Plant-Based and Vegan Diets With Type 2 Diabetes: What the Research Shows and How to Do It Safely

Plant-based eating gets argued about more than almost any other part of diabetes care. Some people hear that a vegan diet can reverse type 2 diabetes. Others hear that beans and rice will spike your numbers and that you need meat to stay healthy. Neither is a safe thing to act on without looking at what the research actually studied. This guide goes through the evidence on vegetarian and vegan diets in type 2 diabetes, what the numbers are for the starchy plant foods that worry people most, which nutrients you have to plan for, and when a big change in how you eat needs a conversation with your care team first.

This is general education, not a treatment plan. Nothing here is medication advice, and no diet is guaranteed to change your blood sugar. If you take diabetes medicine, especially insulin or a sulfonylurea, talk with your clinician before making a large change in how you eat, because your needs can change as your food changes.

Quick answer

• Controlled trials of vegetarian and especially vegan diets in people with type 2 diabetes have, on average, reported a modest lower A1C than comparison diets. The pooled differences in the analyses below ranged from about 0.2 to 0.4 percentage points, and the trials were small and short.
• Large observational studies link healthful plant-based eating (whole grains, legumes, vegetables, fruit, nuts) with a lower risk of developing type 2 diabetes, and unhealthful plant-based eating (refined grains, sweetened drinks, potatoes, sweets) with a higher risk. Observational studies cannot prove cause.
• Plant foods are not carb-free. Beans, lentils, whole grains and fruit all contain carbohydrate, and portion size still matters. What they add is fiber and protein alongside the carbohydrate.
• A vegan diet has no reliable natural source of vitamin B12. The NIH fact sheet says people who eat little or no animal food might not get enough, and that metformin might reduce B12 absorption. Planning for B12 is not optional.
• The American Diabetes Association lists vegetarian and vegan patterns among the eating patterns that can fit diabetes care, and says there is no single plan that works for everyone. The best pattern is one you can follow safely for years.

What counts as plant-based? The terms are not the same

People use “plant-based” loosely, and studies define it differently, which is one reason results vary. Here is how the common terms are used.

When you read a headline about “a plant-based diet,” look for which of these was actually tested. A trial of a low-fat vegan diet with meals provided tells you something different from a survey of people who eat less meat.

What the research shows

The evidence falls into two groups that answer different questions. Trials ask whether changing diet changes a measurement such as A1C. Observational studies ask whether people who already eat a certain way develop diabetes more or less often. Both are useful, and neither alone settles the question.

How to read these numbers

A drop of 0.2 to 0.4 points in A1C is real but modest. These studies do not show that plant-based diets beat every other pattern. In several of them, the comparison group was also following a therapeutic diabetes diet, which can narrow the gap. The 2011 trial shows the pattern clearly: both groups were on calorie-restricted diets with meals provided, so part of the difference may come from weight loss and visceral fat loss, not from the absence of meat itself. The authors of that trial attributed some of the benefit to greater visceral fat loss.

Size and length matter too. The pooled trials involve hundreds of people, not thousands, and most last months, not years. Definitions of “vegan” differed from trial to trial. The vegan meta-analysis authors noted this, along with the fact that the vegan and control diets were not matched in every way other than veganism, so differences in calories and macronutrients may explain part of the effect. I read the conference report of that analysis, not the full published paper, so treat the figures as preliminary.

The observational result is the one most often quoted, and it deserves a careful reading. In the UK Biobank analysis, the people who ate a healthful plant-based pattern developed type 2 diabetes less often, and a large share of that link ran through body weight and waist size: BMI and waist circumference each accounted for about 28% of the association. The authors note that an observational design cannot establish cause, and that diet was measured at only a few points in time. The same analysis found the opposite direction for people whose plant-based diet was made up of refined grains, sugary drinks and sweets.

