OBJECTIVES: The 15197:2003 ISO states that 95% of the glucose results shall fall within ±15 mg/dl of the concentrations ≤75 mg/dl and ±20% at those >75 mg/dl.  The objective was to estimate the 1-year economic impact for the Spanish Health System of blood glucose self-monitoring by using glucose meters with different degrees of accuracy METHODS: A probabilistic model was designed to estimate the clinical and economic outcomes of a type 2 (T2D) or type 1 diabetes (T1D) cohort. A second-order Monte Carlo simulation was run, in order to estimate the frequency of non-detected hypoglycemia and hyperglycemia cases. The frequency of macro and microvascular events, associated with the non-detected readings, was calculated. Finally, the economic impact was assessed for better accuracy levels (±15; ±5%) RESULTS: We estimated a total prevalence of 96,169 type 1 and 3,115,866 type 2 treated diabetic patients in Spain. We included all the T1D patients and those insulin dependents T2D (23.4%). The average annual cost of associated events and those of monitoring blood glucose were estimated at 95,987,193 € for T2D and 11,501,292 € for T1D with a ±20% accuracy. When an accuracy of ±15% was analyzed, the annual costs for T2D were reduced to 88,349,485 € and to 9,180,317 € for T1D, showing a total saving of 9,958,682 € a year. If the accuracy rose up to ±5%, the results were 80,067,179 € in T2D and 6,084,776 € in T1D, for a total saving of 21,336.530 € a year. The total costs reduced by 9.3%, 15.4% and 19.9% with accuracy of ±15%, ±10% and ±5%, respectively, respect of the initial ±20% CONCLUSIONS: This study shows that if the accuracy of the glucose meters raises, several macrovascular and microvascular events and hypoglycemic episodes could be avoided. That can improve patients’ quality of life and reduce significantly the associated costs