Key takeaways
- A small study in the Journal of the Endocrine Society found that people with early type 2 diabetes on a ketogenic diet showed better beta cell function than people on a low-fat diet, measured by how their pancreas made insulin rather than by blood sugar alone.
- That matters because glucose tells you the result, while insulin output tells you how hard the factory is working to get it.
A small study in the Journal of the Endocrine Society found that people with early type 2 diabetes on a ketogenic diet showed better beta cell function than people on a low-fat diet, measured by how their pancreas made insulin rather than by blood sugar alone. That matters because glucose tells you the result, while insulin output tells you how hard the factory is working to get it. It was a small study, so the word may carries real weight, and the useful takeaway is which markers you could ask your doctor to track.
What did the small study on keto beta cell function actually find?
It found that a ketogenic diet lowered a marker of pancreas stress more than a low-fat diet did, and that this was linked to better keto beta cell function. The Endocrine Society’s release says both groups lost a modest amount of weight, yet the keto group cut the proportion of proinsulin it secreted further. The lead researcher at the University of Alabama at Birmingham framed it as a suggestion, not a verdict. That is the honest shape of it.
Nutrition Insight reports the trial included 51 people with type 2 diabetes who followed one diet or the other for 12 weeks. Fifty-one people. Twelve weeks. That is a starting line, not a finish line.
Here is the real problem with the headline. It says people on keto may have a better chance of reversing diabetes. Sit with that for a moment. The study measured a mechanism. It did not follow anyone long enough to show remission, and nobody should read it as a cure.
What do beta cells do?
Beta cells are the cells in your pancreas that make and release insulin, the hormone that moves glucose out of your blood and into your cells. The Endocrine Society describes type 2 diabetes as a disease where those cells cannot secrete enough insulin to control blood sugar.
Let me explain why that framing matters. Most people think type 2 diabetes is only about insulin resistance, the cells ignoring the signal. That is half the story. A review in the medical literature argues that type 2 diabetes never develops without beta cell dysfunction. Resistance pushes the factory to work overtime. The disease shows up when the factory starts to fail.
So the evidence says the pancreas is not a passive bystander. It is the engine room. If the engine room is under strain, your glucose can look acceptable for years while the cost keeps compounding underneath.
Why does insulin output tell you more than a glucose reading?
Because glucose tells you the score, and insulin output tells you how hard the team played to get it. Two people can post the same fasting glucose while one pancreas is coasting and the other is redlining.
The study used the proinsulin to C-peptide ratio. Proinsulin is the raw form of insulin. A stressed beta cell pushes out more of it unfinished. The researchers describe the ratio as a reflection of beta cell stress that has fallen after diet-induced weight loss in earlier work. C-peptide is the useful partner. MedlinePlus explains that providers may use a C-peptide test to gauge insulin levels instead of testing insulin directly.
Here is the test I would put to any diet claim. Does it only move the number you can see, or does it move the machinery behind it? A glucose reading is a snapshot. Insulin output is the workload. If you aren’t testing, you’re guessing.
How much weight should a 51-person study carry?
Enough to raise a good question, not enough to change your plan on its own. Small, short trials are where ideas get tested first, and many of them fade under larger ones.
The bigger body of evidence says the same thing about low-carb eating in diabetes. A meta-analysis of 23 randomised trials found low-carb diets raised remission rates at 6 months, with benefits fading by 12 months and adherence deciding much of the outcome. Another review of 25 trials found HbA1c fell more on carb restriction at 3 and 6 months but not at 12 or 24 months.
The honest answer is that keto can move markers early, and keeping them moved is a different contest. Individual results vary, and a group average cannot tell you which side of it you will land on. That is the entire stakes of tracking your own numbers.
What did the 600-burger experiment teach me about tracking?
It taught me that the body keeps the score whether you read it or not. The film shows Dr. Mike, a fit doctor, eating 20 plain Burger King patties a day for 30 days, no bun, no condiments, water only, with bloodwork before, during and after. You can see how we set it up on the keto vs carnivore page.
