Prior research into multiple sclerosis prognosis has produced conflicting results. This paper presents an original approach in which the disease course is described by the movements of patients through well-defined disease states. A Markov model is proposed to describe these movements and to evaluate the effect of prognostic factors on transitions from state to state. The feasibility and applicability of this model is determined using data on the course of disease in 278 diagnosed patients from Lyon. Patients with older age at onset, females, and those with monosymptomatic onset are found to be at a higher risk of transition to a worse disease state.