应用大型语言模型评估护理质量:监测多动症药物副作用。

IF 6.2 2区 医学 Q1 PEDIATRICS
Yair Bannett, Fatma Gunturkun, Malvika Pillai, Jessica E Herrmann, Ingrid Luo, Lynne C Huffman, Heidi M Feldman
{"title":"应用大型语言模型评估护理质量:监测多动症药物副作用。","authors":"Yair Bannett, Fatma Gunturkun, Malvika Pillai, Jessica E Herrmann, Ingrid Luo, Lynne C Huffman, Heidi M Feldman","doi":"10.1542/peds.2024-067223","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>To assess the accuracy of a large language model (LLM) in measuring clinician adherence to practice guidelines for monitoring side effects after prescribing medications for children with attention-deficit/hyperactivity disorder (ADHD).</p><p><strong>Methods: </strong>Retrospective population-based cohort study of electronic health records. Cohort included children aged 6 to 11 years with ADHD diagnosis and 2 or more ADHD medication encounters (stimulants or nonstimulants prescribed) between 2015 and 2022 in a community-based primary health care network (n = 1201). To identify documentation of side effects inquiry, we trained, tested, and deployed an open-source LLM (LLaMA) on all clinical notes from ADHD-related encounters (ADHD diagnosis or ADHD medication prescription), including in-clinic/telehealth and telephone encounters (n = 15 628 notes). Model performance was assessed using holdout and deployment test sets, compared with manual medical record review.</p><p><strong>Results: </strong>The LLaMA model accurately classified notes that contained side effects inquiry (sensitivity = 87.2, specificity = 86.3, area under curve = 0.93 on holdout test set). Analyses revealed no model bias in relation to patient sex or insurance. Mean age (SD) at first prescription was 8.8 (1.6) years; characteristics were mostly similar across patients with and without documented side effects inquiry. Rates of documented side effects inquiry were lower for telephone encounters than for in-clinic/telehealth encounters (51.9% vs 73.0%, P < .001). Side effects inquiry was documented in 61.4% of encounters after stimulant prescriptions and 48.5% of encounters after nonstimulant prescriptions (P = .041).</p><p><strong>Conclusions: </strong>Deploying an LLM on a variable set of clinical notes, including telephone notes, offered scalable measurement of quality of care and uncovered opportunities to improve psychopharmacological medication management in primary care.</p>","PeriodicalId":20028,"journal":{"name":"Pediatrics","volume":"155 1","pages":""},"PeriodicalIF":6.2000,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Applying Large Language Models to Assess Quality of Care: Monitoring ADHD Medication Side Effects.\",\"authors\":\"Yair Bannett, Fatma Gunturkun, Malvika Pillai, Jessica E Herrmann, Ingrid Luo, Lynne C Huffman, Heidi M Feldman\",\"doi\":\"10.1542/peds.2024-067223\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Objective: </strong>To assess the accuracy of a large language model (LLM) in measuring clinician adherence to practice guidelines for monitoring side effects after prescribing medications for children with attention-deficit/hyperactivity disorder (ADHD).</p><p><strong>Methods: </strong>Retrospective population-based cohort study of electronic health records. Cohort included children aged 6 to 11 years with ADHD diagnosis and 2 or more ADHD medication encounters (stimulants or nonstimulants prescribed) between 2015 and 2022 in a community-based primary health care network (n = 1201). To identify documentation of side effects inquiry, we trained, tested, and deployed an open-source LLM (LLaMA) on all clinical notes from ADHD-related encounters (ADHD diagnosis or ADHD medication prescription), including in-clinic/telehealth and telephone encounters (n = 15 628 notes). Model performance was assessed using holdout and deployment test sets, compared with manual medical record review.</p><p><strong>Results: </strong>The LLaMA model accurately classified notes that contained side effects inquiry (sensitivity = 87.2, specificity = 86.3, area under curve = 0.93 on holdout test set). Analyses revealed no model bias in relation to patient sex or insurance. Mean age (SD) at first prescription was 8.8 (1.6) years; characteristics were mostly similar across patients with and without documented side effects inquiry. Rates of documented side effects inquiry were lower for telephone encounters than for in-clinic/telehealth encounters (51.9% vs 73.0%, P < .001). Side effects inquiry was documented in 61.4% of encounters after stimulant prescriptions and 48.5% of encounters after nonstimulant prescriptions (P = .041).</p><p><strong>Conclusions: </strong>Deploying an LLM on a variable set of clinical notes, including telephone notes, offered scalable measurement of quality of care and uncovered opportunities to improve psychopharmacological medication management in primary care.</p>\",\"PeriodicalId\":20028,\"journal\":{\"name\":\"Pediatrics\",\"volume\":\"155 1\",\"pages\":\"\"},\"PeriodicalIF\":6.2000,\"publicationDate\":\"2025-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Pediatrics\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1542/peds.2024-067223\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"PEDIATRICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Pediatrics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1542/peds.2024-067223","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PEDIATRICS","Score":null,"Total":0}
引用次数: 0

