Epidemiological risk factors for developing long COVID: a rapid review with meta-analysis

Introduction Long COVID, characterized by persistent symptoms after the acute phase of SARS-CoV-2 infection, affects an estimated two million (3.1%) people in the UK. However, the epidemiological risk factors for developing this condition remain poorly understood. This study aimed to review the current literature and conduct a meta-analysis of the risk factors associated with long COVID.

Methods MEDLINE database was searched. Multivariate regression analysis studies exploring sex, age or comorbidities as risk factors were included. All studies required > 100 adult participants with laboratory-confirmed COVID-19 and follow-up at least four weeks following diagnosis. The risk of bias was assessed using the JBI critical appraisal tool. For risk factors that met eligibility criteria for meta-analysis, a provision of pooled estimates via a random-effects meta-analysis was conducted.

Results Of 2585 studies screened, ten studies were included in this review; presenting data on 5,387 patients (mean age 51.6 ± 17.9, 47.5% female). Meta-analysis of seven studies found female sex to be a risk factor for long COVID (OR 1.90, 95% CI 1.66 to 2.18). Meta-analysis of four studies found comorbidities to also be a risk factor (OR 1.46, 95% CI 1.16–1.83), and one study found COPD to be a specific risk factor. Four studies explored older age as a risk factor; just one observed significant results.

Conclusion Despite good quality data demonstrating female sex as a risk factor, evidence for the other risk factors, especially age and specific comorbidities, is heterogeneous. Considering the large burden of disease associated with long COVID, research must continue to further understand this condition.