Community polls: Informal social media polls: SARS-CoV-2 reinfection and Long COVID, and the presence of new eye floaters in Long COVID and ME/CFS

by Patient-Led Research Collaborative

Cite as: Patient-Led Research Collaborative (2023). Community polls: Informal social media polls: SARS-CoV-2 reinfection and Long COVID, and the presence of new eye floaters in Long COVID and ME/CFS, Patient-Generated Hypotheses Journal for Long COVID & Associated Conditions, Vol. 1, 49-55

Abstract

Generating and sharing polls on social media is one way that patient communities can informally test their theories and give researchers ideas about what to further explore. In February 2023, the Patient-Led Research Collaborative conducted two sets of informal polls on Twitter and Mastodon. One question was about Long COVID and reinfections, and the other asked about new eye floaters (myodesopsias) in patients with Long COVID and myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). This brief article describes the results of each poll: 10.2% of Twitter respondents and 20% of Mastodon respondents reported experiencing Long COVID after a reinfection, rather than a first infection, and more than half of Twitter respondents with Long COVID or ME/CFS answered “Yes” or “Maybe” that they noticed new eye floaters as a symptom. Both polls had limitations: Neither was intended to establish incidence or prevalence data, and the poll respondents are likely not representative of the overall Long COVID and ME/CFS communities. Despite this, the results provide potentially useful information for Long COVID and ME/CFS research and policy.


Introduction

Social media has been widely used for health purposes in the last decade, especially during the ongoing COVID pandemic1. Many platforms are important for patients and clinicians alike. A key way that patients source information about their illnesses or test their hypotheses among the patient community is by creating and sharing polls on social media. These polls come with limitations, however. For example, Twitter polls are unlikely to reach a broad sample audience, and the generalization of their results is limited2. Additionally, polls on Twitter are limited to four response options, and until very recently, posts could only be a maximum of 280 characters. Polls on Mastodon face similar constraints: For example, they also have only four response options. Despite the limitations to the conclusions one can draw from these polls, they reveal insights that can help patients and researchers pursue hypotheses through more formalized research avenues. In February 2023, the Patient-Led Research Collaborative conducted two sets of informal polls on Twitter and Mastodon. Both sets of polls were anonymous: Twitter and Mastodon do not reveal who participates in their polls. We conducted these polls to test questions of importance to the Long COVID patient community and/or to build off hypotheses published as part of the first issue of the Patient-Generated Hypotheses Journal.

Long COVID after reinfections

The first set of polls was a question of importance to the Long COVID patient community. We asked whether an initial COVID-19 infection or a reinfection caused the onset of Long COVID symptoms (see Figure 1). There is limited data on reinfections and Long COVID. Bowe et al. (2022) and Hadley et al. (2023) reviewed electronic health records (EHRs) of patients and established that reinfections can cause an onset of Long COVID symptoms3,4. Bowe et al. (2022) found that “the risks of adverse health outcomes increased as the number of infections increased” in the acute and subacute phase (p. 2399). However, in the study, patients were only followed for six months after infection. Additionally, Bowe et al. (2022) included only patients with a positive SARS-CoV-2 test. Patients who may have been infected, but either were not tested or whose test showed a false negative, were not included in the analysis. This may have caused an underestimation of reinfection risks.

A systematic review published by Pecoraro et al. (2021) calculated that up to 58% of COVID patients may initially have a false negative polymerase chain reaction (PCR) SARS-CoV-2 test5. This means that there are limitations to the conclusions that can be drawn from studies which review EHRs for positive tests. Overall, there is limited information about the number of times people have been infected with SARS-CoV-2 and the risks associated with those reinfections. This informal poll collected additional data on the topic.

We asked users on Twitter and Mastodon the following question with results provided in-line and in Figure 1:

Twitter poll

At which infection did you first experience Long COVID symptoms? (n=664)

  • 1st infection: 89.9%
  • 2nd infection: 7.7%
  • 3rd infection: 1.1%
  • 4th infection or higher: 1.4%

Note: The results may not total to 100% due to rounding.

