Kathleen M. McTigue

Background: Acute kidney injury (AKI) is a common, life-threatening complication among hospitalized patients. Timely prediction is crucial for prevention, while accurate prognosis following onset is essential for guiding care. Although electronic health records (EHRs) provide rich longitudinal data, their high dimensionality, heterogeneity, and temporal complexity pose challenges for traditional machine learning. Objective: The aim of this study was to develop and evaluate a transformer-based model for generalizable AKI prediction and prognosis by modeling long-range temporal dependencies and complex relationships in heterogeneous EHR data and to examine its interpretability. Methods: We developed the Biomarker-Aware Transformer for AKI (BAT-AKI), which integrates multidimensional embeddings capturing temporal dynamics, textual descriptions, hierarchical ontology structures, and semantic relationships in EHRs. To enhance sensitivity to AKI-related biomarker fluctuations, BAT-AKI combines masked language modeling with a novel biomarker abnormality-aware pretraining task. The model was pretrained and fine-tuned for 4 tasks: AKI onset, early reversal, recovery at discharge, and 90-day post-AKI mortality. Pretraining and internal validation used data from the National Patient-Centered Clinical Research Network (PCORnet) site, with external validation across 3 additional sites. Results: BAT-AKI achieved the strongest overall discrimination, outperforming baseline models in area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) on most tasks. For AKI onset prediction, it achieved an AUROC of 0.8549 (95% CI 0.8433-0.8658) and an AUPRC of 0.6041 (95% CI 0.5786-0.6292), exceeding the second-best baseline by 2.8% in AUROC and 8.1% in AUPRC. Across prognostic tasks, BAT-AKI achieved AUROCs of 0.6006 (95% CI 0.5703-0.6318), 0.6080 (95% CI 0.5745-0.6397), and 0.8017 (95% CI 0.7708-0.8311) and AUPRCs of 0.7036 (95% CI 0.6690-0.7388), 0.7417 (95% CI 0.7082-0.7765), and 0.5063 (95% CI 0.4422-0.5684) for early reversal, recovery at discharge, and 90-day post-AKI mortality, respectively, corresponding to relative gains of 2.4% to 6% in AUROC and 3.7% to 4.1% in AUPRC over the second-best baseline. Performance remained stable across external cohorts, with AKI onset AUROCs ranging from 0.7754 (95% CI 0.7677-0.7827) to 0.7989 (95% CI 0.7885-0.8092) and AUPRCs ranging from 0.4777 (95% CI 0.4631-0.4924) to 0.5687 (95% CI 0.5495-0.5871), corresponding to relative improvements of 3.7% to 5.3% in AUROC and 16.8% to 25.3% in AUPRC over the transformer baseline. In a low-resource setting using only 2% of the fine-tuning data, BAT-AKI achieved an AUROC of 0.7898 (95% CI 0.7759-0.8036) and an AUPRC of 0.4903 (95% CI 0.4640-0.5178) for AKI onset prediction, exceeding the second-best baseline by 4.2% in AUROC and 10.9% in AUPRC. Visualization of embeddings, attention patterns, and integrated-gradient feature attributions highlighted clinically meaningful relationships and patient-specific trajectories. Conclusions: BAT-AKI provides generalizable AKI prediction and prognosis with consistent discrimination across institutions, while offering interpretability insights into its predictions, supporting its potential for early warning and AKI care optimization.

