Journal of General Internal Medicine
Top medRxiv preprints most likely to be published in this journal, ranked by match strength.
Show abstract
BackgroundShort-video platforms have become increasingly important sources of health information for the general public. However, the informational quality and dissemination patterns of content related to specific therapeutic modalities, such as enhanced external counterpulsation (EECP), remain insufficiently characterized. This study aimed to evaluate the informational quality of EECP-related videos on a short-video platform and to examine the relationship between content quality and user engag...
Show abstract
BackgroundThe physician assistant (PA) workforce has expanded rapidly in the United States, increasing the importance of effective physician-PA collaboration. Although PAs improve patient outcomes and access to care, the determinants of effective collaboration has not been well studied. North Carolina provides a relevant context due to its growing PA workforce and supervisory regulatory structure, in which physicians retain administrative responsibility for PA supervision across practice setting...
Show abstract
IntroductionEighteenth-century medical texts document a formative period in the evolution of clinical reasoning, yet their integration into modern medical education is limited. The traditional approach to learning the history of medicine has naturally focused on passive reading, but new approaches using AI could enable learners to interrogate and simulate the historical diagnostic logic and therapeutic paradigms. More specifically, large language models (LLMs) offer an opportunity to create inte...
Show abstract
ObjectiveAlthough trauma-focused psychotherapies are effective for posttraumatic stress disorder (PTSD), recovery under routine outpatient conditions remains variable. We examined whether participation in a structured Specialty Care (SC) model integrating clinician specialization, flexible treatment density, and coordinated navigation was associated with accelerated PTSD recovery compared with standard outpatient care. MethodsWe conducted a retrospective matched cohort study (2024-2025) of U.S....
Show abstract
IntroductionHealthcare organizations have begun incorporating screening procedures for social determinants of health (SDOH) into care, recognizing the impact these factors can have on health outcomes. We aimed to present methods for evaluating redundancy in the risk information gained across SDOH questions and for evaluating whether demographic biases are present in whether patients were asked SDOH questions and whether they declined to answer them. MethodsSDOH question data were analyzed for 1...
Show abstract
ObjectivesArtificial intelligence (AI)-enabled digital stethoscopes combine phonocardiography and electrocardiography to support detection of cardiac rhythm and structural abnormalities. This study evaluated the feasibility and exploratory diagnostic performance of AI-guided cardiac auscultation during routine general practice consultations and home visits. MethodsIn this prospective feasibility study, 50 consecutive patients aged [≥]65 years underwent AI-assisted auscultation using the Eko ...
Show abstract
BackgroundApproximately 33% of U.S. Veterans live in rural areas, often facing significant barriers to accessing healthcare due to staffing shortages at VA facilities. The Contract Buyout (CBO) program, authorized under the PACT Act of 2022, was designed to address rural healthcare staffing shortages by enabling Veterans Health Administration (VHA) facilities to buy out existing service contracts to work in rural VA facilities. Despite its potential, uptake of the program has been limited, with ...
Show abstract
AbstractAccurate health information is ineffective if patients cannot understand it. Large Language Model (LLM) health research values veridical precision; however, linguistic accessibility remains an under-examined component of output quality and usability. This study investigated two sources of variability in readability classification: differences across LLM systems and across readability metrics. The analysis tested 1,120 data points from seven systems in English and Portuguese, comparing ba...
Show abstract
Smoking Cessation Efforts for Patients with Asthma and COPD IntroductionSmoking cessation can alter the natural history of both COPD and asthma by reducing the frequency and severity of exacerbations and slowing disease progression. Accordingly, the Global Initiative for Asthma and the Global Initiative for Chronic Obstructive Lung Disease recommend that clinicians address smoking cessation at every visit using counseling and pharmacotherapy. MethodsThe Mount Sinai Health System includes seven...
Show abstract
PurposeThe objective of this study was to identify predictors of statin adherence in the primary and secondary prevention of CVD among patients in the first two years after the date of first prescription using real-world data. MethodsThe Electronic Practice Based Research Network Linked Dataset was used in this study. Statin adherence was calculated using a modified proportion of days covered (PDC) formula. Individuals with PDC [≥] 80% during the two years of observation period were consider...
