Analysis of ChatGPT Responses to Ophthalmic Cases: Can ChatGPT Think like an Ophthalmologist?

Objective: Large language models such as ChatGPT have demonstrated significant potential in question-answering within ophthalmology, but there is a paucity of literature evaluating its ability to generate clinical assessments and discussions. The objectives of this study were to (1) assess the accuracy of assessment and plans generated by ChatGPT and (2) evaluate ophthalmologists' abilities to distinguish between responses generated by clinicians versus ChatGPT. Design: Cross-sectional mixed-methods study. Subjects: Sixteen ophthalmologists from a single academic center, of which 10 were board-eligible and 6 were board-certified, were recruited to participate in this study. Methods: Prompt engineering was used to ensure ChatGPT output discussions in the style of the ophthalmologist author of the Medical College of Wisconsin Ophthalmic Case Studies. Cases where ChatGPT accurately identified the primary diagnoses were included and then paired. Masked human-generated and ChatGPT-generated discussions were sent to participating ophthalmologists to identify the author of the discussions. Response confidence was assessed using a 5-point Likert scale score, and subjective feedback was manually reviewed. Main Outcome Measures: Accuracy of ophthalmologist identification of discussion author, as well as subjective perceptions of human-generated versus ChatGPT-generated discussions. Results: Overall, ChatGPT correctly identified the primary diagnosis in 15 of 17 (88.2%) cases. Two cases were excluded from the paired comparison due to hallucinations or fabrications of nonuser-provided data. Ophthalmologists correctly identified the author in 77.9% ± 26.6% of the 13 included cases, with a mean Likert scale confidence rating of 3.6 ± 1.0. No significant differences in performance or confidence were found between board-certified and board-eligible ophthalmologists. Subjectively, ophthalmologists found that discussions written by ChatGPT tended to have more generic responses, irrelevant information, hallucinated more frequently, and had distinct syntactic patterns (all P < 0.01). Conclusions: Large language models have the potential to synthesize clinical data and generate ophthalmic discussions. While these findings have exciting implications for artificial intelligence-assisted health care delivery, more rigorous real-world evaluation of these models is necessary before clinical deployment. Financial Disclosures: The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Você está pensando, refletindo, analisando o que você vê e o que você faz o tempo todo: percepção de residentes de Medicina de Família e Comunidade sobre vídeo feedback / You are thinking, reflecting, analysing what you see and what you do all the time: perception of General Practice trainees about video feedback / Estás pensando, reflexionando, analizando lo que ves y lo que haces todo el tiempo: percepción de los residentes de Medicina Familiar y Comunitaria sobre vídeo feedback

Introdução: A comunicação é reconhecida como uma habilidade central por vários órgãos reguladores internacionais da educação médica. O ensino específico de habilidades de comunicação é fundamental para melhorar a comunicação dos médicos. As técnicas experienciais mostraram superioridade em comparação com os modelos tradicionais. A utilização de consultas reais ajuda os estudantes a visualizar melhor as suas competências de entrevista e a refletir sobre elas. Com os avanços da tecnologia, o uso de consultas médicas gravadas em vídeo tornou-se a abordagem padrão para o ensino da comunicação. No entanto, a eficácia dessa técnica depende do envolvimento ativo dos estudantes. As suas contribuições e comentários dos pares sobre a consulta gravada são essenciais para a aprendizagem. Contudo, a perspectiva do estudante sobre a utilidade dessa abordagem educativa recebeu pouca atenção. Objetivos: Compreender a percepção da aprendizagem dos residentes de medicina de família e comunidade resultante da atividade de vídeo feedback na sua formação profissional. Métodos: Estudo exploratório, qualitativo, realizado com residentes do primeiro ano de medicina de família e comunidade de um programa de residência estabelecido em São Paulo, Brasil. Os participantes foram entrevistados após as sessões educativas, que foram analisadas por meio de análise temática reflexiva. Resultados: A autopercepção de sua prática, o aprendizado de habilidades de comunicação e os ganhos afetivos foram identificados pelos participantes como pontos de aprendizado derivados da atividade de vídeo feedback. Além disso, sobre o aprendizado de habilidades específicas de comunicação, eles mencionaram comunicação não-verbal e verbal, conexões entre teoria e prática, estrutura de consulta e oportunidades para cristalizar conhecimentos. Os ganhos afetivos incluíram sentir-se parte de um grupo, melhora da autoestima, superação de inseguranças, percepção de consultas mais efetivas, reforço do gosto pelo trabalho e reconhecer a necessidade de mais aprendizado. Conclusões: Os ganhos de aprendizagem identificados em nosso estudo levaram a uma experiência de humanidade compartilhada, que permite aos participantes serem mais efetivos técnica e afetivamente com seus pacientes. Além disso, identificamos que a atividade educativa de vídeo feedback pode ser utilizada para outros possíveis fins educacionais além do ensino da comunicação.