Mastodon poll

At which infection did you first experience Long COVID symptoms? (n=607)

  • 1st infection: 80.0%
  • 2nd infection: 14.0%
  • 3rd infection: 3.0%
  • 4th infection or higher: 3.0%

Figure 1. Percentage of Respondents with Long COVID Who Developed Long COVID after a 1st, 2nd, 3rd, or 4th Infection or Higher

We assumed that users who responded to these polls self-identify as having Long COVID. 89.9% of respondents on Twitter and 80.0% of respondents on Mastodon reported experiencing Long COVID after their first infection. 10.2% of Twitter respondents and 20% of Mastodon respondents reported experiencing Long COVID after a reinfection. Sample sizes between the platforms were similar with n=664 on Twitter and n=607 on Mastodon.

Note that this data does not indicate the likelihood of developing Long COVID after each infection. Rather, the data demonstrates after which infection a small sample of people in an online community developed Long COVID. There are important limitations to this data: for example, the poll did not allow respondents to indicate if they developed Long COVID after a first infection, then recovered, then developed it again after a subsequent infection. It is possible that those active in Long COVID social media communities have been sick for many months and are more likely to have developed Long COVID from their first infection. The poll also assumes that everyone who responded has Long COVID, with inclusion criteria not verified. Additionally, many people may not realize that they have Long COVID and therefore would be unable to respond to such a poll on social media.

Overall, however, the results suggest that Long COVID from reinfections does occur, and all people are at some risk of Long COVID, even if their prior infection(s) did not cause it. Bowe et al. (2022) and Hadley et al. (2023) support this conclusion3,4. Bowe et al. (2022) found that negative health outcomes “increased in a graded fashion according to the number of infections” (p. 2399). This has implications for both COVID mitigation policy and research. Transmissibility of SARS-CoV-2 increases frequently from viral mutations, and many people have been infected multiple times. Therefore, it is more important than ever to mitigate transmission and research the effects of multiple infections on mortality and long-term health outcomes, including Long COVID.

Eye floaters

The second set of polls we conducted was in response to Matt Mazewski’s hypothesis, which is included in this publication. The polls ask about the presence of new eye floaters (myodesopsias) in patients with Long COVID and ME/CFS. Because the poll refers to this condition by the name “eye floaters,” for consistency, this article will continue to use that language. As Mazewski hypothesizes, in both Long COVID and ME/CFS, inflammatory triggers may cause collagen degradation that leads to the presence of eye floaters6.

We asked users on Twitter and Mastodon the following questions with results provided in-line. Twitter results are also shown in Figure 2:

Twitter poll

If you have Long COVID, did you notice new eye floaters (spots in your vision that may look to you like black or gray specks, strings, or cobwebs) as one of its symptoms? (n=544 subtracting “see results”)

  • Yes: 37.1%
  • No: 49.5%
  • Maybe: 13.5%

Note: The results may not total to 100% due to rounding.

If you have ME/CFS, did you notice new eye floaters (spots in your vision that may look to you like black or gray specks, strings, or cobwebs) as one of its symptoms? (n=281, subtracting “see results”)

  • Yes: 38.5%
  • No: 45.2%
  • Maybe: 16.3%

Mastodon poll

If you have Long COVID and/or ME/CFS, did you notice new eye floaters (spots in your vision that may look to you like black or gray specks, strings, or cobwebs) as one of your symptoms? (n=76, subtracting “see results”)

  • Yes: 42.9%
  • No: 44.4%
  • Maybe: 12.7%

Note: Mastodon results not shown in Figure 2 due to small sample size.

Figure 2. Existence of New Eye Floaters among Twitter respondents with Long COVID and ME/CFS

More than half of Twitter respondents with Long COVID or ME/CFS answered “Yes” or “Maybe” that they noticed new eye floaters as a symptom on the Twitter polls.

On Mastodon, where there is limited functionality when posting a thread with multiple polls, ME/CFS and Long COVID were grouped together in order to minimize confusion and obtain a larger sample size. Despite this, the sample size (n=76) was too small to include in our graph or analysis. On Twitter, the sample size was larger for the Long COVID question than the ME/CFS question: n=544 (Long COVID) and n=281(ME/CFS).

