How the other half screens: A model for partnerships between student-run free clinics and genetic counseling programs to address disparities in hereditary cancer evaluation.

Genetic medicine is considered a major part of the future of preventative care, offering evidence-based, effective interventions to improve health outcomes and reduce morbidity and mortality, especially regarding hereditary cancer screening. Identification of individuals who would benefit from screening is key to improving their cancer-related healthcare outcomes. However, patients without insurance, of historically underserved races, of lower socioeconomic status, and in rural communities have lower access to such care. Barriers to access lead to populations having higher rates of undetected hereditary cancer, and consequently more severe forms of cancer. With an already-established reach, student-run free clinics can work with genetic counseling training programs to incorporate genetic medicine into their workflow. Such partnerships will (1) make genetic care more accessible with goals of improving patient morbidity, mortality, and health outcomes, (2) offer robust educational experiences for genetic counseling learners, particularly in understanding social determinants of health and barriers to care, and (3) actively combat the growing racial and geographic gaps in genetic care. Our study presents how one student-run free clinic implemented genetic counseling into its primary care workflow to improve access to genetics services. We present two examples of how genetic counseling improved patients' medical care. We also identify obstacles encountered during this program's development, as well as solutions-those we incorporated and possible considerations for other clinics. With the hope that other clinics can use this paper to design similar partnerships, we aim to lessen the gap between sickness and screening.

Results of Leveraging Pharmaceutical Patient Assistance Programs to Expand Access to High Cost Medications in a Student-Run Free Clinic.

High costs make many medications inaccessible to patients in the United States. Uninsured and underinsured patients are disproportionately affected. Pharmaceutical companies offer patient assistance programs (PAPs) to lower the cost-sharing burden of expensive prescription medications for uninsured patients. PAPs are used by various clinics, particularly oncology clinics and those caring for underserved communities, to expand patients' access to medications. Prior studies describing the implementation of PAPs in student-run free clinics have demonstrated cost-savings during the first few years of using PAPs. However, there is a lack of data regarding the efficacy and cost savings of longitudinal use of PAPs across several years. This study describes the growth of PAP use at a student-run free clinic in Nashville, Tennessee over ten years, demonstrating that PAPs can be used reliably and sustainably to expand patients' access to expensive medications. From 2012 to 2021, we increased the number of medications available through PAPs from 8 to 59 and the number of patient enrollments from 20 to 232. In 2021, our PAP enrollments demonstrated potential cost savings of over $1.2 million. Strategies, limitations, and future directions of PAP use are also discussed, highlighting that PAPs can be a powerful tool for free clinics in serving underserved communities.

Patients and Guns: Florida Physicians Are Not Asking.

OBJECTIVES: The number of deaths from gun violence continues to increase in the United States. Despite multiple studies demonstrating that counseling patients leads to safer gun storage, it is not routinely practiced by physicians. There are multiple barriers to discussing firearms with patients. A barrier in Florida, until recently, was a law preventing physicians from asking patients about firearms. The law was overturned in 2017; however, it is unclear whether physicians are aware of this decision. We undertook a survey to study University of Florida faculty physicians' knowledge, attitudes, and practices related to discussing firearms safety. METHODS: The survey consisted of 15 questions related to firearms and counseling. Invitations to participate were e-mailed in 2018 to faculty in general internal medicine, emergency medicine, and surgery within our institution. RESULTS: The response rate was 50% (n = 71/142). The majority of faculty surveyed did not own a gun (56%). Ninety-one percent of faculty surveyed agreed that "gun violence is a public health issue" and 93% agreed that gun safety discussion with patients at risk for suicidal or violent behavior is important. More than half of the respondents (62%) believed they could effectively discuss firearms safety with patients; 73% strongly agreed or agreed that they would discuss gun safety with at-risk patients, whereas 27% were either neutral or disagreed. Fewer still (55%) feel comfortable initiating conversations, and only 5% of participants always talk to at-risk patients about gun safety. Twenty-four percent discussed gun safety most of the time, 30% discussed it sometimes, 32% rarely discussed it, and 9% never discussed it; 76% were aware of the 2017 court decision overturning the physician gag law in Florida. The most-often cited barriers to discussions included lack of time (36%), worry about negative reaction from patient (30%), worry about lack of support from administration (13%), and lack of knowledge (20%). Gun owners and nonowners differed significantly on only two survey items: having taken a firearms safety course (gun owners more likely, relative risk 1.63, 95% confidence interval 1.16-2.29, P = 0.001) and agreeing with gun violence being a public health issue (gun owners being less likely, relative risk 1.24, 95% confidence interval 1.03-1.49, P = 0.006). CONCLUSIONS: Faculty miss opportunities to prevent gun violence despite acknowledging that it is important to do so. More than 40% of the physicians who were surveyed do not counsel at-risk patients about gun safety, citing a lack of knowledge, a persisting belief that asking patients about guns in Florida is illegal, worry about negative patient reactions, and time limitations. Inaction persists despite increased awareness and activism by physicians regarding gun violence. A wider availability of continuing medical education opportunities to learn about firearms counseling should be considered.

