Search

Abstract Existing natural language processing (NLP) methods to convert free-text clinical notes into structured data often require problem-specific annotations and model training. This study aims to evaluate ChatGPT’s capacity to extract information from free-text medical notes efficiently and comprehensively. We developed a large language model (LLM)-based workflow, utilizing systems engineering methodology and spiral “prompt engineering” process, leveraging OpenAI’s API for batch querying ChatGPT. We evaluated the effectiveness of this method using a dataset of more than 1000 lung cancer pathology reports and a dataset of 191 pediatric osteosarcoma pathology reports, comparing the ChatGPT-3.5 (gpt-3.5-turbo-16k) outputs with expert-curated structured data. ChatGPT-3.5 demonstrated the ability to extract pathological classifications with an overall accuracy of 89%, in lung cancer dataset, outperforming the performance of two traditional NLP methods. The performance is influenced by the design of the instructive prompt. Our case analysis shows that most misclassifications were due to the lack of highly specialized pathology terminology, and erroneous interpretation of TNM staging rules. Reproducibility shows the relatively stable performance of ChatGPT-3.5 over time. In pediatric osteosarcoma dataset, ChatGPT-3.5 accurately classified both grades and margin status with accuracy of 98.6% and 100% respectively. Our study shows the feasibility of using ChatGPT to process large volumes of clinical notes for structured information extraction without requiring extensive task-specific human annotation and model training. The results underscore the potential role of LLMs in transforming unstructured healthcare data into structured formats, thereby supporting research and aiding clinical decision-making.

Objective: Aspirin has been used for primary prevention of atherosclerotic cardiovascular disease (ASCVD) for decades, but this indication has become controversial with recent trial data. The 2022 US Preventive Services Task Force (USPSTF) provided a recommendation to consider aspirin use for primary prevention in adults 40–59 years with a 10-year ASCVD risk ≥10 % and not at increased risk of bleeding, yet population estimates for the impact of this recommendation are unknown. The objective of this study is to determine the prevalence and demographics of the US population who meet eligibility criteria for aspirin under the new 2022 USPSTF guidelines. Methods: This is a serial cross-sectional study using data from the 2011-March 2020 National Health and Nutrition Examination Survey (NHANES) database. Individuals aged 40–59 years without a self-reported history of ASCVD were included. 10-year estimated ASCVD risk ≥10 % as calculated by the Pooled Cohort Equations (PCE) and increased bleeding risk determined using variables adapted from USPSTF guidelines were further applied as inclusion and exclusion criteria, respectively. The weighted frequencies of US adults aged 40–59 years qualifying for primary prevention aspirin, subgrouped by gender, age, and race/ethnicity, were calculated. Results: Among 72,840,734 US individuals aged 40–59 years, 7.2 million (10 %) are eligible for consideration of primary prevention aspirin by PCE criteria. Of these, approximately 30 % would be potentially excluded based on increased bleeding risks, resulting in a net eligible cohort of 5 million. This represents 7 % of US adults aged 40–59 years and only 2.6 % of adults ≥18 years. Men, age 50–59 years, and Black race have higher proportions meeting aspirin use eligibility. Conclusions: The overall prevalence of US individuals who qualify for aspirin for primary prevention under the 2022 USPSTF guidelines is modest, with larger proportional eligibility among men, older age, and Black individuals.

Background Dietary supplement use is prevalent in the general US population, but little is known regarding the driving reasons for their use among those with atherosclerotic cardiovascular disease (ASCVD). Methods and Results Data from the National Health and Nutrition Examination Survey (NHANES) from 2017 to March 2020 were used to identify adults with ASCVD. Supplement use was assessed by interviewers using label review, and surveys captured self‐reported reasons for use. Demographic, clinical, medication, and laboratory characteristics were compared between supplement users and nonusers. Among individuals with ASCVD in the National Health and Nutrition Examination Survey (n=965; mean age, 65 years; 56.1% men; 73.7% White individuals), 73.1% reported taking ≥1 dietary supplements, most commonly multivitamins (35.4%), vitamin D (30.8%), and fish oil (19.8%). Of those taking supplements, 47.3% report taking them under the advisement of a health professional. Nearly one fifth (17.9%) reported taking at least 1 supplement for “heart health,” most commonly fish oil (11.1%), followed by CoQ10 (4.2%) and resveratrol (1.5%). Supplement users were older (68 versus 62 years; P=0.003), included more women (45.8% versus 37.7%; P=0.17), were less likely to smoke (11.0% versus 36.4%; P