The Evidence Behind the Claim That Chewing Raises GLP-1

This article previously extended the story from a mouse study all the way to preventing diabetes in humans. It was rewritten in August 2026.

The term GLP-1 now often comes up alongside drug names like Mounjaro and Ozempic. You’ll sometimes see the claim that “you don’t need to rely on medication — chewing well raises your own GLP-1.” This article used to say that too.

First, here is what the study cited as the basis for that claim actually examined.

The Study Behind the Claim

Effects of Differences in Masticatory Dynamics and Chewing Habits on GLP-1 Secretion in Mice (E02)

This is a report held in the academic repository of the Health Sciences University of Hokkaido. The subjects were mice, and it was not published in a peer-reviewed journal.

It compares solid feed (a chewing condition) and liquid feed (a non-chewing condition, n=10 each) matched for calories and nutrients, and reports that active GLP-1 30 minutes after intake was significantly higher in the chewing group (p<0.05). When the vagus nerve is blocked, this difference disappears.

However, in this acute experiment, insulin secretion showed no significant difference under any condition. In a separate experiment with 12 weeks of feeding, fasting active GLP-1 and insulin were significantly higher in the long-term chewing group, and pancreatic β-cell area was also larger. However, there was no difference between the groups in body weight, blood glucose, or the oral glucose tolerance test.

The authors themselves distinguish between a single instance of chewing and chewing as a habit, writing that “chewing did not affect insulin secretion, but chewing habit did affect insulin secretion.”

That is as far as what this report actually states.

The Gap When Applying This to Humans

This article previously went on from there to call this “a safe approach that prevents the fundamental progression of impaired glucose tolerance and type 2 diabetes” and “a fundamental, non-invasive self-intervention program for raising natural blood concentrations of GLP-1 without relying on drug administration.” Both statements have been removed.

Here are the reasons.

• Mice are not humans. A mechanism observed in an animal experiment does not necessarily work the same way in humans. It is not unusual for metabolic interventions that looked promising in mice to fail to reproduce in human trials
• It has not been peer-reviewed. The article treated a report that has not passed third-party verification as if it were an established finding
• It did not track the onset of diabetes. What this report measured was hormone levels and cell area, not whether the mice were less likely to develop diabetes. Human onset was certainly not tracked at all
• It was not compared against medication. This was not a trial comparing GLP-1 receptor agonists with chewing well, so there is no basis for writing that you “don’t need to rely on medication”

The results of studies in humans are mixed. Small trials that specified the actual chewing count have reported increases in active GLP-1. The originals were verified in August 2026 and added to the evidence list.

• E17 (Sonoki et al. 2013) — in healthy volunteers, thirty chews per bite significantly raised active GLP-1 at 30 minutes (p<0.05). But glucose and insulin did not change under any condition, and in fifteen patients with type 2 diabetes there was no significant difference. The authors describe it as a pilot trial
• E18 (Li J et al. 2011) — forty chews per bite gave 11.9% less energy intake than fifteen, with higher GLP-1. This site has not been able to read the full text (paywalled), so significance levels and funding remain unverified
• E19 (Xu J et al. 2015) — gum chewing while fasting gave significantly higher GLP-1 at 30 minutes (p=0.031). But glucose, insulin and GIP showed no significant differences, and the authors phrase their conclusion as reducing the decline in GLP-1 during fasting rather than increasing it

E02’s discussion cited these as “Jie L” and “Jianping X”; the correct names are Li J and Xu J. The third appeared under Endocrine Journal, but was in fact published in Endocrine, a different journal.

What the three have in common is that GLP-1 and satiety differed while glucose and insulin did not.

The Two Trials That Varied Food Form Reach Opposite Conclusions

E05 is a trial that varied not the chewing count but the form of the food (walnuts eaten whole versus as a paste), and it found no significant difference in active GLP-1, blood glucose, or insulin.

E20 (Kamemoto et al. 2024) varied food form in the same way (shredded cabbage versus puree) but found total GLP-1 higher at 45, 60 and 90 minutes, with higher areas under the curve for insulin and GIP. It is a randomised crossover trial in 19 healthy young men.

E20 has to be read together with the following, however.

• GLP-1 was lower at 120 and 180 minutes, and the overall GLP-1 area under the curve showed no significant difference. The shape is a shift in timing rather than an increase
• Glucose showed no difference in area under the curve, and at 30 minutes it was higher in the chewing condition (p=0.007)
• It was funded by Kewpie Corporation, and 3 of the 9 authors are employees of that company. The paper states that the funder was not involved in data collection, analysis, interpretation or writing, but this is a conflict of interest readers should know about

Two trials of the same design reaching different conclusions is the current state of the food-form evidence.

It cannot be written that this has been “confirmed in humans,” nor that “nothing has been found in humans” — that is the accurate state of things.

The Explanation We Removed

This article also used to include an explanation of a pathway in which chewing stimulation travels from the trigeminal nerve to the brainstem, and efferent activity of the vagus nerve activates the intestine’s L cells ahead of time. This has been removed.

The cited report contains no description of directly measuring this pathway. While the article wrote “it is suggested,” the following paragraph swung back to asserting it as fact — calling it “a sophisticated neural input interface that turns on the switch of the entire metabolic factory” — which made the piece read as though it had been confirmed.

If You Are Considering Medication

If you are considering a GLP-1 receptor agonist, you do not need to stop seeing your doctor because of this article.

Prescription medications have their effects and side effects confirmed in human clinical trials, with dosing managed by a physician. This site has not been able to identify a study showing that chewing well can serve as a substitute.

If you need to make specific decisions about your weight or blood glucose, please consult a physician or registered dietitian. This site introduces research on chewing and has not been reviewed by a medical professional.

Further Reading

• How Much Does Increasing Chewing Count Change Postprandial Energy Expenditure?
• Adjusting Convenience Store Meals to Increase Chewing Count (the article covering E05)

---

※ This article is for informational purposes and is not a substitute for diagnosis or treatment.