(1491) Quantifying Differences in PICU Documentation Practices: A Multi-Institution Study

Introduction: Effective communication through the electronic health record (EHR) is crucial for ensuring patient safety. However, redundant information, copying and pasting, and the use of “smart text” to automate note writing have contributed to increasing note length and provider frustration. In this multisite study we hypothesize that there are both site and personnel type differences in PICU note documentation metrics.

Methods: This retrospective observational study was conducted at two academic PICUs (CHOA and URMC). The cohort includes all patients who were admitted to the PICUs from 1/2016 to 6/2023. We calculated documentation and “smart text” usage metrics [e.g., note counts, length, character source, “smart text” usage count] stratified by site and personnel type. We descriptive the metric results and compare between institutions with Wilcoxon rank sum and Chi-square tests.

Results: Our cohort included 84,575 PICU encounters resulting in 2.65m notes (CHOA: 2.0m, URMC: 658k). At both sites, physicians wrote the most notes (51% vs 47%). Total note lengths were significantly longer at CHOA [CHOA: 2929 (IQR 1049, 7033) vs URMC: 1434 (IQR 621, 4343), p< 0.001]. In subset analyses, physicians' notes were significantly longer than those of other personnel at both sites. Median note length increased at both sites from the first to last year of study (CHOA: 13% URMC: 17%). At CHOA, a significantly fewer characters were manually inserted than at URMC (25% vs 33%, p<.001) in addition to significantly more templated characters (47% vs 28%, p<.001). Overall physicians had a lower percentage of manually entered characters compared to the other personnel types (20% vs 45%, p<.001). Across all author types, median number of “smart text” elements differed by site [CHOA: 18 [IQR: 13, 26], URMC: 7 [IQR: 2, 16), p< 0.001]. Physicians used significantly more “smart text” elements per note compared to other personnel (16 vs 6, p<.001).

Conclusions: We demonstrate significant differences in note documentation practices among personnel types across two academic PICUs, suggesting both personnel type and site-specific practices influence documentation habits. Understanding these differences and improving documentation practices can help mitigate note bloat, enhance the quality of clinical notes, and improve patient safety.