Comparing machine learning algorithms for multimorbidity prediction: An example from the Elsa-Brasil study

• Published in: PloS one Vol. 17; no. 10; p. e0275619
• Main Authors: Polessa Paula, Daniela, Barbosa Aguiar, Odaleia, Pruner Marques, Larissa, Bensenor, Isabela, Suemoto, Claudia Kimie, Mendes da Fonseca, Maria de Jesus, Griep, Rosane Härter
• Format: Journal Article
• Language: English
• Published: San Francisco Public Library of Science 07-10-2022; Public Library of Science (PLoS)
• Subjects: Accuracy; Algorithms; Angina pectoris; Applications programs; Asthma; Biology and Life Sciences; Blood pressure; Cancer; Cardiovascular disease; Chains; Chronic illnesses; Classifiers; Comorbidity; Computer and Information Sciences; Diabetes; Heart; Kidney diseases; Learning algorithms; Longitudinal studies; Machine learning; Medicine and Health Sciences; Mental disorders; Metabolic disorders; Migraine; Morbidity; Multimorbidity; Physical Sciences; Prediction models; Quality of life; Questionnaires; Research and Analysis Methods; Risk factors; Support vector machines; Testing; Brazil
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0275619

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.