Objective: Few studies have investigated the maintenance of skills acquired in classroom-based clinician education. Using an advanced simulation-based clinical communication skill training program for postgraduate psychiatry education (ComPsych), we aimed to investigate skill acquisition through assessing changes in competence (abilities) and performance (practice). Methods: Forty trainee psychiatrists (15 males; age range 26-48) participated. Video-recorded Standardized Patient Assessments (SPAs) were delivered twice pre- and post-training to assess learning. Skills were coded by independent psychologists using the Core Communication Skills (CCS) coding system. Simulated patients (SPs) rated trainees' communication performance using the Sim-Patient SPA checklist. Paired t-tests, linear mixed models and logistic mixed models assessed changes in communication skills over time. Results: For SPAs, reliability of coder ratings was deemed acceptable (ICC range 0.67 to 0.87). Mean post-training communication performance significantly increased for skills in agenda setting (p < 0.001), information organization (p < 0.001), empathic skills (p = 0.046), and overall skills performance (p = 0.001). Significant decreases for questioning skills were indicative of reduced reliance on these skills post-training. SPs rated all skillsets higher post-training. A modest relationship was detected between frequency (coded) and (SP-rated) quality of communication skills. Improvements in agenda setting and information organisation skills were retained ~ 6 weeks post-training. Conclusions: Training improved patient-centered communication skills in psychiatry trainees, particularly skills in agenda setting and information organization, with skills retained ~ 6 weeks post-training. There was reduced reliance on questioning skills, which are well utilised generally. The study supports the benefits of this method of communication skills training into postgraduate psychiatry education.