Applied Clinical Informatics最新文献

{"title":"Enhancing Data Quality and Remote Accessibility in Clinical Trials: A SeroSelectTB Case Study in Ethiopia, Tanzania, and South Africa.","authors":"Sasho Najdov, Jordancho Arsov, Jovan Davchev, Tamirat Assefa, Jemrath Bikombo, Debora Kajeguka, Anna Okunola, Welile Nwamba, Aleksandar Josifoski, Carol Holm-Hansen","doi":"10.1055/a-2873-5092","DOIUrl":"10.1055/a-2873-5092","url":null,"abstract":"<p><strong>Background: </strong>Ensuring data quality and maintaining remote accessibility remain key challenges in clinical trials conducted in low-resource settings. Although several software platforms and digital tools attempt to address these challenges, no single, comprehensive solution adequately meets the complex demands of decentralized and intermittently connected trial environments. Consequently, context-specific tools must be developed and adapted to ensure reliable performance under such constraints.</p><p><strong>Objective: </strong>The aim of this study was to enhance data quality and enable uninterrupted participant randomization in settings with limited internet and electricity through the implementation of an automated Quality Control (QC) reporting system and an online and offline participant randomization application.</p><p><strong>Methods: </strong>A centralized automated QC reporting system was developed using a Python-based script, which was integrated with the Research Electronic Data Capture system to extract and process data, and generate a web-based monitoring dashboard, which displayed data summaries, recruitment charts, and flags on missing or inconsistent records. Thereafter, a mobile randomization application built with React Native was developed for online and offline participant randomization, employing block randomization to ensure balanced allocation between study arms and enrollment continuity.</p><p><strong>Results: </strong>The automated QC system reduced data verification time from months to 1 hour, enabling real-time oversight and early error correction. The randomization application maintained balanced study arms and allowed uninterrupted participant enrollment during internet outages. These digital tools improved data completeness, reduced bias and human error, and increased trial efficiency across all study sites.</p><p><strong>Conclusion: </strong>Automation and offline functionality can significantly enhance data quality, efficiency, and integrity in clinical trials conducted in resource-constrained environments. The experience from this study demonstrates the value of combining automated QC and offline randomization tools to ensure robust and continuous study operations.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":"17 2","pages":"365-373"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13287932/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148007202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preambient Artificial Intelligence Clinical Documentation Time for Pediatric Residents: A 3-Year Baseline Observational Study.","authors":"Yahya Almodallal, Natalie Ramsy, Lindsey A Knake, Anna Schmitz","doi":"10.1055/a-2856-4821","DOIUrl":"10.1055/a-2856-4821","url":null,"abstract":"<p><strong>Objectives: </strong>The objective of this study is to characterize pediatric resident documentation time using electronic health record (EHR) audit-log-data and to assess interindividual variability in documentation patterns.</p><p><strong>Methods: </strong>We conducted a retrospective, longitudinal study at an academic children's hospital, analyzing the EHR audit-log-data between July 1, 2021 and June 30, 2024. All clinical notes to which a pediatric resident contributed were included. Results are shown as descriptive statistics and pairwise comparisons of log-transformed continuous variables were performed using Welch's analysis of variance and Games-Howell post hoc testing.</p><p><strong>Results: </strong>Over 3 years, 79 residents contributed to the documentation of 156,898 clinical notes for an average of 2.1 hours per day. The mean (95% confidence interval) total resident time spent on one note was 12.1 (12.0-12.1) minutes. First-year residents contributed to 51.6% of all notes. More than half of resident note-editing time occurred outside scheduled shift hours (54.4%), including 56.3% of ambulatory note time and 53.0% of inpatient note time. Across the study period, monthly documentation time showed substantial month-to-month fluctuation but only small overall trends, with adjusted time-per-note declining significantly over time for most graduating classes.</p><p><strong>Conclusion: </strong>This single-center study quantified pediatric resident EHR documentation time and found that time was highest among postgraduate year-1 residents, frequently extended into nights and weekends, and varied widely between individuals. The data provide a baseline to inform residency-level workflow optimization and to evaluate interventions that aim to reduce documentation time while preserving quality and educational value.