United States

Background:
Clinical trials often struggle to enroll participants. Unnecessarily varied eligibility criteria create barriers to clinicians offering trials, particularly if the variation has limited clinical significance (e.g., eGFR 45 vs. 60; ECOG 1 vs. 2). However, the degree of variation in trial eligibility criteria remains poorly characterized. For these reasons, we analyzed the variability in eligibility criteria of genitourinary cancer trials at our institution.

Methods:
We extracted eligibility criteria from protocol documents for active genitourinary cancer trials at our institution. We grouped and compared eligibility criteria within prostate cancer and bladder cancer trials. 

Results:
We identified 35 active genitourinary cancer trials at our institution. Eligibility criteria were highly varied. For example, in prostate cancer trials (n=25), there were 6 different kidney function criteria (4 eGFR and 2 creatinine cutoffs). For all trials, there were 7 different eGFR cutoffs (25, 30, 40, 45, 50, 51, 60) with marginal practical implications. Other criteria were similarly varied. 

Conclusion: 
Trial eligibility criteria have unnecessary variation. These variations are of dubious clinical significance but increase the complexity of trials and may decrease trial acceptability, adoption, and patient enrollment. Simplifying trials, perhaps by creating standards for trial eligibility criteria, may address these barriers. Such standards could make trials more acceptable, less expensive, and more feasible through diminished administrative burden. 


AMA Research Challenge 2022

Unnecessary Variation in Clinical Trial Eligibility Criteria: a Call for Standardization

**Welcome to EMNLP 2023!**

On behalf of the EMNLP 2023 Organizing Committee, I extend a warm and heartfelt welcome to all of you to the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP). It is with immense pleasure and excitement that we gather here in Singapore, a vibrant hub of innovation and technological advancement.

The conference program is packed with insightful presentations, thought-provoking workshops, and engaging networking opportunities. In addition to the technical sessions, we have also planned several social events that will provide you with the opportunity to connect with your colleagues.

The conference is held in person at the Resorts World Convention Centre in Singapore, and available on-line with the help of Underline.

For those in Singapore, I hope that you will find time to explore the exciting Sentosa Island, Gardens by the Bay, Singapore Botanic Gardens and the many other attractions unique to Singapore.

Of course, we are grateful to our sponsors and partners for their generous support of EMNLP 2023, whose contributions make it possible for us to host this world-class event.

Yuji Matsumoto (RIKEN AIP) 
EMNLP 2023 General Chair

To access the **EMNLP 2023** event page on Underline, you need to register for the Conference. 
Please follow **[this link](https://2023.emnlp.org/registration/)** for more details.

EMNLP 2023

EMNLP 2023 took place in Singapore from Dec 6th to Dec 10th, 2023.

poster

Welcome to the largest national, multi-specialty research event **featuring research from 750 medical students, residents and fellows, and international medical graduates.**


#### Event Details
The 2022 American Medical Association Research Challenge includes a poster symposium hall and a semifinals hall with the top scored research. Log in at any time throughout the event to:
* **View posters and presentations** on a variety of topics and specialties
* **Score posters in the semifinals** to help decide the top five finalists (AMA member exclusive opportunity)
* **Connect with participants** and provide feedback to help advance their research

At the conclusion of this event, the AMA will announce the top five finalists who will advance to the AMA Research Challenge finals, present their research to an elite panel of judges and compete to win a $10,000 grand prize sponsored by Laurel Road.
 

#### Attend Live Forums
**Oct. 20, 6 — 7 p.m. Central**
 **How to plan a career in academic medicine** 
 Learn how to plan a career in academic medicine and get your questions answered. [Access Here](https://underline.io/events/363/lounge)
 
 **Speakers**
![](https://assets.underline.io/uploads/markdown_image/1/image/d5adb133dfc58eb0a6d469e9cedcd5d4.jpg)
Ricardo Correa, MD, EdD
![](https://assets.underline.io/uploads/markdown_image/1/image/5f9437ab87861538bf02b9ca7c0d926e.jpg)
Stuart M. Greenstein, MD
![](https://assets.underline.io/uploads/markdown_image/1/image/05b1b637c99286d558c371f8e8b185ce.jpg)
Stephanie Strohbeen, MD
 
