Predictors of Food Allergy in Infants in Northern Virginia
Sonya V. Chawla
Background: The prevalence of food allergies and other atopic diseases in Western cultures
appears to be on the rise. Recent estimates show the frequency of food allergies in the American
general population to be between 6% - 8%.1 In the past five years, peanut sensitization alone
doubled in young children from the United States and the United Kingdom with a total
prevalence of 1.6% in young children. Similarly, milk and egg allergies appear to be rising with
milk allergies now affecting approximately 2.5% of infants and egg allergy affecting 1.6%-2.6%
of infants.2 The prevalence of atopic dermatitis, a symptom that is often the first sign of a food
allergy, is also increasing, with an overall prevalence of 10% in children.3
Purpose: The purpose of this study is to examine factors that may predict the development of
food allergy in infants. The primary study endpoint will be the correlation between positive food
allergies in patients between the ages of 0-3 and the presence of a family history of atopy
(including allergic rhinitis, eczema, asthma or other food allergies). The secondary endpoint of
the study will be the correlation between food allergies and symptoms of mild, moderate or
severe eczema in the patient. Another secondary endpoint of the study will be to determine the
prevalence of food allergy among eczematous patients and among all patients in this Northern
Virginia pediatric population using ICD-9 codes for Ingested Food Allergy and Eczema.
Methods: With the permission of the physicians at Northern Virginia Pediatrics, a large private
pediatric group in Arlington, VA, the principal investigator PI will perform a retrospective chart
review of 500 patients who presented with eczema and underwent radioallergosorbent testing
(RAST blood testing) for common food allergens. At the Northern Virginia Pediatrics practice,
patients between 0-3 years of age are sent for RAST testing when they present with symptoms of
eczema to identify possible food allergies that may be contributing to symptoms. Each patient
will be assigned a numeric code that will be entered into an Excel spreadsheet. The PI will
record demographic information about each patient (including age, sex and ethnicity), allergic
symptoms upon presentation to the pediatrician (mild, moderate or severe eczema), family
history of atopy (including parent/sibling history of allergic rhinitis, asthma, eczema, food
allergy or true drug allergy) and lab results for RAST testing indicating positive or negative food
allergy. RAST testing is performed for the 5 most common food allergens in infants – peanut,
Grundy J, Matthews S, Bateman B, et al. 2002. Rising prevalence of allergy to peanut in children: data from two
sequential cohorts. J All Clin Immuno. 110:784-8.
Bhombal S, Bothwell M. 2004. Prevalence of elevated total IgE and food allergies in a consecutive series of ENT
pediatric patients. Presented at Ann Mtg of Am Acad of Otolaryngology, NewYork, NY 2004.
Lee L, Burks W. 2006. Food Allergies: prevelance, molecular characterization, and treatment/prevention strategies.
Ann Rev Nut. 26: 539-565.
milk, soy, wheat, and eggs. Data will be recorded directly into the Excel spreadsheet. The Excel
spreadsheet document will be password protected by the PI.
Statistical Analysis: At the outset, a sample size will be outlined that will give adequate power
to the study. P values will be calculated to determine the statistical significance of the
correlation of family history of each atopic condition with positive food allergy in the infant. We
will calculate the degree of correlation between symptoms of mild v. moderate v. severe eczema
and diagnosis of positive food allergy. Finally, the prevalence of food allergy among patients
presenting with eczema and among all patients at this practice will be calculated.
Risks: There is no treatment plan for the study and participants are expected to incur zero risk.
Costs/Funding: Minimal cost would be required for such a study, as no treatments or products
of any type will be necessary to conduct this research. The time of the PI and assistants at the
Northern Virginia facility will be compensated as part of their hourly wages.
Future: If the study is able to identify that specific family history (i.e. allergic rhinitis in 1 or
both parents) has a high degree of correlation with the development of food allergy in infants, it
may empower physicians to better educate adults planning to start a family. If certain family
history may predict the development of food allergies in an infant, then pregnant women may be
educated about the importance of avoiding highly allergenic foods during the third trimester of
pregnancy and during breast feeding, when the child’s immune system is still weak and therefore
susceptible to developing food allergies. Of course, before such advice becomes the mainstay of
practice, the study will have to be conducted in multiple centers encompassing various
geographic regions in order to make general conclusions about the predictors of food allergy in