Anoop Agrawal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Anoop Agrawal has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Ken Masters
Medical Education and Informatics Department, Sultan Qaboos University
7 shared publications - Neil Mehta
Cleveland Clinic Lerner College of Medicine of Case Western Reserve University
7 shared publications - Heather MacNeill
Associate Professor, University of Toronto, Toronto Metropolitan University
6 shared publications - Jennifer Benjamin
Department of Pediatrics, Huffington Department of Education Innovation and Technology, Baylor College of Medicine
6 shared publications - Tyson Pillow
Baylor college of Medicine, Houston, USA
3 shared publications - Sofia Valanci-Aroesty
Royal College of Physicians and Surgeons of Canada
2 shared publications
Latest Publications
- Open Access
- 30 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Promise or Peril: Evaluating the Performance of OpenEvidence in Clinical Decision Making in the Global Context. (Preprint)
BACKGROUND Clinicians face the impossible task of keeping up with evidence-based literature to inform clinical decision making. OpenEvidence holds great promise by providing latest evidence to inform patient care.AI solutions are increasingly used to support physicians during clinical decision making. Clinicians need to critically analyze and evaluate outputs and be aware of biases that exist prior to application, especially when caring for patients in global contexts. OBJECTIVE To evaluate performance of OpenEvidence (OE) in global context. METHODS In this study a sample of 11 clinicians from the United States (n=3) and six other countries (n=8) performed standardized searches for 10 common clinical conditions. Search results were evaluated using the Rubric Assessment Scale and additional questions compared U.S. based results versus other countries. Data were analyzed using descriptive statistics, and Wilcoxon rank-sum test for comparisons of median differences. All statistical tests were two-tailed p-value < .05 was considered statistically significant. RESULTS A total of 80 evaluations showed 88% accuracy and 89% efficiency with significant geographic disparities. U.S. based searches demonstrated alignment with management guidelines (p=0.02), higher perceived overall value (p=0.02) and inclusion of local experts(p<0.0001). The global group reported wider, less relevant differential diagnoses (p=0.03) and demonstrated lack of regional and socio-cultural contexts. CONCLUSIONS OE is a valuable tool but exhibits significant geographic bias, favoring U.S. clinical contexts with limitations for global applicability. There is critical need for training AI models on diverse and representative datasets to ensure accurate performance. Clinicians should critically appraise AI outputs prior to application to model responsible use of AI. CLINICALTRIAL NA
- 30 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessPromise or Peril: Evaluating the Performance of OpenEvidence in Clinical Decision Making in the Global Context. (Preprint)
BACKGROUND Clinicians face the impossible task of keeping up with evidence-based literature to inform clinical decision making. OpenEvidence holds great promise by providing latest evidence to inform patient care.AI solutions are increasingly used to support physicians during clinical decision making. Clinicians need to critically analyze and evaluate outputs and be aware of biases that exist prior to application, especially when caring for patients in global contexts. OBJECTIVE To evaluate performance of OpenEvidence (OE) in global context. METHODS In this study a sample of 11 clinicians from the United States (n=3) and six other countries (n=8) performed standardized searches for 10 common clinical conditions. Search results were evaluated using the Rubric Assessment Scale and additional questions compared U.S. based results versus other countries. Data were analyzed using descriptive statistics, and Wilcoxon rank-sum test for comparisons of median differences. All statistical tests were two-tailed p-value < .05 was considered statistically significant. RESULTS A total of 80 evaluations showed 88% accuracy and 89% efficiency with significant geographic disparities. U.S. based searches demonstrated alignment with management guidelines (p=0.02), higher perceived overall value (p=0.02) and inclusion of local experts(p<0.0001). The global group reported wider, less relevant differential diagnoses (p=0.03) and demonstrated lack of regional and socio-cultural contexts. CONCLUSIONS OE is a valuable tool but exhibits significant geographic bias, favoring U.S. clinical contexts with limitations for global applicability. There is critical need for training AI models on diverse and representative datasets to ensure accurate performance. Clinicians should critically appraise AI outputs prior to application to model responsible use of AI. CLINICALTRIAL NA
