Ayo-Ajibola Oluwatobiloba, Julien Catherine, Lin Matthew E, Riddell Jeffrey, Duan Naihua, Kravitz Richard L
Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA.
Department of Head and Neck Surgery, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, CA, USA.
J Gen Intern Med. 2025 Feb 10. doi: 10.1007/s11606-025-09406-9.
ChatGPT has quickly gained popularity as a source of online health information (OHI). However, it is unclear how having a usual source of primary care (USPC) is related to OHI-seeking.
Explore how having a USPC and other characteristics thought to affect access-to-care influence the use of ChatGPT and other OHI forms.
Cross-sectional national survey.
Adult members of ResearchMatch, a non-profit affiliate of the National Institutes of Health, between June and August 2023.
The survey evaluated demographics, health characteristics, and OHI-seeking behaviors, including ChatGPT usage. OHI sources were categorized as "passive" (Google, Wikipedia, WebMD) and "interactive" (forums, Q&A sites, ChatGPT). Descriptive statistics, t-tests, and chi-square tests compared users by USPC status. Multiple logistic regression estimated adjusted effects on ChatGPT use.
Of 21,499 adults invited to participate in the survey, 2406 (11.2%) responded. Among respondents, 56% reported having a USPC. Those with a USPC, compared to those without, were older, spoke English as their primary language, had higher income, and had more formal education (all p<.001). Participants with a USPC were more likely to use passive OHI (OR 2.46, 95% CI 1.55-3.90, p<.001) and less likely to use interactive OHI (OR 0.73, 95% CI 0.60-0.89, p=.002) or ChatGPT (OR 0.56, 95% CI 0.44-0.71, p<.001). Age over 50 (OR 0.11, 95% CI 0.06-0.20, p<.001), non-White race (OR 0.51, 95% CI 0.38-0.70, p<.001), very good or better health (OR 0.71, 95% CI 0.55-0.92, p=.009), and college education (OR 0.61, 95% CI 0.39-0.97, p=.035) were inversely related to ChatGPT use.
In this national survey of patients participating in a clinical research matching service, those with regular primary care access relied less on ChatGPT, suggesting that a personal primary care relationship may attenuate the need or motivation to use AI-derived OHI.
ChatGPT作为在线健康信息(OHI)来源迅速受到欢迎。然而,尚不清楚拥有常规初级保健来源(USPC)与寻求OHI之间的关系。
探讨拥有USPC以及其他被认为会影响医疗服务可及性的特征如何影响ChatGPT及其他OHI形式的使用。
全国性横断面调查。
2023年6月至8月期间,美国国立卫生研究院的非营利附属机构ResearchMatch的成年成员。
该调查评估了人口统计学特征、健康特征以及寻求OHI的行为,包括ChatGPT的使用情况。OHI来源分为“被动型”(谷歌、维基百科、WebMD)和“互动型”(论坛、问答网站、ChatGPT)。描述性统计、t检验和卡方检验按USPC状态对用户进行比较。多元逻辑回归估计对ChatGPT使用的调整效应。
在受邀参与调查的21499名成年人中,2406人(11.2%)做出了回应。在受访者中,56%的人报告有USPC。与没有USPC的人相比,有USPC的人年龄更大,以英语为主要语言,收入更高,接受的正规教育更多(所有p<0.001)。有USPC的参与者更有可能使用被动型OHI(比值比[OR]2.46,95%置信区间[CI]1.55 - 3.90,p<0.001),而使用互动型OHI(OR 0.73,95% CI 0.60 - 0.89,p = 0.002)或ChatGPT(OR 0.56,95% CI 0.44 - 0.71,p<0.001)的可能性较小。50岁以上(OR 0.11,95% CI 0.06 - 0.20,p<0.001)、非白人种族(OR 0.51,95% CI 0.38 - 0.70,p<0.001)、健康状况非常好或更好(OR 0.71,95% CI 0.55 - 0.92,p = 0.009)以及受过大学教育(OR 0.61,95% CI 0.39 - 0.97,p = 0.035)与ChatGPT的使用呈负相关。
在这项对参与临床研究匹配服务的患者进行的全国性调查中,有定期初级保健服务的人对ChatGPT的依赖程度较低,这表明个人与初级保健的关系可能会减弱使用人工智能衍生的OHI的需求或动机。