To describe the pregabalin exposure-adverse event (AE) (dizziness) relationship in patients with generalized anxiety disorder, separate models were developed for the incidence of AE and for the conditional severity of AE, given that an AE has occurred using patient data from six clinical studies. The incidence component was modeled using a nonlinear logistic regression model. The conditional severity component was modeled as an ordered categorical variable with a proportional odds model to capture the severity on any given day. A Markov element was introduced to account for the correlation between neighboring observations. The proportional odds model including a time course of appearance and disappearance of AE could adequately describe the time course of probability of dizziness. Incorporating a transition model including Markov elements improved the model fit and greatly improved the predictability of the time course of probability of dizziness.