Introducing gluten to a baby when celiac runs in the family

Short answer

The 2024 update from the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) says introducing gluten at any time from completed 4 months to 12 months does not change a child’s cumulative incidence of celiac disease, though earlier introduction may lead to earlier seroconversion and celiac disease27, 29. The 2016 paper it updates was derived from families with a first-degree relative with celiac disease26; ESPGHAN calls any recommendation on the amount of gluten unjustifiable for now28, and the NIDDK says blood relatives of people with celiac disease should talk with their doctor about being tested4.

Published by Gluten Verdict · updated · 14 sources, each linked below

Reviewed by Mark Peter, Registered dietitian (RD), on

A high chair tray with puffs, a spoon and a bib on a sunlit kitchen floor

Key points

  • PreventCD gave 16- to 24-week-old babies with a first-degree relative gluten or placebo and found similar celiac disease rates at 3 years8, 9; CELIPREV compared first gluten at 6 and 12 months, and the differences seen at 2 years were no longer significant at 5 years12, 13.
  • ESPGHAN’s 2024 update says breastfeeding does not reduce the risk and that earlier introduction may lead to earlier seroconversion and celiac disease27.
  • How much gluten, and which gluten foods first, are open questions: ESPGHAN calls any recommendation on amount unjustifiable for now28.
  • Two pooled reviews report different results: a 2016 meta-analysis found 25% higher risk with late (after 6 months) versus recommended (4 to 6 months) introduction18, while a 2023 update found age at gluten introduction was not associated with cumulative risk22.
  • The NIDDK says doctors in the US do not routinely screen for celiac disease and that blood relatives should talk with their doctor about being tested4; it also says doctors do not recommend starting a gluten-free diet before diagnostic testing5.

What a family history means for a baby’s risk

Gluten is a protein found naturally in wheat, barley and rye7. The NIDDK says you are more likely to develop celiac disease if someone in your family has it1. It also says about 30 percent of people carry the gene variants DQ2 or DQ8 and only about 3 percent of them develop the disease2.

One source of numbers for children with an affected first-degree relative is PreventCD, a birth cohort of 944 genetically predisposed children with 10 years of follow-up30. Summarizing the 2022 analysis, the Celiac Disease Foundation says the risk had been considered 5-10%, and that the study found 7.5% at age three, 16.6% at age eight and 17.5% at age 1033.

In that analysis, 135 of 944 children developed celiac disease by a median follow-up of 8.3 years, more often in girls and in children with two copies of HLA-DQ2 (8-year cumulative incidence 35.4%, against at most 18.2% in the other HLA groups)31. Coeliac Australia adds that most of the risk for coeliac disease relates to the genes the child is born with42.

What the two randomized trials and the TEDDY cohort found

Three studies on early gluten in children at higher genetic or family risk
StudyWho was in itWhat was comparedWhat it found
PreventCD (N Engl J Med, 2014)944 children positive for HLA-DQ2 or HLA-DQ8 with at least one first-degree relative with celiac disease8100 mg of immunologically active gluten daily from 16 to 24 weeks of age, or placebo8Celiac disease in 5.2% of children by age 3: 5.9% in the gluten group and 4.5% in the placebo group (hazard ratio 1.23; 95% CI 0.79 to 1.91)9. The authors concluded the small quantities did not reduce the risk by age 310
CELIPREV (N Engl J Med, 2014)832 newborns with a first-degree relative with celiac disease12First gluten at 6 months or at 12 months12At 2 years the 6-month group had more celiac disease autoimmunity (16% vs 7%) and overt celiac disease (12% vs 5%); at 5 years the differences were no longer significant (autoimmunity 21% vs 20%, overt disease 16% and 16%)13
TEDDY (JAMA, 2019)Children carrying HLA genotypes linked to type 1 diabetes and celiac disease, followed in Finland, Germany, Sweden and the United States; an observational cohort15Gluten intake estimated from 3-day food records from 6 months to age 516Higher gluten intake in the first 5 years was associated with increased risk of celiac disease autoimmunity and celiac disease17

The trial authors’ own conclusions are narrow. CELIPREV’s: neither delayed introduction nor breast-feeding modified the risk among at-risk infants, though later introduction was associated with a delayed onset of disease, and a high-risk HLA genotype was an important predictor14. PreventCD reported that breast-feeding, exclusive or not, did not significantly influence the outcome either11.