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":" ","pages":"289-296"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13287930/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147786699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Structured Framework for Electronic Health Record Optimization: The EASY Program.","authors":"Obeid Shafi, Daniel Liu, James S Magee, Ashley Antipolo, Feliciano Yu","doi":"10.1055/a-2831-5615","DOIUrl":"10.1055/a-2831-5615","url":null,"abstract":"<p><p>Despite widespread adoption, electronic health records (EHRs) continue to present persistent challenges related to usability, workflow inefficiencies, and clinician burden. Structured EHR optimization programs that address these issues in a scalable and replicable manner remain limited in the literature.This study aimed to describe the development, implementation, and outcomes of the EASY (Eliminate, Automate, Standardize, and Simplify, Y'all) EHR Optimization program-designed to enhance EHR usability, provider satisfaction, and workflow efficiency through a multidisciplinary, data-informed approach.EASY employs a structured three-phase model (people, process, technology), integrating user-centered design principles and institutional tools such as clinician observations, GROSS (Getting Rid of Stupid Stuff) surveys, Epic EHR analytics (Signal and Tune-Up reports), and SOAARR (Subjective, Objective, Artifacts, Assessment, Recommendations, Results) documentation. The program evolved to adopt a sprint-based implementation cycle (4-6 weeks) to maintain engagement, reduce scope creep, and support iterative improvements.EASY yielded both user-reported and analytic benefits. Providers described improved satisfaction, greater alignment with clinical workflows, and enhanced transparency through structured communication. Signal data demonstrated measurable improvements in documentation efficiency, ordering practices, and reduced time spent in specific EHR functions. Targeted usability enhancements, ranging from quick wins to complex builds, were data-informed and guided by clinician feedback. The EASY program reflects best practices in EHR optimization, aligning with literature that emphasizes provider engagement, data-informed design and decision-making, and multidisciplinary collaboration. Its modular design and analytic rigor support adaptability across varied clinical settings, though implementation success is contingent on sustained provider involvement and informatics infrastructure.EASY offers a practical, scalable model for structured EHR optimization. It provides a replicable framework and actionable strategies for informatics teams seeking to improve EHR usability, reduce burden, and foster provider-centered innovation.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":" ","pages":"153-161"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13008491/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147476021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of Clinical Informatics Interest and the Impact of Advanced Electronic Health Record Training on Trainee Proficiency.","authors":"Tiranun Rungvivatjarus, Weena Joshi, Mario Bialostozky","doi":"10.1055/a-2873-3262","DOIUrl":"10.1055/a-2873-3262","url":null,"abstract":"<p><strong>Background: </strong>While electronic health records (EHRs) have been almost universally adopted in academic health care settings, physician training has lagged in efficient use and application. Especially for trainees, EHR training is often confined to basic EHR onboarding, with few opportunities to develop advanced skills. Moreover, literature on advanced EHR training is sparse, and curricula remain largely unstandardized.</p><p><strong>Objectives: </strong>This study surveys pediatric fellows and medical students (MS) on their experience, attitude, and interest regarding clinical informatics. Additionally, we assessed the change in trainees reported EHR proficiency after receiving data literacy and self-service EHR data extraction training.</p><p><strong>Methods: </strong>This cross-sectional study surveyed pediatric trainees at a large tertiary pediatric health care system and first-year MS at the University of California, San Diego School of Medicine. An electronic survey was distributed between July and November 2022 to trainees and repeated 14 months after Epic SlicerDicer training. Descriptive statistics were used to analyze responses.