Neighborhood Social Characteristics and Chronic Disease Outcomes: Does the Geographic Scale of Neighborhood Matter?

Malia Jones, University of California, Los Angeles

Empirical research on the relationship between characteristics of the local environment and health outcomes depends upon a valid specification of geospatial scale of neighborhoods. This paper compares results of operationalizing neighborhood at four spatial scales derived from US census geography. Multilevel mulitvariate models show how neighborhood social characteristics, operationalized at each scale, predict individual-level hypertension and obesity in a sample of Los Angeles County adults (n for individuals=900, n for neighborhoods=65). Descriptive results show that neighborhood characteristics are similar across scales, but that census tracts do not accurately or consistently capture the scale at which meaningful variation in neighborhood characteristics occur. In multivariate models, the main effects of neighborhood poverty on chronic disease are not driven by the selection of census tracts as the areal unit for neighborhoods. Creating alternative neighborhood scales based on census geography does not appear to perform substantially and consistently better and involves considerable time investment.