Possession of epinephrine auto-injectors by Canadians with food allergies.

To the Editor: Although there is unanimous agreement that epinephrine is the first-line treatment for anaphylaxis, many with food allergy have not been prescribed an epinephrine auto-injector (EAI). As part of our nationwide Canadian study on the prevalence of food allergy, households from the 10 Canadian provinces were randomly selected from the electronic white pages and were telephoned between May 2008 and March 2009. Households self-reporting an allergy to peanut, tree nut, fish, shellfish, and/ or sesame were recontacted within 4 months of the telephone survey and asked whether the individual(s) with allergy currently had an EAI. There was no differentiation between EAI formulations currently available in Canada (EpiPen; King Pharmaceuticals Canada, Mississauga, Ontario, Canada, and Twinject; Paladin Labs Inc, St-Laurent, Quebec, Canada). Two categories of respondents with allergy were defined: (1) those reporting a convincing history of an IgE-mediated allergic reaction* and/or a physician diagnosis of an allergy to peanut, tree nut, fish, shellfish, or sesame, termed the probable group, and (2) those reporting a physician diagnosis of an allergy to peanut, tree nut, fish, shellfish, or sesame, termed the diagnosed group. Multivariate logistic regression models were performed for each group of respondents to identify factors associated with having an EAI; multiple imputation techniques were used to adjust for missing data for the low-income variable. Both models were hierarchical using the following household-level variables: postsecondary education of household respondent (attained college/university degree), low-income household, marital status of household respondent (married/living with partner), urban location of household, and birthplace of household respondent (not born in Canada). The following individual-level data of the allergic participants were also included: age (<18 years), sex, type of allergy (peanut, tree nut, or sesame), multiple allergies (allergy to >1 of peanut, tree nut, sesame, fish, or shellfish), age at most severe reaction, treatment with epinephrine during most severe reaction, multiple allergic reactions, and self-report of diagnostic allergy testing. Of 10,596 households contacted, 3,666 responded (35% participation rate), of which 3,613 completed the entire interview, representing 9,667 individuals. Of these 9,667 individuals, 310 (3.2%) were considered to have a probable food allergy to at least one of the following: peanut, tree nut, fish shellfish, and/or sesame. Of those with probable food allergies, 261 (84%) could be recontacted and queried on the EAI (convincing history only, n5 63; diagnosis only, n5 38; convincing history and diagnosis,