ABC | Volume 110, Nº5, May 2018

Original Article Ramires et al Prevalence of metabolic syndrome in Brazil Arq Bras Cardiol. 2018; 110(5):455-466 In this perspective, based on 2013 NHS data, this study objective was to estimate MS prevalence and its components on Brazilian population aged ≥ 18 years and its association with sociodemographic, behavioral and biological variables. Methods Design and sample of the study NHS is a household-based cross-sectional survey with representativeness of Brazilian adult population conducted between August and December 2013. NHS was a population survey on health and its determinants, carried out in Brazil by the Brazilian Institute of Geography and Statistics (IBGE) allied to the Ministry of Health. 9 The sampling process was organized by clusters in three stages, where the primary sampling unit (PSU) was composed by census sectors, secondary unit was households and tertiary unit, inhabitants 18 years old or older. Within each stage, participants were selected through simple random sampling. It was considered the corresponding households weight, the dweller selection probability, non‑response adjustments by sex and calibration by population totals, by sex and estimated age classes, with all the dwellers weight. A detailed methodology applied description for NHS-2013 was previously published. 9 To describe Brazilian population health conditions, NHS was constituted by thematic modules that addressed health and lifestyle individual perception, chronic diseases presence, as well as sociodemographic information. For the study purposes, individuals with data on the factors that make up MS were selected, that is, diabetes mellitus and hypercholesterolemia self-referred medical diagnosis, in addition to blood pressure (BP) values and waist circumference measurements (WC). Among the 69.954 domiciles occupied with resident selected for NHS-2013 interview, 60,202 individuals 18 years old or older were interviewed, representing response rate of selected residents of 86%. 9 We excluded 800 pregnant women from the total, generating a final sample composed of 59.402 individuals. The National Commission of Ethics in Research (CONEP) approved the NHS project in June 2013, under Opinion N°. 328,159. All the interviewees who agreed to participate in the investigation signed the term of free and informed consent. 9 Outcome variable In this study, MS outcome variable was defined according to the international cardiology consensus harmonization proposal (IDF / NHLBI / AHA / WHF / IAS / IASO), characterized by the presence of three out of five metabolic risk factors. 4 In our work, once NHS did not provide biochemical data, MS classification was made considering at least the presence of three of four situations available in database: 1) self-reported diabetes diagnosis; 2) BP values considered borderline for Systemic Hypertension diagnosis (HAS) (systolic ≥130 mmHg and/or diastolic ≥ 85 mmHg); 3) WC values above the cut-off point established as abdominal obesity threshold for South America population (men ≥ 90 cm and women ≥ 80 cm); and, 4) self-reported hypercholesterolemia diagnosis. A trained team using a calibrated digital device evaluated BP. Individuals needed to be at rest and were oriented to empty their bladders, not smoke or drink during the 30-minute period prior to measurement and not perform any physical activity during the period of one hour before the measurement. BP measurements were made with individuals in sitting position, after having rested for at least five minutes. The individuals were instructed to stay relaxed and leaning against the back of the chair, not crossing their legs and leaving their left arm free of clothes and leaning on a table at the same level as their chest or heart. Three BP measurements were evaluated, with two minutes intervals between them. For the present study, the mean between the second and the third measures was used. In this technique, systolic and diastolic pressures were calculated by algorithms from the maximum oscillation point corresponding to average BP. 9 As for anthropometric measurements, weight (kg), height (m) and WC (cm) were measured - having as reference for the perimeter the midpoint between last rib and iliac crest, being used, respectively, portable electronic scale, portable stadiometer, flexible and inelastic measuring tape with 0.1 cm precision. The procedures for anthropometry realization followed the same protocol used in the Brazilian Institute of Geography and Statistics (IBGE) 2008-2009 Family Budget Survey (FBS). 9 Independent variables Independent variables selection was performed based on distal and proximal determinants conceptual model, developed to show sociodemographic, behavioral and comorbid multiple factors impact for MS on population health status. 10 Sociodemographic factors used included: sex, age (18‑59 years and ≥ 60 years), schooling (≤ 8 years and > 8 years), skin color (white / non-white), marital status (living with partner or not), macro-region (South / Southeast /Center-West and North/Northeast). Macro regions were dichotomized in order to contrast Brazilian development poles, with the South, Southeast and Central West regions being the most developed. Behavioral variables were self‑perceived health, considering the combination of "very good" and "good" responses, defining as reference "regular", "poor" and "very poor" responses. For variable physical activity construction (PA), the following information were considered regarding free time and volume, effort duration and intensity for physical activity, using three months preceding the day of the interview as reference period for PAquestions in the questionnaire. Effort intensitywas obtained by conversion of physical exercise type or sport reported in vigorous or moderate PA as determined by the Physical Activity Compendium (PAC). 11 Effort duration in each session was expressed in minutes and divided into three groups: < 19; 20 to 29; > 30 minutes. Weekly frequency was determined by PA practice day number per week and, for analysis purpose, was divided into 0 to 2 days, 3 to 4 days and > 5 days per week. 456