Better estimates of food allergy prevalence

Better estimates of food allergy prevalence
lianne4
Lianne Soller

For years, experts have relied upon telephone surveys to estimate the number of Canadians with food allergies. But how do you accurately measure the prevalence of food allergies when people do not answer the phone or refuse to be surveyed?

An innovative paper by AllerGen researchers is the first to demonstrate that adjustment for nonresponse can lead to important changes in estimating food allergy prevalence. “Adjusting for nonresponse bias corrects overestimates of food allergy prevalence” was published online in January 2015, in the Journal of Allergy and Clinical Immunology: In Practice.

The paper’s findings were based upon data from a 2010‒2011 AllerGen-funded study, “Surveying Prevalence of Food Allergy in All Canadian Environments”, which surveyed 5,734 Canadian households about food allergies. The research team then adjusted for nonresponse bias by gathering information from households who refused or could not be reached to complete the study.

“This research is the first to consider the effect of non-response bias in the estimation of food allergy prevalence, and we have clearly demonstrated that doing so is crucial in developing accurate estimates,” says Lianne Soller, the paper’s first author and an AllerGen trainee.