OBJECTIVE： To introduce Pharmaceutical Care Network Europe （PCNE） classification system to develop medication therapy management （MTM）， and to investigate the application of PCNE classification system in solving drug-related problems （DRPs） in type 2 diabetic patients and the effect of it on clinical outcomes. METHODS： The patients with type 2 diabetes diagnosed in endocrinology department of our hospital from Jul. 10， 2018 to Oct. 31， 2018 were randomly divided into clinical pharmacist-led intervention （“physician-pharmacist-nurse” mode） group and control group receiving only traditional medical services （“physician-nurse” mode）. According to PCNE classification， the number of DRPs found in the pharmaceutical intervention group， the types of problems， causes， the types of interventions， acceptance for interventions and outcomes were analyzed and evaluated. Drug compliance （the highest score is 8） and HbA1c compliance （＜7％） were compared between 2 groups during hospitalization （or at the discharge） and 3 months after discharge. RESULTS： Totally 76 cases were included （40 cases in pharmaceutical intervention group and 36 cases in control group）. During hospitalization， 51 DRPs were found in the pharmaceutical intervention group， among which 42 problems were related to the effectiveness of treatment， mainly due to improper usage and dosage （23 problems）； the types of intervention was mainly aimed at the patient level （24 problems）. 38 problems received intervention （acceptance rate was 74.51％） and 32 problems （62.75％） were completely solved. Compared with those at admission， after following up for 3 months patients with low score （6 points） in the drug compliance of the pharmaceutical intervention group decreased from 26 to 8 （P＜0.000 1）， patients with medium score （6-8 points） increased from 10 to 22 （P＝0.006 2）， patients with high score （8 points） increased from 4 to 10， and drug compliance improved significantly， while there was no significant change in drug compliance in the control group. Compared with those at the discharge， after 3 months’ follow-up， the HbA1c compliance rate of the pharmaceutical intervention group increased from 25.00％ to 77.50％， and that of the control group increased from 25.00％ to 55.56％. There were statistical differences （P＜0.000 1）， and HbA1c compliance rate of the pharmaceutical intervention group was significantly higher than that of the control group. CONCLUSIONS： In the practice of MTM service， clinical pharmacists use PCNE classification system to collect， analyze， intervene， solve and evaluate DRPs systematically. The service mode can provide reference for standardizing pharmaceutical care mode.