Over the years, an increasing number of studies have focused on the impact on public health of non-transmissible diseases. These conditions (which include, e.g., cardiovascular and oncological diseases) are characterized by the fact that they are driven by the exposure to behavioral risk factors (such as smoking, sedentary lifestyle, alcohol consumption) and not by human-to-human interactions. In this paper, we focus on the impact of three behavioral risk factors: smoking, sedentary lifestyle and alcohol consumption. We construct a Markovian model that characterizes each agent by an independent Markov chain, whose states describe age, gender, exposure to risk factors and health of the individual.