JBJesse Burk-Rafel

Jesse Burk-Rafel

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jesse Burk-Rafel has a total of 171 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Nudging Toward Precision Medical Education: Linking Diagnostic Exposures to Tailored Learning

    by Jesse Rafel, Daniel J Sartori, Helen Finkelstein, Omar Moussa, Marc M Triola

    Problem Authentic patient encounters are the raw material of clinical learning, yet the educational resources learners receive are rarely keyed to the diagnoses in front of them, creating temporal and cognitive gaps. Precision medical education (PME) proposes delivering the right resource to the right learner at the right moment, but practical implementation in the clinical learning environment remains limited. Approach We developed DxMentor, an electronic health record (EHR)-integrated platform that captures each learner's daily inpatient diagnostic exposures from documented International Classification of Diseases, Tenth Revision (ICD-10) codes. Artificial intelligence (AI) is used to match each diagnosis to an educator-curated formulary of micro-learning resources and board-style questions, and to PubMed-derived primary and synthesis literature converted into plain-language evidence summaries. A personalized email "nudge" is delivered before morning rounds, copying supervising attendings for residents, with engagement tracked longitudinally. We report implementation outcomes from July 2024-April 2026. Outcomes DxMentor evaluated 32,846 encounters from 335 medical students and 346 internal medicine residents, delivering 17,340 nudges containing 63,754 didactic resources, 17,038 question sets, and 23,594 summarized articles for approximately $390 in AI token costs. In a benchmarking sample, 91.5% (366/400) of diagnosis-resource pairs were rated relevant by physician-educators. Overall, 78.5% (12,393/15,793) of nudges were opened and 11.3% (1,955/17,340) had at least one click. Engagement was higher among residents than students (open: 80.7% vs. 65.8%; click-through: 12.7% vs. 3.1%; both P < .001), with substantial between-learner variability. Next Steps Email opens and clicks are engagement proxies rather than measures of learning. We are therefore linking nudges to educational outcomes, testing alternative recommendation strategies and timing, and expanding to additional specialties and ambulatory and surgical settings.

  • Development and validation of TRainee Attributable & Automatable Care Evaluations in Real-time (TRACERs)

    by Jesse Burk-Rafel, Stefanie S Sebok-Syer, Ian Larson, Sally A Santen, Eduardo Iturrate, Judee Richardson, Holly A Caretta-Weyer, Matthew Kelleher, Seth W Overla, Jason Keller, et al.

    Purpose: To develop TRainee Attributable & Automatable Care Evaluations in Real‑time (TRACERs) for inpatient diabetes management, collect validity evidence for their use in formative assessment, and explore performance variation across residents and institutions. Method: In 2023, a multi‑institutional team created 2 TRACERs based on type 2 diabetes guidelines—discourage bolus‑only insulin (TRACER #1) and encourage basal (± bolus) insulin (TRACER #2)—for internal medicine residents at 3 large residency programs. Residents were attributed to inpatient admissions based on placing the most medication orders in the first 12 hours. Structured queries extracted 35 discrete variables from the electronic health record. Two experts per institution reviewed random admissions (July-August 2022) to establish criterion validity. A retrospective cohort (July 2020-June 2023) added validity evidence. Results: Automated extraction achieved ≥96% sensitivity and ≥95% specificity when compared to manual review. Among 615 residents attributed to 6192 admissions of patients with type 2 diabetes at high risk for hyperglycemia, TRACER #1 occurred in 42.6% (1689/3965) of admissions at Program A, 28.9% (408/1410) at Program B, and 26.7% (218/817) at Program C. TRACER #2 occurred in 44.9% (367/817) of Program C admissions versus 24.2% (959/3965) at Program A and 28.2% (397/1410) at Program B (all P < .001). Four resident-level insulin‑ordering profiles were identified—consistent basal-bolus insulin use (most guideline-concordant), basal-predominant, bolus-predominant, and bolus-only (most guideline-discordant)—with between‑resident variation exceeding between‑program differences. Longitudinally, cohort-level trends masked opposing individual trajectories—some residents improved with exposure while others worsened—and performance tertiles were distinguishable early in training. Conclusions: TRACERs revealed substantial resident‑ and institution‑level variation in insulin‑ordering practices, including guideline deviations, demonstrating potential for real‑time formative feedback. Multi‑institutional implementation highlighted scalability barriers, including electronic health record heterogeneity and workflow‑dependent attribution, underscoring the need for continued refinement and broader validation.