 **Oct. 22, 11 a.m. — Noon Central**
 **How to get published**
 Learn tips to get your research published and get your questions answered. [Access Here](https://underline.io/events/363/lounge)
 
 **Speakers**
![](https://assets.underline.io/uploads/markdown_image/1/image/8ce6235e957d31cb5e6f9097cc6f6972.jpg)
Jessica Li, MD/PhD Candidate
![](https://assets.underline.io/uploads/markdown_image/1/image/9950ef2e0237f54cb3d8280334d42491.jpg)
Charles Lopresto, DO

### Sponsor
AMA would like to thank Laurel Road for sponsoring a $10,000 grand prize for the winner of the AMA Research Challenge. Laurel Road is not involved in the judging or scoring process to select the winner. The winner will be announced on Dec. 7 during the AMA Research Challenge finals. Payment of the grand prize will be directly from Laurel Road. See the AMA Research Challenge Official Rules for terms and conditions.

![](https://assets.underline.io/uploads/markdown_image/1/image/611e331d4a0eba0920f240cfc04b4d77.png)

**There is no registration fee for this event! **

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals.

**Please use the link below to register for free if you have not already done so:**

https://www.ama-assn.org/about/events/ama-research-challenge

Registration Is Required

Welcome to the largest national, multi-specialty research event featuring research from 750 medical students, residents and fellows, and international medical graduates.


Background: Electronic health records (EHR) were implemented to improve patient care, reduce healthcare disparities, engage patients and families, improve care coordination, and maintain privacy and security. Unfortunately, the mandated use of EHR has also resulted in significantly increased clerical and administrative burden, with physicians spending an estimated three-fourths of their daily time interacting with the EHR, which negatively affects within-clinic processes and contributes to burnout. In-room scribes are associated with improvement in all aspects of physician satisfaction and increased productivity, though less is known about the use of other technologies such as GG, Natural Language Processing (NLP) and Machine-Based Learning (MBL) systems. Given the need to decrease administrative burden on clinicians, particularly in the utilization of the EHR, there is a need to explore the intersection between varying degrees of technology in the clinical encounter and their ability to meet the aforementioned goals of the EHR.

Aims: The primary aim is to determine predictors of overall perception of care dependent on varying mechanisms used for documentation and medical decision-making in a routine clinical encounter. Secondary aims include comparing the perception of individual vignettes based on demographics of the participants and investigating any differences in perception questions by demographics of the participants.

Methods: Video vignettes were shown to 498 OhioHealth Physician Group patients and to ResearchMatch volunteers during a 15-month period following IRB approval. Data included a baseline survey to gather demographic and background characteristics and then a perceptual survey where patients rated the physician in the video on 5 facets using a 1 to 5 Likert scale. The analysis included summarizing data of all continuous and categorical variables as well as overall perceptions analyzed using multivariate linear regression with perception score as the outcome variable.

Results: Univariate modeling identified sex, education, and type of technology as three factors that were statistically significantly related to the overall perception score. Males had higher scores than females (p=0.03) and those with lower education had higher scores (p<0.001). In addition, the physician documenting outside of the room encounter had statistically significantly higher overall perception scores (mean=22.2, p<0.001) and the physician documenting in the room encounter had statistically significantly lower overall perception scores (mean=15.3, p<0.001) when compared to the other vignettes. Multivariable modeling identified all three of the univariably significant factors as independent factors related to overall perception score. Adjusted least squares means (LSM) were calculated and males had higher scores than females (LSM=20.7 vs. 19.9, p=0.03). Those with lower education had higher scores (p<0.001). Specifically, high school education had higher scores than associate/bachelor education (LSM=21.6 vs. 19.9, p=0.0002) and high school education had higher scores than master/higher education (LSM=21.6 vs. 19.5, p<0.0001). Differences in technology were found with the physician documenting outside of the room encounter having significantly higher overall perception scores (LSM =22.6) and physician documenting in the room encounter having significantly lower overall perception scores (LSM =15.7) compared to the other vignettes. No differences between age groups were found on the individual perception scores. Males had higher scores than females on ‘The doctor clearly explained the diagnosis and treatment to the patient’ and ‘The doctor was sincere and trustworthy’. High school education had higher scores than associate/bachelor and master/higher on all five individual perception scores.