The book records what the numbers did along the way. By Day 29 his ARX numbers were record lows for him. None of that showed up as one tidy glucose figure. It showed up across a dozen measures, which is why we quantified literally everything. The lab we used refused to be part of the film, so I paid for the testing myself.
Then the months that followed. After the burgers came a keto-pescatarian reset month, in which Dr. Mike recovered. On a later month of grass-finished beef he declined again, though less. The film suggests one diet label tells you very little; the details of what is on the plate, and who is eating it, decide the result. What his final labs looked like is in the film, and I will not spoil it here.
Which markers can you ask your doctor to track?
Ask about markers that show both your glucose and your insulin output, and ask to repeat them on the same schedule. The point is a trend, not a single draw.
- Fasting glucose. The snapshot. Read our note on fasting glucose in ketosis before you panic about a high morning number.
- HbA1c. MedlinePlus notes it reflects average blood glucose over roughly the past three months.
- Fasting insulin and C-peptide. The workload behind the glucose.
- Proinsulin, if your doctor thinks it is useful. It is the marker this study leaned on, and it is not on a routine panel.
- Triglycerides, waist and blood pressure. The wider picture, which we cover in scoring your five markers.
In my experience, the people who stick with a change are the ones who can see it on paper. Your doctor decides what is appropriate for you, especially if you take diabetes medication, because changing carbs can change what that medication does.
Frequently asked questions
Is the proinsulin to C-peptide ratio a standard blood test?
No, it is not part of a routine panel, and many clinics do not order it. It is mainly a research marker of beta cell stress. A doctor can usually order fasting insulin and C-peptide more easily, and those two give a practical view of how much insulin your pancreas is producing alongside your glucose.
Did the keto group in the study lose more weight?
According to the Endocrine Society’s summary, both groups lost a modest amount of weight on average. The difference the researchers highlighted was in the proinsulin ratio, not the scale. That is why they suggested the beta cell changes might occur without substantial weight loss, though a small trial cannot settle that question.
Can people with type 1 diabetes apply this study?
No. The trial enrolled people with type 2 diabetes, where beta cells still make insulin but struggle to keep up. In type 1 diabetes the immune system destroys most beta cells, so the question of easing their workload is different. Anyone with type 1 should work through diet changes with their own care team.
What foods did Dr. Mike eat during the burger month?
Only plain Burger King patties and water, 20 patties a day for 30 days, with no bun and no condiments. The patty was chosen in a blind taste test against McDonald’s, Wendy’s and a grass-fed patty. The grass-fed patty won overall, and Burger King was the favourite fast-food patty.
Were the burger patties ever tested in a lab?
No. Every lab I approached either refused or quoted a five-figure minimum to analyse the fast-food meat, so the patties themselves were never lab-tested. What we did measure was Dr. Mike: his blood, his heart rate variability, his brain scans and his fitness, at set points before, during and after the month.
The Bottom Line
A small study suggests keto may ease the strain on beta cells in early type 2 diabetes, and it measured that strain by looking at insulin output rather than glucose alone. Take the method home, not the headline. Ask your doctor which of these markers fit your situation, get a baseline, and retest on a schedule. Boring is good. Repeated numbers beat one dramatic reading every time.
If you want to see what tracking an experiment in numbers looks like when a fit doctor puts his own body on the line, watch The Keto Project on Prime Video, or read how it started in 600 Burgers in 30 Days. How Dr. Mike’s numbers ended up is the part only the film shows. What will you do with your own?
Sources
- Keto diet may improve beta cell function in people with type 2 diabetes (Endocrine Society)
- Study finds keto diet improves beta-cell function in type 2 diabetes (Nutrition Insight)
- Importance of Beta Cell Function for the Treatment of Type 2 Diabetes
- C-Peptide Test (MedlinePlus)
- Blood Glucose Test (MedlinePlus)
- Efficacy and safety of low and very low carbohydrate diets for type 2 diabetes remission
- Effect of dietary carbohydrate restriction on glycemic control in adults with diabetes
This article is for information only and is not medical advice. Individual results vary. Speak to a qualified clinician before changing how you eat.
The whole experiment is in the film. 600 burgers, 30 days, every blood test — and the ending we never post.
Watch The Keto Project on Amazon