Validation of a Neighborhood SES Index in a Managed Care Organization

Abstract

Neighborhood socioeconomic status (SES), beyond individual SES, has been associated with health behaviors and outcomes. We constructed a neighborhood SES index using readily available US Census variables and studied associations of the index with self-reported education, formal education, and neighborhood residential characteristics among adult enrollees of a managed care organization (MCO). A neighborhood SES index was constructed for MCO enrollees using a principal components analysis of 7 Census measures of their residential tracts or block groups. Quartiles of resulting neighborhood SES scores were matched with self-reported education, household income, and neighborhood residential characteristics obtained on 3 surveys of MCO adults ranging in age from 18 to 99. Cross-tabulations of neighborhood SES with self-reported measures were used to assess concordance. Consistent with other non-MCO populations, distributions of the neighborhood SES index and self-reported education, household income, and other neighborhood residential characteristics were concordant. The presence of high SES MCO enrollees in low SES neighborhoods and low SES MCO enrollees in high SES neighborhoods was, however, common. MCO enrollees living in low SES neighborhoods also reported neighborhood problems (crime, unattended dogs) known to affect walkability more frequently than those living in high SES neighborhoods. A simple, valid neighborhood SES index can be constructed for MCO populations using readily available US Census variables. Although individual and neighborhood SES are on average concordant, discrepancies between individual and neighborhood SES suggest that other factors may influence how area-based SES contributes to health behaviors and outcomes in an MCO population.