Conclusion: The study found sex, education, and type of technology were significant indicators for overall perception of varying technologies used for documentation and medical decision-making in a routine clinical encounter. Importantly, the vignette depicting the least interaction with the EHR received the most positive overall perception score, while the vignette depicting the physician utilizing the EHR during the interaction received the least positive overall perception score. This suggests patients most value having the full attention of the physician and feel less strongly about differentiating the logistics of data transcription and medical decision-making, provided they feel engaged during the interaction. Therefore, the authors suggest maximizing face-to-face time in the integration of technology into the clinical encounter, allowing for increased perceptions of personal attention in the patient-physician interaction.

Assessment of Patient Perceptions of Technology and theUse of Machine-Based Learning in a Clinical Encounter

Background: 
• Sleep disruption occurs in two-thirds of patients in the hospital due to the unfamiliar and stressful environment characterized by illness and discomfort, a preponderance of noise and light, and around-the-clock interventions. 
• Effects of sleep deprivation include decreased immune function, higher heart rate and blood pressure, increased insulin resistance, decreased cognitive function and alertness, increased rate of falls and delirium, increased length of hospital stay, and impaired social interactions.
• Clustering care allows patients to receive uninterrupted rest and has demonstrated positive effects on sleep quality and quantity. 
• Rest improves outcomes, length of stay, satisfaction scores, and fall rates and reduces the need for medications 

Methods
This quality improvement study is being conducted on a general Medicine floor in Queens Hospital Center, New York. To minimize patient interruption during the night hours, designated hours of uninterrupted sleep (11 pm-5 am) for all admitted patients were implemented. 
Interventions: 
• Avoiding drawing labs, checking finger stick blood sugar, and taking vitals or any patient care between these hours in non-emergency situations.
• Avoiding giving medication during this time unless it was an emergency.
• Minimizing the amount of noise and light and silencing unnecessary sounds to ensure a proper environment for patients to get sound sleep. 
A PICOT (population, intervention, comparison, outcome, and time frame) questionnaire has been developed to administer surveys to discharged patients upon leaving the hospital about their sleep quality, injury or falls, overall concerns, and satisfaction


Results
Intervention or Uninterrupted sleep cycle (%)/ Control or interrupted sleep cycle (%)
1.Woken up an appropriate amount of time during the night- 82.1/70
2.Believed that overnight awakenings did not significantly affect them the next morning- 77.8/60
3.Believe health concerns have improved significantly-46.4/20
4.Likely to visit the same hospital for future health concerns -57.1/55


Conclusion
• The areas of the hospital where the sleep cycle of the patients was not interrupted showed a significant improvement and rectification of their health concerns.
• This should encourage hospital administration to implement improved plans for uninterrupted sleep cycles while promoting better healthcare.






Better Sleep,Better Healing:Sleep Improvement Project at Hospital Setting

Background: According to the World Health Organization (WHO), 1 in 6 people will be ages 60 or older by 2030. With existing age-related preconceptions such as older adults being “frail” or a “burden”, it is crucial for physicians to understand how to care for older patients, especially when they are marginalized by both age and other factors like gender, immigration status, and income. Resident physicians have limited and varied geriatric and social justice training. The objective of this study is to develop a social justice curriculum with a module focusing on senior care.

Methods: Literature reviews were conducted on PubMed articles to summarize important factors when providing care to older adults. Search terms included: “ageism” AND “elder care” (n=32 results), “best practices geriatrics” (n=2,614 results), “screening for elder abuse” (n=932 results), and “family support” AND “elders” (n=2,812 results). Of these, 20 articles were selected to compile major themes into a document that is provided to residents during the curriculum.

Results: Key senior care and social justice practices are: (1) collaborate and create goals important to the patient, (2) use age-inclusive language, (3) communicate with others involved in patient’s care to establish an interprofessional team, (4) assess fall risk when necessary and provide recommendations to reduce falls given a patient’s environmental context, (5) eliminate unnecessary medications and provide financial resources for medications, (6) screen for elder abuse.

