Ziauddeen, Nida, Gurdasani, Deepti, O'Hara, Margaret E., Hastie, Claire, Roderick, Paul, Yao, Guiqing and Alwan, Nisreen (2021) Characteristics of Long Covid: findings from a social media survey. medRxiv.
Abstract
Many people are not recovering for months after being infected with SARS-CoV-2. Long Covid has emerged as a major public health concern that needs defining, quantifying, and describing. We aimed to explore the initial and ongoing symptoms of Long Covid following SARS-CoV-2 infection and describe its impact on daily life in people who were not admitted to hospital during the first two weeks of the illness. We co-produced a survey with people living with Long Covid. We collected the data through an online survey using convenience non-probability sampling, with the survey posted both specifically on Long Covid support groups and generally on social media. The criteria for inclusion were adults with lab-confirmed (PCR or antibody) or suspected COVID-19 managed in the community (non-hospitalised) in the first two weeks of illness. We used agglomerative hierarchical clustering to identify specific symptom clusters, and their demographic and functional correlates.
We analysed data from 2550 participants with a median duration of illness of 7.7 months (interquartile range (IQR) 7.4-8.0). The mean age was 46.5 years (standard deviation 11 years) with 82.8% females and 79.9% of participants based in the UK. 89.5% described their health as good, very good or excellent before COVID-19. The most common initial symptoms that persisted were exhaustion, chest pressure/tightness, shortness of breath and headache. Cough, fever, and chills were common initial symptoms that became less prevalent later in the illness, whereas cognitive dysfunction and palpitations became more prevalent later in the illness. 26.5% reported lab-confirmation of infection. The biggest difference in ongoing symptoms between those who reported testing positive and those who did not was loss of smell/taste. Ongoing symptoms affected at least 3 organ systems in 83.5% of participants.
Most participants described fluctuating (57.7%) or relapsing symptoms (17.6%). Physical activity, stress and sleep disturbance commonly triggered symptoms. A third (32%) reported they were unable to live alone without any assistance at six weeks from start of illness. 16.9% reported being unable to work solely due to COVID-19 illness. 66.4% reported taking time off sick (median of 60 days, IQR 20, 129). 37.0% reported loss of income due to illness, and 64.4% said they were unable to perform usual activities/duties.
Acute systems clustered broadly into two groups: a majority cluster (n=2235, 88%) with cardiopulmonary predominant symptoms, and a minority cluster (n=305, 12%) with multisystem symptoms. Similarly, ongoing symptoms broadly clustered in two groups; a majority cluster (n=2243, 88.8%) exhibiting mainly cardiopulmonary, cognitive symptoms and exhaustion, and a minority cluster (n=283, 11.2%) exhibited more multisystem symptoms. Belonging to the more severe multisystem cluster was associated with more severe functional impact, lower income, younger age, being female, worse baseline health, and inadequate rest in the first two weeks of the illness, with no major differences in the cluster patterns when restricting analysis to the lab-confirmed subgroup.
This is an exploratory survey of Long Covid characteristics. Whilst it is important to acknowledge that it is a non-representative population sample, it highlights the heterogeneity of persistent symptoms, and the significant functional impact of prolonged illness following confirmed or suspected SARS-CoV-2 infection. To study prevalence, predictors and prognosis, research is needed in a representative population sample using standardised case definitions (to include those not lab-confirmed in the first pandemic wave).
This record has no associated files available for download.
More information
Identifiers
Catalogue record
Export record
Contributors
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.