United States

Background

Electronic health record (EHR) notes are critical to medical decision making but have become unwieldy to manually sift through due to their many aims: communication, compliance, and billing. Consequently, clinicians spend more time navigating EHRs than with patients, leading to burnout. 

Prior analysis focuses on static completed notes, and few have explored granular reading and writing activities. Clinicians describe time-varying purposes of EHR use: rapid sense-making of new patients and re-familiarization with existing patients. We introduce a framework to dynamically retrieve relevant notes at different points during a patient’s visit to streamline discovery of relevant history and satisfy changing information needs.

Methods

We analyze 48,192 note-writing sessions in the high-acuity emergency department (ED). We design a binary classification task of whether to retrieve a past note during the next session given the current writing context. Audit logs, which track reading and writing activity in the EHR, provide labels for machine learning. We encode patient and clinician information, note metadata, and a bag of words representation for note text. We train logistic regression models to predict the top ten most relevant notes for each session. 

For quantitative evaluation, we measure precision@10, recall@10, and AUC. Additionally, two physicians conducted chart review for five random patients. Each clinician was given the patient&#39;s age, sex, and chief complaint, and assigned relevance for past notes. We compared their relevance judgments with the predicted relevant notes.

Results

Our learned models’ quantitative performance surpassed that of deterministic baselines and achieved an AUC of 0.963. For the qualitative evaluation, for four out of five visits, the top ranked predicted note surfaced during the first session of the patient’s visit was determined to be directly helpful to the patient’s current presentation by both clinicians. For three of four visits, showing only the top ranked predicted note was sufficient information that both clinicians needed to proceed with patient care.

Conclusion

We present a novel framework that updates retrieval suggestions dynamically as clinical context changes, thus helping clinicians find relevant information more quickly at different points during the patient’s visit. It also surfaces difficult-to-find information and suggests patterns a clinician may not have initially deemed relevant, prompting them to reassess a diagnosis they initially anchored on and evaluate potential biases. Demonstrating that our methods perform well in the demanding ED setting is a promising proof of concept that can translate to other clinical settings and data modalities.

AMA Research Challenge 2024 

Conceptualizing Machine Learning for Dynamic Information Retrieval of EHR Notes

## Welcome to EMNLP 2022!
I am delighted to welcome you to EMNLP 2022! I believe this conference has been complicated beyond any precedent. Over the past year, it’s been thrilling to see the organization team approach each new puzzle with creativity and enthusiasm. We hope that those participating in Abu Dhabi as well as those joining remotely will leave the conference feeling newly inspired by the program and newly connected to our ever-growing community. Following EMNLP 2021 and major NLP conferences since, EMNLP 2022 is “hybrid,” serving both virtual and in-person participants.

Our key innovations for EMNLP 2022 include:

* EMNLP 2022 is “hybrid” in a second sense, as well: we allowed both direct and rolling review paper submissions, building on the pilot experiment of EMNLP 2021, which considered a small number of ARR submissions. 
* Familiar from NAACL but new to EMNLP, we’ve added an industry track.
* During the conference, “portals” will link virtual poster sessions to in-person conference participants during poster sessions each day.
* The first ACL-family conference in the United Arab Emirates.

 *Message from Noah A. Smith, University of Washington and Allen Institute for AI, Seattle, Washington, USA* 
***EMNLP 2022 General Chair***
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/9eec7d4a287ee18c278b08229290aa83.png)](https://drive.google.com/file/d/1OlPv6QBeo62VVTughj2jkiLeyHd1WnUt/view)
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/a3db7a768409f05192210d98601edb25.png)](https://emnlp2022.rocket.chat/)

To access this site you need to register. Please register [here](https://2022.emnlp.org/registration/).

Register here

EMNLP 2022

Welcome!
EMNLP 2022 will take place in Abu Dhabi from December 7th to December 11th, 2022. And it will be held in hybrid mode, both online and offline.