Background: Patient narratives often aim to foster empathy, sustain attention, and translate lived experiences into patient-centered research priorities; however, while the content of these stories has been widely examined, audience engagement, which can influence whether narratives achieve these aims, has received less attention. Currently, a standardized method to assess engagement with unstructured patient narratives is lacking. We aimed to validate an adapted narrative engagement scale and examine potential differences in engagement based on storyteller demographics and health topics. Methods: We analyzed audio narratives from patients and caregivers collected by [Platform Name Blinded for Review]. Participants ( n = 646) shared unstructured health-related stories. We adapted the Busselle and Bilandzic narrative engagement scale, originally designed for visual media, to assess three dimensions: narrative understanding, attentional focus, and emotional engagement. Confirmatory factor analysis (CFA) was conducted on a subset of narratives ( n = 174) to verify the factor structure. For the remaining narratives ( n = 472), we calculated factor scores and performed Mann-Whitney U tests to compare engagement across demographic groups and health topics. Results: Most participants were female, White, and college-educated. The adapted scale demonstrated preliminary evidence of acceptable psychometric performance. Overall, stories were rated as easy to understand and able to maintain audience focus. Demographic factors were not significantly associated with engagement scores. Notably, narratives of COVID-19 were significantly easier ( p <0.05) to understand and pay attention to, but less emotionally engaging than non-COVID-19 stories, whereas injury- or wound-related narratives elicited higher emotional engagement. Engagement scores remained consistent across other health topics. Conclusions: This study provides preliminary validation evidence for the adapted narrative engagement measure for evaluating audience engagement with unstructured patient audio narratives across diverse health concerns. Future efforts should focus on capturing narratives from underrepresented groups to strengthen generalizability. This adapted measure can help researchers systematically evaluate not only what patient stories communicate, but also how audience understand, attend to, and emotionally engage with those stories, supporting the meaningful use of patient narratives in patient-centered research.

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have reshaped the clinical approach to managing obesity and type 2 diabetes. As approved indications have expanded, use of GLP-1RAs has increased rapidly in the United States. Although randomized trials demonstrate strong efficacy, many questions remain about their optimal use in clinical practice. Real-world data (RWD) from electronic health records, registries, insurance claims, and other sources offer a promising avenue to address these questions. However, concerns about data quality, selection bias, and incomplete ascertainment of medication use and outcomes pose significant challenges to the validity of the resulting evidence. In May 2025, the National Institute of Diabetes and Digestive and Kidney Diseases convened experts from regulatory agencies, payer organizations, and academia to explore these challenges. This second of 2 synopsis articles on the workshop summarizes the discussion around the strengths and limitations of various RWD sources and methodological approaches to strengthen causal inference and generalizability. Presenters highlighted pragmatic clinical trials and target trial emulation as strategies to generate stronger real-world evidence (RWE) that is relevant to both clinical practice and policy. The workshop underscored that careful attention to study design, data limitations, and analytic approach is essential to yield RWE that informs clinicians, patients, payers, and policymakers.

Background: Existing work to understand adults’ health care experiences has focused on the analysis of patient feedback provided as written responses to after-visit surveys or social media discourse. Often, such written feedback has been studied using natural language processing techniques, such as topic detection and sentiment analysis, to provide coarse-grained insights. Storytelling is a powerful form of communication and may provide insights into factors contributing to gaps in health care outcomes and avenues for improvement. In addition, studying health care experiences using natural language processing techniques has been limited to patients. The experiences of stakeholders, such as caregivers and health care providers, remain underexplored. Objective: We extract fine-grained insights from health care experiences through narratives collected from patients, caregivers, and health care providers using large language models (LLMs). Topic detection, together with hierarchical summarization of long-form stories from individuals, offers fine-grained insights. Furthermore, the study demonstrates that generated summaries can be evaluated using the LLM-as-a-judge framework and validates the outcomes through comparisons with 2 domain experts. Methods: Fifty automatically transcribed stories of African American experiences were used to identify topics in their experiences using the latent Dirichlet allocation (LDA) technique. Stories about a given topic were summarized using an open-source LLM-based hierarchical summarization approach. Topic summaries were generated by summarizing across story summaries for each story that addressed a given topic. The generated topic summaries were rated for fabrication, accuracy, comprehensiveness, and usefulness by the GPT-4 model; its reliability was validated against the original story summaries by 2 domain experts. Results: Whisper-based automatic transcription of audio narrations achieved a Levenshtein score of 6%. Twenty-six topics were identified using LDA and labeled using the LLM in the 50 African American stories. The GPT-4 ratings suggest that topic summaries were free from fabrication, highly accurate, comprehensive, and useful. The reliability of GPT ratings compared to expert assessments showed moderate-to-high agreement (Bennett S -score of 0.65 or higher). Our approach identified African American experience-relevant topics, such as health behaviors, interactions with medical team members, caregiving, and symptom management, among others. Such insights could help researchers learn from unstructured datasets in an efficient manner—leveraging the communicative power of storytelling. Conclusions: The use of LDA and LLMs to identify and summarize the experiences of African American individuals suggests a variety of possible avenues for health research and possible clinical improvements to support patients and caregivers, thereby improving health outcomes.

