How to Reverse Type 2 Diabetes: Diet, Weight, and More

Type 2 diabetes can be put into remission, and for many people the primary route is weight loss. In the landmark DiRECT trial, almost half of participants achieved remission within a year through a structured diet program run in ordinary primary care clinics. The key mechanism involves clearing excess fat from the liver and pancreas, which restores the body’s ability to regulate blood sugar. But “reversal” is more complicated than a single headline suggests, and the likelihood of success depends on how long you have had diabetes, how much weight you lose, and whether you can keep it off.

Why Researchers Say “Remission” Instead of “Cure”

An international expert panel proposed that the standard term should be “remission” rather than “reversal” or “cure.” The agreed-upon definition is an HbA1c below 6.5% measured at least three months after stopping all glucose-lowering medications.¹ The word “remission” was chosen deliberately. It signals that the underlying susceptibility has not disappeared: if you regain the weight or the metabolic conditions that triggered diabetes in the first place, blood sugar levels can climb back up. That does not mean the improvement is trivial. Going into remission means your body is handling glucose normally without medication, which reduces your risk of complications and can transform daily life. But it is an ongoing state to maintain, not a one-time fix.

The Fat Inside Your Organs

The central insight behind modern diabetes remission research is that excess fat stored inside the liver and pancreas drives the disease. A research group led by Roy Taylor at Newcastle University showed that when people with type 2 diabetes followed a very low calorie diet, fat inside the liver dropped dramatically within weeks, and pancreatic fat declined as well. As that organ fat cleared out, both insulin production by the pancreas and the liver’s response to insulin returned toward normal.²

The proposed explanation, sometimes called the twin-cycle hypothesis, describes a vicious loop. Too much fat in the liver increases insulin resistance, which makes the body pump out more insulin. That excess insulin drives even more fat production in the liver, and the liver starts exporting more fat into the bloodstream. When that fat reaches the pancreas, it gradually poisons the insulin-producing beta cells. Break the cycle by removing the excess organ fat, and the beta cells can recover, at least in many people.³ This is why weight loss works when medications alone often do not achieve remission: the drugs manage blood sugar, but they do not necessarily clear the fat that is causing the problem.

How Much Weight Loss Is Enough

The DiRECT trial gave the clearest picture of the relationship between weight loss and remission. Participants followed a total diet replacement program of about 825 to 853 calories per day using formula shakes for 12 to 20 weeks, then gradually reintroduced food, with ongoing support to maintain weight loss.⁴ At one year, about 46% of the intervention group achieved remission, compared to 4% in the control group receiving standard care.⁵

The dose-response relationship was striking. Among people who gained weight or stayed the same, nobody achieved remission. Those who lost 5 to 10 kg had about a one-in-three chance. Those who lost 10 to 15 kg had better than a coin flip. And among people who lost 15 kg or more, roughly 86% went into remission.⁵ The message is clear: more weight loss means a much higher chance of remission, and the threshold for a realistic shot seems to be somewhere around 10 kg for most people.

Low-Carb and Ketogenic Diets

You do not have to use meal-replacement shakes to lose enough weight for remission. Low-carbohydrate and ketogenic diets have attracted substantial interest because they lower blood sugar directly by reducing carbohydrate intake, and they tend to produce significant weight loss. A trial comparing a low-carbohydrate ketogenic diet to a low-glycemic-index diet in people with type 2 diabetes found that the ketogenic group had larger drops in HbA1c and body weight, and that diabetes medications were reduced or eliminated in over 95% of the ketogenic diet group versus about 62% in the comparison group.⁶

Reviews of the evidence generally confirm that very low-carb diets help reduce the need for insulin and oral diabetes medications.⁷ The practical challenge is sustainability. Ketogenic diets are restrictive, and long-term adherence is lower than most people expect when they start. The evidence suggests that whatever dietary approach reliably produces and maintains sufficient weight loss is the one most likely to lead to remission. The specific macronutrient composition matters less than the total amount of weight lost and kept off.

