What Does COVID-19 Recovery Actually Look Like?
An Analysis of the Prolonged COVID-19 Symptoms Survey by Patient-Led Research Team
Generated from survey data organized by decentralized team of COVID-19 patients, exported on May 2, 2020 (640 Responses)
Report Released: May 11th, 2020 by
Report created and written by volunteers from the COVID-19 Body Politic Slack Group including: Gina Assaf, Hannah Davis, Lisa McCorkell, Hannah Wei., O’Neil Brooke, Athena Akrami, Ryan Low, Jared Mercier, and Adetutu A.
Survey Authors and Contributors Include: Gina Assaf., Tina L., Annie C., Monica S., Jared Mercier, Lauren N., Noel H., JD Davids, and Susie.
The goal of this research is to capture and share a bigger picture of the experiences of patients suffering from COVID-19 with prolonged symptoms using a data driven approach. The survey content and research analysis are “patient-centric,” conducted through participatory type research. Survey questions and symptoms were aggregated and curated by patients themselves with expertise in research and survey design. Analysis was also conducted by patients themselves with expertise in both quantitative and qualitative data analysis. This approach is especially important for COVID-19 because patients experiencing symptoms are in need of timely research and content relevant to them that is not currently available due to the novelty of this virus.
It is important to note that this survey targeted patients with prolonged symptoms (patients with symptoms for over two weeks), though it does include a few responses from patients with shorter symptom duration. The survey was primarily distributed to the Body Politic COVID-19 Support Group on Slack, the Facebook Survivor Corps group, and through other personal social media accounts. The survey received 640 responses and was open from April 21st to May 2nd, 2020.
About a quarter of respondents (23.1%) tested positive for COVID-19, 27.5% tested negative, and the remainder (47.8%) were not tested. In our analysis, we included all responses regardless of testing status. Due to the severe lack of testing available in many areas and the prevalence of false negatives, we do not believe people’s experiences with COVID-19 symptoms should be discounted because they did not receive a positive test result. When disaggregating the data, we found that the main difference between people who tested positive and people who tested negative was not necessarily the symptoms they experienced (as most symptoms were similar), but how early in their illness they were able to be tested. We believe that current testing practices are not capturing a large subset of COVID-19 patients and that more investigation into this is required. If we limited the analysis to only respondents who received a positive test result, we could lose valuable information, and would be doing a disservice to the thousands, if not hundreds of thousands, of people with prolonged symptoms who were unable to access a test early enough in their illness, if at all. We believe future research must consider the experiences of all people with COVID-19 symptoms, regardless of testing status, in order to better understand the virus and underscore the importance of early and widespread testing.
When considering the results of this survey, it is important to keep in mind that this sample is not representative of all COVID-19 patients. Sampling bias is at work here: both in who would be willing and able to take a survey, and who would have exposure to the survey. We consider this sample to be disproportionately, white, cis-gender female and U.S.-based; we plan to intentionally conduct broader outreach to create a subsequent version of the survey and report with a more diverse group of respondents. Further, unless indicated, we have not completed significance testing on our findings. Therefore, our results should not be taken as being representative of the COVID-19 experience.
With that in mind, our patient-led analysis on COVID-19 experience is the first of its kind. We believe our participatory research may shed light on the short and long term effects of this virus that is currently underrepresented in academic and media reports. While the medical community’s primary focus is on those with acute severe conditions, respondents of this survey represent those with mild to moderate symptoms with a prolonged recovery who may or may not have received medical support in their region. Our data analysis, alongside anecdotal reports in the Body Politic support group, captures the following:
* Impacts of the timing of testing on the status of results
* Cyclical symptoms experienced unexpectedly for over six weeks at time of response
* The nature, severity, and recovery time course of symptoms week by week
* An analysis on differences in time course of symptoms reported by patients with positive and negative test results
* Internal and external stigmas experienced by patients in recovery
* Impacts on lifestyle including physical activity and self-isolation
We recognize and want to emphasize that COVID-19 appears differently in each person, and we hope that this report can help medical professionals and researchers understand more about this virus from the patients’ perspective, and be a jumping off point for further research. We also hope that our analysis will give family, friends, employers, coworkers, and the general public better insight into some of the experiences of people living with COVID-19.
