Abstract
OBJECTIVE: To validate the General Medication Adherence Scale (GMAS) in Brazilian Portuguese for hypertensive patients. METHODS: The GMAS-English was translated into Brazilian Portuguese and adapted for cultural appropriateness by a translation process and expert panel. A cross-sectional study was conducted in northeast Brazilian cardiology divisions of public and private hospitals, interviewing hypertensive patients. Reliability was assessed using Cronbach's alpha, intraclass correlation, and Pearson's correlation. Convergent validity was tested against the BMQ using chi-square. Criterion validity was assessed by comparing GMAS with blood pressure control using chi-square. RESULTS: The GMAS was translated and adapted according to standard procedures. In a validation study with 167 hypertensive patients, Cronbach's alpha was 0.79, and Pearson's correlation showed significant test-retest reliability (p < 0.001). Convergent validity with BMQ was significant (p < 0.001), with 89.4 % sensitivity for behaviors considered adherent (High adherence and good adherence), but between the strata that measure low adherence (Partial adherence, low adherence and very low adherence), the specificity rate was 50 %. Criterion validity between GMAS and blood pressure control was not observed. CONCLUSION: The Brazilian Portuguese version of the GMAS exhibited good consistency and reproducibility, modest agreement with BMQ scale and did not demonstrate acceptable criterion validity for hypertensive patients.