Association Between Residential Redlining and Atopic Dermatitis in a National Birth Cohort Consortium

Authors

• Alexandra Sitarik, Henry Ford HealthFollow
• Sarah Mandel
• Patrick Ryan
• Lisa Arkin
• Diane Gold
• James Gern
• Anne Marie Singh

Recommended Citation

Sitarik A, Mandel S, Ryan P, Arkin L, Gold D, Gern J, Singh A. Association Between Residential Redlining and Atopic Dermatitis in a National Birth Cohort Consortium. J Allergy Clin Immunol 2026; 157(2):AB182.

Document Type

Conference Proceeding

Publication Date

2-10-2026

Publication Title

J Allergy Clin Immunol

Keywords

adverse outcome, asthma, atopic dermatitis, birth cohort, census tract, child, cohort analysis, cohort effect, conference abstract, controlled study, drug therapy, ethnicity, female, human, major clinical study, male, maternal smoking, neighborhood, phenotype, social segregation, special situation for pharmacovigilance

Abstract

Rationale: Historically redlined neighborhoods associate with adverse health outcomes, such as asthma. However, their contribution to disparities in atopic dermatitis (AD) is unknown. Methods: Birth addresses from the Children's Allergy & Asthma Data Repository were categorized into grades A (“Best”), B (“Still Desirable”), C (“Definitely Declining”), or D (“Hazardous”) using 1930s Home Owners’ Loan Corporation (HOLC) maps. No grade was assigned if <50% of the birth address’s census tract was within HOLC boundaries (likely rural or suburban locations); mixed grade was assigned if no single grade accounted for ≥50% of the census tract. AD was defined as current physician diagnosis at ages 1 and 5. AD phenotypes throughout childhood were also investigated. Logistic regression with random cohort effects was utilized, along with multiple imputation (N=5237). Results: After adjusting for sex, race, ethnicity, parental history of AD, maternal education, and maternal smoking, children born in grade D neighborhoods had 1.44 times higher odds of 1-year AD, relative to children born in A+B grade neighborhoods (OR (95% CI) = 1.44 (1.03, 2.00); p = 0.031). Additionally, children with no grade assigned had 1.57 times higher odds of 1-year AD vs. A+B (OR (95% CI) = 1.57 (1.15, 2.13); p = 0.004). No associations were observed for 5-year AD. For AD phenotype, children with mixed grade assignment had 3.08 times higher odds of early AD with potential reoccurrence vs. A+B (OR (95% CI) = 3.08 (1.09, 8.73); p = 0.034). Conclusions: The historical racist practice of redlining associates with present day risk of early AD.

Volume

157

Issue

2

First Page

AB182