摘要

目的:评估大型语言模型(LLM)在衡量临床医生对监测注意缺陷/多动障碍(ADHD)儿童用药后副作用的实践指南的依从性方面的准确性。方法:对电子健康记录进行回顾性人群队列研究。队列包括2015年至2022年间在社区初级卫生保健网络中诊断为ADHD的6至11岁儿童和2次或更多的ADHD药物治疗(兴奋剂或非兴奋剂处方)(n = 1201)。为了确定副作用查询的文档,我们对所有与ADHD相关的临床记录(ADHD诊断或ADHD药物处方)进行了培训、测试和部署了开源LLaMA (LLaMA),包括门诊/远程医疗和电话记录(n = 15628份记录)。与手动医疗记录审查相比,使用保留和部署测试集评估模型性能。结果:LLaMA模型对含有副作用查询的笔记进行了准确分类(在保留集上灵敏度= 87.2,特异度= 86.3,曲线下面积= 0.93)。分析显示,在患者性别或保险方面没有模型偏差。首次处方时平均年龄(SD)为8.8(1.6)岁;有无副作用记录的患者的特征基本相似。电话就诊的副作用问询率低于诊所/远程医疗就诊的副作用问询率(51.9% vs 73.0%)。结论:在一组可变的临床记录(包括电话记录)中部署LLM,提供了可扩展的护理质量测量,并发现了改善初级保健精神药理学药物管理的机会。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Applying Large Language Models to Assess Quality of Care: Monitoring ADHD Medication Side Effects.

Objective: To assess the accuracy of a large language model (LLM) in measuring clinician adherence to practice guidelines for monitoring side effects after prescribing medications for children with attention-deficit/hyperactivity disorder (ADHD).

Methods: Retrospective population-based cohort study of electronic health records. Cohort included children aged 6 to 11 years with ADHD diagnosis and 2 or more ADHD medication encounters (stimulants or nonstimulants prescribed) between 2015 and 2022 in a community-based primary health care network (n = 1201). To identify documentation of side effects inquiry, we trained, tested, and deployed an open-source LLM (LLaMA) on all clinical notes from ADHD-related encounters (ADHD diagnosis or ADHD medication prescription), including in-clinic/telehealth and telephone encounters (n = 15 628 notes). Model performance was assessed using holdout and deployment test sets, compared with manual medical record review.

Results: The LLaMA model accurately classified notes that contained side effects inquiry (sensitivity = 87.2, specificity = 86.3, area under curve = 0.93 on holdout test set). Analyses revealed no model bias in relation to patient sex or insurance. Mean age (SD) at first prescription was 8.8 (1.6) years; characteristics were mostly similar across patients with and without documented side effects inquiry. Rates of documented side effects inquiry were lower for telephone encounters than for in-clinic/telehealth encounters (51.9% vs 73.0%, P < .001). Side effects inquiry was documented in 61.4% of encounters after stimulant prescriptions and 48.5% of encounters after nonstimulant prescriptions (P = .041).

Conclusions: Deploying an LLM on a variable set of clinical notes, including telephone notes, offered scalable measurement of quality of care and uncovered opportunities to improve psychopharmacological medication management in primary care.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Pediatrics
Pediatrics 医学-小儿科
CiteScore
12.80
自引率
5.00%
发文量
791
审稿时长
2-3 weeks
期刊介绍: The Pediatrics® journal is the official flagship journal of the American Academy of Pediatrics (AAP). It is widely cited in the field of pediatric medicine and is recognized as the leading journal in the field. The journal publishes original research and evidence-based articles, which provide authoritative information to help readers stay up-to-date with the latest developments in pediatric medicine. The content is peer-reviewed and undergoes rigorous evaluation to ensure its quality and reliability. Pediatrics also serves as a valuable resource for conducting new research studies and supporting education and training activities in the field of pediatrics. It aims to enhance the quality of pediatric outpatient and inpatient care by disseminating valuable knowledge and insights. As of 2023, Pediatrics has an impressive Journal Impact Factor (IF) Score of 8.0. The IF is a measure of a journal's influence and importance in the scientific community, with higher scores indicating a greater impact. This score reflects the significance and reach of the research published in Pediatrics, further establishing its prominence in the field of pediatric medicine.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信