A lot of people think they have food allergies, but they're likely wrong.
That's partly because it's easy to confuse common food-related problems like lactose intolerance or celiac disease with an allergy. But it's also because there are a lot of tests promoted for food allergies that don't measure up.
That includes a blood test called IgG analysis, which is often promoted on the Internet as a way to identify problem foods. It sounds almost exactly like an IgE test, which is used by allergists to measures the immune system's response. (A true allergy is when the immune system reacts to something that's not a real threat.)
An article in the current Canadian Medical Association Journal reminds doctors that IgG tests aren't reliable tests of food allergies.
It follows the guidelines on diagnosing food allergies from the National Institute of Allergy and Infectious Disease in 2010, which lumped IgG testing with more than a dozen "nonstandardized and unproven procedures" that should be avoided. A positive IgG test, the expert panel said, could cause people to avoid foods that they can safely eat.
That got us here at The Salt wondering what does work.
To find out, we called up Scott Sicherer, a professor of pediatrics at the Jaffe Food Allergy Institute at the Mount Sinai School of Medicine.
The NAIAD guidelines say that a "food challenge test" is the gold standard for food allergies, and Sicherer agrees. To do one, he feeds patients small amounts of a suspect food over the course of an hour, in his office (for safety reasons). "Ultimately the proof is in the pudding. If someone is able to eat a food and not get sick from it, they don't have a problem with it," he says.
He does use the IgE test at times; it measures immunoglobulin E, a substance that's part of the body's allergic response. But it's not very specific. For example, about 8 percent of people have a positive IgE test to peanuts, Sicherer says, but most of them can safely eat peanuts. Skin prick tests are also used for food allergies, but also less than perfect.
"The most important test of all is - can you guess what it is?" Sicherer asks. "The most important test of all is the medical history."
That's the medical version of a detective's case work: asking a patient what she ate and when she ate it. Doctor and patients then match up symptoms and food to figure out if food is causing the problem.
"If someone comes in and says 'I'm gaining a lot of weight, do you think it's a food allergy? I'd say 'Gee, gaining weight is not a symptom of a food allergy.'"
Confusion over what constitutes a food allergy is common among physicians, according to a review article that was published in 2010 in the Journal of the American Medical Association. So it's no wonder that we ordinary eaters get confused, too.
Even the numbers on food allergy are all over the map. The JAMA article estimated that about 1 to 2 percent of Americans have food allergies. A survey released last year found that 8 percent of children have some sort of food allergy, and that 39 percent of those allergies can cause severe reactions. But the NIAID says that 5 percent of children and 4 percent of adults are more reasonable.