Faculty Bibliography

IMPORTANCE/OBJECTIVE:Patients with poor glycaemic control have a high risk for major cardiovascular events. Improving glycaemic monitoring in patients with diabetes can improve morbidity and mortality. OBJECTIVE:To assess the effectiveness of a patient portal message in prompting patients with poorly controlled diabetes without a recent glycated haemoglobin (HbA1c) result to have their HbA1c repeated. DESIGN/METHODS:A pragmatic, randomised clinical trial. SETTING/METHODS:A large academic health system consisting of over 350 ambulatory practices. PARTICIPANTS/METHODS:Patients who had an HbA1c greater than 10% who had not had a repeat HbA1c in the prior 6 months. EXPOSURES/METHODS:A single electronic health record (EHR)-based patient portal message to prompt patients to have a repeat HbA1c test versus usual care. MAIN OUTCOMES/RESULTS:The primary outcome was a follow-up HbA1c test result within 90 days of randomisation. RESULTS:The study included 2573 patients with a mean (SD) HbA1c of 11.2%. Among 1317 patients in the intervention group, 24.2% had follow-up HbA1c tests completed within 90 days, versus 21.1% of 1256 patients in the control group (p=0.07). Patients in the intervention group were more likely to log into the patient portal within 60 days as compared with the control group (61.2% vs 52.3%, p<0.001). CONCLUSIONS:Among patients with poorly controlled diabetes and no recent HbA1c result, a brief patient portal message did not significantly increase follow-up testing but did increase patient engagement with the patient portal. Automated patient messages could be considered as a part of multipronged efforts to involve patients in their diabetes care.

Inflammation is the body's response to infection, trauma or injury and is activated in a coordinated fashion to ensure the restoration of tissue homeostasis and healthy physiology. This process requires communication between stromal cells resident to the tissue compartment and infiltrating immune cells which is dysregulated in disease. Clinical innovations in patient diagnosis and stratification include measures of inflammatory activation that support the assessment of patient prognosis and response to therapy. We propose that (i) the recent advances in fast, dynamic monitoring of inflammatory markers (e.g., cytokines) and (ii) data-dependent theoretical and computational modeling of inflammatory marker dynamics will enable the quantification of the inflammatory response, identification of optimal, disease-specific biomarkers and the design of personalized interventions to improve patient outcomes - multidisciplinary efforts in which biomedical engineers may potentially contribute. To illustrate these ideas, we describe the actions of cytokines, acute phase proteins and hormones in the inflammatory response and discuss their role in local wounds, COVID-19, cancer, autoimmune diseases, neurodegenerative diseases and aging, with a central focus on cardiac surgery. We also discuss the challenges and opportunities involved in tracking and modulating inflammation in clinical settings.

Objectives Improve concordance between patient end-of-life preferences and code status orders by incorporating data from a state registry with Clinical Decision Support (CDS) within the electronic health record (EHR) to preserve patient autonomy and ensure that patients receive care that aligns with their wishes. Methods Leveraging a Health Information exchange (HIE) interface between the New York State Medical Orders for Life-Sustaining Treatment (eMOLST) registry and the EHR of our academic health system, we developed a bundled CDS intervention that displays eMOLST information at the time of code status ordering and provides an in-line alert when providers enter a resuscitation order discordant with wishes documented in the eMOLST registry. To evaluate this intervention, we performed a segmented regression analysis of an interrupted times series to compare percentage of discordant orders before and after implementation among all hospitalizations for which an eMOLST was available. Results We identified a total of 3648 visits that had an eMOLST filed prior to inpatient admission and a code status order placed during admission. There was a statistically significant decrease of discordant resuscitation orders of -5.95% after the intervention went live, with a relative risk reduction of 25%, [95% CI: -9.95%, -1.94%, p=0.009] in the pre- and post-intervention period. Logistic regression model after adjusting for co-variates showed an average marginal effect of -5.12% after the intervention [CI =-9.75%, -0.50%, p=0.03]. Conclusions Our intervention resulted in a decrease in discordant resuscitation orders. This study demonstrates that accessibility to eMOLST data within the provider workflow supported by CDS can reduce discrepancies between patient end-of-life wishes and hospital code status orders.

