OBJECTIVES: Approximately 65% of patients diagnosed with hypertension are not well controlled and two third of patients need to be treated by two or more drugs for achieving target blood pressure. The aim of this study was to describe antihypertensive treatment patterns in real life settings by using Markov chain model METHODS: Data concerning prescriptions of patients with diagnosed hypertension were obtained from a large primary care survey conducted in 2012 using CSD Longitudinal Patient Database. Patients selected were treated with sartans, prescribed alone or in combination with amlodipine and/or hydrochlorothiazide (HCT), either in free associations or in fixed combinations. A Markov chain model M1 with 4-states (A: single sartan, B: sartan+ amlodipine, C: sartan + HCT, D: sartan + amlodipine + HCT) was proposed to model transitions from one treatment to another over time. A second chain, M2, with 6 states was also studied, in which dual-therapies were divided into two states depending on whether the combination was free or fixed. Age and sex were included as covariates. R packages and MSM DIAGRAM were used. RESULTS: