Background: Tropospheric concentrations of ozone and particulate matter (PM) have increased significantly since preindustrial times. These pollutants have been demonstrated to cause negative health impacts, including cardiovascular and respiratory mortality. Previous estimates of the global burden of air pollution on premature human mortality have been based on compilations of global air quality measurements. Here we use results from a global atmospheric chemistry and transport model (MOZART-2) simulation of ozone and PM concentrations in 1860,the preindustrial base case, and 2000, the present day case, to drive human mortality estimates. This model-based method allows estimation of mortality in regions for which ground-level concentration measurements are unavailable and omitted from estimates based on direct air quality measurements. In addition, by simulating the preindustrial base case, we are able to isolate the mortality due to anthropogenic pollution and avoid making assumptions for background ozone and PM concentrations.