2026 Edelman Trust Barometer Special Analysis: Health Decisions in the Age of Information Overload
Special analysis of the 2026 Edelman Trust Barometer health data set produced with the Yale School of Public Health, surveying 12,998 adults in 13 countries between 28 February and 11 March 2026 on how people weigh advice from clinicians against artificial intelligence tools, social-media creators, family, friends and people with similar conditions. Nearly half of respondents believe that someone without medical training who is fluent with AI can perform at least one health task (deciding who needs care, basic procedures, recommending treatment or medicines, diagnosing) as well as or better than a doctor, rising to 59% among 18 to 34-year-olds; diagnosis is the task people are least willing to hand over (16%). When a doctor and another source disagree, only 51% always follow the doctor. Education and access to care did not correlate with holding divisive health beliefs, and naming AI platforms as a health influence rises from 13% among people holding no divisive beliefs to 46% among those holding three or more.
Publisher
Edelman Trust Institute, with the Yale School of Public Health
Published
15 Sept 2026
Added
today
DOI
—
Key Findings
- 49% globally (59% of 18 to 34-year-olds) believe an AI-fluent person without medical training can do at least one health task as well as or better than a doctor; by task: deciding whether someone needs to see a provider 26%, basic procedures 19%, choosing treatment or medication 19%, diagnosis 16% (task breakdown from press coverage of the same release)
- When recommendations from a doctor and another source conflict, 51% always follow the doctor, 34% say it depends, 15% always follow the other source
- Health-care providers still lead uncredentialed sources on diagnosis (50% vs 23%) and long-term treatment (34% vs 18%); uncredentialed sources lead on diet and nutrition and on wellness and longevity (35% vs 26%)
- AI platforms are named as an influence on health decisions by 13% of people who hold no divisive health beliefs and 46% of those who hold three or more; friends and family rise from 51% to 71%, people with similar conditions from 27% to 59%
- Nearly one in three say an uncredentialed person (creator, spiritual leader) has a major influence on their health decisions; formal training from a trusted institution is the top marker of a legitimate expert (71%)
Methodology Notes
Online survey, n=12,998 adults in 13 countries, fieldwork 28 February to 11 March 2026 (per Yale News); analysis of previously unpublished items from the April 2026 Special Report: Trust and Health data set (16 countries, 16,000 respondents). The AI-fluency item was asked of half the sample (press). Country list per press coverage: Australia, Brazil, Canada, France, Germany, India, Indonesia, Japan, Mexico, South Africa, South Korea, UK, US; Yale News confirms Australia, Brazil, India, South Africa and the US. Yale News states publication on 15 September 2026; Yale says further analyses will be submitted for peer review. Belief measures, not performance measures. VERIFICATION ROUTE: the edelman.com report PDF returns 403 to every client tried; findings taken from the co-publisher's article (news.yale.edu, 17 September 2026, quoting Dean Megan Ranney), a Yale SPH co-author's write-up (Matthew Facciani, 17 September 2026) and trade coverage (O'Dwyer's PR News 16 September; World of DTC Marketing).
Sources
Edelman report PDF (403 to crawlers; see verification)(opens in a new tab) (primary)
Yale News (co-publisher) on the report(opens in a new tab) (17 Sept 2026)
Write-up by a Yale SPH collaborator on the analysis(opens in a new tab) (17 Sept 2026)
Edelman landing page (403 to crawlers)(opens in a new tab) (15 Sept 2026)
O'Dwyer's PR News coverage(opens in a new tab) (16 Sept 2026)
World of DTC Marketing key points(opens in a new tab) (17 Sept 2026)
Archived snapshot (Wayback Machine)(opens in a new tab) — preserved against link rot
Topics
Tags
Cite This
APA
Edelman Trust Institute, with the Yale School of Public Health. (2026). 2026 Edelman Trust Barometer Special Analysis: Health Decisions in the Age of Information Overload. https://www.edelman.com/sites/g/files/aatuss191/files/2026-09/2026%20Edelman%20Trust%20Barometer%20Special%20Analysis%20with%20the%20Yale%20School%20of%20Public%20Health_FINAL.pdf
Related Insights
Large Language Models in the UK: Public Use, Trust, and Attitudes
Oxford Internet Institute, University of Oxford · 9 Jul 2026
Trust in Generative AI for Health Information Consumption and the Effect of Learned Dependency: Randomized Controlled Experimental Study
Journal of Medical Internet Research (JMIR Publications); University of Arizona · 16 Sept 2026