Despite questionable benefits for weight loss and a poorly understood propensity for causing weird rashes, the ketogenic diet has remained an article of faith for gym rats, fitness influencers, and ambitious nerds hoping to one day look like they’ve been cast in a Marvel movie.
But Keto’s defenders faced a serious setback this March with the retraction of a controversial peer-reviewed study—walked back to little fanfare earlier this spring, but now the subject of a detailed postmortem by Avery Orrall at the science news nonprofit Retraction Watch.
The retracted study asserted that arterial plaque build-up recorded in the circulatory systems of 100 otherwise healthy participants, all following ketogenic diets, somehow bore no relationship to the elevated cholesterol levels associated with keto’s low-carb, high-fat weight-loss strategy. Those counterintuitive findings, naturally, had been fiercely debated ever since the study was first published by the Journal of the American College of Cardiology: Advances on April 7, 2025.
‘A collective mess’
As UC Berkeley metabolism researcher Kevin Klatt told Wired back then, the ensuing drama was “a collective mess” exacerbated by the social media prowess of the study’s coauthor, entrepreneur Dave Feldman, “a conflicted party with no training in the biomedical sciences.”
Medical professionals who spoke to Retraction Watch were even harsher in their assessment. Michael Mindrum, an internal medicine specialist and an assistant professor with Dalhousie University, described Feldman and his coauthor Nicholas Norwitz as “social media influencers” who would “try to fit whatever data is there into their ‘narrative.’”
‘A number of anomalies’
Norwitz, a Harvard- and Oxford-trained doctor, has since taken credit for the retraction, asserting on Substack that he and his colleagues had actually asked the journal to take down the article themselves—after Feldman performed “a deeper analysis and found a number of anomalies.”
Nevertheless, their critics had been leveling allegations of “selective” data reporting and questionable statistical methods since nearly the beginning, including a letter published by JACC: Advances one month after their study ran. According to the letter’s authors, public health researchers Miguel López-Moreno and José Francisco López-Gil, the study appeared to use questionably serious control groups to test against its population of keto dieters, “often involving very small subgroups (e.g., n = 17).”
Undaunted by the haters, Norwitz, Feldman, and other members of their team have already published a preprint reanalyzing data from their retracted paper. They’re currently promoting their ongoing keto research on X.com.
