Prevalence of long COVID in the general adult population according to different definitions and socio-demographic and infection characteristics. A nationwide random sampling survey in France in autumn 2022.

Prévalence du COVID long dans la population adulte générale selon différentes définitions et selon les caractéristiques de l’infection et sociodémographique. Une enquête nationale par sondage aléatoire à l'automne 2022.

Publié le 27 mai 2024

As early as April 2020, subjects infected with SARS-CoV-2 reported persistent and very diverse symptoms of a condition described as "long COVID". Since then, several definitions of the condition have been proposed, mainly based on symptoms and the time elapsed since infection.

In October 2021, the World Health Organization (WHO) proposed a definition of what it called "post-COVID-19 condition". This symptom-based definition was established by a group of experts using the Delphi consensus methodology. A broad spectrum of symptoms was identified in the course of this study, ranging from very common symptoms (fatigue and breathlessness) to more specific symptoms such as taste and smell disorders, or non-specific symptoms such as joint pain and menstrual disorders.

However, a number of other definitions are still used in population-based surveys, producing highly variable estimates of the prevalence of long COVID, which further complicates public health decision-making.

An initial study carried out by Santé publique France in March 2022 estimated the prevalence of long COVID at 4% of the general French adult population. The second study, which has just been published [1] in the journal Clinical Microbiology and Infection, carried out in collaboration with teams from National Institute of Health and Medical Research (Inserm) and the hospital network AP-HP, has updated this estimate and clarified the main socio-demographic determinants of the condition, taking into account several definitions and severity thresholds, as well as the perception of having had long COVID. It was carried out in autumn 2022, after the major waves of infections by Omicron variants of SARS-CoV-2.

3 questions for: Joël Coste, Non-Communicable Diseases and Trauma Division, Santé publique France

Visuel d'illustration

The epidemiology of long COVID is complicated by the fact that so many definitions are used around the world. In your study, you used several: why, and what do they teach you?

We have in fact adopted the WHO definition of post-COVID-19 condition, which was used in the first study carried out at the beginning of 2022, in order to be able to compare the situation in France before and after the major waves of SARS-CoV-2 infection linked to the Omicron variant. In order to compare our results with those obtained in the United Kingdom and the United States, we have also used the definitions of these two countries. In addition, to assess the burden of long COVID more accurately, we distinguished several forms of post-COVID-19 condition according to the level of impact (high, moderate, no impact) of the selected symptoms on activities of daily life. Finally, we asked survey participants whether they "believed they had been affected by a long form of Covid-19" (reported long COVID or perceived long COVID).

The concordance of the categorisations derived from the different definitions appeared weak or moderate, with Kappa coefficients (worth 0 in the case of concordance due to chance alone and 1 in the case of perfect concordance) ranging from 0.18 to 0.59. For example, 47% and 29% of people who met the criteria for the American (National Centre for Health Statistics, US-NCHS) and English (Office for National Statistics, UK-ONS) definitions also met the WHO definition of post-COVID-19 condition. Similarly, less than half (43%) of people meeting the WHO's definition thought they had been affected by long COVID. These discrepancies, which affected both sexes and all age groups, led us to present prevalence estimates according to the different definitions used to date in population-based studies. The WHO's definition of post-COVID-19 condition offers us the advantage of being able to distinguish between different levels of impact of symptoms on daily activities.

What are your conclusions about the burden of long COVID? Have you identified the groups most at risk of developing long COVID?

The prevalence of post-COVID-19 condition, according to the WHO definition, has been estimated at 4.0% of the French adult population and at 2.4% and 1.2% respectively if we consider the forms with moderate and high impacts on people's daily lives. Among people with post-COVID-19 condition, 31% had been experiencing symptoms for more than twelve months, and 22% for more than eighteen months. In addition, the prevalence of long COVID reported was 7.1%, and 7.6% and 13.4% for the American and English definitions, respectively.
It was higher among women (5.4%) and lower among the elderly (2.2% for the over 65s and 1.7% for the over 75s) and people living alone (2.5%). These differences were observed for all the definitions used. They mainly reflect higher exposure to SARS-CoV-2 among women and lower exposure among the elderly and people living alone.

