OBJECTIVE: To determine the variables associated with the prescription of long-term antidepressants (more than three years) to patients in a primary health area. DESIGN: Descriptive observational study. SITE: Basic health area of primary care. PARTICIPANTS: A sample of 315 participants assigned to a basic health area, who have been prescribed at least one antidepressant during a calendar year. INTERVENTIONS: Institutional information sources. MAIN MEASUREMENTS: Those variables that can be associated with a prolonged prescription of antidepressants. The selection of the patients was carried out by simple random sampling. Analysis using Chi-Square contrast for the comparison of proportions. Multivariate binary logistic regression to observe possible associations in the set of variables. RESULTS: The study analyzed a majority of women (75.2%) with a mean age of 61.7 years (SD=38.18). SSRIs (Selective Serotonin Reuptake Inhibitors) were the most commonly used antidepressants (50.5%), especially sertraline. Dysthymia was the main diagnosis (17%). Prolonged duration of treatment was associated with SSRIs and dysthymia. Logistic regression showed relevant factors: dysthymia, age, number of antidepressants, and SSRI use. CONCLUSION: There is a high prevalence of prescription beyond three years. It is observed how dysthymic disorder is associated with a prolonged duration of treatment, as well as age, SSRI use, antidepressant change, and the total number of antidepressants prescribed.