Background: Insufficient response to monotreatment for depression is a common phenomenon in clinical practice. Even so, evidence indicating how to proceed in such cases is sparse. Methods: This study looks at the second phase of a sequential treatment algorithm, in which 103 outpatients with moderately severe depression were initially randomized to either short-term supportive psychodynamic therapy (PDT) or antidepressants. Patients who reported less than 30% symptom improvement after 8 weeks were offered combined treatment. Outcome measures were the Hamilton Depression Rating Scale (HAM-D), the Clinical Global Impression of Severity and Improvement, the SCL-90 depression subscale and the EuroQOL questionnaire. Results: Despite being nonresponsive, about 40% of patients preferred to continue with monotherapy. At treatment termination, patients initially randomized to PDT had improved more than those initially receiving antidepressants, as indicated by the HAM-D and the EuroQOL, independently of whether the addition was accepted or not. Conclusions: Starting with psychotherapy may be preferable in mildly and moderately depressed outpatients. For patients who receive either PDT or antidepressants, combined therapy after early nonresponse seems to be helpful. Nevertheless, this sequential strategy is not always preferred by patients.