24h movement behavior and metabolic syndrome (24h-mesyn) risk score: uma ferramenta para autoavaliação do risco de síndrome metabólica para universitários em regiões de baixa renda

Abstract

The objective of this study was to develop and validate a nomogram model for predicting individual risk of developing metabolic syndrome in low-income regions, based on user-friendly self-reported lifestyle and anthropometric information. The current study was conducted with data corresponding to the baseline of the prospective 24-hour movement behavior and metabolic syndrome cohort. We used data from 496 university students in two Brazilian cities located in the country's poorest states, with Gini indices of 0.56 and 0.43 for Imperatriz and Miracema do Tocantins, respectively. We adopted as potential predictors: demographic variables (biological sex, age and race/ethnicity) and economic variables (family income, student profession), school variables (during primary and secondary education), 24-hour movement behavior (physical activity, behavior sedentary and sleep) and body weight. Based on the predictors for abdominal obesity identified via multilevel logistic regression, we created nomogram models according to the cumulative level of information (school, behavioral, and body weight levels). We used the area under the receiver operating characteristic curve, AUROC, with a reference value of 0.70, to evaluate the discrimination ability of the predictions; the Hosmer-Lemeshow test for model calibration; and bootstrapping (B = 150) for internal validity. And, we assessed the external validity of the model in a subsample of 375 students with metabolic syndrome data. We identified 114 (23.0%) students who were abdominally obese. We found at least ten variables associated with the primary outcome: score A) biological sex, age, physical structure for Physical Education classes, participation in sports competitions (during elementary school); failure, Physical Education as a preferred curricular component, Physical Education classes per week; participation in sports training (during high school); score B) adherence to movement behaviors within 24 hours; C score) and body weight. All three proposed nomograms showed acceptable performances (score A, AUROC: 0.70 [95% CI: 0.64-0.75]; score B, AUROC: 0.70 [95% CI: 0.65-0. 76]; and C score: 0.94 [95% CI: 0.92-0.96)]. Calibration assessment showed reasonable consistency (p > 0.05) in our models. In internal validation, we observed an increase in the predictive capacity for scores A (AUROC = 0.71) and B (AUROC = 0.71), and a decrease for score C (AUROC = 0.86) for predicting metabolic syndrome. Thus, the 24h-MESYN risk scores serve as valid tools for self-assessment of the risk of developing metabolic syndrome among young people from low-income regions.