The Effects of Leadership Curricula With and Without Implicit Bias Training on Graduate Medical Education: A Multicenter Randomized Trial.

PURPOSE: To determine whether a brief leadership curriculum including high-fidelity simulation can improve leadership skills among resident physicians. METHOD: This was a double-blind, randomized controlled trial among obstetrics-gynecology and emergency medicine (EM) residents across 5 academic medical centers from different geographic areas of the United States, 2015-2017. Participants were assigned to 1 of 3 study arms: the Leadership Education Advanced During Simulation (LEADS) curriculum, a shortened Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS) curriculum, or as active controls (no leadership curriculum). Active controls were recruited from a separate site and not randomized to limit any unintentional introduction of materials from leadership curricula. The LEADS curriculum was developed in partnership with the Council on Resident Education in Obstetrics and Gynecology and Council of Residency Directors in Emergency Medicine as a novel way to provide a leadership toolkit. Both LEADS and the abbreviated TeamSTEPPS were designed as six 10-minute interactive web-based modules.The primary outcome of interest was the leadership performance score from the validated Clinical Teamwork Scale instrument measured during standardized high-fidelity simulation scenarios. Secondary outcomes were 9 key components of leadership from the detailed leadership evaluation measured on 5-point Likert scales. Both outcomes were rated by a blinded clinical video reviewer. RESULTS: One hundred ten obstetrics-gynecology and EM residents participated in this 2-year trial. Participants in both LEADS and TeamSTEPPS had statistically significant improvement in leadership scores from "average" to "good" ranges both immediately and at the 6-month follow-up, while controls remained unchanged in the "average" category throughout the study. There were no differences between LEADS and TeamSTEPPS curricula with respect to the primary outcome. CONCLUSIONS: Residents who participated in a brief structured leadership training intervention had improved leadership skills that were maintained at 6-month follow-up.

Impact of levels of simulation fidelity on training of interns in ACLS.

AIM: Identifying the modality and fidelity of simulation that offers the greatest benefit to the learner is critical to Advanced Cardiac Life Support (ACLS) training. Our hypothesis is that participants who receive ACLS training on high-fidelity mannequins will perform better than those trained on low-fidelity mannequins. METHODS: The study was performed in the context of an ACLS Initial Provider course for new postgraduate year 1 residents and involved 3 training arms: (1) low-fidelity, (2) mid-fidelity, and (3) high-fidelity. Educational outcomes were evaluated by written scores, student evaluations of the course, and expert rater reviews of megacode performance. RESULTS: A convenience sample of 54 subjects was randomized to 1 of the 3 training arms. All 3 groups significantly improved based on written posttest scores (P < 0.0001); however, pretest to posttest improvement among the 3 training arms was not significantly different: low-fidelity = 42.3 (95% CI, 35.7-48.9); mid-fidelity = 41.3 (95% CI, 34.7-47.9); high-fidelity = 40.8 (95% CI, 34.3-47.5; P = 0.95). All participants felt the simulator environment was realistic regardless of level of fidelity. Participants in the high-fidelity group were less likely to feel comfortable in the simulator environment (P = 0.0045). Clinical performance as assessed by expert raters' megacode scores was better for high-fidelity (66.3) than mid-fidelity (60.1) (P = 0.04). CONCLUSION: Overall, there was no difference among the 3 groups in test scores or perceived instructor or course quality; however, subjects trained on high-fidelity mannequins performed better than those trained on mid-fidelity with respect to megacode performance.