Put together, a fair summary is this: well-planned vegetarian and vegan diets appear to be a safe and reasonable option for many adults with type 2 diabetes, and they may lower A1C and body weight by modest amounts. They are not the only option, and the research does not support promising any individual a particular result.

What diabetes and nutrition organizations say

The American Diabetes Association says on its eating for diabetes management page that there is no single eating plan that works for everyone, and describes vegetarian and vegan patterns among the options, noting they may help people reduce diabetes risk, lower A1C, lose weight or lower LDL cholesterol. It also stresses food choices you will want to stick with over time over fast-weight-loss fad diets.

A published review titled “Nutrition concerns and health effects of vegetarian diets,” whose abstract record I read at Andrews University, concluded that a well-planned vegetarian diet that includes fortified foods can be nutritionally adequate for adults and children. It named vitamin B12, vitamin D, omega-3 fatty acids, calcium, iron and zinc as the nutrients of concern, and said supplements and fortified foods can offer a useful shield against deficiency. I could only read the abstract, not the full review, and I could not open the Academy of Nutrition and Dietetics position paper itself.

Plant foods still have carbohydrate: the numbers

The most common worry about plant-based eating with diabetes is that beans, lentils, grains and fruit are all carbohydrate foods. That is true. The practical question is how much carbohydrate comes with how much fiber and protein, and how big your portion is. The table uses label data from the USDA FoodData Central database and from a lentil industry brochure. I could not use the USDA website directly, so I read the records through a public mirror that credits USDA.

Notes: the black bean and chickpea rows are store-brand label data in the USDA branded database, last updated 2019 to 2020, and brands vary. The lentil brochure showed two panels for whole and split lentils and I could not tell which panel belonged to which type, so the fiber figure is given as a range of 4 or 9 g and I treat it as unverified. Quinoa, tofu, tempeh, oats and fruit are not in the table because I could not verify per-serving figures for them from a fetchable source; check their labels, or see our guides to oatmeal and bananas.

Portion math with beans and rice

Scaling the label figures: a full cup of the black beans in the table is about 44 g of carbohydrate, and a full cup of the chickpeas about 40 g. Those are the same amounts you would count for a cup of rice or a few slices of bread. The fiber is the difference: beans carry about 27% of their carbohydrate weight as fiber and chickpeas about 35%, compared with about 9% for brown rice and 5% for millet.

Here is how that plays out in a rice-and-beans bowl. A bowl with 1 cup of brown rice and 1 cup of black beans comes to about 90 g of carbohydrate and 16 g of fiber. A bowl with half a cup of each, plus a generous pile of non-starchy vegetables, comes to about 45 g of carbohydrate and 8 g of fiber. That is a 45 g difference from the same ingredients, entirely from portion size. Counting the vegetables as zero is a simplification, since they contain a little carbohydrate too.

Swapping grains also changes less than people expect. By the USDA records, a cup of millet has about 4.8 g less carbohydrate than a cup of brown rice and about half the fiber. Neither is a low-carb food. If your goal is a lower-carbohydrate plate, the portion usually matters more than which grain it is. Our guide to white rice and blood sugar and our potato guide walk through the same kind of math for other starches, and our label-reading guide covers how to read the numbers on a package.

Getting enough protein without meat

Protein is rarely the hard part for people eating plant-based, but it takes more planning than adding a chicken breast. A Mayo Clinic page on planning a vegetarian diet lists soy products, legumes, nuts, seeds and whole grains as sources, and eggs and dairy for lacto-ovo eaters. The table gives verified per-serving protein where I have it.

Two practical points. First, most of the high-protein plant foods also carry carbohydrate, so adding more beans to get protein adds carbohydrate too. Some soy foods may carry less carbohydrate per serving, which is why labels matter. Second, if you are eating plant-based to manage weight as well as A1C, protein helps with fullness, and spreading it through the day is a reasonable habit to discuss with a dietitian.

The nutrients you have to plan for

This is the part of plant-based eating that cannot be skipped. A well-planned diet can cover these nutrients. A casual one often does not.