Introduction: Communication is recognized as a central skill by various international medical education regulatory bodies. Specific teaching on communication skills is important to enhance doctors' communication. Experiential techniques appear to be superior compared to traditional models. Real-life consultation helps trainees visualize their interview skills and reflect on them. Upgraded by technology, the use of video-recorded medical visits became the standard approach for communication teaching. However, the effectiveness pf this technique relies on trainees' active involvement. Their inputs and peer feedback on the recorded consultation are essential to learning. Despite its importance, their perspective on the usefulness of video feedback in medical education has received limited attention. Objective: To understand the perception of learning among general practice trainees as a result of the video feedback activity in their vocational training. Methods: An exploratory, qualitative study, conducted with first-year general practice trainees from an established training program in São Paulo, Brazil. Participants were interviewed after educational session, which were analyzed using reflexive thematic analysis. Results: Self-perception of their practice, communication skills learning, and affective gains were identified by participants as learning points derived from the video feedback activity. Furthermore, for specific communication skills learning, they mentioned nonverbal and verbal communication, theory and practice connections, consultation structure and opportunities for crystallizing knowledge. Affective gains included feeling part of a group, improving self-esteem, overcoming insecurities, perception of more effective consultations, reinforcing fondness for their work, and need for more learning. Conclusions: The learning gains identified in our study led to an experience of common humanity, which allowed participants to be more technically and affectively effective with their patients. Also, we identified that the video feedback educational activity can be used for other possible educational purposes, beyond the teaching of communication.

Introducción: La comunicación es reconocida como una habilidad fundamental por varios organismos reguladores internacionales de educación médica. La enseñanza específica de habilidades de comunicación es importante para mejorar la comunicación de los médicos. Las técnicas experienciales parecen ser superiores a los modelos tradicionales. El uso de consultas reales ayuda a los estudiantes a visualizar y reflexionar mejor sobre sus habilidades de entrevista. Actualizado por la tecnología, el uso de consultas médicas grabadas en video se ha convertido en el enfoque estándar para la enseñanza de la comunicación. Sin embargo, para que la técnica funcione, la participación de los estudiantes es crucial. Sus contribuciones y comentarios de los compañeros sobre la consulta grabada son esenciales para el aprendizaje. Sin embargo, la perspectiva de los estudiantes sobre la utilidad de este enfoque educativo ha recibido poca atención. Objetivos: Comprender la percepción del aprendizaje por parte de los residentes de medicina de familia y comunitaria como resultado de la actividad de vídeo feedback en su formación profesional. Métodos: Estudio cualitativo exploratorio realizado con residentes de primer año de medicina familiar y comunitaria de un programa de residencia establecido en São Paulo, Brasil. Los participantes fueron entrevistados después de una sesión educativa, que fueron analizados mediante análisis temático reflexivo. Resultados: La autopercepción de su práctica, el aprendizaje de habilidades comunicativas y las ganancias afectivas fueron identificadas por los participantes como puntos de aprendizaje derivados de la actividad de vídeo feedback. Además, sobre el aprendizaje de habilidades comunicativas específicas, mencionaron la comunicación verbal y no verbal, las conexiones entre la teoría y la práctica, la estructura de consulta y las oportunidades para cristalizar conocimientos. En cuanto a las ganancias afectivas, relataron sentirse parte de un grupo, mejora de la autoestima, superación de las inseguridades, percepción de consultas más efectivas, refuerzo del gusto por el trabajo y necesidad de más aprendizaje. Conclusión: Los logros de aprendizaje identificados en nuestro estudio llevaron a una experiencia de humanidad compartida, que permite a los participantes ser técnica y afectivamente más efectivos con sus pacientes. Además, identificamos que la actividad educativa de vídeo feedback puede ser utilizada para otros posibles fines educativos, además de la enseñanza de la comunicación.

