Associação entre estilo de vida e tempo de exposição à TV: uma abordagem baseada em machine learning e modelos de regressão
DOI:
https://doi.org/10.14571/brajets.v18.n4.1338-1348Palavras-chave:
comportamento sedentárioio, tempo de tela, hábitos alimentares, modelagem estatísticaResumo
O tempo excessivo em frente à televisão tem sido associado ao comportamento sedentário e a riscos metabólicos, sendo influenciado por fatores como nível de atividade física, hábitos alimentares e características socioeconômicas. Este estudo empregou técnicas de aprendizado de máquina e modelagem estatística para identificar os principais preditores do tempo semanal de exposição a TV entre escolares do ensino médio. Foram analisadas variáveis categóricas, incluindo sexo, nível de atividade física, consumo alimentar e fatores socioeconômicos, utilizando o software R. A seleção de variáveis foi realizada por meio dos métodos Random Forest e Recursive Feature Elimination, seguida de análises estatísticas com regressão de Poisson e Quasi-Poisson. Os resultados indicaram que o tempo de exposição a computador, videogame ou outros dispositivos aos finais de semana e o sexo masculino apresentaram maior relevância preditiva para o tempo semanal de exposição a TV. O consumo de refrigerantes mostrou associação estatisticamente significativa no modelo Quasi-Poisson, porém com baixa importância preditiva no modelo Random Forest. Conclui-se que fatores comportamentais e sociodemográficos estão associados ao tempo de exposição a TV em escolares, e que o consumo de refrigerantes deve ser interpretado com cautela como potencial fator associado, reforçando a necessidade de investigações adicionais.
Referências
Alosaimi, N., Sherar, L. B., Griffiths, P., & Pearson, N. (2023). Clustering of diet, physical activity and sedentary behaviour and related physical and mental health outcomes: a systematic review. BMC Public Health, 23(1), 1572. https://doi.org/10.1186/s12889-023-16372-6
Bejarano, C. M., Carlson, J. A., Conway, T. L., Saelens, B. E., Glanz, K., Couch, S. C., Cain, K. L., & Sallis, J. F. (2021). Physical Activity, Sedentary Time, and Diet as Mediators of the Association Between TV Time and BMI in Youth. American Journal of Health Promotion, 35(5), 613–623. https://doi.org/10.1177/0890117120984943
Couronné, R., Probst, P., & Boulesteix, A.-L. (2018). Random forest versus logistic regression: a large-scale benchmark experiment. BMC Bioinformatics, 19(1), 270. https://doi.org/10.1186/s12859-018-2264-5
Davis, J. N., Asigbee, F. M., Markowitz, A. K., Landry, M. J., Vandyousefi, S., Khazaee, E., Ghaddar, R., & Goran, M. I. (2018). Consumption of artificial sweetened beverages associated with adiposity and increasing HbA1c in Hispanic youth. Clinical Obesity, 8(4), 236–243. https://doi.org/10.1111/cob.12260
Epifânio, S. B. O., Silveira, J. A. C. da, Menezes, R. C. E. de, Marinho, P. M., Brebal, K. M. de M., & Longo-Silva, G. (2020). Análise de série temporal do consumo de bebidas açucaradas entre adultos no Brasil: 2007 a 2014. Ciência & Saúde Coletiva, 25(7), 2529–2540. https://doi.org/10.1590/1413-81232020257.19402018
Fan, H., Yan, J., Yang, Z., Liang, K., & Chen, S. (2022). Cross-sectional associations between screen time and the selected lifestyle behaviors in adolescents. Frontiers in Public Health, 10. https://doi.org/10.3389/fpubh.2022.932017
