Study population and setting
This preprint provides extensive data on the long-term (through 7 months) experience and recovery process of individuals with confirmed or suspected COVID-19, using data collected from online surveys administered from September 6 to November 25, 2020. Survey links were disseminated over social media and through COVID-19 patient support groups. The study included 3,762 adults (79% women, 85% white, age range: 30-60 years) from 56 countries (41% US, 35% UK and Ireland), who self-reported symptoms consistent with COVID-19, and whose illness lasted longer than 28 days (with symptom onset occurring between December 2019 and May 2020). Participants were not required to have diagnostic or antibody testing for SARS-CoV-2. Participants reported symptom severity (none, very mild, mild, moderate, severe, very severe) and presence of symptoms during specific time intervals after initial symptom onset (weekly intervals for first 4 weeks, then monthly intervals through month 7) for 66 symptoms.
Summary of Main Findings
Among the 3,762 participants, 2,330 were ever tested for SARS-CoV-2; only 1,020 had a positive RT-PCR, antigen, or antibody test. Symptom prevalence and trends were similar between those who ever tested positive for SARS-CoV-2 and those who did not; loss of smell or taste was more common among those who tested positive. 3,505 (93%) participants were still experiencing symptoms at the time of survey completion, including 2,454 respondents who had experienced symptoms for at least 6 months, and 257 recovered. Among those with symptoms lasting more than 3 months, symptom peak occurred in month 2 (mean number of symptoms reported=17.2, 95% CI: 16.5-17.8). When patients were asked to list their most debilitating symptoms, the top three responses were fatigue, breathing issues, and cognitive dysfunction. The most prevalent and persistent symptom by week 28 was fatigue with a prevalence around 80% (all symptom prevalence estimates were conditional on no recovery having yet occurred by a given time); followed by brain fog/memory issues and other neurological sensations (~60%); headaches, insomnia, shortness of breath, palpitations, and tachycardia each had a prevalence near 40% though week 28. The authors identified 3 clusters of symptoms by the trajectory of their self-reported prevalence over time. Those in cluster 1 (including diarrhea, loss of appetite, vomiting, runny nose, sore throat, dry cough, and fever) were most common during early illness onset and then decreased over time. Those in cluster 2 (including chest pain/tightness, abdominal pain, muscle aches, confusion, headaches, chills, and fatigue) gradually increased during the first few weeks with a slow decrease or no change over time. Those in cluster 3 (including palpitations, dermatologic effects, constipation, hearing/vision alterations, joint pain, brain fog, and memory issues) increased over the first two months and then persisted. Only 27% of unrecovered patients and 49% of recovered patients reported working as many hours as they had done prior to becoming ill.
This study included a relatively large number of participants and collected extensive data on a wide range of symptom data, including severity and trends over time.
This a preprint that has not yet been peer reviewed, so all findings should be interpreted with caution. All data were based on self-reported symptoms and COVID-19 testing results, which are subject to the standard limitations of self-reported data. Importantly, participants self-selected into the study, which limits generalizability and may induce a sampling bias toward individuals who actively sought support groups for their condition. The sample is primarily female and primarily white, which is generally not representative of the larger population of individuals with COVID-19. Persistent symptom trends should not be interpreted as prevalences among all individuals who had COVID-19, but rather represent the experience of a highly selected group of individuals who had symptoms lasting more than 28 days and elected to participate in the online survey. The majority of participants did not receive a positive RT-PCR, antigen, or antibody test for COVID-19, so it is possible that other conditions contributed to the symptoms experienced.
This is the largest known report of patients experiencing symptoms consistent with “long COVID,” with detailed data on duration and severity of symptoms in the months following confirmed or suspected COVID-19.
This review was posted on: 22 January 2021