Sixty-five young adults with remitted major depressive disorder (MDD) were followed for 18 months. Recurrence of MDD was reported by 41.5% of the initial sample and 49.1% of those who completed the study (n = 53). Survival analyses were used to identify predictors of recurrence so that individuals at greatest risk could be targeted for intervention. Potential predictors included measures of comorbid psychopathology (Axis II pathology, and current and lifetime nonmood Axis I diagnoses), depression-specific clinical features (number of episodes, past treatment, and suicidality), and self-reported cognitive and interpersonal constructs (hope, dysfunctional attitudes, and interpersonal problems). Only personality pathology (specifically, the total dimensional and Cluster B dimensional scores on the International Personality Disorder Examination) significantly predicted hazard of recurrence.