Conclusion: There are key evidence-based knowledge, attitudes, and skills that focus on unique healthcare needs of older adults who may also be marginalized by society with intersecting identities (e.g., refugees, unhoused individuals). Training resident physicians with this focus may allow them to better advocate for their patients’ goals. Next steps are to evaluate the effectiveness of this curriculum.

Development of a Social Justice Curriculum: Best Practices for Senior Care and Age-Inclusive Medicine

Title: Validating the Utility of Standard Operating Procedures In a Student-Run Free Clinic
Background: Student-Run Free Clinics are common at medical schools across the country. These clinics face unique challenges with continuity due to annual changes in leadership. In this pilot study, we aimed to stabilize the leader transition process by creating standard operating procedures (SOPs) for each position in the MedZou Specialty Clinic.
Methods: MedZou Specialty Clinic SOPs for 2021 and 2022 were assessed via survey. Participants rated quality of SOPs and their self-assessed preparedness. All ratings were measured on a scale from 1 to 5, with 1 representing the lowest quality or least prepared and 5 representing the highest quality or most prepared. An optional free-response was also included.
Results: In 2021, only 7 of 11 student leaders received a SOP during training. Upon study completion in 2022, all 11 incoming student leaders received a SOP. There was an observed increase in average quality (P<.001) and self-assessed preparedness (P= .006) between cohorts.
Conclusion: Providing SOPs during the MedZou training period correlated with increased satisfaction and confidence in incoming student leaders. Implementation of SOPs for all MedZou clinics may be helpful for creating smooth transitions for future leaders.

Validating the Utility of Standard Operating Procedures in a Student-Run Free Clinic

Background: The rising prevalence of physician burnout in recent years has drawn significant attention. The COVID-19 pandemic has resulted in additional strain on healthcare staff working tirelessly for extended hours, amplifying physical exhaustion and emotional stress levels. Physicians in training (residents) have been considered to be at increased risk of burnout, even before the pandemic. With limitations in social interactions and travel restrictions, the possibility of interventions to prevent burnout has significantly reduced. We conducted a study to assess the impact of the COVID19 pandemic on self-reported burnout among internal medicine residents at our hospital.

Methods: The Oldenburg Burnout Inventory was used as the screening tool. Residents filled out a mid-year questionnaire in December 2019 [pre-COVID period] and in December 2021 [COVID period]. The survey was voluntary, anonymous, and confidential. The inventory included a total of 16 questions divided into disengagement and exhaustion categories. Residents scored the questions between 1 through 4 [strongly agree to strongly disagree]. Overall, burnout scores were characterized as high [score >2.17], medium [1.0-2.17], and low [<1] for the disengagement category. In the exhaustion category, burnout scores were characterized as high [>2.57], medium [1.67-2.57], and low [<1.67]. The degree of burnout among residents before and during the COVID pandemic was compared based on cohort/ level of training. pre-COVID pandemic and during the pandemic. A higher score indicated suggested more burnout among residents. The one-way ANOVA test and unpaired T-test were used for statistical analyses, as appropriate.

Results: 75 responses were obtained during the pre-COVID period, and 43 responses were obtained during the COVID period. Comparing the first-year trainees [PGY1], pre-COVID and during COVID, there reported more burnout [p<0.0001] during the COVID pandemic. During COVID, first-year trainees [PGY1] reported significantly more burnout as compared to PGY2 [p=0.025] and PGY3 [p=0.0018] trainees with higher burnout scores. The level of burnout was not statistically significant between PGY2 and PGY3 trainees [p=0.95]. In the disengagement category, The residents scored highest in questions concerning disengagement over time, the question being if this was the only type of work they can see themselves doing. In the exhaustion category, residents reported that they felt tired before arriving at work and needed a longer time to relax to feel better after work. They felt worn out easily quickly, not having enough energy for leisure activities.

Conclusion: The data suggest that there is comparably more burnout among PGY1 with learning a new healthcare system and electronic medical records(EMR)s, adjusting to new work schedules and living conditions as possible contributing factors. In comparison, concern for future career/fellowship applications is a potential source of concern for PGY2. Our observation of increased burnout among the current PGY1 who started their training during the pandemic raises questions about how well they were prepared to face a healthcare crisis. Meanwhile, during their second and third years of training, residents have gathered more skills and knowledge which enhance their resilience and ability to cope with unprecedented demands on the healthcare system.