Show abstract
We developed and validated a self-administered clinical vignette platform powered by a large language model (LLM), deployed through a SurveyCTO web survey, to measure primary health care provider competencies in Vietnam. In a pilot focus group, nine physicians rated LLM-simulated patient interactions as realistic (mean 3.78/5) and user-friendly. In the validation phase, 22 providers completed 132 vignette interactions across ten clinical scenarios in Vietnamese. Essential diagnostic checklist sc...
Show abstract
Community health workers (CHWs) in low-resource settings deliver variable-quality care. This study used OpenAIs o3 and Googles Gemini Flash 2.5 to evaluate whether large language models (LLMs) listening to CHW-patient interactions could generate accurate referral decisions. Across 150 participating Rwandan CHWs, 429 encounters were recorded (in Kinyarwanda) and then processed by LLMs. CHWs demonstrated high referral accuracy (97.9% [95% CI: 96.1%-98.9%]), and OpenAIs o3 performed similarly to C...
Show abstract
BackgroundBehavioral telemetry--the analysis of clinical actions NOT taken--may identify care process failures associated with adverse outcomes. While missed nursing care predicts outcomes in survey-based studies, objective EHR-derived measures are lacking. We hypothesized that missing routine cognitive assessment in ICU patients with low acute physiologic derangement would predict mortality independent of illness severity. MethodsRetrospective cohort study using MIMIC-IV (2008-2022, Beth Israe...
Show abstract
BackgroundHypertension guidelines recommend the use of automated BP devices over manual devices to reduce observer bias, such as terminal digit preference. We aimed to evaluate systematic differences in BP readings and the association with incident stroke according to type of measurement. MethodsUsing de-identified electronic health record data from Optum Labs Data Warehouse from primary care visits in 2024, we classified providers BP measurement method using proportion of odd terminal digit pr...
Show abstract
BackgroundAdvanced Practice Providers (APPs) in emergency and urgent care settings experience high burnout rates, yet limited research examines cognitive factors influencing professional fulfillment. The Empowerment Dynamic framework suggests outcome-oriented thinking may protect against burnout compared to problem-oriented patterns. ObjectiveTo examine relationships between cognitive mindset orientation, professional fulfillment, and burnout among APPs while providing preliminary validation of...
Show abstract
BackgroundArtificial intelligence chatbots (AICs) are increasingly being integrated into scholarly publishing, with the potential to automate routine editorial tasks and streamline workflows. In traditional, complementary, and integrative medicine (TCIM) publishing, editorial and peer review processes can be particularly complex due to diverse methodologies and culturally embedded knowledge systems, presenting unique opportunities and challenges for AIC adoption. MethodsAn anonymous, online cro...
Show abstract
IntroductionPhysician assistant (PA) programs face persistent challenges in recruiting and retaining clinical preceptors due to time constraints, administrative burden, lack of compensation, and limited training. Additional pressures, such as health care consolidation, program expansion, clinician burnout, and financial implications of paid clinical sites, further strain preceptorship capacity. This study examines motivators and barriers influencing clinicians willingness to precept PA students....
Show abstract
BackgroundThe 340B Drug Pricing Program was established to expand access to care for low-income and uninsured patients by allowing safety-net hospitals and clinics to purchase outpatient drugs at discounted prices. Over time, the program has expanded substantially, raising questions about whether participating hospitals are meeting the programs intended objectives. MethodsUsing 2023 hospital financial data from the RAND Corporation, we conducted cross-sectional descriptive comparisons of 340B a...
Show abstract
BackgroundDementia affects over 57 million people worldwide. UK and international policy position personalised, conversation-based care planning as central to post-diagnostic support. However, delivery in primary care is inconsistent, and many practitioners lack dementia-specific communication training. Existing evidence focuses on single roles or settings, leaving a gap in understanding how communication operates across the primary care workforce. AimsTo identify what helps and hinders effecti...
Show abstract
Natural language processing was applied to 3,586 Australian health practitioner tribunal decisions (1999-2026) to identify patterns in professional misconduct, outcomes, and temporal trends at a scale impractical through manual analysis. A text classification approach categorised 2,428 disciplinary decisions across seven misconduct types with acceptable accuracy for the major categories (per-class F1 0.47-0.82). Boundary violations were the most prevalent misconduct type (30.2%), followed by dis...