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 Table 5 – Bivariate analysis and multivariable models for factors associated with metabolic syndrome among Brazilian men according to hierarchical levels of exposure variables studied - National Health Survey (NHS), Brazil,2013 Variables Bivariate analysis Model 1 Model 2 Model 3 PR CI 99% PR CI 99% PR CI 99% PR CI 99% Age 18-59 years 1 1 1 1 ≥ 60 years 3,23 2,55- 4,08 2,74 2,17- 3,46 2,07 1,61- 2,67 2,60 2,04- 3,31 Education > 8 years 1 ≤ 8 years 1,24 0,99- 1,54 Conjugal situation Do not live with a partner 1 1 1 1 Live with a partner 2,30 1,82- 2,92 1,81 1,43- 2,30 1,74 1,37- 2,20 1,48 1,17- 1,88 Housing region NE/N 1 1 1 1 SE/S/CW 1,63 1,31- 2,02 1,49 1,20- 1,85 1,74 1,40- 2,14 1,57 1,28- 1,94 Skin color White 1 Non-white† 0,67 0,54- 0,84 Physical activity Active 1 Inactive 3,02 0,84- 10,90 Health self-perception Very good/good 1 1 1 Regular – Very bad 3,17 2,51- 4-01 2,72 2,12- 3,50 2,59 2,01- 3,33 Stroke No 1 Yes 3,15 1,99- 4,98 Another CVD‡ No 1 Yes 3,17 2,29- 4,39 Overweight No 1 1 Yes 3,27 2,45- 4,37 3,58 2,73- 4,70 Depression No 1 1 Yes 2,19 1,49- 3,23 1,41 0,98- 2,02 Chronic renal failure No 1 1 Yes 3,57 2,41- 5,28 1,85 1,23- 2,76 PR: prevalence ratio; CI 99%: confidence interval 99%; N: norte; NE: northeast; SE: southeast; S: south; CW: central-west; CVD: cardiovascular disease; (*) Yellow, indigenous, brown, black; (‡) infarction, angina, heart failure or other. 462