The aim of this research is to determine the factors affecting the level of out-of-pocket health expenditures as a proportion of total health expenditure in OECD countries by using decision tree method and to classify countries according to these variables. The research population comprised the 2015 data of 36 OECD member countries. In the research, the proportion of out-of-pocket health expenditures in total health expenditures of the countries as a dependent variable and the per capita gross domestic product, the ratio of the population covered by the health insurance, the ratio of the population living in the rural area, the ratio of the population of 0-14 years and 65 years of age, the proportion of the population with at least bachelor degree, the mortality rate due to cardiovascular disease, cancer, diabetes or respiratory diseases, the ratio of the population that perceives the health status as bad, domestic government health expenditure as a proportion of the total expenditure and the number of labor force as independent variables are used. The decision tree model was built using the Orange data mining program using the CART algorithm. As a result of the analyzes, it was found that the most important factor affecting the out-of-pocket health expenditures was the domestic government health expenditure as a proportion of the total expenditure and the other factors were the ratio of the population living in the rural areas, the ratio of the population 65 years and older, the proportion of the population with at least bachelor degree and the number of labor force. According to the model, it was found that the countries gathered in 9 classes. It is thought that the results obtained from the research will provide health policy makers and planners with important evidence-based information in the decisions to be taken regarding the determination of the level of health expenditures.