What is the influence of lifestyle habits and socioeconomic factors on the occurrence of prostate cancer in Brazil?

• Autor(a) principal: Souza, Marco Antonio de
• Data de Publicação: 2023
• Outros Autores: Monteiro, Camila Nascimento, Barros, Cláudia Renata dos Santos
• Tipo de documento: preprint
• Idioma: por
• Título da fonte: SciELO Preprints
• Texto Completo: https://preprints.scielo.org/index.php/scielo/preprint/view/7566
• topic: câncer de próstata; estilo de vida; estudos transversais; Aprendizado de máquina; prostate cancer; lifestyle; Cross-Sectional Studies; machine learning; cáncer de próstata; estilo de vida; Estudios Transversales; aprendizaje automático

Objective: the investigation of physical, lifestyle and socioeconomic features that may be associated with the occurrence of prostate cancer in Brazil. Methods: a microdata base referring to the 2019 National Health Survey in Brazil was used, with the selection of 42,799 male individuals; this group was analyzed using statistical methods and machine learning modeling (logistic regression and decision tree). Results: the models applied allowed us to identify with a good level of accuracy individuals with prostate cancer diagnosis (DCP), in addition to groups with specific features more strongly associated with such a disease. Conclusion: the models indicate a significant influence of socioeconomic, physical and dietary factors on the frequency of DCP in the analyzed group. The high level of accuracy and sensitivity of the models demonstrates the potential of machine learning methods for predicting DCP.