Trend and Co-Occurrence Network of COVID-19 Symptoms From Large-Scale Social Media Data: Infoveillance Study

    December 2022
    Jiageng Wu, Lumin Wang, Yining Hua, Minghui Li, Li Zhou, David W. Bates, Jie Yang
    Image of study
    TLDR Social media data can help track and predict COVID-19 symptoms and trends.
    This study analyzed 471,553,966 COVID-19-related tweets from February 1, 2020, to April 30, 2022, identifying 201 symptoms across 10 body systems. It found a strong correlation between self-reported symptoms and new COVID-19 infections, with a 1-week leading trend. Symptom prevalence shifted from respiratory to musculoskeletal and nervous symptoms over time. Differences between Delta and Omicron strains were noted, with Omicron showing fewer severe and typical COVID-19 symptoms. The co-occurrence network highlighted potential comorbidity risks and disease progressions. The study demonstrates that social media data can effectively complement clinical research by providing a comprehensive view of pandemic symptoms.
    Discuss this study in the Community →

    Research cited in this study

    7 / 7 results