On the recent systematic food allergy review

All over the media last week were reports that the prevalence of food allergies is grossly overestimated.  That is indeed good news, let’s have a quick look at the systematic review that led to these statements (1).

The review focused on foods which account for over 50% of food allergies: cow’s milk, hen’s egg, peanut, tree nuts, fish, and shellfish.  The authors looked through 1,216 papers and through an undoubtedly arduous quality assessment, reduced the number to be included in the review to 182, 72 of which are related to the foods listed previously.

In short, there are several critical limitations that make conclusions and treatment of food allergies difficult.  First, there is no standard definition of a food allergy.  Not all studies provide a definition and not all include non IgE-mediated immune responses.  Studies also use different methods to diagnose a food allergy; in general there are: self-report, a skin prick test, food-specific IgE determination, or a double-blind, placebo-controlled food challenge, which is considered the gold standard.  However, because of resources and risk, and a lack of criteria for a positive test, it is not frequently used.

From data pooling, self-reporting of food allergies generally shows a greater incidence than the 3 methods listed above.  The authors note a need for standardizing tests for a clearer picture.  Data also show a recent increase in allergy prevalence, though it is difficult to determine if this reflects an increased awareness, or a true increase.  As stated by the authors, food allergies affect more than 1% or 2% but less than 10% of the US population.Clearly this is a messy area to interpret.

The authors conclude that as far as management techniques for food allergies, we still need more studies.  They reviewed areas including: elimination diets, immunotherapy, food substitutions/alterations, diets in breastfeeding women, medical or pharmacologic therapies, probiotics, education, prevention, breastfeeding, maternal diet during pregnancy or lactation, and special infant diets.

I am surprised at the lack of efficacy of many of these strategies.  Perhaps the most surprising is that the authors could only identify 1 randomized, controlled trial on the elimination diet, though it is apparently the therapy of choice.  Also, you may recall from media headlines some neat work lately on immunotherapy with peanuts.  However, this method is still being researched and cannot yet be medically used.  The authors note still unknown safety issues.

Finally, skin prick tests or serum IgE determination to foods risk overdiagnosis, especially in patients with nonspecific symptoms.  According to the authors, these patients with a positive test for either have less than a 50% chance of actually having a food allergy, which is concerning.  Subsequent, unnecessary nutritional changes may reduce diet quality.

I did not realize this topic of study was so convoluted.  This seems like an important paper to highlight the desperate need for more standardization of, well, everything.  As quoted from the NY Times, it is part of “a large project organized by the National Institute of Allergy and Infectious Diseases to try to impose order on the chaos of food allergy testing. An expert panel will provide guidelines defining food allergies and giving criteria to diagnose and manage patients. They hope to have a final draft by the end of June.

In short, even with over 1,000 papers on food allergies, we are still far from a proper understanding of prevalence and effective treatments.

Reference

1. Chafen, J., Newberry, S., Riedl, M., Bravata, D., Maglione, M., Suttorp, M., Sundaram, V., Paige, N., Towfigh, A., Hulley, B., & Shekelle, P. (2010). Diagnosing and Managing Common Food Allergies: A Systematic Review JAMA: The Journal of the American Medical Association, 303 (18), 1848-1856 DOI: 10.1001/jama.2010.582

  • http://members.authorsguild.net/fleckenstein/blog.htm Alexa Fleckenstein M.D.

    The main problem with that article was that it is pointing out that not every “allergy” is a real allergy. But undoubtedly, there are many people who have reactions to food which can't be classified as allergies – let's call them intolerances. And those intolerances are not even addressed in the article! Yet real people (patients) live with these really annoying intolerances – and they get nothing but a sneer that they problems don't count. Alexa Fleckenstein M.D., physician, author.

  • http://members.authorsguild.net/fleckenstein/blog.htm Alexa Fleckenstein M.D.

    The main problem with that article was that it is pointing out that not every “allergy” is a real allergy. But undoubtedly, there are many people who have reactions to food which can't be classified as allergies – let's call them intolerances. And those intolerances are not even addressed in the article! Yet real people (patients) live with these really annoying intolerances – and they get nothing but a sneer that they problems don't count. Alexa Fleckenstein M.D., physician, author.