Contextual Predictors of Cumulative Biological Risk: Segregation and Allostatic Load


  • Data and coding information used in this analysis are available for replication purposes from Anna Bellatorre. This article was partially supported by Award R01MD004025 from the National Institute on Minority Health and Health Disparities and NHLBI Center Grant 1 P50ESO12383. The content is solely the responsibility of the authors and does not necessarily represent the official views of NIMHD, NHLBI, or NIH. This article was also partially supported by research support funds from the San Diego State University Research Foundation. The authors gratefully acknowledge the programming assistance of Ms. Aimee Bower of the RAND Corporation.

Direct correspondence to Anna Bellatorre, Department of Sociology, University of Nebraska–Lincoln, 731 Oldfather Hall, Lincoln, NE 68588 〈〉.



Segregation is considered to be a fundamental cause of race/ethnic disparities in health. However, very few studies have tested whether levels of segregation are related to health outcomes using multilevel data and appropriate methodologies. In this study, we investigate the relationships between two distinct dimensions of segregation and allostatic load to determine whether the experiences of individuals in segregated neighborhoods are related to allostatic load as a possible predisease indicator.


To test our hypotheses, we utilized publicly available data from the National Health and Nutrition Examination Survey III, 1988–1994. We utilized random-intercept hierarchical generalized Poisson regression models to conduct our multivariate analyses.


We find that inflammatory response is related to both the evenness and exposure domains of segregation. That is, both the unequal distribution of minority groups over areal units as well as the degree of potential contact between minority and majority group members are related to these predisease pathways.


In this study, we build on prior research by Massey (2004) to investigate the relationships between two distinct dimensions of segregation and allostatic load. Our results indicate that segregation is a significant predictor of allostatic load, net of individual-level characteristics.