Faculty Bibliography

BACKGROUND:Poor documentation quality can significantly affect healthcare 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. OBJECTIVES/OBJECTIVE: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. METHODS: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 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 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. RESULTS:64 inpatient progress notes were included, representing 55% female patients with a median age of 73. GPT-4 feedback was non-inferior to physician feedback in all measures: understandability (mean 1.27, 95% CI 0.73 to 1.8, P < 0.001), usefulness (mean 2.09, 95% CI 1.27 to 2.91, P < 0.001), acceptability (mean 2.07, 95% CI 1.33 to 2.81, P < 0.001), and impartiality (mean -0.20, 95% CI -0.52 to 0.12, P < 0.001). CONCLUSIONS:This study shows that an LLM can be leveraged to generate note quality feedback that is non-inferior to expert clinician feedback.

INTRODUCTION/BACKGROUND:Idiopathic acute pancreatitis (IAP) accounts for up to 20% of acute pancreatitis cases despite thorough initial evaluation. Endoscopic ultrasound (EUS) and magnetic resonance cholangiopancreatography (MRCP) are commonly used as second-line imaging modalities when IAP etiologies are unclear, yet their comparative diagnostic performance remains uncertain. We conducted a systematic review and meta-analysis to assess the diagnostic yield of EUS versus MRCP in identifying etiologies of IAP. METHODS:A systematic search of PubMed, EMBASE, and Google Scholar (January 2000-April 2025) identified English-language studies comparing EUS and MRCP diagnostic performance in IAP. Two reviewers independently extracted data and assessed study quality using the Newcastle-Ottawa Scale. Pooled relative risks (RR) were calculated using a random-effects model, with subgroup, sensitivity, and publication bias analyses done per PRISMA 2020. Subgroup analyses assessed diagnostic yield by etiology-biliary disease (cholelithiasis, choledocholithiasis, microlithiasis, and sludge), pancreatic divisum, intraductal papillary mucinous neoplasm (IPMN), and malignancy (pancreatic adenocarcinoma and periampullary cancer). RESULTS: = 0.0%). CONCLUSION/CONCLUSIONS:EUS surpasses MRCP in diagnosing IAP biliary etiologies and has an overall higher diagnostic yield. Prior studies suggest that EUS may be oversensitive in diagnosing CP, which may be consistent with the increased rate of diagnosis by EUS in our data. There was some signal that EUS had a higher diagnostic yield for cancer, which may highlight a potential role in identifying occult malignancy in IAP evaluation. In practice, choosing EUS vs. MRCP depends on resource availability, but a patient-centered approach that integrates modality strengths with clinical profiles can improve diagnostic accuracy and prognostic outcomes.

BACKGROUND AND AIMS/OBJECTIVE:Endoscopic full-thickness resection (EFTR) is an established, safe technique for the resection of appendiceal orifice (AO) neoplasms. Post-EFTR appendicitis is a recognised adverse event. There are no systematic reviews and a paucity of literature which has assessed outcomes especially with respect to delayed appendicitis, mucocele, or fistula formation. We aimed to evaluate efficacy of EFTR for AO lesions. PATIENTS AND METHODS/METHODS:Consecutive AO lesions referred for consideration of EFTR were prospectively studied. Multiple data points were recorded including technical success, EFTR histopathological data, adverse events, and follow-up surveillance data by colonoscopy. Surveillance CT was performed due to concern of potential mucocele from the obstructed remnant appendix. RESULTS:Over a 4 year period to July 2023, 37 AO lesions were referred to a tertiary center for consideration of EFTR. EFTR was attempted in 35 (95%) lesions. Most lesions were small [median size 10mm, interquartile range (IQR) 10-15mm], Paris 0-IIa morphology (n=32, 91%) with serrated histopathology (n=17, 49%). R0 resection was achieved in most EFTR cases (n=30/35, 86%). Adverse events included appendicitis (n=4, 11%) and delayed bleeding (n=2, 6%). At 6-month (IQR 4-6 months) surveillance colonoscopy, there was 1 (3%) case of residual lesion. This was successfully treated endoscopically, confirmed on a second surveillance colonoscopy. There was one case of appendicitis of the remnant at 7 months. At surveillance CT abdomen/pelvis (median 15 months, IQR 7-37 months), 2/17 (12%) fistulas were identified. Both of these patients had presumed adhesions due to abdominal surgery prior to EFTR. CONCLUSIONS:In conclusion, EFTR is an effective technique for the curative resection of select, small (<15mm) Paris 0-IIa AO lesions. Appendicitis is a relatively common adverse event but often managed conservatively. The long-term significance post-EFTR fistulas remains unclear. Caution should be exercised when considering EFTR in a patient with prior regional surgery.

