Fishing for Validation: Understanding Promises and Challenges of a Private Social Media Group for COVID-19 Long-Hauler Patients

This article summarizes the paper “Fishing for Validation: Understanding Promises and Challenges of a Private Social Media Group for COVID-19 Long-Hauler Patients” about the characteristics of a private social media group that was dedicated to marginalized COVID-19 long-hauler community. This paper will be presented at the 26th ACM Conference on Computer-Supported Cooperative Work and Social Computing, a top venue for social computing scholarship. It will also be published in the journal Proceedings of the ACM (PACM).

The COVID-19 pandemic, a global health crisis, has infected more than 301 million people and killed more than 5.48 million globally since late January 2020. Although COVID-19 vaccines are found to be effective at protecting people against severe infection and death, new variants of COVID-19 virus raised concerns against the efficacy of the vaccines and the antibodies from previous infections. The pandemic that has continued to spread for more than two years has no longer remained a health crisis; instead, it has become a human, economic, and social turmoil for many communities. At the time of this global crisis, it is critical to understand how social media impacts people’s lives.

During the pandemic social media are being used to spread unapproved cures, misleading rumors, and conspiracy theories about the origin of the Coronavirus. More than a quarter of the most viewed videos on COVID-19 on YouTube contained misleading information, representing more than 62 million views that created unnecessary health anxiety. However, not all information consumed from social media was harmful. Scientists and healthcare professionals used social media to circulate critical information about the pandemic and the vaccinations instantaneously to millions of people. Amid long-term social distancing, social media helped people to maintain social connections and ward off social isolation. In this paper, we made further contribution in this domain by exploring how social media groups may become the primary source of assurance and validation for communities, especially at the time of global crisis when the basic needs and requirements are hard to fulfill.

Who are COVID-19 Long-Haulers?

To examine the impact of social media groups, we focused on COVID-19 long-haulers community on Facebook. Who belongs to the COVID-19 long-haulers community? Typically, mild or moderate COVID-19 symptoms last about two weeks for most people. However, some individuals experience lingering health problems even when they have recovered from the acute phase of the illness. Standard COVID tests cannot find any live Coronavirus in these patients’ bodies. However, they still experience a wide range of symptoms regularly for months. People living with post-COVID symptoms for at least four weeks are identified as “COVID-19 long haulers”, ”long-COVID”, or “post-acute sequelae of SARS-CoV-2 infection” as referred by the National Institutes of Health (NIH).

Our Contributions:

• First, long-hauler patients did not exist before this pandemic. Studying this community has given us a unique opportunity to understand how trust and empathy among community members can be formed entirely on social media when other forms of communication are hard to reach.
• Second, we contributed to the emerging topic of studying the role of social media in seeking health information. When misinformation and disinformation on online platforms have become a primary concern while seeking health information, our study explored how social media groups may establish a sense of mutual trust and emotional dependence.
• Third, we aimed to go beyond the analysis of symptoms reported by long-hauler patients in their social-media group. We learned why and how the COVID-19 long-hauler group named “COVID-19 Long Haulers Discussion Group” was formed on Facebook and its impact on the members coping with their long-hauler conditions during difficult times.
• Our results contributed to the existing literature on cooperative AI. They showed how an AI-supported social media platform like Facebook could contribute to the formation of an online community and motivate the members to cooperate to overcome emotional and physiological challenges during a crisis.

What did we do?

We conducted a mixed-method study that took two main approaches. First, from Facebook, we collected the long-hauler group’s posts and comments (186,860 entries) from November 3rd, 2020, to February 8th, 2021. To identify the topics of discussion in this group, we applied a combination of qualitative annotation and LDA topic modeling. Our analysis allowed us to acquire a baseline understanding of the group’s discussion topics on social media. This knowledge helped us interview the group members in the second stage. We interviewed 23 members of the group, including the admin. We asked them to reflect on their experiences of being a part of the group.

Key Findings:

• We identified two primary topics: 1) discussions regarding long-hauler symptoms and 2) discussions not related to long-hauler symptoms. These two main topics contained nine sub-topics. The discussions regarding symptoms often asked for assurance and validation from the community that their lingering, unusual symptoms were not completely unheard of by other group members. In contrast, the posts and comments that were not about symptoms often shared the journey to recovery to keep the hope up among the members of the community.
• Our interviews found that the Facebook group helped its members find validation that their long-COVID symptoms were not related to any psychosomatic conditions; instead, they were caused by their physiological conditions. The group’s private setting helped them feel secure and protected where no one would take advantage of them for sharing their vulnerabilities or judge them because of their unexplained conditions. The group helped them frame their questions to their doctors and primary care physicians in a way that often assisted them in diagnosing the root causes of their symptoms faster.

We learned how social media groups’ intimate, private settings might serve as essential support systems for online communities. The community can stand for each other at times when regular resources, infrastructures, and even friends and family members may not be able to provide adequate support. Our findings have widespread implications for examining online community designs and functioning to better support ad-hoc groups’ needs. We also drew attention to the limitations and complexities of such groups, which have great potential but may not serve all the needs of the communities as the communities grow bigger and older. We believe this research would stir up discussion on new avenues for public health intervention to better address the requirements of minority groups during critical times such as a global pandemic.