</p><p><strong>Results: </strong>In total, 33/57 fellows and 93/419 MS (26.5% response rate) completed the survey. Interest in clinical informatics was reported by 53% of fellows and 23% of MS. Most respondents agreed informatics is important in clinical care, research, health care management, and continuing education. Many reported receiving insufficient advanced EHR training. Before training, 33% of fellows and 8% of MS reported proficiency in basic EHR accessing/navigating/editing skills. None of the fellows reported proficiency in data visualization/extraction. Following training, both MS and fellows reported increased proficiency in data visualization/extraction.</p><p><strong>Conclusion: </strong>Pediatric MS and fellows recognize the value of informatics in modern health care. Opportunities for advanced EHR training are limited for trainees and a single training session demonstrated improve proficiency. Broadening access to advanced EHR training across all stages of training could better equip future clinicians for real-world practice.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":" ","pages":"357-364"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13287954/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147876521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Variation in Measures of Electronic Health Record Use Outside Scheduled Hours: A Cross-Sectional Study of Academic Primary Care Physicians.","authors":"Adam Rule, Jeffrey J Baltus, Mark A Micek, Christine A Sinsky, Brian G Arndt","doi":"10.1055/a-2859-0028","DOIUrl":"10.1055/a-2859-0028","url":null,"abstract":"<p><strong>Background: </strong>The amount of time ambulatory physicians spend in an electronic health record (EHR) outside scheduled hours has grown in recent years and been linked to burnout. Measures provided by EHR vendors can help health systems track EHR use outside scheduled hours, but different measures track different periods of EHR use and none currently tracks all EHR use outside patient scheduled hours. How much vendor-derived measures differ from one another or from total EHR use outside patient scheduled hours is unclear.</p><p><strong>Objective: </strong>The objective of this study is to compare measures of EHR use outside scheduled hours.</p><p><strong>Methods: </strong>We collected data on 195 academic primary care physicians' EHR use and clinic schedules between February and July 2025. We measured each physician's EHR use outside patient scheduled hours (i.e., <i>Work Outside of Work</i>) and compared this investigator-derived measure to three vendor-derived measures of EHR use outside scheduled hours.</p><p><strong>Results: </strong>Study physicians averaged 6.0 hours of total EHR use and 2.9 hours of EHR use outside patient scheduled hours (i.e., <i>Work Outside of Work</i>) for every eight patient scheduled hours. The vendor-derived measures of <i>Pajama Time, Time on Unscheduled Days</i> plus <i>Time Outside Scheduled Hours</i>, and <i>Time Outside of Clinic Hours</i> averaged 0.9, 2.2, and 2.6 hours per eight patient scheduled hours, respectively.</p><p><strong>Conclusion: </strong>Almost half of study physicians' EHR use occurred outside patient scheduled hours. Common vendor-derived measures captured varying amounts of EHR use outside scheduled hours and all were less than total EHR use outside patient scheduled hours. This variation could affect calculations used to inform EHR training, program evaluation, or expectations for patient scheduled hours. Health systems should thus exercise caution when using vendor-derived measures to estimate total EHR use outside patient scheduled hours and continue to work with EHR vendors to develop scalable measures of EHR use that support diverse operational applications.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":"17 2","pages":"307-314"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13287947/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147934586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Industry Electives in Clinical Informatics Fellowship: Early Experiences from a Multi-Institution Survey.","authors":"Nicholas Genes, Priyanka Solanki, Joseph Kannry, Raman Khanna, Dara Mize, Veena Lingam, Robert W Turer, Michael G Leu","doi":"10.1055/a-2820-3029","DOIUrl":"10.1055/a-2820-3029","url":null,"abstract":"<p><p>Clinical informatics (CI) fellowship training equips physicians to design, implement, and evaluate health information systems in support of patient care. While core curricula emphasize academic health system experiences, fellows may benefit from exposure to industry settings where much health technology innovation originates.This study aimed to characterize the structure, perceived value, and logistical challenges of industry electives among CI fellowship programs and to synthesize best practices for integrating these experiences into training.We surveyed current and former CI fellows and their program directors from two Accreditation Council for Graduate Medical Education (ACGME)-accredited programs between September 2024 and March 2025. To be included, fellows were required to complete at least 4 weeks of an industry elective. Free-text responses were analyzed using inductive thematic analysis. A