New solid studies: the advice on gluten for infants needs to be changed!

Babies eating a roll

Future Gluten Intolerants?

This is a question that I frequently get and that many parents of infants struggle with: Is it important for infants to eat gluten, ie bread and hot cereal, early in life?

Even today the official guidelines encourage parents to introduce foods with wheat early to reduce the risk of gluten intolerance. This is what the Swedish guidelines for infants include:

If the infant is given small amounts of gluten while still nursing, the risk that the child will be gluten intolerant is reduced. At no later than six months, and no earlier than four months, you should start giving the infant some gluten-containing foods… For example, you can let the infant have a bite of white bread or crackers or a small spoon of hot cereal or wheat-based formula a couple of times a week… After six months gradually increase the amount. 

This assertive advice is unfortunately based only on uncertain statistics from questionnaire studies, i.e. observational studies. Such statistics prove nothing. The guideline-issuing authorities have a troublesome ability to sound certain without enough supporting evidence.

So is the advice above good or bad? Nobody knew before, but now this has finally been tested seriously.

The other week two critical studies were published in the New England Journal of Medicine – the world’s most respected medical science journal. For the first time studies were designed to test whether the advice works.

Study 1

In the first study, 944 children with an elevated risk for celiac disease were randomized to receive either gluten or placebo between 4 and 6 months of age. They really tested in the best way  whether early gluten exposure prevents gluten intolerance.

The result? It did not help. The children given gluten had just as high a frequency of gluten intolerance, and there was even a non-significant trend towards their showing gluten intolerance more often!

A whopping 5.9 % of the children introduced to gluten early became gluten intolerant, as compared to 4.5 % of those who could avoid gluten early on.

This first test now completely refutes the advice to give gluten early on, it doesn’t work. But it gets worse for the official recommendations…

Study 2

In the second study 832 children with celiac disease in the family were randomized to receive advice on gluten from either 6 months or 12 months of age.

The result? Clear cut. At two years of age 12 % of the children given advice on gluten from 6 months of age were gluten intolerant. Only 5 percent of those who didn’t introduce gluten until 12 months of age developed gluten intolerance – less than half, a difference that was highly statistically significant.

Advice on postponing gluten at least until 12 months of age more than halved the risk of becoming gluten intolerant!

The Journal’s Comment

In a comment to the studies the journal says that it will now be difficult to advise introducing gluten early on – it just seems to be wrong. The advice to introduce gluten while still nursing also seems to be incorrect – we’re not seeing any benefit to this either.

Background: early gluten exposure produced an epidemic of gluten intolerance.

There’s a sad back drop to the story. In the 80’s, Sweden started to give advice that resulted in larger amounts of gluten in infant formula. The result was a disastrous increase in gluten intolerance and that’s when they backed off to today’s advice.

However, the new studies show that they haven’t backed enough yet.

Conclusion: More gluten, more gluten intolerance

The results from these first studies, when gluten introduction was clear: it doesn’t help. Instead, all evidence indicates that the official guidelines instead hurt babies. Parents who follow this advice to introduce gluten early on run a higher risk of getting gluten intolerant children! The advice will likely have to be changed.

When and how should parents introduce gluten to their children in order to avoid gluten intolerance? Judging from the best science we have so far, the simple answer may be this:

The less gluten the better and the later gluten is introduced the better.


Gluten Makes a Growing Number of Swedes Sick

South Park Runs a Gluten-Free Episode!

The Studies

Membership only (abstract free):

NEJM: Randomized Feeding Intervention in Infants at High Risk for Celiac Disease

NEJM: Introduction of Gluten, HLA Status, and the Risk of Celiac Disease in Children

Editorial: NEJM: The Missing Environmental Factor in Celiac Disease


  1. FrankG
    It seems counterintuitive that we woud even need to introduce infants to some foods in a certain way, in an attempt to reduce intolerance.... surely if infants display an intolerance to a food, then the common-sense poistion would be that they should not be eating it?

    Is it preordained somewhere that all humans WILL eat gluten... or die trying!?!

    Replies: #5, #7
  2. Vala
    What about sprouted bread like ezekiel bread? It does contain gluten but no flour. I know for myself that when I eat flour based gluten products I feel an inflammation in my gut, but eating sprouted bread is fine. I know they are processed very differently so I can't help to wonder if some gluten sensitivity comes down to the way we process flour these day.
    Reply: #15
  3. Yufina
    How about this:

    "In order to digest grains, your body needs to make use of an enzyme called amylase. Amylase is the enzyme responsible for splitting starches. And, guess what? Babies don’t make amylase in large enough quantities to digest grains until after they are a year old at the earliest. Sometimes it can take up to two years. You see, newborns don’t produce amylase at all. Salivary amylase makes a small appearance at about 6 months old, but pancreatic amylase (what you need to actually digest grains) is not produced until molar teeth are fully developed! First molars usually don’t show up until 13-19 months old, on average.