A Comparison of Burnout Among Internal Medicine Residents in Different Post-Graduate Years of Training in a Community Hospital During the COVID-19 Pandemic

Background: Supplementary medical school learning resources, including textbooks, United States Medical Licensing Examinations (USMLE) preparatory materials, and internet searches have all been shown to lack equitable representation of brown and black skin images. This creates a knowledge gap for learners and decreases providers’ confidence managing skin disease in all populations. Subsequently, patients with non-white skin experience delayed or missed diagnoses, which contributes to known skin health disparities. Here, we aimed to create a standardized course auditing system to assess the use of brown and black skin images in the preclinical medical education.

Methods: We performed a cross-sectional audit of the 2020-2021 preclinical curriculum at the University of Pittsburgh School of Medicine (UPSOM). All human images in the learning material for 34 first- and second-year medical school courses were analyzed. Skin color was categorized as light/white, medium/brown, and dark/black using the Massey-Martin NIS Skin Color Scale.

Results: In total, 1,660 unique images were analyzed; 71.3%, (n=1,183) were light/white, 16.1% (n=267) were medium/brown and 12.7% (n=210) were dark/black. Dermatologic images of skin, hair, nail, or mucosal disease made up 62.1% (n=1,031) of the images and 68.1% (n=702) of the dermatologic images were light/white. Images of infectious diseases were more frequently presented in darker skin colors (χ2(2) = 15.46, p<.001). 

Conclusion: White skin was the standard used for visual learning images. These results further substantiate the lack of representation of brown and black skin in medical education. Using our efforts as a model, we propose guidelines to achieve visual learning equity. First, institutions should create a taskforce to implement a similar course auditing system. Medical school educators should be informed to present skin conditions on a spectrum of skin colors (white, brown, and black skin) to demonstrate differences in disease manifestations based on constitutive skin type. Resources with high quality images of brown and black skin should be made accessible for editing learning material. Course content should be audited regularly with feedback and support given to instructors. It is paramount that we analyze and revise our medical school curricula to ensure the next generation of physicians are educated to care for all patients.

Achieving Visual Learning Equity: A Course Auditing System of Skin Color of Images in Preclinical Medical Education

Addressing Instances of Racism, Discrimination, and Microaggression in Medical Education in Puerto Rico: A Symposium in Diversity, Equity and Inclusion

Background 

While the number of medical school applicants has increased, disparities among underrepresented groups in medicine continue to widen. Previous research has examined medical school applicants; however, limited literature examines the medical school re-applicant population. This study compares first-time applicants (FTA) to re-applicants (RA) from 2014 to 2022 by race/ethnicity. 

Methods 

Historical data was requested and acquired for RA and FTA by race/ethnicity (Alone or in Combination) from the AAMC matriculation file. In total, 411,770 applicants between 2014 and 2022 were analyzed by race/ethnicity, respectively. Of those applicants, there were 302,345 (73.4%) FTA and 109,425 (26.6%) RA. Descriptive statistics were calculated. 

Results 

The percentage change from 2014 to 2022 for FTA amongst these groups was: American Indian/Alaska Native (AI/AN): +51.8%; Asian: +53.3%; Black/African American (B/AA): +88.2%; Hispanic/Latino/Spanish Origin (H/L/SO): +70.9%; Native Hawaiian/Other Pacific Islander (NH/OPI): +36.8%; White: +16.7%; Other: +29.3%. The percentage change from 2014 to 2022 for RA amongst these groups was: AI/AN: +58.4%; Asian: +40.0%; B/AA: +73.4%; H/L/SO: +50.8%; NH/OPI: +68.2%; White: +13.0%; Other: +7.9%. 

Towards the end of the reporting period, there are similar upward trends for FTA who identified as Asian, B/AA, H/L/SO, and White. This was not seen for the FTA amongst other racial/ethnic categories, which were relatively stable over time (AI/AN, NH/OPI, Other). When examining RA, there is a similar upward trend towards the end of the reporting period among those who identified as Asian, B/AA, and White. There was no upward trend amongst those who identified as H/L/SO even though there was an increase in applicants. The other groups trended stable over time (AI/AN, H/L/SO, NH/OPI, Other). 

