Abstract Rationale Intensive care unit (ICU) patients face a high risk of morbidity and mortality, often requiring patients and their families to navigate complex decisions regarding treatment options and goals of care. Effective end-of-life (EOL) communication skills and showing empathy during difficult conversations are essential for delivering patient- and family-centered care around EOL. In this study, we implemented a dedicated EOL communication skills training program and evaluated its impact on participants’ perceived skills and attitudes toward EOL conversations. Methods This prospective, qualitative educational intervention study was conducted from March to October 2025 at Norwalk Hospital, a community-based teaching hospital in Norwalk, CT. Internal medicine residents and medical students completing ICU rotations were enrolled. The intervention was a single 1-hour EOL communication training session adapted from a curriculum developed by the Beth Israel Deaconess Medical Center Shapiro Institute for Education and Research. The session was facilitated by pulmonary/critical care fellows and emphasized four core competencies: agenda-setting, the ask-tell-ask framework, clear information delivery, and recognition and response to emotion. Participants completed pre- and post-session surveys evaluating communication skills, perceived comfort, and attitudes toward EOL conversations using a 5-point Likert scale. Survey administration time points illustrated in Figure 1. Statistical significance was set at α = 0.01 (two tailed). Effect sizes were calculated as mean change divided by retrospective pre-intervention standard deviation with ≥0.8 considered a large effect. Results Eighteen participants completed the EOL communication training session and surveys. The cohort included 3 fourth-year medical students and 15 internal medicine residents (PGY-1 n = 5, PGY-2 n = 7, PGY-3 n = 3). Pre-intervention, most participants rated their communication skills and comfort with EOL discussions as poor or fair. Post-intervention responses demonstrated significant improvements in perceived confidence across all domains. Overall confidence in skills increased from 17% pre-intervention to 62% post-intervention with the greatest relative improvement observed among medical students and first-year residents. The domains seeing highest impact were knowledge of language to avoid during difficult conversations (effect size 1.89; p 0.01) and confidence in delivering bad or distressing news to families (effect size 1.27; p 0.01). Conclusion Findings suggest that even brief, structured communication training can meaningfully enhance trainee preparedness for and confidence in EOL communication in the ICU. Given the importance of EOL discussions in critical care, broader implementation and longitudinal reinforcement of core competencies may further enhance confidence, reduce decisional conflict, and support higher-quality patient-centered care. Our future research will examine durability of initial intervention and role for repeat training. This abstract is funded by: None
Margolskee et al. (2026) studied this question.