collected notification on the Structured Medical Record [SMR] in Subjective, Pre-appreciation, Objective, Appreciation and Post-appreciation [SPOAP] and on Not Structured Medical Record [NSMR] . It is a retrospective, analytic study, including 910 DM. Executed in four primary health centers, at Monastir governorate in 2010. As regards methodology, we conducted a sampling at 3 degrees. The first draw of the month, the second on the weeks, the third is systematic type of medical records with a step of sounding of 2. We collected information about patient's socio-demographic characteristics, the contact patterns, clinical examination of the data, assumptions and diagnostics procedures. We used chi2 test to compare the distribution between SMR and NSMR at the Threshold of 5%. Four hundred and one SMR [44%] and 509 [56%] NSMR were included. The contact patterns was noted on 44% of NSMR and 93% of SMR [< 10-4] . The physical examination had been noted on 67% of SMR and 8% of NSMR [p < 10-4] , the hypotheses diagnoses on 72% of SMR and 31% of NSMR [p < 10-4] . The conducts had been noted on 98% of SMR and 95% of NSMR [p < 0, 045] . The distribution of the motives for contacts, physical acts, hypotheses diagnoses and therapeutic families were different between SMR and NSMR. The medical records was adequate in 52% of SMR and in 2% of NSMR [p < 10-4] . The use of SMR improves the notification and the care continuity