Fernāte, A., Vazne, Ž., Zuša, A., Bula-Biteniece, I., Dravniece, I., Grants, J., Žīdens, J., & Jakovļeva, M. (2024). Adult physical activity, sedentary behaviour and sleep quality in the digital transformation era. Society. Integration. Education. Proceedings of the International Scientific Conference, 2, 540–549. https://doi.org/10.17770/sie2024vol2.7790
Gong, W.-J., Fong, D. Y.-T., Wang, M.-P., Lam, T.-H., Chung, T. W.-H., & Ho, S.-Y. (2019). Increasing socioeconomic disparities in sedentary behaviors in Chinese children. BMC Public Health, 19(1), 754. https://doi.org/10.1186/s12889-019-7092-7
Hu, X., Drenowatz, C., Duncan, M., Bao, R., Chen, S., He, J., & Tang, Y. (2023). Physical education, muscle strengthening exercise, sport participation and their associations with screen time in adolescents. Frontiers in Public Health, 11. https://doi.org/10.3389/fpubh.2023.1100958
Khadidja, L. (2024). The Digital Family Culture and Its Influence on Child Electronic Guidance: A Case Study of Selected Families. Pakistan Journal of Life and Social Sciences (PJLSS), 22(2). https://doi.org/10.57239/PJLSS-2024-22.2.00613
Knopf, A. (2025). Screen time associated with substance use in young adolescents. The Brown University Child and Adolescent Behavior Letter, 41(2), 1–5. https://doi.org/10.1002/cbl.30843
Lopes, G. C. D. (2024). Screen, brain and behavior: neuroscientific evidence about the negative effects of prolonged use. South Florida Journal of Health, 5(4), e4756. https://doi.org/10.46981/sfjhv5n4-004
Lourenço, C. L. M., Christofoletti, M., Malta, D. C., & Mendes, E. L. (2021). Associação entre tempo excessivo frente à TV e índice de massa corporal em adolescentes brasileiros: uma análise de regressão quantílica da PeNSE, 2015. Ciência & Saúde Coletiva, 26(11), 5817–5828. https://doi.org/10.1590/1413-812320212611.28352020
Lourenço, C. L. M., Silva Filho, R. C. dos S., Hauser, E., Barbosa, A. R., & Mendes, E. L. (2020). Cluster and simultaneity of modifiable risk factors for cardiovascular diseases in adolescents of Southeast Brazil. Motriz: Revista de Educação Física, 26(2). https://doi.org/10.1590/s1980-6574202000020033
Medina, C., Jáuregui, A., Campos-Nonato, I., & Barquera, S. (2018). Prevalencia y tendencias de actividad física en niños y adolescentes: resultados de Ensanut 2012 y Ensanut MC 2016. Salud Pública de México, 60(3, may-jun), 263. https://doi.org/10.21149/8819
Molleri, N., Gomes Junior, S. C., Marano, D., & Zin, A. (2023). Survey of the Adequacy of Brazilian Children and Adolescents to the 24-Hour Movement Guidelines before and during the COVID-19 Pandemic. International Journal of Environmental Research and Public Health, 20(9), 5737. https://doi.org/10.3390/ijerph20095737
Mori, T., Oishi, K., Aoki, T., Ito, Y., Ikeue, K., Yamaguchi, H., Hanano, H., Yamamoto, Y., & Ishii, K. (2023). Association Between Adolescent Screen Time And Neighborhood Socioeconomic Factors. Medicine & Science in Sports & Exercise, 55(9S), 520–521. https://doi.org/10.1249/01.mss.0000984656.01422.6e
MUTHURI, S. K. et al. Correlates of objectively measured overweight/obesity and physical activity in Kenyan school children: results from ISCOLE-Kenya. BMC Public Health, v. 14, n. 1, p. 436, 2014. DOI: https://doi.org/10.1186/1471-2458-14-436.