Aprendizagem baseada em projeto e a formação médica: relato de experiência introduzindo questões para o debate interprofissional / Project-based learning and medical education: experience report introducing issues for interprofessional debate / El aprendizaje basado en proyecto y la formación médica: informe de experiencia que introduce cuestiones para el debate interprofesional

A aprendizagem baseada em projeto orientada pelos fundamentos da educação interprofissional é um modelo que pode contribuir para a formação de relacionamentos interpessoais, criatividade, empatia e colaboração na educação médica, por meio de uma colaboração mútua com profissionais de saúde da rede. Muito se fala da efetividade desse método no campo do ensino e aprendizagem médica, mas há a necessidade de incluir a importância do desenvolvimento de habilidades interprofissionais, com equipes colaborativas, em ações extensionistas, diante das necessidades locais no contexto da atenção primária, pensando na melhoria dos resultados de saúde. O objetivo deste trabalho é apresentar um relato de experiência de aprendizagem baseada em projeto de estudantes de Medicina no contexto da Estratégia Saúde da Família. Participaram deste trabalho estudantes do Módulo Integração Ensino, Serviço e Comunidade da Faculdade de Medicina da Universidade Federal dos Vales do Jequitinhonha e Mucuri que executaram, em colaboração com uma equipe interprofissional o projeto sobre a saúde do homem. Como resultado da análise qualitativa do feedback entre os integrantes, observaram-se mudanças no comportamento dos estudantes, com melhorias na comunicação, empatia e nas relações interpessoais, por meio do trabalho colaborativo com a equipe interprofissional. Esta experiência poderá ser adaptada para implementar o ensino e aprendizagem no projeto pedagógico orientado pela educação interprofissional na atenção primária.

Project-based learning guided by the fundamentals of interprofessional education is a model that can contribute to the formation of interpersonal relationships, creativity, empathy and collaboration within medical education, through mutual collaboration with health professionals in the health network. Much has been said about the effectiveness of this method in medical teaching and learning, but there is a need to include the importance of developing interprofessional skills, with collaborative teams, within extension actions, in view of local needs in the context of primary care, thinking about the improved health outcomes. The objective of this work was to present a report of a project-based learning experience of medical students in Family Health Strategy. Students from the Teaching, Service and Community Integration Module of the Faculty of Medicine of Universidade Federal dos Vales do Jequitinhonha e Mucuri participated in this work, executing in collaboration with an interprofessional team a project about men's health. As a result of the qualitative analysis of the feedback among the members, changes in student behavior were observed with improvements in communication, empathy and interpersonal relationships through collaborative work with the interprofessional team. This experience can be adapted to implement teaching and learning in the pedagogical project guided by interprofessional education in primary care.

El aprendizaje basado en proyectos y guiado por los fundamentos de la educación interprofesional es un modelo que puede contribuir a la formación de relaciones interpersonales, creatividad, empatía y colaboración dentro de la educación médica, a través de la colaboración mutua con los profesionales de la salud en la red de salud. Mucho se habla de la efectividad de este método dentro de la enseñanza y el aprendizaje médico, pero es necesario incluir la importancia del desarrollo de habilidades interprofesionales, con equipos colaborativos, dentro de las acciones de extensión, frente a las necesidades locales en el contexto de la atención primaria, pensando sobre los mejores resultados de salud. El objetivo de este trabajo es presentar un informe de experiencia de aprendizaje basado en proyectos de estudiantes de medicina en la Estrategia de Salud Familiar. Participaron en este trabajo estudiantes del Módulo Integración Enseñanza, Servicio y Comunidad de la Facultad de Medicina de la Universidade Federal dos Vales do Jequitinhonha e Mucuri que ejecutaron en colaboración con un equipo interprofesional el proyecto sobre la salud del hombre. Como resultado del análisis cualitativo de la retroalimentación entre los integrantes, se observaron cambios en el comportamiento de los estudiantes con mejoras en la comunicación, la empatía y las relaciones interpersonales a través del trabajo colaborativo con el equipo interprofesional. Esta experiencia puede adaptarse para implementar la enseñanza y el aprendizaje en el proyecto pedagógico guiado por la educación interprofesional en atención primaria.

Medical faculty's perception of students' mental health: a qualitative survey study / Percepción de los profesores de medicina sobre la salud mental de los estudiantes: un estudio cualitativo / Percepção dos docentes de medicina sobre a saúde mental dos estudantes: um estudo qualitativo

This study sought to explore the perspective of medical faculty on the mental health of their students. This qualitative study based on a focus group is part of a longitudinal research that studied the mental health of Brazilian students. One group was conducted with faculty employed at a medical school. Topics discussed covered the concept of mental health and medical education. Six professors participated in one group. The mental health of medical students is a construct that encompasses emotional aspects, ability to solve problems and multiple facets of a human being, according to the participants. Artistic practices, moments of socialization and leisure were perceived as stimulating students' good mental health. Excessive demands generate competitiveness and the teacher's expectation of the student's good performance based on their own experience can harm the student's mental health. Participants also highlighted that a pedagogical reformulation that makes sense for the student's learning process is necessary to update traditional curricula. Medical students' mental health is influenced by experiences and exchanges during the medical school, mainly between professor and student, understood as necessary and inherent to the process of becoming physician. The findings of this study show the need for curriculum changes in the medical education process and updating teacher training for good practices that reinforce good mental health.