TEDDY looked at how much gluten children ate, not when the first gluten was given. It shows an association in an observational cohort, not the result of assigning babies to amounts15. The NIDDK describes related research as suggesting that children with a genetic predisposition who eat more gluten in early childhood may have a greater risk3.

How two pooled reviews compare

Two systematic reviews of early feeding and celiac disease
ReviewWhat it pooledOn timing of first glutenLimits and open points the authors state
Journal of Pediatrics, 2016 (epub 2015)1,982 studies identified, 15 eligible; the meta-analysis covered 2 randomized trials, 10 cohort studies and 1 case-control study18A 25% increase in celiac disease risk with late (after 6 months) versus recommended (4 to 6 months) introduction (risk ratio 1.25; 95% CI 1.08 to 1.45)18. The authors found no evidence that early introduction raises risk19More studies are needed that control for potential confounders and that evaluate environmental factors in low-risk families19
Alimentary Pharmacology & Therapeutics, 2023 (2022 update)36 publications (30 studies)20Age at gluten introduction was not associated with cumulative risk; two randomized trials suggested earlier introduction was associated with earlier appearance of celiac autoimmunity, meaning positive antibody tests22, 21Insufficient evidence on the amount of gluten tied to higher risk; results inconsistent on whether feeding practices change the risk that different HLA genotypes carry23

The 2016 review pooled two trials with eleven observational studies and flagged confounding as a limit18, 19. The 2023 update concluded that, for people at genetic risk, breastfeeding and age at gluten introduction have no effect on cumulative incidence during childhood, with some evidence that the amount eaten may matter24. ESPGHAN’s 2024 position paper, published after both, takes the same line on timing27.

What ESPGHAN and two celiac charities say

Published positions on first gluten for babies with celiac in the family
SourceWhat it saysWhat it leaves open
ESPGHAN position paper, 2016Gluten may be introduced anytime between 4 and 12 completed months; in high-risk children earlier introduction is associated with earlier celiac disease autoimmunity and celiac disease, but the cumulative incidence in later childhood is similar25. The advice applies to all infants, though derived from families with a first-degree relative with celiac disease26The optimal amount of gluten to introduce at weaning has not been established43
ESPGHAN update, 2024Statements essentially similar to 2016; any time between completed 4 and 12 months does not affect cumulative incidence; breastfeeding does not reduce risk27Any recommendation on the gluten amount is currently unjustifiable; no specific guidance on which gluten-containing foods to introduce28
Coeliac UK (charity)Gluten can be introduced from six months old, with no benefits for delaying it; once a baby is established on solid foods, gluten should be eaten regularly36, 37The Introducing gluten passage of its page gives no amount44.
Coeliac Australia (national society)Current evidence suggests introduction should not be delayed, even for infants at higher risk such as when one or both parents have coeliac disease; introduce as for any other child39. It adds there is no evidence that avoiding gluten altogether helps40It says definitive recommendations are difficult38

The wording differs: ESPGHAN gives a window, Coeliac UK a start at six months25, 36.

