Learning Health Systems最新文献

{"title":"Opportunities and Challenges in Using National EHR Networks for AI in Learning Health Systems.","authors":"Polina V Kukhareva, Ramkiran Gouripeddi, Niels Peek, Kensaku Kawamoto","doi":"10.1002/lrh2.70090","DOIUrl":"10.1002/lrh2.70090","url":null,"abstract":"<p><strong>Background: </strong>National electronic health record (EHR) networks can support learning health systems (LHSs) by enabling large-scale data aggregation, monitoring, and benchmarking, but their capacity to produce trustworthy and locally deployable machine learning and artificial intelligence (ML/AI) models remains uncertain. We characterized major US national EHR networks and examined barriers to ML/AI development and deployment across the LHS cycle.</p><p><strong>Methods: </strong>We conducted an environmental scan combining PubMed searches with reviews of network websites, governance documents, and federal and vendor white papers through September 2025. Eligible networks aggregated patient-level EHR data nationally. We abstracted scale, settings, data domains, harmonization approaches, and access models, identified ML/AI studies, and mapped barriers onto a seven-step LHS-AI cycle from data capture to implementation and monitoring.</p><p><strong>Results: </strong>We identified 23 national EHR networks spanning federal and academic consortia, vendor-led consortia, commercial aggregators, and practice-based research networks, covering fewer than 1 million to more than 200 million patients. Most used common data models to standardize inputs. We identified 34 ML/AI studies, but only a small subset was prospectively evaluated or integrated into clinical workflows. Common barriers included heterogeneous data capture, privacy and linkage constraints, residual variation despite harmonization, limited representativeness, need for local recalibration, and sociotechnical challenges related to implementation and evaluation.</p><p><strong>Conclusion: </strong>National EHR networks offer critical infrastructure for LHSs but currently function primarily as research rather than ML/AI platforms. Addressing data, implementation, and evaluation barriers to enable ML/AI development and subsequent local deployment is essential to realizing their potential.</p>","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 Suppl 1","pages":"e70090"},"PeriodicalIF":2.3,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13240372/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148200539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Large Language Models for Summarizing Advance Care Planning Information From Goals of Care Notes in the EHR.","authors":"Ninad Ekbote, Melody Akhondzadeh, Ross Graham, John Bell, Delee Glasser, Robert El-Kareh, Andrew Chua, Aaron Boussina, Arin Tai-Seale, Lily Poursoltan, Karandeep Singh, Parag Agnihotri, Ming Tai-Seale","doi":"10.1002/lrh2.70086","DOIUrl":"10.1002/lrh2.70086","url":null,"abstract":"<p><strong>Objectives: </strong>Embedding systematic, structured data extraction within electronic health records (EHR) is vital for improved real-time insights into care delivery. This study evaluates the feasibility of using large language models (LLMs) to extract structured advance care planning (ACP) information from unstructured Goals of Care (GoC) clinical notes in the EHR.</p><p><strong>Materials and methods: </strong>A sample of 100 de-identified GoC notes was manually annotated by clinicians across four ACP categories: Patient Priorities, Code Status, Decision Maker, and Documentation. Two LLMs (Mistral 24.07 and LLaMA 3.1) were prompted to extract structured outputs without domain-specific fine-tuning. Model outputs were compared to human annotations using cosine similarity of BioBERT embeddings.</p><p><strong>Results: </strong>Mistral 24.07 achieved high semantic similarity in Code Status (0.814), Documentation (0.781), and Patient Priorities (0.770), but lower alignment in Decision Maker (0.609).</p><p><strong>Conclusions: </strong>LLMs can effectively extract structured ACP information, particularly in well-documented categories, suggesting potential for scalable, data-driven feedback loops that improve the provision of care. However, accuracy challenges remain, and further refinement is needed for nuanced qualitative content categories.</p>","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 Suppl 1","pages":"e70086"},"PeriodicalIF":2.3,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13240055/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148200531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 S1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148094067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 S1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148094071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect of Using Personalized Estimates of Diabetes Risk During Primary Care Visits for People With Prediabetes.","authors":"Natalia Olchanski, Christine E Skiro, Kristen Sonon, Sarah Kaus, Ericka C Holmstrand, Elizabeth L Ciemins, John K Cuddeback, Francis R Colangelo, David M Kent","doi":"10.1002/lrh2.70087","DOIUrl":"10.1002/lrh2.70087","url":null,"abstract":"<p><strong>Background: </strong>To assess acceptability, feasibility, and effectiveness of incorporating individualized risk prediction into clinical assessment, decision making, and communication of risk of type 2 diabetes, with and without preventive interventions, in patients with prediabetes.