Pathway to the common measurement of urban health: a data envelopment analysis (DEA) approach

Abstract

This study examined the extent to which a generic set of indicators would permit intercity health comparisons. Central to this objective was the identification of consistent readily available indicators that would aid in the measurement of human health in urban environments. Specifically, this study sought to identify and classify health indicators related to the determinants of human health in urban environments, to compare the influence of each indicator on and among urban areas, to determine the relationship between income inequality and urban health, and similarly, between racial and ethnic diversity and urban health. Using readily-available data, this study employed an input-oriented Data Envelopment Analysis (DEA) to establish a benchmark to measure the relative health of a Metropolitan Statistical Area (MSA) along with an identified peer group and peer weights relative for each sub-optimal MSA. Of the 188 MSAs analyzed, the average effectiveness score was 0.91. Overall, 38% of the sample operated below the 0.90 level of efficiency, 51% operated at 0.90 but less than 1.00, and 11% were deemed to be effective with a score of 1.00. While no statistically significant relationship was found between income inequality and urban health, a statistically significant positive correlation was found between racial and ethnic diversity and urban health. With the establishment of a benchmark, a relative peer group, and peers weights, intercity and intracity exemplars may be identified rendering insight into best practices, pragmatic target setting and resource allocation, and the effectiveness of policy and environmental changes over time.