Which questions the sources treat as settled, and which stay open

Settled and open, as these sources read
QuestionWhere the sources standStatus
Does breastfeeding lower the risk?ESPGHAN 2024: no, at any amount or duration27. The 2023 update agrees for children at genetic risk24; PreventCD and CELIPREV reported no significant effect11, 14Consistent across the sources read
Does the month of first gluten, within 4 to 12 months, change the cumulative incidence?ESPGHAN 2024 says no for any time from completed 4 to 12 months27; the 2023 update found age at introduction not associated with cumulative risk22. Earlier introduction may lead to earlier seroconversion and celiac disease27Consistent on cumulative incidence; earlier-seroconversion point noted
Is late introduction (after 6 months) linked to higher risk?The 2016 meta-analysis found a risk ratio of 1.2518; the 2023 update found no association with age at introduction22The two reviews report different results
Should gluten be avoided altogether?Coeliac Australia: no evidence that it helps, and not recommended40One charity’s position
Does the amount of gluten matter?Observational cohorts link a higher amount to higher risk (TEDDY17; ESPGHAN 2024’s summary of observational studies45), but the 2023 update finds insufficient evidence to determine the amount23 and ESPGHAN calls any amount recommendation currently unjustifiable28Open
Which gluten foods first?No specific guidance28Open
Do feeding practices change the risk that HLA genotypes carry?Results were inconsistent23Open

When testing is discussed, and what the sources say about gluten before testing

The NIDDK says doctors in the United States do not routinely screen people for celiac disease, and that blood relatives of people with celiac disease should talk with their doctor about their chances to see if they should be tested4.

On when to screen, the sources are open. The authors of a PreventCD prediction analysis wrote that screening is recommended for children with affected first-degree relatives but that how often, and from what age, was unknown30; the Celiac Disease Foundation likewise says it is unclear when to screen these children and how frequently35. They concluded that these children should be screened early in life, including HLA-DQ2/8 typing, and, if genetically predisposed, get further personalized screening advice from their Prediction application32. The Celiac Disease Foundation describes that model as using age, gender, HLA-DQ genes and the number of affected first-degree relatives34. That is one research group’s conclusion, not a rule.

A gene result is not a diagnosis. The NIDDK says having DQ2 or DQ8 alone does not mean you have celiac disease, and most people with these gene variants do not develop it6. Our guides on celiac genetic testing and on the celiac blood test cover the two kinds of test.

Both the NIDDK and Coeliac UK tie testing to eating gluten. The NIDDK says doctors do not recommend starting a gluten-free diet before diagnostic testing because it can affect test results5, and Coeliac UK notes that coeliac disease can only be diagnosed once gluten is in the diet37.

On symptoms, Coeliac Australia says that if symptoms suggestive of coeliac disease develop, such as diarrhoea, testing should be discussed with the child’s doctor41. Our guide to celiac symptoms in children covers the symptoms the sources describe.

This page reports what the sources say. It does not tell you whether or when to test your baby; that is a conversation for the child’s doctor.

Questions to bring to the child’s pediatric gastroenterologist

  • Which relatives in our family count as first-degree for this baby, and how was each diagnosed?
  • Do you recommend HLA typing for my baby, and what would a result change?
  • If you recommend blood tests, at what ages and how often?
  • Do you follow ESPGHAN’s 2024 position paper, or another society’s guidance, on the timing of first gluten?
  • Which symptoms should lead us to call before the next planned test?
  • What should we do about gluten in the diet while testing is still planned?

Verdicts in this guide

Questions people ask

Should a baby with celiac in the family avoid gluten?

Coeliac Australia says there is no evidence that avoiding gluten altogether is helpful, and that the approach is not recommended40.

Does breastfeeding protect against celiac disease?

ESPGHAN’s 2024 update says breastfeeding, at any amount, exclusive or of any duration, does not reduce the risk of developing celiac disease27.

Is the risk for a child with a first-degree relative with celiac disease 5 to 10 percent?

The Celiac Disease Foundation says that was the figure until recently, and that PreventCD’s 10-year data found 7.5% at age three, 16.6% at age eight and 17.5% at age 10 for children with an affected first-degree relative33.

Do doctors in the US routinely screen people for celiac disease?

The NIDDK says doctors in the United States do not routinely screen people for celiac disease; blood relatives should talk with their doctor to see if they should be tested4.

Will my baby’s genes be known before solids start?

ESPGHAN’s 2016 position paper notes that genetic risk alleles are generally not known in an infant at the time of solid food introduction, so its recommendations apply to all infants26.

More in Women, pregnancy and babies

Sources