App-based symptom tracking to optimize SARS-CoV-2 testing strategy using machine learning

• Published in: PloS one Vol. 16; no. 3; p. e0248920
• Main Authors: Dantas, Leila F, Peres, Igor T, Bastos, Leonardo S L, Marchesi, Janaina F, de Souza, Guilherme F G, Gelli, João Gabriel M, Baião, Fernanda A, Maçaira, Paula, Hamacher, Silvio, Bozza, Fernando A
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 25-03-2021; Public Library of Science (PLoS)
• Subjects: Adult; Age; Anosmia; Anosmia - etiology; Biology and life sciences; Brazil; Computer and Information Sciences; Continuity (mathematics); Coronaviruses; Coryza; Cough; COVID-19; COVID-19 - complications; COVID-19 - diagnosis; COVID-19 - virology; COVID-19 Testing; Demographics; Demography; Diarrhea; Disease transmission; Distance learning; Dyspnea - etiology; Electronic devices; Engineering and Technology; Epidemics; False Negative Reactions; False Positive Reactions; Female; Fever; Fever - etiology; Gender; Health risks; Heart rate; Humans; Industrial engineering; Infections; Learning; Learning algorithms; Machine Learning; Male; Medicine and Health Sciences; Middle Aged; Mobile Applications; Myalgia; Nausea; Pharyngitis; Physical Sciences; Prediction models; Public health; Registries; Research and Analysis Methods; Retrospective Studies; Rhinitis; Risk; SARS-CoV-2 - isolation & purification; Self Report; Severe acute respiratory syndrome; Severe acute respiratory syndrome coronavirus 2; Sleep; Smartphones; Smell; Social Sciences; Statistical analysis; Variables; Viral diseases; Wearable technology; Brazil; United States > US; Rio de Janeiro Brazil
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0248920

Tests are scarce resources, especially in low and middle-income countries, and the optimization of testing programs during a pandemic is critical for the effectiveness of the disease control. Hence, we aim to use the combination of symptoms to build a predictive model as a screening tool to identify people and areas with a higher risk of SARS-CoV-2 infection to be prioritized for testing.
We performed a retrospective analysis of individuals registered in "Dados do Bem," a Brazilian app-based symptom tracker. We applied machine learning techniques and provided a SARS-CoV-2 infection risk map of Rio de Janeiro city.
From April 28 to July 16, 2020, 337,435 individuals registered their symptoms through the app. Of these, 49,721 participants were tested for SARS-CoV-2 infection, being 5,888 (11.8%) positive. Among self-reported symptoms, loss of smell (OR[95%CI]: 4.6 [4.4-4.9]), fever (2.6 [2.5-2.8]), and shortness of breath (2.1 [1.6-2.7]) were independently associated with SARS-CoV-2 infection. Our final model obtained a competitive performance, with only 7% of false-negative users predicted as negatives (NPV = 0.93). The model was incorporated by the "Dados do Bem" app aiming to prioritize users for testing. We developed an external validation in the city of Rio de Janeiro. We found that the proportion of positive results increased significantly from 14.9% (before using our model) to 18.1% (after the model).
Our results showed that the combination of symptoms might predict SARS-Cov-2 infection and, therefore, can be used as a tool by decision-makers to refine testing and disease control strategies.