Validação de instrumento de determinação da adesão farmacoterapêutica para hipertensão arterial

Abstract

Background: Non-adherence to treatment is an important risk factor for therapeutic failure, complications and increased costs in chronic diseases, such as hypertension. Among the proposed methods for determining adherent behavior, self-report adherence scales are widely used in research and clinical practice because they have low cost, easy applicability and consider subjective aspects of the individual. Objective: To perform the cross-cultural adaptation and validation into Portuguese language of the Simplified Medication Adherence Questionnaire (SMAQ) and Brief Evaluation of Medication Influences and Beliefs (BEMIB) therapeutic adherence scales for patients with hypertension. Methodology: International methodological recommendations for cross-cultural adaptation were followed. The semantic equivalence was attributed by the authors of the original scales. The pretest was conducted as a pilot with 20 patients from the target population. To certify reliability and validity, a multicenter cross-sectional study was conducted in five outpatient care units in Maceio/AL and Aracaju/SE, between January and July 2019. 117 patients over 18 years old were recruited using antihypertensive drugs. Internal consistency (Cronbach's alpha) was tested as a reliability parameter. Criterion and construct validity were verified by concurrent validation and validation by known groups. Patients were classified as “adherent” and non-adherent” as recommended by each measure. Results: The participants had a mean age of 56.6 years (SD = 10.7 years), mostly female (72.6%) and selfdeclared brown color/race (60.7%). Regarding education, 73 patients did not reach high school, 12% of whom were illiterate. The average number of antihypertensive drugs prescribed per patient was 1.87 (SD = 0.87). Regarding the adherence profile attributed by each scale, 79.5% (n = 86) of the patients were considered non-adherent by the SMAQ; by BEMIB, 80.3% (n = 94) and by MGL, 64.1% (n = 75). Satisfactory internal consistency was found for SMAQ (Cronbach's alpha = 0.63) and insufficient for BEMIB (Cronbach's alpha = 0.22). Satisfactory correlation coefficient was found for SMAQ and the Morisky-GreenLevine test as external criterion (r = 0.56, p <0.001). No correlation was observed between BEMIM and the criterion (r = 0.07, p = 0.40). There was no significant correlation of scales with blood pressure control. The sensitivity of SMAQ was 75.3% and BEMIB 78.0%; specificity was 29.5% for SMAQ and 15.9% for BEMIB; Positive predictive value was 63.9% for SMAQ and 60.6% for BEMIB and negative predictive value was 41.9% for SMAQ and 30.4% for BEMIB. The average social desirability was 21.9 (SD = 4.08). Conclusion: The SMAQ version proved to be valid and reliable for determining adherence to pharmacotherapy in patients with hypertension, an unconfirmed result for BEMIB. The scales showed more ability to detect non-adherent patients, but with low specificity, possibly influenced by the high social desirability.