IJCS | Volume 33, Nº3, May / June 2020

273 Espeche et al. Adherence to antihypertensive therapy in Argentina Int J Cardiovasc Sci. 2020; 33(3):272-277 Original Article Medication adherence is defined by the World Health Organization (WHO) as “the degree to which a person’s behavior corresponds with the agreed recommendations from a health care provider.” 10 The WHO estimates the prevalence of non-adherence to antihypertensive medication to be 30-50%, depending on differences in drug-class, type of prevention and methods used to measure adherence. 6,7,11 In Argentina, a study based on a non-population sample showed a rate of adherence of῀26%. 12 Remarkably, non-adherence was associated with higher cardiovascular risk. 13 Factors that influence adherence could be classified into five dimensions: 1. patient-related factors (inadequate beliefs or skills), 2. socioeconomic-related factors (poor health literacy or low social support), 3. condition- related factors (presence of comorbidities), 4. therapy- related factors (complex drug regimen) and 5. health system/health-care team-related factors (inadequate communication with health-care provider). 2 There are direct and indirect methods to assess adherence. While direct methods have the advantage of having greater accuracy, the cost, availability, and accessibility of these methods make their use unlikely in current practice. Conversely, indirect methods, such as the Morisky-Green test, are easy to use in daily medical practice (Table 3). 14 Thus, this test could be used to evaluate adherence in real-world settings. The aim of this study was to determine the prevalence of adherence in hypertensive patients treated by physicians in several cities in Argentina. Methods A multicenter cross-sectional study was conducted in eight cities of Argentina (Tandil, La Plata, El Calafate, Santiago del Estero, San Miguel de Tucumán, Rosario, Misiones and Buenos Aires) between March andAugust 2018 using a prospectively designed protocol. Each city was represented by a single centre. These cities are the capitals or the most important cities of six provinces (BuenosAires, Santiago del Estero, Santa Fe, Santa Cruz, Tucuman andMisiones), in the North, South and Central regions of the country. The studywas conducted on consecutive hypertensive patients seen in the general practice office who had been under pharmacological treatment for at least six months. Blood pressure (BP) measurements were performed by the physician in a single visit (regardless of the purpose of the visit), using an OMRON HEM 705 CP device (OMRON HEALTHCARE Co., Kyoto, Japan). Two measures were taken, with a one-minute interval between measurements, and the mean of the measures was defined as office blood pressure. Body weight was determined with subjects wearing light clothes and no shoes. Height was measured without shoes using a metallic tape, and body mass index (BMI) was calculated. In addition, history of dyslipidemia, diabetes, cardiovascular diseases, and smoking was recorded. The level of adherence was assessed using theMorisky questionnaire. 14 A non-adherent patient was defined as a patient who answered positively to any one of the questions. Controlled hypertensive patients were defined as those individuals whose BP was < 140-90 mmHg. The type and number of antihypertensive drugs and the use of fixed-dose combinations were recorded. The patients willing to participate signed an informed consent form. Individuals were divided into controlled and uncontrolledhypertension using the traditional definition 4 and in “non-adherent” and “adherent” according to the Morisky test. Statistical analysis The study sample had a normal distribution determined by Test Shapiro Wilk. Baseline continuous variables (age, BMI, systolic BP, and diastolic BP) were expressed as mean ± standard deviation (SD) and were compared using an independent t-test. Categorical variables (sex, current smokers, adherence, diabetes, dyslipidemia, hypertension control, and previous cardiovascular event) were expressed as percentages and compared using the χ 2 test. To identify the variables independently associated with non-adherence, a forward stepwise binary regression logistic model was performed and the results expressed as odds ratio (OR) with 95% of confidence interval (95% CI). All tests were two-tailed, and P values < 0.05 were considered statistically significant. All statistical analyses were performed using SPSS 18.0. Results A total of 862 individuals aged 61 ± 14 years of age, 53% women, from the eight cities (nearly 100 individuals per center) were included. Most patients (79,1%) had health insurance and similar socioeconomic