- 8 August 2025
- Published by Taylor & Francis in Medical Teacher
- p. 1-3
Addressing educational overload with generative AI through dual coding and cognitive load theories
Health professions education faces a critical challenge: the volume and complexity of medical knowledge has outpaced the cognitive limits of learners. Cognitive load theory indicates that traditional text-heavy instruction overloads the working memory. Current educational materials fail to leverage the dual coding theory which proposes that content should be presented through both verbal and visual channels. Generative AI tools enable the creation of multimodal educational content. Platforms such as ChatGPT, Gemini, NotebookLM, and HeyGen enable the creation of multimodal educational content - audio summaries, interactive mind maps, infographics, narrated videos, and voice-based interactions - designed to improve information processing in the working memory. Multimodal AI-generated content potentially enhances comprehension, retention and learning efficiency, and provides scalable, easily updated resources. These tools align with Generation Z learning preferences and can be integrated into existing curricula to transform text heavy content into multimodal engaging content. These tools can be used by faculty and learners,with minimal technical expertise or time investment. To ensure success, institutions must provide faculty and learners training, with consistent access, and evaluate outcomes. Pilot programs with iterative feedback can guide thoughtful implementation. By aligning educational strategies with cognitive science principles, generative AI can play a transformative role in addressing educational overload and creating more effective, engaging learning environments.
- 8 August 2025
- Published by Taylor & Francis in Medical Teacher
- p. 1-3
Addressing educational overload with generative AI through dual coding and cognitive load theories
Health professions education faces a critical challenge: the volume and complexity of medical knowledge has outpaced the cognitive limits of learners. Cognitive load theory indicates that traditional text-heavy instruction overloads the working memory. Current educational materials fail to leverage the dual coding theory which proposes that content should be presented through both verbal and visual channels. Generative AI tools enable the creation of multimodal educational content. Platforms such as ChatGPT, Gemini, NotebookLM, and HeyGen enable the creation of multimodal educational content - audio summaries, interactive mind maps, infographics, narrated videos, and voice-based interactions - designed to improve information processing in the working memory. Multimodal AI-generated content potentially enhances comprehension, retention and learning efficiency, and provides scalable, easily updated resources. These tools align with Generation Z learning preferences and can be integrated into existing curricula to transform text heavy content into multimodal engaging content. These tools can be used by faculty and learners,with minimal technical expertise or time investment. To ensure success, institutions must provide faculty and learners training, with consistent access, and evaluate outcomes. Pilot programs with iterative feedback can guide thoughtful implementation. By aligning educational strategies with cognitive science principles, generative AI can play a transformative role in addressing educational overload and creating more effective, engaging learning environments.
- 10 January 2025
- Published by Taylor & Francis in Medical Teacher
- Vol. 47 (6) , 949-954
Twelve tips on applying AI tools in HPE scholarship using Boyer’s model
AI has changed the landscape of health professions education. With the hype now behind us, we find ourselves in the phase of reckoning, considering what's next; where do we start and how can educators use these powerful tools for daily teaching and learning. We recognize the great need for training to use AI meaningfully for education. Boyer's model of scholarship provides a pedagogical approach for teaching with AI and how to maximize these efforts towards scholarship. By offering practical solutions and demonstrating their usefulness, this Twelve tips article demonstrates how to apply AI towards scholarship by leveraging the capabilities of the tools. Despite their potential, our recommendation is to exercise caution against AI dependency and to role model responsible use of AI by evaluating AI outputs critically with a commitment to accuracy and scrutinize for hallucinations and false citations.
