Long-term patient-reported symptoms of COVID-19: an analysis of social media data

• DOI: 10.1101/2020.07.29.20164418
• Publication Date: 2020-08-01T12:27:57Z
• Authors: Juan M. Banda; Gurdas Viguruji S...; Osaid H. Alser; Daniel Prieto-Alh...
• Keywords: Globe; Longitudinal data; Identification; 2019-20 coronavirus outbreak; Pandemic

Abstract As the COVID-19 virus continues to infect people across globe, there is little understanding of long term implications for recovered patients. There have been reports persistent symptoms after confirmed infections on patients even three months initial recovery. While some these documented follow-ups clinical records, or participate in longitudinal surveys, datasets are usually not publicly available standardized perform analyses them. Therefore, a need use additional data sources continued follow-up and identification latent that might be underreported other places. In this work we present preliminary characterization post-COVID-19 using social media from Twitter. We combination natural language processing clinician reviews identify self-reported set Twitter users.