Introduction: The relationship between redlining and diabetes prevalence has been understudied. Residents of redlined tracts experience risk factors (inadequate diet, lack of exercise, etc.) at higher rates, which can precipitate increases in diabetes cases, especially in low-income neighborhoods. We observe the statistical association between redlining regions and diabetes prevalence and establish causality through access quantification.

Methods: Historic Redlining Scores (HRS) were collected from the Inter-university Consortium for Political and Social Research. HRS were weighted averages of HOLC A, B, C, and D grades from the 1930s. Through connection with the U.S Census Tract, we obtained diabetes rates from Houston, Miami, and Cleveland, cities with prevalent adult diabetes cases and discriminatory housing policies. Generalized Linear Regression (LR) computed the correlation between HRS and diabetes rates from 2018-2020. USDA Food Access Research Dataset was used to deduce access disparity for low-income residents in the three cities.

Results: LR models for 2018-2020 demonstrated a significant positive association between HOLC grade and diabetes rate of 3.35, 3.22, and 3.17 (all p < 2e-16), respectively. Furthermore, there is a 35% decrease in food accessibility for low-income residents within a half-mile for grade C/D tracts compared to grade A/B tracts, with over half (54%) of residents in C/D tracts without access to supermarkets within a half-mile.

Conclusion: The significant positive relationship between historic redlining rates and diabetes rates strongly suggests the effects of discriminatory policies continue to adversely affect minority communities. We believe the lack of adequate healthy food access results in minority demographics more heavily relying on processed and fast food, predisposing them to a higher risk of diabetes. Increasing infrastructure for healthy food access presently may alleviate diabetes rates in the next few decades.

Disclosure

S.R.Shah: None. T.Martheswaran: None. S.R.Shah: None. A.Kumar: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.