OBJECTIVES: To identify and evaluate medication errors in terms of clinical impact and costs avoided by the pharmaceutical intervention. METHODS: The pharmaceutical monitoring in patients with hematologic diseases was performed since July 2014 to July 2015 to identify medication errors and classify them according to Otero et al. (2000). A descriptive analysis was performed including central and dispersion measures to detail their distribution and to estimate the incidence and prevalence. Each error was evaluated to identify whether it was related to the patient´s safety and /or had a clinical and economic consequence. The most common clinical consequence was determined based on scientific information of the drugs involved. The cost estimate for complications avoided was taken of Diagnosis Related Groups 2015 of Mexican Institute of Social Security expressed in US dollars. RESULTS: A total of 345 events of medication errors were detected; the prevalence and incidence estimated were 12% and 25% respectively. In the sample 49 (58%) patients were men; the mean age was 48.7 ±16.18 years. Non-Hodgkin lymphoma had the most frequent errors, in total 115 (33%). The most common errors were detected during the prescription with 137 (40%) events; the major cause was lapse/distracting in 161 (47%) errors. The most frequent error type was medication omission 143 (41%). All errors were categorized without damage. A total 316 pharmaceutical interventions were performed, preventing 47 clinical complications, this results in annual avoided cost of 97,247.92 USD, the cost saving amount per avoided complications were 2,069.10 USD and a saving per patient of 1,144.09 USD. CONCLUSIONS: Early detection of medication errors by pharmacists intervention improves the safety of pharmacotherapy, ensures the effectiveness and prevents issues to the patient, furthermore, reduces costs by optimizing resources, providing just the necessary therapy and avoiding complications which could increase direct medical cost.