AIM/OBJECTIVE:To describe the self-reported mental health of nurses from 35 countries who worked during the COVID-19 pandemic. BACKGROUND:There is little occupationally specific data about nurses' mental health worldwide. Studies have documented the impact on nurses' mental health of the COVID-19 pandemic, but few have baseline referents. METHODS:A descriptive, cross-sectional design structured the study. Data reflect a convenience sample of 9,387 participants who completed the opt-in survey between July 31, 2022, and October 31, 2023. Descriptive statistics were run to analyze the following variables associated with mental health: Self-reports of mental health symptoms, burnout, personal losses during the pandemic, access to mental health services, and self-care practices used to cope with pandemic-related stressors. Reporting of this study was steered by the STROBE guideline for quantitative studies. RESULTS:Anxiety or depression occurred at rates ranging from 23%-61%, with country-specific trends in reporting observed. Approximately 18% of the sample reported experiencing some symptoms of burnout. The majority of nurses' employers did not provide mental health support in the workplace. Most reported more frequently engaging with self-care practices compared with before the pandemic. Notably, 20% of nurses suffered the loss of a family member, 35% lost a friend, and 34% a coworker due to COVID-19. Nearly half (48%) reported experiencing public aggression due to their identity as a nurse. CONCLUSIONS:The data obtained establish a basis for understanding the specific mental health needs of the nursing workforce globally, highlighting key areas for service development. IMPLICATIONS FOR NURSING POLICY/CONCLUSIONS:Healthcare organizations and governmental bodies need to develop targeted mental health support programs that are readily accessible to nurses to foster a resilient nursing workforce.

Ensuring reliability of Large Language Models (LLMs) in clinical tasks is crucial. Our study assesses two state-of-the-art LLMs (ChatGPT and LlaMA-2) for extracting clinical information, focusing on cognitive tests like MMSE and CDR. Our data consisted of 135,307 clinical notes (Jan 12th, 2010 to May 24th, 2023) mentioning MMSE, CDR, or MoCA. After applying inclusion criteria 34,465 notes remained, of which 765 underwent ChatGPT (GPT-4) and LlaMA-2, and 22 experts reviewed the responses. ChatGPT successfully extracted MMSE and CDR instances with dates from 742 notes. We used 20 notes for fine-tuning and training the reviewers. The remaining 722 were assigned to reviewers, with 309 each assigned to two reviewers simultaneously. Inter-rater-agreement (Fleiss' Kappa), precision, recall, true/false negative rates, and accuracy were calculated. Our study follows TRIPOD reporting guidelines for model validation. For MMSE information extraction, ChatGPT (vs. LlaMA-2) achieved accuracy of 83% (vs. 66.4%), sensitivity of 89.7% (vs. 69.9%), true-negative rates of 96% (vs 60.0%), and precision of 82.7% (vs 62.2%). For CDR the results were lower overall, with accuracy of 87.1% (vs. 74.5%), sensitivity of 84.3% (vs. 39.7%), true-negative rates of 99.8% (98.4%), and precision of 48.3% (vs. 16.1%). We qualitatively evaluated the MMSE errors of ChatGPT and LlaMA-2 on double-reviewed notes. LlaMA-2 errors included 27 cases of total hallucination, 19 cases of reporting other scores instead of MMSE, 25 missed scores, and 23 cases of reporting only the wrong date. In comparison, ChatGPT's errors included only 3 cases of total hallucination, 17 cases of wrong test reported instead of MMSE, and 19 cases of reporting a wrong date. In this diagnostic/prognostic study of ChatGPT and LlaMA-2 for extracting cognitive exam dates and scores from clinical notes, ChatGPT exhibited high accuracy, with better performance compared to LlaMA-2. The use of LLMs could benefit dementia research and clinical care, by identifying eligible patients for treatments initialization or clinical trial enrollments. Rigorous evaluation of LLMs is crucial to understanding their capabilities and limitations.

BACKGROUND/UNASSIGNED:Hospital mergers remain common, but their influence on healthcare quality varies. Data on effects of European hospital mergers are ill defined, and academic hospitals in particular. This case study assesses early quality of care changes in two formerly competing Dutch academic hospitals that merged on June 6, 2018. METHODS/UNASSIGNED:Statistical process control and interrupted time series analysis were performed. All adult, non-psychiatric patients, admitted between 01-03-2016 and 01-10-2022 were eligible for analysis. Primary outcome measure was all cause in-hospital mortality (or hospice), secondary outcomes were unplanned 30-day readmissions to same hospital, length of stay, and patients' hospital rating. Data were obtained from electronic health records, and patient experience surveys. FINDINGS/UNASSIGNED:The mean (SD) age of the 573 813 included patients was 54·3 (18·9) years. The minority was female (277 817, 48·4 %), and most admissions were acute (308 597, 53·8 %). No merger related change in mortality was found in the first 20 months post-merger (limited to the pre-Covid-19 era). For this same period, the 30-day readmission incidence changed to a downward slope post-merger, and the length of stay shortened (immediate level-change -3·796 % (95 % CI, -5·776 % to -1·816 %) and trend-change -0·150 % per month (95 % CI, -0·307 % to 0·007 %)). Patients' hospital ratings seemed to improve post-merger. INTERPRETATION/UNASSIGNED:In this quality improvement study, a full- and gradual post-merger integration strategy for a Dutch academic hospital merger was not associated with changes in in-hospital mortality, and yielded slight improved results for secondary quality of care outcomes.

