Gary Taubes argues that type 2 diabetes should be viewed not simply as a disease of elevated blood sugar, but as a disorder closely connected to insulin resistance and the body's difficulty handling carbohydrates.
Traditional treatment can successfully lower blood glucose with medications. Taubes asks a different question: if carbohydrates are the nutrient that most directly raises blood glucose, why not substantially reduce the amount of carbohydrate a person with diabetes eats?
He argues that carbohydrate restriction can lower blood glucose, reduce insulin requirements, help with weight loss, and—in some people—contribute to remission of type 2 diabetes.
Low-carbohydrate diets are now recognized as one evidence-based option for managing type 2 diabetes, although they aren't the only effective dietary approach.
Taubes' work is particularly interesting because he examines the history of diabetes treatment and argues that approaches used before the discovery of insulin—and later largely abandoned—may deserve another look.
His central question is worth considering:
Instead of only treating high blood sugar, can reducing dietary carbohydrate help address the metabolic problem itself?
Gary Taubes, an investigative science journalist and author, has spent decades questioning conventional thinking about obesity, nutrition, and diabetes.
In his book Rethinking Diabetes: What Science Reveals About Diet, Insulin, and Successful Treatments, Taubes examines the history of diabetes treatment and asks whether modern medicine has overlooked an important idea:
If people with diabetes have difficulty handling carbohydrates, shouldn't reducing carbohydrates be a central part of treatment?
Taubes believes the answer deserves far more attention.
Type 2 diabetes is characterized by elevated blood glucose and usually substantial insulin resistance.
Conventional treatment can include dietary changes, exercise, weight loss, oral medications, newer drugs such as GLP-1 receptor agonists and SGLT2 inhibitors, and sometimes insulin.
These treatments can be highly effective.
Taubes' criticism isn't that lowering blood glucose with medication is useless. Clearly, controlling high glucose is important.
Instead, he asks whether treatment has sometimes focused too heavily on controlling the resulting high blood sugar while paying insufficient attention to the dietary carbohydrate that produces much of the glucose entering the bloodstream after meals.
One of the fascinating parts of the history Taubes describes concerns diabetes treatment before insulin became available in the 1920s.
Physicians knew that carbohydrates increased glucose in the urine and blood.
Because they had few effective treatments, severely restricting carbohydrate was sometimes part of diabetes management.
For people with Type 1 diabetes, however, diet could not replace the missing insulin. The discovery of insulin was lifesaving and transformative.
As insulin and other medications became available, dietary treatment evolved considerably.
Taubes argues that some potentially useful lessons about carbohydrate restriction were gradually forgotten.
This distinction is essential.
People with Type 1 diabetes produce little or no insulin and require insulin treatment.
People with Type 2 diabetes usually still produce insulin—often large amounts in the earlier stages—but their tissues have become resistant to its effects.
Carbohydrate restriction may reduce glucose excursions and insulin requirements in both conditions, but it does not eliminate the need for insulin in Type 1 diabetes.
Taubes' argument about potentially reversing or achieving remission from diabetes applies primarily to Type 2 diabetes.
Taubes sometimes describes Type 2 diabetes as a condition of carbohydrate intolerance.
The phrase is deliberately simple.
Carbohydrates—particularly sugars and rapidly digested starches—have the greatest immediate effect on blood glucose.
If someone's glucose regulation is impaired, a high-carbohydrate meal presents a substantial metabolic challenge.
Taubes therefore asks:
Why prescribe large quantities of the nutrient that a person has the most difficulty metabolizing?
That question forms much of the logic behind low-carbohydrate treatment.
Reducing dietary carbohydrate generally reduces post-meal glucose excursions.
That means the body needs less insulin to deal with those meals.
For someone taking glucose-lowering medication, improved glucose levels may also mean that medication requirements decline.
Weight often falls as well, particularly when carbohydrate restriction reduces hunger and total food intake.
Improvements may include:
Lower blood glucose
Lower HbA1c
Lower triglycerides
Weight loss
Reduced medication requirements
Improved insulin sensitivity
The magnitude of improvement varies considerably among individuals.
The word “reversed” can cause confusion.
A more precise medical term is remission.
Type 2 diabetes remission generally means that blood glucose has fallen below the diabetic range for a defined period without glucose-lowering medication.
Remission is possible.
It has been demonstrated not only with low-carbohydrate diets but also with substantial weight loss achieved through other dietary interventions and bariatric surgery.
This tells us something important:
Type 2 diabetes isn't inevitably progressive.
For at least some people, substantial metabolic improvement can change the course of the disease.
Taubes emphasizes insulin and carbohydrate restriction, but weight loss itself is also important.
Research has shown that reducing excess fat stored in the liver and pancreas can markedly improve glucose regulation in some people with Type 2 diabetes.
This is one area where different theories may lead to similar practical advice.
Taubes emphasizes lowering carbohydrates and insulin demand.
Other researchers emphasize energy balance and reducing excess organ fat.
Both approaches can result in weight loss and improved insulin sensitivity.
Low-carbohydrate diets are now supported by a substantial body of clinical research.
Trials have found improvements in glucose control, triglycerides, weight, and medication requirements.
The American Diabetes Association and other professional organizations now recognize reducing overall carbohydrate intake as one possible evidence-based dietary strategy.
That represents an important change from the period when very-low-carbohydrate diets were often dismissed outright.
However, low-carb isn't the only dietary approach supported by evidence.
Mediterranean-style diets, calorie-restricted diets, and other eating patterns can also improve Type 2 diabetes, particularly when they produce sustained weight loss and better food quality.
The American Diabetes Association (ADA) now recognizes low-carbohydrate eating as one evidence-based option for people with diabetes. Its 2026 Standards of Care recommends considering reduced carbohydrate intake for some adults to improve blood glucose control and emphasizes that eating plans should be individualized.
