ABC | Volume 110, Nº1, January 2018

Original Article Pitanga et al Physical activity and cardiovascular risk factors Arq Bras Cardiol. 2018; 110(1):36-43 The purpose of this paper was to verify the association between LTPA and/or CPA with different cardiovascular risk scores in the cohort from Longitudinal Study of Adult Health (ELSA-Brasil). Methods Population and sample ELSA-Brasil is a cohort study of 15,105 economically active or retired people of both genders, aged 35-74, from six teaching and research institutions in the cities of Salvador, Vitória, Belo Horizonte, Rio de Janeiro, São Paulo, and Porto Alegre, whose methodological details have been previously described. 18,19 For the present study, all baseline participants (2008-2010) who answered the questionnaires about PA were selected, as long as they had the information required to calculate cardiovascular risk scores. After excluding participants who reported previous myocardial infarction, stroke, peripheral vascular disease, and heart failure, the sample was formed with 13,721 participants (45.3% males, 54.7% females). ELSA-Brasil was approved by the National Commission for Research Ethics (CONEP) and by all Ethics Committees of the research centers involved. All participants signed the informed consent form, assuring secrecy and confidentiality to data. Data production Data were collected by a team of interviewers and trained evaluators, all of them certified by a quality control committee 19 and able to carry out the study protocol at the ELSA-Brasil Research Center. Face-to-face interviews were conducted with standardized and previously validated questionnaires. Evaluation of physical activity The International Physical Activity Questionnaire (IPAQ) was applied to identify and quantify PA, consisting of questions about the frequency and duration of physical activities at work (moderate and vigorous walking), while commuting, in domestic activities, and in leisure time. 20 ELSA-Brasil only addressed leisure time and commuting activities. PA was measured in minutes per week by multiplying weekly frequency by each event’s duration of each. For the purpose of this study, participants were classified as to leisure-time activities as follows: • sedentary (< 10 min/week, any PA); • (≥ 10 min to < 150 min/week of walking, moderate PA and/or 10 min to < 60 min/week of vigorous PA and/or 10 min to < 150 min/week of any combination of walking, moderate and vigorous PA); • physically active (≥ 150 min/week of walking , moderate PA and/or ≥ 60 min/week of vigorous PA and/or ≥ 150 min/week of any combination of walking, moderate and vigorous PA); • very active (≥ 150 min/week of vigorous PA, or ≥ 60 min/week of vigorous PA plus 150 min/week of any combination of walking and moderate PA). For dichotomized analyzes, participants sorted as sedentary and not very active were considered insufficiently active, and active participants were those sorted as physically active and very active. Commuting PA was categorized as insufficiently active (< 150 min/week of walking and/or cycling) and physically active (150 min/week of walking and/or cycling). Evaluation of cardiovascular risk Five cardiovascular risk scores were used. Two of them were proposed by Wilson et al. 8 and aimed to estimate the risk of coronary artery disease. Variables used were: age, systolic and diastolic blood pressure (BP), high-density lipoprotein (HDL-C), diabetes, smoking, and total cholesterol in the first; and age, systolic and diastolic BP, HDL-C, diabetes, smoking and low‑density lipoprotein (LDL-C) in the second. The third and fourth scores, proposed by D'Agostino et al., 9 aimed to identify patients at high risk for any initial atherosclerotic event (coronary heart disease, cerebrovascular diseases, peripheral vascular disease, and heart failure), using following variables: age, treated and untreated systolic and diastolic BP, HDL-C, body mass index (BMI), diabetes, smoking, and total cholesterol in the third; and age, treated and untreated systolic and diastolic BP, HDL-C, diabetes, smoking, and BMI in the fourth. The fifth score, indicated by ACC and AHA, 10 aimed at estimating the risk for atherosclerotic diseases. The variables used were: age, treated and non‑treated systolic BP, total cholesterol, HDL-C, smoking, and diabetes. All cardiovascular risk scores were calculated for ELSA-Brasil participants, with detailed scoring scheme previously reported. 8-10 Participants with scores ≥ 20% were considered at high risk for future cardiovascular events. 21 Evaluation of covariables BP was obtained with a validated oscillometric device (Omron HEM-705CPINT) after a five-minute rest, with the subject sitting in a quiet and temperature-controlled room (20-24°C). Three measurements were taken at 1-min intervals each. The mean of the last two BP measurements was calculated and used in our analysis. Definition of diabetes was based on self-reported information and laboratory exams. Patients were considered to have been diagnosed if they had been previously informed by a physician that they had diabetes or if they had used medication for diabetes in the last two weeks. Patients not previously diagnosed with diabetes were classified as having diabetes when fasting plasma glucose level was ≥ 7.0 mmol/L, two‑hour post-load glucose was ≥ 11.1 mmol/L, or glycated hemoglobin (HbA1c) was ≥ 6.5%. 22,23 Participants were sorted as hypertensive if systolic blood pressure (SBP) was ≥ 140 mmHg, diastolic blood pressure (DBP) was ≥ 90 mmHg or if they had taken any medication to treat hypertension in the last two weeks. Total cholesterol and HDL-C were determined by the enzymatic colorimetric method. LDL-C was calculated by the Friedewald equation. 37