- 10 January 2025
- Published by Taylor & Francis in Medical Teacher
- Vol. 47 (6) , 949-954
Twelve tips on applying AI tools in HPE scholarship using Boyer’s model
AI has changed the landscape of health professions education. With the hype now behind us, we find ourselves in the phase of reckoning, considering what's next; where do we start and how can educators use these powerful tools for daily teaching and learning. We recognize the great need for training to use AI meaningfully for education. Boyer's model of scholarship provides a pedagogical approach for teaching with AI and how to maximize these efforts towards scholarship. By offering practical solutions and demonstrating their usefulness, this Twelve tips article demonstrates how to apply AI towards scholarship by leveraging the capabilities of the tools. Despite their potential, our recommendation is to exercise caution against AI dependency and to role model responsible use of AI by evaluating AI outputs critically with a commitment to accuracy and scrutinize for hallucinations and false citations.
- 26 October 2024
- Published by Taylor & Francis in Medical Teacher
- Vol. 47 (5) , 788-790
Pedagogy and generative artificial intelligence: Applying the PICRAT model to Google NotebookLM
Healthcare educators (HPE) are challenged by rapid developments in Generative Artificial Intelligence (GenAI) tools. They need a standardized model to evaluate these new tools and to guide them in pedagogically-sound integration in the curriculum. PICRAT is an educational model designed specifically to help teachers meet this challenge. NotebookLM is a new multi-featured GenAI tool to help teachers and learners in education and research. Its newest feature allows automatic generation of an engaging podcast (called audio overview) from uploaded education or research content. Using the example of NotebookLM and, specifically, the auto-podcast feature, we illustrate how HPE can use the PICRAT model to evaluate GenAI tools for technology integration. We discuss how this model can be utilized as a standardized approach for evaluation and implementation of GenAI tools in health professions education.
- 26 October 2024
- Published by Taylor & Francis in Medical Teacher
- Vol. 47 (5) , 788-790
Pedagogy and generative artificial intelligence: Applying the PICRAT model to Google NotebookLM
Healthcare educators (HPE) are challenged by rapid developments in Generative Artificial Intelligence (GenAI) tools. They need a standardized model to evaluate these new tools and to guide them in pedagogically-sound integration in the curriculum. PICRAT is an educational model designed specifically to help teachers meet this challenge. NotebookLM is a new multi-featured GenAI tool to help teachers and learners in education and research. Its newest feature allows automatic generation of an engaging podcast (called audio overview) from uploaded education or research content. Using the example of NotebookLM and, specifically, the auto-podcast feature, we illustrate how HPE can use the PICRAT model to evaluate GenAI tools for technology integration. We discuss how this model can be utilized as a standardized approach for evaluation and implementation of GenAI tools in health professions education.
- Top Cited Papers
- 29 January 2024
- Published by Taylor & Francis in Medical Teacher
- Vol. 46 (6) , 752-756
Twelve tips on creating and using custom GPTs to enhance health professions education
The custom GPT is the latest powerful feature added to ChatGPT. Non-programmers can create and share their own GPTs ("chat bots"), allowing Health Professions Educators to apply the capabilities of ChatGPT to create administrative assistants, online tutors, virtual patients, and more, to support their clinical and non-clinical teaching environments. To achieve this correctly, however, requires some skills, and this 12-Tips paper provides those: we explain how to construct data sources, build relevant GPTs, and apply some basic security.
- 29 January 2024
- Published by Taylor & Francis in Medical Teacher
- Vol. 46 (6) , 752-756
Top Cited PapersTwelve tips on creating and using custom GPTs to enhance health professions education
The custom GPT is the latest powerful feature added to ChatGPT. Non-programmers can create and share their own GPTs ("chat bots"), allowing Health Professions Educators to apply the capabilities of ChatGPT to create administrative assistants, online tutors, virtual patients, and more, to support their clinical and non-clinical teaching environments. To achieve this correctly, however, requires some skills, and this 12-Tips paper provides those: we explain how to construct data sources, build relevant GPTs, and apply some basic security.
