OBJECTIVES: Evaluate the possibility of extracting data from the EMR used by the Health Services of Colima, Mexico and use it to assess the quality of care and prescription patterns for Diabetes prevention and control in outpatient clinics. METHODS: A copy of the entire EMR database, including personal identification variables, was obtained from the Health Services of Colima. A data verification and validation process was carried out including checking for EMR duplicity using Structured Query Language (SQL) and phonetic algorithms.  A flat table for each patient’s encounter with the health services was constructed in order to have a longitudinal record along with vital signs, diagnostic and control tests as well as drugs prescribed. Each encounter was then coded to reflect in a single character string the main variables of diabetes care: number of visit after diagnosis, glucose measurement, drugs administered, as well as eye and feet examination. RESULTS: The EMR in Colima initiated its operation in 2005 as a pilot in 3 clinics, in 2010 it covered about 50% of the state’s clinics (55) and in 2013 reached 100%: 117 clinics. A total of 393,398 records were extracted and consolidated with 2,271,251 outpatient visits in the period 2005 – march 2014. The age and sex structure of the population in the EMR was very similar to that of the 2010 population center for the state.  Eleven percent of the population eighteen or older in the database was diagnosed with diabetes; only 45% of these had at least one glucose exam and only 16% were taking insulin. The most frequent medication prescribed was metformine. CONCLUSIONS: The use of data from EMR is suitable to evaluate quality of care and prescription patterns in the prevention and control of diabetes mellitus. Continuing monitoring established quality parameters and prescription may help to improve quality.