In medical forecasting, there are often challenges in which it is necessary to assess the risk that is continuous for a long time, and important events can occur more than once. One of the ways of solving problems of this type is the use of Markov models.Markov models suggest that the patient is always in one of the finite numbers of discrete states of health, called the Markov states. All events are modeled as a transition from one state to another. In order for the Markov chain to end, it

more »

... contain at least one absorbing state from which the patient can't pass into other states. In medical models, such a condition is the death of the patient, as this is the only condition from which the patient can't escape. The purpose of this work is to build a Markov chain for predicting the prevalence of coronary heart disease in UkraineTo test the predictive quality, the model was built since 1996. The data obtained for the model for 1996-2012 were compared with known epidemiological data. The average prediction error is 3.2 %. We predict an increase in the incidence of coronary heart disease to 35 041.2 per 100 000 population in 2025.