IMPORTANCE/UNASSIGNED:Hospital course (HC) summarization represents an increasingly onerous discharge summary component for physicians. Literature supports large language models (LLMs) for HC summarization, but whether physicians can effectively partner with electronic health record-embedded LLMs to draft HCs is unknown. OBJECTIVES/UNASSIGNED:To compare the editing effort required by time-constrained resident physicians to improve LLM- vs physician-generated HCs toward a novel 4Cs (complete, concise, cohesive, and confabulation-free) HC. DESIGN, SETTING, AND PARTICIPANTS/UNASSIGNED:Quality improvement study using a convenience sample of 10 internal medicine resident editors, 8 hospitalist evaluators, and randomly selected general medicine admissions in December 2023 lasting 4 to 8 days at New York University Langone Health. EXPOSURES/UNASSIGNED:Residents and hospitalists reviewed randomly assigned patient medical records for 10 minutes. Residents blinded to author type who edited each HC pair (physician and LLM) for quality in 3 minutes, followed by comparative ratings by attending hospitalists. MAIN OUTCOMES AND MEASURES/UNASSIGNED:Editing effort was quantified by analyzing the edits that occurred on the HC pairs after controlling for length (percentage edited) and the degree to which the original HCs' meaning was altered (semantic change). Hospitalists compared edited HC pairs with A/B testing on the 4Cs (5-point Likert scales converted to 10-point bidirectional scales). RESULTS/UNASSIGNED:Among 100 admissions, compared with physician HCs, residents edited a smaller percentage of LLM HCs (LLM mean [SD], 31.5% [16.6%] vs physicians, 44.8% [20.0%]; P < .001). Additionally, LLM HCs required less semantic change (LLM mean [SD], 2.4% [1.6%] vs physicians, 4.9% [3.5%]; P < .001). Attending physicians deemed LLM HCs to be more complete (mean [SD] difference LLM vs physicians on 10-point bidirectional scale, 3.00 [5.28]; P < .001), similarly concise (mean [SD], -1.02 [6.08]; P = .20), and cohesive (mean [SD], 0.70 [6.14]; P = .60), but with more confabulations (mean [SD], -0.98 [3.53]; P = .002). The composite scores were similar (mean [SD] difference LLM vs physician on 40-point bidirectional scale, 1.70 [14.24]; P = .46). CONCLUSIONS AND RELEVANCE/UNASSIGNED:Electronic health record-embedded LLM HCs required less editing than physician-generated HCs to approach a quality standard, resulting in HCs that were comparably or more complete, concise, and cohesive, but contained more confabulations. Despite the potential influence of artificial time constraints, this study supports the feasibility of a physician-LLM partnership for writing HCs and provides a basis for monitoring LLM HCs in clinical practice.

BACKGROUND AND STUDY AIMS/UNASSIGNED:EUS-guided gallbladder drainage (EUS-GBD) using lumen apposing metal stents (LAMS) has excellent technical and short-term clinical success for acute cholecystitis (AC). The goals of this study were to determine the long-term clinical outcomes and adverse events (AEs) of EUS-GBD with LAMS. PATIENTS AND METHODS/UNASSIGNED:A multicenter, retrospective study was conducted at 18 US tertiary care institutions. Inclusion criteria: any AC patient with attempted EUS-GBD with LAMS and minimum 30-day post-procedure follow-up. Long-term clinical success was defined as absence of recurrent acute cholecystitis (RAC) > 30 days and long-term AE was defined as occurring > 30 days from the index procedure. RESULTS/UNASSIGNED:<0.01) were associated with RAC. AEs occurred in 38 of 109 patients (34.9%) at any time, and in 10 of 109 (9.17%) > 30 days from the index procedure. Most long-term AEs (7 of 109; 6.42%) were LAMS-specific. No technical or clinical factors were associated with occurrence of AEs. LAMS were removed in 24 of 109 patients (22%). There was no difference in RAC or AEs whether LAMS was removed or not. CONCLUSIONS/UNASSIGNED:EUS-GBD with LAMS has a high rate of long-term clinical success and modest AE rates in patients with AC and is a reasonable destination therapy for high-risk surgical candidates.

Treatment of anastomotic ulcers, also known as marginal ulcers, is challenging, especially when established techniques have failed. PuraStat is a biocompatible synthetic peptide gel that is indicated for hemostasis of bleeding in the gastrointestinal tract and vascular anastomoses. We aim to evaluate the feasibility of PuraStat in the setting of nonhealing anastomotic ulcers when used alongside standard therapies. This is a multicenter case series of adult patients who had PuraStat applied with a follow-up repeat endoscopy. Nine out of 10 patients showed clinical improvement. We concluded that PuraStat is an effective agent to aid in healing of anastomotic ulcer.

Electronic health records have become an integral part of modern health care, but their implementation has led to unintended consequences, such as poor note quality. This case study explores how NYU Langone Health leveraged artificial intelligence (AI) to address the challenge to improve the content and quality of medical documentation. By quickly and accurately analyzing large volumes of clinical documentation and providing feedback to organizational leadership and individually to providers, AI can help support a culture of continuous note quality improvement, allowing organizations to enhance a critical component of patient care.