Abstract
Patient decompensation necessitating rapid response team (RRT) care in the hospital setting involves complex medical decision making, strong leadership skills, and precise communication where every second matters. However, RRT outcomes can vary based on leader training, knowledge, and experience. We designed five digital, condition-specific, guided algorithms to improve RRT care and compared user survey data among three physician cohorts across the clinical training spectrum to assess the practicality of real-world usage in a small feasibility study.Guided algorithms to common RRT scenarios, including tachycardia, bradycardia, hypotension, hypoxia, and altered mental status, were used by 157 physicians at our institution across three Internal Medicine user cohorts (1: end-of-year PGY-2-5 residents, 2: new PGY-2 residents, and 3: attending hospitalists) from April to December 2024. Survey data from 28 respondents were compared across cohorts using Kruskal-Wallis and Dunn statistical analyses.Survey responses demonstrated consistently high scores across cohorts regarding improvement in patient care, improved RRT leader experience, improved confidence, reduced stress/cognitive load, potential for standardization of care, and likelihood of recommendation to a colleague. Interestingly, new PGY-2 residents rated ease of navigation at 7/10 compared to 10/10 by attending hospitalists (p = 0.016).Digital, guided RRT algorithms are a practical and effective tool for enhancing physician care delivery during inpatient rapid response events across all levels of training. High survey scores across cohorts warrant consideration for broader implementation. Variation in ease of navigation scores highlights the importance of tailoring information flow and usability features to less experienced users. Overall, these algorithms show promise as valuable adjuncts during acute care delivery in high-stakes clinical settings.