Large Language Model Influence on Management Reasoning: A Randomized Controlled Trial.

Importance: Large language model (LLM) artificial intelligence (AI) systems have shown promise in diagnostic reasoning, but their utility in management reasoning with no clear right answers is unknown. Objective: To determine whether LLM assistance improves physician performance on open-ended management reasoning tasks compared to conventional resources. Design: Prospective, randomized controlled trial conducted from 30 November 2023 to 21 April 2024. Setting: Multi-institutional study from Stanford University, Beth Israel Deaconess Medical Center, and the University of Virginia involving physicians from across the United States. Participants: 92 practicing attending physicians and residents with training in internal medicine, family medicine, or emergency medicine. Intervention: Five expert-developed clinical case vignettes were presented with multiple open-ended management questions and scoring rubrics created through a Delphi process. Physicians were randomized to use either GPT-4 via ChatGPT Plus in addition to conventional resources (e.g., UpToDate, Google), or conventional resources alone. Main Outcomes and Measures: The primary outcome was difference in total score between groups on expert-developed scoring rubrics. Secondary outcomes included domain-specific scores and time spent per case. Results: Physicians using the LLM scored higher compared to those using conventional resources (mean difference 6.5 %, 95% CI 2.7-10.2, p<0.001). Significant improvements were seen in management decisions (6.1%, 95% CI 2.5-9.7, p=0.001), diagnostic decisions (12.1%, 95% CI 3.1-21.0, p=0.009), and case-specific (6.2%, 95% CI 2.4-9.9, p=0.002) domains. GPT-4 users spent more time per case (mean difference 119.3 seconds, 95% CI 17.4-221.2, p=0.02). There was no significant difference between GPT-4-augmented physicians and GPT-4 alone (-0.9%, 95% CI -9.0 to 7.2, p=0.8). Conclusions and Relevance: LLM assistance improved physician management reasoning compared to conventional resources, with particular gains in contextual and patient-specific decision-making. These findings indicate that LLMs can augment management decision-making in complex cases. Trial registration: ClinicalTrials.gov Identifier: NCT06208423; https://classic.clinicaltrials.gov/ct2/show/NCT06208423.

Influence of a Large Language Model on Diagnostic Reasoning: A Randomized Clinical Vignette Study.

Importance: Diagnostic errors are common and cause significant morbidity. Large language models (LLMs) have shown promise in their performance on both multiple-choice and open-ended medical reasoning examinations, but it remains unknown whether the use of such tools improves diagnostic reasoning. Objective: To assess the impact of the GPT-4 LLM on physicians' diagnostic reasoning compared to conventional resources. Design: Multi-center, randomized clinical vignette study. Setting: The study was conducted using remote video conferencing with physicians across the country and in-person participation across multiple academic medical institutions. Participants: Resident and attending physicians with training in family medicine, internal medicine, or emergency medicine. Interventions: Participants were randomized to access GPT-4 in addition to conventional diagnostic resources or to just conventional resources. They were allocated 60 minutes to review up to six clinical vignettes adapted from established diagnostic reasoning exams. Main Outcomes and Measures: The primary outcome was diagnostic performance based on differential diagnosis accuracy, appropriateness of supporting and opposing factors, and next diagnostic evaluation steps. Secondary outcomes included time spent per case and final diagnosis. Results: 50 physicians (26 attendings, 24 residents) participated, with an average of 5.2 cases completed per participant. The median diagnostic reasoning score per case was 76.3 percent (IQR 65.8 to 86.8) for the GPT-4 group and 73.7 percent (IQR 63.2 to 84.2) for the conventional resources group, with an adjusted difference of 1.6 percentage points (95% CI -4.4 to 7.6; p=0.60). The median time spent on cases for the GPT-4 group was 519 seconds (IQR 371 to 668 seconds), compared to 565 seconds (IQR 456 to 788 seconds) for the conventional resources group, with a time difference of -82 seconds (95% CI -195 to 31; p=0.20). GPT-4 alone scored 15.5 percentage points (95% CI 1.5 to 29, p=0.03) higher than the conventional resources group. Conclusions and Relevance: In a clinical vignette-based study, the availability of GPT-4 to physicians as a diagnostic aid did not significantly improve clinical reasoning compared to conventional resources, although it may improve components of clinical reasoning such as efficiency. GPT-4 alone demonstrated higher performance than both physician groups, suggesting opportunities for further improvement in physician-AI collaboration in clinical practice.

