Five years after the start of COVID-19 pandemic and the communications challenges, divisions, and false claims that followed, less than half of the public say they have at least some confidence in the federal government’s health agencies like the Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration (FDA) to carry out many of their core responsibilities, a new KFF Tracking Poll on Health Information and Trust finds.
The low levels of confidence apply to federal agencies’ ability to ensure the safety and effectiveness of prescription drugs (46% express at least some confidence) and vaccines (45%), and to respond to outbreaks of infectious diseases (42%). Even fewer (32%) express at least “some” confidence in the agencies to act independently without interference from outside interests.
When it comes to vaccines specifically, close to six in 10 adults have at least “a fair amount” of trust in either the FDA (57%) or CDC (59%) to provide reliable information. That leaves about four in 10 who say they trust the two institutions “not much” or “not at all” when it comes to vaccines (43% for the FDA, 41% for the CDC).
The overall level of trust in each case is similar to where it stood in September 2023, though the poll reveals significant partisan shifts as the second Trump administration and Health and Human Services Secretary Robert F. Kennedy Jr. have started to change vaccine policies and messaging.
The shares of Democrats who say they trust the FDA and CDC have fallen by nearly 20 percentage points since 2023, while the shares of Republicans who trust each of the agencies have increased by about 10 points. While those shifts have narrowed a partisan divide on trust, Democrats remain more likely than Republicans to say they trust each agency’s vaccine information.
“There are remarkably low levels of trust in the nation’s scientific agencies, shaped by partisan perspectives, and that presents a real danger for the country if and when another pandemic hits,” KFF President and CEO Drew Altman said.
Amid the Trump administration’s ongoing changes to vaccine policy that could affect COVID-19 vaccines relying on mRNA technology, the poll finds that nearly half (45%) of the public say they’ve heard or read about the false claim that the mRNA vaccines such as those used for COVID-19 can change their DNA.
Very few (3%) say the false claim is “definitely true,” while a quarter (24%) say it is “definitely false.” Most are unsure, saying the claim is “probably true” (26%) or “probably false” (45%).
About a third (32%) of adults say vaccines that use mRNA technology are “generally safe” compared to about one in six (16%) who say they are “generally unsafe.” At the same time, about half of the public (52%) say they do not know enough about this technology to say. At least one in five Republicans (23%) and independents (18%) say they think mRNA vaccines are unsafe compared to far fewer Democrats (3%).
Other findings include:
- Six in 10 adults (60%) say that under the Trump administration, federal health agencies are not paying enough attention to science. This includes most Democrats (73%) and independents (63%) but less than half of Republicans (45%).
- Large majorities of the public are at least somewhat confident in the safety of most vaccines, including those for measles, mumps and rubella (83%), the flu (74%), and, among those ages 50 and older, pneumonia (82%) and shingles (79%). A narrower majority (56%) has confidence in the safety of COVID-19 vaccines, reflecting low confidence among Republicans (30%).
The poll is part of KFF’s Health Information and Trust Initiative, which is aimed at tracking health misinformation in the U.S., analyzing its impact on the American people, and mobilizing media to address the problem.
Designed and analyzed by public opinion researchers at KFF, this survey was conducted April 8-15, 2025, online and by telephone among a nationally representative sample of 1,380 U.S. adults in English and in Spanish. The margin of sampling error is plus or minus 3 percentage points for the full sample. For results based on other subgroups, the margin of sampling error may be higher.