A Centers for Disease Control and Prevention ad promoting COVID-19 vaccines appeared on a website claiming those same vaccines caused widespread harm. The American Heart Association's message about cardiovascular health ran alongside articles questioning established medical science. Between 2021 and 2024, government health agencies and medical organizations spent nearly $37 million advertising on websites that actively spread health misinformation—the very thing these groups exist to combat.

The Accidental Alliance

The numbers tell a strange story. Yale researchers examined 11 news websites flagged for spreading health misinformation and found that about 10% of their total advertising revenue—roughly $37 million out of $336 million—came from the organizations supposedly fighting false health claims. The Department of Health and Human Services alone spent more than $623,000 on these placements. The Alzheimer's Association contributed over $89,000. Pharmaceutical giants Pfizer and Eli Lilly bought space on the same sites.

This wasn't a coordinated campaign. Nobody at the CDC sat down and decided to fund a platform spreading vaccine conspiracy theories. Instead, it happened through programmatic advertising—automated systems that place ads across thousands of websites based on algorithms, audience demographics, and bid prices. A byzantine process, as NewsGuard and Comscore described it, where brands often have no idea where their ads actually appear.

The organizations involved uniformly called these placements "unintentional" when confronted. They pointed to keyword-exclusion filters designed to prevent exactly this problem. They emphasized the "dynamic nature of digital advertising" that makes complete control nearly impossible. All true. And all beside the point.

The Trust Tax

When you see a CDC ad on a website claiming vaccines cause autism, what registers isn't the ad's content. It's the implicit endorsement. The visual association suggests legitimacy. If a government health agency advertises here, the reasoning goes, maybe this site isn't so fringe after all.

This creates a feedback loop. Advertising revenue helps misinformation sites expand their reach and polish their presentation. Better production values attract larger audiences. Larger audiences generate more ad revenue. The cycle continues, funded partly by the very institutions tasked with promoting accurate health information.

The misinformation industry now earns an estimated $2.6 billion annually from advertising. For every $2.16 sent to legitimate newspapers, US advertisers send $1 to misinformation websites. More than 4,000 top brands have advertised on sites publishing COVID-19 misinformation alone. The money isn't just flowing—it's gushing.

The Credibility Paradox

Some marketing experts argue that reaching vaccine-hesitant audiences requires meeting them where they are. If people skeptical of public health messaging frequent certain websites, shouldn't health organizations advertise there? The logic has surface appeal: you don't convince people by preaching to the choir.

But this misunderstands how credibility works. When the American Heart Association's ad appears next to an article questioning vaccine safety, it doesn't just reach a skeptical audience—it validates the platform hosting that audience. The association lends its reputation to a site actively undermining the scientific consensus it claims to uphold.

Neeraj Patel, the fourth-year Yale medical student who led the research, put it simply: these websites are "directly at odds with the missions of these organizations." The contradiction isn't subtle. It's definitional.

The Automation Excuse

The "we didn't know" defense carries less weight than organizations claim. Programmatic advertising may be complex, but it's not uncontrollable. Exclusion lists exist. Manual placement options exist. More rigorous vetting processes exist. These tools require more effort and money than automated systems, but that's the cost of maintaining institutional integrity.

When HHS spokesman Andrew Nixon blamed the placements on "the Biden administration" and called it "a serious failure," he at least acknowledged the problem. Most organizations offered variations of "we're reviewing our processes" without explaining why those processes failed in the first place or what specific changes would prevent future failures.

The Alzheimer's Association and American Heart Association both emphasized their commitment to evidence-based information. Both promised to strengthen controls. Neither explained why controls sufficient to prevent this obvious conflict didn't already exist.

Rebuilding What Algorithms Broke

Public health operates on trust. When that trust erodes, people die. They skip vaccines. They ignore treatment guidelines. They turn to unproven remedies promoted by the same sites that health organizations inadvertently funded.

The solution isn't complicated, just expensive. Organizations must choose between the efficiency of programmatic advertising and the integrity of their missions. They can't have both. Manual ad placement costs more. Excluding entire categories of websites reduces reach. Rigorous vetting slows campaigns. These are features, not bugs.

The Yale study, published April 1, 2026, in JAMA Network Open, offers a clear measurement of the problem. Now comes the harder part: deciding whether institutional credibility is worth the premium of advertising that doesn't accidentally bankroll your opposition. The answer should be obvious. The fact that it apparently isn't tells you everything about how we got here.