Background: Minor and mild-major depression are highly prevalent in primary care. There is insufficient evidence for the effectiveness of antidepressants in the treatment of minor and mild-major depression. We compared the effectiveness of usual primary care treatment, with or without antidepressants, in minor and mild-major depression. Methods: Apragmatic patient-randomized equivalence trial with 52 weeks follow-up was conducted in The Netherlands. In total, 59 primary care physicians (PCPs) recruited and treated 181 adult patients with minor or mild-major depression. Patients were randomized to four consultations within 3 months of usual care plus antidepressants (UCandAD) or usual care alone (UCnoAD). The Montgomery Åsberg Depression Rating Scale (MADRS) was used to assess changes in severity of depressive symptoms. The predefined equivalence margin was set at five points. Multilevel analysis was used to analyze the data. Secondary outcome measures were the Short-Form 36 (SF-36), and the Client Satisfaction Questionnaire (CSQ-8). Results: Patients received on average 3.0 (SD 1.4) 15-min consultations within 3 months with (n = 85) or without paroxetine (n = 96). Equivalence of UCandAD and UCnoAD was demonstrated in the intention-to-treat analyses as well as the per-protocol analysis after 6 weeks, but not at 13, 26 and 52 weeks follow-up. No statistical differences in effectiveness between treatment groups were found in the intention-to-treat analysis. No differences in the physical and mental functioning (SF-36) were found between the treatment groups. Patients allocated to UCandAD were slightly more satisfied with their treatment at 13 weeks follow-up (but not at 52 weeks follow-up) than patients allocated to UCnoAD. Preliminary analyses suggested that subgroups such as patients with mild-major (instead of a minor) depression might benefit from antidepressant treatment. Patients who were assigned to their preferred treatment (in particular to UCnoAD) were more often compliant and had better clinical outcomes. Conclusion: UCandAD was as effective as UCnoAD over the first 6 weeks, but not at 13, 26, and 52 weeks. However, superiority of either treatment could not be demonstrated either. The question whether antidepressants add any clinical effect to usual care remains unresolved. We recommend future studies to look for subgroups of patients who may benefit from antidepressants. © 2007 Hermens et al; licensee BioMed Central Ltd.