Objective: To examine the durability of improvement in child and maternal symptoms 6 months after termination of child-parent psychotherapy (CPP). Method: Seventy-five multiethnic preschool-age child-mother dyads from diverse socioeconomic backgrounds were randomly assigned to (1) CPP or (2) case management plus community referral for individual treatment. Children were 3 to 5 years old. Follow-up assessments were conducted 6 months after the end of a 1-year treatment period. Mothers completed the Child Behavior Checklist and the Symptom Checklist Revised to assess child's and mother's symptoms. Results: For treatment completers, general linear model (GLM) repeated-measures analyses support the durability of CPP with significant group x time interactions for children's total behavior problems and mothers' general distress. Intent-to-treat analyses revealed similar findings for children's behavior problems but were not significant for maternal symptoms. Conclusions: The findings provide additional evidence of the efficacy and durability of CPP with this population and highlight the importance of a relationship focus in the treatment of traumatized preschoolers. (Contains 1 table and 1 figure.)