301 - NICU Progress Note Practice Survey to Inform Documentation Standardization and AI Integration

Publication Number: 1287.301

Kelsey A. Simek, University of Utah School of Medicine, Salt Lake City, UT, United States; Nicholas R. Carr, Intermountain Health, University of Utah, Sandy, UT, United States; Antonio J. Hernandez, The University of Texas Health Science Center at San Antonio Joe R. and Teresa Lozano Long School of Medicine, San Antonio, TX, United States; Ameena Husain, University of Utah School of Medicine, Salt Lake City, UT, United States; Lindsey A. Knake, University of Iowa Roy J. and Lucille A. Carver College of Medicine, Iowa City, IA, United States; Shama Patel, Nationwide Children's Hospital, Columbus, OH, United States

Background: Artificial intelligence (AI) is rapidly assuming administrative tasks including clinical progress note generation. AI is most effective when the task is standardized. Best practices for daily progress note generation in the neonatal intensive care unit (NICU) are not well defined. Standardization of NICU progress note generation is crucial to ensure clinically relevant AI outputs.

Objective: We aim to explore key domains of NICU progress notes: authorship, organization, and the perceived time burden of documentation.

Design/Methods: We conducted a multi-institutional survey of neonatology providers with dual informatics expertise selected for their unique qualifications and interest in this topic. Participants were identified by study authors based on the participant’s activity in national level neonatal informatics settings. The Nationwide Children’s Hospital institutional review board reviewed and exempted the study. Study results were analyzed with descriptive statistics.

Results: Thirty-four participants were invited from 30 institutions and 29 (85%) responded to the survey (Table 1). Select results are detailed in Table 2. Many respondents reported shared progress note authorship and the majority identified advanced practice providers as the primary author. However, most respondents believed the primary author should be the attending neonatologist. Standard templates and copy-forward functionality were used by most, though there was a split between systems-based and problem-based assessment and plan organization. Many felt they spend too much time on documentation and that documentation significantly impacts time spent doing other important tasks. However, most respondents reported rarely or never writing notes after ending their shift.

Conclusion(s): Our survey study of an enriched sample of neonatology provider-informaticists reveals variation in documentation authorship and organization, and significant perceived time burden despite usual time-savers like templated notes and copy forward functionality. EHR vendor AI tools may provide an avenue for documentation standardization and facilitation of acceptable AI generated progress notes which could address the significant time burden imposed by documentation. Future research should incorporate a broad sample representative of the entire neonatology provider community. We speculate further progress in this area will prepare our field to effectively leverage AI-generated documentation and enhance clinician well-being in neonatal settings.

Table 1: Respondant Characteristics
Table 1.pdf

Table 2: Select Survey Results
Table 2.pdf