Trust Didn't Collapse, It Fractured
Global datasets reveal what's actually happening in views of health and science.
Updated May 31, 2026 | Reviewed by Tyler Woods
You may have heard that trust in science is collapsing. The evidence suggests something different: most people still trust scientists, but many no longer trust themselves to navigate a chaotic information environment and know whom to believe.
The Good News: Trust in Scientists is Holding
When we look at trust in scientists, we still see relatively high levels globally. Last year, I was part of a massive global study published in Nature Human Behaviour , a collaboration involving over 200 researchers across 68 countries and nearly 72,000 respondents.
The findings were, in some ways, encouraging. Trust in scientists globally remains moderate to high. No country showed low overall trust. About 78 percent of respondents believe scientists are well qualified, and 75 percent trust scientific methods as the best way to determine truth.
So any sweeping narrative of a global crisis of trust in science is not quite right. Most people, in most places, still believe in the enterprise of science. That is worth stating plainly, but it is not the whole story.
The Bad News: The Sources We Find Credible are Fracturing
The 2026 Edelman Trust Barometer Special Report on Trust and Health paints a more complicated picture. Across 16 countries and 16,000 respondents, confidence in people’s own ability to navigate health information and make informed decisions dropped 10 points in a single year. That decline was significant in 14 of the 16 countries surveyed, and it cut across gender , income, education , and political orientation.
Trust in media to accurately report health information remains 11 points below where it stood before the COVID pandemic. Two in three people globally feel their country is fragmented over health issues. And this isn’t abstract: 70 percent of respondents hold at least one of the six “divisive” health beliefs tested, including vaccine skepticism, fluoride fears, and claims about raw milk.
Importantly, the individuals holding these views are not disengaged or uninformed. They consume more health news, consult AI frequently, and seek out more sources than people who hold none of these beliefs. The problem is not ignorance or lack of information. It is an overwhelming and fragmented information environment that may be quietly eroding people’s confidence in their own judgment.
What does this look like in practice? Around the same time these data were released, Katelyn Jetelina of Your Local Epidemiologist wrote something that made the picture feel especially immediate. Her piece on the hantavirus response focused on what happens when trust fractures and people have nowhere stable to land.
Jetelina argued that pandemic trauma , mistrust , weak leadership , and a fear -amplifying media environment combined to distort public perceptions of risk. While social media reflected spiraling anxiety , only 20 percent of survey respondents were actually concerned.
The Distinction That Matters
Here is what I think these two datasets, taken together, are telling us. There is a meaningful difference between trusting scientists in the abstract and trusting the information ecosystem in practice. Those are two separate things, and they appear to be moving in opposite directions.
People can believe scientists are competent, and most do, while still feeling uncertain about what to eat, whether to vaccinate their children, what AI is telling them, and whom to believe when credible-sounding sources conflict. The 68-country study and the Edelman report are measuring different constructs. One captures something closer to institutional legitimacy. The other captures something more like navigational confidence. And it is the second that is in trouble.
This connects directly to my own research on identity and misinformation. When people feel overwhelmed and uncertain, they do not become neutral information processors. They fall back on social identity as a navigation tool. Health beliefs become markers of who you are and whom you trust, not simply conclusions drawn from evidence. Edelman’s 2026 research also found 70 percent of people globally are hesitant or unwilling to trust someone who differs from them in values, facts, or background.
That is how we arrive at a world where those divisive beliefs cut across education levels and political affiliations. When you cannot sort the evidence, you sort the messenger.
You can watch this happen with the CDC in the United States. Under the last administration, trust in the agency split along the usual partisan lines, with Democrats at around 70 percent and Republicans at around 51% . Now, with Robert F. Kennedy Jr. running the agency that sets vaccine policy, Democratic trust has dropped nearly 10 points, and Republican trust has not risen to replace it. The result is an agency, unchanged in name and function, that fewer than half of Americans now trust. The lesson is not that one side has the CDC right. It is that trust in the institution was never only about the institution. It tracked who was in charge, and whether that person felt like one of us.
Where the Trust Actually Lives
Here is the thread that ties the hopeful findings together. Trust has not disappeared. It has relocated. It has moved away from distant institutions and toward the people we already know.
The Edelman report makes this concrete: the personal doctor remains the single most trusted voice, no matter how many divisive beliefs a person holds. New US data says the same thing. KFF’s latest tracking poll finds doctors and health care providers far more trusted than government agencies, pharmaceutical companies, or food companies. Confidence in the person in the exam room is holding even as confidence in the institution erodes. Trust continues to be found hyper-locally, where relationships have been established.
The same pattern shows up in unexpected places. In the 68-country study, religiosity was positively associated with trust in scientists, and the link was strongest in Muslim-majority countries. That cuts against what many Western observers assume, and it points somewhere useful. Faith leaders are not obstacles to trust in science. In many communities, they are among its strongest carriers. The pastor, the imam, and the family doctor are not always competing with science. They can be channels through which science reaches people who would never read a journal or a press release.
And the population on the other end is more reachable than the doom narrative suggests. Among people holding many divisive health beliefs from the Edelman data, 72 percent say they would accept a health recommendation that contradicts their current views if they heard it from trusted sources two or more times. Only 11 percent say they never would. KFF finds the same thing in what it calls the “malleable middle.” On the myth that mRNA vaccines alter your DNA , only 3 percent are certain it is true. Most people are not locked in. They are waiting to hear something credible, more than once, from someone they trust.
The Path Forward Is Not Simply Showing Up
This is the part that is easy to get half right. Scientists do need to be present in digital spaces. When we are absent, the void does not stay empty. The loudest and most extreme voices fill it, and the algorithm rewards them for it. Presence is the price of entry.
But presence is not trust, and broadcasting more information into a saturated environment can deepen the fragmentation rather than fix it. The trusted messengers work because they are in a relationship with the people they serve. The doctor listens before prescribing. The pastor knows the family. The reciprocity is the critical factor, not the volume.
Which means the work is two-way. Earning trust is not a communications problem to be solved with better content and wider reach. It is built through dialogue, where scientists and institutions listen to communities, take their concerns seriously, partner with the people those communities already trust, and stay accountable when they get things wrong. We have spent years asking how to make people listen to us. The more useful question is how we show people that we are listening to them.
The Bigger Picture and Moving Forward
The question is not only whether people trust science, but whether they trust themselves, and the systems around them, to translate science into decisions that matter in their everyday lives.
I have written before about what rebuilding trust requires , and it is not better fact-checking or a viral campaign. It begins with warmth and humility in individual interactions. It grows through genuine community partnerships with the pastors, teachers, and local figures people already trust. And it requires institutional reform: greater transparency, accountability, and a willingness to admit mistakes. That framework feels more urgent with each new health crisis.
We have spent years asking whether people trust science. That was never quite the right question. Most people do trust scientists. What many have lost is confidence in their own ability to make sense of a chaotic information world and decide whom to believe. You do not restore that confidence by broadcasting more facts into the noise. You restore it the way it was lost: person by person, in the relationships where trust still lives.
A version of this post also appears on Misguided: The Newsletter.
Share this post Facebook Bluesky Linkedin Email
There was a problem adding your email address. Please try again.
By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy
Matthew Facciani, Ph.D., is a Research Scientist at the Yale School of Public Health.
Get the help you need from a therapist near you–a FREE service from Psychology Today.
This article is part of the Bringwise Psychology Journal — daily insights on human behavior, mental health, and personal growth.