Intermittent Fasting

Intermittent fasting, which involves cycling between periods of eating and extended fasting, has also shown promise for blood sugar control and weight loss in people with type 2 diabetes. One randomized trial tested a 5:2 pattern (five days of normal eating, two days of meal replacement at reduced calories) against standard medications. Over 16 weeks, the fasting group had the largest drop in HbA1c and lost more weight than groups taking metformin or empagliflozin.⁸

The catch is what happens after the initial period ends. A systematic review found that while intermittent fasting consistently improved blood sugar and body composition over 12 to 24 weeks, follow-up at 12 to 18 months did not show sustained advantages, and most studies found no meaningful difference between intermittent fasting and simply eating fewer calories every day.⁹ Intermittent fasting can work well for people who find it easier to skip meals on certain days than to eat less every day. But it is not a metabolic shortcut: the benefits track with how much weight you lose, not with the fasting pattern itself.

Bariatric Surgery

For people with more severe obesity or those who have not been able to lose enough weight through diet alone, bariatric surgery is the most effective single intervention for diabetes remission. A meta-analysis of randomized trials found that at one year after surgery, about 53% of patients achieved remission compared to roughly 5% with medical management alone. That advantage persisted at two, three, and even five or more years.¹⁰ Surgery also reduced the risk of microvascular complications like kidney and eye disease.

A single-center study following patients for five years reported that about 80% achieved remission at some point after surgery and 60% maintained remission at the five-year mark. Among those who initially achieved remission, about one in four eventually relapsed, with most relapses occurring between the fourth and fifth year.¹¹ Surgery is not a permanent guarantee either, but it produces the largest and most durable weight loss of any current approach, which is why its remission rates are the highest.

Weight Loss Medications

The newer generation of weight loss drugs, particularly GLP-1 receptor agonists and dual-acting drugs like tirzepatide, are changing the conversation about diabetes remission. In the SURMOUNT-2 trial, tirzepatide produced average weight loss of roughly 13 to 15% over 72 weeks in people with both obesity and type 2 diabetes, with about 80% of participants on the drug losing at least 5% of their body weight.¹² That degree of weight loss puts many people in the range where remission becomes likely, based on what the DiRECT data showed.

The open question is whether remission achieved with these drugs lasts when the drug is stopped. Most trials show significant weight regain after discontinuation, which suggests that for many people, medication-assisted remission may require ongoing treatment. That is still a meaningful improvement over managing diabetes with insulin and multiple daily medications, but it changes the framing from “reversal” to long-term management with a different set of tools.

Who Is Most Likely to Succeed

The single strongest predictor of whether someone achieves remission is how long they have had diabetes. A Swedish registry study of bariatric surgery patients found that each additional year of diabetes duration reduced the odds of being free from diabetes medication, both at two and five years after surgery. Pre-surgery insulin use, older age, and higher baseline HbA1c also predicted worse outcomes, while higher BMI at baseline actually predicted better outcomes (because there was more weight available to lose).¹³

A trial specifically comparing people with shorter versus longer diabetes duration found that even with similar amounts of weight loss, those with longer-standing diabetes had a 32% lower remission rate. Still, half of those with longer duration achieved remission, which means it is not hopeless for people diagnosed years ago.¹⁴ The reason duration matters is that beta cells gradually lose their capacity to recover. Early in the disease, the cells are stressed but still functioning. After years of high blood sugar and fat exposure, some of those cells die off permanently. The earlier you act, the more beta cell function you can reclaim.

The Personal Fat Threshold

One common misconception is that you have to be very overweight to develop type 2 diabetes, or that only severely obese people can benefit from weight-loss-based remission strategies. The personal fat threshold hypothesis challenges that assumption. It proposes that each person has their own individual threshold for how much fat their body can handle before metabolic problems develop, and that threshold is independent of BMI. Some people develop diabetes at a BMI of 24; others never develop it at a BMI of 40.¹⁵

A study tested this directly by putting people with type 2 diabetes and a BMI below 27 through repeated weight loss cycles. After losing an average of about 6.5% of their body weight, 70% achieved sustained diabetes remission. Their liver fat fell to normal, and beta cell function improved substantially, even though these were people who were not traditionally considered overweight enough to benefit from weight loss interventions.¹⁶ If you have type 2 diabetes and are not particularly heavy, the evidence suggests you may still carry concealed excess fat in your liver and pancreas, and losing a relatively modest amount of weight can make a real difference.

