لأغراض تعليمية فقط، وليست نصيحة طبية. استشر طبيبك قبل تغيير نظامك الغذائي أو أي دواء.
Diabetes on a plant based diet: what the doctors say
Diabetes on a plant based diet: what McDougall, Barnard, Ornish, Kahleova and other doctors say and found in their studies, with sources.
Updated . For education only, not medical advice.
Our view
The low-fat plant based doctors on this site see type 2 diabetes mainly as a disease of the rich Western diet and the weight it brings, not of bad genes or of eating potatoes and rice123. Neal Barnard, Peter Rogers and Alan Goldhamer explain it through fat stored inside muscle and liver cells, which blunts the signal of insulin, so that blood sugar stays high even when the body makes plenty of insulin456.
The strongest evidence they offer comes from randomized trials with a comparison group: Barnard’s NIH-funded trial in type 2 diabetes, in which A1c fell on both diets but about three times as much on the low-fat vegan one, and Hana Kahleova’s 2024 trial of 58 people with type 1 diabetes, where daily insulin needs fell on the vegan diet and not on a portion control diet789. Nelson, summing up Barnard’s trial in 2026, gave its length as 22 weeks with a check at 74 weeks, when both groups had improved10. Much of the rest, from Walter Kempner’s 100 rice diet patients to the clinic reports of McDougall, Pritikin, Ornish and Rip Esselstyn, comes without a comparison group and shows what happened, not what would have happened otherwise1112131415.
We give most weight to the low-fat clinicians’ explanation, while noting that their own trial diets also favored low glycemic foods such as beans, fruit and green vegetables716. Several of them call type 2 diabetes reversible or even curable, but each also says that medicines change quickly on such a diet and must be adjusted by the patient’s own doctor121718.
Where they agree
They agree that diabetes follows the Western way of eating: Denis Burkitt saw it rise in Pacific Islanders within two generations of moving to New Zealand, and both Burkitt and Jeff Nelson pointed to the Pima of Arizona, whose rates soared after their diet changed19320. Caldwell Esselstyn counts diabetes among the slow injuries that the usual American diet piles up over decades, and David Jenkins tied traditional plant foods to a time when the disease was almost unknown2122.
They agree that starch is not the cause: McDougall moves patients onto rice, corn and potatoes, Novick blames weight, inactivity and a diet heavy in refined foods, oil and cheese rather than the potato, and Barnard says sugar is not the main culprit in type 2 diabetes23224. Meat comes up on the other side of the ledger, with Barnard linking both processed and unprocessed meat to diabetes risk and Campbell linking higher animal protein intake to higher blood sugar and A1c2526.
They also agree that blood sugar can shift within days of a diet change, which is why McDougall, Rogers and Ornish all tell people on diabetes medicines to involve their doctor before they start121718. Goldhamer notes that blood sugar begins to improve as soon as the food changes, long before the extra weight is gone6.
Where they differ
The doctors part ways on what in the diet does the most harm, on type 1 diabetes, on fasting, on how much they promise, and their evidence ranges from randomized trials to observations without a comparison group242728298.
Fat, sugar or glycemic impact
Barnard and Rogers put fat first: Barnard says removing it lets the fat inside cells shrink and insulin work again, and Rogers, drawing on Gerald Shulman’s research at Yale, describes fat lodging in muscle, then liver, then the pancreas30531. Goldhamer cited a study in which healthy medical students, fed a very fatty diet for just two days, tested in the diabetic range for glucose32. Jenkins instead stressed foods that raise blood glucose less, and Ornish’s 2014 column asked for fewer sugars and refined carbohydrates rather than more fat or animal protein3314. Doug Lisle said in 2026 that food addiction programs which single out sugar and flour while allowing oil and animal foods are simply wrong34.
Plant based but still high in fat
In 2026 Nelson responded to a cardiologist whose insulin resistance appeared after she had eaten plant based for years by saying the trials that helped people with diabetes used very low-fat diets3536. When he analysed her cookbook, the typical recipe drew about 12% of its calories from fat, and many went well beyond that37. Rip Esselstyn makes a related point about coconut oil, whose saturated fat he says promotes insulin resistance38.
Type 1 diabetes
Kahleova’s 2024 trial randomly assigned 58 people with type 1 diabetes to a low-fat vegan diet or a portion control diet for 12 weeks; daily insulin dropped by about 12 units in the vegan group and stayed flat in the other, while A1c fell in both groups with no significant difference8939. In 1982 Pritikin said a paper about to be published showed insulin doses cut by 40% in type 1 patients40. Campbell holds a different idea about its origin: in 2014 he described cow’s milk protein fragments prompting antibodies in infants that then attack the pancreas, and in 2021 he said this makes type 1 harder to reverse274142. Goldhamer wrote that people with type 1 can often stabilize on much less insulin, though his center does not fast them because they make little or none of their own4328.
Fasting and salt
Goldhamer’s center, by his 2011 article, often used supervised water only fasting for type 2 patients whose blood sugar resisted diet alone, but not for type 128. Kempner looked elsewhere: in 1993 he reasoned that, because salt and adrenal hormones help raise blood sugar, keeping salt as low as possible should bring high blood sugar down, and he said this had proved useful44.
Cure, reversal or improvement
McDougall called type 2 diabetes curable in 2011, and in 2026 Rogers said every case of type 2 can be cured1229. Ornish in 2012 called it preventable or even reversible for most people, Kahleova in 2024 said it can be reversed in many cases, and Barnard in 2022 said his trial suggested it need not be a one-way street45467. The trial data are more modest than the words: in Barnard’s type 2 trial both groups improved, and in Kahleova’s type 1 trial A1c moved about the same on both diets1039.
