The aim of this study was to investigate the effectiveness of long-term psychoanalytic and psychodynamic psychotherapies. In a prospective, randomized outcome study, psychoanalytic (mean duration: 39 months, mean dose: 234 sessions) and psychodynamic (mean duration: 34 months, mean dose: 88 sessions) therapy were compared at post-treatment and at one-, two-, and three-year follow-up in the treatment of patients with a primary diagnosis of unipolar depression. All treatments were carried out by experienced psychotherapists. Primary outcome measures were the Beck Depression Inventory and the Scales of Psychological Capacities, and secondary outcome measures were the Global Severity Index of the Symptom Checklist 90-R, the Inventory of Interpersonal Problems, the Social Support Questionnaire, and the INTREX Introject Questionnaire. Interviewers at pre- and post-treatment and at one-year follow-up were blinded; at two- and three-year follow-up, all self-report instruments were mailed to the patients. Analyses of covariance, effect sizes, and clinical significances were calculated to contrast the groups. We found significant outcome differences between treatments in terms of depressive and global psychiatric symptoms, personality functioning, and social relations at three-year follow-up, with psychoanalytic therapy being more effective. No outcome differences were found in terms of interpersonal problems. We concluded that psychoanalytic therapy associated with its higher treatment dose shows longer-lasting effects.