Keeping Remission Long-Term

Achieving remission is one challenge. Maintaining it is another, and the data here are more sobering. The DiRECT trial followed participants for five years. By that point, average weight loss in the intervention group had declined to about 6 kg (down from 10 kg at one year), and only about 13% remained in remission. Among those who had been in remission at two years, roughly a quarter were still in remission at five years.¹⁷ The two-year results already showed the trend: about 36% of the intervention group was still in remission at 24 months, closely tied to how much weight they had maintained.⁴

These numbers are not a failure of the approach so much as a reflection of how difficult sustained weight loss is for anyone, with or without diabetes. The people who stayed in remission were the ones who kept the weight off. What this means practically is that remission is not a single event you achieve and then forget about. It requires permanent changes in how you eat and how much support you get. The DiRECT five-year extension also showed that participants who had a stronger initial response, losing more weight and achieving lower HbA1c in the first year, were more likely to maintain remission and to participate in ongoing follow-up.¹⁸

Why Remission Is Worth Pursuing Even If It Does Not Last Forever

Even temporary remission appears to produce lasting benefits. A large observational study found that people with type 2 diabetes who achieved remission had a 25% lower risk of cardiovascular disease, about a 32% lower risk of microvascular complications like eye and kidney damage, and about a 22% lower risk of macrovascular complications like heart attack and stroke over seven years of follow-up, compared to those who never achieved remission.¹⁹ The period of normal blood sugar, even if it eventually ends, seems to give the body a window to recover from some of the damage that high glucose causes to blood vessels and nerves.

This finding reframes the conversation. Even if you achieve remission for two or three years and then relapse, you are not back to square one. You have spent years with better metabolic health, and that time matters. The goal is not perfection. It is doing as much as you can for as long as you can.

Psychological Hurdles and Daily Realities

The research on remission tends to focus on biology, but in practice, the biggest barriers are behavioral. A study using real-time tracking of mood and eating in people with type 2 diabetes found that higher levels of fatigue, food cravings, and feeling energized (which sounds counterintuitive, but being “pumped up” before a meal sometimes leads to impulsive eating) all predicted dietary lapses. Eating out was another strong predictor. On days when participants had a dietary lapse, their blood glucose levels were measurably higher, and daily stress independently pushed glucose levels up regardless of diet.²⁰

This is not a minor footnote. The people who maintained remission in the DiRECT trial were not metabolic unicorns. They were people who managed to sustain difficult dietary changes over years, which means they found ways to cope with cravings, social eating, stress, and the relentless pull of old habits. Structured support, whether from a clinical program, a dietitian, or even a digital platform, makes a real difference. A pilot of the DiRECT model adapted for Australian primary care found that the approach was feasible outside the UK and that participants were willing to pay for the formula diet products themselves, suggesting genuine buy-in when the program is accessible.²¹

Gut Bacteria and Emerging Research

One of the more interesting areas of emerging science is the role of gut bacteria in diabetes remission. A trial of a low-calorie diet in people with obesity and type 2 diabetes found that after 12 weeks, participants showed significant shifts in their gut microbiome, including increases in beneficial bacteria like Bifidobacterium and Lactobacillus.²² Separately, a study of patients undergoing bariatric surgery found that the specific gut bacteria present before surgery predicted who would go on to achieve diabetes remission afterward.²³

This research is still in its early stages and has not yet produced any gut-targeted treatment that reliably improves diabetes outcomes. But it helps explain why people with seemingly similar weight loss sometimes have very different metabolic responses: their gut ecosystems may be processing nutrients differently. It also raises the possibility that future treatments could combine weight loss with microbiome-targeted approaches for better remission rates.

Young People With Type 2 Diabetes

Type 2 diabetes in teenagers and young adults is rising, and the disease tends to be more aggressive in younger people, progressing faster and responding less well to standard medications. A small pilot study tested whether the same very-low-calorie approach used in adults could work in young people. Those who stuck with the diet lost a meaningful amount of weight, saw their liver fat cut by more than half, and most no longer met the diagnostic criteria for type 2 diabetes after 34 weeks.²⁴ The catch was adherence. Nearly half the participants could not stick with the diet, which highlights the need for programs specifically designed for younger patients, with different support structures and realistic expectations about compliance. The biology of remission appears to work the same way in young people, but the behavioral challenges are different and arguably steeper.