Global prevalence of post-COVID-19 condition (Long COVID): a systematic review and meta-analysis of observational studies

Samal Kassymbek, Aigul Abduldayeva, Aigul Abduldaeva, Nikolay Safonov

Frontiers in Public Health, 2026

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.3389/fpubh.2026.1839646
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Background Long COVID is an umbrella term for persistent, relapsing, or newly developed health problems after SARS-CoV-2 infection, whereas post-COVID-19 condition (PCC) refers more specifically to the World Health Organization clinical case definition. Reported prevalence varies substantially because of differences in terminology, operational definitions, study populations, follow-up duration, hospitalization status, and symptom ascertainment. This systematic review and meta-analysis synthesized global observational evidence on Long COVID/PCC prevalence and examined major sources of variability. Methods A systematic review and meta-analysis was conducted according to PRISMA 2020 principles. PubMed/MEDLINE, Scopus, Web of Science, and the WHO COVID-19 Global Literature Database were searched for studies published from 1 January 2020 to 23 February 2026. Observational studies were eligible if they included individuals with confirmed or probable SARS-CoV-2 infection, reported Long COVID/PCC or persistent post-COVID symptoms at least 4 weeks after acute infection, and provided numerator and denominator data. Terminology and operational case definitions were extracted separately. Prevalence estimates were pooled using a random-effects model with logit transformation and back-transformation. Heterogeneity was assessed using Cochran's Q, I 2 , tau 2 , and prediction intervals. Subgroup and sensitivity analyses were performed. Results Twenty-two studies contributing 27 prevalence estimates and more than 200,000 participants were included. The primary estimate-level pooled prevalence was 30.8% (95% CI 26.8–35.0). Heterogeneity was extreme: Q = 8031.9, df = 26, p < 0.001; I 2 = 99.7%; tau 2 = 0.252 on the logit scale. The prediction interval was 14.0–54.8%. Pooled prevalence was higher among hospitalized cohorts (37.9%, 95% CI 29.5–47.1) than among non-hospitalized, community-based, or mixed cohorts (26.2%, 95% CI 22.0–30.9). WHO-defined PCC yielded lower prevalence (22.8%, 95% CI 14.3–34.4) than broader symptom-based definitions (39.7%, 95% CI 30.5–49.5). Cohort-level sensitivity analysis yielded a similar prevalence of 31.0% (95% CI 26.8–35.4). Conclusion Long COVID/PCC represents a substantial post-acute public health burden. However, because heterogeneity was extreme, the pooled prevalence should be interpreted as a descriptive summary of heterogeneous observational evidence rather than as a single precise global rate. Standardized definitions, harmonized surveillance, transparent reporting, rehabilitation pathways, and multidisciplinary care models are needed.