- 2 January 2024
- Published by Taylor & Francis in Postgraduate Medicine
- Vol. 136 (1) , 44-51
Levels and drivers of burnout during the COVID-19 pandemic among a diverse group of resident physicians
ObjectivesEvidence has shown significant impacts of the COVID-19 pandemic on physicians. We hypothesized that these effects would impact surgical and non-surgical resident education differently, with non-surgical specialties being more heavily impacted by frontline work and surgical specialties losing elective cases.MethodsWe examined well-being and burnout among resident physicians in surgical and non-surgical specialties during the peak of the COVID-19 pandemic using the Mayo Physician Well-Being Index (WBI).ResultsCompleted surveys were received from 110 residents, 55% of whom were in a surgical training program. 35% of respondents were identified as 'at risk' for burnout. Increased demands from work (adj. OR 3.79, 95% CI 1.50, 9.59, p = 0.005) was associated with an increased likelihood for being 'at risk' compared to those without increased demands. Odds of having increased stress level were higher amongst residents with fear/anxiety of the unknown (adj. OR 4.21, 95% CI 1.63, 10.90, p = 0.003) and more demands outside work (adj. OR 10.54, 95% CI 2.63, 42.16, p = 0.001) but lower amongst residents with more time for studying (OR 0.23, 95% CI 0.09, 0.64, p = 0.005). Risk for burnout was not significantly different between surgical and non-surgical specialties when adjusting for increased demands from work (adj. OR 1.43, 95% CI 0.60, 3.37, p = 0.0.418).ConclusionPerceived effects of the COVID-19 pandemic upon residents' educational experience was mixed: reduced clinical volume had a negative impact, while increased time for study was perceived favorably. These findings suggest potential strategies and targets to mitigate the stress and burnout of a future crisis, whether large or small, among surgical and non-surgical trainees.
- 2 January 2024
- Published by Taylor & Francis in Postgraduate Medicine
- Vol. 136 (1) , 44-51
Levels and drivers of burnout during the COVID-19 pandemic among a diverse group of resident physicians
ObjectivesEvidence has shown significant impacts of the COVID-19 pandemic on physicians. We hypothesized that these effects would impact surgical and non-surgical resident education differently, with non-surgical specialties being more heavily impacted by frontline work and surgical specialties losing elective cases.MethodsWe examined well-being and burnout among resident physicians in surgical and non-surgical specialties during the peak of the COVID-19 pandemic using the Mayo Physician Well-Being Index (WBI).ResultsCompleted surveys were received from 110 residents, 55% of whom were in a surgical training program. 35% of respondents were identified as 'at risk' for burnout. Increased demands from work (adj. OR 3.79, 95% CI 1.50, 9.59, p = 0.005) was associated with an increased likelihood for being 'at risk' compared to those without increased demands. Odds of having increased stress level were higher amongst residents with fear/anxiety of the unknown (adj. OR 4.21, 95% CI 1.63, 10.90, p = 0.003) and more demands outside work (adj. OR 10.54, 95% CI 2.63, 42.16, p = 0.001) but lower amongst residents with more time for studying (OR 0.23, 95% CI 0.09, 0.64, p = 0.005). Risk for burnout was not significantly different between surgical and non-surgical specialties when adjusting for increased demands from work (adj. OR 1.43, 95% CI 0.60, 3.37, p = 0.0.418).ConclusionPerceived effects of the COVID-19 pandemic upon residents' educational experience was mixed: reduced clinical volume had a negative impact, while increased time for study was perceived favorably. These findings suggest potential strategies and targets to mitigate the stress and burnout of a future crisis, whether large or small, among surgical and non-surgical trainees.
- Open Access
- 4 October 2023
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 26, e53344
Reflections from the pandemic: is Connectivism the panacea for clinicians? (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). The COVID 19 pandemic and the recent increased interest in generative artificial intelligence (GAI)I highlight the need for interprofessional communities collaborating to find solutions to complex problems. An example of this was how the siloed healthcare community was unaware of the airborne transmission of SARSCoV-2 .This delay in learning about the airborne transmission of the virus cost lives. Although constructivist learning approach is essential for knowledge building through experiential learning, this mistake could have been avoided if the healthcare community adopted a connectivist approach which is more suited for knowledge acquisition in digitally hyperconnected and networked world. We propose that healthcare communities adopt the theory of connectivism to supplement the traditional constructivist approach to keep current, up-to-date and to amplify accurate information .information. Learning in connectivism includes interacting with experts from other disciplines and creating nodes of accurate and accessible information, while deciphering misinformation from facts. Autonomy, connectedness, diversity and openness are foundational for learners to thrive in this connectivist learning environment.
