Among children with food allergy, do sociodemographic factors and healthcare use differ by severity?

ABSTRACT

Among children with food allergy, we aim to describe differences in allergy severity by sociodemographic characteristics and potential differences in healthcare characteristics according to food allergy severity. Using the 2007 National Survey of Children's Health, we identified children with food allergies based on parental report (n = 4,657). Food allergic children were classified by the severity of their food allergy, as either mild (n = 2,333) or moderate/severe (n = 2,285). Using logistic regression, we estimated the odds of having moderate/severe versus mild food allergy by sociodemographic characteristics and the odds of having selected healthcare characteristics by food allergy severity. Among children with food allergy, those who were older (ages 6 through 17 years) and those who had siblings were more likely to have moderate/severe allergy compared to their younger and only-child counterparts. There were no significant differences in severity by other sociodemographic characteristics. Children with a moderate/severe food allergy were more likely to report use of an Individual Education Plan (OR = 1.88 [1.31, 2.70]) and to have seen a specialist than those with mild food allergy. Among younger children with food allergy, those with moderate/severe food allergy were more likely to require more services than is usual compared with those with mild allergy. Associations between allergy severity and health care-related variables did not differ significantly by race/ethnicity, income level, or maternal education. We report few differences in allergy severity by sociodemographic characteristics of food allergic children. In addition, we found that associations between allergy severity and use of health related services did not differ significantly by race/ethnicity or poverty status among children with food allergy. Given the importance of food allergy as an emerging public health issue, further research to confirm these findings would be useful.