Este estudio buscó explorar la perspectiva de los profesores de medicina sobre la salud mental de sus estudiantes. Este estudio cualitativo basado en un grupo focal es parte de una investigación longitudinal que estudió la salud mental de estudiantes brasileños. Un grupo se llevó a cabo con profesores empleados en una escuela de medicina. Los temas tratados abarcaron el concepto de salud mental y educación médica. Seis docentes participaron en un grupo. La salud mental de los estudiantes de medicina es un constructo que abarca aspectos emocionales, capacidad de resolución de problemas y múltiples facetas del ser humano, según los participantes. Las prácticas artísticas, los momentos de socialización y el ocio fueron percibidos como estimulantes de la buena salud mental de los estudiantes. Las exigencias excesivas generan competitividad y la expectativa del docente sobre el buen desempeño del estudiante basándose en su propia experiencia puede perjudicar la salud mental del estudiante. Los participantes también resaltaron que es necesaria una reformulación pedagógica que tenga sentido para el proceso de aprendizaje del estudiante para actualizar los currículos tradicionales. La salud mental de los estudiantes de medicina está influenciada por las experiencias y los intercambios durante la carrera de medicina, principalmente entre profesor y estudiante, entendidos como necesarios e inherentes al proceso de convertirse en médico. Los hallazgos de este estudio muestran la necesidad de cambios curriculares en el proceso de formación médica y de actualización de la formación docente hacia buenas prácticas que refuercen la buena salud mental.

Este estudo buscou explorar a perspectiva dos docentes de medicina sobre a saúde mental de seus alunos. Este estudo qualitativo baseado em grupo focal faz parte de uma pesquisa longitudinal que estudou a saúde mental de estudantes brasileiros. Um grupo foi conduzido com professores empregados em uma faculdade de medicina. Os temas discutidos abrangeram o conceito de saúde mental e educação médica. Seis professores participaram de um grupo. A saúde mental dos estudantes de medicina é um construto que engloba aspectos emocionais, capacidade de resolução de problemas e múltiplas facetas do ser humano, segundo os participantes. As práticas artísticas, os momentos de socialização e de lazer foram percebidos como estimuladores da boa saúde mental dos estudantes. Exigências excessivas geram competitividade e a expectativa do professor pelo bom desempenho do aluno com base na própria experiência pode prejudicar a saúde mental do aluno. Os participantes destacaram também que é necessária uma reformulação pedagógica que faça sentido para o processo de aprendizagem do aluno para atualizar os currículos tradicionais. A saúde mental dos estudantes de medicina é influenciada pelas experiências e trocas durante o curso de medicina, principalmente entre professor e aluno, entendidas como necessárias e inerentes ao processo de tornar-se médico. Os achados deste estudo mostram a necessidade de mudanças curriculares no processo de formação médica e de atualização da formação docente para boas práticas que reforcem a boa saúde mental.

Leadership Development and Its Association With Retention of Surgical Versus Nonsurgical Faculty at a Large Academic Health Center.

INTRODUCTION: Leadership development programs are the key for engaging physicians as partners in health-care administration. These programs have become common; however, one potential consequence of leadership training is attrition. METHODS: Using a prospectively collected database, Kaplan-Meier curves were used to analyze attrition of physician faculty following an intramural leadership program at an academic health center. RESULTS: Six cohorts totaling 165 faculty successfully completed our leadership program. Sixty-one faculty (37%) were from surgical departments (Group A) and 104 (63%) from nonsurgical departments (Group B). Thirty-six program graduates (21.8%) departed the institution at a mean follow-up of 68 mo for an annualized attrition rate of 3.6%. Attrition rates for Group B were significantly greater than for Group A (P = 0.05). Most faculty left for career advancement opportunities elsewhere. CONCLUSIONS: Our annualized attrition rate for leadership trained faculty was 3.6%, well below the attrition rate of our academic health center (5.3%) over a similar time period, and the 2018 national physician attrition rate of 7.8%. Strategies to improve retention are discussed.

