Symptom clusters in Covid19: A potential clinical prediction tool from the COVID Symptom study app

  1. Carole H. Sudre
  2. Karla A. Lee
  3. Mary Ni Lochlainn
  4. Thomas Varsavsky
  5. Benjamin Murray
  6. Mark S. Graham
  7. Cristina Menni
  8. Marc Modat
  9. Ruth C E Bowyer
  10. Long H. Nguyen
  11. David A. Drew
  12. Amit D. Joshi
  13. Wenjie Ma
  14. Chuan-Guo Guo
  15. Chun-Han Lo
  16. Sajaysurya Ganesh
  17. Abubakar Buwe
  18. Joan Capdevila Pujol
  19. Julien Lavigne du Cadet
  20. Alessia Visconti
  21. Maxim B Freidin
  22. Julia S. El-Sayed Moustafa
  23. Mario Falchi
  24. Richard Davies
  25. Maria F. Gomez
  26. Tove Fall
  27. M. Jorge Cardoso
  28. Jonathan Wolf
  29. Paul W. Franks
  30. Andrew T. Chan
  31. Tim D. Spector
  32. Claire J Steves
  33. Sébastien Ourselin
1 evaluations Published on
Read the full article Related papers
This article on Sciety

Abstract

As no one symptom can predict disease severity or the need for dedicated medical support in COVID-19, we asked if 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 May 1-May 28th, 2020. Using the first 5 days of symptom logging, the ROC-AUC 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.

One sentence summary

Longitudinal clustering of symptoms can predict the need for respiratory support in severe COVID-19.

Related articles

Related articles are currently not available for this article.