The Beliefs about Medicines Questionnaire (BMQ) allows the assessment of cognitive representations encompassing patients' medication beliefs about taking medication for their disease in different cultures. To determine the construct validity and reliability of the BMQ questionnaire adapted to Colombian hypertensive patients. An instrumental psychometric study was carried out in a sample of 238 hypertensive patients with an average age of 65 years (SD= 11.4), with a predominance of women (70%). Construct validity was assessed by exploratory and confirmatory factor analysis. Reliability was calculated using Cronbach's alpha coefficient method. A reduced version of 16 items was obtained; in the BMQ-General section the 7 items were grouped into two factors that explained 64% of the common variance and good fit (χ2= 61.46; gl = 13; p = 0.000; CFI = 0.917; NNFI = 0.89; CFI=0.917; SRMR=0.054; RMSEA = 0.125; IC 90% [0,10, 0,16]). In the BMQ-Specific the 9 items grouped into two factors explained 63.17% of the common variance with an acceptable fit (χ2= 122.4; gl = 26; p = 0.000; CFI = 0.88; NNFI = 0.84; CFI=0.88; SRMR=0.106; RMSEA = 0.125; IC 90% [0.10, 0.15]). The Cronbach's alpha reliability for the BMQ-General and Specific was 0.82 and 0.78, respectively. The version of the BMQ adapted to Colombian hypertensive patients has adequate psychometric characteristics and its use is recommended in research.