Background

Electronic health record (EHR) notes are critical to medical decision making but have become unwieldy to manually sift through due to their many aims: communication, compliance, and billing. Consequently, clinicians spend more time navigating EHRs than with patients, leading to burnout. 

Prior analysis focuses on static completed notes, and few have explored granular reading and writing activities. Clinicians describe time-varying purposes of EHR use: rapid sense-making of new patients and re-familiarization with existing patients. We introduce a framework to dynamically retrieve relevant notes at different points during a patient’s visit to streamline discovery of relevant history and satisfy changing information needs.

Methods

We analyze 48,192 note-writing sessions in the high-acuity emergency department (ED). We design a binary classification task of whether to retrieve a past note during the next session given the current writing context. Audit logs, which track reading and writing activity in the EHR, provide labels for machine learning. We encode patient and clinician information, note metadata, and a bag of words representation for note text. We train logistic regression models to predict the top ten most relevant notes for each session. 

For quantitative evaluation, we measure precision@10, recall@10, and AUC. Additionally, two physicians conducted chart review for five random patients. Each clinician was given the patient's age, sex, and chief complaint, and assigned relevance for past notes. We compared their relevance judgments with the predicted relevant notes.

Results

Our learned models’ quantitative performance surpassed that of deterministic baselines and achieved an AUC of 0.963. For the qualitative evaluation, for four out of five visits, the top ranked predicted note surfaced during the first session of the patient’s visit was determined to be directly helpful to the patient’s current presentation by both clinicians. For three of four visits, showing only the top ranked predicted note was sufficient information that both clinicians needed to proceed with patient care.

Conclusion

We present a novel framework that updates retrieval suggestions dynamically as clinical context changes, thus helping clinicians find relevant information more quickly at different points during the patient’s visit. It also surfaces difficult-to-find information and suggests patterns a clinician may not have initially deemed relevant, prompting them to reassess a diagnosis they initially anchored on and evaluate potential biases. Demonstrating that our methods perform well in the demanding ED setting is a promising proof of concept that can translate to other clinical settings and data modalities.

poster

Welcome to the **largest national, multi-specialty research** event featuring research from more than 750 medical students, residents, fellows, and international medical graduates.

Use the buttons below or navigation menu to access the event.
 

[![](https://assets.underline.io/markdown_image/1/image/20e1e77aa954efeac7a15d84c85eb383.jpg)](https://underline.io/events/468/sessions?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/068522d81aca0b68a369787c7dfccd1b.jpg)](https://underline.io/events/468/posters?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/d3971ece79d5e5575b178bdc2c8bbfae.jpg)](https://underline.io/events/468/help-desk)
 

#### Event Details

The 2024 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 presented by Laurel Road. 

### Grand Prize 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 winner selection process. The winner will be announced on February 20, 2025 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)](https://www.laurelroad.com/)

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

**There is no registration fee for this event! **

If you have used the AMA Research Challenge platform previously, please click the ‘Log-in’ button below. If you do not remember your password, please click “forgot password” and follow prompts to reset your password. After resetting your password, please refresh your webpage to access the platform.

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals. 

**Please use this link [here](https://www.ama-assn.org/member-benefits/events/ama-research-challenge) to register for free.**

Registration Is Required

Join the AMA Research Challenge poster symposium and semifinals to explore top research posters and score posters in the semifinals (an AMA member exclusive opportunity).

Abstract Title: 
Sex Disparities in a Telestroke Network: From Treatment Times to Outcomes

Background: The structure of the stroke care delivery system in rural South Carolina has been changed by the Medical University of South Carolina tele-stroke program with >40 spoke sites with improved treatment times and outcomes overall but disparities at the level of primary stroke centers and acute stroke ready hospitals have not been well studied. The aim of this study is to retrospectively analyze data from the tele-stroke network for sex differences in stroke care processes at spoke sites to identify disparities in treatment times and outcomes between men and women. 

Methods: Data were collected and analyzed for 21,549 patients between 1/1/2012 and 12/31/2022 for whom a telestroke was paged out. Statistical analysis was completed using SPSS version 28 (IBM corporation, Armonk, NY). Variables with continuous outcomes were assessed using independent samples median tests and categorical variables were assessed using X².

