NIHR Signal Type 2 diabetes can be reversed with a low-calorie diet

Published on 13 February 2018

Nearly half of people given a formula replacement diet of 830 calories per day for three to five months, followed by food reintroduction, went into remission from type 2 diabetes. They were supported to achieve and maintain weight reduction by primary care nurses or dieticians.

This trial involved 298 adults who had been diagnosed with type 2 diabetes within six years. Those that stuck with the program were more likely to lose weight (average 10kg was lost) and to go into remission compared to usual care. More than two-thirds of them were also able to stop both diabetic and high blood pressure tablets.

These results were seen after one year. The challenge will be to see if the results can be maintained over the planned four year follow-up period. If successful, the program could be easily replicated by other GP surgeries with minimal training requirements.

  •   Diabetes, Medicines, Obesity and nutrition, Primary care
Type 2 diabetes can be reversed with a low-calorie diet

Why was this study needed?

Type 2 diabetes is a growing epidemic, costing the NHS £23.7 billion per year. By 2016 there were nearly 3.6 million people diagnosed with diabetes in the UK and a further one million estimated to have undiagnosed type 2 diabetes. Rates of diabetes appear to be increasing.

Bariatric surgery has been shown to put people into remission from diabetes. However, this is not a feasible solution for all people with type 2 diabetes, and there can be serious complications.

This study aimed to see if a very low-calorie diet program delivered in primary care could lead to at least 15kg weight loss and remission of type 2 diabetes.

What did this study do?

The DiRECT cluster-randomised trial involved 49 GP practices in Scotland and Tyneside. The practices were assigned to either provide an intense weight management program or usual care according to NICE guidelines for overweight adults with type 2 diabetes. The 298 participants had been diagnosed within the last six years and were not on insulin injections.

The program involved stopping diabetic and high blood pressure medications and going on a formula replacement diet. This consisted of 825 to 853 calories per day for three to five months. Food was then reintroduced over two to eight weeks and walking encouraged through the use of a pedometer. Structured support for long-term weight loss maintenance was provided by a nurse or dietitian.

This was a well-conducted study, and the results are likely to be reliable.

What did it find?

By 12 months:

  • Weight loss of 15kg or more was achieved by 24% of the intervention group compared to none of the usual care group. Average weight loss was 10kg in the intervention group versus 1kg in the usual care group (adjusted difference ‑8.8kg, 95% confidence interval [CI] ‑10.3 to ‑7.3).
  • Diabetes remission occurred in 46% of the intervention group compared to 4% of the usual care group. It only occurred in people who had lost weight.
  • Remission was 20 times more likely with the intervention (odds ratio 19.7, 95% CI 7.8 to 49.8).
  • Better adherence to the regime increased the amount of weight loss and the chance of diabetes remission.
  • Diabetes medication did not need to be reintroduced for 74% of the intervention group. This was compared to 18% of the usual care group who were able to stop them. Blood pressure tablets were also no longer required for 68% of the intervention group compared to 39% of the usual care group.

What does current guidance say on this issue?

The NICE guideline on type 2 diabetes (2015) recommends the same dietary advice that is given to the general population. This includes eating a high-fibre, low glycaemic index, low-fat diet with plenty of fruit and vegetables. They encourage weight loss and exercise but do not specify very low-calorie diets to go into remission.

What are the implications?

This study shows that it is possible to achieve remission from type 2 diabetes by dieting. Admittedly this was a very motivated group - but the results show that it can be done, and so should encourage more people to give it a go.

Resource implications include the eight hours of training the nurse or dietitian received. This included cognitive behavioural therapy techniques to facilitate behaviour change. The likely increased number of regular appointments would also need to be factored in. Nevertheless, this is a promising relatively low resource intervention which is likely to be cost-saving in the long run if the results can be maintained.

Citation and Funding

Lean MEJ, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT): an open-label, cluster-randomised trial. Lancet. 2017. [Epub ahead of print].

This project was funded by Diabetes UK.


Diabetes UK. State of the Nation 2016: Time to take control of diabetes. London: Diabetes UK; 2016.

NICE. Type 2 diabetes in adults: management. NG28. London: National Institute for Health and care Excellence; 2015.

Expert commentary

The idea that type 2 diabetes can be put into remission with weight loss is not new, but the DiRECT trial shows that this is possible, and safe, in primary care in the UK, at least for those early in the course of the disease. However, maintaining weight loss is difficult, and the long-term outcomes of this on-going trial will be important.

In the meantime, those patients who are interested in taking the approach used in DiRECT to help manage their diabetes should be supported to do so.

John Wilding, Professor of Medicine & Honorary Consultant Physician, Obesity and Endocrinology Research, Theme Lead for Metabolism and Nutrition, Institute of Ageing & Chronic Disease, University Hospital Aintree


The structured support included cognitive behavioural therapy techniques. The time in each phase - diet replacement, food reintroduction and maintenance - was tailored to the individual. Increased exercise was encouraged, aiming for up to 15,000 steps per day. However, this was not achieved by many as there was no difference in reported exercise in either group between baseline and 12 months.