UC Berkeley

Background

Asthma disproportionately affects communities of colour in the USA, but the underlying factors for this remain poorly understood. In this study, we assess the role of historical redlining as outlined in security maps created by the Home Owners' Loan Corporation (HOLC), the discriminatory practice of categorising neighbourhoods on the basis of perceived mortgage investment risk, on the burden of asthma in these neighbourhoods.

Methods

We did an ecological study of HOLC risk grades and asthma exacerbations in California using the security maps available for the following eight cities: Fresno, Los Angeles, Oakland, Sacramento, San Diego, San Jose, San Francisco, and Stockton. Each census tract was categorised into one of four risk levels (A, B, C, or D) on the basis of the location of population-weighted centroids on security maps, with the worst risk level (D) indicating historical redlining. We obtained census tract-level rates of emergency department visits due to asthma from CalEnviroScreen 3.0. We assessed the relationship between risk grade and log-transformed asthma visit rates between 2011 and 2013 using ordinary least squares regression. We included potential confounding variables from the 2010 Census and CalEnviroScreen 3.0: diesel exhaust particle emissions, PM_2·5, and percent of the population living below 2 times the federal poverty level. We also built random intercept and slope models to assess city-level variation in the relationship between redlining and asthma.

Findings

In the 1431 census tracts assessed (64 [4·5%] grade A, 241 [16·8%] grade B, 719 [50·2%] grade C, and 407 [28·4%] grade D), the proportion of the population that was non-Hispanic black and Hispanic, the percentage of the population living in poverty, and diesel exhaust particle emissions all significantly increased as security map risk grade worsened (p<0·0001). The median age-adjusted rates of emergency department visits due to asthma were 2·4 times higher in census tracts that were previously redlined (median 63·5 [IQR 34·3] visits per 10 000 residents per year [2011-13]) than in tracts at the lowest risk level (26·5 [18·4]). In adjusted models, redlined census tracts were associated with a relative risk of 1·39 (95% CI 1·21-1·57) in rates of emergency department visits due to asthma compared with that of lowest-risk census tracts.

Interpretation

Historically redlined census tracts have significantly higher rates of emergency department visits due to asthma, suggesting that this discriminatory practice might be contributing to racial and ethnic asthma health disparities.

Funding

National Heart Lung Blood Institute.