IMIA

Digital Identity as Patient Safety Infrastructure

Henry Bair

Health care has invested billions in interoperability, analytics, and generative artificial intelligence (AI). Yet a more basic layer remains fragile: knowing, at a level of assurance proportional to clinical risk, who is interacting with the health system and what they are authorized to do. Digital identity is often framed as an information technology security feature; in practice, however, it is a care delivery infrastructure.

https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-2880-9387

Effect of Speech Recognition Software on Provider Documentation Characteristics Within an Electronic Health Record System

McClane Howland, Zoe Lockhart, Sunit P. Jariwala

Objectives This study evaluates the effect of speech recognition software Dragon Medical One (DMO) on physician documentation burden and EHR workflows using Epic Signal usage metrics.

Methods We conducted a longitudinal cohort study of physicians at a single academic medical center between December 2019 and November 2023. We compared 291 DMO adopters with 2,828 non-adopters using a staggered-adoption difference-in-differences design with two-way fixed effects and standard errors clustered at the physician level. We analyzed 23 Epic Signal metrics spanning documentation methods, time metrics, and workflow efficiency. We applied Benjamini-Hochberg correction for multiple comparisons and validated parallel trends assumptions through event study analysis. Heterogeneous effects were assessed through interaction models with demeaned baseline physician characteristics.

Results Among 3,119 physicians (291 adopters, 2,828 non-adopters), DMO adoption shifted documentation composition from manual typing to voice recognition (+2.67 percentage points; q < 0.001) with corresponding decreases in manual (−1.98 percentage points; q = 0.002) and SmartTool-based documentation (−1.89 percentage points; q = 0.004). Time in notes per appointment decreased by 0.50 minutes (q = 0.004) and time outside scheduled hours decreased by 2.74 minutes per day (q = 0.009). Of 23 outcomes 6 survived false discovery rate correction at q < 0.05. Event study analyses confirmed parallel pre-treatment trends for 22 of 23 outcomes. Physicians with higher baseline documentation time experienced larger reductions in time in notes per day and time in system per day following DMO adoption (interaction q < 0.01).

Conclusion Adoption of speech recognition software was associated with a shift from manual to voice-based documentation, reduced time in notes per appointment, and reduced time outside scheduled hours. Effects were most pronounced among physicians with the highest baseline documentation burden, suggesting that those with the greatest room for improvement benefit most from this technology.
https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-2882-9924

Beyond the Model: Practical Insights from Monitoring Predictive Models across Diverse Clinical Workflows

John J. Hanna, Christopher R. Dennis, Andrew O. Johnson, Lauren N. Cooper, Christopher S. Evans, Christoph Ulrich Lehmann, Richard J. Medford

Objectives To support the artificial intelligence (AI) lifecycle in an integrated academic health system, we implemented a modular monitoring system to oversee electronic health record (EHR) vendor-provided clinical predictive models. This case study describes the lessons learned using this modular system to support sustainable oversight of the deployed predictive models across their lifecycle.

Methods We developed a modular monitoring system using automated data pipelines refreshed daily to support longitudinal oversight of four clinical predictive models developed by our EHR vendor (Epic Systems). An interactive monitoring application was designed to bridge the technical–operational gap for stakeholders involved in AI lifecycle decisions. Interactive reports include automated and assisted threshold-independent and threshold-dependent performance measures, flag rates, and problem-specific metrics crafted to support operational decision-making.

Results Through the lenses of people, process, and technology, we describe lessons learned from the real-world implementation of a monitoring system. Effective oversight required clearly defined dyadic ownership (technical and operational) and embedded monitoring activities within existing AI governance decision-making processes across all AI lifecycle stages. Monitoring extended beyond performance drift to include alert burden, workflow alignment, and problem-level signals. Modularity enabled rapid adaptation as models and workflows evolved. Silent alert simulations proved valuable for selecting preimplementation thresholds that aligned with clinical capacity and workflows but were limited in predicting postimplementation performance when the models faced real-world human–alert interactions.

Conclusion Implementing a modular, governance-aligned monitoring system enabled sustained oversight of vendor-provided clinical predictive models and shifted monitoring beyond traditional technical performance metrics toward user-centric and operationally meaningful measures. These findings highlight the importance of integrated monitoring infrastructure as a core component of responsible AI governance in clinical settings.

https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-2876-1059

Generative AI in Admission Notes and Diagnostic Completeness: A Pilot Study

Alfredo Camargo Rodrigues, Jason M. Misurac, Lindsey A. Knake, Kevin Barker, James M. Blum

Background Admission history and physical (H&P) notes influence inpatient documentation, risk adjustment, and reimbursement. In high-acuity settings, time constraints and fragmented chart data contribute to under-documentation of comorbidities. Generative AI systems that draft admission notes from existing electronic health record data may improve documentation completeness, but real-world inpatient performance remains insufficiently characterized.

Objective This study aimed to evaluate whether AI-generated admission notes identify documentation-relevant diagnoses supported by the medical record but not explicitly captured in provider-authored admission notes.

