Pagali, Sandeep R.

ABSTRACT Background Mobility functional status is inconsistently recorded in electronic health records. Recent advances in Large Language Models (LLMs) enable automated extraction of functional information from unstructured clinical notes. Leveraging a validated functional LLM with high performance in mobility extraction, we aimed to study the association between LLM‐derived mobility functional status and clinical outcomes, including emergency department (ED) visits, hospitalizations, and mortality. Methods Individuals aged 60 years and older enrolled in the Mayo Clinic Study of Aging, with documented clinical notes in 2022 were included. Individuals without any identifiable functional information were excluded. The LLM‐derived mobility functional status categorized mobility as impaired or unimpaired, across the five mobility domains as classified by International Classification of Functioning, Disability, and Health. Clinical outcomes in 2023 were extracted from the Rochester Epidemiology Project database. Logistic regression models assessed associations between mobility impairment and outcomes, adjusted for age, sex, Elixhauser comorbidity index, and education. Results Three thousand three hundred thirty‐seven individuals were screened, of whom three hundred forty‐three had no functional information. Two thousand nine hundred ninety‐four individuals were analyzed. 79.4% had at least one mobility impairment. Higher mobility impairment is associated with older age and greater Elixhauser comorbidity. Individuals with 3 or more impairments had an increased adjusted odds of ED visits (Adjusted Odds Ratio (aOR): 2.18 (1.64, 2.90)) and hospitalizations (aOR: 2.35 (1.51, 3.65)) compared to unimpaired individuals ( p < 0.001), while mortality did not demonstrate a significant association with mobility impairment. Conclusion LLM‐derived mobility functional status assessment was associated with ED visits and hospitalization, but not with all‐cause mortality after adjustment. Prospective validation and evaluation of incremental discrimination, calibration, and clinical utility are needed before clinical implementation.

Objective Designation of a Medical Power of Attorney (MPOA) is a key component of advance care planning (ACP) that ensures patient preferences are honored when decision-making capacity is not present. This study aimed to quantify MPOA documentation rates and identify demographic and clinical factors associated with documentation presence. Methods We conducted a retrospective observational study using electronic health record data from 979 610 adult hospitalization encounters representing 459 365 unique patients within a large U.S. healthcare system from January 1, 2020, through December 31, 2021. All hospitalized patients aged ≥18 years were included. The primary outcome was MPOA documentation. Demographic and clinical variables were analyzed using multivariable logistic regression. Statistical significance was defined as P < .05. Results Among 979 610 encounters, 260 752 encounters (26.6%) had documented MPOA. Male sex, White race, older age (67.0 vs 54.3 years), and unmarried status were associated with a higher MPOA documentation rate (all, P < .001). In multivariable analyses, male sex (adjusted odds ratio [AOR], 1.07; 95% CI, 1.06-1.08) and unmarried status (AOR range, 1.41-1.75) were associated with higher odds of MPOA documentation. Dementia (AOR, 1.90; 95% CI, 1.86-1.93) and malignancy (AOR, 2.22; 95% CI, 2.19-2.25) were associated with higher odds, while discharge against medical advice was associated with lower odds (AOR, 0.91; 95% CI, 0.88-0.94) of MPOA documentation. Conclusions MPOA documentation rates were low, highlighting missed opportunities to identify alternate decision makers when a patient lacks capacity. Targeted interventions to promote surrogate decision-maker designation and documentation are needed, especially in inpatient care settings.

Abstract Objective This study evaluated sequential multi-site fine-tuning of large language models (LLMs), simulating incremental deployment across institutions for extracting mobility functional status from unstructured clinical notes. Materials and Methods We assembled a corpus of 600 clinical notes from 3 institutions, yielding 3810 annotated note sections (1200 training; 2610 testing). Building upon our previously established LLM mobility extraction pipeline, which utilized optimized prompt engineering, we applied parameter-efficient low-rank adaptation (LoRA) fine-tuning. We evaluated sequential multi-site fine-tuning, where the model is trained locally and only the updated LoRA weights are transferred to the next site. We compared this against joint fine-tuning, which requires pooling all multi-site patient data centrally before training. Performance was measured by micro F1-scores for 2 distinct subtasks: mobility extraction and impairment classification. Results Sequential multi-site fine-tuning yielded comparable results to centralized joint fine-tuning for mobility extraction (F1, 0.861 vs 0.894) and impairment classification (F1, 0.912 vs 0.900). Both sequential and joint fine-tuning outperformed an untuned 70B model. Additionally, single-site LoRA models showed robust cross-site generalization. Discussion Sequential multi-site fine-tuning offers a resource-friendly approach to adapt LLMs to clinical environments. By retaining previously acquired knowledge without severe interference, this strategy allows for local fine-tuning at each institution without the need to pool or share sensitive patient data required for joint fine-tuning. Conclusion Sequential multi-site fine-tuning provides a viable solution for the scalable, incremental deployment of LLMs across institutions, showing promising generalizability for clinical information extraction from electronic health record narratives.

