Infection is a major complication associated with implanted medical devices. This is due, in part, to the ability of foreign material to markedly potentiate the infectivity of some microbial pathogens. Although data in the literature suggest that low doses of some microorganisms have the potential to produce clinically significant infections, the dose-response relationship for this effect has been poorly characterized, especially at low doses. To more accurately assess the risk posed by exposure to microbially contaminated devices, dose-response data has been obtained in our laboratories to characterize the dose-response relationship for S. aureus on a variety of suture materials implanted subcutaneously in mice. Mathematical dose-response models which have been used to estimate the risk of infection from ingestion of waterborne pathogens (e.g., exponential and Beta-Poisson models) were applied using data from our studies to estimate the probability of infection to low doses (< 20 organisms) of pathogens on medical devices. Estimates of the probability of infection from a given dose of an organism, and information on the frequency of device use, may allow the FDA to develop a science-based framework for the derivation of Sterility Assurance Levels (SALs) for medical devices.