Evaluation of GPT-4 ability to identify and generate patient instructions for actionable incidental radiology findings.

OBJECTIVES: To evaluate the proficiency of a HIPAA-compliant version of GPT-4 in identifying actionable, incidental findings from unstructured radiology reports of Emergency Department patients. To assess appropriateness of artificial intelligence (AI)-generated, patient-facing summaries of these findings. MATERIALS AND METHODS: Radiology reports extracted from the electronic health record of a large academic medical center were manually reviewed to identify non-emergent, incidental findings with high likelihood of requiring follow-up, further sub-stratified as "definitely actionable" (DA) or "possibly actionable-clinical correlation" (PA-CC). Instruction prompts to GPT-4 were developed and iteratively optimized using a validation set of 50 reports. The optimized prompt was then applied to a test set of 430 unseen reports. GPT-4 performance was primarily graded on accuracy identifying either DA or PA-CC findings, then secondarily for DA findings alone. Outputs were reviewed for hallucinations. AI-generated patient-facing summaries were assessed for appropriateness via Likert scale. RESULTS: For the primary outcome (DA or PA-CC), GPT-4 achieved 99.3% recall, 73.6% precision, and 84.5% F-1. For the secondary outcome (DA only), GPT-4 demonstrated 95.2% recall, 77.3% precision, and 85.3% F-1. No findings were "hallucinated" outright. However, 2.8% of cases included generated text about recommendations that were inferred without specific reference. The majority of True Positive AI-generated summaries required no or minor revision. CONCLUSION: GPT-4 demonstrates proficiency in detecting actionable, incidental findings after refined instruction prompting. AI-generated patient instructions were most often appropriate, but rarely included inferred recommendations. While this technology shows promise to augment diagnostics, active clinician oversight via "human-in-the-loop" workflows remains critical for clinical implementation.

Factors Associated with Growth in a Cohort of Children with Complex Biventricular Congenital Heart Disease.

OBJECTIVES: To evaluate patterns and determinants of longitudinal growth among children requiring complex biventricular repair for congenital heart disease, as well as to assess for associations of growth with early feeding modality, comorbidities, postoperative complications, and socioeconomic characteristics. STUDY DESIGN: A single-institution retrospective cohort study was performed in children born February 1999 to March 2009 with complex congenital heart disease who underwent biventricular repair before age 4 years, defined by Risk Adjustment in Congenital Heart Surgery-1 category 3-5. Clinical characteristics, height, weight, and body mass index (BMI) from ages 2-12 years were collected by chart review. Neighborhood-level socioeconomic data were identified using a geographic information system approach. The adjusted association of covariates with growth outcomes was estimated using multivariable linear regression models using generalized estimating equations. RESULTS: Compared with population growth curves, the cohort (n = 150) trended toward early decrease in age-adjusted weight and height. Early tube feeding was significantly associated with decreased BMI before adolescence (-0.539; 95% CI -1.02, -0.054; P = .029). In addition, other clinical and perioperative characteristics had significant associations with growth, including low birth weight, preoperative tube feeds, need for multiple bypass runs, and diagnosis of feeding disorder. CONCLUSIONS: Early childhood growth in children with complex biventricular repair may be impaired. Early tube feeding was associated with decreased BMI over the course of early childhood, which may indicate a need for continued close nutrition follow-up and support even beyond the duration of tube feeds.

Emergency medicine resident perspectives on well-being during COVID-19: a qualitative study.

BACKGROUND: Emergency medicine (EM) clinician well-being has been negatively impacted throughout the COVID-19 pandemic. Resident physicians are particularly vulnerable yet less is known about their perspectives. METHODS: The objective of this study was to use qualitative methods to understand EM residents' perspectives on well-being during COVID-19. EM residents at an urban, academic institution in the USA were recruited via email and participated in virtual, semi-structured interviews between November 2020 and February 2021. Interviews were conducted by a trained qualitative researcher, recorded, transcribed and de-identified by a third party vendor. All transcripts were double coded by two trained study team members using thematic analysis to identify the themes and interviews were stopped when no new themes emerged. RESULTS: Seventeen semi-structured interviews were conducted until thematic saturation was reached with residents in their first 4 years of training: 6 postgraduate year (PGY)-1 (35%), 6 PGY-2 (35%), 2 PGY-3 (12%) and 3 PGY-4 (18%). Five themes were identified: (1) isolation from peers in training contrasting with a collective call to action, (2) desire for increased acknowledgement and structured leadership support, (3) concerns about personal needs and safety within the clinical environment, (4) fear of missed educational opportunities and lack of professional development and (5) need for enhanced mental and physical health resources. CONCLUSIONS: This qualitative study elucidated factors inside and outside of the clinical environment which impacted EM resident well-being. The findings suggest that programme and health system leadership can focus on supporting peer-to-peer and faculty connections, structured guidance and mentorship on resident career development and develop programmes which bolster resident on-shift support and acknowledgement. These lessons can be used by training programmes to better support residents, but the generalisability is limited due to the single-centre design and participation.

Defining well-being: A case-study among emergency medicine residents at an academic center: A qualitative study.

BACKGROUND: Healthcare-associated burnout is linked to reduced quality of care, decreased patient experience, and higher cost. The National Academy of Medicine has emphasized the importance of supporting clinician well-being across healthcare; however, well-being is poorly defined, especially early in emergency medicine training. OBJECTIVES: The primary objective of this study was to explore and understand the attitudes, beliefs, and perspectives of emergency medicine (EM) resident physicians surrounding well-being. A secondary objective was to identify priority areas of focus to promote a culture of well-being for EM trainees. APPROACH: We conducted semi-structured focus groups of EM resident physicians at an urban, academic institution with a 4-year training curriculum. Focus group interviews were transcribed and constructivist aggregated themes were identified using content analysis with a constant comparative coding approach. RESULTS: Seventeen EM residents participated in semi-structured qualitative focus groups (PGY1 = 6, PGY2 = 6, PGY3 = 2, PGY4 = 3). Six key themes related to well-being emerged spanning clinical and nonclinical areas: (1) a focus on basic needs being met, (2) on-shift operational structure, (3) individual feedback, (4) feeling valued for clinical contributions, (5) a sense of community within the clinical environment, and (6) a sense of personal ownership over time. CONCLUSIONS: Shifting the focus for medical trainees away from mitigating burnout and toward proactively promoting well-being is important. Understanding the perspectives and key themes in how EM residents define well-being can help support trainees early in their careers. Using qualitative methods, this study identified six key themes that can guide trainees, educational leaders, and academic hospital systems as they work toward building a culture of well-being early in graduate medical education.