Persistent symptoms following SARS-CoV-2 infection in a random community sample of 508,707 people

Summary Background Long COVID, describing the long-term sequelae after SARS-CoV-2 infection, remains a poorly defined syndrome. There is uncertainty about its predisposing factors and the extent of the resultant public health burden, with estimates of prevalence and duration varying widely. Methods Within rounds 3–5 of the REACT-2 study, 508,707 people in the community in England were asked about a prior history of COVID-19 and the presence and duration of 29 different symptoms. We used uni-and multivariable models to identify predictors of persistence of symptoms (12 weeks or more). We estimated the prevalence of symptom persistence at 12 weeks, and used unsupervised learning to cluster individuals by symptoms experienced. Findings Among the 508,707 participants, the weighted prevalence of self-reported COVID-19 was 19.2% (95% CI: 19.1,19.3). 37.7% of 76,155 symptomatic people post COVID-19 experienced at least one symptom, while 14.8% experienced three or more symptoms, lasting 12 weeks or more. This gives a weighted population prevalence of persistent symptoms of 5.75% (5.68, 5.81) for one and 2.22% (2.1, 2.26) for three or more symptoms. Almost a third of people (8,771/28,713 [30.5%]) with at least one symptom lasting 12 weeks or more reported having had severe COVID-19 symptoms (“significant effect on my daily life”) at the time of their illness, giving a weighted prevalence overall for this group of 1.72% (1.69,1.76). The prevalence of persistent symptoms was higher in women than men (OR: 1.51 [1.46,1.55]) and, conditional on reporting symptoms, risk of persistent symptoms increased linearly with age by 3.5 percentage points per decade of life. Obesity, smoking or vaping, hospitalisation, and deprivation were also associated with a higher probability of persistent symptoms, while Asian ethnicity was associated with a lower probability. Two stable clusters were identified based on symptoms that persisted for 12 weeks or more: in the largest cluster, tiredness predominated, while in the second there was a high prevalence of respiratory and related symptoms. Interpretation A substantial proportion of people with symptomatic COVID-19 go on to have persistent symptoms for 12 weeks or more, which is age-dependent. Clinicians need to be aware of the differing manifestations of Long COVID which may require tailored therapeutic approaches. Managing the long-term sequelae of SARS-CoV-2 infection in the population will remain a major challenge for health services in the next stage of the pandemic. Funding The study was funded by the Department of Health and Social Care in England. Research in context Evidence before this study Recent systematic reviews have documented the wide range of symptoms and reported prevalence of persistent symptoms following COVID-19. A dynamic review of Long COVID studies (NIHR Evidence) in March 2021 summarised the literature on the prevalence of persistent symptoms after acute COVID19, and reported that most studies (14) were of hospitalised patients, with higher prevalence of persistent symptoms compared with two community-based studies. There was limited evidence from community studies beyond 12 weeks. Another systematic review reported a median of over 70% of people with symptoms lasting at least 60 days. A review of risk factors for Long COVID found consistent evidence for an increased risk amongst women and those with high body mass index (BMI) but inconsistent findings on the role of age and little evidence concerning risks among different socioeconomic or ethnic groups which are often not well captured in routine healthcare records. Long COVID is increasingly recognised as heterogenous, likely underpinned by differing biological mechanisms, but there is not yet consensus on defining subtypes of the condition. Added value of this study This community-based study of over half a million people was designed to be representative of the adult population of England. A random sample of adults ages 18 years and above registered with a GP were invited irrespective of previous access to services for COVID-19, providing an estimate of population prevalence that was representative of the whole population. The findings show substantial declines in symptom prevalence over the first 12 weeks following Covid-19, reported by nearly one fifth of respondents, of whom over a third remained symptomatic at 12 weeks and beyond, with little evidence for decline thereafter. Risk factors identified for persistent symptoms (12 weeks or more) suggestive of Long COVID confirm some previous findings - an increased risk in women, obese and overweight individuals and those hospitalised for COVID-19, with strong evidence for an increasing risk with age. Additional evidence was found for an increased risk in those with lower income, smoking or vaping and healthcare or care home workers. A lower risk was found in those of Asian ethnicity. Clustering identified two distinct groups of individuals with different symptom profiles at 12 weeks, highlighting the heterogeneity of clinical presentation. The smaller cluster had higher prevalence of respiratory and related symptoms, while for those in the larger cluster tiredness was the dominant symptom, with lower prevalence of organ-specific symptoms. Implications of available evidence There is a high prevalence of persistent symptoms beyond 12 weeks after acute COVID-19, with little evidence of decline thereafter. This highlights the needs for greater support for patients, both through specialised services and, for those from low-income settings, financial support. The understanding that there are distinct clusters of persistent symptoms, the most common of which is dominated by fatigue, is important for the recognition and clinical management of the condition outside of specialised services.