CDC Vaccines Seminar Spotlights Professor Emily Martin as Another Academic Fake

When you suspect a scientific charlatan, always check their citations.

6 minute read

A scandal has been brewing the last several months at the Centers for Disease Control and Prevention (CDC) where researchers were caught publishing poor quality studies—called test negative design—to promote vaccines. Following weeks of controversy, the CDC hosted a seminar last week to hash out and debate which studies are best to test vaccine effectiveness.

I’m not running a biostatistics seminar here, but if you’re interested in learning more about all the problems and biases with test negative design, see this interview I did with Oxford’s Carl Heneghan or take a look at this 2-page memo now circulating inside the Department of Health and Human Services: The Problematic Test-Negative Design.

The CDC seminar pitted Martin Kulldorff, a long-term vaccine advisor to the CDC and now the chief science advisor to the administration, against two scientists chosen by the CDC. The first to speak was Emily Martin at the University of Michigan, followed by Natalie Dean at Emory University.

However, I discovered that Emily Martin has a history of using poor quality research to draw scientific conclusions. To promote masks in a 2022 essay, Martin cited a study by Cochrane researcher Tom Jefferson. But Martin cited an early 2007 version of Jefferson’s Cochrane work, ignoring three updates that Jefferson published with Cochrane that used higher quality studies.

When I called up Jefferson and told him what Martin did, he started laughing and said that Martin doesn’t know what she’s doing. “She purposely cited the early version that she agreed with,” Jefferson said. “It’s citation bias. Nobody has ever been prosecuted for it, but they should be.”

Martin did not respond to repeated requests for comment.

CDC research favors Big Pharma

The CDC scandal kicked off back in April, when the acting CDC director discovered that CDC scientists are using test negative studies to promote vaccines as effective. The scandal erupted in the pages of the Washington Post when CDC employees leaked these internal deliberations to try and win this scientific debate in the media and allow themselves to continue using bad research.

I interviewed Oxford’s Carl Heneghan last week for The DisInformation Chronicle Podcast, to discuss the scandal and why federal scientists should use better research techniques to give the public real advice about vaccines.

Heneghan is a physician-researcher, who has spent decades in the field of evidence-based medicine, which seeks out the highest quality peer-reviewed research to guide patient care.

“Your CDC has a particular problem—ideology,” Heneghan told me on The DisInformation Chronicle podcast. “They’re not testing whether something works; they’re trying to prove what they think works. In doing so, test negative designs are really good to do that.”

Heneghan said that the CDC needs to get away from observational research, like test negative studies, and start running randomized controlled trials (RCTs) which are more objective and less prone to manipulation by a researcher’s own biases. We then discussed why RCTs are better at determining whether something works—like a drug, medical device, or any other procedure or treatment—and why observational studies that the CDC uses are rife with potential bias.

CDC reckoning on low quality studies

I didn’t have time to watch the entire debate, but I did get a chance to see Emily Martin defend the CDC current practices using test negative studies, instead of higher quality RCTs. Intrigued at Martin’s ability to ignore RCTs in favor of low quality observational research, I looked her up and found she had published a piece at The Conversation arguing “masks work” to stop viruses.

“We are infectious disease epidemiologists and researchers, and we have spent our careers focused on understanding how viruses spread and how best to stop them,” Martin wrote in 2022.

Well, I know Martin’s essay is false, because I have interviewed Jefferson who wrote the series of Cochrane Reviews that found little-to-no evidence that masks stop viruses. Cochrane uses a systematic review process, which requires Cochrane authors to gather all the published studies available on a topic under consideration, and then reference the best available evidence, meaning RCTs. However, RCTs can be supplemented with observational studies if not enough RCTs have been completed.

Jefferson explained to me that Cochrane released their first review on masks in 2006, and updated that original review in 2007, 2010, 2011, 2020, and 2023. Each update improved the original review with better, higher quality research, as scientists published more and more RCTs studying masks.

So here’s an interesting question: How was Martin able to conclude that ‘masks work’ when we know Cochrane has reviewed the highest quality mask studies and concluded that they don’t?

Answer: Martin cited an early version that Cochrane published in 2007 and ignored the updates published in 2010, 2011, and 2020.

Here’s how Martin did this.

In her 2022 essay for The Conversation, Martin claims:

Long before the COVID-19 pandemic, researchers were studying the effectiveness of masks at reducing transmission of other respiratory viruses. Meta-analyses of viral spread during the original SARS epidemic in 2002-2003 showed that one infection was averted for every six people wearing a mask, and for every three people who were wearing an N95 mask.

Click on the link, and the source Martin cites as evidence that masks stop one infection for every six people wearing a mask takes you to the Cochrane Review on masks completed in 2007 and then published in the BMJ in 2008. As you can see the lead author is Tom Jefferson.

When Cochrane published this 2007 review almost two decades ago, very few RCTs had been done on masks, so the Cochrane authors had to rely on lower quality observational studies—the type of studies that Martin and CDC scientists favor. But as the years progressed, Jefferson and his fellow Cochrane scientists updated that review, referencing more and more RCTs on masks as they became available.

In short, the 2010 update was better than the 2007 Cochrane mask review, and the 2011 update was better than the 2010. And so on. By the time of the 2020 update, the Cochrane authors had enough RCTs that they no longer needed to rely on low quality observational studies.

Each update also found less and less evidence that “masks work” to stop viruses. By 2023, the Cochrane authors concluded that wearing a mask “probably makes little or no difference” in stopping viruses and wearing an N95 respirator “probably makes little or no difference” as well.

When I contacted Jefferson and asked him why Martin wrote a 2022 essay that relied on the 2007 Cochrane report and ignored the latest update from 2020, Jefferson said Martin cherry-picked the early version that referenced low quality studies because it gave her the answer she wanted. He then started laughing. “She chose the version that told her what she wanted to know,” Jefferson said. “The old versions are only there if you want to understand the history.”

The hilarity of Martin citing Jefferson’s research, and Jefferson saying that Martin doesn’t understand his research, reminded me of the Marshall McLuhan scene in Annie Hall.

Annie (Diane Keaton) and Alvy (Woody Allen) are standing in line to see a movie when an academic standing behind them begins bloviating in a loud, obnoxious voice about media philosopher Marshall McLuhan. Alvy starts complaining about the blowhard professor who retorts that he teaches a class on media at Columbia University. “I think my insights into McLuhan have a great deal of validity,” the professor says in his defense.

“I heard what you are saying,” says McLuhan, stepping into the scene. “You know nothing of my work. How you ever got to teach a course in anything is totally amazing.”