Results: Of 10,713 (49.7%) males and 10,836 (50.3%) females, females were older (68.29±15.5 vs 66.09±13.7, p<0.01) and more likely to be African American (20.4% vs16.8%, p<0.05) as compared to males. Females were more likely to get thrombolytic therapy (29.06% vs 27.2%, p<0.05) but there was no difference in endovascular therapy between both sexes. Treatment times showed that door-to-registration timing (F: 18±27 vs M: 17±26, p<0.05) and door-to-telestroke page timing (F: 22±27 vs M: 20±26, p<0.05) was significantly higher in females compared to males with ultimately door to needle time having the expected disparity between men and women (F: 59±36 vs M: 55±35, p<0.05). Women also had significantly higher mean discharge NIHSS (M: 4.17±6.20 vs F: 4.61±6.70 (p<0.001) and mRS score (M:2.24±1.93 vs F:2.43±1.95 (p<0.001)). More males were discharged home (62.4% vs 60.4%, (p<0.001), and more females were discharged to nursing homes (8.1% vs 6.2%, p<0.001) and hospice (3.9% vs 2.9%, p<0.001) post admission.

Conclusion: Females are older at stroke onset, have longer treatment times, and have poorer outcomes and discharge dispositions. Our findings highlight the opportunities to ameliorate sex disparities in developing telestroke networks in the United States. Future studies should target these disparities to improve outcomes.

Sex Differences in a Telestroke Network: Form Risk Factors to Outcomes

Introduction: Solid-organ transplant recipients (SOTRs) have a higher burden of gynecological cancers as compared to the general population and are likely to develop high-risk oncogenic HPV types due to immunosuppressive therapy. The objective of this study is to evaluate HPV vaccine awareness and uptake among SOTRs, and current attitudes toward the vaccine amongst SOTR care providers.

Methods: This cross-sectional prospective study surveyed SOTR patients and their medical providers on HPV awareness and vaccination (IRB ID: APP-23-00356) at the USC Dermatology clinic and Keck Hospital between April 2023 to May 2024. 

Results: Forty percent (36/91) of patients completed the survey, of whom 18 (50%) knew that HPV is preventable with vaccination and twenty-two (61.1%) participants believed they were at an increased risk for HPV infection. Four (11.1%) participants said their providers recommended the HPV vaccine and four patients (11.1%) reported having received the vaccine. 

Fifty-nine percent of providers (17/ 29) completed the survey, of whom seven (41.2%) providers recommend the HPV vaccine for their SOTR patients. Five (29.4%) providers discuss HPV vaccination status with their SOTR patients, and two (11.8%) stated that they offer the vaccine in their clinic.

Conclusions: Despite a majority of patients believing themselves to be at increased risk for HPV infection, only a minority of patients reported receiving HPV vaccine. Correspondingly, few SOTR care providers offered the HPV vaccine to patients. Our survey highlights the need to increase HPV vaccine access and improve provider guidance for HPV vaccination in SOTRs.


HPV Awareness and Vaccination Among Solid Organ Transplant Patients and Providers

This research investigates the role of time since trauma (TST) in refining trauma team activation (TTA) criteria at a level I trauma center. A retrospective analysis of 1,261 patients presenting after motor vehicle accidents from 2016 to 2021 was conducted, focusing on the relationship between TST and post-emergency department (ED) disposition. The study found that TTA rates decreased with increasing TST, yet patients with a TST of two or more hours showed higher rates of admission, ICU utilization, and surgical interventions. These findings challenge the assumption that TTA need diminishes over time and suggest that current criteria may benefit from refinement to account for longer TST.

