Clostridium difficile diarrhea is a major cause of morbidity and mortality in hospitals. However, the number of cases in an outbreak is usually relatively small. This precludes many traditional statistical methods of modeling epidemics. Stochastic models are designed to deal with small numbers and are promising methods of understanding C. difficile epidemiology. This is illustrated by a reversible jump Markov chain Monte Carlo model based on the herd immunity hypothesis of C. difficile outbreaks.