Search

Depression affects 264 million persons in the world, accounting for some 4.3% of the global burden of disease. Current studies indicate that the decrease in the consumption of omega-3 food sources is associated with the increasing incidence of depression. The study aims to assess the association between the consumption of omega-3 and the maintenance and incidence of depressive episodes in adults (39–64 years) and elderly adults (>65 years). This was a longitudinal study using data from the baseline and first follow-up wave of the Longitudinal Study of Adult Health (ELSA-Brasil). Depressive episodes were obtained with the Clinical Interview Schedule Revised (CIS-R), and food consumption was measured with the Food Frequency Questionnaire (FFQ). Logistic regression was used to analyze associations between the consumption of omega-3 and depressive episodes. Fatty acids from the omega-3 family showed a protective effect against the maintenance of depressive episodes. In relation to incidence, the estimates suggest that the higher the consumption of omega-3 acids, the lower the risk of developing depressive episodes, and significant associations were found between the consumption of omega-3 and alpha-linolenic acid. Dietary consumption of omega-3, DHA, EPA, DPA, and alpha linolenic fatty acids may have a protective effect against the maintenance and incidence of depressive episodes.

Instituto Brasileiro de Geografia e Estatística. Escola Nacional de Ciências Estatísticas, Rio de Janeiro, RJ, Brasil. Universidade do Estado do Rio de Janeiro. Instituto de Nutrição. Rio de Janeiro, RJ, Brasil. Fundação Oswaldo Cruz. Escola Nacional de Saúde Pública. Rio de Janeiro, RJ, Brasil. Universidade de São Paulo. Faculdade de Medicina da Universidade de São Paulo & Hospital Universitário. Departamento de Medicina Interna. São Paulo, SP, Brasil. Universidade de São Paulo. Faculdade de Medicina. Departamento de Clínica Médica. Divisão de Geriatriia. São Paulo, SP, Brasil. Fundação Oswaldo Cruz. Escola Nacional de Saúde Pública. Departamento de Epidemiologia. Rio de Janeiro, RJ, Brasil. Fundação Oswaldo Cruz. Instituto Oswaldo Cruz. Laboratório de Educação em Saúde e Meio Ambiente. Rio de Janeiro, RJ, Brasil. Background Multimorbidity is a worldwide concern related to greater disability, worse quality of life, and mortality. The early prediction is crucial for preventive strategies design and integrative medical practice. However, knowledge about how to predict multimorbidity is limited, possibly due to the complexity involved in predicting multiple chronic diseases. Methods In this study, we present the use of a machine learning approach to build cost-effective multimorbidity prediction models. Based on predictors easily obtainable in clinical practice (sociodemographic, clinical, family disease history and lifestyle), we build and compared the performance of seven multilabel classifiers (multivariate random forest, and classifier chain, binary relevance and binary dependence, with random forest and support vector machine as base classifiers), using a sample of 15105 participants from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). We developed a web application for the building and use of prediction models. Results Classifier chain with random forest as base classifier performed better (accuracy = 0.34, subset accuracy = 0.15, and Hamming Loss = 0.16). For different feature sets, random forest based classifiers outperformed those based on support vector machine. BMI, blood pressure, sex, and age were the features most relevant to multimorbidity prediction. Conclusions - Our results support the choice of random forest based classifiers for multimorbidity prediction.