Time Delay in Motor Vehicle Accident Arrival Time: A Critical Analysis of Trauma Team Activation

Abstract Title
A Survey on Disabilities and Disability Awareness in General Surgery Residents in the US

Background
Disabilities pose challenges for medical professionals, including surgical residents, impacting their training, performance, and overall well-being. Despite national statistics highlighting the high prevalence of disabilities among adults in the US, there remains a notable gap in understanding how these conditions impact surgical residency programs, particularly regarding the disclosure of disabilities and the utilization of accommodations. This study compared disability prevalence among general surgery residents with national benchmarks, explored disclosure barriers, and assessed disability policies in residency programs.

Methods
This study utilized a descriptive cross-sectional survey conducted among 198 general surgery residents across 323 programs located in urban, suburban, and rural settings. Participants completed a detailed 33-item questionnaire that encompassed demographic data, disability status, practices of disclosure, usage of accommodations, and awareness levels regarding disability policies and training within their residency programs. The survey was administered anonymously via email over two months, from February to April 2024.

Results
Among the respondents, 10.6% reported having physical disabilities, while 20.2% reported mental disabilities; the majority, 68.7%, reported no disabilities. Notably, 62.9% of residents with disabilities had not disclosed their status to program directors, primarily citing fears of discrimination and concerns over the burden of proof. Residents expressed significant apprehension regarding the transparency of accommodation requests and potential biases in evaluations. Moreover, a substantial portion (53.3%) reported receiving no prior training on interacting with individuals with disabilities, and 75.3% were unaware of their residency program's specific disability policies. These findings underscore critical deficiencies in disability
awareness, policy communication, and support mechanisms within surgical residency programs across the United States.

Conclusion
This study highlights pervasive gaps in disability disclosure and support within US surgical residency programs, emphasizing the urgent need for reform. Despite about 1 in 3 surgical residents reporting a disability, disability support seems lacking, impacting the health and wellbeing of residents. Recommendations include integrating comprehensive disability-related training into residency curricula, enhancing transparency of disability policies during the application process, and establishing clear protocols for residents to request accommodations. Addressing these challenges is crucial not only for promoting inclusivity and reducing stigma but also for enhancing professional success and overall well-being of surgical residents with disabilities. Such improvements have the potential to elevate standards of patient care and enhance surgical outcomes. Future research should aim to expand the scope of investigation to include a larger and more diverse sample, as well as explore intersectional factors influencing the accommodation needs of residents with disabilities.

A Survey on Disabilities and Disability Awareness in General Surgery Residents in the US

Background
Antisocial Personality Disorder (ASPD) is one of the most stigmatized psychiatric diagnoses in the public sphere, contrasting with Major Depressive Disorder (MDD) and Generalized Anxiety Disorder (GAD), which are among the least stigmatized. ASPD stigma extends to healthcare, where many providers hold negative attitudes towards patients with ASPD, affecting their willingness to work with this population. Understanding whether medicine perpetuates this stigma is crucial for early intervention in medical training. Given the role of language in perpetuating stigma, as evident in studies on Opioid Use Disorder, this study explored how language contributes to negative attitudes about ASPD in healthcare, an area largely overlooked in stigma research, Psychiatry, and medical education. It aimed to assess medical education resources for stigmatizing language and compare how they reflect public stigma patterns, where ASPD is more stigmatized than MDD and GAD.

Methods
Seven widely used medical education resources were analyzed for their depiction of ASPD, MDD, and GAD. Since many medical students favor third-party resources for studying and board exams, the analysis included Boards and Beyond, AMBOSS, First Aid, First Aid for Psychiatry Clerkship, and APA, as well as in-house materials from Warren Alpert Medical School and the DSM-5. The analysis focused on two stigma components identified by Phelan and Link: labeling and status loss, particularly through clinical pessimism. Fisher's Exact Test compared ASPD to MDD and GAD across labeling, treatment difficulty, and treatment omission.