Conclusion 

Racial/Ethnic disparities persist among reapplicants to medical school. Without remediation, this concerning gap will continue to result in lower numbers of H/L/SO matriculants, who are key to creating diverse, equitable, and inclusionary systems of health care. Future research should use qualitative methods to explore barriers to and reasons for not reapplying, as well as include a larger number of schools across geographical regions. Future studies should also include focus on interventions to improve opportunities for reapplicants to be admitted to medical school successfully.

An Analysis of U.S.Medical School Re-Applicants by Race/Ethnicity

Background: Given the increasing impact artificial intelligence (AI) has in the field of medicine, physician leaders have called for AI education in medical training, even at the undergraduate medical education level. This study aims to assess the perceptions of US medical students on AI in medicine and to survey the most preferred teaching methods and AI topics. 

Methods: A mixed-methods survey was sent to all 169 US MD/DO schools asking for help with survey distribution in May 2021. Thirteen 5-point Likert scale questions assessed the perceptions of medical students about AI in medicine. We also asked how many hours students want to spend learning about AI per month. Next, we asked students to select their preferred learning modalities and AI topics. Finally, we included a free response section to capture any additional thoughts. 

Results 390 medical students (age: 26 ± 3) from 17 US medical schools participated in the survey. 351 (90%) agreed that “AI will take a significant role in medicine” and 347 (89%) “want to learn what medical students should know about AI.” Furthermore, only 60 (16%) agreed that learning about AI will significantly detract themselves from the existing medical school curriculum. 197 (52%) chose to spend 1-2 hours per month to learn about AI, while 123 (33%) chose 3+ hours. Despite this, only 34 (9%) said that their school has opportunities to learn about AI. Considering this knowledge, the most popular learning modalities students preferred were short lectures (264, 70%), preclinical electives (180, 48%), and Q&A panels (167, 44%). The three most popular AI topics are fundamental concepts of AI (247, 65%), when to use AI (227, 60%), and the pros and cons of AI (224, 59%). Also, 23 free text responses were received with a mix of opinions on the extreme ends. 

Conclusion: US medical students understand and are excited for the potential impact of AI in medicine. Current resources on AI are insufficient in medical school curriculums. Formal curricular offerings on the basics and the clinical utility of AI are good first steps to teach medical students about AI.

Artificial Intelligence Education in Medical School: What do Students Want to Know?

Background
One of the manifestations of the long-standing call for increased diversity and recruitment of underrepresented racial/ethnic minorities in medicine (URMiMs) are pre-medical pathway programs. One such program is the Bridge to Medicine (BTM) program, which boosts knowledge in healthcare careers and provides academic resources to URMiM high school students (HSS). While other pathway programs have demonstrated efficacy in promoting career awareness and equipping students with professional developmental skills, there are fewer programs and studies that evaluate the impact of these programs on students’ career-related perceptions and interests. Our study hypothesized that the BTM program would improve measures of career certainty and decision among URMiM HSS.
Methods
Career perceptions and interests in the STEMM field were evaluated using the Career Decision Scale (CDS), Career Thoughts Inventory (CTI), STEM Semantics Survey, and STEM Career Interest Survey (STEM CIS). Correlational analyses and MANOVAs were performed to evaluate differences between scores of career indecision and uncertainty across demographic variables.
Results
Correlational analyses and descriptive statistics revealed short-term successes in improving career certainty and decision. However, MANOVA’s revealed differences in race, gender, and grade level post participation in the BTM program.
Conclusion
Various measures of career certainty and decision CTI and CDS subscales highlighted significant changes in student career perceptions from the start to the end of the BTM program with an overall increase in career interest and certainty and decrease in career indecision by the end of the BTM program. Significant differences in CTI CA scores across racial/ethnic groups highlight the discrepancy in career uncertainty within students of racial/ethnic minority groups and the lack of consideration of systemic barriers that likely underlie these differences in career perceptions. 


Premium content

Next from AMA Research Challenge 2022

Assessment of Patient Perceptions of Technology and theUse of Machine-Based Learning in a Clinical Encounter

Similar lecture

AutoTrial: Prompting Language Models for Clinical Trial Design