Vitamin B12

The NIH Office of Dietary Supplements consumer fact sheet on vitamin B12 puts the recommended daily amount for adults at 2.4 micrograms. It says people who eat little or no animal foods, including vegetarians and vegans, might not get enough from their diets, and that B12 occurs naturally only in animal foods. Fortified cereals and nutritional yeast are among the fortified options it lists. Mayo Clinic’s guidance says B12 deficiency may go undetected in people who eat a vegan diet and that it is important for vegans to consider supplements, B12-enriched cereals and fortified soy products.

There is a second reason to pay attention if you have type 2 diabetes. The same NIH fact sheet says metformin might reduce B12 absorption and lower blood levels. So a person who eats vegan and takes metformin has two separate reasons to watch B12. Our guide to metformin, vegetarian diets and B12 risk goes through this. Which form and amount of B12 supplement is right for you, and how often to test your level, are questions for your clinician and the label, not for a blog post.

Vitamin D, omega-3 fats, calcium, iodine, iron and zinc

• Vitamin D: Mayo notes fortified milks, some cereals and some margarines as sources, and says people with limited fortified foods and sun exposure may want to discuss supplements with a health care provider.
• Omega-3 fats: Vegetarians who skip fish may fall short on DHA and EPA. Walnuts and flaxseed provide a different omega-3 (the Mayo page notes this), and fortified products are an option. Whether you need an algae-based supplement is something to ask about.
• Calcium: Dark greens like kale and broccoli and calcium-fortified foods are plant options. Check the label on plant milks and tofu for whether they are calcium-fortified.
• Iodine: Iodized salt is an easy source. Without seafood, dairy or iodized salt, deficiency is a risk, according to Mayo. This is worth raising with your clinician if you eat little salt.
• Iron and zinc: Mayo notes that iron and zinc from plants are absorbed less easily, and that eating vitamin C-rich foods alongside iron helps. Beans, lentils, whole grains, soy and nuts are sources.

If you are planning a vegan diet, a reasonable first step is a visit with a registered dietitian who works with diabetes and a blood test to see where you start. I did not find a single authoritative number for every one of these nutrients that applies to everyone, so I am deliberately not listing doses.

Plant-based does not mean healthy: the refined-carb trap

The UK Biobank result above is the clearest warning. An unhealthful plant-based pattern was linked to a higher risk of type 2 diabetes, not a lower one. A diet made of white bread, pasta, sweetened drinks, chips, fries, cookies and meat substitutes is technically vegan, and it will not behave like a diet built on beans, vegetables, intact grains and nuts.

Common traps include fruit juice and smoothies that deliver a lot of sugar with little fiber, vegan desserts and snack bars that use syrups, and big portions of refined grains or potatoes. Our guide to sweeteners and “sugar-free” labels can help when a package is making health claims. Reading the carbohydrate, fiber and added sugar lines is just as important on vegan products as on any other.

A sample day, with the carb math

This sample is built only from foods where I have a number, so it is partial. It leaves out vegetables, tofu, seasonings, drinks and snacks. Think of it as a way to see how carbohydrate is spread out, not as a meal plan for you.

Those three meals add up to about 125 g of carbohydrate from the counted foods, and about 19 g of fiber from the chickpeas, bread, rice and beans (the oats and banana fiber are not included). The oat figure comes from our oatmeal guide, the banana figure of roughly 27 g for a medium banana from our banana guide, and the bread figure from the USDA values in our bread guide. Half a banana is my scaling of the medium figure.

For a week of dinners, a rotation like lentil soup with a side salad, a tofu and vegetable stir-fry over a measured scoop of rice, bean chili, a chickpea and vegetable curry, a tempeh bowl, a big salad with beans and nuts, and a vegetable and bean pasta night gives variety without a new recipe every day. Those are ideas, not numbers. Check each recipe’s carbohydrate and portion yourself.

