OBJECTIVES: To introduce the application of stochastic simulation models in drug safety and demonstrate how the population impact of a drug’s safety profile, and other “what if” scenarios, can be quantified. METHODS: The patient group is defined according to the epidemiology of the condition. This group will form the cohort that will be followed up over a specified period. Baseline risks of the events and death, in the absence of drug exposure, are then assigned to each patient according to age, sex and other relevant risk factors. These parameters are obtained from available studies. Random times to each event and death are generated for each patient by applying a model derived from the exponential distribution; the unique parameter is the risk of each event. Case fatality is randomly assigned. Following a non-fatal event during the simulated follow-up, the risk of recurrent and related events is updated. This cohort provides the expected number of events and forms a comparator cohort. Subsequently, scenarios of drug exposure, or channeling associated with drug use, are created and compared with the comparator cohort. RESULTS: Drug exposure scenarios are modeled by applying relative risks “RR” to each patient’s baseline risks. The RRs associated with drug exposure may be sought from studies or may represent “what if” scenarios. Channeling can be modeled by altering the composition of the patient group. Tabular and graphical summaries of the net effect of drug exposure can then be created. CONCLUSIONS: This approach incorporates relevant epidemiological data into a single framework, offers the opportunity of evaluating potential drug safety issues and may be applied to other aspects of drug risk-benefit.