OBJECTIVES/UNASSIGNED:Accelerating demand for patient messaging has impacted the practice of many providers. Messages are not recommended for urgent medical issues, but some do require rapid attention. This presents an opportunity for artificial intelligence (AI) methods to prioritize review of messages. Our study aimed to highlight some patient portal messages for prioritized review using a custom AI system integrated into the electronic health record (EHR). MATERIALS AND METHODS/UNASSIGNED:We developed a Bidirectional Encoder Representations from Transformers (BERT)-based large language model using 40 132 patient-sent messages to identify patterns involving high acuity topics that warrant an immediate callback. The model was then implemented into 2 shared pools of patient messages managed by dozens of registered nurses. A primary outcome, such as the time before messages were read, was evaluated with a difference-in-difference methodology. RESULTS/UNASSIGNED: = 396 466), an improvement exceeding the trend was observed in the time high-scoring messages sit unread (21 minutes, 63 vs 42 for messages sent outside business hours). DISCUSSION/UNASSIGNED:Our work shows great promise in improving care when AI is aligned with human workflow. Future work involves audience expansion, aiding users with suggested actions, and drafting responses. CONCLUSION/UNASSIGNED:Many patients utilize patient portal messages, and while most messages are routine, a small fraction describe alarming symptoms. Our AI-based workflow shortens the turnaround time to get a trained clinician to review these messages to provide safer, higher-quality care.

IMPORTANCE/UNASSIGNED:Virtual patient-physician communications have increased since 2020 and negatively impacted primary care physician (PCP) well-being. Generative artificial intelligence (GenAI) drafts of patient messages could potentially reduce health care professional (HCP) workload and improve communication quality, but only if the drafts are considered useful. OBJECTIVES/UNASSIGNED:To assess PCPs' perceptions of GenAI drafts and to examine linguistic characteristics associated with equity and perceived empathy. DESIGN, SETTING, AND PARTICIPANTS/UNASSIGNED:This cross-sectional quality improvement study tested the hypothesis that PCPs' ratings of GenAI drafts (created using the electronic health record [EHR] standard prompts) would be equivalent to HCP-generated responses on 3 dimensions. The study was conducted at NYU Langone Health using private patient-HCP communications at 3 internal medicine practices piloting GenAI. EXPOSURES/UNASSIGNED:Randomly assigned patient messages coupled with either an HCP message or the draft GenAI response. MAIN OUTCOMES AND MEASURES/UNASSIGNED:PCPs rated responses' information content quality (eg, relevance), using a Likert scale, communication quality (eg, verbosity), using a Likert scale, and whether they would use the draft or start anew (usable vs unusable). Branching logic further probed for empathy, personalization, and professionalism of responses. Computational linguistics methods assessed content differences in HCP vs GenAI responses, focusing on equity and empathy. RESULTS/UNASSIGNED:A total of 16 PCPs (8 [50.0%] female) reviewed 344 messages (175 GenAI drafted; 169 HCP drafted). Both GenAI and HCP responses were rated favorably. GenAI responses were rated higher for communication style than HCP responses (mean [SD], 3.70 [1.15] vs 3.38 [1.20]; P = .01, U = 12 568.5) but were similar to HCPs on information content (mean [SD], 3.53 [1.26] vs 3.41 [1.27]; P = .37; U = 13 981.0) and usable draft proportion (mean [SD], 0.69 [0.48] vs 0.65 [0.47], P = .49, t = -0.6842). Usable GenAI responses were considered more empathetic than usable HCP responses (32 of 86 [37.2%] vs 13 of 79 [16.5%]; difference, 125.5%), possibly attributable to more subjective (mean [SD], 0.54 [0.16] vs 0.31 [0.23]; P < .001; difference, 74.2%) and positive (mean [SD] polarity, 0.21 [0.14] vs 0.13 [0.25]; P = .02; difference, 61.5%) language; they were also numerically longer (mean [SD] word count, 90.5 [32.0] vs 65.4 [62.6]; difference, 38.4%), but the difference was not statistically significant (P = .07) and more linguistically complex (mean [SD] score, 125.2 [47.8] vs 95.4 [58.8]; P = .002; difference, 31.2%). CONCLUSIONS/UNASSIGNED:In this cross-sectional study of PCP perceptions of an EHR-integrated GenAI chatbot, GenAI was found to communicate information better and with more empathy than HCPs, highlighting its potential to enhance patient-HCP communication. However, GenAI drafts were less readable than HCPs', a significant concern for patients with low health or English literacy.