PCORnet ^® , a national resource funded by the Patient-Centered Outcomes Research Institute, is designed to enhance the nation’s capacity to conduct efficient, patient-centered health research. The robust and adaptable PCORnet infrastructure can be leveraged to support a variety of study designs. Within this framework, PCORnet ^® Studies, a distinguished subset of research studies that meets specific criteria and approval, serve as exemplary models of patient-centered research, standing out for their rigorous adherence to criteria that elevate the quality and impact of research. Over the past 10 years, over 300 studies have been completed using the PCORnet infrastructure, including 58 studies that have earned the distinction of being a meritorious PCORnet ^® Study, of which 19 have been completed. We explore several of these efforts, highlighting the study archetypes supported by the PCORnet infrastructure, as well as the clinical therapeutic areas of these studies, funding sources, and breadth of PCORnet institutional users. We also provide lessons learned from our accumulated experience that translates the PCORnet infrastructure into a continuously learning framework and highlight unique capabilities of the PCORnet infrastructure supporting innovation in future trials.

Health plans and public payers (ie, Medicare and Medicaid) maintain records of millions of transactions between patients and health care providers from claims submitted by hospitals, clinics, pharmacies, and other care settings. A major use of claims data in research is to supplement the information contained in patients’ medical records. PCORnet is a large, distributed “network of networks” funded by the Patient-Centered Outcomes Research Institute (PCORI) to improve the nation’s capacity to efficiently conduct definitive health research. Seventy-eight partner health systems nationwide map clinical data from their electronic health records to the PCORnet ^® Common Data Model (CDM) so that the data may be efficiently used for research purposes. The ability to link data from electronic health records in the PCORnet infrastructure with complementary data from other sources, such as health insurance claims, further enhances the capacity for comparative clinical effectiveness research (CER). This commentary showcases the health plan partnerships of 3 PCORnet ^® Clinical Research Networks (CRNs)—REACHnet, PaTH, and OneFlorida+—that enhance the capacity for CER involving linked clinical and claims data. We describe the transferable regulatory and technical infrastructures in place to efficiently link data for research purposes. To demonstrate these partnerships and data linkage in action, we also discuss research use cases pertaining to weight-related outcomes and diabetes that align with payers’ interests in chronic disease management.