The impact of intersectionality on the experience of medical trainees.

Diverse trainees have different training experiences than non-diverse trainees; however, it has not been demonstrated if the number of diverse traits (DTs) (i.e. intersectionality) contributes to worse experiences. Our objective was to understand if the number of DTs a trainee has impacts their training experience. We distributed a survey to medical trainees at Mass General Brigham from 15/12/23 to 15/4/24. DTs were based on race/ethnicity, gender-identity, sexual orientation, birthplace, disability, and highest parental education. A total of 134 responses were obtained. Respondents with 1 DT were more likely to report dissatisfaction with the quality of their training compared to those with 0 DT (P = 0.03). However, this did not remain true for those with 2 or 3+ DT (P = 0.39, P = 0.59). Only respondents with 1 DT disagreed that they had similar opportunities for academic success to those of their peers (P = 0.03) and agreed that they had to work harder than others to be valued equally (P < 0.01). Respondents with 3+ DTs felt that their institution was not diverse (P < 0.01) and racist (P < 0.01) compared to respondents with zero DTs. Therefore, trainees with DTs had different training experiences than trainees with zero DTs, but the number of DTs did not consistently correlate with the quality of their experience. We believe this data is important for administrators to understand how intersectionality effects the trainee experience. Additionally, we believe this data can be used to gauge disparities in education and create an avenue to address issues, such as discrimination, microaggressions, and low retention rate of diverse applicants.

Two-Dimensional Deaths? A Discourse Analysis of Patient Death in Preclinical Tutorial Cases at a Canadian Medical School.

Introduction: The prospect of death is everywhere, but seldom directly addressed, in undergraduate medical education (UGME). Despite calls for UGME curricula to address the complex social and emotional aspects of death and dying, most curricula focus on biomedical, legal, and logistical aspects, or concentrate these topics within palliative care content and/or in simulations with simulated patients and manikins. We aimed to add to death education scholarship by exploring the complexities of death and dying within two dimensional simulations-i.e., in the text-based cases used in Case-Informed-Learning (CIL). Method: We conducted a critical discourse analysis exploring how death and dying were discursively constructed in the formal, planned curriculum at one medical school. We used two methods: (1) Document Analysis: We developed a template to analyze 127 cases regarding their discursive constructions of death and dying; (2) Longitudinal Interviewing: We conducted semi-structured interviews with a cohort of 12 medical students, twice annually throughout their medical program (total 92 interviews). We collectively analyzed data, attuning to how the format, content, and purpose of each case discursively constructed death and dying. Results: There were 127 tutorial cases included in the undergraduate, pre-clerkship case-informed curriculum. In the five (4%) cases featuring a patient who dies, death and dying were discursively constructed as: (1) predictable; (2) a plot device; (3) a cautionary tale; (4) an epilogue; (5) deliberate and careful; and (6) an absence. Very few cases highlighted death and dying in their titles, learning objectives, or questions, and where it did feature, it was framed a biomedical fact or outcome. Only one case allowed for a nuanced, in-depth and open-ended discussion of patient death and dying, but it was scheduled at a time that prevented meaningful engagement. This glossing over the complexities of death was identified as a missed opportunity by students, who, as their clinical placements loomed, were eager to broach this topic in detail with tutors and other teaching faculty. Discussion: Death was often a conspicuous absence in this CIL curriculum. In the few cases that featured the death of the main patient character, multiple discourses were mobilized that worked together to construct death as something that happens elsewhere, outside the parameters of core curriculum. In other words, death happens-predictably, slowly, as a means to an end and the result of moral failures, in the case or somewhere in the future-but was not the primary concern. To deepen engagement with these subjects in CIL, we encourage medical educators to attend to representations of patient death by considering the format, content, purpose, and timing of these cases. Conclusion: Carefully rendered cases thoughtfully embedded in the curriculum offer tremendous potential. We suggest nuanced cases featuring patient death, with plenty of space and time for discussion, reflection, and storytelling may help address gaps in formal UGME preclinical curricula addressing death and dying.

The Use of Silicone Simulators for Bile Duct Anastomosis Education in Medical Conferences for the Purpose of Improving the Canadian Medical Education Directives for Specialists (CanMEDS) Competencies.