Leveraging large language models for generating responses to patient messages-a subjective analysis.

OBJECTIVE: This study aimed to develop and assess the performance of fine-tuned large language models for generating responses to patient messages sent via an electronic health record patient portal. MATERIALS AND METHODS: Utilizing a dataset of messages and responses extracted from the patient portal at a large academic medical center, we developed a model (CLAIR-Short) based on a pre-trained large language model (LLaMA-65B). In addition, we used the OpenAI API to update physician responses from an open-source dataset into a format with informative paragraphs that offered patient education while emphasizing empathy and professionalism. By combining with this dataset, we further fine-tuned our model (CLAIR-Long). To evaluate fine-tuned models, we used 10 representative patient portal questions in primary care to generate responses. We asked primary care physicians to review generated responses from our models and ChatGPT and rated them for empathy, responsiveness, accuracy, and usefulness. RESULTS: The dataset consisted of 499 794 pairs of patient messages and corresponding responses from the patient portal, with 5000 patient messages and ChatGPT-updated responses from an online platform. Four primary care physicians participated in the survey. CLAIR-Short exhibited the ability to generate concise responses similar to provider's responses. CLAIR-Long responses provided increased patient educational content compared to CLAIR-Short and were rated similarly to ChatGPT's responses, receiving positive evaluations for responsiveness, empathy, and accuracy, while receiving a neutral rating for usefulness. CONCLUSION: This subjective analysis suggests that leveraging large language models to generate responses to patient messages demonstrates significant potential in facilitating communication between patients and healthcare providers.

Using large language model to guide patients to create efficient and comprehensive clinical care message.

OBJECTIVE: This study aims to investigate the feasibility of using Large Language Models (LLMs) to engage with patients at the time they are drafting a question to their healthcare providers, and generate pertinent follow-up questions that the patient can answer before sending their message, with the goal of ensuring that their healthcare provider receives all the information they need to safely and accurately answer the patient's question, eliminating back-and-forth messaging, and the associated delays and frustrations. METHODS: We collected a dataset of patient messages sent between January 1, 2022 to March 7, 2023 at Vanderbilt University Medical Center. Two internal medicine physicians identified 7 common scenarios. We used 3 LLMs to generate follow-up questions: (1) Comprehensive LLM Artificial Intelligence Responder (CLAIR): a locally fine-tuned LLM, (2) GPT4 with a simple prompt, and (3) GPT4 with a complex prompt. Five physicians rated them with the actual follow-ups written by healthcare providers on clarity, completeness, conciseness, and utility. RESULTS: For five scenarios, our CLAIR model had the best performance. The GPT4 model received higher scores for utility and completeness but lower scores for clarity and conciseness. CLAIR generated follow-up questions with similar clarity and conciseness as the actual follow-ups written by healthcare providers, with higher utility than healthcare providers and GPT4, and lower completeness than GPT4, but better than healthcare providers. CONCLUSION: LLMs can generate follow-up patient messages designed to clarify a medical question that compares favorably to those generated by healthcare providers.

Leveraging Large Language Models for Generating Responses to Patient Messages.

Objective: This study aimed to develop and assess the performance of fine-tuned large language models for generating responses to patient messages sent via an electronic health record patient portal. Methods: Utilizing a dataset of messages and responses extracted from the patient portal at a large academic medical center, we developed a model (CLAIR-Short) based on a pre-trained large language model (LLaMA-65B). In addition, we used the OpenAI API to update physician responses from an open-source dataset into a format with informative paragraphs that offered patient education while emphasizing empathy and professionalism. By combining with this dataset, we further fine-tuned our model (CLAIR-Long). To evaluate the fine-tuned models, we used ten representative patient portal questions in primary care to generate responses. We asked primary care physicians to review generated responses from our models and ChatGPT and rated them for empathy, responsiveness, accuracy, and usefulness. Results: The dataset consisted of a total of 499,794 pairs of patient messages and corresponding responses from the patient portal, with 5,000 patient messages and ChatGPT-updated responses from an online platform. Four primary care physicians participated in the survey. CLAIR-Short exhibited the ability to generate concise responses similar to provider's responses. CLAIR-Long responses provided increased patient educational content compared to CLAIR-Short and were rated similarly to ChatGPT's responses, receiving positive evaluations for responsiveness, empathy, and accuracy, while receiving a neutral rating for usefulness. Conclusion: Leveraging large language models to generate responses to patient messages demonstrates significant potential in facilitating communication between patients and primary care providers.