A Bayesian network is developed to embed the probabilistic reasoning dependencies of the demographics on the incidence of infectious diseases. Influenza epidemics occur every year in both hemispheres during the winter. The Bayesian learning paradigm is used to create synthetic data sets that simulate an outbreak of influenza for a geographic area. The Bayesian prior and posterior probabilities can be altered to represent an outbreak for various demographics in different geographic regions. Epidemic curves are generated, via time series analysis of the data sets, for the temporal flow of influenza on different variants of the demographics. The analysis of the demographic-based epidemic curves facilitates in the identification of the risk levels among the different demographic sections. Spread vaccination lowers the impact of the epidemic, depending on the efficacy of the vaccine. Our model is equipped to analyze the effects of spread vaccination and design vaccination strategies, that optimize the use of public health resources, by identifying high-risk demographic groups. Our results show that application of the vaccine in the order of risk levels will further lower the epidemic impact as compared to uniform spread vaccination.