跳到正文
原文
Medical Xpress·· 6 天前AI 评分45

药物安全信息需保持可访问性以应对AI搜索

Drug safety information must remain accessible as patients move from websites to AI

AI 导读

研究发现,338个美国处方药官方网站的移动性能评分中位数仅为56分,20.4%的页面存在视觉稳定性问题,影响患者获取重要药物信息。随着患者越来越多地使用AI系统获取药物信息,药物安全信息的呈现方式变得尤为重要。AI搜索可能会整合来自不同来源的信息,而这些信息可能无法保持原始网站上的详细程度、背景或风险与益处的平衡。

正文
Drug safety information must remain accessible as patients move from websites to AI
AI-powered interfaces are changing how patients access prescription drug information. Credit: AI-generated illustration

A prescription drug warning can be accurate and still be difficult for a patient to use. What matters is not only the information itself but also how and where people receive it. This becomes especially important as patients increasingly use AI systems to find answers about medications, sometimes without visiting the original source.

Working with patient-facing pharmaceutical technologies, I have seen how different priorities shape digital products. Marketing teams focus on communicating benefits, regulatory teams focus on compliance, and technology and design teams must combine these requirements into an interface that works for patients. These goals are not always easy to balance.

The FDA requires prescription drug promotion to present benefits and risks with fair balance. But the presentation of this information creates its own challenges.

A good example is important safety information (ISI), one of the most challenging elements of pharmaceutical websites. ISI is essential for communicating drug risks, but its presentation creates a difficult design problem. Long safety sections or panels that stay visible while scrolling can take up valuable screen space and make navigation harder, especially on mobile devices. The FDA requires appropriate risk disclosure, but it does not prescribe one universal interface for presenting ISI.

Drug safety information must remain accessible as patients move from websites to AI
Automated testing revealed widespread accessibility issues across 338 U.S. prescription drug websites. Credit: Max Shestov

Website design can obscure safety information

In a 2017 study, FDA researchers found that where risk information appears on a drug website affects how well consumers remember it. People recalled fewer risks when the information was placed on a separate page instead of the homepage.

The placement of safety information is therefore more than a visual design decision. It can influence what patients remember about a drug's risks. To investigate the technical side of this problem, we conducted a study of 338 official U.S. prescription drug brand homepages, evaluating performance, accessibility, visual stability and readability. Our work is published in the journal Universal Access in the Information Society.

The findings were concerning. The median mobile performance score was only 56 out of 100. Just 2.7% of homepages had no issues detected by automated checks against basic accessibility standards, and 20.4% of pages with available real-user data showed poor visual stability, meaning that content could unexpectedly shift while loading.

We did not measure patient understanding or FDA compliance. However, the results reveal technical barriers that can make important medication information harder to access and use. Repeatable measurements like these could also help public health agencies identify digital barriers that deserve further investigation.

Drug information moves into AI search

But the way patients access information is changing faster than the traditional website model.

In October 2026, Adweek reported that Google had begun testing prescription drug advertisements in AI Mode and AI Overviews in the United States. A Google spokesperson confirmed that these are standard text or shopping ads displayed in labeled sponsored sections.

The advertisements are separate from AI-generated answers, but both can now become part of the same information experience.

On their own websites, pharmaceutical companies control the content and its presentation. In AI-powered search, patients may receive a summary combining information from different sources, presented through an interface the manufacturer does not control.

The larger issue is not whether AI will replace pharmaceutical websites. From an HCI perspective, we need to look at the broader information system involving patients, AI technologies, medical sources and advertising platforms. The question is what happens to drug information as it moves between these systems.

A company may carefully review its safety information, but that does not guarantee that every AI-generated explanation will preserve the same details, context or balance between benefits and risks.

Evaluating safety across interfaces

I believe we need to expand how we evaluate pharmaceutical digital communication. Testing whether information is accurate at its original source is important, but we should also examine how it is presented, transformed and understood across different interfaces.

This is where I see the next direction for research. We need to study whether AI systems preserve essential safety information, distinguish medical evidence from commercial messages and help patients make informed decisions.

Our research shows that even websites directly controlled by pharmaceutical companies have measurable accessibility and performance problems. As AI adds new layers between the original information and the patient, evaluating the complete information experience becomes even more important.

Ensuring that drug information is accurate at its source is essential. Ensuring that patients can access and understand it wherever they encounter it deserves equal attention.

This story is part of Science X Dialog, where researchers can report findings from their published research articles. Visit this page for information about Science X Dialog and how to participate.

More information

Max Shestov et al., Regulated medication information, accessible web delivery? A cross-sectional evaluation of United States prescription drug brand homepages, Universal Access in the Information Society (2026). DOI: 10.1007/s10209-026-01394-9

Key medical concepts

Prescription Drugs

Who's behind this story?

Lisa Lock

Lisa Lock

BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021. Full profile →

Robert Egan

Robert Egan

Bachelor's in mathematical biology, Master's in creative writing. Well-traveled with unique perspectives on science and language. Full profile →

Max Shestov is an MS Computer Science student at Georgia Tech and a technology lead working at the intersection of health care, technology, and AI, with a focus on human-computer interaction and patient-facing digital technologies.

Citation: Drug safety information must remain accessible as patients move from websites to AI (2026, November 13) retrieved 13 November 2026 from https://medicalxpress.com/news/2026-11-drug-safety-accessible-patients-websites.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

来源:Medical Xpress · medicalxpress.com