"Too much wheat and gluten in early stages of infancy raises risk of coeliac disease in children at risk of the condition," reports the Mail Online.
Researchers looked at the diets of 6,605 children from Sweden, Finland, Germany and the US, all of whom had genetic variants that put them at higher risk of developing autoimmune conditions like coeliac disease, where the immune system starts to attack the body's own tissues.
Gluten is a protein found in grains including wheat, rye and barley. Despite claims to the contrary, it's an essential component of a healthy diet for most people.
In people with coeliac disease, gluten triggers the immune system to produce antibodies that attack the lining of the gut wall, meaning they cannot absorb nutrients.
Coeliac disease is not a food intolerance, it's an autoimmune condition (where the immune system mistakenly attackes healthy tissue). The only treatment is a lifelong gluten-free diet.
Between 1 in 5 and 1 in 10 of this higher risk sample developed coeliac disease, a higher rate than the 1 in 100 population average.
Researchers looked at children from birth to the age of 5, as coeliac disease often starts in early childhood.
They found children who ate more than the average amount of gluten were a little more likely to get the disease.
But this does not mean gluten definitely caused their disease. The type of study means we cannot tell that and there also may be other factors, aside from gluten, that contribute towards their condition.
For example, some children could have been on low- or no-gluten diets as a precaution, but would have otherwise developed coeliac disease.
The study paves the way for more research into the best diet for children with a genetic vulnerability to coeliac disease.
The study was carried out by researchers from 13 universities, hospitals and healthcare institutions in Sweden, the US, Germany and Finland.
It was funded by the US National Institutes of Health, the Centers for Disease Control and Prevention, and JDRF, a US-based diabetes charity previously known as the Juvenile Diabetes Research Foundation.
It was published in the peer-reviewed Journal of the American Medical Association (JAMA).
The Mail Online carried a balanced and accurate report of the study. The website made it clear that the study involved children known to have a higher risk of developing an autoimmune condition, so they were not representative of children in general.
This was a cohort study.
Cohort studies are good ways to spot patterns between risk factors (such as gluten in the diet) and outcomes (such as coeliac disease), but cannot tell us for sure that 1 causes the other.
Other factors may be involved.
The Environmental Determinants of Diabetes in the Young (TEDDY) study was set up to look at the link between environmental and genetic risk factors for coeliac disease and type 1 diabetes.
Like other autoimmune conditions, both diseases are associated with certain variants of histocompatibility complex (HLA) genes.
These are a group of genes that carry instructions for making immune cells. Variants in these instructions can trigger autoimmune conditions.
More than 8,000 children with HLA gene types linked to coeliac disease and type 1 diabetes were recruited at birth from 6 clinical centres in Finland, Germany, Sweden and the US.
The researchers asked parents to record their children's diets over 3 days at intervals, when the children were aged 6, 9, 12, 18, 24, 30 and 36 months.
From the food records, researchers calculated the amount of gluten children ate per day.
They also looked at how much gluten they ate as a proportion of their overall diet and compared with their body weight.
The researchers looked at the development of coeliac in 2 ways.
They first followed up children with annual blood tests until they were 5 years old to look for antibodies that attack the gut lining (tissue transglutaminase autoantibodies), which suggests they're having an adverse reaction to gluten.
Once a child had developed these antibodies (on 2 consecutive samples), they were said to have coeliac disease autoimmunity, but did not yet have a diagnosis.
A diagnosis of coeliac disease requires a gut biopsy showing inflammation or, in this study, 2 blood tests showing high levels of antibodies.
The researchers then compared children's chances of developing coeliac disease autoimmunity (antibodies) or diagnosed coeliac disease according to their gluten intake.
Researchers took account of confounding factors, including the child's country of residence, sex, type of genetic variant, overall calorie intake and family history of coeliac disease.
Children who ate more gluten than the average child in the study were more likely to develop coeliac autoimmunity or coeliac disease.
Out of the 6,605 children in the analysis, 1,216 (18%) developed antibodies. Coeliac disease was diagnosed in 447 children (7%).
Most developed antibodies or coeliac disease between the ages of 2 and 3.
The researchers calculated:
Similar results were found for coeliac disease diagnosis:
The researchers said: "Higher gluten intake during the first 5 years of life was associated with increased risk of coeliac disease autoimmunity and coeliac disease among genetically predisposed children."
They said a trial of different amounts of gluten in early childhood among genetically at-risk children "would be warranted to confirm our findings".
This study adds to our knowledge about how coeliac disease may develop in children with genetic variants linked to the disease, and how diet in early childhood appears to be linked with that.
But it does not tell us what young children should be eating. Most people do not have genetic variants linked to coeliac disease, so the results do not affect them.
For those who do, this observational study still does not allow us to be sure that the amount of gluten eaten is what caused the disease.
We do not know enough about the diets of the young children in this study, and this will have been primarily guided by their parents or carers.
Knowing that their child was at higher risk of developing coeliac disease may have influenced the foods that some gave.
This means the "average" intake of gluten in this sample may have been much lower than the general population average.
Children who were assessed to be at higher risk because they were eating more than this may not have been eating excessively high amounts of gluten at all.
They may have just have been eating more like the typical amount most children eat.
Meanwhile, some children in this study who did not develop coeliac disease may have been given little or no gluten by their parents, but would have developed the disease had they been exposed to more.
The study has other limitations. Because diet was reported by the children's parents, it may not be completely accurate.
Also, the amount of gluten in food like sauces and cakes had to be estimated, so may not be accurate either.
There's been a lot of media attention around people claiming to have a gluten intolerance and sensitivity, but this is not the same thing as having an autoimmune condition.
Coeliac disease is actually fairly rare, affecting only around 1 in 100 people in the UK population.
Gluten can form a normal part of a healthy diet for most people.