation and generalization of research results based on these data sources. In the specific case of depression, we used these databases to determine the incidence of depression among diabetic individuals in Quebec. This required the use of a previously validated algorithm (validated in another province) that we modified to define and identify the cases of depression in the RAMQ databases. We observed an incidence of depression of 9.47 persons years over a follow-up of 8 years. Finally, we assessed the impact of depression on adherence and persistence with antidiabetic drugs as well as the factors that affect patients’ use of these drugs. Our results suggest that depression has a negative impact on the use of antidiabetic drugs and allow the identification of possible solutions.