You’re caught up.

Home/Issues/Issue 11/Study

Randomised crossover trial PubMed records 11 March 2026 as the publication date. Volume 50, issue 6, pages 1228 to 1235.

Four hours of eating a day produced a four hundred calorie deficit, and the hunger hormones answered the way they answer any weight loss

AI narration, generated on first listen
Journal
International Journal of Obesity 50(6):1228 to 1235
Authors
Kramer CK, Zinman B, Feig DS, Retnakaran R
Institution
the Leadership Sinai Centre for Diabetes at Mount Sinai Hospital and the University of Toronto
Published
11 March 2026
Source
PMID 41814085 · DOI 10.1038/s41366-026-02042-1
Design
Randomised crossover trial. Six weeks of twenty hours fasting with a four hour eating window against six weeks of standard lifestyle, with a four week washout, and oral glucose tolerance testing at four visits.
Sample
Thirty nine adults with overweight or obesity and type 2 diabetes.

Why they ran it

The interesting claim about fasting is not that it removes weight, it is that it removes weight differently, without the counter regulation that makes weight loss hard to hold. The authors note that time restricted eating shifts substrate use toward lipolysis and ketogenesis, and say this raises the possibility of a different effect on energy balance regulation. Then they measure the regulation.

Drawn from the paper's introduction and stated aim.

Randomised crossover. Thirty nine adults with overweight or obesity and type 2 diabetes did six weeks of twenty hours fasting with a four hour eating window and six weeks of standard lifestyle, in random order, separated by a four week washout. A seventy five gram oral glucose tolerance test at baseline and at six, ten and sixteen weeks.

Participants reported eating 384 calories a day less during the fasting period, with a standard deviation of 488 and a p value below 0.001, and no difference in macronutrient distribution. The parent trial achieved a 3.86 percent reduction in body weight.

Fasting leptin fell by 2,445 with a standard error of 885, p equals 0.009, and the leptin response across the glucose challenge fell too, p below 0.001.

Fasting ghrelin rose by 28 picograms per millilitre, standard error 11.3, p equals 0.02. That comparison is after the fasting period against before it, inside the same arm.

The ghrelin response across the glucose challenge, compared between the two arms, did not differ, p equals 0.35. Neither did glucagon at p equals 0.99, GLP-1 at p equals 0.98, or peptide YY at p equals 0.61.

The authors conclude that this does not prevent the physiologic compensatory changes associated with weight reduction.

The numbers

Self reported intake during fastingminus 384 calories per day, standard deviation 488, p below 0.001
Fasting leptinminus 2,445, standard error 885, p equals 0.009
Leptin response across the glucose challengereduced, p below 0.001
Fasting ghrelin, within arm, after against beforeplus 28 pg/mL, standard error 11.3, p equals 0.02
Ghrelin response across the challenge, between armsp equals 0.35, no difference
Glucagon, GLP-1, peptide YY between armsp equals 0.99, 0.98 and 0.61, no differences
Protocol20 hours fasting, 4 hour window, 6 weeks per arm, 4 week washout
Weight change in the parent trialminus 3.86 percent

Why this might happen

Proposed by the authors This is the explanation the authors offer in their discussion. This study did not test it.

The authors open with the case for a difference, that time restricted eating shifts substrate use toward lipolysis and ketogenesis. They then measure the counter regulation and find the ordinary pattern: leptin down, fasting ghrelin up, incretin and satiety peptides unchanged. Their conclusion names the negative directly, that this does not prevent the physiologic compensatory changes associated with weight reduction.

Drawn from the paper's stated conclusion.

What this does not show

  • It does not show fasting compensates worse than ordinary dieting. There was no ordinary dieting arm. The comparator is standard lifestyle, so the result is about whether fasting escapes compensation, not about whether it escapes it better or worse than anything else.
  • The ghrelin rise is a within arm comparison. Plus 28 picograms per millilitre at p equals 0.02 is after the fasting period against before it. The between arm comparison across the glucose challenge was not significant at p equals 0.35. Those are different questions and only the first one moved.
  • The calorie deficit is self reported. Minus 384 calories a day with a standard deviation of 488 is a soft number. It should be read as approximate and as evidence that a deficit existed rather than as a measurement of its size.
  • It does not describe people without type 2 diabetes. Every participant had overweight or obesity and type 2 diabetes, and the hormonal set point in that group is not the general case.

Where this leaves us

A four hour eating window produced its weight loss the ordinary way, and the body answered it the ordinary way.

Thirty nine adults with overweight or obesity and type 2 diabetes, against their own standard lifestyle.

The same measurements with a matched calorie restriction arm, which would separate fasting from weight loss rather than from doing nothing.

Caveats worth holding

  • Intake was self reported, not measured.
  • The comparator is standard lifestyle rather than a matched calorie restriction.
  • Thirty nine people in a crossover is adequate for leptin and thin for the gut peptides.

Newsletter

Each issue by email, when the newsletter launches. Leaving your address puts you on the list, nothing is sent yet.