This technical report explores the potential of including silicone bile duct simulators for the purpose of completing a bile duct anastomosis (BDA) in medical conferences. The purpose is to target the need for exposure to more surgical skills and to contribute to the Canadian Medical Education Directives for Specialists (CanMEDS) requirements, as per the Royal College of Physicians and Surgeons of Canada.  Data collection was completed at the 2023 Canadian Conference for the Advancement of Surgical Education (C-CASE) in Montreal, Canada. For several years, the quality improvement feedback received at the end of these conferences suggested a few areas of improvement, one of which was related to the concept of return on investment (ROI). The participants spend a considerable amount of funds to travel to the conferences but feel that the only measurable gains are at a research capacity and thus only relate to two CanMEDS competencies. By leveraging C-CASE, the aim is to enhance students' educational experience during events they already intend to attend. Initially, students participated in a five-part simulation workshop and engaged in a think-aloud protocol (TAO). From there, nine participants were recruited for a focus group to further understand the perceived educational value and feedback on both the simulators and the conference structure.

History and Social Implications of the Pulse Oximeter.

The pulse oximeter is a portable, bedside tool that allows for the measurement of oxygen saturation in a patient's red blood cells. The technology is based on oxygenated and deoxygenated hemoglobin absorbing light at different wavelengths. The device calculates the ratio of oxygenated to deoxygenated hemoglobin in the blood, and an algorithm produces a percentage oxygen saturation value. Due to its portability and ease of use, it is a ubiquitous medical tool that is commonly used in medical practice. This paper reviews the history and evolution of this tool, and the scientific laws behind oximetry. It also introduces the importance of the pulse oximeter and its basic functions. In addition, the limitations of pulse oximetry are discussed, especially as they pertain to pigmented skin.

Bridging the Telehealth Digital Divide With Collegiate Navigators: Mixed Methods Evaluation Study of a Service-Learning Health Disparities Course.

BACKGROUND: Limited digital literacy is a barrier for vulnerable patients accessing health care. OBJECTIVE: The Stanford Technology Access Resource Team (START), a service-learning course created to bridge the telehealth digital divide, trained undergraduate and graduate students to provide hands-on patient support to improve access to electronic medical records (EMRs) and video visits while learning about social determinants of health. METHODS: START students reached out to 1185 patients (n=711, 60% from primary care clinics of a large academic medical center and n=474, 40% from a federally qualified health center). Registries consisted of patients without an EMR account (at primary care clinics) or patients with a scheduled telehealth visit (at a federally qualified health center). Patient outcomes were evaluated by successful EMR enrollments and video visit setups. Student outcomes were assessed by reflections coded for thematic content. RESULTS: Over 6 academic quarters, 57 students reached out to 1185 registry patients. Of the 229 patients contacted, 141 desired technical support. START students successfully established EMR accounts and set up video visits for 78.7% (111/141) of patients. After program completion, we reached out to 13.5% (19/141) of patients to collect perspectives on program utility. The majority (18/19, 94.7%) reported that START students were helpful, and 73.7% (14/19) reported that they had successfully connected with their health care provider in a digital visit. Inability to establish access included a lack of Wi-Fi or device access, the absence of an interpreter, and a disability that precluded the use of video visits. Qualitative analysis of student reflections showed an impact on future career goals and improved awareness of health disparities of technology access. CONCLUSIONS: Of the patients who desired telehealth access, START improved access for 78.7% (111/141) of patients. Students found that START broadened their understanding of health disparities and social determinants of health and influenced their future career goals.

Optimizing ChatGPT's Interpretation and Reporting of Delirium Assessment Outcomes: Exploratory Study.

