To the Editor:
Food allergy is a growing and serious health problem in the US(1). However, the relationships between demographic characteristics, socioeconomic status (SES), and food allergy are unclear. Racial disparities have been identified in the prevalence of food specific IgE(2) and in the clinical practice of allergy-related healthcare(3).
To evaluate socioeconomic and racial/ethnic disparities in food allergy, we investigated associations of race/ethnicity and SES factors during pregnancy with both food-specific IgE and allergic-type symptoms to common food allergens (milk, egg, peanut, soy, and wheat) in adolescents in a large Boston-area cohort.
The study participants were 1,148 adolescents from Project Viva, a prospective pre-birth cohort. Participant’s mothers were enrolled at their initial prenatal visit (between April 1999 and July 2002). Further details about the cohort, exposure and outcome assessment, and statistical analyses are described in this article’s online repository.
We examined the following indicators of SES: maternal education, smoking by the mother during pregnancy and by anyone in the home during the child’s first year of life, household income, and median household income from the neighborhood census tract. We determined the race/ethnicity of the adolescent by asking the mother to choose one or more racial/ethnic groups for her child
For those participants with sufficient blood samples to measure specific IgE levels by Phadia ImmunoCAP, we examined sensitivity to each of these 5 common food allergens (milk, egg, peanut, soy, and wheat). We also evaluated IgE-mediated symptoms (hives, wheeze, dyspnea, itchy throat, cough, rhinitis, eczema, nausea/vomiting, diarrhea, loss of consciousness) following food intake, as reported by maternal questionnaire. We assessed associations of adolescent race/ethnicity and SES with allergy symptoms and allergen-specific IgE using logistic regression models. For high prevalence outcomes (peanut and wheat-specific IgE, with prevalence of >27% among non-Hispanic black and “other” participants), we also determined prevalence ratios. All models adjusted for age, sex, race/ethnicity and parental history of atopy.
See online repository Table E1 for participant characteristics. Of the 647 adolescents with sufficient blood samples, 31% had food specific IgE >0.35 k/UL to at least one of the 5 foods (See Table E2 in this article’s online repository).
Associations of race/ethnicity and SES indicators with allergic sensitization are reported in Table 1. Non-Hispanic black teens had higher risk of sensitization to each of the 5 foods, reaching statistical significance for peanut (prevalence ratio 2.29, 95% CI 1.54, 3.39) and wheat (prevalence ratio 1.64, 95% CI 1.15, 2.34).
Table 1: Associations of Race/Ethnicity and SES with Food-Specific IgE.
Data from 647 adolescents participating in Project Viva who had IgE samples
| Milk | Egg | Peanut | Soy | Wheat | |
|---|---|---|---|---|---|
| Adjusted odds ratio or prevalence ratio* (95% CI) for food specific IgE >0.35 k/UL | |||||
| Adolescent race/ethnicity | |||||
| . Non-Hispanic White | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| . Non-Hispanic Black | 1.42 (0.75, 2.68) | 1.78 (0.84, 3.77) | 2.29 (1.54, 3.39) | 1.68 (0.86, 3.28) | 1.64 (1.15, 2.34) |
| . Hispanic | 1.55 (0.73, 3.27) | 1.75 (0.72, 4.26) | 1.28 (0.70, 2.23) | 1.32 (0.55, 3.12) | 0.66 (0.33, 1.31) |
| . Other | 0.96 (0.39, 2.37) | 1.80 (0.74, 4.39) | 2.25 (1.41, 3.58) | 1.94 (0.87, 4.31) | 1.45 (0.92, 2.28) |
| Maternal education | |||||
| . <College | 3.00 (1.46, 6.16) | 1.53 (0.68, 3.46) | 1.70 (0.94, 3.07) | 1.31 (0.63, 2.69) | 1.23 (0.69, 2.17) |
| . College graduate | 1.68 (0.85, 3.35) | 1.30 (0.61, 2.75) | 1.53 (0.89, 2.63) | 1.13 (0.58, 2.19) | 1.73 (1.06, 2.81) |
| . Graduate degree | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Smoked during pregnancy (yes v. no) | 0.76 (0.31, 1.84) | 0.86 (0.32, 2.31) | 1.80 (0.97, 3.34) | 1.76 (0.84, 3.71) | 1.37 (0.73, 2.54) |
| Annual household income | |||||
| . <$40,000 | 1.42 (0.63, 3.20) | 1.33 (0.51, 3.42) | 1.37 (0.67, 2.77) | 1.22 (0.48, 3.13) | 0.63 (0.30, 1.32) |
