Neighbourhood racial/ethnic residential segregation and cardiometabolic risk: the multiethnic study of atherosclerosis. Issue 1 (29th September 2018)

• Record Type: Journal Article
• Authors: Mayne, Stephanie L; Hicken, Margaret T; Merkin, Sharon Stein; Seeman, Teresa E; Kershaw, Kiarri N; Do, D Phuong; Hajat, Anjum; Diez Roux, Ana V
• Is Part Of: Journal of epidemiology and community health. Volume 73:Issue 1(2019)
• Page Start: 26
• Page End: 33
• Publication Date: 2018-09-29
• Subjects: longitudinal studies -- neighborhood/place -- social epidemiology; Public health -- Periodicals; Epidemiology -- Periodicals; 614.4
• DOI: 10.1136/jech-2018-211159
• Languages: English
• ISSNs: 0143-005X

Abstract

Abstract : Background: Racial residential segregation has been linked to adverse health outcomes, but associations may operate through multiple pathways. Prior studies have not examined associations of neighbourhood-level racial segregation with an index of cardiometabolic risk (CMR) and whether associations differ by race/ethnicity. Methods: We used data from the Multi-Ethnic Study of Atherosclerosis to estimate cross-sectional and longitudinal associations of baseline neighbourhood-level racial residential segregation with a composite measure of CMR. Participants included 5015 non-Hispanic black, non-Hispanic white and Hispanic participants aged 45–84 years old over 12 years of follow-up (2000–2012). We used linear mixed effects models to estimate race-stratified associations of own-group segregation with CMR at baseline and with the rate of annual change in CMR. Models were adjusted for sociodemographics, medication use and individual-level and neighbourhood-level socioeconomic status (SES). Results: In models adjusted for sociodemographics and medication use, high baseline segregation was associated with higher baseline CMR among blacks and Hispanics but lower baseline CMR among whites. Individual and neighbourhood-level SES fully explained observed associations between segregation and CMR for whites and Hispanics. However, associations of segregation with CMR among blacks remained (high vs low segregation: mean difference 0.17 SD units, 95% CI 0.02 to 0.32; medium vs low segregation: mean difference 0.18 SD units, 95% CI 0.03 to 0.33). Baseline segregation was not associated with change in CMR index scores over time. Conclusion: Associations of own-group racial residential segregation with CMR varied by race/ethnicity. After accounting for SES, living in a more segregated neighbourhood was associated with greater risk among black participants only.