- 4 October 2023
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 26, e53344
Open AccessReflections from the pandemic: is Connectivism the panacea for clinicians? (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). The COVID 19 pandemic and the recent increased interest in generative artificial intelligence (GAI)I highlight the need for interprofessional communities collaborating to find solutions to complex problems. An example of this was how the siloed healthcare community was unaware of the airborne transmission of SARSCoV-2 .This delay in learning about the airborne transmission of the virus cost lives. Although constructivist learning approach is essential for knowledge building through experiential learning, this mistake could have been avoided if the healthcare community adopted a connectivist approach which is more suited for knowledge acquisition in digitally hyperconnected and networked world. We propose that healthcare communities adopt the theory of connectivism to supplement the traditional constructivist approach to keep current, up-to-date and to amplify accurate information .information. Learning in connectivism includes interacting with experts from other disciplines and creating nodes of accurate and accessible information, while deciphering misinformation from facts. Autonomy, connectedness, diversity and openness are foundational for learners to thrive in this connectivist learning environment.
- Open Access
- 4 October 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Reflections From the Pandemic: Is Connectivism the Panacea for Clinicians? (Preprint)
UNSTRUCTURED The COVID-19 pandemic and the recent increased interest in generative artificial intelligence (GenAI) highlight the need for interprofessional communities’ collaboration to find solutions to complex problems. A personal narrative experience of one of the authors compels us to reflect on current approaches to learning and knowledge acquisition and use solutions to the challenges posed by GenAI through social learning contexts using connectivism. We recognize the need for constructivism and experiential learning for knowledge acquisition to establish foundational understanding. We explore how connectivist approaches can enhance traditional constructivist paradigms amid rapidly changing learning environments and online communities. Learning in connectivism includes interacting with experts from other disciplines and creating nodes of accurate and accessible information while distinguishing between misinformation and accurate facts. Autonomy, connectedness, diversity, and openness are foundational for learners to thrive in this learning environment. Learning in this environment is not just acquiring new knowledge as individuals but being connected to networks of knowledge, enabling health professionals to stay current and up-to-date. Existing online communities with accessible GenAI solutions allow for the application of connectivist principles for learning and knowledge acquisition.
- 4 October 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessReflections From the Pandemic: Is Connectivism the Panacea for Clinicians? (Preprint)
UNSTRUCTURED The COVID-19 pandemic and the recent increased interest in generative artificial intelligence (GenAI) highlight the need for interprofessional communities’ collaboration to find solutions to complex problems. A personal narrative experience of one of the authors compels us to reflect on current approaches to learning and knowledge acquisition and use solutions to the challenges posed by GenAI through social learning contexts using connectivism. We recognize the need for constructivism and experiential learning for knowledge acquisition to establish foundational understanding. We explore how connectivist approaches can enhance traditional constructivist paradigms amid rapidly changing learning environments and online communities. Learning in connectivism includes interacting with experts from other disciplines and creating nodes of accurate and accessible information while distinguishing between misinformation and accurate facts. Autonomy, connectedness, diversity, and openness are foundational for learners to thrive in this learning environment. Learning in this environment is not just acquiring new knowledge as individuals but being connected to networks of knowledge, enabling health professionals to stay current and up-to-date. Existing online communities with accessible GenAI solutions allow for the application of connectivist principles for learning and knowledge acquisition.
