Abstract
BACKGROUND: Food allergy has been shown to negatively impact children's mental health and quality of life. However, its impact on school performance is unknown. OBJECTIVE: We aimed to investigate whether food allergy, severe and nonsevere, is associated with school performance when accounting for measured and unmeasured familial factors. METHODS: This was a register-based cohort study, with sibling controls, including all children born in Sweden 2001-5 (n = 456,164) with food allergy information based on hospital visits and prescriptions, grades, and national test results from all Swedish schools and confounders. Primary exposure was food allergy severity (none, nonsevere, or severe) in school years 7-9, and the primary outcome was total grades from year 9, with secondary exposures/outcomes also at younger ages. The primary outcome was analyzed by linear regression and, for sibling/twin control analyses, fixed effect linear regression. Results were replicated in a twin cohort (n = 31,609). RESULTS: In unadjusted and analyses adjusted for measured confounders, children with severe food allergy appeared to have better total grades than children without food allergy (β(unadjusted) = 10.6 [95% confidence interval (CI), 8.6, 12.6] and β(adjusted) = 5.5 [95% CI, 3.7, 7.4]). When also adjusting for unmeasured confounders shared by siblings, the difference was close to null and statistically nonsignificant (β(sibling) = 1.6 [95% CI, -1.5, 4.7]; for nonsevere food allergy, β(sibling) = -0.0 [95% CI, -2.2, 2.1]). The twin cohort results were similar. CONCLUSIONS: We found no consistent evidence of a negative effect of food allergy, either severe or nonsevere, on school performance when adjusting for measured and unmeasured confounders shared by siblings.