Long COVID Prevalence and Risk Factors: A Systematic Review and Meta-Analysis of Prospective Cohort Studies.

Background: Long COVID, or post-acute sequelae of SARS-CoV-2 infection (PASC), affects millions globally, with persistent symptoms impacting quality of life. This meta-analysis synthesizes prospective cohort studies to estimate the prevalence of Long COVID symptoms and identify risk factors. Methods: We systematically searched PubMed for prospective cohort studies (2020-2025) on Long COVID, focusing on prevalence and risk factors. Studies with ≥100 participants and follow-up ≥3 months were included. Data were extracted on symptom prevalence (e.g., fatigue, dyspnoea) and risk factors (e.g., sex, hospitalization). Random-effects models were used to pool prevalence and odds ratios (OR). Risk of bias was assessed using the Newcastle-Ottawa Scale (NOS). Results: Fourteen prospective studies (n = 168,679) were included. Pooled prevalence of Long COVID was 18.0% (95% CI: 12.5-23.5%, I2 = 9.8%) among survivors followed for ≥6 months. Fatigue (41.0%, 95% CI: 33.2-49.4%) and dyspnoea (22.5%, 95% CI: 15.6-29.8%) were the most common symptoms. Female sex (OR = 1.52, 95% CI: 1.25-1.92) and prior hospitalization (OR = 2.35, 95% CI: 1.98-2.90) were significant risk factors. High heterogeneity (I2 > 90%) was noted. Conclusions: Long COVID affects over one-fifth of SARS-CoV-2 survivors, with fatigue and dyspnoea persisting in many. Female sex and severe acute infection increase risk. Standardized definitions and longer follow-up are needed.