Neshteruk, C. D., Tripicchio, G. L., Lobaugh, S., Vaughn, A. E., Luecking, C. T., Mazzucca, S., & Ward, D. S. (2021). Screen Time Parenting Practices and Associations with Preschool Children’s TV Viewing and Weight-Related Outcomes. International Journal of Environmental Research and Public Health, 18(14), 7359. https://doi.org/10.3390/ijerph18147359
Neta, A. da C. P. de A., Farias Júnior, J. C. de, Ferreira, F. E. L. de L., Aznar, L. A. M., & Marchioni, D. M. L. (2024). Association between sedentary behavior, diet and nutritional status in adolescents: baseline results from the LONCAAFS Study. Ciência & Saúde Coletiva, 29(4). https://doi.org/10.1590/1413-81232024294.17082022
Oliveira, G. A. L., Santos Gonçalves, V. S., Nakano, E. Y., & Toral, N. (2024). Consumption of ultra-processed foods and low dietary diversity are associated with sedentary and unhealthy eating behaviors: A nationwide study with Brazilian Schoolchildren. PLOS ONE, 19(1), e0294871. https://doi.org/10.1371/journal.pone.0294871
Patil, K. V., Yesugade, K. D., & Naikwadi, K. B. (2024). A Study on Regression Based Machine Learning Models to Predict the Student Performance. Journal of Engineering Education Transformations, 38(2), 177–186. https://doi.org/10.16920/jeet/2024/v38i2/24200
Pearson, N., & Biddle, S. J. H. (2011). Sedentary Behavior and Dietary Intake in Children, Adolescents, and Adults. American Journal of Preventive Medicine, 41(2), 178–188. https://doi.org/10.1016/j.amepre.2011.05.002
Raggio Luiz, R., & Magnanini, M. M. F. (2000). A Lógica Da Determinação Do Tamanho Da Amostra Em Investigações. Cadernos Saúde Coletiva, 8, 9–28. https://edisciplinas.usp.br/pluginfile.php/4116370/mod_resource/content/1/DeterminaçãoamostraRonir2000_2.pdf
Sahoo, S. (2024). The impact of streaming services on youth television viewing habits and media literacy. ShodhKosh: Journal of Visual and Performing Arts, 5(1). https://doi.org/10.29121/shodhkosh.v5.i1.2024.946
Silva, K. S. da, Lopes, A. D. S., Hoefelmann, L. P., Cabral, L. G. de A., De Bem, M. F. L., Barros, M. V. G. de, & Nahas, M. V. (2013). Projeto COMPAC (comportamentos dos adolescentes catarinenses): aspectos. Revista Brasileira de Cineantropometria e Desempenho Humano, 15(1). https://doi.org/10.5007/1980-0037.2013v15n1p1
Stiglic, N., & Viner, R. M. (2019). Effects of screentime on the health and well-being of children and adolescents: a systematic review of reviews. BMJ Open, 9(1), e023191. https://doi.org/10.1136/bmjopen-2018-023191
Thomas, G., Bennie, J. A., De Cocker, K., Ireland, M. J., & Biddle, S. J. H. (2020). Screen-based behaviors in Australian adolescents: Longitudinal trends from a 4-year follow-up study. Preventive Medicine, 141, 106258. https://doi.org/10.1016/j.ypmed.2020.106258
Vilardell-Dávila, A., Martínez-Andrade, G., Klünder-Klünder, M., Miranda-Lora, A. L., Mendoza, E., Flores-Huerta, S., Vargas-González, J. E., Duque, X., & Vilchis-Gil, J. (2023). A Multi-Component Educational Intervention for Addressing Levels of Physical Activity and Sedentary Behaviors of Schoolchildren. International Journal of Environmental Research and Public Health, 20(4), 3003. https://doi.org/10.3390/ijerph20043003
Vizcaino, M., Buman, M., DesRoches, T., & Wharton, C. (2020). From TVs to tablets: the relation between device-specific screen time and health-related behaviors and characteristics. BMC Public Health, 20(1), 1295. https://doi.org/10.1186/s12889-020-09410-0
Downloads
Publicado
Edição
Secção
Licença
Direitos de Autor (c) 2025 Rodrigo Mercês Reis Fonseca, Bruna Maria Palatino Ferreira, Cristiane dos Santos Silva, Rafaelle Dayanne Barros, Hector Luiz Rodrigues Munaro

Este trabalho encontra-se publicado com a Licença Internacional Creative Commons Atribuição 4.0.
The BRAJETS follows the policy for Open Access Journals, provides immediate and free access to its content, following the principle that making scientific knowledge freely available to the public supports a greater global exchange of knowledge and provides more international democratization of knowledge. Therefore, no fees apply, whether for submission, evaluation, publication, viewing or downloading of articles. In this sense, the authors who publish in this journal agree with the following terms: A) The authors retain the copyright and grant the journal the right to first publication, with the work simultaneously licensed under the Creative Commons Attribution License (CC BY), allowing the sharing of the work with recognition of the authorship of the work and initial publication in this journal. B) Authors are authorized to distribute non-exclusively the version of the work published in this journal (eg, publish in the institutional and non-institutional repository, as well as a book chapter), with acknowledgment of authorship and initial publication in this journal. C) Authors are encouraged to publish and distribute their work online (eg, online repositories or on their personal page), as well as to increase the impact and citation of the published work.