Results
The analysis revealed significant stigma toward ASPD, with 85.7% (n=6) of the materials using labels like "bad," "manipulative," and "deceitful." Notably, "bad" was used as a mnemonic. In contrast, none of the materials for MDD or GAD employed such labels (p = 0.046). Regarding status loss, 42.9% (n=3) of ASPD materials emphasized treatment difficulty, compared to none for MDD or GAD (p = 0.031). Additionally, 71.4% (n=5) of ASPD resources omitted information on available treatments, while 42.9% (n=3) did so for MDD and GAD. Although the difference in treatment omission was not statistically significant (p = 1.0), the trend still reflects a bias.

Conclusion
The findings suggest that medical education promotes ASPD stigma, mirroring the disproportionate stigma seen in the general public. The use of stigmatizing language indicates that medical education may perpetuate negative attitudes among healthcare providers. Addressing this stigma is crucial to improving care and outcomes for individuals with ASPD. Further research is needed to explore causality and develop strategies to reduce stigma in medical training.

Stigma in Psychiatry Education: A Comparative Analysis of ASPD, MDD, and GAD

ChatGPT and other large language models (LLM) present a promising avenue for enhancing medical education due to its advanced text comprehension and generation capabilities. Motivated by the skyrocketing use of ChatGPT and the lack of curricular mention or discussion on how to best utilize LLMs in medical school, this study investigates the integration of ChatGPT into medical education through a single-session workshop.
Methods
Fifty-nine students from the Emory School of Medicine, 80% first- and second-year medical students and 20% physician assistant students, voluntarily participated in a ChatGPT workshop—a one hour session with learning objectives to understand the framework and construction of a LLM (ChatGPT) and develop comfort in using ChatGPT within medical contexts by generating relevant responses to medical queries in simulated scenarios. Thirty minutes was allocated toward a lecture style presentation and another thirty minutes to hands-on use of ChatGPT. Surveys administered before, immediately after, and eight weeks post-workshop assessed participants' comfort with and likelihood of using ChatGPT, as well as their perceptions of its utility in their future medical careers, ranked on a Likert scale.
Results
ChatGPT usage rose from “never used” and “rare” usage for 76% of participants to usage at least one time per week for 82% of participants at 8 weeks post-workshop, with 23% using ChatGPT four or more days per week. Survey results revealed a significant increase in participants' comfort with using ChatGPT immediately after the workshop, with this improvement persisting at the eight-week follow-up (p<0.05). Moreover, participants reported a significantly higher likelihood of utilizing ChatGPT in their future medical practice after attending the workshop (p<0.05) with the greatest proportion of individuals, 64%, using ChatGPT for explanations of pathophysiology.
Conclusion
This workshop highlights the effectiveness and need of education and training in fostering understanding and utilization of AI technologies like ChatGPT. Despite initial hesitancy, participants demonstrated increased comfort and willingness to incorporate ChatGPT into their medical education and practice. The study underscores the potential of integrating ChatGPT into medical curricula to enhance learning experiences, promote interdisciplinary collaboration, and foster critical thinking skills among students. Additionally, it emphasizes the need for further research to explore the long-term impact and ethical considerations of integrating AI technologies into medical education. Overall, integrating ChatGPT into medical education holds significant implications for the future of healthcare training and practice.

Integrating ChatGPT into Medical Education: A Workshop-Based Approach

Background
Clinical simulations have an important role in medical education, allowing students to gain realistic clinical experience in a didactic manner to improve patient interaction and clinical reasoning skills. However, access to simulations is often limited by resources including physical space in simulation centers and time with standardized patients. ChatGPT is a large language model (LLM) artificial intelligence software that is freely accessible online, that has garnered significant attention in the medical community for its ability to generate relatively accurate medical information. We evaluated the ability of ChatGPT to create high-fidelity virtual clinical simulations and provide simulated patient information and feedback across several clinical scenarios.

