Rethinking race/ethnicity, income, and childhood asthma: racial/ethnic disparities concentrated among the very poor

OBJECTIVE: Past studies of the prevalence of childhood asthma have yielded conflicting findings as to whether racial/ethnic disparities remain after other factors, such as income, are taken into account. The objective of this study was to examine the association of race/ethnicity and family income with the prevalence of childhood asthma and to assess whether racial/ethnic disparities vary by income strata. METHODS: Cross-sectional data on 14,244 children aged <18 years old in the 1997 National Health Interview Survey were examined. The authors used logistic regression to analyze the independent and joint effects of race/ethnicity and income-to-federal poverty level (FPL) ratio, adjusting for demographic covariates. The main outcome measure was parental report of the child having ever been diagnosed with asthma. RESULTS: Bivariate analyses, based on weighted percentages, revealed that asthma was more prevalent among non-Hispanic black children (13.6%) than among non-Hispanic white children (11.2%; p<0.01), but the prevalence of asthma did not differ significantly between Hispanic children (10.1%) and non-Hispanic white children (11.2%; p=0.13). Overall, non-Hispanic black children were at higher risk for asthma than non-Hispanic white children (adjusted odds ratio OR]=1.20; 95% confidence interval CI] 1.03, 1.40), after adjustment for sociodemographic variables, including the ratio of annual family income to the FPL. Asthma prevalence did not differ between Hispanic children and non-Hispanic white children in adjusted analyses (adjusted OR=0.85; 95% CI 0.71, 1.02). Analyses stratified by income revealed that only among children from families with incomes less than half the FPL did non-Hispanic black children have a higher risk of asthma than non-Hispanic white children (adjusted OR=1.99; 95% CI 1.09, 3.64). No black vs. white differences existed at other income levels. Subsequent analyses of these very poor children that took into account additional potentially explanatory variables did not attenuate the higher asthma risk for very poor non-Hispanic black children relative to very poor non-Hispanic white children. CONCLUSIONS: Non-Hispanic black children were at substantially higher risk of asthma than non-Hispanic white children only among the very poor. The concentration of racial/ethnic differences only among the very poor suggests that patterns of social and environmental exposures must overshadow any hypothetical genetic risk.