Cost and shopping tips

I did not verify any prices, so there are no dollar figures here. A few general points are worth knowing. Canned beans and lentils are convenient, and their labels list sodium, so compare brands. Frozen vegetables and fruit without added sauces or sugar are a practical way to keep produce around. Fortified plant milks and B12 supplements do cost money, so build them into the plan from the start. Our grocery store guides show how to read labels at specific stores.

Safety: when to slow down and ask first

If you take diabetes medicine

Large changes in food can change how much glucose your body is dealing with. If you take insulin or a sulfonylurea in particular, a quick drop in carbohydrate or calories without a plan with your care team can raise the chance of low blood sugar. This article does not give dosing or adjustments, and nobody should adjust medication from reading a web page. Make the plan first, then change the diet.

If you have kidney disease

The National Kidney Foundation article on plant-based diets and type 2 diabetes says replacing meat with soy and other vegetable proteins may lower kidney disease risk for people with type 2 diabetes, but it calls the research preliminary and based on small studies. It also says that if you have been told to limit potassium or phosphorus, or are on dialysis, you should review your diet with your dietitian or nephrologist. Many plant foods that are good for most people, such as beans, nuts and some fruits and vegetables, are also high in potassium or phosphorus. I did not find a source that gives safe amounts for everyone, so this is a clinician-first topic.

If you have a history of disordered eating

Cutting out entire food groups can be risky for anyone with a history of disordered eating. If that is you, a plant-based diet is a decision to make with a clinician or therapist who knows your history, not on your own.

If you are pregnant or planning to be

Nutrient needs change in pregnancy, and this article does not cover them. Talk to your obstetric care team before changing your diet.

How this plays out in different situations

Prediabetes

The research linking healthful plant-based eating with lower diabetes risk is observational, so it is a reason to consider it, not a promise. If you are trying one of these patterns, the most useful signs are probably the boring ones: body weight and waist size, because they explained a large share of the link in the UK Biobank study, and your next A1C. See our diet for type 2 diabetes guide for the general food approach.

Type 2 diabetes managed with food only

For someone not on medication, there is less risk of medicine-related lows, but you still want a baseline A1C, B12 and a plan for the nutrient gaps above. A small number of controlled trials suggest a modest A1C benefit, so a reasonable approach is to try the pattern for a defined period with a check-in rather than all-or-nothing.

Type 2 diabetes on medication

Talk with your prescriber first. If you are on metformin, mention B12 testing. If you are on insulin or a sulfonylurea, ask specifically how the plan should handle lower blood sugar as your food changes. This is the situation where a registered dietitian and your prescriber working together matter most.

If you use a CGM

A continuous glucose monitor can show how your own body responds to a bowl of lentils or a plate of rice and beans, which no study average can. The data is useful but easy to over-read. See our guide to reading your CGM data and our post-meal glucose tracker for a structured way to test one change at a time. Whether a reading is a problem for you is a question for your care team.

Common mistakes

• Going vegan without a B12 plan. The deficiency can build quietly. See the B12 section and the NIH fact sheet.
• Replacing meat with refined carbs. Bread, pasta and potatoes in place of protein turns “plant-based” into the unhealthful pattern the research warns about.
• Assuming “vegan” or “plant-based” on a label means low sugar or low carbohydrate. Read the numbers.
• Treating beans as free food. They are good food, and they are carbohydrate food. Count them.
• Changing diet and medication at the same time without a plan. Make the plan with your care team.
• Testing everything at once. If you use a meter or CGM, change one meal at a time so you can tell what moved the number.

Myth vs fact

Questions to bring to your clinician or dietitian

• Is a vegetarian or vegan pattern safe for me given my medications and kidney function?
• How should we plan for low blood sugar if I change what I eat?
• Should I have my B12, vitamin D and iron checked before I start, and how often afterward?
• Do I need a B12 supplement or other supplement, and which form and amount?
• How many carbohydrate servings per meal fits my goals if beans and grains become staples?
• What would you want to see in my next A1C and other labs to know this is working?
• Can you refer me to a registered dietitian who works with diabetes and plant-based eating?