Methods
We prompted ChatGPT to generate seven clinical scenarios (fractured arm/intimate partner violence, acute appendicitis, aortic dissection, acetaminophen overdose, acute seizure/metastatic lung cancer, COPD exacerbation, and tension pneumothorax) and create interactive simulations including simulated patient history, physical exam discoveries, lab findings, and imaging descriptions. Each scenario was concluded with feedback and repeated in triplicate to assess consistency. Simulation scenarios were graded for fidelity, evaluated by accuracy, realism, and consistency on a 5-point Likert scale (1-2 = poor, 2-3 = low, 3-4 = moderate, 4-5 = high) by three independent physicians holding the academic rank of Full Professor. A one-sample Wilcoxon signed-rank test was used to assess simulation performance based on a threshold criteria of greater than three out of five points. Fleiss’ Kappa was used to assess interrater reliability. All hypothesis testing was performed using a significance level of p<0.05.

Results
Of the seven simulations evaluated, four were rated high fidelity, two were rated borderline high fidelity, and one was rated moderate fidelity regarding simulation accuracy, realism, and consistency. The mean overall score was 4.0, indicating overall high accuracy, realism, and consistency, significantly higher than hypothesized threshold of 3.0 (p < 0.001). There was a high degree of interrater agreement (ordinal weighted Fleiss κ, 0.763; 95%CI: 0.25-1.0, p<0.05).

Conclusion
Overall, ChatGPT was able to consistently generate accurate and consistent virtual clinical simulations. Our study highlights a framework for a novel use case of LLMs in medical education and demonstrates its utility in providing increased accessibility to simulations for medical students. Although LLMs have known limitations in information accuracy, with the rapid advancement of artificial intelligence technology and advent LLMs that can be trained on specific resources, the use of LLMs such as ChatGPT for further simulation-based medical education represents a promising frontier for medical education.

Evaluating the Utility of a Large Language Model in Simulation-Based Medical Education

Background
Implicit bias, or unconscious attitudes towards minoritized groups, among healthcare providers can contribute to disparities in care, health status, and mortality for minoritized patients. One approach to mitigating the negative effects of these biases on patient care is by implementing anti-racism/anti-bias training into medical education. 

Methods
Rising third year medical students participated in a student-led implicit bias and microaggression training program, including experiential learning through standardized patient (SP) encounters and self-reflection via student-led debrief sessions. The SP encounters simulated instances of xenophobia and perceived language barriers in a hospital setting, in which students were expected to address microaggressions in real time utilizing the VITALS (Validate, Inquire, Take time, Assume the best, Leave opportunities, Speak up for others) framework. Participants completed surveys on attitudes regarding implicit bias prior to participating in the program (T1), after the VITALS video presentation (T2), and after the SP encounter and debrief sessions (T3); surveys were adapted from the VITALS study. 

Results
200 students participated in the program of which 135 linked survey sets were collected for all three time points for data analysis. Results showed that from T1 to T3, there were significant improvements in empathy with targets of bias, comfort in sharing experiences and knowledge and skills to counteract microaggressions. There were no differences by race/ethnicity, language exposure, or religious adherence.

Conclusion
Our implicit bias training, equipped students with the tools and practice needed to interrupt microaggressions in the clinical setting and helped improve their understanding and likelihood of engaging in anti-racism allyship.

Efficacy of Standardized Patient Encounters to Reinforce Implicit Bias Training in the Clinical Setting

Background
In recent years, active learning methods have gained attention for improving material retention and teamwork, especially in education. Despite such benefits, learner engagement remains a challenge, prompting exploration into innovative solutions like game-based instruction. By using game principles, educators can sustain learner interest and encourage participation. Building on these principles, the Texas Health Presbyterian Dallas Internal Medicine Residency program implemented a “House Cup” system where points were awarded for educational activities to promote friendly competition and active learning.

Methods
Residents were evenly distributed into six teams. Points were earned based on criteria including: number of admissions on call days, attendance at lectures, Medical Knowledge Self Assessment Program (MKSAP) review, and the number of procedures performed. Data from January through March of 2022, 2023, and 2024 was collected. Surveys were utilized to gather information on the time spent on MKSAP preparation before and after implementation. Qualitative variables were analyzed with chi-square tests; quantitative variables with ANOVA or Kruskal-Wallis. Statistical significance was determined at an alpha level of 0.05.