    Undigested grains wreak havoc on your baby’s intestinal lining. It can throw off the balance of bacteria in their gut and lead to lots of complications as they age including: food allergies, behavioral problems, mood issues, and more."

  4. robert
    "surely if infants display an intolerance to a food, then the common-sense poistion would be that they should not be eating it?"

    Too simple, too obvious.

    And lets not forget all the money that would be lost. No, not good for business at all. Aren't babies entitled to reaping the "benefits" of having those healthy whole-grains as early as possible? Future loyal customers of big food and big pharma.

  5. murray
    Frank, there is a significant "if" in your comment. I believe the prior thinking was that the "if" is not satisfied (that is, intolerance to gluten is not the default state) and that just as babies train their immune system from exposure to proteins in the womb mediated by the mother, so the infant immune system better recognizes friend from foe through early exposure in the presumed resilient training phase.

    What is overlooked in that view, in my opinion, is that early introduction of some proteins might be an overdose that creates an intolerance that would otherwise not develop. An alternative view, therefore, is that an older infant has higher innate tolerance to gluten and, so long as gluten is kept modest within tolerance appropriate to stage of development, intolerance may not develop.

    Analogies and epidemiology are great for generating plausible hypotheses, but plausibility does not imply fact. The too-early-exposure-at-too-great-a-dose hypothesis was equally plausible, and this new research indicates it is more likely fact than the train-earlier hypothesis. The remaining issue is when and how much is too soon and too much for each individual's stage of development? Too uncertain for me. Our family is gluten free.

    Reply: #8
  6. FrankG
    Agreed... it makes sense that infants have ages when different foods may be more appropriately and safely introduced.

    On the other hand, I daresay we can build a "tolerance" to many things... even arsenic.

    Perhaps we are in this mess today because for some reason we stopped listening to the traditional keepers of this knowledge: refined over thousands of years by trial and error, then passed down from Grandmother, to Mother, to Daughter (not meaning to be sexist but rather acknowledging the vital importance of Mothers). We stopped listening to our Grandmothers and started listening to "nutritionists".

    That compounded by facts like the gluten of today's "staff of life" bearing little relationship to the wheat proteins of old; in both quantity and quality.

    Reply: #9
  7. Galina L.
    Not all traditions are healthy, many cultures basically worship bread. In Russia cream of wheat is a traditional baby food.
    Reply: #10
  8. FrankG
    "In Russia cream of wheat is a traditional baby food." Going back how many generations? And what kind of wheat was traditionally used?
  9. Galina L.
    I don't know for how long that tradition lasts, before modern times it was more wide-spread in an affluent part of a society, peasants fed their children pre-chewed rye bread wrapped into a cloth (it would solve the amylase deficiency issue). My mom (77 yo) told me that her mom didn't have enough milk and grandma supplementaly fed her rye bread soaked in a soup made with beef broth and sauerkraut. As far as I know the most-spread variety of a wheat in Russia is low in gluten.
  10. Cindy C
  11. Anne
    Gluten proteins do get into the breast milk. I know a number of people who found that their child's colic disappeared when mom stopped eating gluten. Dr. Rodney Ford, a pediatritian/allergest/GI said he has seen the same thing. Perhaps nursing mothers should be told to eliminate gluten from their diet. I sure wish I had known that - my 2nd child had terrible colic and as an adult is very gluten intolerant.
  12. murray
    Galina, I am reading a new book, The Amazons: Lives and Legends of Warrior Women across the Ancient World, by Professor Adrienne Mayor. Very good, so far. She notes that among the various Scythian nomadic horse peoples, who ranged from north of Greece, around the Black Sea, across Ukraine and southern Russia to Eastern China, the cavalry warrior women in the tribes (who apparently did not nurse) would feed their babies fermented mare's milk (kumis) or bone marrow. The Scythians, of course, were predominantly consumers of meat and not grass seeds, such as wheat.
    Reply: #16
  13. murray
    "Dietary gluten and the development of type 1 diabetes"

  14. Galina L.
    Stephan Guyenet blogged about the adaptation to eating grains among Europeans I found Interesting graphs there.
    I look like a mix between European and Asian person, so I do think I have genes from tribal people who made living herding their cattle. I tolerate wheat and starches worse than other people I know.
  15. Aretaeus Wellness
    It seems everyone is forgetting the fact that babies are first introduced to Gluten at conception. This might explain why it is estimated that between 10-35% of people with Down Syndrome have Celiac Disease.

Leave a reply

Reply to comment #0 by

Older posts