Older studies
Kempner’s 1958 paper followed 100 diabetic patients from 1944 to 1955 without a comparison group: fasting blood sugar fell in 63 and rose in 15, and insulin was stopped in 18 while 9 needed more11. In 1982 Pritikin reported from his center, in a Diabetes Care paper, that three quarters of the insulin-treated type 2 patients no longer needed it after four weeks13. His earlier study at the Long Beach VA hospital did have a control group, 19 patients in each arm, with diabetics among them47.
What each expert says
JMJohn McDougall
McDougall links type 2 diabetes tightly to being overweight, and sees the insulin resistance behind it as the body’s defense against storing more fat, with insulin output often around double the normal148. He moves patients onto starches, vegetables and a little exercise, reports taking many off large insulin doses on the first day under his care, and tells everyone else to see their own doctor first2312.
Esselstyn lists diabetes with high blood pressure, heart disease and stroke as the eventual result of decades of the usual American diet21. He tells of his father, who had a first heart attack at 43 and type 2 diabetes by 6349. In a 2009 talk he warned of teenagers in New York developing type 2 diabetes from soda and obesity50.
Campbell’s main point on type 1 diabetes is his account of cow’s milk protein provoking antibodies in infants that end up attacking the pancreas275141. In 2021 he said this early reaction makes type 1 harder to reverse, and that diets higher in animal protein push up blood sugar and A1c4226.
Barnard explains type 2 diabetes as fat inside muscle and liver cells blocking insulin, and says his NIH-funded team began testing a low-fat vegan diet against it in 2003430. In that trial A1c improved about threefold more on the vegan diet than on a conventional diabetes diet, and he describes one early participant who lost 60 pounds, stopped his medicines and reached an A1c of 5.3, down from 9.57. In 2025 he linked both processed and unprocessed meat to diabetes risk25.
Ornish argues that lifestyle can beat a pill because the body adjusts blood sugar from minute to minute, and that larger changes make it likelier that medicines can be lowered under a doctor’s care5218. A 2014 column reports that over a thousand program participants with type 2 diabetes lowered HbA1c from above 7 to under it, figures that come without a comparison group14. In a 2014 interview he said lowering blood sugar through diet prevents complications better than doing it with drugs53.
Goldhamer wrote in 2011 that whole natural foods and moderate aerobic exercise let most type 2 patients reach normal blood sugar without drugs43. In December 2025 he said people with type 2 diabetes actually carry extra insulin, and that their resistance to it stems from the same diet that made them heavy6.
Rogers sees type 2 diabetes as a disease of diet that no pill will cure, and his 2026 book Prevent and Reverse Diabetes opens with a note asking anyone on medicines to tell their doctor, because doses usually need to come down5417. He describes fat filling muscle, liver and finally the pancreas531.
Novick wrote in 2010 that the potato does not explain the rise in diabetes, pointing instead to excess weight, inactivity and a Western diet of refined foods, oil, cheese and extra calories2. In a 2000 update he wrote about insulin-resistant people who feared carbohydrates yet thrived once they followed the health center’s program55. In a 2012 talk he judged the nut trials in diabetes to show mixed and minimal effects56.
Nelson contrasted the Pima of Arizona, who developed widespread diabetes on processed, fatty food, with their Tarahumara relatives in Mexico who stayed lean on starch20. In 2026 he pointed to Barnard’s type 2 trial and to a type 1 trial whose diet held fat near 10% of calories, arguing that insulin resistance on a plant based diet suggests too much fat1036.
In a 2018 seven-day pilot with about 60 people and no comparison group, Rip Esselstyn reported fasting blood sugar down about five points on average, and said eight of the 23 who started the week with prediabetes or type 2 diabetes ended it with normal fasting values5715. In 2022 he said fiber steadies blood sugar and pointed a prediabetes questioner to the Mastering Diabetes founders5859.
Diabetes is the center of Kahleova’s research at Barnard’s Physicians Committee, including the 2024 type 1 trial in which the vegan group’s insulin needs fell by about 28%609. She co-wrote the 2023 European nutrition recommendations for diabetes, which place plant foods such as vegetables, legumes, fruit and whole grains at the top, and contrasted them with US advice centered on calories and carbohydrates6162. A 2026 release from her organization reports a 16-week trial in type 2 diabetes in which liver fat fell by a third, without calorie counting; the release gives no comparison group6364.
Kempner proposed his rice diet in 1954 for diabetics showing early heart, eye, kidney or circulation damage65. His 1958 paper on 100 patients, without a comparison group, found insulin could be stopped in 18 and lowered in 4211. In 1993 he wrote that patients found it easier to blame heredity than to give up a rich, salty diet44.
In 1982 Pritikin said diabetes deaths fell during the war together with heart deaths, and claimed his program could free nearly all type 2 patients from insulin and pills in a month or two6667. For support he named a Diabetes Care paper from his center that year and a type 1 paper then in press1340.
Jenkins, known for the glycemic index, said in 2006 that diets causing smaller rises in blood glucose are what one wants in diabetes33. In his group’s view, lowering the glycemic load helps most those with big glucose swings, insulin resistance or a thick waist68.
Burkitt placed diabetes with the Western diseases seldom seen in traditional societies, and argued from the Pima that lifestyle, not genes, drives it193. He recommended peas and beans for their soluble fiber69.
The doctors above agree that diet changes can move blood sugar quickly, and McDougall, Rogers and Ornish all say medicines then have to be adjusted by a doctor121718. If you have diabetes or prediabetes, talk with your own doctor before you change how you eat, and keep taking your medicines, insulin included, exactly as your doctor directs. Any change to a dose is your doctor’s decision, made with regular blood sugar checks. This page is education, not medical advice.
Each small number links to its reference below: the video at the exact second, or the page where it was said.