ABSTRACT: Communication about end-of-life care is a core clinical skill. Simulation-based training improves skill acquisition, but effects on patient-reported outcomes are unknown.To assess the effects of a communication skills intervention for internal medicine and nurse practitioner trainees on patient- and family-reported outcomes.Randomized trial conducted with 391 internal medicine and 81 nurse practitioner trainees between 2007 and 2013 at the University of Washington and Medical University of South Carolina.Participants were randomized to an 8-session, simulation-based, communication skills intervention (N?=?232) or usual education (N?=?240).Primary outcome was patient-reported quality of communication (QOC; mean rating of 17 items rated from 0-10, with 0?=?poor and 10?=?perfect). Secondary outcomes were patient-reported quality of end-of-life care (QEOLC; mean rating of 26 items rated from 0-10) and depressive symptoms (assessed using the 8-item Personal Health Questionnaire [PHQ-8]; range, 0-24, higher scores worse) and family-reported QOC and QEOLC. Analyses were clustered by trainee.There were 1866 patient ratings (44% response) and 936 family ratings (68% response). The intervention was not associated with significant changes in QOC or QEOLC. Mean values for postintervention patient QOC and QEOLC were 6.5 (95% CI, 6.2 to 6.8) and 8.3 (95% CI, 8.1 to 8.5) respectively, compared with 6.3 (95% CI, 6.2 to 6.5) and 8.3 (95% CI, 8.1 to 8.4) for control conditions. After adjustment, comparing intervention with control, there was no significant difference in the QOC score for patients (difference, 0.4 points [95% CI, -0.1 to 0.9]; P?=?.15) or families (difference, 0.1 [95% CI, -0.8 to 1.0]; P?=?.81). There was no significant difference in QEOLC score for patients (difference, 0.3 points [95% CI, -0.3 to 0.8]; P?=?.34) or families (difference, 0.1 [95% CI, -0.7 to 0.8]; P?=?.88). The intervention was associated with significantly increased depression scores among patients of postintervention trainees (mean score, 10.0 [95% CI, 9.1 to 10.8], compared with 8.8 [95% CI, 8.4 to 9.2]) for control conditions; adjusted model showed an intervention effect of 2.2 (95% CI, 0.6 to 3.8; P?=?.006).Among internal medicine and nurse practitioner trainees, simulation-based communication training compared with usual education did not improve quality of communication about end-of-life care or quality of end-of-life care but was associated with a small increase in patients' depressive symptoms. These findings raise questions about skills transfer from simulation training to actual patient care and the adequacy of communication skills assessment.clinicaltrials.gov Identifier: NCT00687349.