Abstract
OBJECTIVE: To examine the association between (1) local political party, (2) urban policies, measured by spending on local programmes, and (3) income inequality with premature mortality in large US cities. DESIGN: Cross sectional ecological study. OUTCOME MEASURES: All cause death rates and death rates attributable to preventable or immediate causes for people under age 75. PREDICTOR MEASURES: Income inequality, city spending, and social factors. SETTING: All central cities in the US with population equal to or greater than 100 000. RESULTS: Income inequality is the most significant social variable associated with preventable or immediate death rates, and the relation is very strong: a unit increase in the Gini coefficient is associated with 37% higher death rates. Spending on police is associated with 23% higher preventable death rates compared with 14% lower death rates in cities with high spending on roads. CONCLUSIONS: Cities with high income inequality and poverty are so far unable to reduce their mortality through local expenditures on public goods, regardless of the mayoral party. Longitudinal data are necessary to determine if city spending on social programmes reduces mortality over time.