Results
Prior to the “House Cup” system (pre-group), there was an average of 5.30 admissions (± 1.15). Following implementation (post-group), admissions rose to an average of 7.87 (± 1.68). Lecture attendance remained stable, with a median of 83 (IQR: 70.75-93) in the pre-group and 85 (IQR: 77-93) in the post-group. MKSAP questions completion and review time in the post-group (mode 51+ questions, 4-5 hours) increased compared to the pre-group (mode 0 questions, 0-1 hours). Number of procedures performed in 2024 (54) surpassed both 2022 (7) and 2023 (16).

Conclusion
Our aim was to assess the impact of game-based learning techniques on material retention, engagement, and teamwork within resident education. Integrating game-based learning into educational activities led to increased participation in call day admissions, lecture attendance, MKSAP review, and number of procedures performed. Implementing game-based learning resulted in improved performance across all metrics except lecture attendance, which was already high. These promising results benefit our residency program and offer insights for others seeking to incentivize and motivate residents


Level Up: Exploring the Effectiveness of Game Based Learning in Graduate Medical Education

Abstract Title: Association Between Utilization of Gender-Affirming Care and Insurance Payer, State, and State Policy

Background: Gender-affirming surgery/hormone therapy (GAS/GAHT) play an essential role in improving quality of life for many transgender and gender-diverse (TGD) individuals. Despite increased demand for GAS/GAHT, TGD individuals face many obstacles to care. We investigated the association between utilization rates of GAS/GAHT and insurance payer, state, and state-level nondiscrimination policies, which prohibit health plans from precluding care based on gender identity or expression. With the disparate legislative climate, it is imperative for stakeholders to understand how these factors influence national utilization rates of GAS/GAHT.

Methods: We retrospectively analyzed a cohort of TGD individuals of all ages to characterize GAS/GAHT utilization between 2010-2020 using the PearlDiver-Mariner Database, which contains all-payer claims for 151 million patients in all 50 states and DC. We used binary logistic regressions to evaluate associations between insurance payers (private, Medicaid, Medicare) and GAS/GAHT, while testing for effect modification by individual states. Furthermore, we designed a difference-in-differences analysis to compare GAS/GAHT utilization among those who did and did not live in states that passed policies prohibiting healthcare access discrimination by insurance payers.

Results: 77,366 individuals were included in the study. The distribution of insurance payers was as follows: 77% private, 16% Medicaid, 7% Medicare. GAS/GAHT utilization was highest among the privately insured and lowest among the Medicare subgroups. Individuals covered by Medicaid (OR=0.78,95% CI=0.73-0.83) or Medicare (OR=0.69,95% CI=0.63-0.75) were less likely to undergo GAS compared to the privately insured. Similar differences in GAHT utilization were observed. Effect modification by state was evident in both outcomes. For example, differences in GAHT utilization rates by insurance payers were smaller among those living in Illinois (38% private, 37% Medicaid, 33% Medicare) compared to Idaho (35% private, 15% Medicaid, 23% Medicare). Quasi-experimental analysis suggested that nondiscrimination policies were associated with increased national all-payer treatment utilization, especially GAS, which more than doubled after implementation.

Conclusion: Our study highlights three main findings: 1) Across all states, individuals covered by Medicaid/Medicare received GAS/GAHT less often compared to the privately insured, 2) Some individual states had minimal insurance payer discrepancies in GAS/GAHT utilization, and 3) While these insurance payer differences in utilization remain at the national level, individuals living in states that implemented nondiscrimination policies experienced an increase in GAS/GAHT post-implementation. Overall, nondiscrimination policies implemented in 21 states and DC increased GAS/GAHT utilization, but challenges at the insurance payer and individual state level continue to threaten TGD individuals’ healthcare and lives.

Premium content

Downloads

Next from AMA Research Challenge 2024

Sex Differences in a Telestroke Network: Form Risk Factors to Outcomes

Similar lecture

Automatic Patient Note Assessment without Strong Supervision