New CDC Messaging May Be Eroding Trust in Vaccines, Survey Finds

The CDC’s shift to an uncertainty-based approach when conveying the scientific evidence on vaccines and autism could already be increasing vaccine hesitancy and strengthening agreement with science-denial strategies, according to an online survey of U.S. adults.

In the survey of 2,989 people, those who read the CDC’s controversial new statement on vaccines and autism (“The claim ‘vaccines do not cause autism’ is not an evidence-based claim because studies have not ruled out the possibility that infant vaccines cause autism”) were more likely to agree that vaccines might cause adverse events compared with those who read the agency’s previous consensus-based statement (Welch two-sample t test: t= -7.31, P<0.001).

In addition, respondents who read the new uncertainty-based statement said they were more uncertain that vaccines are safe (t=-6.39, P<0.001) and were less inclined to get recommended vaccines (t=2.30, P=0.022) compared with people reading the previous statement (“Studies have shown that there is no link between receiving vaccines and developing autism spectrum disorder”).

Those who read the uncertainty-based statement were also less likely to trust that the CDC is capable of handling critical situations well and correctly, and they were more likely to support science-denial strategies such as impossible evidentiary standards and cherry-picking scientific findings, according to researchers led by Robert Böhm, PhD, of the University of Vienna.

“Our findings offer a cautionary signal about the public health consequences of shifting official communication policies from a consensus-based to an uncertainty-based framing of the vaccine-autism link,” Böhm and colleagues wrote in Science. “If such shifts in communication become more frequent and widespread, they may gradually erode the formation of science-based attitudes and decision-making.”

Proponents of the uncertainty-focused approach may argue that it increases scientific transparency. But “when uncertainty-focused language is applied to issues widely regarded as scientifically settled yet persistently contested in public discourse, it may act less as constructive transparency than as a salient cue for doubt,” they noted.

Lori Handy, MD, associate director of the Vaccine Education Center in Philadelphia, told MedPage Today, “There are clearly times when it’s important that we express uncertainty in public health communication.” During the COVID-19 pandemic, for example, “we communicated with extreme confidence about things we were not yet certain of, and that was in itself a problem.”

In contrast, there are topics — such as whether the measles-mumps-rubella (MMR) vaccine causes autism — “where we are very sure of the scientific answer within the scientific community,” she said.

“We’re seeing the word transparent kind of get a new definition,” Handy noted. “But you can transparently say, ‘I am certain the MMR vaccine does not cause autism, and that’s the most transparent I can be.'”

The study also had a control group that read neither CDC statement. In this group, survey responses generally fell between those of the uncertainty-based and the consensus-based groups. That demonstrated that the consensus-based statement “tends to mitigate uncertainty and promotes vaccination intentions,” the researchers said, while the uncertainty-based statement “has the opposite effect.”

The researchers conducted their survey from December 2025 to January 2026. Participants were quota-representative of the U.S. population in terms of age, gender, and political orientation. Participants used a 7-point Likert scale to respond to a series of statements after reading either of the CDC statements on vaccines and autism or no statement at all.

Compared with those who read the consensus-based statement, exposure to the uncertainty-based statement was linked with increased endorsement of a composite of science-denial strategies: promotion of fake, nonpertinent experts; reliance on logical fallacies; imposition of impossible evidentiary standards; selective citation of scientific findings; and conspiracy-based reasoning (t=-4.43, P<0.001).

Reading the uncertainty-based vaccine statement did not lead to significantly increased polarization, nor did it significantly alter participants’ expectations about the vaccination intentions of others. The researchers also found no partisan differences in how participants responded to the two statements, even though Republicans in general had higher risk perceptions, greater uncertainty, and lower vaccination intentions than Democrats.

Study limitations included the survey’s self-selection process, which could amplify experimental effects. Participants also may have been exposed to the statements before the survey.