This version of the survey is now closed. We have listed future plans for this research we hope to address with more resources in the section on Future Work.
Highlights from the Survey Analysis
* Almost half of the respondents (47.8%) were not tested (denied testing or did not have access to a test) and the other half were nearly split between those who tested negative (yet still presented COVID-19 symptoms) and those who tested positive for COVID-19.
* Respondents who reported testing positive were tested earlier in their illness on average (by day 10) than those who reported testing negative (on average tested by day 16). This difference between the two populations is statistically significant. This could reinforce the importance of testing symptomatic people early. See the Testing Section for additional information.
* Over half of respondents (57.8%) listed at least one pre-existing condition, with the most prevalent conditions being asthma and vitamin D deficiency. Our analyses suggest pre-existing asthma might prolong recovery time. See the Pre-existing Conditions Section, and Recovery Timecourse Section for additional information.
* The vast majority of participants with symptoms experienced fluctuations both in the type (70% reporting) and intensity (89% reporting) of symptoms over the course of being symptomatic. See the Nature and Severity of Symptoms Section for additional information.
* At the time they took the survey, 90.6% of the respondents had not recovered (Recovery was Self-Interpreted for this Survey). For the 60 respondents who had recovered, the average length of time of being symptomatic was 27 days. The respondents who had not recovered had been experiencing symptoms for an average of 40 days, with a large proportion experiencing symptoms for 5-7 weeks. Our “survival analysis”, shows that the chance of full recovery by day 50 is smaller than 20%. See the Recovery Timecourse Section for additional information.
* On Symptoms (See the Symptoms Section for additional information)
* A slightly elevated temperature (above 98.6°F but below 100.1°F) was more commonly reported than higher temperatures (100.1°F and above).
* In addition to the more widely recognized COVID-19 symptoms such as fever/elevated temperature, cough, and shortness of breath, the other symptoms that were highly reported by participants included fatigue (varying in severity), brain fog/concentration challenges, chills/sweats, trouble sleeping, and a loss of appetite. It is interesting to note that neurological symptoms were consistently reported by patients for eight weeks, specifically brain fog/concentration challenges and trouble sleeping.
* Patients from this survey indicate that their first week of experiencing symptoms had milder/fewer symptoms than weeks 2 and 3. Gastrointestinal symptoms and chills/sweats seem to occur more often in weeks 1-2 and respiratory symptoms appear relatively consistent through weeks 3-4.
* Top 10 symptoms reported over the course of the 8 weeks by this group ranked by highest to lowest are: mild shortness of breath, mild tightness of chest, moderate fatigue, mild fatigue, chills or sweats, mild body aches, dry cough, elevated temperature (98.8-100), mild headache, and brain fog/concentration challenges.
* For this group, when comparing respondents that tested positive versus negative, only 2 of 62 symptoms (loss of smell and loss of taste) came back statistically significant, even when accounting for the point at which the respondent was tested. All other symptoms, including dry cough, loss of appetite, shortness of breath, tightness of chest, elevated temperature, fever (100.1 or greater), GI symptoms, lung burn, elevated heart rate and tachycardia, brain fog/concentration challenges, and dizziness, were not statistically different between the groups reporting positive test results and negative test results.
* Neurological symptoms are being underreported in the media. These include brain fog, concentration challenges, memory loss, seizures, dizziness and problems with balance, various forms of insomnia, and others. Brain fog and concentration challenges were a more common symptom than cough during most weeks, as was insomnia.
* Most of the respondents were not hospitalized. Many did visit the ER/urgent care to seek treatment, but they were not admitted