Understanding the Complex Relationship Between One's Body, Eating, Exercise, and Gender-Affirming Medical Care Among Transgender and Nonbinary Adolescents and Young Adults.

Purpose: Gender dysphoria has been linked to body dissatisfaction, which can affect an individual's eating and exercise habits and increase their risk for disordered eating. The prevalence of eating disorders among transgender and nonbinary (TGNB) adolescents and young adults (AYA) ranges from 5% to 18% and studies have found a higher risk of disordered eating among these AYA in comparison to their cisgender peers. However, there is minimal research on why TGNB AYA are at higher risk. The aim of this study is to understand unique factors that define a TGNB AYA's relationship between their body and food, how this relationship may be affected by gender-affirming medical care, and how these relationships may contribute to disordered eating. Methods: A total of 23 TGNB AYA were recruited from a multidisciplinary gender-affirming clinic to participate in semistructured interviews. Transcripts were analyzed using Braun and Clarke's theory of thematic analysis (2006). Results: The average age of participants was 16.9 years. Forty-four percent of participants identified as having a transfeminine gender identity, 39% transmasculine, and 17% nonbinary/gender fluid. Five themes emerged regarding TGNB participants' relationship to food and exercise: gender dysphoria and control over one's body, societal expectations of gender, mental health and safety concerns, emotional and physical changes with gender-affirming medical care, and recommended resources for TGNB AYA. Conclusion: By understanding these unique factors, clinicians can provide targeted and sensitive care when screening and managing disordered eating among TGNB AYA.

Gender Diverse Youth on Fertility and Future Family: A Qualitative Analysis.

PURPOSE: Gender-affirming treatment for transgender and nonbinary adolescents has been shown to decrease anxiety, depression, and suicidality, but treatments have medical consequences. Specifically, hormone replacement and pubertal blocking may impact patients' fertility and childbearing capabilities. We interviewed gender diverse adolescents regarding their thoughts on family and fertility. METHODS: We completed semistructured interviews with 23 gender diverse adolescents recruited from the Seattle Children's Gender Clinic. Interviewees included transfeminine, transmasculine, and nonbinary youth. Interviews were recorded, transcribed, and analyzed using Braun and Clarke's theory of thematic analysis, a flexible framework for qualitative analysis. RESULTS: Gender diverse adolescents have myriad views on fertility, but four main themes were identified: (1) an interest in future family, including ideas regarding adoption and biological children; (2) barriers to fertility, including cost and procedure-related dysphoria; (3) factors unique to the developmental stage of adolescents, including the age discordance of making fertility decisions as a teenager and parental influence on decision-making; and (4) suggestions for clinicians approaching fertility counseling with adolescents considering hormone therapy. CONCLUSIONS: Many gender diverse youth asserted an interest in building families, although the process of fertility preservation remains fraught. Relative to other studies, our participants were hopeful, imaginative, and interested in having children. Participants wanted to receive specific counseling on fertility, to receive help navigating the logistics of fertility preservation, and to be listened to when their hopes for children (or no children) were stated. Further research is needed to create care paradigms that address fertility of transgender youth in an affirming, developmentally appropriate manner.

How Are Transgender and Gender Nonconforming Youth Affected by the News? A Qualitative Study.

