OBJECTIVES: Diabetes Mellitus (DM) is a chronic-degerenative disease in the population  connected to a high risk of onset of  chronic complications and co-morbidities. Despite that, very low data are available on the cost associated. The objective of this study is to identify the available information about the epidemiology of the disease and to estimate the average annual cost incurred by the National Health Service (NHS) and by the Society for the treatment of diabetes in Italy. METHODS: A probabilistic prevalence cost of Illness model was developed in order to calculate  an aggregate measure of the economic burden associated with the disease in terms of direct medical (drug, hospitalization, monitoring and AE) and indirect costs (absenteeism and early retirement) . A systematic review of the literature was conducted to determine both the epidemiological and economic data. Furthermore, a one-way and probabilistic sensitivity analysis with 5,000 Monte Carlo simulations were performed in order to test the robustness of the results and define a CI 95%.   RESULTS: The model estimated  a prevalence of 2.6 million of patients  with drug therapies in Italy. The total economic burden of diabetic patients in Italy amounted to € 20.3 billion years (95% IC: € 18.61 - € 22.29 billion). Of these, 54% are associated with indirect costs (95% CI € 10.10 - € 11.62 billion) and 46% are attributable only to the direct costs (95% IC: € 8.11 - € 11 06 billion).  CONCLUSIONS: This is a first study in Italy that aimed to estimate direct and indirect cost of diabetes with probabilistic prevalence approach. As might be expected, the lack of information means that the real burden of Diabetes is partly underestimated, especially about indirect costs. However, this approach can be very useful for policy makers in order to understand the economic aspects in the treatment of diabetes in Italy.