ABC | Volume 113, Nº4, October 2019

Original Article Santiago et al. Prevalence of hypertension and associated factors Arq Bras Cardiol. 2019; 113(4):687-695 of hypertension and determine its associated factors in the adult population from the semi-arid region of the state of Pernambuco, Northeastern Brazil. Methods This is a population-based cross-sectional study conducted with male and female adults (aged 20 to 59 years) residing in the semi-arid region of Pernambuco. The study population was determined by cluster sampling. Pernambuco is subdivided into 12 development regions (DR), six of which correspond to the semi-arid zone. Among them, three were randomly selected in the first stage of the sampling process. Next, one city was chosen from each DR: Serra Talhada (DR 4), Custódia (DR 12), and Belém de São Francisco (DR 1). Next, five census tracts were drawn per city with urban/rural distribution based on data from the 2010 Census. Lastly, 350 households were randomly selected to form a representative sample of the population in the semi-arid region of Pernambuco. The sample comprised all adult residents of the selected homes who were present at the time of data collection. Individuals with any physical limitation that hindered the anthropometric evaluation, debilitating diseases, and who declined to participate were excluded from the study. All stages of the selection process were performed using lists of random numbers generated with the aid of the EPITABLE tool of the Epi Info statistical package, version 6.04 (CDC/WHO, Atlanta, GE, USA). The fieldwork occurred between July and September 2015 by a team of researchers who had previously undergone training for the administration of data collection instruments. A pilot study was conducted with 30 families in a city not selected for the main study to put into practice the logistics of the fieldwork and test the data collection instruments. The sample size was calculated a posteriori assuming a 20% estimated prevalence of hypertension in the northeast region of Brazil, 8 a 5% sampling error, a 95% confidence interval, and a factor of 1.5 to compensate for the design effect of the cluster sampling. Moreover, 10% was added to compensate for possible dropouts, leading to a total of 410 individuals. The following demographic and socioeconomic characteristics and respective categories were collected: gender (male or female), age in years (20 to 29, 30 to 39, 40 to 49, 50 to 59), ethnicity (white or multiracial/black), schooling (never studied, primary school, high school/ university), employment status (works or does not work), and place of residence (urban or rural area). Data collection followed the guidelines of the Instituto Brasileiro de Geografia e Estatística (IBGE – Brazilian Institute of Geography and Statistics). 12 Economic class was categorized based on the Brazilian Economic Classification Criteria of the Associação Brasileira de Empresas de Pesquisa (ABEP – Brazilian Market Research Association): 13 upper/middle (A1, A2, B1, B2, C1, C2) and lower (D, E) classes. We collected the following behavioral characteristics: alcohol intake in the previous 30 days (yes or no); active smoking (smoker/ex-smoker or never smoked); passive smoking (yes or no; individuals who do not actively smoke but are frequently in contact with cigarette smoke from people at home, work, or school/university); and addition of salt to food after preparation (never, sometimes/almost always). Physical activity level was determined using the International Physical Activity Questionnaire (IPAQ) validated for use in Brazil, 14 which enables the classification of individuals as sedentary/ insufficiently active or active/very active. 14 Anthropometric data were collected in duplicate following the guidelines of the World Health Organization. 15 Body mass was measured using a digital scale (TANITA™, model BF‑683W). Height was measured using a portable stadiometer (Alturaexata™). Waist circumference (WC) was measured at the midpoint between the last rib and the iliac crest with a flexible, non-elastic metric tape (Sanny™). When a difference greater than 0.5 cmwas found between the two height andWC values, the participant was measured a third time, and the two closest results were considered to calculate the arithmetic mean. The following health and nutritional characteristics and respective categories were collected: body mass index (BMI) (not overweight when < 25 kg/m² and overweight when ≥ 25 kg/m²); 16 WC (normal when < 80 for women and < 94 cm for men and increased when ≥ 80 cm for women and ≥ 94 cm for men); 16 waist-to-height ratio (normal when < 0.52 for men and < 0.53 for women and increased when ≥ 0.52 for men and ≥ 0.53 for women); 17 and food security evaluated using the Brazilian Food and Nutritional Insecurity Scale, 18 which enabled classifying the homes into the following categories: food security, mild food insecurity and moderate/ severe food insecurity. Blood samples were collected through a venous puncture after a 10-hour fast. The analyses to determine the levels of fasting blood glucose, triglycerides, and total cholesterol used the Accutrend GCT [Roche Diagnóstica, Brazil], which allows immediate readings. The components of the biochemical profile were fasting blood glucose [normal when < 100 mg/dL and reduced glucose tolerance/diabetes mellitus (DM) when ≥ 100 mg/dL or when the individual used a hypoglycemic medication], 19 triglycerides (normal when < 150 mg/dL and high when ≥ 150 mg/dL), 20 and total cholesterol (normal when < 190 mg/dL and high when ≥ 190 mg/dL). 20 Regarding the outcome variable, blood pressure (BP) was measured in duplicate using the auscultation method (Glicomed™ sphygmomanometer, model CE-0483), followed by the calculation of the arithmetic mean of the results. The procedures to prepare the individuals for BP measurement followed the recommendations of the Brazilian Society of Cardiology: 6 make sure that the individual rested for at least five minutes in a calm environment; did not have a full bladder, had not practiced physical exercise in the previous 60 minutes, had not consumed alcohol, coffee, or food in the previous hour, and had not smoked in the previous 30 minutes; and was seated at the time of the measurement, with the legs uncrossed, feet flat on the floor, and arm at the height of the heart. The criteria to diagnose hypertension was based on the Seventh Brazilian Hypertension Guidelines, 6 which classify an individual with hypertension when the systolic BP is ≥ 140 and/or diastolic BP is ≥ 90. We also considered hypertensive individuals who declared having a previous diagnosis and were under treatment with antihypertensive medications. 688