PURPOSE: News stories about transgender and gender nonconforming (TGNC) people have become more common in recent years and TGNC youth may be disparately affected by this information compared with their cisgender peers. The aim of this study was to understand how TGNC youth react to news stories about TGNC and other gender and sexual minority (GSM) identifying persons. METHODS: As part of a larger study, 23 TGNC youth were recruited from the Seattle Children's Gender Clinic to participate in a semistructured interview. Interviews were audio recorded, transcribed, and thematically analyzed. RESULTS: Participating youth ranged in age from 13 to 19 years (mean = 16.9 years). Of the 23 participants, 43% identified as having a transfeminine gender identity, 39% transmasculine, and 17% nonbinary/gender fluid. Four main themes were identified: (1) news coverage of current political climate affects gender transition; (2) negative news coverage of GSM people contributes to concerns about mental well-being and safety; (3) geographical location affects perception of news; and (4) positive news coverage of TGNC people increases visibility and hope. CONCLUSIONS: Exposure to negative news about GSM people may contribute to increased levels of stress among TGNC youth. With increased visibility, participants described frustration associated with inaccurate portrayals of the TGNC community in the news; however, with increased visibility, there is also a growing sense of shared community and opportunity for acceptance of TGNC people. Given the scope of responses to negative and positive news on GSM people, TGNC youth may benefit from increased support to promote resilience when interpreting the news.

A consensus-based process identifying physical therapy and exercise treatments for patients with degenerative meniscal tears and knee OA: the TeMPO physical therapy interventions and home exercise program.

BACKGROUND: Knee osteoarthritis (OA) is prevalent and often associated with meniscal tear. Physical therapy (PT) and exercise regimens are often used to treat OA or meniscal tear, but, to date, few programs have been designed specifically for conservative treatment of meniscal tear with concomitant knee OA. Clinical care and research would be enhanced by a standardized, evidence-based, conservative treatment program and the ability to study the effects of the contextual factors associated with interventions for patients with painful, degenerative meniscal tears in the setting of OA. This paper describes the process of developing both a PT intervention and a home exercise program for a randomized controlled clinical trial that will compare the effectiveness of these interventions for patients with knee pain, meniscal tear and concomitant OA. METHODS: This paper describes the process utilized by an interdisciplinary team of physical therapists, physicians, and researchers to develop and refine a standardized in-clinic PT intervention, and a standardized home exercise program to be carried out without PT supervision. The process was guided in part by Medical Research Council guidance on intervention development. RESULTS: The investigators achieved agreement on an in-clinic PT intervention that included manual therapy, stretching, strengthening, and neuromuscular functional training addressing major impairments in range of motion, musculotendinous length, muscle strength and neuromotor control in the major muscle groups associated with improving knee function. The investigators additionally achieved agreement on a progressive, protocol-based home exercise program (HEP) that addressed the same major muscle groups. The HEP was designed to allow patients to perform and progress the exercises without PT supervision, utilizing minimal equipment and a variety of methods for instruction. DISCUSSION: This multi-faceted in-clinic PT program and standardized HEP provide templates for in-clinic and home-based care for patients with symptomatic degenerative meniscal tear and concomitant OA. These interventions will be tested as part of the Treatment of Meniscal Tear in Osteoarthritis (TeMPO) Trial. TRIAL REGISTRATION: The TeMPO Trial was first registered at clinicaltrials.gov with registration No. NCT03059004 on February 14, 2017. TeMPO was also approved by the Institutional Review Board at Partners HealthCare/Brigham and Women's Hospital.

Willingness of older adults to participate in a randomized trial of conservative therapies for knee pain: A prospective preference assessment.

BACKGROUND: In preparation for a trial of physical therapy (PT) for patients with degenerative meniscal tear and knee osteoarthritis, we conducted a prospective preference assessment -- a methodology for estimating the proportion of eligible subjects who would participate in a hypothetical randomized trial. METHODS: We identified patients seeking care from the practices of five orthopedic surgeons. Patients completed a survey asking about their willingness to participate in a hypothetical trial, their treatment preferences, their knee pain, and demographic variables. RESULTS: We approached 201 eligible patients, of whom 67% (95% confidence interval [CI] 60%, 73%) completed questionnaires. Of these, 24% (95% CI 17%, 31%) were definitely and 39% (95% CI 31%, 47%) were probably willing to participate in the trial. Thirty-three percent (95% CI 23%, 43%) of subjects with no treatment preference were definitely willing to participate as compared to 9% (95% CI 1%, 17%) with treatment preference (p = .001). Patients with higher educational attainment also stated a greater willingness to participate than those with less education (p = .06). In multivariable logistic regression analysis, those with no treatment preferences had greater adjusted odds of stating they would definitely participate than those with a defined treatment preference (OR 5.2, 95% CI 1.7, 16.2), while subjects with an associate's degree or greater were more likely to state they would definitely participate than those with less education (OR 3.9, 95% CI 1.1, 14.1). CONCLUSION: In this prospective preference assessment, 63% (95% CI 55%, 71%) of subjects with degenerative meniscal tear expressed willingness to participate in a trial of PT modalities. Individuals with no treatment preferences were more likely to state they would participate than were those with higher education. This methodology can help investigators estimate recruitment rates, anticipate generalizability of the trial sample and create strategies to facilitate enrollment.

