Dr. Axe Wire · Study decoded · gut

Lab Model Shows Acid-Blocking Drugs Shift How Starch and Protein Break Down

A lab digestion model found suppressed stomach acid sped up starch breakdown but left more protein undigested than normal acid levels.

By Dr. Axe Editorial Team · October 29, 2026 · 1 source · Watch

At a glance

What they found: Under acid suppression, starch broke down more in the lab model, 53% versus 38% with normal acid, but protein broke down less, 85% versus 94%.
Who was studied: Cells in a lab dish.
What it doesn't show: Whether it works the same way in people.

A lab model simulating acid-suppressing drugs found they sped up starch digestion but left more protein undigested, according to a study published October 19, 2026 in Food & Function.

53% vs 38%starch breakdown

85% vs 94%protein breakdown

3protein sandwiches tested

Researchers built a dynamic in vitro gastric digestion system mimicking stomach churning and acid secretion, then digested three protein sandwiches (whey, egg white, almond) plus a bread control. Each meal ran twice: pH 0.8 with adjusting acid secretion, mimicking a healthy stomach, and pH 1.8 with fixed secretion at half normal volume, mimicking acid-suppression therapy like proton pump inhibitors.

After 180 minutes in the simulated stomach, starch breakdown averaged 53% under acid suppression versus 38% under normal acid across all meals. Protein breakdown stayed low in the stomach phase for every meal, averaging 8% for protein sandwiches and 23% for bread, with little difference between settings. After simulated small intestinal digestion, total starch breakdown became similar between conditions, but total protein breakdown reached 94.2% under normal acid versus 84.7% under acid suppression.

Researchers attributed faster starch breakdown under acid suppression to reduced inactivation of salivary amylase, which works best at higher pH and is normally shut off quickly by strong acid. Stomach acid also activates pepsin for protein digestion, so a less acidic environment can leave protein less broken down by the time food reaches the small intestine.

Starch Breakdown After 180 Minutes: Acid Suppression vs Normal Acid

Food & Function, 2026, Sun et al.

This is a mechanical simulation, not a study of people taking acid-suppressing medication, and doesn't report effects on absorption, blood sugar, or protein status in a living body. Testing in people on actual acid-suppressing drugs is the logical next step.

Dr. Axe’s Take

This is a lab digestion model, not a clinical trial, but it lines up with what I'd expect from years of looking at how stomach acid drives nutrient breakdown. Acid activates pepsin to unlock protein and shuts down the amylase that's already breaking down carbs. Long-term acid suppression can leave protein less digested and starch breaking down faster than it should. If you're on these medications, keep taking them as prescribed and talk with your doctor first. At home, I'd focus on chewing thoroughly, choosing easily digestible proteins like eggs and bone broth, and supporting stomach acid naturally with bitter foods and lemon water before meals.

— Dr. Axe

Take the Cellular Health Quiz →

Study summary

Who: A lab-based dynamic gastric digestion model, not human or animal subjects.
Design: In vitro digestion simulation comparing a normal-acid setting (pH 0.8, variable secretion) with an acid-suppressed setting (pH 1.8, fixed secretion at 50% volume), tested on three protein sandwiches (whey, egg white, almond) and a bread control.
Dose: Simulated gastric digestion for 180 minutes, followed by static small intestinal digestion.
Primary result: Gastric starch breakdown after 180 minutes: 53% (acid-suppressed) vs 38% (normal acid) across all meals.
Secondary: Total protein breakdown after small intestinal digestion: 94.2% (normal acid) vs 84.7% (acid-suppressed). Gastric-phase protein breakdown averaged 8% for protein sandwiches and 23% for bread, similar in both conditions. Total starch breakdown became similar between conditions after the small intestinal phase.
Funding / conflicts: Funder listed in the journal record: University of California. Conflicts not reported in the abstract.
Date: Oct 19, 2026