Where this fits with other diets

Plant-based is one of several patterns that research has studied in type 2 diabetes. For how it compares with others, see our comparison of diabetes diets, and our guides to keto and low-carb diets and the Mediterranean diet. Our diabetes food guide links all of our single-food guides in one place.

Frequently asked questions

Can a vegan diet lower A1C in type 2 diabetes?

In pooled analyses of small, short trials, vegetarian and vegan diets lowered A1C by roughly 0.2 to 0.4 percentage points on average compared with other diets. That is a modest average, individual results vary, and the trials do not promise a result for any one person.

Can a vegan diet reverse type 2 diabetes?

The studies I read do not claim that. They report modest average improvements in A1C and weight. Whether any individual could reduce medication on a diet change is a decision for them and their prescriber.

Are beans and lentils OK with diabetes?

They contain carbohydrate. A half cup of the canned black beans in our table has 22 g of carbohydrate and 6 g of fiber, and a half cup of chickpeas has 20 g and 7 g. Portion size matters, and they are not carb-free. Many dietitians still include them because of their fiber and protein.

Do I need a B12 supplement if I am vegan?

The NIH says plant foods do not naturally contain B12 and that vegetarians and vegans might not get enough. Mayo Clinic says it is important for vegans to consider supplements or fortified foods. What form and amount is right for you should come from your clinician or the label.

Is B12 a bigger concern if I take metformin?

The NIH fact sheet says metformin might reduce B12 absorption and lower blood levels. Combining a vegan diet with metformin gives two reasons to ask about B12 testing.

Is a plant-based diet safe if I have kidney disease?

It depends on your stage and what your care team has told you to limit. The National Kidney Foundation says that if you have been told to limit potassium or phosphorus, you should review your diet with a dietitian or nephrologist.

Is a plant-based diet better than keto or Mediterranean for diabetes?

The ADA says no single pattern works for everyone. Each has trial evidence and each has trade-offs. See our comparison of diabetes diets.

Can a plant-based diet raise blood sugar?

Any diet can if it is built on refined grains, sweetened drinks and large portions. An observational study linked unhealthful plant-based eating with higher diabetes risk. The quality of the plant foods matters.

Should I change my diabetes medication if I go vegan?

Do not adjust medication yourself. Talk with your prescriber before a large diet change, especially if you take insulin or a sulfonylurea, because your needs can change.

Where do I start?

A reasonable start is a visit with a registered dietitian, a baseline lab check, and a plan for B12 and other nutrients. Then change one meal at a time and track how you feel and what your readings show.

Sources

• American Diabetes Association: eating for diabetes management
• NIH Office of Dietary Supplements: vitamin B12 consumer fact sheet
• Mayo Clinic: planning a healthy vegetarian diet
• National Kidney Foundation: plant-based diets and type 2 diabetes
• Review abstract: nutrition concerns and health effects of vegetarian diets
• Academy research brief: vegetarian dietary patterns and type 2 diabetes
• Meta-analysis of vegetarian diets and glycemic control
• Report on vegan-diet meta-analysis (ECO 2022)
• MEDLINE record for the 2011 vegetarian diet trial
• Diabetes on the Net: healthy vs unhealthy plant diets (UK Biobank)
• Lentils Canada nutrition brochure
• USDA SR Legacy 168875 via GetFoodFacts: brown rice
• USDA SR Legacy 168871 via GetFoodFacts: millet
• USDA branded record via GetFoodFacts: black beans
• USDA branded record via GetFoodFacts: chickpeas

About the author: Charles Kirkland is the founder of Metabolic Clarity Labs and author of The Journey Back. He is not a physician. This is general education, not a medically reviewed treatment plan.

Sources checked October 31, 2026. Discuss personal decisions with a qualified clinician. Do not start, stop, or change prescribed treatment based on this article.