By Dr. Sarah Taylor-Black
Many people seem to be aware that the rates of food allergy are increasing in the U.S. For example, the rate of peanut allergy increased from 0.6% in children in 1997, to 1.2% in 2002, to 2.1% in 2008 [1]. The most recent estimate of overall food allergy in the U.S. seems to be 8.0% [2]. But you may be wondering, “How did they get that number? How do we really know how many kids have food allergy?”
That 8% number comes from a study done at the Northwestern University Feinberg School of Medicine, who sent an electronic survey out to a sample of households that had access to the internet, and received 40,104 answers. This is a truly impressive number of responses for a research study. Of those 40,104 children, it was reported that 8% had a convincing or confirmed food allergy [2]. This is a very good study, looking at large numbers of patients from a population that was more or less representative of a sample of US households (at least US households with internet access). Many studies on food allergy prevalence are based on self-reporting, such as this one. The problem, of course, is how do you really know that the child had a food allergy? Sure, somebody in the household is saying that the child is allergic, but how can you be sure all those children really are allergic? After all, lots of people attribute to allergies conditions that aren’t allergic in nature. Or to foods that don’t activate the immune system.
One of the other ways that researchers look at rates of food allergies is by looking in the blood. These types of studies have led to headlines such as this one from WebMD, “African-American Kids May Have More Food Allergies” which Dr. Ehrlich referred to in his January 5th post. For example, a recent article in Pediatrics reported higher rates of food sensitization in black children [3]. However, food allergy sensitization is not food allergy. A positive blood test does not always mean that the person actually has a problem with the food if they eat it. Again, we are left with guessing if those children have food allergy, really. This is why doctors should not broadly check allergy blood tests for foods that have never presented a problem for the patient, which can lead to many false positive results as well as unnecessary anxiety and food avoidance.
Food allergy is a complicated diagnosis to make, and it takes a careful history of past reactions, as well as skin testing and blood testing to confirm it. And, even with these three things, the diagnosis is not always certain, in which case a food challenge can be done. A food challenge, when the doctor carefully and slowly feeds the food to the child to see if there is a reaction, is considered the gold standard test in food allergy. (Component testing may provide a middle ground or even a game-changing alternative, but until it is more generally available, we’re stuck with the old methods.)
So why is it so hard to figure out how many kids have food allergy, really? The reason is that to know if a child truly is allergic, you have to have an allergist experienced with food allergies examine and test the children, or look at least look at the allergy medical records, if available, for everyone. Food allergy research centers can’t do it even for a population living nearby. We have to set priorities. For most food allergy studies, a patient must come into the hospital for an exam by the doctor, get testing done, as well as be fed the allergenic food (this may require the patient to have an IV placed). This can take anywhere from 3 to 6 hours per food, if not longer, depending on the outcome. Just think of the 40,104 children from the internet-based study. Over 3,200 of them reported a food allergy. To examine, test and review the medical records of all of those children would require the time of at least one doctor, a team of nurses, lab work and MONEY. Of course, funding is usually the criterion to ultimately decide which studies get done. When you think about all the resources required, it makes sense why methods used to study the rates of food allergy usually involve telephone or internet-based surveys, or blood samples from national studies, such as NHANES (the National Health and Nutrition Examination Survey).
This is research on an industrial scale, and like any industrial-scale effort, you need to find economies, including the time-honored technique of statistical analysis. Unfortunately, big numbers don’t help individual patients, which is why we need clinical allergists as well as academicians. Whatever the true extent of food allergies, whether it’s closer to 3 million as some studies show, or 6 million as others do, there’s no doubt that this represents a massive public health problem, really.
Dr. Sarah A. Taylor-Black was educated at Dartmouth College and Brown Medical School and trained in Internal Medicine at Columbia Presbyterian Medical Center. She board certified in Internal Medicine. She is currently an allergy and immunology fellow at Mt. Sinai Medical Center. She is fellow in-training member of the American Academy of Allergy, Asthma & Immunology, and the American College of Allergy, Asthma & Immunology.
1. Sicherer S, Munoz-Furlong A, Godbold J, and Sampson H. US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up. J Allergy Clin Immunol 2010;125:1322-1326.
2. Gupta RS, Springston EE, et al. The Prevalence, Severity, and Distribution of Childhood Food Allergy in the United States. Pediatrics. 2011;128:e9-e17.
3. Kumar R, Hui-Ju T, Hong X, et al. Race, Ancestry, and Development of Food-Allergen Sensitization in Early Childhood. Pediatrics 2011;128(4):e821-29.
Photo: fullerton.edu