Oat milk and rice milk in glass carafes beside small piles of whole oat groats and raw rice grains

#025

62 Shakes Are Flagged for Gluten. None of Them Contain a Gluten Grain.

62 Shakes Are Flagged for Gluten. None of Them Contain a Gluten Grain.

62 Shakes Are Flagged for Gluten. None of Them Contain a Gluten Grain.

No wheat, barley, or rye appears anywhere in our 201 recipes, so we went looking for where 62 gluten flags actually come from.

No wheat, barley, or rye appears anywhere in our 201 recipes, so we went looking for where 62 gluten flags actually come from.

MOMS Test Kitchen

MOMS Test Kitchen

September 1, 2026

September 1, 2026

In this LAB note

Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS

Question

Why do 62 of our 201 production recipes carry a gluten allergen flag when not one of them contains wheat, barley, or rye?

Hypothesis

We assumed the flags were a data-entry artifact. A whole-food shake built from legumes, egg, produce, rice, and plant milk has no obvious route to gluten, so our working expectation was that searching the ingredient strings would surface a forgotten grain - a sprouted-wheat powder, a barley-malt sweetener - sitting quietly in about 60 recipes. If that search came back empty, we expected the flags would trace cleanly to one or two ingredients that are gluten-free as plants but not as commercial products.

Method

We searched all six ingredient slots - Liquid, Body, Fruit & Vegetable, Protein, Functional, Flavor - across all 201 production recipes for wheat, barley, rye, spelt, farro, and seitan. We then took the 62 recipes carrying a Gluten allergen flag and tried to attribute each flag to a specific ingredient, by asking which ingredients appear in flagged recipes and never in unflagged ones. Any flag we could not attribute was counted as a failure of the attribution, not explained away. Published context came from two contamination surveys. Koerner and colleagues (2011) measured gluten in 133 samples of the Canadian commercial oat supply. Thompson, Lee, and Grace (2010) tested 22 single-ingredient naturally gluten-free grains, seeds, and flours bought at retail in the United States, including rice products, using R5 ELISA.

Results

Zero of 201 recipes contain wheat, barley, rye, spelt, farro, or seitan. Not one. The 62 flags then split three ways. Forty-three recipes contain oats - 42 as Oat Milk in the liquid slot, one as Oatmeal in the body slot - and every one of those 43 is flagged. EY-W1-ENERGY-012-V1 (Blazing Dawn; Oat Milk 176.8ml, 397.5 kcal, 32.1g protein, 14.4g fiber) lists Gluten as its only allergen, and its full ingredient list is oat milk, produce, legumes, and egg white. EY-W2-DETOX-003-V1 (Clean Slate; Oat Milk 150ml + Coconut Water 195ml, 380.4 kcal, 31.8g protein, 12.5g fiber) is the same shape. That matches the literature: Koerner et al. found roughly 88% of 133 commercial oat samples exceeded 20 mg/kg gluten, with a dedicated gluten-free variety the only sample set that came back consistently negative. The oat plant is not the problem. The oat supply chain is. Another 14 flags land on recipes whose only plausible source is Rice Milk. MY-W3-GUT-004-V1 (Rooted Calm; Rice Milk 150ml, 396.5 kcal, 33.7g protein, 12.8g fiber) and MY-W3-GUT-006-V1 (Settled Mind; Rice Milk 156ml, 384.9 kcal, 33.9g protein, 8.3g fiber) both carry Gluten with no grain of any kind elsewhere in the recipe. Thompson, Lee, and Grace found 7 of 22 naturally gluten-free grain and flour products above 20 ppm, with results ranging up to 2,925 ppm. So 57 of 62 flags trace to two ingredients that are botanically gluten-free and only arrive that way if the processor keeps them separated from wheat.

What Failed

The attribution does not close, and the gap is larger than we would like. Rice Milk appears in 28 recipes and only 14 of them are flagged. The unflagged half are not different drinks: MY-W1-COOL-010-V1 (Cool Ease; Rice Milk 143ml, 382.2 kcal, 32.8g protein) is flagged for Tree Nuts and not for gluten, while MY-W3-GUT-004-V1 with Rice Milk 150ml is flagged for gluten. Same body type, same liquid, 7ml apart, opposite label. We could not find a rule that separates them. Five more flagged recipes contain neither oat nor rice milk at all. EI-W1-CLEANSE-001-V1 is coconut milk, pumpkin, carrot, pineapple, chickpeas, lentils, egg white, and maple syrup, and it carries Gluten with nothing in it that could plausibly supply gluten. The honest reading is that 43 of our 62 gluten flags are consistent and defensible and 19 are not. Our allergen column has been recording a judgement call rather than a rule, and until we ran this analysis nobody could have told you which of the two any given row was. That is a labeling defect, and it is ours.

What We Learned

Gluten in our menu is a property of processing, not of the plant. Where the story is legible, oat milk and rice milk are the whole story, and both are ingredients a shopper would read as gluten-free from the label alone. The second lesson is duller and more useful: the allergen column cannot be derived from the ingredient list. An ingredient's allergen status belongs to a specific supplier's specific production run, which is exactly the kind of fact that degrades quietly when it is maintained by hand, one recipe row at a time.

What Changed at MOMS

The gluten flag is no longer set per recipe. It is set once per sourced ingredient in the ingredient table and inherited by every recipe that uses it, so oat milk and rice milk carry their status in one place and all 201 recipes stay consistent by construction. The 19 rows we could not attribute were re-derived from that table rather than preserved as written. Alongside it, an oat or rice ingredient can only be marked gluten-free if the supplier provides a certificate of analysis showing under 20 ppm - we no longer reason from the plant.

References

Koerner TB, Cleroux C, Poirier C, Cantin I, Alimkulov A, Elamparo H. Gluten contamination in the Canadian commercial oat supply. Food Additives & Contaminants: Part A, 28(6):705-710, 2011. | Thompson T, Lee AR, Grace T. Gluten contamination of grains, seeds, and flours in the United States: a pilot study. Journal of the American Dietetic Association, 110(6):937-940, 2010.

Keep reading

Not sure which shake fits you? We match the formulation to your body type.

Find your body type