BACKGROUND: Generative artificial intelligence (AI) and large language models, such as OpenAI's ChatGPT, have shown promising potential in supporting medical education and clinical decision-making, given their vast knowledge base and natural language processing capabilities. As a general purpose AI system, ChatGPT can complete a wide range of tasks, including differential diagnosis without additional training. However, the specific application of ChatGPT in learning and applying a series of specialized, context-specific tasks mimicking the workflow of a human assessor, such as administering a standardized assessment questionnaire, followed by inputting assessment results in a standardized form, and interpretating assessment results strictly following credible, published scoring criteria, have not been thoroughly studied. OBJECTIVE: This exploratory study aims to evaluate and optimize ChatGPT's capabilities in administering and interpreting the Sour Seven Questionnaire, an informant-based delirium assessment tool. Specifically, the objectives were to train ChatGPT-3.5 and ChatGPT-4 to understand and correctly apply the Sour Seven Questionnaire to clinical vignettes using prompt engineering, assess the performance of these AI models in identifying and scoring delirium symptoms against scores from human experts, and refine and enhance the models' interpretation and reporting accuracy through iterative prompt optimization. METHODS: We used prompt engineering to train ChatGPT-3.5 and ChatGPT-4 models on the Sour Seven Questionnaire, a tool for assessing delirium through caregiver input. Prompt engineering is a methodology used to enhance the AI's processing of inputs by meticulously structuring the prompts to improve accuracy and consistency in outputs. In this study, prompt engineering involved creating specific, structured commands that guided the AI models in understanding and applying the assessment tool's criteria accurately to clinical vignettes. This approach also included designing prompts to explicitly instruct the AI on how to format its responses, ensuring they were consistent with clinical documentation standards. RESULTS: Both ChatGPT models demonstrated promising proficiency in applying the Sour Seven Questionnaire to the vignettes, despite initial inconsistencies and errors. Performance notably improved through iterative prompt engineering, enhancing the models' capacity to detect delirium symptoms and assign scores. Prompt optimizations included adjusting the scoring methodology to accept only definitive "Yes" or "No" responses, revising the evaluation prompt to mandate responses in a tabular format, and guiding the models to adhere to the 2 recommended actions specified in the Sour Seven Questionnaire. CONCLUSIONS: Our findings provide preliminary evidence supporting the potential utility of AI models such as ChatGPT in administering standardized clinical assessment tools. The results highlight the significance of context-specific training and prompt engineering in harnessing the full potential of these AI models for health care applications. Despite the encouraging results, broader generalizability and further validation in real-world settings warrant additional research.

Beyond Morbidity and Mortality Conference: How Do We Learn From Special Cases?

OBJECTIVE: Physicians experience scary cases in the course of usual medical practice. Cases of near misses, legal and ethical dilemmas, or unique clinical challenges are great sources of education. However, there is no format for presentation and dissemination of cases that do not meet criteria for morbidity and mortality (M&M) conferences. The Scary Cases Conference is an innovative educational forum for scrutiny and analysis of these challenging clinical cases. Scary Cases differs from traditional Morbidity and Mortality conferences as it explores outcomes beyond the scope of medical or surgical errors. METHODS: From 2011 to 2021: 11 regional and 10 national "Otolaryngology Scary Cases" conferences and mini-seminars were held. The cases presented were analyzed for case specialty, area of management deemed troublesome, and compared to M&M conference criteria. RESULTS: 187 cases were presented. 62% percent of cases included traditional medical problems, whereas 21% involved legal issues, and 17% focused on ethical dilemmas. For the cases with medical problems, 31% involved airway obstruction, 17% nerve injuries, and 17% malignancy. 49% of cases would have met criteria for presentation at traditional M&M conferences. Of all the "scary cases," 25% were near misses and 26% represented ethical or legal dilemmas which would not be classified as morbitidy, mortality, or near miss. CONCLUSIONS: The Scary Cases provides a forum conducive to learning amongst peers and experts. It allows medical and surgical specialties to share the cases deemed most impactful. The M&M conference would only include half of such cases, but could be expanded beyond the traditional scope in the future.

Fine-Tuning Large Language Models to Enhance Programmatic Assessment in Graduate Medical Education.

Background: Natural language processing is a collection of techniques designed to empower computer systems to comprehend and/or produce human language. The purpose of this investigation was to train several large language models (LLMs) to explore the tradeoff between model complexity and performance while classifying narrative feedback on trainees into the Accreditation Council for Graduate Medical Education subcompetencies. We hypothesized that classification accuracy would increase with model complexity. Methods: The authors fine-tuned several transformer-based LLMs (Bidirectional Encoder Representations from Transformers [BERT]-base, BERT-medium, BERT-small, BERT-mini, BERT-tiny, and SciBERT) to predict Accreditation Council for Graduate Medical Education subcompetencies on a curated dataset of 10 218 feedback comments. Performance was compared with the authors' previous work, which trained a FastText model on the same dataset. Performance metrics included F1 score for global model performance and area under the receiver operating characteristic curve for each competency. Results: No models were superior to FastText. Only BERT-tiny performed worse than FastText. The smallest model with comparable performance to FastText, BERT-mini, was 94% smaller. Area under the receiver operating characteristic curve for each competency was similar on BERT-mini and FastText with the exceptions of Patient Care 7 (Situational Awareness and Crisis Management) and Systems-Based Practice. Discussion: Transformer-based LLMs were fine-tuned to understand anesthesiology graduate medical education language. Complex LLMs did not outperform FastText. However, equivalent performance was achieved with a model that was 94% smaller, which may allow model deployment on personal devices to enhance speed and data privacy. This work advances our understanding of best practices when integrating LLMs into graduate medical education.