consensus-driven process was used to generate practical considerations for program design.Eight fellows reported on industry electives at non-health-center sites such as startups, vendors, and standards bodies. Their responses revealed four themes: (1) Enhanced skill development and exposure to technologies and workflows not available in academic settings; (2) logistical barriers, including limited institutional support, short duration, and complex legal agreements; (3) tangible deliverables such as dashboards, analytic tools, abstracts, and grants; and (4) professional networking that often shaped career trajectories, with some fellows receiving job offers. Practical considerations included identifying partner sites, designating supervisors, negotiating agreements early, defining objectives and deliverables, and addressing financial and logistical support.Industry electives provide career-shaping experiences for CI fellows, expanding exposure to innovation and fostering collaboration between academia and industry. With clear objectives, aligned competencies, and institutional support, these rotations can strengthen training and prepare fellows for diverse roles across health care and technology.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":" ","pages":"177-182"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12991848/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147318805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Object-Centric Process Mining of Laparoscopic Cholecystectomy Workflows: Identifying Bottlenecks and Optimization Opportunities.","authors":"Ufuk Celik","doi":"10.1055/a-2832-9369","DOIUrl":"10.1055/a-2832-9369","url":null,"abstract":"<p><p>This study applies object-centric process mining (OCPM) techniques to analyze laparoscopic cholecystectomy (LC) procedures. Traditional process mining techniques are limited in analyzing workflows involving multiple interacting entities (patients, surgeries, anesthesia physical status, etc.).This study aimed to construct an Object-Centric Event Log (OCEL) from LC procedures, discover multientity process patterns, identify workflow bottlenecks, and analyze how patient complexity affects perioperative dynamics.An OCEL representing 1,186 LC cases (1,108 performed, 78 cancelled) from the University of California, Irvine Medical Center (UCIMC) was analyzed, using PM4Py to obtain object-centric directly-follows graphs (OC-DFG) and variant explorer. Statistical comparisons examined intensive care unit (ICU) admission and the American Society of Anesthesiologists (ASA) classification effects on perioperative duration using Mann-Whitney U and Kruskal-Wallis tests.Process discovery revealed a 92% adherence to the reference clinical pathway. ICU-admitted patients (<i>n</i> = 353, 31.6%) demonstrated significantly longer perioperative durations than non-ICU patients (<i>n</i> = 765, median 13.83 vs. 8.85 hours, <i>p</i> < 0.001, Cohen's <i>d</i> = 0.81). The ASA rating showed no significant effect on total perioperative time (<i>p</i> = 0.824). Major bottlenecks included booking-to-operating room (OR) transfer (median 7.2 hours), preanesthesia preparation delays (28 minutes in 5.6% of cases), and postanesthesia care unit (PACU) discharge (median 19.5 hours, interquartile range [IQR]: 6.3-42.1 hours). Pathway completeness was 99.1% with minimal documentation errors. Handoff efficiency varied substantially, with OR-to-PACU transfers occurring rapidly (median: 6 minutes) but PACU-to-discharge transitions exhibiting extreme variability.OCPM enables multiperspective insights invisible to traditional case-centric approaches. While intraoperative phases function efficiently, preoperative scheduling and postoperative discharge represent primary bottlenecks. The high ICU admission rate (31.6%) likely reflects institutional case mix and data classification practices rather than true critical care needs. Targeted interventions addressing preoperative scheduling optimization, discharge bottlenecks, and real-time monitoring could substantially improve surgical throughput.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":" ","pages":"204-212"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13065364/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147460671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging a Large Language Model to Generate Quality Improvement Feedback for Clinical Notes.","authors":"Christopher J Kim, Joseph Gelfinbein, Nihan Gencerliler, Nusrat Jahan, Jahnavi Udaikumar, Lauren M Heery, Adam Goodman, Sarah Ng, Joel Attard, Sharmin Asha, Jesse Burk-Rafel, Benedict V Guzman, Katherine A Hochman, Paul A Testa, Jonah Feldman","doi":"10.1055/a-2851-0739","DOIUrl":"10.1055/a-2851-0739","url":null,"abstract":"<p><strong>Background: </strong>Poor documentation quality can significantly affect health care operations, but the feedback process for clinicians to improve clinical notes is time-consuming and often insufficient. Large language models (LLMs) such as Generative Pre-trained Transformer 4 (GPT-4) have the potential to streamline this process.