BACKGROUND Acute kidney injury (AKI) is a common, life-threatening complication among hospitalized patients. Timely prediction is crucial for prevention, while accurate prognosis following onset is essential for guiding care. Although electronic health records (EHRs) provide rich longitudinal data, their high dimensionality, heterogeneity, and temporal complexity pose challenges for traditional machine learning. OBJECTIVE The aim of this study was to develop and evaluate a transformer-based model for generalizable AKI prediction and prognosis by modeling long-range temporal dependencies and complex relationships in heterogeneous EHR data and to examine its interpretability. METHODS We developed the Biomarker-Aware Transformer for AKI (BAT-AKI), which integrates multidimensional embeddings capturing temporal dynamics, textual descriptions, hierarchical ontology structures, and semantic relationships in EHRs. To enhance sensitivity to AKI-related biomarker fluctuations, BAT-AKI combines masked language modeling with a novel biomarker abnormality-aware pretraining task. The model was pretrained and fine-tuned for 4 tasks: AKI onset, early reversal, recovery at discharge, and 90-day post-AKI mortality. Pretraining and internal validation used data from the National Patient-Centered Clinical Research Network (PCORnet) site, with external validation across 3 additional sites. RESULTS BAT-AKI achieved the strongest overall discrimination, outperforming baseline models in area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) on most tasks. For AKI onset prediction, it achieved an AUROC of 0.8549 (95% CI 0.8433-0.8658) and an AUPRC of 0.6041 (95% CI 0.5786-0.6292), exceeding the second-best baseline by 2.8% in AUROC and 8.1% in AUPRC. Across prognostic tasks, BAT-AKI achieved AUROCs of 0.6006 (95% CI 0.5703-0.6318), 0.6080 (95% CI 0.5745-0.6397), and 0.8017 (95% CI 0.7708-0.8311) and AUPRCs of 0.7036 (95% CI 0.6690-0.7388), 0.7417 (95% CI 0.7082-0.7765), and 0.5063 (95% CI 0.4422-0.5684) for early reversal, recovery at discharge, and 90-day post-AKI mortality, respectively, corresponding to relative gains of 2.4% to 6% in AUROC and 3.7% to 4.1% in AUPRC over the second-best baseline. Performance remained stable across external cohorts, with AKI onset AUROCs ranging from 0.7754 (95% CI 0.7677-0.7827) to 0.7989 (95% CI 0.7885-0.8092) and AUPRCs ranging from 0.4777 (95% CI 0.4631-0.4924) to 0.5687 (95% CI 0.5495-0.5871), corresponding to relative improvements of 3.7% to 5.3% in AUROC and 16.8% to 25.3% in AUPRC over the transformer baseline. In a low-resource setting using only 2% of the fine-tuning data, BAT-AKI achieved an AUROC of 0.7898 (95% CI 0.7759-0.8036) and an AUPRC of 0.4903 (95% CI 0.4640-0.5178) for AKI onset prediction, exceeding the second-best baseline by 4.2% in AUROC and 10.9% in AUPRC. Visualization of embeddings, attention patterns, and integrated-gradient feature attributions highlighted clinically meaningful relationships and patient-specific trajectories. CONCLUSIONS BAT-AKI provides generalizable AKI prediction and prognosis with consistent discrimination across institutions, while offering interpretability insights into its predictions, supporting its potential for early warning and AKI care optimization. CLINICALTRIAL

Background: The PCORnet ^® infrastructure was funded by PCORI in 2014 to streamline clinical trials, increase patient-centered research, and generate knowledge that leads to improved health care and outcomes. In this paper, we summarize the significant achievements of the infrastructure over the last decade as well as recent accomplishments. We also provide an update on the expanded patient population who receive care at sites participating in PCORnet ^® Clinical Research Networks (CRNs) and share priorities for the future. Methods: The electronic health records of 71 health systems participating in PCORnet ^® CRNs as of July 2024 were queried, and data were analyzed to describe 10 common health conditions, stratified by demographic characteristics of age, sex, race, ethnicity, and an index of social deprivation. Results: Out of over 100M total patients with activity in the last 10 years, health systems participating in PCORnet ^® CRNs had over 47 million unique patients with at least one encounter in 2023. The most common chronic conditions among these patients were hypertension (18%), anxiety disorders (10%), type 2 diabetes (8%), and asthma (5%). Over 20% of patients receiving care at a site participating in PCORnet were in the top 50% of metrics for area deprivation. The PCORnet infrastructure supported 51 PCORnet ^® studies, all of which met established guidelines for use of the PCORnet ^® Common Data Model (CDM), patient-engagement, and commitment to return of results. Discussion: PCORnet ^® CRNs represent a diverse and expanding patient population and often include data on the socioeconomic status of the communities. Through continued efforts to engage communities and patients and national-scale research, the PCORnet ^® infrastructure can help improve care and outcomes for patients affected by common and rare conditions.