tetano
Editor, Senior Moderator
BMJ Open
. 2026 May 18;16(5):e104115.
doi: 10.1136/bmjopen-2025-104115.
Assessing community resilience during the first COVID-19 wave peak in a megacity in eastern China: a cross-sectional study using social media data
Lanyue Zhang[SUP] 1 [/SUP], Shouchuang Zhang[SUP] 1 [/SUP], Yuehui Wei[SUP] 1 [/SUP], Danijela Gasevic[SUP] 2 [/SUP], Stella Talic[SUP] 2 [/SUP], Jiayi Weng[SUP] 3 [/SUP], Jun Zhang[SUP] 4 [/SUP], Larry Z Liu[SUP] 5 6 [/SUP], Weiyan Jian[SUP] 7 8 [/SUP]
Affiliations
Objectives: In the face of pandemics resulting from infectious disease, improving community resilience has become increasingly vital. China's sudden exit from Zero-COVID policy in December 2022 triggered a surge in COVID-19 cases, compounded by medication shortages due to earlier restrictions, creating a public health crisis. This study assesses community resilience during the first post-Zero-COVID infection wave (8 December 2022 to 7 January 2023), focusing on adaptation mechanisms, resource mobilisation, protective behaviours and medicine access, using real-time social media data to capture these dynamics.
Methods: This cross-sectional study analysed all geotagged COVID-19-related posts on Sina Weibo-China's largest public microblogging platform-collected from a megacity in eastern China with over 10 million residents, covering 8 December 2022 to 7 January 2023. Posts were obtained through a data purchase agreement with Sina Weibo and comprised publicly available content. Machine learning and natural language processing were applied to classify posts across four dimensions: content, responder, response type and time. Community resilience was assessed using the community-level response ratio, response speed and sentiment expressed in interactions related to medicine-seeking posts.
Results: 26 973 posts were analysed, of which 12 152 (45.05%) were help-seeking. Among these help-seeking posts, 11 236 (92.46%) specifically sought COVID-19 medicine, of which 8495 (75.61%) of these medicine-seeking posts received community support. Over 4 weeks, community responses comprised >70% of all replies (the rest were from market-based responders, government and NGOs). More than half of the community responses occurred within an hour, and the emotional state at the community level was the most stable and consistently positive, indicating a high level of prompt community engagement.
Conclusion: Communities in the sample area consistently exhibited prompt and proactive responses during the health crisis, with community responses accounting for the majority of interactions on medicine-seeking posts. The power of community mutual aid can significantly enhance responsiveness to public health emergencies. Such insights suggest that strengthening community resilience is crucial in designing more effective disaster response strategies.
Keywords: COVID-19; Health policy; Internet; Public health; Social Media.
. 2026 May 18;16(5):e104115.
doi: 10.1136/bmjopen-2025-104115.
Assessing community resilience during the first COVID-19 wave peak in a megacity in eastern China: a cross-sectional study using social media data
Lanyue Zhang[SUP] 1 [/SUP], Shouchuang Zhang[SUP] 1 [/SUP], Yuehui Wei[SUP] 1 [/SUP], Danijela Gasevic[SUP] 2 [/SUP], Stella Talic[SUP] 2 [/SUP], Jiayi Weng[SUP] 3 [/SUP], Jun Zhang[SUP] 4 [/SUP], Larry Z Liu[SUP] 5 6 [/SUP], Weiyan Jian[SUP] 7 8 [/SUP]
Affiliations
- PMID: 42150823
- DOI: 10.1136/bmjopen-2025-104115
Objectives: In the face of pandemics resulting from infectious disease, improving community resilience has become increasingly vital. China's sudden exit from Zero-COVID policy in December 2022 triggered a surge in COVID-19 cases, compounded by medication shortages due to earlier restrictions, creating a public health crisis. This study assesses community resilience during the first post-Zero-COVID infection wave (8 December 2022 to 7 January 2023), focusing on adaptation mechanisms, resource mobilisation, protective behaviours and medicine access, using real-time social media data to capture these dynamics.
Methods: This cross-sectional study analysed all geotagged COVID-19-related posts on Sina Weibo-China's largest public microblogging platform-collected from a megacity in eastern China with over 10 million residents, covering 8 December 2022 to 7 January 2023. Posts were obtained through a data purchase agreement with Sina Weibo and comprised publicly available content. Machine learning and natural language processing were applied to classify posts across four dimensions: content, responder, response type and time. Community resilience was assessed using the community-level response ratio, response speed and sentiment expressed in interactions related to medicine-seeking posts.
Results: 26 973 posts were analysed, of which 12 152 (45.05%) were help-seeking. Among these help-seeking posts, 11 236 (92.46%) specifically sought COVID-19 medicine, of which 8495 (75.61%) of these medicine-seeking posts received community support. Over 4 weeks, community responses comprised >70% of all replies (the rest were from market-based responders, government and NGOs). More than half of the community responses occurred within an hour, and the emotional state at the community level was the most stable and consistently positive, indicating a high level of prompt community engagement.
Conclusion: Communities in the sample area consistently exhibited prompt and proactive responses during the health crisis, with community responses accounting for the majority of interactions on medicine-seeking posts. The power of community mutual aid can significantly enhance responsiveness to public health emergencies. Such insights suggest that strengthening community resilience is crucial in designing more effective disaster response strategies.
Keywords: COVID-19; Health policy; Internet; Public health; Social Media.