The extent of interindividual variation in ancestral proportions among Latinos presents the opportunity to correlate global ancestry with quantitative traits of asthma, such as FEV1
, a surrogate for asthma severity, and bronchodilator response to albuterol. Differences in ancestry proportion may partly explain the dramatic differential in these traits between Puerto Ricans and Mexicans. To address this hypothesis, we used regression models to test for association between individual admixture (IA) estimates and the quantitative distributions of asthma severity, as defined by FEV1
and bronchodilator responsiveness (ΔFEV1
) among Puerto Ricans and Mexicans (11
). Age, sex, asthma duration, regular use of asthma medication, socioeconomic status (SES), and body mass index (BMI) were entered in the model as covariates. To adjust for potential environmental interactions, secondhand exposure to environmental tobacco smoke (ETS) and birthplace were also incorporated into the models as covariates. We used a forward stepwise procedure to select covariates for each model.
We found significant relationships between ancestry and asthma severity in our Mexican American population. European ancestry was associated with more severe asthma (defined by lower baseline FEV1 [Pearson r = –0.211, p = 0.0051; Spearman r = −0.228, p = 0.0024]). European ancestry remained a significant predictor of baseline FEV1 in a multivariate regression model after adjustment for age, sex, asthma duration, regular use of asthma medication, ETS exposure, birthplace, SES, recruitment site, and BMI. The forward stepwise regression model identified age and ETS exposure as the only other significant effectors of baseline FEV1. A decrease of 1.7% (95% CI, 0.6–2.8%) in baseline FEV1 was observed per 10% increase in European ancestry. Because our Mexican subjects with asthma, on average, had 45% European, 52% Native American, and only 3% African ancestry, there was a strong negative correlation between Native American and European ancestry. Therefore, any negative association with European ancestry would be expected to have a positive association with Native American ancestry. As expected, in models in which Native American ancestry was the main predictor, Native American ancestry was associated with milder asthma (defined as higher measures of baseline FEV1 [Pearson r = 0.176, p = 0.0197; Spearman r = 0.186, p = 0.0139]). After correction for the aforementioned potential confounders, Native American ancestry remained significantly associated with higher baseline FEV1 values.
We also tested the association between ancestry and asthma severity as a qualitative trait, comparing the proportion of ancestry among subjects with mild versus severe asthma, as defined by asthma medication use and clinical symptoms. We found an association between higher European ancestry and asthma severity among Mexicans with asthma. For each 10% increase in European ancestry, there was an approximate 37% risk increase for severe asthma. Adjustment for age, sex, asthma duration, ETS exposure, birthplace, and BMI did not affect the association. Conversely, Native American ancestry was associated with decreased risk of asthma severity, with an OR of 0.66 (95% CI, 0.54–0.81, p = 0.00006) for having severe asthma versus mild asthma per 10% increase in Native American admixture. Both of the associations between European ancestry and lower baseline FEV1 and increased clinical severity were significant after adjustment for multiple hypothesis testing. These associations between ancestry and asthma-related phenotypes provide supporting evidence for the use of admixture mapping method (described below) to identify genes associated with asthma-related traits among Latino populations.