Efficacy of medical education combined with extended care on adherence to inhaled glucocorticoids and clinical effects in patients with bronchial asthma.

OBJECTIVE: To observe the clinical efficacy of medical education combined with extended care in patients with bronchial asthma and its effect on adherence to inhaled glucocorticoids. METHODS: Ninety-eight patients with bronchial asthma were divided into the control group and the experimental group, n = 49, by utilizing the random number table method. The control group was given routine education and care as well as routine out-of-hospital instructions, and the experimental group was given medical education and extended care based on the control group. Asthma disease knowledge mastery, asthma control, quality of life, medication adherence and lung function were compared between both groups, and the number of asthma attacks and re-hospitalizations were recorded. RESULTS: The experimental group performed higher scores of health knowledge, asthma control test and quality of life, rate of complete adherence, forced expiratory volume in one second (FEV1), peak expiratory flow rate, and FEV1/forced vital capacity. The number of asthma attacks and the times of re-hospitalizations were lower in the experimental group (all P < 0.05). CONCLUSION: Medical education combined with extended care can improve bronchial asthma patients' mastery of asthma disease knowledge, effectively control patients' conditions, enhance patients' quality of life and lung function, increase patients' adherence to inhaled glucocorticoids, and reduce the recurrence of bronchial asthma patients.

"Lives and times": The case for qualitative longitudinal research in anatomical sciences education.

Qualitative longitudinal research (QLR) focuses on changes in perceptions, interpretations, or practices through time. Despite longstanding traditions in social science, QLR has only recently appeared in anatomical sciences education (ASE). While some existing methodology papers guide researchers, they take a narrow view of QLR and lack specificity for ASE. This discursive article aims to (1) describe what QLR is and its benefits, its philosophies, methodologies and methods, considerations, and quality indicators, and (2) critically discuss examples of QLR in ASE. Underpinned by relativist ontology and subjectivist epistemology, time can be understood as fluid/subjective or fixed/objective. QLR is a flexible, creative, and exploratory methodology, often associated with other methodologies. Sampling is typically purposive, with repeated and recursive data collection methods, and complex three-strand analyses (themes, cases, and time), enabling cross-sectional and longitudinal analyses. QLR involves ethical, relationship, analytical, dissemination, and funding considerations. Key quality indicators relate to qualitative research as well as temporal aspects. Most of the nine ASE papers reviewed explored changes in anatomy learners, but few labeled their methodology as QLR. Just under half described their sampling as purposive, most employed pre-planned and standardized repeated interviews, analyzed their data cross-sectionally, and utilized qualitative data analysis software. Most cited the confirmability and transferability of their studies, but few cited credibility and dependability elements. Study timeframes and tempos were generally clear, but details of longitudinal retention/attrition were often lacking, and longitudinal data analysis was not often conducted. We therefore provide recommendations for the conduct of QLR in ASE.

Efficacy of large language models and their potential in Obstetrics and Gynecology education.

Objective: The performance of large language models (LLMs) and their potential utility in obstetric and gynecological education are topics of ongoing debate. This study aimed to contribute to this discussion by examining the recent advancements in LLM technology and their transformative potential in artificial intelligence. Methods: This study assessed the performance of generative pre-trained transformer (GPT)-3.5 and -4 in understanding clinical information, as well as its potential implications for obstetric and gynecological education. Obstetrics and gynecology residents at three hospitals underwent an annual promotional examination, from which 116 of the 170 questions over 4 years (2020-2023) were analyzed, excluding 54 questions with images. The scores achieved by GPT-3.5, -4, and the 100 residents were compared. Results: The average scores across all 4 years for GPT-3.5 and -4 were 38.79 (standard deviation [SD], 5.65) and 79.31 (SD, 3.67), respectively. For groups R1, R2, and R3, the cumulative annual average scores were 79.12 (SD, 9.00), 80.95 (SD, 5.86), and 83.60 (SD, 6.82), respectively. No statistically significant differences were observed between the scores of GPT-4.0 and those of the residents. When analyzing questions specific to obstetrics, the average scores for GPT-3.5 and -4.0 were 33.44 (SD, 10.18) and 90.22 (SD, 7.68), respectively. Conclusion: GPT-4 demonstrated exceptional performance in obstetrics, different types of data interpretation, and problem solving, showcasing the potential utility of LLMs in these areas. However, acknowledging the constraints of LLMs is crucial and their utilization should augment human expertise and discernment.