Potential confounders in the analysis of Brazilian adolescent’s health: a combination of machine learning and graph theory
| Autor: | Guilherme Augusto Zimeo Morais, João Ricardo Sato, Amanda Yumi Ambriola Oku, Ana Paula Arantes Bueno, André Fujita |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Male
medicine.medical_specialty Adolescent Health Toxicology and Mutagenesis Adolescent Health Poison control TEORIA DOS GRAFOS Suicide prevention Article Occupational safety and health Machine Learning 03 medical and health sciences 0302 clinical medicine Injury prevention medicine Humans Social inequality 030212 general & internal medicine 030505 public health Public health machine-learning public health Public Health Environmental and Occupational Health Human factors and ergonomics graph Nutrition Assessment Social Class Socioeconomic Factors network Female 0305 other medical science Psychology Centrality Social psychology Brazil |
| Zdroj: | Repositório Institucional da USP (Biblioteca Digital da Produção Intelectual) Universidade de São Paulo (USP) instacron:USP International Journal of Environmental Research and Public Health Volume 17 Issue 1 |
| Popis: | The prevalence of health problems during childhood and adolescence is high in developing countries such as Brazil. Social inequality, violence, and malnutrition have strong impact on youth health. To better understand these issues we propose to combine machine-learning methods and graph analysis to build predictive networks applied to the Brazilian National Student Health Survey (PenSE 2015) data, a large dataset that consists of questionnaires filled by the students. By using a combination of gradient boosting machines and centrality hub metric, it was possible to identify potential confounders to be considered when conducting association analyses among variables. The variables were ranked according to their hub centrality to predict the other variables from a directed weighted-graph perspective. The top five ranked confounder variables were &ldquo gender&rdquo &ldquo oral health care&rdquo intended education level&rdquo and two variables associated with nutrition habits&mdash eat while watching TV&rdquo and &ldquo never eat fast-food&rdquo In conclusion, although causal effects cannot be inferred from the data, we believe that the proposed approach might be a useful tool to obtain novel insights on the association between variables and to identify general factors related to health conditions. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|