  • Open Access

    Institutional Standing and Trainee Outcomes in the 2025 US Residency Match

    by Joseph I. Turner, Ariel Arias, Jesse Burk-Rafel, Eric K. Oermann

    Importance: The transition from medical school to residency forms a national training network, yet its large-scale structure and implications for trainee outcomes remain poorly characterized. Objective: To evaluate the US residency match as a network and assess how institutional position relates to residency placement, educational debt, and specialty choice. Design: Cross-sectional analysis of publicly reported 2025 residency match outcomes. Setting: 107 US MD-granting medical schools and 301 residency institutions with available match data. Participants: 14,616 US MD students matching into residency in 2025 (convenience sample). Exposure: Institutional position within the residency match network, quantified using PageRank network centrality. The relative strength of each school’s graduating class was defined as the median centrality of residency destinations across graduates (placement score). Main Outcomes and Measures: Residency placement outcomes, mean medical school debt at graduation, and specialty choice (primary care vs surgical specialties) in relation to institutional position within the residency match network. Network-derived measures were also compared with NIH funding, residency reputation, and student selectivity. Results: Among 14,616 US MD students matched across 107 medical schools and 301 residency institutions (approximately 73.5% of total US MD cohort), network-derived measures of institutional influence closely aligned with benchmarks of institutional standing such as NIH funding, residency reputation, and student selectivity (Spearman ρ = 0.72–0.86; all p < .001). Graduate outcomes varied systematically across institutions. Graduates of highly connected medical schools were more likely to match into highly connected residency programs (87.3% for top-quintile vs 41.0% for bottom-quintile schools). Schools with higher placement scores had graduates with lower educational debt, reduced entry into primary care, and increased entry into surgical or competitive specialties. Compared with bottom-decile schools, top-decile schools (stratified by placement score) had 37% lower mean graduate debt, 24% lower primary care entry, and 75% higher surgical specialty entry. Higher educational debt was not associated with entry into higher-compensated specialties. Conclusions and Relevance: The residency match network reflects a hierarchical structure of institutional standing. Graduates of higher- and lower-positioned medical schools experience systematically different residency placement outcomes. These findings provide a population-level, behavior-based perspective on institutional influence and its relationship to training pathways. KEY POINTS: Question: How does institutional standing relate to residency placement, debt burden, and specialty choice? Findings: In this cross-sectional analysis of 14,616 US MD students entering residency, institutional influence derived from match data aligned with traditional metrics like NIH funding. Graduates of highly connected medical schools matched into highly connected residency programs. These well-connected schools also showed lower student debt, less primary care entry, and higher placement into surgical fields. Higher average debt was not associated with greater entry into higher-paying specialties. Meaning: The residency match operates as a hierarchy, linking a medical school’s position to career pathways and graduate outcomes.

