OBJECTIVES: Pneumococcal vaccination programmes change the natural course of infection as the number of susceptible subjects decrease along time. Markovian and Discrete Event Simulation models enable to capture the whole vaccination effect as infection rate remains constant. A differential equation dynamic model based on Anderson and May work was developed to describe how pneumococcal vaccination modifies the extension of the disease in the susceptible population. METHODS: Measure of epidemiological effectiveness was the number of contagions avoided by the preventive intervention. No assumptions on mortality and life-years-gained were considered. The nonlinear ordinary differential equations system proposed was dS(t)/dt = -β*I(t)*S(t)+γ*I(t)-V(t) dI(t)/dt = + β*I(t)*S(t)-γ*I(t) where: t = moment in time (months); I(t) and S(t) = number of infective and susceptible subjects at each time t; β = transmission coefficient; γ = natural withdrawal coefficient. The first order derivatives with respect to t, dI(t)/dt and dS(t)/dt, indicate the instantaneous variation rates in time of infective and susceptible, while V(t) indicates the number of vaccinated individuals at each time. Study time horizon was five years. Spanish 65-years-old cohort annually vaccinated was 318,000 subjects. The parameters to populate the model came from Spanish Ministry of Health database (CMBD) and published data. RESULTS: Over a 5-year period, the number of avoided contagions derived from the implementation of the vaccination strategy would be 83,844 with a clear cumulative profile (1,922 on the 1st year; 7,874 on the 2nd; 15,748 on the 3rd; 24,683 on the 4th and 33,617 on the 5th). CONCLUSIONS: Dynamic models should be used to assess the impact of vaccination programs for infectious diseases where the infection strength varies along time. The goodness of fit of this pneumococcal dynamic model was high and captured health outcomes more easily than alternative modelling methodologies.