Symptom clusters in COVID-19: A potential clinical prediction tool from the COVID Symptom Study app
Carole H. Sudre, Karla A. Lee, Mary Ní Lochlainn, Thomas Varsavsky, Benjamin Murray, Mark S. Graham, Cristina Menni, Marc Modat, Ruth C. E. Bowyer, Long H. Nguyen, David A. Drew, Amit D. Joshi, Wenjie Ma, Chuan‐Guo Guo, Chun‐Han Lo, Sajaysurya Ganesh, Abubakar Buwe, Joan Capdevila Pujol, Julien Lavigne du Cadet, Alessia Visconti, Maxim B. Freidin, Julia S. El-Sayed Moustafa, Mario Falchi, Richard Davies, Maria F. Gomez, Tove Fall, M. Jorge Cardoso, Jonathan Wolf, Paul W. Franks, Andrew T. Chan, Tim D. Spector, Claire J. Steves, Sébastien Ourselin
Abstract
As no one symptom can predict disease severity or the need for dedicated medical support in coronavirus disease 2019 (COVID-19), we asked whether documenting symptom time series over the first few days informs outcome. Unsupervised time series clustering over symptom presentation was performed on data collected from a training dataset of completed cases enlisted early from the COVID Symptom Study Smartphone application, yielding six distinct symptom presentations. Clustering was validated on an independent replication dataset between 1 and 28 May 2020. Using the first 5 days of symptom logging, the ROC-AUC (receiver operating characteristic - area under the curve) of need for respiratory support was 78.8%, substantially outperforming personal characteristics alone (ROC-AUC 69.5%). Such an approach could be used to monitor at-risk patients and predict medical resource requirements days before they are required.