• depression;
  • antidepressive agents;
  • fluoxetine;
  • lithium carbonate;
  • desipramine

Objective:  Patients with major depressive disorder often show only partial or no response to antidepressants, necessitating next-step interventions such as dose increase or augmentation. Factors moderating response to these next-step interventions are not well-studied.

Method:  In this randomized, double-blind investigation of next-step treatments in 101 outpatients who failed to respond to fluoxetine 20 mg for 8 weeks, the impact of depressive course and sociodemographic factors on likelihood of treatment response following dose increase or lithium or desipramine augmentation was examined.

Results:  After controlling for depression severity at baseline, current marriage and earlier onset of depression were associated with greater likelihood of response in a logistic regression. Intervention strategy was not predictive of response.

Conclusion:  Marital status and earlier onset of depression may be clinically useful in predicting outcome following any next-step intervention for treatment resistance, rather than with particular strategies.