- 1 January 2018
- Published by Elsevier in The American Journal of Medicine
- Vol. 131 (1) , 107-114
X + Y Scheduling Models in Internal Medicine Residency Programs: A National Survey of Program Directors' Perspectives
- 1 January 2018
- Published by Elsevier in The American Journal of Medicine
- Vol. 131 (1) , 107-114
X + Y Scheduling Models in Internal Medicine Residency Programs: A National Survey of Program Directors' Perspectives
- Open Access
- 1 January 2018
- Published by Taylor & Francis in Medical Education Online
- Vol. 23 (1) , 1542922
Impact of the final adjective in the Medical Student Performance Evaluation on determination of applicant desirability
Background: The Medical Student Performance Evaluation (MSPE) is a primary source of information used by residency programs in their selection of trainees. The MSPE contains a narrative description of the applicant’s performance during medical school. In 2002, the Association of American Medical Colleges’ guideline for preparation of the MSPE recommended inclusion of a comparative summative assessment of the student’s overall performance relative to his/her peers (final adjective). Objective: We hypothesize that the inclusion of a final adjective in the MSPE affects a reviewer’s assessment of the applicant’s desirability more than the narrative description of performance and designed a study to evaluate this hypothesis. Design: Fifty-six faculty members from the Departments of Pediatrics and Medicine with experience reviewing MSPEs as part of the intern selection process reviewed two pairs of mock MSPE letters. In each pair, the narrative in one letter was superior to that in the other. Two final adjectives describing relative class ranks were created. Each subject was first presented with a pair of letters with mismatched final adjective (study), i.e., the letter with the stronger narrative was presented with the weaker final adjective and vice versa. The subject was then presented with a second pair of letters without final adjectives (control). Subjects ranked the relative desirability of the two applicants in each pair. Results: The proportion of rankings congruent with the strength of the narratives under study and control conditions were compared. Subjects were significantly less likely to rank the applicants congruent with the strength of the narratives when the strength of the final adjectives conflicted with the strength of the narrative; 42.9% of study letters were ranked congruent with the narrative versus 82.1% of controls (p = 0.0001). Conclusion: The MSPE final adjective had a greater impact than the narrative description of performance on the determination of applicant desirability. Abbreviations: MSPE: Medical Student Performance Evaluation; AAMC: Association of American Medical Colleges; BCM: Baylor College of Medicine
- 1 January 2018
- Published by Taylor & Francis in Medical Education Online
- Vol. 23 (1) , 1542922
Open AccessImpact of the final adjective in the Medical Student Performance Evaluation on determination of applicant desirability
Background: The Medical Student Performance Evaluation (MSPE) is a primary source of information used by residency programs in their selection of trainees. The MSPE contains a narrative description of the applicant’s performance during medical school. In 2002, the Association of American Medical Colleges’ guideline for preparation of the MSPE recommended inclusion of a comparative summative assessment of the student’s overall performance relative to his/her peers (final adjective). Objective: We hypothesize that the inclusion of a final adjective in the MSPE affects a reviewer’s assessment of the applicant’s desirability more than the narrative description of performance and designed a study to evaluate this hypothesis. Design: Fifty-six faculty members from the Departments of Pediatrics and Medicine with experience reviewing MSPEs as part of the intern selection process reviewed two pairs of mock MSPE letters. In each pair, the narrative in one letter was superior to that in the other. Two final adjectives describing relative class ranks were created. Each subject was first presented with a pair of letters with mismatched final adjective (study), i.e., the letter with the stronger narrative was presented with the weaker final adjective and vice versa. The subject was then presented with a second pair of letters without final adjectives (control). Subjects ranked the relative desirability of the two applicants in each pair. Results: The proportion of rankings congruent with the strength of the narratives under study and control conditions were compared. Subjects were significantly less likely to rank the applicants congruent with the strength of the narratives when the strength of the final adjectives conflicted with the strength of the narrative; 42.9% of study letters were ranked congruent with the narrative versus 82.1% of controls (p = 0.0001). Conclusion: The MSPE final adjective had a greater impact than the narrative description of performance on the determination of applicant desirability. Abbreviations: MSPE: Medical Student Performance Evaluation; AAMC: Association of American Medical Colleges; BCM: Baylor College of Medicine