Racial and economic segregation and diabetes mortality in the USA, 2016-2020

Abstract

Background: The purpose of this study was to examine the association between racial and economic segregation and diabetes mortality among US counties from 2016 to 2020.

Methods: We conducted a cross-sectional ecological study that combined county-level diabetes mortality data from the National Vital Statistics System and sociodemographic information drawn from the 2016-2020 American Community Survey (n=2380 counties in the USA). Racialized economic segregation was measured using the Index Concentration at the Extremes (ICE) for income (ICE_income), race (ICE_race) and combined income and race (ICE_combined). ICE measures were categorised into quintiles, Q1 representing the highest concentration and Q5 the lowest concentration of low-income, non-Hispanic (NH) black and low-income NH black households, respectively. Diabetes was ascertained as the underlying cause of death. County-level covariates included the percentage of people aged ≥65 years, metropolitan designation and population size. Multilevel Poisson regression was used to estimate the adjusted mean mortality rate and adjusted risk ratios (aRR) comparing Q1 and Q5.

Results: Adjusted mean diabetes mortality rate was consistently greater in counties with higher concentrations of low-income (ICE_income) and low-income NH black households (ICE_combined). Compared with counties with the lowest concentration (Q1), counties with the highest concentration (Q5) of low-income (aRR 1.96; 95% CI 1.81 to 2.11 for ICE_income), NH black (aRR 1.32; 95% CI 1.18 to 1.47 for ICE_race) and low-income NH black households (aRR 1.70; 95% CI 1.56 to 1.84 for ICE_combined) had greater diabetes mortality.

Conclusion: Racial and economic segregation is associated with diabetes mortality across US counties.

Keywords: DIABETES MELLITUS; ETHNIC GROUPS; Health inequalities; POVERTY; SOCIAL CLASS.

Figure 1.

Rate Ratio Comparing Diabetes Mortality Rates at the County Level based on comparing Least (Q1) to Most (Q5) Privileged^a Index of Concentration at the Extremes, United States, 2016-2020 Note. ICE=Index of Concentrations at the Extreme. Data on county level ICE measures came from the US Census American Community Survey 2016-2020 5-year file. Diabetes mortality data are from the 2016-2020 National Vital Statistics System. ^aEach ICE measure was categorized into quintiles: Q1 (highest concentration of low income, Black, and low-income Black households, respectively) and Q5 (lowest concentrations)