  • The first step in visual diagnosis: a study of novices developing the ability to distinguish normal from abnormal cases

    by So-Young Oh, J. Burk-Rafel, I. Reinstein, R. Hatala, P. W. M. Van Gerven, F. W. J. M. Smeenk, M. V. Pusic


  • Top Cited Papers

    Macy Foundation Innovation Report Part II: From Hype to Reality: Innovators’ Visions for Navigating AI Integration Challenges in Medical Education

    by Brian C. Gin, Kate LaForge, Jesse Md Burk-Rafel, Christy K. Boscardin

    Purpose: Artificial intelligence (AI) promises to significantly impact medical education, yet its implementation raises important questions about educational effectiveness, ethical use, and equity. In the second part of a 2-part innovation report, which was commissioned by the Josiah Macy Jr. Foundation to inform discussions at a conference on AI in medical education, the authors explore the perspectives of innovators actively integrating AI into medical education, examining their perceptions regarding the impacts, opportunities, challenges, and strategies for successful AI adoption and risk mitigation. Method: Semistructured interviews were conducted with 25 medical education AI innovators—including learners, educators, institutional leaders, and industry representatives—from June to August 2024. Interviews explored participants’ perceptions of AI’s influence on medical education, challenges to integration, and strategies for mitigating challenges. Transcripts were analyzed using thematic analysis to identify themes and synthesize participants’ recommendations for AI integration. Results: Innovators’ responses were synthesized into 2 main thematic areas: (1) AI’s impact on teaching, learning, and assessment, and (2) perceived threats and strategies for mitigating them. Participants identified AI’s potential to enact precision education through virtual tutors and standardized patients, support active learning formats, enable centralized teaching, and facilitate cognitive offloading. AI-enhanced assessments could automate grading, predict learner trajectories, and integrate performance data from clinical interactions. Yet, innovators expressed concerns over threats to transparency and validity, potential propagation of biases, risks of over-reliance and deskilling, and institutional disparities. Proposed mitigation strategies emphasized validating AI outputs, establishing foundational competencies, fostering collaboration and open-source sharing, enhancing AI literacy, and maintaining robust ethical standards. Conclusions: AI innovators in medical education envision transformative opportunities for individualized learning and precision education, balanced against critical threats. Realizing these benefits requires proactive, collaborative efforts to establish rigorous validation frameworks; uphold foundational medical competencies; and prioritize ethical, equitable AI integration.


  • Large Language Model–Augmented Strategic Analysis of Innovation Projects in Graduate Medical Education

    by Abigail Ford Winkel, Jesse Burk-Rafel, Kyla Terhune, Brian T. Garibaldi, Ami L. DeWaters, John Patrick T. Co, John S. Andrews


  • How Data Analytics Can Be Leveraged to Enhance Graduate Clinical Skills Education

    by Brian T. Garibaldi, McKenzie Hollon, Michelle I. Knopp, Abigail Ford Winkel, Jesse Burk-Rafel, Holly A. Caretta-Weyer


  • Open Access

    Artificial intelligence based assessment of clinical reasoning documentation: an observational study of the impact of the clinical learning environment on resident documentation quality

    by Verity Schaye, David J. DiTullio, Daniel J. Sartori, Kevin Hauck, Matthew Haller, Ilan Reinstein, Benedict Guzman, Jesse Burk-Rafel

    Objective measures and large datasets are needed to determine aspects of the Clinical Learning Environment (CLE) impacting the essential skill of clinical reasoning documentation. Artificial Intelligence (AI) offers a solution. Here, the authors sought to determine what aspects of the CLE might be impacting resident clinical reasoning documentation quality assessed by AI. In this observational, retrospective cross-sectional analysis of hospital admission notes from the Electronic Health Record (EHR), all categorical internal medicine (IM) residents who wrote at least one admission note during the study period July 1, 2018– June 30, 2023 at two sites of NYU Grossman School of Medicine’s IM residency program were included. Clinical reasoning documentation quality of admission notes was determined to be low or high-quality using a supervised machine learning model. From note-level data, the shift (day or night) and note index within shift (if a note was first, second, etc. within shift) were calculated. These aspects of the CLE were included as potential markers of workload, which have been shown to have a strong relationship with resident performance. Patient data was also captured, including age, sex, Charlson Comorbidity Index, and primary diagnosis. The relationship between these variables and clinical reasoning documentation quality was analyzed using generalized estimating equations accounting for resident-level clustering. Across 37,750 notes authored by 474 residents, patients who were older, had more pre-existing comorbidities, and presented with certain primary diagnoses (e.g., infectious and pulmonary conditions) were associated with higher clinical reasoning documentation quality. When controlling for these and other patient factors, variables associated with clinical reasoning documentation quality included academic year (adjusted odds ratio, aOR, for high-quality: 1.10; 95% CI 1.06–1.15; P <.001), night shift (aOR 1.21; 95% CI 1.13–1.30; P <.001), and note index (aOR 0.93; 95% CI 0.90–0.95; P <.001). AI can be used to assess complex skills such as clinical reasoning in authentic clinical notes that can help elucidate the potential impact of the CLE on resident clinical reasoning documentation quality. Future work should explore residency program and systems interventions to optimize the CLE.


