A105-24 Sensitivity of African Genetic Ancestry-lung Function Associations to Socio-environmental Model Specification in Black and Puerto Rican Children

Background

African genetic ancestry (AFR), a biological measure derived from genome-wide variation, is associated with lower lung function. Although genetic ancestry has been proposed as a biologically informative complement to race in clinical and epidemiological models, it remains unclear whether ancestry-lung function associations reflect intrinsic biological effects or socio-environmental factors.

Objective

To examine the association between AFR and lung function with adjustment for individual and combined socio-environmental factors in Black and Puerto Rican children.

Methods

We analyzed fixed complete-case samples comprising 823 Black and 851 Puerto Rican children from the SAGE and GALA asthma studies. Lung function was assessed using FEV₁ and FVC z-scores based on Global Lung Function Initiative race-neutral equations. African ancestry (per 10% increase) was evaluated in linear regression models with sequential adjustment for Area Deprivation Index (ADI), pollution mixtures summarized by principal component analysis, and a composite socioeconomic status (SES) construct derived from multiple indicators, evaluated individually and jointly. Acculturation constructs were evaluated only among Puerto Rican children.

Results

Higher African ancestry was associated with lower lung function in both populations (p < 0.01). Among Black children, AFR coefficients were highly stable across model specifications (base model: FEV₁ z β = −0.150, SE = 0.026; FVC z β = −0.146, SE = 0.024), with minimal change after adjustment for ADI, pollution, and SES, individually or jointly. Correlations between AFR and socio-environmental factors were weak, as were correlations between these factors and lung function. Among Puerto Rican children, AFR associations—particularly for FVC—varied across model specifications. While AFR-FEV₁ associations remained stable across adjustment strategies (base model: β = −0.146, SE = 0.036; fully adjusted model: β = −0.146, SE = 0.037), the AFR-FVC association was smaller in magnitude in the fully adjusted model (β = −0.166, SE = 0.037) relative to individual adjustment (β = −0.186, SE = 0.036). In this cohort, AFR showed modest correlations with ADI, pollution, SES, and acculturation, alongside stronger correlations between socio-environmental factors and lung function.

Conclusions

African genetic ancestry-lung function associations are population-specific and sensitive to socio-environmental adjustment. These findings suggest that genetic ancestry-lung function associations reflect population-specific alignment between genetic ancestry and measured socio-environmental factors, rather than a stable, intrinsic biological effect. Substituting genetic ancestry for race does not eliminate sensitivity to socio-environmental structure, underscoring the need for explicit, population-specific socio-environmental representation when interpreting ancestry-associated differences in lung function.

This abstract is funded by: National Institutes of Health, American Lung Association, UCSF Department of Pediatrics, The Catholic University of Korea