Among people whose infection with SARS-CoV-2 dated back at least three months, the prevalence of post-COVID-19 condition according to the WHO definition was 8%, ranging from 5.3% (men) to 10.2% (women), 14.9% (jobseekers) and 18.6% (after hospitalisation for COVID-19); 21% of these people had been infected during the Delta wave and 53% during the Omicron waves.
While the prevalence of post-COVID-19 condition remained at 4% in the general population between the beginning and end of 2022, the prevalence of post-COVID-19 condition calculated among participants who reported at least one infection in that year fell sharply: from 30% observed at the beginning of 2022 to 8% at the end of the year. The 2022 Omicron variants (or the context of their occurrence in vaccinated people) have in fact been associated, in several studies, with a lower risk of post-COVID-19 condition than earlier variants. This explains why the prevalence of long Covid in the general population has remained stable (4%), while the proportion of the French population infected with SARS-CoV-2 rose from 13% to 48% from the beginning to the end of 2022 (infections dating at least three months).

What can this new knowledge contribute to prevention and care for these people? Why is epidemiological monitoring of this condition important?

Our analyses enabled us to identify many and varied risk factors for long COVID, of a demographic, social, occupational, pathological (co-morbidities), behavioural and perceptual nature, and linked to SARS-CoV-2 infection. The multitude of factors suggests that this condition should not only be considered as a direct complication of SARS-CoV-2 infection, but conceived and represented in a broad network of contextual, medical, psychological, professional and social factors that increase the individual risk of long COVID, over and above coronavirus infection. In addition to vaccination and certain behaviours that protect against SARS-CoV-2 infection, these contextual factors should be given greater weight in strategies aimed at limiting the burden of long COVID in the population. Analyses are still underway to clarify the impact of the condition on quality of life and identify population groups, whether clinical (identifiable by particular symptoms) or socio-demographic, at risk of severe deterioration in quality of life.

Despite the stabilisation of prevalence in France in 2022, the question of epidemiological monitoring for long COVID still arises, particularly for post-COVID-19 condition, although monitoring has become difficult given the sharp decline in the use of tests to detect SARS-CoV-2 infection. The latest prevalence estimates obtained in the United States, where monthly monitoring of long COVID continues, show that the phenomenon was not in decline at the start of 2024. The question of epidemiological monitoring or surveillance arises particularly in the case of prolonged forms (31% of cases), and those whose symptoms have a high or very high impact on daily activities (30%), and probably on care and social protection systems.

The WHO definition of post-COVID-19

Post-COVID-19 corresponds to people presenting the following four criteria:

  • A history of probable or confirmed infection with SARS-CoV-2 dating back at least three months 
  • Symptoms that have persisted for at least two months
  • Symptoms that cannot be explained by another diagnosis
  • Symptoms that generally have an impact on daily activities

Learn more: Soriano, J. B., S. Murthy, J. C. Marshall, P. Relan, and J. V. Diaz. 2022. A clinical case definition of post-COVID-19 condition by a Delphi consensus. Lancet Infect Dis 22 (4):e102-e107. doi: 10.1016/s1473-3099(21)00703-9.

Further information:

  • Different definitions of long COVID:
    • Soriano, J. B., S. Murthy, J. C. Marshall, P. Relan, and J. V. Diaz. 2022. "A clinical case definition of post-COVID-19 condition by a Delphi consensus." Lancet Infect Dis 22 (4):e102-e107. doi: 10.1016/s1473-3099(21)00703-9.
    • Nehme, M., O. Braillard, G. Alcoba, S. Aebischer Perone, D. Courvoisier, F. Chappuis, and I. Guessous. 2021. "COVID-19 Symptoms: Longitudinal Evolution and Persistence in Outpatient Settings." Ann Intern Med 174 (5):723-725. doi: 10.7326/m20-5926.
    • Robineau, O., E. Wiernik, C. Lemogne, X. de Lamballerie, L. Ninove, H. Blanché, J. F. Deleuze, C. Ribet, S. Kab, M. Goldberg, G. Severi, M. Touvier, M. Zins, and F. Carrat. 2022. "Persistent symptoms after the first wave of COVID-19 in relation to SARS-CoV-2 serology and experience of acute symptoms: A nested survey in a population-based cohort." Lancet Reg Health Eur 17:100363. doi: 10.1016/j.lanepe.2022.100363.