HIV Prevention and Treatment Information from Four Artificial Intelligence Platforms: A Thematic Analysis.

IF 2.4 2区 医学 Q2 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH
Stephen Beegle, Luis A Gomez, Jason T Blackard, Bingfang Yan, Jaime Robertson, Kevin T Fedders, Shaina Horner, Ramsey Miller, Chavez R Rodriguez, Abby Atreya, Jennifer L Brown
{"title":"HIV Prevention and Treatment Information from Four Artificial Intelligence Platforms: A Thematic Analysis.","authors":"Stephen Beegle, Luis A Gomez, Jason T Blackard, Bingfang Yan, Jaime Robertson, Kevin T Fedders, Shaina Horner, Ramsey Miller, Chavez R Rodriguez, Abby Atreya, Jennifer L Brown","doi":"10.1007/s10461-025-04786-9","DOIUrl":null,"url":null,"abstract":"<p><p>Health information is highly accessible with the prominence of artificial intelligence (AI) platforms, such as Chat Generative Pre-Trained Transformer (ChatGPT). Within the context of human immunodeficiency virus (HIV), it is paramount to understand and evaluate the information being provided by AI platforms concerning the safety, side effects, and efficacy of medications to prevent and treat HIV. Prompts (n = 38) requesting information regarding HIV medication use for prevention and treatment were inputted into three AI-based Large Language Models (LLMs; ChatGPT 3.5, ChatGPT 4.0, Google Bard [now Gemini]) and one chatbot (HIV.gov Chatbot) on four consecutive weeks. Outputs (n = 608) were recorded verbatim, weekly by platform. Qualitative analyses using a conventional content analysis coding approach examined key themes in responses; response comprehensiveness was rated via the number of themes represented in a response. Core themes emerged across prompts. A recommendation to speak with a medical professional for further information was the most common theme across platforms. Organ/bone side effects were the most prevalent side effect. Responses pointed to medication efficacy to prevent and treat HIV. ChatGPT 4.0 provided the most comprehensive responses across platforms, while the HIV.gov Chatbot gave the least comprehensive information. Health information on HIV medication safety, side effects, and efficacy is widely available using AI platforms. Results indicate that AI responses typically included recommendations to consult a medical professional to personalize care. The efficacy of medications was never questioned across AI platforms. Future research directions for AI use within the context of HIV prevention and care are provided.</p>","PeriodicalId":7543,"journal":{"name":"AIDS and Behavior","volume":" ","pages":""},"PeriodicalIF":2.4000,"publicationDate":"2025-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"AIDS and Behavior","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s10461-025-04786-9","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0

Abstract

Health information is highly accessible with the prominence of artificial intelligence (AI) platforms, such as Chat Generative Pre-Trained Transformer (ChatGPT). Within the context of human immunodeficiency virus (HIV), it is paramount to understand and evaluate the information being provided by AI platforms concerning the safety, side effects, and efficacy of medications to prevent and treat HIV. Prompts (n = 38) requesting information regarding HIV medication use for prevention and treatment were inputted into three AI-based Large Language Models (LLMs; ChatGPT 3.5, ChatGPT 4.0, Google Bard [now Gemini]) and one chatbot (HIV.gov Chatbot) on four consecutive weeks. Outputs (n = 608) were recorded verbatim, weekly by platform. Qualitative analyses using a conventional content analysis coding approach examined key themes in responses; response comprehensiveness was rated via the number of themes represented in a response. Core themes emerged across prompts. A recommendation to speak with a medical professional for further information was the most common theme across platforms. Organ/bone side effects were the most prevalent side effect. Responses pointed to medication efficacy to prevent and treat HIV. ChatGPT 4.0 provided the most comprehensive responses across platforms, while the HIV.gov Chatbot gave the least comprehensive information. Health information on HIV medication safety, side effects, and efficacy is widely available using AI platforms. Results indicate that AI responses typically included recommendations to consult a medical professional to personalize care. The efficacy of medications was never questioned across AI platforms. Future research directions for AI use within the context of HIV prevention and care are provided.

四个人工智能平台的艾滋病防治信息专题分析
由于人工智能(AI)平台的突出,例如聊天生成预训练转换器(ChatGPT),健康信息可以高度访问。在人类免疫缺陷病毒(HIV)的背景下,了解和评估人工智能平台提供的有关预防和治疗HIV药物的安全性、副作用和有效性的信息至关重要。要求提供艾滋病药物预防和治疗信息的提示(n = 38)被输入到三个基于人工智能的大型语言模型(LLMs)中;ChatGPT 3.5, ChatGPT 4.0,谷歌Bard[现在的Gemini])和一个聊天机器人(HIV.gov chatbot)连续四周。输出(n = 608)逐字记录,每周按平台记录。定性分析使用传统的内容分析编码方法检查关键主题的回应;回答的综合性是通过回答中所代表主题的数量来评定的。核心主题出现在提示中。建议与医疗专业人员交谈以获取进一步信息是各平台最常见的主题。器官/骨骼副作用是最常见的副作用。结果显示药物治疗对预防和治疗HIV的疗效。ChatGPT 4.0提供了最全面的跨平台回答,而HIV.gov聊天机器人提供的信息最不全面。关于艾滋病毒药物安全性、副作用和疗效的健康信息可通过人工智能平台广泛获取。结果表明,人工智能的反应通常包括咨询医疗专业人员以提供个性化护理的建议。在人工智能平台上,药物的功效从未受到质疑。提出了人工智能在艾滋病预防和护理领域的未来研究方向。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
AIDS and Behavior
AIDS and Behavior Multiple-
CiteScore
6.60
自引率
13.60%
发文量
382
期刊介绍: AIDS and Behavior provides an international venue for the scientific exchange of research and scholarly work on the contributing factors, prevention, consequences, social impact, and response to HIV/AIDS. This bimonthly journal publishes original peer-reviewed papers that address all areas of AIDS behavioral research including: individual, contextual, social, economic and geographic factors that facilitate HIV transmission; interventions aimed to reduce HIV transmission risks at all levels and in all contexts; mental health aspects of HIV/AIDS; medical and behavioral consequences of HIV infection - including health-related quality of life, coping, treatment and treatment adherence; and the impact of HIV infection on adults children, families, communities and societies. The journal publishes original research articles, brief research reports, and critical literature reviews. provides an international venue for the scientific exchange of research and scholarly work on the contributing factors, prevention, consequences, social impact, and response to HIV/AIDS. This bimonthly journal publishes original peer-reviewed papers that address all areas of AIDS behavioral research including: individual, contextual, social, economic and geographic factors that facilitate HIV transmission; interventions aimed to reduce HIV transmission risks at all levels and in all contexts; mental health aspects of HIV/AIDS; medical and behavioral consequences of HIV infection - including health-related quality of life, coping, treatment and treatment adherence; and the impact of HIV infection on adults children, families, communities and societies. The journal publishes original research articles, brief research reports, and critical literature reviews.5 Year Impact Factor: 2.965 (2008) Section ''SOCIAL SCIENCES, BIOMEDICAL'': Rank 5 of 29 Section ''PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH'': Rank 9 of 76
×
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:604180095
Book学术官方微信