Epidemiological characterization of drug overconsumption: the example of antidepressants.

The authors present a pharmaco-epidemiological method for the characterization of drug overconsumption. Using the database of a French regional health insurance office, a comparison was made between patients receiving doses higher than recommended and all patients receiving normal doses of two antidepressant drugs (tianeptine and milnacipran). Logistic regression models were used to identify the variables related to overconsumption. The selected variables were different for each antidepressant and provided a good characterization of overconsumption features: insufficient effectiveness and severity of the condition for milnacipran and abuse/misuse for tianeptine. Characterization of overconsumption of antidepressants can be performed using logistic regression models. This method of analysis, which is seldom used in the context of pharmacodependence, makes it possible to explain posology increase.