There are limitations to this informal poll. Given that a relatively small number of people on social media responded, it is not a representative sample of the larger Long COVID or ME/CFS populations. It is also not intended to provide incidence or prevalence data. Further, those who experienced eye floaters may have been more likely to answer the question, which may have skewed the data. The poll also assumes that everyone who responded has Long COVID or ME/CFS, but inclusion criteria were not verified. Despite these limitations, the response still indicates that many in the Long COVID and ME/CFS population do experience new eye floaters, and this could be an important topic for future research that is currently overlooked.  As Mazewski discusses in his hypothesis, it is also possible that having eye floaters is a risk factor for Long COVID and ME/CFS.

Conclusion

Informal social media polls, while limited in many ways, are important for exploring health-related phenomena that patient groups may notice more readily than the general public. Creating these polls and disseminating results may allow patients and researchers to pursue areas of particular importance sooner than they otherwise would have. This benefits both patients and the larger medical community.

The role of reinfections in Long COVID is of great importance to the Long COVID community as well as to the public at large. The poll results suggest that while most people who are active in the Long COVID communities on Twitter and Mastodon experienced Long COVID symptoms at their first infection, reinfections did cause an onset of Long COVID symptoms for a significant number of people (10.2-20%). Bowe et al. (2022) established that multiple infections with SARS-CoV-2 increase the risk for adverse health outcomes, including Long COVID3. There is a need for widespread public health warnings about the risks of COVID reinfection, not only in regard to mortality, but also to Long COVID.

In our second set of polls, more than half of poll respondents with Long COVID or ME/CFS answered “Yes” or “Maybe” that they experienced new eye floaters as a symptom, indicating that this is perhaps an avenue worthy of future research, despite the limitations of polls discussed previously. The topic serves as one example of an area of study where patients are, perhaps, more aware of a health phenomenon than many scientists or researchers may be. Continuing to develop hypotheses, testing them informally via social media polls, and then disseminating the results to a larger audience is important for all patient communities, but especially for the patient communities of Long COVID and ME/CFS. These conditions are chronically underfunded and still not well-known among the medical community, despite the fact that Long COVID and ME/CFS affect tens of millions of people. Informal social media polls and wider dissemination of their results may be one way to move the needle toward more public awareness, research, and funding for Long COVID and ME/CFS.


References

1 Chen, J., & Wang, Y. (2021). Social media use for health purposes: Systematic review. Journal of Medical Internet Research, 23(5), e17917. https://doi.org/10.2196/17917

2 Mondal, S., Modak, P. K., Selim, M., Mondal, H., Baidya, C., Hribersek, M., Singla, R. K., Shen, B., & Atanasov, A. G. (2022). Twitter poll as a medium for questionnaire-based health survey: An experience of a pilot study on the preference of systems of medicine for various health conditions. Cureus, 14(9). https://doi.org/10.7759/cureus.28767

3 Bowe, B., Xie, Y., & Al-Aly, Z. (2022). Acute and postacute sequelae associated with SARS-CoV-2 reinfection. Nature Medicine, 28(11), 2398–2405. https://doi.org/10.1038/s41591-022-02051-3

4 Hadley, E. E., Yoo, Y. J., Patel, S., Zhou, A. G., Laraway, B., Wong, R., Preiss, A. J., Chew, R., Davis, H., Chute, C. G., Pfaff, E. R., Loomba, J., Haendel, M. A., Hill, E. L., & Moffitt, R. A. (2023). SARS-CoV-2 reinfection is preceded by unique biomarkers and related to initial infection timing and severity: An N3C RECOVER EHR-based cohort study. MedRxiv. https://doi.org/10.1101/2023.01.03.22284042

5 Pecoraro, V., Negro, A., Pirotti, T., & Trenti, T. (2021). Estimate false‐negative RT‐PCR rates for SARS‐CoV‐2. A systematic review and meta‐analysis. European Journal of Clinical Investigation, 52(2). https://doi.org/10.1111/eci.137066 Proal, A. D., & VanElzakker, M. B. (2021). Long COVID or post-acute sequelae of COVID-19 (PASC): An overview of biological factors that may contribute to persistent symptoms. Frontiers in Microbiology, 12. https://doi.org/10.3389/fmicb.2021.698169