In its mission to protect human health and the environment, the United States (U.S.) Environmental Protection Agency (EPA) implemented the NO_x Budget Trading Program (NBP) to reduce the emissions of nitrogen oxides (NO_x) in the Eastern U.S., with the intent of reducing ambient concentrations of both NO_x and the secondarily-formed ozone chemicals. These pollutants and their precursors can be transported downwind, contributing to pollutant levels at locations much farther from the emission sources, potentially impacting human health in downwind areas. This study investigates the health risks in New York State (NYS) from exposure to polluted air parcels transported from the Midwest. Back-trajectories are performed from several sites within NYS for ten summers (June, July and August from 1997 to 2006) to identify days that the air parcel passed through the Ohio River Valley (ORV) region within 48 hours back in time. The ORV zone is defined as a boundary encompassing relatively high-emitting power plants in the Midwest and is used as an indicator variable to represent the transport of ozone from this area into NYS in an epidemiology analysis. The ORV zone variable and the daily maximum 8-hr average ozone concentrations are then used as the main health effects in a Generalized Additive Model (GAM) to investigate potential associations between these two variables and respiratory-related hospital admissions. The results of the analysis indicate that the risk of being admitted to the hospital for a respiratory-related illness on days that air parcels are transported from the Midwest is elevated in NYS sub-regions 2, 3 and 6. In addition, the risk of respiratory-related hospital admission from exposure to ozone is elevated in Regions 2 and 4. Two time periods before the implementation of the NBP and after the implementation of the NBP (summers of 1997 – 2000 and 2004 – 2006, respectively) were also examined, but an analysis of the number of summers needed to conduct the analysis revealed that there are not enough years of data to discern a difference in the health signal between the two time periods. According to this analysis, the health effect measures become inconsistent (as compared to applying the model using the data from all 10 summers) at about 5 summers (552 days) or less.