</p><p><strong>Methods: </strong>We integrated a prediction model into the clinical workflow at a U.S. health care organization. We conducted patient and provider focus groups and pre- and post-dissemination surveys among 2500 patients with prediabetes who had primary care visits between May 2018 and December 2019. We compared rates of progression to type 2 diabetes at 3 years between the intervention group and a propensity score-matched cohort of patients who received usual care.</p><p><strong>Results: </strong>Prior to implementing the predictive model, 41.6% of providers and 63.8% of patients felt confident or very confident in their ability to estimate the risk of progression to diabetes for individual patients. After personalized risk information was made available, this increased to 92.8% for providers and 66.9% for patients. People with prediabetes who had a primary care visit where their care team had access to personal risk of developing type 2 diabetes assessed by the EHR-based prediction model were significantly less likely to progress to diabetes within 3 years, compared to a propensity-score-matched cohort who received usual care in the same health system without individualized risk estimates (19.5% vs. 27.6%, <i>p</i> = 0.042).</p><p><strong>Conclusions: </strong>Used at the point of care during a primary care visit, the EHR-based diabetes risk calculator helped providers prioritize patients for diabetes preventive interventions, facilitated communication, and improved health outcomes among patients with prediabetes.</p>","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 Suppl 1","pages":"e70087"},"PeriodicalIF":2.3,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13240528/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148200549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 S1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148094068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 S1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148094070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Engaging Arab/MENA Communities in Learning Health Systems: Insights and Guidance for Future Research on AI and Health Equity.","authors":"Reema Hamasha, Dalya Saleem, Kerry A Ryan, Sean Tan, Morgan Sielaff, Adnan Hammad, Mona Makki, Jodyn Platt","doi":"10.1002/lrh2.70091","DOIUrl":"10.1002/lrh2.70091","url":null,"abstract":"<p><strong>Introduction: </strong>The integration of artificial intelligence (AI) tools in healthcare offers significant opportunities to improve patient care. However, underrepresented groups such as the Arab/Middle Eastern North African (MENA) community in the United States have historically been excluded in health data and conversations regarding AI tool implementation and development. We present our experience with the Arab/MENA community exploring attitudes about the use of AI in healthcare, focusing on our engagement and recruitment efforts as well as relevance for learning health systems science.</p><p><strong>Methods: </strong>We conducted a virtual democratic deliberation session (<i>n</i> = 33) in partnership with the Arab Community Center for Economic and Social Services (ACCESS) in Michigan, as part of a larger study involving five sessions (<i>n</i> = 159). In partnership with ACCESS staff, we collaboratively developed study materials and recruited Arab/MENA community members to share their perspectives on AI in healthcare. Qualitative thematic analysis was used to identify the community's perspectives, priorities, and barriers to the use of AI in healthcare.</p><p><strong>Results: </strong>The deliberation session highlighted four key themes related to the use of AI tools: (1) transparency in AI development was viewed as essential to building community trust, (2) human connection, with concerns that increased reliance on AI could replace empathy and weaken patient-provider interactions, (3) the role of healthcare providers, with preference on providers using AI as a supportive tool rather than replacing direct care, and (4) representation due to concerns over whether AI systems would reflect the experiences and needs of the Arab/MENA community in healthcare.</p><p><strong>Conclusions: </strong>Community-based partnerships are essential for advancing responsible AI implementation in healthcare and building a learning health system. Our experiences highlight the importance of transparency, cultural sensitivity, and meaningful community involvement to build trust and address the needs of underrepresented groups as AI evolves.</p>","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 Suppl 1","pages":"e70091"},"PeriodicalIF":2.3,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13239094/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148200484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 S1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148094064","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Catching the Wave: AI In Health Services Research.","authors":"Sarah M Greene, Lucy A Savitz","doi":"10.1002/lrh2.70092","DOIUrl":"10.1002/lrh2.70092","url":null,"abstract":"","PeriodicalId":43916,"journal":{"name":"Learning Health Systems","volume":"10 Suppl 1","pages":"e70092"},"PeriodicalIF":2.3,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13240361/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148200512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}