Prevalence of adherence to antihypertensive treatment: a comparison of drug biochemical analysis and the self-report scale

Background: Assessing adherence to antihypertensive medication remains a major challenge for healthcare professionals. The objective of this study was to compare the evaluation of adherence to antihypertensive medication between direct and indirect methods, and to identify variables associated with nonadherence. Methods: < 0.20 in the bivariate analysis were included in the multiple logistic regression model. The significance level was set at 5%. Cohen's Kappa coefficient (CKC) was used to assess agreement between the methods. Results: The sample consisted predominantly of women (61.7%), white individuals (63.2%), and married participants (52.8%), with a mean age of 65 ± 13.3 years and median education of 12.0 (7-12) years. Participants were using an average of 3.2 ± 1.3 antihypertensive drugs, and 69.2% had controlled blood pressure (BP). The prevalence of adherence according to the direct method was 32.4%, while 90.1% were adherent according to the indirect method. Agreement between the direct and indirect methods was low (CKC = 0.014). Variables independently associated with nonadherence according to the direct method were regular physical activity (odds ratio [OR], 0.370; 95% confidence interval [CI], 0.158-0.853), proteinuria (OR, 2.406; 95% CI, 1.066-5.751), number of antihypertensive medications (OR, 1.570; 95% CI, 1.127-2.225), and systolic BP (OR, 1.051; 95% CI, 1.010-1.097). Indirect adherence assessment was associated only with the presence of common mental disorders (OR, 1.163; 95% CI, 1.035-1.307). Conclusions: The prevalence of adherence to antihypertensive drug treatment assessed by the indirect method was substantially higher than that observed with the direct method. The results were inconsistent between the 2 approaches. Moreover, the direct method identified a broader range of variables associated with nonadherence.