Traditional Poster Round
Background: : Learning to function as a leader on an inter-professional code team is an important component of residency training (1). Baseline data demonstrated residents in our program often did not assume this role because they lacked self-confidence in their code leadership ability. We therefore developed a Resident Code Team Leadership Evaluation Tool (RCTLET) to compare resident self-perception to both facilitator assessment of their performance and a Target Level of Performance (TLP).
Research Question: : Our two primary research questions were: 1) can we develop a reliable tool to provide direct observational feedback to residents acting as code team leaders across multiple modalities (i.e. mock codes, various other simulated experiences, and live events, etc.), and 2) can we use that tool to compare facilitator evaluation of resident performance to resident self-assessment and TLP for graduating pediatric residents, pre-determined by local content experts.
Methodology: : Tool development: A literature search and local content expert input were used to identify eight elements pertinent to code leadership: role identification, communication, knowledge sharing, delegation, reevaluation, crowd control, assessment, and management (2,3,4). The RCTLET rated each element on a 5-point Likert scale, with descriptors outlining low, medium, and high performance. Thirteen experienced local content experts completed the RCTLET based on their expected TLP for graduating pediatric residents.
Tool implementation: Following mock codes, residents completed the RCTLET as a self-assessment, and two facilitators completed the RCTLET based on direct observation. We compared resident and facilitator assessments. Mean and median scores were compared with univariate and Kruskal-Wallis ANOVA, respectively. Pairwise comparisons were made with Mann-Whitney U-tests and independent t-tests. Effect size was calculated with Cohen’s d. Inter-item reliability was assessed with Cronbach alpha. Inter-rater reliability was determined by intraclass correlation coefficient.
Results: : We compared resident and facilitator assessments for 29 events. There was a statistically significant difference (d=0.8, p<.015) in the overall RCTLET average between facilitators (M(SD)=4.0(.6)) and residents (3.6(.5)). Significant differences (all p<.05) were reported for delegation, crowd control, assessment, and management, with facilitators scoring residents higher than resident self-assessment. The inter-item reliability for the eight RCTLET elements was α=0.84. The intraclass correlation between residents and facilitators for the RCTLET average was ICC (2,1)=0.52 (p<.001).
Discussion/Conclusions: : Facilitators rated resident code leadership performance higher than residents rated themselves. The RCTLET provided reliable data that can be used to provide formative feedback to residents aimed at improving their confidence and code team leadership skills. The RCTLET can be used to provide formative feedback to residents across a variety of simulated and live code scenarios to improve residence confidence in their code team leadership skills.