Intermittent Fasting and Type 2 Diabetes: The Guideline Debate

Three researchers have argued in The Lancet Diabetes & Endocrinology that clinical practice guidelines should now list intermittent fasting as one of several lifestyle options for type 2 diabetes. Their article, published online on 26 June 2026, is a Personal View: an opinion piece, not a guideline, and no guideline has changed. Two studies from June 2026 help put it in context.

Read this first

If you have diabetes or take medication that lowers blood sugar, do not start fasting on your own. Going without food while on these medicines can cause hypoglycemia (dangerously low blood sugar), and doses often need to be adjusted before and during any fasting plan. That is a job for the doctor who knows your treatment. The trials described here were research under supervision, not a treatment to try alone.

This is general information, not medical advice. People who are pregnant or breastfeeding, have a history of eating disorders, take other regular medication, or are under 18 should also talk to a doctor before fasting.

What the Personal View argues

The authors, from the University of Illinois Chicago, the University of Minnesota and the Harvard T H Chan School of Public Health, start from a gap. They count more than 225 clinical trials of intermittent fasting, nearly 40 in people with diabetes or prediabetes. Yet neither the American Diabetes Association Standards of Care nor the International Diabetes Federation Global Clinical Practice Recommendations recommend it.

Their reading of the evidence, as summarized in the abstract:

• Type 2 diabetes. Different forms of intermittent fasting, including time-restricted eating, the 5:2 diet and the fasting mimicking diet, reduce HbA1c (typically by 0.3 to 1.2%), fasting glucose, 24-hour glucose and body weight relative to usual eating patterns.
• Compared with calorie restriction. Intermittent fasting is as effective for improving glycemic control. As effective, not more effective.
• Hypoglycemia. It does not increase the risk of hypoglycemia "when medications are adjusted using simple rules". That is fasting with medication managed by a clinician.
• Prediabetes. Intermittent fasting "might" reduce glycemia.
• Type 1 and gestational diabetes. There are insufficient data to recommend it.

From this they argue that the evidence "has sufficiently matured" to include intermittent fasting in guidelines as one of several evidence-based lifestyle options for type 2 diabetes. (HbA1c is a blood test that reflects average blood sugar over time.)

Who is making the argument

A Personal View is the authors' interpretation, not a new trial. The record carries these declarations of interest:

• All three authors received research grants from the US National Institutes of Health, and one also from the US Department of Defense.
• One wrote a diet book about every-other-day eating.
• One received continuous glucose monitors from a manufacturer for her research and is an associate editor at a diabetes journal.
• One chairs the safety monitoring committee of another intermittent fasting study and, from late spring 2026, will receive consulting fees from a company for content knowledge on intermittent fasting, circadian rhythms and nutrition.

None of this makes the argument wrong. It does mean the piece is advocacy by people invested in the field.

What two June 2026 studies found

A trial in adults at risk of type 2 diabetes

A non-inferiority trial published online in Diabetologia on 6 June 2026 compared time-restricted eating with individualized dietetic guidance at two Australian clinical research institutes. The participants did not have diabetes: they were adults with overweight or obesity and a high score on an Australian diabetes risk tool, with an average HbA1c of 5.8% at the start.

One hundred and twenty-four people were assigned to time-restricted eating (a self-selected 9-hour window, last eating occasion by 19:00) and 123 to dietetic guidance.

Time point	Difference in HbA1c between groups (95% CI)	What the authors concluded
4 months	-0.02% (-0.07 to 0.03), p = 0.40	Time-restricted eating was non-inferior, but not superior
12 months	0.05% (-0.02 to 0.11), p = 0.14	Non-inferiority could no longer be concluded

The non-inferiority margin was 0.10%, and at 12 months the range crossed it.

The sentence that matters most comes next in the abstract: the absolute changes in HbA1c were small and not clinically meaningful in either group at either time point. Adverse events were minor and did not differ between groups. The authors, who declare no relationships that might bias the work, suggest it may be a practical short-term alternative when dietetic support is limited or a person prefers it.

A meta-analysis in type 2 diabetes and prediabetes

A meta-analysis in the June 2026 issue of the Asia Pacific Journal of Clinical Nutrition pooled 14 randomized trials (899 patients) published before 30 September 2025.

• Against eating as usual (4 studies, 316 patients): intermittent fasting did better on HbA1c (standardized mean difference -0.64, 95% CI -1.04 to -0.24), fasting blood glucose (-0.39, -0.64 to -0.15) and body weight (-0.30, -0.55 to -0.05).
• Against continuous calorie restriction (10 studies, 583 patients): the effects were similar.
• Dropout: 21.5% with intermittent fasting, 26.2% with continuous calorie restriction and 15.9% with usual eating.

These are standardized mean differences, not percentage points of HbA1c. The comparison against usual eating rests on only four studies. The authors declare no conflict of interest.

How the pieces fit together

In people with type 2 diabetes or prediabetes, the pooled trials found that intermittent fasting improved blood sugar compared with eating as usual and performed about as well as calorie restriction. That fits the Personal View's actual request: fasting as one option among several, not a better one.

The Diabetologia trial adds a limit. In at-risk adults without diabetes, neither time-restricted eating nor dietitian-led advice changed HbA1c meaningfully over 12 months.

What this means if you fast

If you do not have diabetes, a fasting schedule is one reasonable option, and the best choice is the one you can keep. Our guide to choosing a fasting schedule and the schedules overview can help.

If you do have diabetes, the useful step is a conversation. Ask your doctor whether fasting fits your treatment. Any medication changes belong in that appointment.

QoFast is a fasting tracker app for iPhone and Android for intermittent fasting plans such as 16:8, 18:6, 20:4, OMAD and custom schedules. It gives general information, not medical advice, and is not a tool for managing diabetes. It is free to download with an optional Pro subscription: App Store and Google Play.

FAQ

Do diabetes guidelines recommend intermittent fasting? According to the Personal View, the American Diabetes Association Standards of Care and the International Diabetes Federation recommendations do not include it. The authors are arguing that they should.

Is it safe to fast with diabetes? Not without medical supervision. The Personal View's hypoglycemia statement applies when medications are adjusted, which requires your doctor.

Does time-restricted eating lower blood sugar in people at risk? In the Diabetologia trial, HbA1c changes were not clinically meaningful in either group.

Sources

• Varady KA, Chow LS, Peterson CM. "Intermittent fasting to treat diabetes: time to update clinical practice guidelines." The Lancet Diabetes & Endocrinology, published online 26 June 2026. DOI: 10.1016/S2213-8587(26)00119-1, PubMed 42361830.
• Parr EB, et al. "Time-restricted eating versus dietetic guidance on glycaemic outcomes in adults at risk of type 2 diabetes: a non-inferiority randomised clinical trial." Diabetologia, published online 6 June 2026. DOI: 10.1007/s00125-026-06762-x, PubMed 42251202.
• Chen J, et al. "Efficacy of intermittent fasting on blood glucose and weight in type 2 diabetes and prediabetes: A comparison with ad libitum and continuous energy restriction diets." Asia Pacific Journal of Clinical Nutrition, June 2026, 35(3):398-411. DOI: 10.6133/apjcn.202606_35(3).0003, PubMed 42242847.