Objectives: The electroconvulsive therapy for the acute management of severe agitation in dementia (ECT-AD) trial is a multisite study evaluating the effectiveness of ECT for severe agitation in individuals with advanced dementia. Given that participants lack decisional capacity because of advanced cognitive impairment, the trial required a structured and ethically sound approach to obtaining informed consent and assent across jurisdictions with differing legal requirements, including Massachusetts, New York, Georgia, Minnesota, and Michigan. This paper describes the approach used in the study. Methods: We describe the consent framework developed for the ECT-AD trial, including procedures for identifying legally authorized representatives, complying with state-specific laws for ECT consent, obtaining surrogate consent for research participation, and assessing patient assent and dissent using a standardized tool. Conclusions: The approach used in this study may serve as a practical model for future research involving individuals without decision-making capacity and legally complex interventions such as ECT.

Abstract Clinical narrative text contains crucial patient information, yet reliable extraction remains challenging due to linguistic variability, documentation habits, and differences across care settings. Large language models (LLMs) have shown strong accuracy on clinical information extraction (IE), but their reproducibility (stability under repeated runs) and robustness (stability under small, natural prompt variations) are less consistently quantified, despite being central to clinical deployment. In this study, we evaluate three open-weight LLMs representing distinct modeling choices: a dense general-purpose model (Llama 3.3), a mixture-of-experts (MoE) general-purpose model (Llama 4), and a domain-tuned medical model (MedGemma). We focus on binary clinical IE aligned with four mobility classes from the International Classification of Functioning, Disability and Health (ICF) framework. Using a controlled experimental design, we quantify (1) intra-prompt reproducibility across repeated sampling and (2) inter-prompt robustness across paraphrased prompts. We jointly report predictive performance (F1-score) and stability (Fleiss’ Kappa). And we test factor effects using three-way ANOVA with post-hoc comparisons. Results show that increasing temperature generally degrades agreement, but the magnitude depends on model and task; furthermore, prompt paraphrasing can substantially reduce stability, with particularly large drops for the MoE model. Finally, we evaluate a practical mitigation, self-consistency via majority voting, which improves κ substantially and often improves or preserves F1-score, at the cost of additional inference. Together, these findings provide a reproducible framework and concrete recommendations for evaluating and improving LLM reliability in clinical IE.

ABSTRACT Advanced chronic diseases, or multicomplexity in older adults presents unique challenges. Transthyretin cardiac amyloidosis (ATTR‐CA) is one such scenario where heart failure is a common presentation, and management remains challenging. We describe the challenges involved in the diagnosis of ATTR‐CA in older adults, which has implications on treatment options. We discuss the application of comprehensive geriatric assessment (CGA) and shared decision making (SDM) in the context of these challenges. Based on geriatric medicine principles, an innovative framework for applying SDM in ATTR‐CA patients is proposed, including consideration of the patient priorities care approach. The proposed framework emphasizes assessing functionality, frailty, and life expectancy to help categorize risk. This framework can be applied in various advanced chronic diseases or multicomplexity. Based on risk categorization, treatment burden, and alignment with values and preferences, management pathways are suggested for each risk category. Incorporating CGA and SDM, the proposed framework supports patient‐centered care, ensuring that clinical recommendations are tailored to each older adult's unique needs and goals. In ATTR‐CA, a collaboration between cardiology and geriatric medicine provides significant value in managing older adults. The need to prevent age‐related bias in clinical decision‐making exists across all health conditions, and the proposed framework allows for a thorough evaluation of multimorbidity, frailty, disability, and patient preferences. While ATTR‐CA is used as a prototype, this integrated approach can be applied across all health conditions and is essential for delivering holistic care, improving communication, and aligning treatment plans with patient values.