Methods In this single-center retrospective pilot study, we reviewed 22 matched pairs of AI-generated and provider-authored admission H&P notes at a large academic medical center. We assessed principal diagnosis concordance and identified net-new secondary diagnoses. Secondary diagnoses identified by the AI but absent from provider-authored notes were adjudicated by a clinical documentation improvement (CDI) team using standard institutional criteria.

Results AI-generated notes aligned with the provider-authored principal diagnosis in 91% of cases (20/22). The CDI team adjudicated 104 AI-identified secondary diagnoses, of which 97% (101/104) were supported for documentation. Ninety-four diagnoses were net-new, quality-relevant conditions not documented in provider-authored notes (median: 4.5 per admission). Net-new diagnoses were observed across all provider types, with numerically higher counts among advanced practice providers and residents; however, the sample was too small to support inferential comparisons. Despite its modest sample size, this pilot demonstrated high principal diagnosis concordance, though the AI misclassified the principal diagnosis in two cases. AI-generated drafts identified additional CDI-supported diagnoses not captured by providers, though narrative quality and factual accuracy were not evaluated.

Conclusion These findings highlight the potential of generative AI to surface documentation-relevant information at admission and underscore the importance of human oversight in AI-assisted documentation workflows. Larger multicenter studies are needed to assess generalizability and safety.
https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-2876-0998

Applying User-Centered Design to Develop a Prescriber Feedback Tool in Acute Outpatient Care Settings at the Veterans Health Administration

Shilo Anders, Carrie Reale, Thomas Reese, Russ Beebe, Robert Winter, Dax Westerman, Jesse O. Wrenn, Jin H. Han, Milner Staub, Melissa Rubenstein, Michael J. Ward, Michael E. Matheny

Objectives Acute care providers lack an easy way to assess their prescribing practices and track future-related care for their patients. Thus, we conducted design evaluations and subject matter expert (SME) design sessions in a user-centered design (UCD) approach to develop an audit and feedback tool that provides individualized, scalable prescribing feedback to clinical providers about their antibiotic and nonsteroidal anti-inflammatory drug (NSAID) prescriptions in unplanned care settings (e.g., emergency department and urgent primary care).

Methods A UCD approach was conducted with 11 individual interviews through two rounds of formative testing, focusing on interface design efficiency, effectiveness, and visualization interpretability. We conducted several design sessions with SMEs, prescribers in emergency and primary care medicine, where the design team asked the SMEs to comment and do a walk-through with various design prototypes for the tool, and then further iterate on new designs. Feedback about different user interface designs was obtained from future tool users in usability evaluation sessions where a provider interacted with the prototype through think-aloud, guided by a semistructured interview outline.

Results Through two rounds of usability evaluations, key usability issues were identified with the navigation, language, and interpretation of the data presented. This led to substantial interface design changes prior to implementation that improved usability and usefulness, as evidenced by a decrease in the number of usability issues found during the second round of evaluation. Participants appreciated the concept and usefulness of the tool presented; however, during usability sessions, they identified important optimizations, clarifications, and changes for improvement.

Conclusion Key generalizable findings include user preferences for nonjudgmental framing of prescribing, and a desire for intuitive presentation and summarization of recent care delivered to support actionable feedback. Required changes during UCD underscore the importance of this type of usability evaluation during tool ideation and development.
https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-2866-4361

Development and Application of a Nurse-Led Clinical Decision Support System for Safe Intravenous Medication Administration: A Nonrandomized Controlled Trial
Fuling Zhang, Wenjuan Yun, Lili Qin

Background Intravenous medication administration is a high-risk clinical procedure, where medication errors can lead to adverse consequences. Evidence-based clinical practice guidelines provide recommendations for the administration and monitoring of intravenous infusions. These guidelines are being increasingly integrated into clinical decision support systems (CDSS). The development of CDSS should emphasize nurses as core users, closely align with their clinical workflows, and ultimately create practical, user-friendly tools through thoughtful interface design, functional logic, and intelligent alert mechanisms.

Objectives We aimed to design and develop a clinical decision support tool based on the Data-Information-Knowledge-Wisdom model, which minimizes infusion errors by providing real-time alerts and standardizing workflows.

Methods A nonrandomized trial (May–July 2024) in a tertiary hospital compared traditional practices (n = 1,204) with a CDSS (n = 1,207) using 300 clinical rules and a personal digital assistant interface. Outcomes included error rates, severity, nurse satisfaction, and efficiency.

Results The CDSS reduced errors by 56.8% (16.69–7.21%, p < 0.001), eliminated severe errors (Level 3–4), improved nurse satisfaction (mean: 69.1/85 on a 17–85 scale), and reduced prescription processing time by 41%.

Conclusion This nurse-led CDSS enhances infusion safety and efficiency, offering a scalable solution. Artificial intelligence-driven predictive error detection could further optimize outcomes.
https://www.thieme-connect.com/products/ejournals/abstract/10.1055/a-2867-0618