JMIR Publications

ABSTRACT

Background:

For an emergent pandemic, such as COVID-19, the statistics of symptoms based on hospital data may be biased or delayed due to the high proportion of asymptomatic or mild-symptom infections who were not recorded in hospitals. Meanwhile, the difficulty of accessing large-scale clinical data also limits many researchers to conduct timely research.

Objective:

Given the wide coverage and promptness of social media, this study aimed to present an efficient workflow to track and visualize the dynamic characteristics and co-occurrences of symptoms for the pandemic from large-scale and long-term social media data.

Methods:

This retrospective study included 471,553,966 COVID-19-related tweets from Feb 1, 2020, to Apr 30, 2022. The dynamic evolutions of COVID-19 symptoms over time and virus strain (Delta and Omicron) were analyzed from the perspectives of weekly new cases, overall distribution, and temporal prevalence of reported symptoms. A co-occurrence symptom network was developed and visualized to investigate inner relationships among symptoms and affected body systems.

Results:

This study identified 201 COVID-19 symptoms and grouped them into 10 affected body systems. The weekly quantity of self-reported symptoms has a high consistency and one-week leading trend (0.8802, P<.001) with new COVID-19 infections. We found the prevalence difference of symptoms between Delta period and Omicron period: less severe symptoms (coma and dyspnea), more flu-like symptoms (throat pain and nasal congestion), and less typical COVID symptoms (anosmia and taste altered) (All P<.001). Network analysis reveals the co-occurrences among symptoms and systems corresponding to specific disease progressions, including palpitations and dyspnea, alopecia and impotence.

Conclusions:

This study identified more and milder COVID-19 symptoms than clinical research and characterize the dynamic symptom evolution based on 400 million tweets over 27 months. The symptom network reveals potential comorbidity risk and prognostic disease progressions. These demonstrate that the cooperation of social media and a well-designed workflow can depict a holistic picture of pandemic symptoms to complement clinical studies.