</p><p><strong>Objectives: </strong>This study aimed to determine whether an LLM can generate feedback to improve the medical contingency and discharge planning (MCDP) component of clinical documentation that is non-inferior to feedback by physicians.</p><p><strong>Methods: </strong>A cross-sectional study of GPT-4 feedback and physician feedback on inpatient progress notes was conducted. A random sample of 64 inpatient progress notes identified by the validated artificial intelligence (AI) Audit Tool as having a low likelihood of containing MCDP was included from adult general medicine patients hospitalized at New York University Langone Health (NYULH) in December 2023. Both the GPT-4 model and attending physicians generated feedback on these inpatient progress notes. A/B testing was then conducted on the measures of understandability, usefulness, acceptability, and impartiality. Evaluations employed 5-point Likert scales that were converted to 10-point bidirectional interval scales for interpretability, ranging from -10 (human suggestions significantly better) to +10 (GPT-4 suggestions significantly better), with a non-inferiority threshold set to -1 for the primary endpoint.</p><p><strong>Results: </strong>Sixty-four inpatient progress notes were included, representing 55% female patients with a median age of 73 years. GPT-4 feedback was non-inferior to physician feedback in all measures: Understandability (mean, 1.27, 95% CI: 0.73-1.8, <i>p</i> < 0.001), usefulness (mean, 2.09, 95% CI: 1.27-2.91, <i>p</i> < 0.001), acceptability (mean, 2.07, 95% CI: 1.33-2.81, <i>p</i> < 0.001), and impartiality (mean, -0.20, 95% CI: -0.52 to 0.12, <i>p</i> < 0.001).</p><p><strong>Conclusion: </strong>This study shows that an LLM can be leveraged to generate note quality feedback that is non-inferior to expert clinician feedback.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":" ","pages":"252-258"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13287944/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147693173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Workflow for Artificial Intelligence-Enhanced Patient Messaging Services.","authors":"Matthew R Allen, Vijay M Tiyyala, Karthik Ramesh, Nimit Desai, Job Shiach, Mark Dredze, John W Ayers","doi":"10.1055/a-2852-9026","DOIUrl":"10.1055/a-2852-9026","url":null,"abstract":"<p><strong>Background: </strong>Current artificial intelligence (AI) integration in patient messaging often relies on generating AI drafts for clinician review, yet this workflow has achieved limited effect.</p><p><strong>Objective: </strong>This study aimed to describe and evaluate a novel, clinician-first workflow for patient messaging where AI enhances a clinician-generated draft.</p><p><strong>Methods: </strong>Using 268 patient questions from public data, we compared physician-only responses, AI-only responses, and AI-enhanced responses.</p><p><strong>Results: </strong>Responses were ranked on overall preference and the CREATE TRUST framework. AI-enhanced responses were significantly preferred overall, ranking first in 38.8% of evaluations (average rank 1.69; <i>p</i> < 0.01), outperforming both AI-only (27.6%; 2.29) and physician-only (25.5%; 2.11) responses. AI-enhanced responses ranked highest for Understandable (44.5%) and Tailored (39.4%). AI-only responses ranked highest for Thorough (71.0%) and Empathic (69.8%), while physician-only responses ranked highest in Authentic (90.9%; all <i>p</i> < 0.01). Safety analysis identified consequential additions in 3.36% and omissions in 1.12% of AI-enhanced messages.</p><p><strong>Conclusion: </strong>A novel workflow-based on the Clinical Action Support framework-where AI enhances a clinician's draft may offer an improved approach to AI implementation in patient messaging services.</p>","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":"17 2","pages":"269-274"},"PeriodicalIF":2.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13287956/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147786676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Erratum on Pagination Correction: Industry Electives in Clinical Informatics Fellowship: Early Experiences from a Multi-Institution Survey.","authors":"Nicholas Genes, Priyanka Solanki, Joseph Kannry, Raman Khanna, Dara Mize, Veena Lingam, Robert W Turer, Michael G Leu","doi":"10.1055/a-2843-2793","DOIUrl":"10.1055/a-2843-2793","url":null,"abstract":"","PeriodicalId":48956,"journal":{"name":"Applied Clinical Informatics","volume":"17 2","pages":"e1"},"PeriodicalIF":2.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13148862/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147844957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}