The ADA also recognizes Mediterranean and other healthy eating patterns and does not recommend one diet for everyone. It notes that very-low-carbohydrate diets can improve A1C in the short term, while evidence of an advantage over other diets becomes less clear over longer periods.
The American Diabetes Society (ADS) is a newer organization that takes a more strongly low-carbohydrate approach to diabetes prevention and treatment. It identifies carbohydrate restriction as the central component of its nutrition guidelines and argues that reducing carbohydrates can lower blood glucose and A1C, improve Insulin Resistance, promote weight loss, and in some people help achieve remission of type 2 diabetes.
Its leadership includes physicians and researchers associated with metabolic and low-carbohydrate approaches, including Ken Berry, M.D., Ben Bikman, Ph.D., and Mariela Glandt, M.D. The ADS position is considerably more favorable toward carbohydrate restriction than conventional diabetes guidelines.
A common objection to low-carbohydrate diets is that reducing carbohydrates often means eating more fat.
Low-carb diets frequently improve triglycerides and HDL cholesterol.
LDL cholesterol is less predictable.
It falls in some people, changes little in others, and rises—sometimes substantially—in others.
This matters because LDL-containing particles may play a role in atherosclerotic cardiovascular disease. Although this is more in dispute than many people realize.
All major cardiovascular risk factors deserve attention.
Any diet has to work outside a research study.
Some people find carbohydrate restriction surprisingly easy because hunger diminishes and they enjoy the foods they eat.
Others find avoiding bread, pasta, rice, potatoes, desserts, and other carbohydrate-rich foods too restrictive.
Taubes argues that the appropriate comparison isn't whether a low-carb diet is effortless.
It is whether someone finds it more sustainable and effective than the alternatives.
Long-term adherence matters more than winning an argument about which diet is theoretically best.
This deserves special emphasis.
Someone with diabetes who substantially reduces carbohydrate intake can experience a rapid fall in blood glucose.
If that person continues taking the same amount of insulin or certain glucose-lowering medications, dangerously low blood glucose can occur.
Medication requirements may need to be adjusted promptly.
That should be done with a healthcare professional familiar with managing medication during carbohydrate restriction.
Do not stop or reduce diabetes medication on your own.
The diabetes-treatment landscape has also changed dramatically.
GLP-1 receptor agonists and SGLT2 inhibitors can provide benefits extending beyond glucose lowering for appropriate patients, including cardiovascular and kidney benefits.
This is important when evaluating Taubes' historical argument.
The choice isn't necessarily diet or medication.
For many people the best treatment may combine effective dietary change, physical activity, weight management, and appropriately selected medications.
The objective is better health, not ideological purity about how it is achieved.
Taubes has long argued that conventional nutrition science underestimated the role of carbohydrates and insulin in obesity and metabolic disease. His work helped revive scientific and public interest in low-carbohydrate diets.
Low-carbohydrate diets have often faced skepticism, particularly over their long-term effects on cardiovascular health. However, clinical studies have shown that these diets can improve several important metabolic markers, often including lower triglycerides, higher HDL cholesterol, improved blood glucose, and weight loss. LDL cholesterol, however, can rise substantially in some people.
Critics also argue that low-carbohydrate diets can be difficult to maintain. That varies considerably by individual. Many people find them sustainable and report that reduced hunger and greater satiety make it easier to eat less without deliberately restricting calories.
The larger debate over Taubes’s work concerns how much of obesity can be explained by carbohydrates and insulin specifically, rather than by a combination of diet quality, calorie intake, hormones, genetics, environment, and other factors.
What I Find Valuable About Taubes
Taubes' greatest contribution may be his willingness to examine the history of accepted medical ideas.
Scientific consensus is important, but consensus can change.
That doesn't mean established advice should automatically be distrusted.
It means we should remain willing to ask:
What is the evidence?
How strong is it?
What assumptions are we making?
And what happens when we test alternatives?
Those are useful questions well beyond diabetes.
Gary Taubes asks us to reconsider how we think about Type 2 diabetes.
High blood glucose is important, but it is part of a larger metabolic disorder involving insulin resistance and impaired carbohydrate metabolism.
For some people, substantially reducing dietary carbohydrate can improve blood glucose, reduce medication requirements, promote weight loss, and in some cases help achieve diabetes remission.
It isn't the only effective strategy.
But it is a legitimate one.
And Taubes' fundamental question remains compelling:
If your body has difficulty handling carbohydrates, shouldn't reducing carbohydrates be one of the options considered?
Taubes's book examining the history of diabetes treatment and the evidence for carbohydrate restriction.
Taubes writes about nutrition science, diabetes, obesity, research methods, and the difficulties of reaching reliable conclusions in nutrition research.
The American Diabetes Association describes several evidence-based eating patterns for diabetes management, including Mediterranean, low-carbohydrate, and very-low-carbohydrate approaches, while emphasizing that nutrition should be individualized.
The American Diabetes Society presents the scientific case for carbohydrate restriction as a central strategy for improving blood glucose, Insulin Resistance,
and other metabolic problems associated with type 2 diabetes.
A video discussion of the ideas in Rethinking Diabetes, including insulin, carbohydrates, and the history of diabetes treatment.
Dr. Eric Westman interviews Dr. Jason Fung about type 2 diabetes, Insulin Resistance, and why reducing dietary carbohydrates and using intermittent fasting may help some people substantially improve—and sometimes reverse—the metabolic abnormalities associated with type 2 diabetes.
Dr. William Davis interviews Dr. Eric Westman about using carbohydrate restriction to improve Insulin Resistance and type 2 diabetes, and discusses cholesterol, cardiovascular risk, and the role of low-carbohydrate diets in metabolic health.