| . $40,001-70,000 | 1.29 (0.68, 2.48) | 0.91 (0.40, 2.08) | 1.61 (0.92, 2.81) | 2.02 (1.05, 3.89) | 1.30 (0.78, 2.18) |
| . >$70,000 | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Any smokers at home in first year (yes v. no) | 0.80 (0.36, 1.78) | 0.44 (0.15, 1.32) | 0.69 (0.34, 1.37) | 1.17 (0.54, 2.56) | 0.86 (0.46, 1.62) |
| Census tract median household income (lowest quartile v. upper 3 quartiles) | 1.24 (0.65, 2.38) | 1.59 (0.75, 3.39) | 0.78 (0.44, 1.36) | 0.90 (0.44, 1.81) | 0.72 (0.41, 1.25) |
All models adjusted for adolescent age, sex, and race/ethnicity and parental history of atopy
Prevalence ratio used in models with common outcomes instead of odds ratio (associations of adolescent race/ethnicity with peanut and wheat IgE)
Adolescents whose mothers had less than a college degree vs. a graduate degree had higher odds of sensitization to milk (OR 3.00, 95% CI 1.46, 6.16) and adolescents whose mothers had a college degree vs. a graduate degree had higher odds of sensitization to wheat (OR 1.73, 95% CI 1.06, 2.81). Associations between low maternal education and each of the other 3 foods were also positive however confidence intervals were wide and included 1.0. Adolescents from homes with annual household income ≤$40,000 vs. >$70,000 also had greater odds of sensitization to milk, egg, peanut, and soy, although comparisons did not reach statistical significance. Similarly, this pattern was observed for milk, peanut, soy, and wheat in adolescents from households with incomes $40,001-$70,000 vs. >$70,000, reaching statistical significance for soy (OR 2.02, 95% CI 1.05, 3.89). Maternal prenatal smoking was associated with a higher odds of peanut sensitization (OR 1.80; 95% CI 0.97, 3.34), but the lower confidence limit did cross 1.0. Otherwise exposure to smoking and living in lowest quartile of census tract income were not associated with allergic sensitization.
The prevalence of food allergy symptoms is reported in online repository E2. Of the 1,145 participants, 83 (7.2%) reported allergic-type symptoms to any food. Associations of race/ethnicity and SES with allergic symptoms upon food ingestion are shown in Table 2. There were no convincing patterns of association for any measures of SES with food allergy symptoms. However, non-Hispanic black adolescents had a higher odds of maternal reporting of allergic symptoms to peanut (OR 2.41, 95% CI 1.12, 5.17) compared to non-Hispanic white adolescents.
Table 2: Associations of Adolescent Race/Ethnicity and Maternal SES with Allergic Symptoms.
Data from 1145 adolescents participating in Project Viva who had allergic symptoms data
| Milk | Egg | Peanut | Wheat | |
|---|---|---|---|---|
| Adjusted OR (95% CI) for allergic symptoms | ||||
| Adolescent race/ethnicity | ||||
| . Non-Hispanic White | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| . Non-Hispanic Black | 1.41 (0.38, 5.29) | 2.66 (0.63,11.28) | 2.41 (1.12, 5.17) | 0.72 (0.09, 6.03) |
| . Hispanic | NA | NA | 1.44 (0.48, 4.35) | NA |
| . Other | 1.52 (0.32, 7.16) | 1.35 (0.16,11.66) | 4.14 (1.90, 8.99) | 2.29 (0.45,11.50) |
| Maternal education | ||||
| . <College | 1.85 (0.38, 9.14) | 0.44 (0.07, 2.98) | 0.82 (0.34, 1.99) | 0.85 (0.14, 5.18) |
| . College graduate | 2.23 (0.56, 8.84) | 0.60 (0.13, 2.80) | 1.76 (0.85, 3.66) | 0.77 (0.17, 3.52) |
| . Graduate degree | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Smoked during pregnancy (yes v. no) | 1.67 (0.36, 7.63) | 1.29 (0.16,10.58) | 1.16 (0.44, 3.08) | 1.25 (0.15,10.22) |
| Annual household income | ||||
| . <$40,000 | 0.57 (0.06, 5.57) | NA | 0.34 (0.12, 1.01) | 2.14 (0.32,14.13) |
| . $40,000-70,000 | 2.67 (0.86, 8.27) | 0.99 (0.18, 5.40) | 0.15 (0.03, 0.64) | NA |
| . >$70,000 | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Any smokers at home during first year (yes v. no) | 1.06 (0.23, 4.95) | 0.89 (0.10, 7.68) | 0.90 (0.36, 2.22) | 0.91 (0.11, 7.56) |
| Census tract median household income (lowest quartile v. upper 3 quartiles) | 0.78 (0.17, 3.64) | 0.18 (0.02, 1.79) | 1.29 (0.61, 2.74) | 0.44 (0.04, 4.47) |
All models adjusted for adolescent age, sex, and race/ethnicity and parental history of atopy. NA indicates insufficient prevalence of outcome to estimate OR (95% CI).
