ABC | Volume 112, Nº6, June 2019

Original Article Scheer et al Exercise at school and cardiovascular risk factors Arq Bras Cardiol. 2019; 112(6):775-781 Regarding our sensitivity analysis in the logistic regression model, it is important to note that the magnitude of association between the school policy and being overweight increased for girls (OR: from 1.49 [1.13–1.98] to 1.89 [1.13–2.75]) and decreased for boys (OR: from 1.49 [1.13–1.98] to 1.09 [0.71‑1.69]), when compared to that of the fully adjustedmodel. With respect to the association between the school policy and pre-hypertension/hypertension, the magnitude of association slightly decreased for girls (OR: from 1.86 [1.36–2.54] to 1.66 [1.10–2.51]) and increased for boys (OR: from1.86 [1.36–2.54] to 2.20 [1.37–3.54]), with no difference in lipid profile. 29 Despite the imbalance between the sexes, which motivated the sensitivity analysis, there may be a rationale to explain the changes in the magnitude of the point-estimation. Previous findings support that, overall, girls are more likely to be overweight than boys – perhaps because boys are more physically active and less sedentary. 30 Nevertheless, data from NHANES revealed that the prevalence of risk factors for CVD and metabolic syndrome was higher in boys than in girls, 31 which is in line with he statement that boys are more likely to be pre-hypertensive/hypertensive than girls. 32 This apparently counterintuitive fact could be supported by the present experiment. Among girls, 606 were overweight, and 603 were diagnosed with pre- or hypertension. Among boys, 404 were overweight and 502 were pre-hypertensive or hypertensive. These numbers indicate that other risk factors, such as family history, may be associated with elevated BP – since at least 98 boys had this condition despite the absence of overweight. Therefore, the interactions between school environment and sex are not sufficient to explain the differences in both overweight and pre-hypertension/hypertension observed in the fully adjusted model and in the model stratified by sex. However, this analysis supports the fact that sex may explain the prevalence of overweight and pre-hypertension/ hypertension in adolescents regardless of PA status. Limitations and future directions The results of this study need to be interpreted in light of some limitations. First, because of the cross-sectional design, we cannot establish a causal relationship between school environment (i.e., school PA policy) and the development of CVD risk factors. Second, we did not measure the level of PA of the participants to verify whether they met the PA recommendations of the school policy – this is why we chose school, instead of PA level, as the exposure. Third, no adjustment was made for socioeconomic status, parental influence, or dietary intake, which might have some influence on the development of CVD risk factors. Finally, the categorization of some continuous variables for analysis cannot be disregarded as a limitation, even though we used thresholds commonly reported in the literature. On the other hand, the hypothesis raised by our study needs to be addressed in intervention-based studies, like multicenter, cluster randomized controlled trials. We addressed, with a valuable sample size, the prevalence of some CVD risk factors in the school environment of a middle-income country in a chronological age-homogeneous sample. The use of higher PA doses, like that of the OEG policy, need to be tested and confirmed in future experiments. Conclusion The prevalence of pre-hypertension/hypertension and overweight differedbetween schools, and students attendingOEG had lower proportion and odds of developing CVD risk factors. Author contributions Conception and design of the research and Acquisition of data: Scheer C, Belém LJ, Fabiano LCC, Pinheiro LT; Analysis and interpretation of the data: Scheer C, Helal L, Belém LJ, Fabiano LCC, Pinheiro LT, Stein R; Statistical analysis: Scheer C, Helal L, Belém LJ, Fabiano LCC, Pinheiro LT; Obtaining financing: Scheer C, Stein R; Writing of the manuscript and Critical revision of the manuscript for intellectual content: Scheer C, Ferrari F, Helal L, Stein R. Potential Conflict of Interest No potential conflict of interest relevant to this article was reported. Sources of Funding This study was funded by Fundação Pró-Coração and partially funded by Fundo de Incentivo à Pesquisa e Eventos (FIPE/HCPA). Study Association This study is not associatedwith any thesis or dissertationwork. Ethics approval and consent to participate This study was approved by the Ethics Committee of the Instituto Brasileiro de Cardiologia under the protocol number CAAE 14549513.1.0000.5272. All the procedures in this study were in accordance with the 1975 Helsinki Declaration, updated in 2013. Informed consent was obtained from all participants included in the study. 779