OBJECTIVES: The sequence mining techniques can effectively explore the relation between medicine and outcome along the temporal dimension. The first objective is to applied the temporal association rules to the hypertension patient database and check the patterns for current practice and consequent outcome. The second objective is to extract the rules generate from history record and check the evidence for doctor decision support and for cost-effective analysis. METHODS: First all the records of patients who were diagnosed with hypertension were extracted from database from a special built CDM database in hospital. Then all patients were assigned with an ID and all the interested measurements (medicine, diagnosis, blood pressure records, etc.) were arranged in sequence with related time stamp. The sequence mining technique (SPADE, Sequential Pattern Discovery using Equivalence classes) was applied to the hypertension patients and identify the frequent temporal medication patterns. The medication patterns were further investigated with the corresponding outcome measure and the medicine cost. RESULTS: