Abstract
BACKGROUND: During the COVID-19 pandemic, an increasing number of patients have reported persistent symptoms after recovery, a phenomenon known as long COVID. These symptoms may persist for weeks or months, affecting the patient's daily life and health. To systematically understand the long-term impact of long COVID, this study conducted a systematic review and meta-analysis. This study aims to determine the long-term effects of long COVID by identifying, evaluating and summarising the incidence and duration of persistent symptoms after the acute phase of COVID-19. METHOD: We searched PubMed, Embase, and medRxiv up to August 2021 for articles and preprints presenting original research on the symptoms of long COVID. Following title/abstract and full-text screening, based on the PICOS framework, we excluded articles that did not clearly report on diagnoses, reported on symptoms lasting less than four weeks, lacked epidemiological data, or did not provide complete data. We assessed the bias of included studies using the Newcastle-Ottawa Scale. For effects reported in more than two studies, we performed meta-analysis of prevalence, and also counted the duration of each symptom. RESULTS: We included 19 observational studies in the meta-analysis, through which we determined the incidence and duration of five common long COVID symptoms, including cognitive/memory/attention disorders (36%, unreported duration), fatigue (34%, 5.5 months), mental health problems (including anxiety and depression, 31%, 3.5-3.8 months), and dyspnoea (24%, 6.52 months) and chest pain (23%, 2 months). CONCLUSIONS: The symptoms of long COVID usually persist after the acute phase of COVID-19. The clustering of long COVID symptoms provides a direction for studying the aetiology, diagnosis, and management of post-COVID conditions. Our findings provide important baseline data for the prevention and treatment of long COVID.