Dependent Variables. This study applied Data Envelopment Analysis (DEA) methodology to create efficiency scores as the measure for hospital efficiency for each hospital in each study year. DEA is a non-parametric methodology used to measure the technical efficiency of decision-making units when the production process involves multiple inputs and outputs. Briefly, DEA identifies an optimal frontier of inputs and outputs on the basis of the actual performance of decision-making units to measure the relative efficiency of one unit versus another. We used a variable return-to-scale specification of the model to allow for the possibility of varying and extreme returns to scale from inputs and outputs among study hospitals. Further, an input-oriented approach was chosen because we assumed that hospitals have more control over inputs than they do over outputs.