IMPORTANCE/UNASSIGNED:Large language models (LLMs) are crucial for medical tasks. Ensuring their reliability is vital to avoid false results. Our study assesses two state-of-the-art LLMs (ChatGPT and LlaMA-2) for extracting clinical information, focusing on cognitive tests like MMSE and CDR. OBJECTIVE/UNASSIGNED:Evaluate ChatGPT and LlaMA-2 performance in extracting MMSE and CDR scores, including their associated dates. METHODS/UNASSIGNED:Our data consisted of 135,307 clinical notes (Jan 12th, 2010 to May 24th, 2023) mentioning MMSE, CDR, or MoCA. After applying inclusion criteria 34,465 notes remained, of which 765 underwent ChatGPT (GPT-4) and LlaMA-2, and 22 experts reviewed the responses. ChatGPT successfully extracted MMSE and CDR instances with dates from 742 notes. We used 20 notes for fine-tuning and training the reviewers. The remaining 722 were assigned to reviewers, with 309 each assigned to two reviewers simultaneously. Inter-rater-agreement (Fleiss' Kappa), precision, recall, true/false negative rates, and accuracy were calculated. Our study follows TRIPOD reporting guidelines for model validation. RESULTS/UNASSIGNED:For MMSE information extraction, ChatGPT (vs. LlaMA-2) achieved accuracy of 83% (vs. 66.4%), sensitivity of 89.7% (vs. 69.9%), true-negative rates of 96% (vs 60.0%), and precision of 82.7% (vs 62.2%). For CDR the results were lower overall, with accuracy of 87.1% (vs. 74.5%), sensitivity of 84.3% (vs. 39.7%), true-negative rates of 99.8% (98.4%), and precision of 48.3% (vs. 16.1%). We qualitatively evaluated the MMSE errors of ChatGPT and LlaMA-2 on double-reviewed notes. LlaMA-2 errors included 27 cases of total hallucination, 19 cases of reporting other scores instead of MMSE, 25 missed scores, and 23 cases of reporting only the wrong date. In comparison, ChatGPT's errors included only 3 cases of total hallucination, 17 cases of wrong test reported instead of MMSE, and 19 cases of reporting a wrong date. CONCLUSIONS/UNASSIGNED:In this diagnostic/prognostic study of ChatGPT and LlaMA-2 for extracting cognitive exam dates and scores from clinical notes, ChatGPT exhibited high accuracy, with better performance compared to LlaMA-2. The use of LLMs could benefit dementia research and clinical care, by identifying eligible patients for treatments initialization or clinical trial enrollments. Rigorous evaluation of LLMs is crucial to understanding their capabilities and limitations.

BACKGROUND:During the COVID-19 pandemic, acute respiratory infection (ARI) antibiotic prescribing in ambulatory care markedly decreased. It is unclear if antibiotic prescription rates will remain lowered. METHODS:We used trend analyses of antibiotics prescribed during and after the first wave of COVID-19 to determine whether ARI antibiotic prescribing rates in ambulatory care have remained suppressed compared to pre-COVID-19 levels. Retrospective data was used from patients with ARI or UTI diagnosis code(s) for their encounter from 298 primary care and 66 urgent care practices within four academic health systems in New York, Wisconsin, and Utah between January 2017 and June 2022. The primary measures included antibiotic prescriptions per 100 non-COVID ARI encounters, encounter volume, prescribing trends, and change from expected trend. RESULTS:At baseline, during and after the first wave, the overall ARI antibiotic prescribing rates were 54.7, 38.5, and 54.7 prescriptions per 100 encounters, respectively. ARI antibiotic prescription rates saw a statistically significant decline after COVID-19 onset (step change -15.2, 95% CI: -19.6 to -4.8). During the first wave, encounter volume decreased 29.4% and, after the first wave, remained decreased by 188%. After the first wave, ARI antibiotic prescription rates were no longer significantly suppressed from baseline (step change 0.01, 95% CI: -6.3 to 6.2). There was no significant difference between UTI antibiotic prescription rates at baseline versus the end of the observation period. CONCLUSIONS:The decline in ARI antibiotic prescribing observed after the onset of COVID-19 was temporary, not mirrored in UTI antibiotic prescribing, and does not represent a long-term change in clinician prescribing behaviors. During a period of heightened awareness of a viral cause of ARI, a substantial and clinically meaningful decrease in clinician antibiotic prescribing was observed. Future efforts in antibiotic stewardship may benefit from continued study of factors leading to this reduction and rebound in prescribing rates.