Rethinking Diabetes with Gary Taubes
What this video covers
Physician and low-carb advocate Dr. Ken Berry interviews Taubes about his 2023 book Rethinking Diabetes, which Taubes describes as a history of how diabetes came to be treated the way it is — and an argument that the foundations of that treatment were never as solid as the profession believes. The book’s method, as Taubes explains it, is archival: with modern internet repositories, he says, researchers can now see roughly 950 pieces of a thousand-piece jigsaw puzzle that the physicians who set diabetes policy were assembling from only 50 to 100.
The emotional center of the interview is an anecdote Taubes uses to frame the book’s thesis. A journalist named Ross Wollen, diagnosed with type 1 diabetes at 36, listened to his endocrinologist explain the treatment plan — get half your calories from carbohydrates, then take insulin to metabolize them — and replied, in effect: you are telling me carbohydrates are toxic to me and insulin is the antidote, so I should eat the toxin and take the antidote — why don’t I just not eat the toxin? Taubes presents this as the question conventional diabetes therapy has never honestly confronted: why prescribe carbohydrate consumption plus insulin, rather than carbohydrate restriction?
Taubes is candid about why the book is a hard sell. It is written, he says, for precisely the people least likely to read it — clinicians whose careers were built on the assumptions it challenges — and he invokes the difficulty even accomplished professionals have in questioning premises their success rests on. Berry, a sympathetic host, mostly facilitates. Note that Berry’s own channel carries a standing disclaimer that his videos are not medical advice, and nothing in this interview should be taken as a treatment recommendation.
Key points
- Taubes argues that modern diabetes therapy rests on assumptions locked in when physicians could see only a fraction of the relevant evidence.
- He presents the book as a history of the diet-drug-disease relationship in diabetes, concluding with what he says the evidence would have supported had it all been visible.
- The “poison and antidote” anecdote frames his central challenge: why do guidelines have diabetics eat carbohydrates and then medicate the consequences, rather than restrict carbohydrates?
- Taubes acknowledges the book targets clinicians invested in the conventional model, and discusses why challenging career-defining assumptions is so difficult.
- He connects the diabetes story to his longer-running critique of nutrition science’s evidentiary standards.
- Taubes traces the “eat the toxin, take the antidote” logic to 1971, when the American Diabetes Association issued nutrition guidelines telling patients with diabetes to get over 50% of their calories from carbohydrates — a recommendation, he argues, that only made sense once insulin existed to medicate the consequences.
- He adds the second half of the antidote’s story: insulin therapy’s consequences — weight gain, possibly hypertension, possibly premature death — “none of which would happen if you just didn’t eat the toxin.” Children diagnosed at 10 and put on insulin at 12, he says, are now dying in their twenties of atherosclerosis.
- His numbers on the scale of the problem: roughly 2 million Americans had diabetes 60 years ago; today about 30 million are diagnosed and another 9 million undiagnosed.
- Berry’s interview is advocacy-adjacent rather than adversarial; the book’s historical claims and its therapeutic implications remain contested.