Disparities in Total Knee Replacement: Population Losses in Quality-Adjusted Life-Years Due to Differential Offer, Acceptance, and Complication Rates for African Americans.

OBJECTIVE: Total knee replacement (TKR) is an effective treatment for end-stage knee osteoarthritis (OA). American racial minorities undergo fewer TKRs than whites. We estimated quality-adjusted life-years (QALYs) lost for African American knee OA patients due to differences in TKR offer, acceptance, and complication rates. METHODS: We used the Osteoarthritis Policy Model, a computer simulation of knee OA, to predict QALY outcomes for African American and white knee OA patients with and without TKR. We estimated per-person QALYs gained from TKR as the difference between QALYs with current TKR use and QALYs when no TKR was performed. We estimated average, per-person QALY losses in African Americans as the difference between QALYs gained with white rates of TKR and QALYs gained with African American rates of TKR. We calculated population-level QALY losses by multiplying per-person QALY losses by the number of persons with advanced knee OA. Finally, we estimated QALYs lost specifically due to lower TKR offer and acceptance rates and higher rates of complications among African American knee OA patients. RESULTS: African American men and women gain 64,100 QALYs from current TKR use. With white offer and complications rates, they would gain an additional 72,000 QALYs. Because these additional gains are unrealized, we call this a loss of 72,000 QALYs. African Americans lose 67,500 QALYs because of lower offer rates, 15,800 QALYs because of lower acceptance rates, and 2,600 QALYs because of higher complication rates. CONCLUSION: African Americans lose 72,000 QALYs due to disparities in TKR offer and complication rates. Programs to decrease disparities in TKR use are urgently needed.

Participant-Reported Benefits of Involvement in an Adaptive Sports Program: A Qualitative Study.

BACKGROUND: Although participation in adaptive sports offers numerous benefits for persons with disabilities, a substantial number of eligible persons do not take part. Previous studies have identified personal and environmental factors that promote or inhibit adaptive sports participation. However, these studies have considered a relatively narrow range of factors. OBJECTIVE: To use qualitative research techniques to identify novel factors that influence participation in a community-based adaptive sports program. DESIGN: Qualitative focus group study. SETTING: Community-based adaptive sports programs affiliated with a rehabilitation hospital network. PARTICIPANTS: Participants were recruited from among 134 adults who registered for the sports program in 2013-2014. Participants with mobility or sensory impairment, absence of cognitive impairment, and English proficiency were included. The 90 former participants with adequate contact information were contacted, and 17 participated in the focus groups. METHODS: Two moderators led each of 3 audio-recorded focus groups using a moderator's guide. We conducted a thematic analysis of transcript data to identify perceived benefits, barriers, and facilitators of participation. RESULTS: Our analysis identified 5 themes: physical well-being and health/safety; interpersonal and social relationships; intrapersonal and beliefs/attitudes; physical environment; and access. Participants experienced participation both as physically beneficial and as transformative in terms of how they view themselves. However, programs drew on limited personal resources and sometimes presented a perceived risk of injury. Finding information about and transportation to programs was a challenge. Participants formed an informal community that modeled what athletes with disabilities are capable of, helping to overcome initial doubts. CONCLUSIONS: To gain the benefits of participation, athletes overcame significant barriers, several of which may be modifiable, including transportation and difficult-to-find information about program offerings. The importance of community and raising awareness of athletes' own abilities suggests a key role for mentorship. Our study contributes to the understanding of experiential aspects that motivate participation in adaptive sports. LEVEL OF EVIDENCE: IV.