How common are food allergies? What's the best way to make a diagnosis? And how best should people manage the illness? In short, doctors don't know.

A review of 72 food allergy studies between 1988 and 2009 found that research is lacking on all of these questions, confusing doctors and the patients they treat.

For example, there's no agreed-upon definition for food allergies and no clear best way to diagnose them according to the research, published in today in the Journal of the American Medical Association.

Of course, all that makes it tremendously difficult to pin point who has a legitimate food allergy and who doesn't. Researchers think food allergies affect more than 1 or 2 percent -- but less than 10 percent -- of the U.S. population. Whether the prevalence of food allergies are increasing, well that's not clear either.

In terms of specific treatment, researchers found there is not enough study on the effectiveness of elimination diets and immunotherapy. With infants at high risk of developing milk allergies, it's unclear what treatment is most effective. Giving infants hydrolyzed formula may work to prevent allergy, but more research is needed. The same is true of giving infants probiotics in addition to breast milk or hypoallergenic formula -- results remain unclear, the study states.