select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to study looking at early introduction of gluten and coeliac disease indicators in children

A study, published in JAMA Pediatrics, looked at the early introduction of gluten and coeliac disease indicators in children.


Prof Alastair Sutcliffe, Professor of General Paediatrics, UCL, said:

Does the press release accurately reflect the science?

“Yes but it understates the fact that this was a secondary analysis from the internationally recognised of studies called EAT.  In other words this finding could be random as the study was not powered to detect this result.

Is this good quality research?  Are the conclusions backed up by solid data?

“This is good quality research by a bona fide group of researchers.

How does this work fit with the existing evidence?

“There is little evidence around the topic and thus this study is both novel and welcomed.

Have the authors accounted for confounders?


What are the implications in the real world?

“A further fully powered a priori primary hypothesis study needs to be done before we can say feed excess gluten early to a group of children at risk.

Is there any overspeculation?



Dr Baptiste Leurent, Assistant Professor in Medical Statistics, London School of Hygiene & Tropical Medicine, said:

“This is a secondary analysis of a randomised trial which looked at the early introduction of allergenic food (before six months).  In this sub-study, the authors compared the prevalence of coeliac disease in the two groups, and found that seven (1.4%) of children in the “standard feeding” group had developed coeliac disease by 3 years, and none in the “early introduction” group.

“This was a well conducted study, and because the data come from a randomised trial, the two groups of children can be considered identical, and any difference seen is likely due to the early introduction of allergenic food.

“However, this comparison is based on very small numbers, and could have happened by chance alone.  Typically, a result that has more than 5% chance to have happened by chance alone is not considered sufficient evidence – here we are just below this threshold, at 2%.  In other words, even if food introduction had no effect, there was around a 2% chance for these seven children to end up in the same group.

“Also, the trial was not designed to answer this question, making the finding less reliable.  For example, although less likely, the difference could have been due to other allergenic ingredients introduced early (milk, peanut, etc.).

“Other studies have found an effect toward the other direction, and is it is therefore too early to suggest any change in the baby’s diet to the public.  However, it is an interesting finding, which definitely needs further exploration, as the study authors acknowledge.”


‘Early Gluten Introduction and Celiac Disease in the EAT Study: A Prespecified Analysis of the EAT Randomized Clinical Trial’ by Kirsty Logan et al. was published in JAMA Pediatrics at 16:00 UK time on Monday 28 September 2020.

DOI: 10.1001/jamapediatrics.2020.2893


Declared interests

Dr Baptiste Leurent: “No conflict of interest.”

None others received.

in this section

filter RoundUps by year

search by tag