  • Open Access

    Large Language Model–Based Assessment of Clinical Reasoning Documentation in the Electronic Health Record Across Two Institutions: Development and Validation Study

    by Verity Schaye, David DiTullio, Benedict Vincent Guzman, Scott Vennemeyer, Hanniel Shih, Ilan Reinstein, Danielle E Weber, Abbie Goodman, Danny T Y Wu, Daniel J Sartori, et al.

    Background: Clinical reasoning (CR) is an essential skill; yet, physicians often receive limited feedback. Artificial intelligence holds promise to fill this gap. Objective: We report the development of named entity recognition (NER), logic-based and large language model (LLM)–based assessments of CR documentation in the electronic health record across 2 institutions (New York University Grossman School of Medicine [NYU] and University of Cincinnati College of Medicine [UC]). Methods: The note corpus consisted of internal medicine resident admission notes (retrospective set: July 2020-December 2021, n=700 NYU and 450 UC notes and prospective validation set: July 2023-December 2023, n=155 NYU and 92 UC notes). Clinicians rated CR documentation quality in each note using a previously validated tool (Revised-IDEA), on 3-point scales across 2 domains: differential diagnosis (D0, D1, and D2) and explanation of reasoning, (EA0, EA1, and EA2). At NYU, the retrospective set was annotated for NER for 5 entities (diagnosis, diagnostic category, prioritization of diagnosis language, data, and linkage terms). Models were developed using different artificial intelligence approaches, including NER, logic-based model: a large word vector model (scispaCy en_core_sci_lg) with model weights adjusted with backpropagation from annotations, developed at NYU with external validation at UC, NYUTron LLM: an NYU internal 110 million parameter LLM pretrained on 7.25 million clinical notes, only validated at NYU, and GatorTron LLM: an open source 345 million parameter LLM pretrained on 82 billion words of clinical text, fined tuned on NYU retrospective sets, then externally validated and further fine-tuned at UC. Model performance was assessed in the prospective sets with F1-scores for the NER, logic-based model and area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) for the LLMs. Results: At NYU, the NYUTron LLM performed best: the D0 and D2 models had AUROC/AUPRC 0.87/0.79 and 0.89/0.86, respectively. The D1, EA0, and EA1 models had insufficient performance for implementation (AUROC range 0.57-0.80, AUPRC range 0.33-0.63). For the D1 classification, the approach pivoted to a stepwise approach taking advantage of the more performant D0 and D2 models. For the EA model, the approach pivoted to a binary EA2 model (ie, EA2 vs not EA2) with excellent performance, AUROC/AUPRC 0.85/ 0.80. At UC, the NER, D-logic–based model was the best performing D model (F1-scores 0.80, 0.74, and 0.80 for D0, D1, D2, respectively. The GatorTron LLM performed best for EA2 scores AUROC/AUPRC 0.75/ 0.69. Conclusions: This is the first multi-institutional study to apply LLMs for assessing CR documentation in the electronic health record. Such tools can enhance feedback on CR. Lessons learned by implementing these models at distinct institutions support the generalizability of this approach.