Precisely when babies first eat cereals may affect their odds of subsequently developing diabetes. Two studies suggest that giving cereals to diabetes-susceptible infants within 3 months of birth greatly enhances their risk for type 1 diabetes. One of these studies also concludes that waiting 7 months or more before introducing cereal carries similar risk. Other researchers say the intriguing new leads don't yet warrant changes in baby-care guidelines.
For years, scientists have been investigating environmental factors that might contribute to autoimmune disorders having known genetic components. These include type 1 diabetes and celiac disease, an autoimmune response to the cereal protein gluten. Some research has linked diabetes risk to young infants' consuming foods or formulas that contain proteins from cow's milk (SN: 6/26/99, p. 404: http://www.sciencenews.org/sn_arc99/6_26_99/fob2.htm). Other studies have found no such association.
To explore whether the timing of food introduction contributes to the risk of developing type 1 diabetes or celiac disease, researchers followed 1,610 German babies who had a family history of type 1 diabetes. They tracked the children from birth to an average of 6.5 years of age, noting when they began to eat various foods and whether they developed certain antibodies that tend to precede the onset of each autoimmune disorder.
Infants who ate gluten before the age of 3 months were 5.2 times as likely to develop diabetes-related antibodies as the other infants were, Anette-G. Ziegler of Hospital München-Schwabing in Munich, Germany, and her colleagues report in the Oct. 1 Journal of the American Medical Association. Their data indicate that neither dairy products nor foods that lack gluten affect diabetes risk and that celiac disease isn't connected to specific foods in a child's early diet.
Meanwhile, Jill M. Norris of the University of Colorado Health Sciences Center in Denver and her colleagues focused on type 1 diabetes risk. They followed 1,183 children whose genes or family history placed them at high risk of developing the disease.
As did the European researchers, Norris and her colleagues found that diabetes-associated antibodies are unrelated to the start of dairy consumption. They also conclude that the antibodies appear more frequently in infants who consume cereals before 3 months of age than in those who first eat cereals between 3 and 7 months of age. The groups differed by a factor of 4.32. Risk for diabetes-linked antibodies didn't depend on whether the cereals contained gluten.
Moreover, Norris and her coworkers say, children who are at least 7 months old when they first ingest cereals are 5.36 times as likely to show diabetes-linked antibodies as are those who start cereals during the 3-to-7-month window. The European team didn't find this pattern in their own data.
The new findings should be regarded with "cautious interest," says Mark Atkinson of the University of Florida in Gainesville. Beyond the data on the health effects of adding cereal to the infant diet at different times, he says, both studies undermine the purported link between diabetes and milk's introduction. But an ongoing international effort to clarify the role of milk-based infant formulas in type 1 diabetes should continue, he says.
The lead researcher of that study, Mikael Knip of the University of Helsinki in Finland, says, "My view is that early introduction of complex foreign proteins [from cereals or from cow's milk] probably increases the risk." However, he points out that the strength of associations between specific dietary factors and type 1 diabetes may vary across populations.
If you have a comment on this article that you would like considered for publication in Science News, send it to firstname.lastname@example.org. Please include your name and location.
Mark A. Atkinson
Department of Pathology
University of Florida
College of Medicine
1600 SW Archer Road
Gainesville, FL 32610-0275
Department of Pediatrics
Hospital for Children and Adolescents
University of Helsinki
P.O. Box 281
Jill M. Norris
Department of Preventive Medicine and Biometrics
University of Colorado, Denver
Health Sciences Center
4200 East Ninth Avenue
Denver, CO 80262
Diabetes Research Institute
Koelner Platz 1