In this prospective pre-birth cohort residing in the Northeastern U.S., there was a pattern of higher odds of sensitization to all 5 foods studied in association with non-Hispanic black race. Lower household income was associated with higher odds of sensitization to soy, and lower maternal educational attainment was associated with higher odds of sensitization to milk and wheat.
Current theories to explain the increasing prevalence of food allergies have focused on hygiene, microbiome, diet, and allergen exposure(4), but have not generally focused on racial/ethnic or socioeconomic disparities underlying this growing childhood health problem. One possible explanation may be early childhood diet. While the LEAP randomized trial demonstrated a reduced risk of developing a peanut allergy in high risk infants if peanuts are introduced at an early age(5), CDC data has shown that early childhood nut consumption varies by race and SES. Specifically, non-Hispanic black youths and those from lower income households consume fewer nuts than white children and those in higher income households(6). These dietary patterns may help explain the higher prevalence of nut allergy observed in our study and others among non-Hispanic black youth.
Our findings are in agreement with other studies in the U.S. that have found a higher prevalence of food sensitization among non-Hispanic blacks(1,7) and in those from lower SES homes(7), although interestingly self-report of food allergy has been found to be less common in these groups(8). Despite an overall decline in food allergy prevalence with age, we demonstrate that racial disparities in food allergen sensitization persist into the teenage years.
Together with evidence of racial disparities in diagnosis and treatment of food allergies(2,3,7) our findings indicate a need to educate both the scientific and lay community about demographic risk factors for food allergy in adolescents.
Supplementary Material
Clinical Implications:
Our study indicates the existence of racial/ethnic and socioeconomic differences in the prevalence of food allergy among adolescents. These results further the concern that food allergy may be under-diagnosed and under-treated among racial minorities and economically disadvantaged youth.
Acknowledgments
Funding: This work was supported by the US National Institutes of Health (K23ES026204, R01HD034568, R01AI102960, UH3OD023286, R01 AI118833 and U19 AI136053), the American Thoracic Society Foundation and the American Lung Association.
Conflict of Interest: We confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome (see funding for grant information).
References:
- 1.Gupta RS, Springston EE, Warrier MR, Smith B, Kumar R, Pongracic J, et al. The prevalence, severity, and distribution of childhood food allergy in the United States. Pediatrics. 2011. July;128(1):e9–17. [DOI] [PubMed] [Google Scholar]
- 2.Branum AM, Lukacs SL. Food allergy among children in the United States. Pediatrics. 2009. December;124(6):1549–55. [DOI] [PubMed] [Google Scholar]
- 3.Hannaway PJ, Connelly ME, Cobbett RM, Dobrow PJ. Differences in race, ethnicity, and socioeconomic status in schoolchildren dispensed injectable epinephrine in 3 Massachusetts school districts. Ann Allergy Asthma Immunol. 2005. August;95(2):143–8. [DOI] [PubMed] [Google Scholar]
- 4.Sicherer SH, Sampson HA. Food allergy: Epidemiology, pathogenesis, diagnosis, and treatment. J Allergy Clin Immunol. 2014. February;133(2):291–307; quiz 308. [DOI] [PubMed] [Google Scholar]
- 5.Sicherer SH, Munoz-Furlong A, Burks AW, Sampson HA. Prevalence of peanut and tree nut allergy in the US determined by a random digit dial telephone survey. J Allergy Clin Immunol. 1999. April;103(4):559–62. [DOI] [PubMed] [Google Scholar]
- 6.Nielson S, Herrick K, Akinbami L, Ogden C. Nut Consumption , 2009-2012 U.S. Youth. U.S. Center for Disease Control. NCHS Data Brief: No 238; 2016. Available from: https://www.cdc.gov/nchs/data/databriefs/db238.pdf [PubMed]
- 7.Liu AH, Jaramillo R, Sicherer SH, Wood RA, Bock SA, Burks AW, et al. National prevalence and risk factors for food allergy and relationship to asthma: results from the National Health and Nutrition Examination Survey 2005-2006. J Allergy Clin Immunol. 2010. October;126(4):798–806. e13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McGowan EC, Matsui EC, Peng R, Salo PM, Zeldin DC, Keet CA. Racial/ethnic and socioeconomic differences in self-reported food allergy among food-sensitized children in National Health and Nutrition Examination Survey III. Ann Allergy Asthma Immunol. 2016. November;117(5):570–572.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.