We studied the relation between individual and neighborhood socioeconomic characteristics and the probability of:

• – new long-duration antidepressant treatment;

• – early antidepressant discontinuation.

We followed two cohorts of inhabitants of Marseille (aged 18–64 years) covered by the National Health Insurance Fund (NHIF) for 2.5 years. In the first cohort (316,412 individuals in 2008), we studied new long-duration antidepressant treatments (≥ 4 antidepressants prescription claims within 6 months after the index claim, and none in the 6 months before). The second cohort was restricted to the 14,518 individuals with a new episode of antidepressant treatment prescribed by a private GP in 2008–2009 to study early treatment discontinuation: < 4 antidepressant prescription claims in the 6 months following the index claim. We developed a deprivation index at the neighborhood level (census block) and used multivariate multilevel logistic models adjusted for consultations with GPs and psychiatrists. In the second cohort, analyses were further adjusted on GPs characteristics.

First cohort: the probability of new long-duration antidepressant treatments was negatively associated with both individual low income and neighborhood deprivation. Second cohort: low income, and prescribers’ clientele composition (high proportion of disadvantaged patients) were independently associated with an increased risk of early discontinuation. A significant interaction was found between low income and gender.

Our results add further evidence supporting the existence of inequalities in antidepressant treatment at both the individual, GP and neighborhood levels, and that these inequalities occur principally during the processes of care. Inequalities in antidepressant continuation are more pronounced among women. Further research is warranted to improve our understanding of their mechanisms.