As the world becomes increasingly urbanized, growing populations are exposed to poor ambient air quality and at risk of the associated health outcomes. Urban air quality is affected both by local sources of air pollution and sources outside city borders. Policy-makers who develop air quality policies need to know whether it is most effective to focus on local policies or to spend resources fostering larger regional air quality management cooperation. Identifying the fraction of air pollution exposure from emissions as a function of distance from the city is a critical element of air quality management design. We estimate the health burden associated with exposure to fine particulate matter (PM2.5), ozone (O3), and nitrogen dioxide (NO2) from county-level anthropogenic sources in and around 14 US cities; this analysis is a test-bed to conduct future global analyses. We use adjoint sensitivities calculated from the chemical transport model GEOS-Chem, high resolution satellite-derived surface concentrations of PM2.5 and NO2, and health impact assessment methods. For the 70.2 million people living in these cities, we estimate that 27,740 PM2.5- and O3-related premature deaths and 126,600 NO2-related new asthma cases were attributable to air pollution exposure in 2011. Development within the GEOS-Chem adjoint framework enables sectoral attribution and policy analysis in addition to the rote assessment of impact. We find that 70% of deaths and nearly 100% of these asthma cases were attributable to anthropogenic emissions. There is great variability in the sources of the anthropogenically-related health impacts; within-urban emissions make up 5% in Austin to 56% in Los Angeles and Phoenix (median: 31%) of urban premature deaths and 18% in Austin to 82% in Los Angeles (median: 59.5%) of new asthma cases, with the remaining portions attributable to emissions from outside the urban area. For each city, we estimate the air quality related health benefits associated with the adoption of a vehicle-miles-traveled fee in that city and in multiple spatial regions surrounding the city. The findings suggest that the proportion of urban air pollution that is regional is greater for premature deaths than new asthma cases and for the eastern US than the western US.