Abstract
BACKGROUND: Characterizing the paradigm and impact of long COVID is crucial for addressing this worldwide health challenge. This study aimed to investigate the prevalence of long COVID one year after primary Omicron infection and characterize differences in long-term health consequence between participants with persistent long COVID and those who fully recovered. METHODS: This a community-based cross-sectional study conducted from December 2023 to March 2024 at the China-Japan Friendship Hospital and 16 administrative districts in Beijing. 12,789 participants infected with Omicron between December 2022 and January 2023 were recruited through stratified multistage random sampling and included in the final analysis. Of them, 376 participants with persistent long COVID and 229 without long COVID were matched for further physical examinations. The primary outcome was the prevalence of long COVID one year after infection. Secondary outcomes included muscle strength, exercise capacity, health-related quality of life (HRQoL), mental health, work status, laboratory tests, and examinations. FINDINGS: Among 12,789 participants (media [IQR] age, 48.4 [37.3 to 61.4] years; 7817 females [61.1%]), 995 of them (7.8%) experienced long COVID within one year, with 651 (5.1%) having persistent symptoms. Fatigue (598/995 [60.1%]) and post-exertional malaise (367/995 [36.9%]) were the most common symptoms. Brain fog had the lowest resolution proportion as 4.2% within one year. The odds of long COVID increased with reinfections (odds ratios for one reinfection 2.592 [95% CI: 2.188 to 3.061]; two or more: 6.171 [3.227 to 11.557]; all p < 0.001). Participants with persistent long COVID had markedly lower muscle strength (upper-limb: 26.9 ± 12.4 vs. 29.1 ± 14.5 Kg; lower-limb: 40.0 [27.0 to 62.0] vs. 43.0 [28.0 to 59.0] s), worse exercise capacity and poorer HRQoL, and meaningful difference in laboratory tests results compared to those without long COVID. They also exhibited significantly higher proportions of abnormal lung function (FEV(1) %pred<80%: 13.0% vs. 2.0%; DLco %pred<80%: 32.7% vs. 19.9%) and lung imaging abnormalities (23.5% vs. 13.6%). INTERPRETATION: The considerable health burden of long COVID and the progression of neurological symptoms following Omicron infection warrant close monitoring. Utilizing professional questionnaires and developing reliable diagnostic tools are necessary for improving diagnosis and treatment of long COVID. FUNDING: This work was supported by Beijing Research Center for Respiratory Infectious Diseases (BJRID2024-012), Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (2022-I2M-CoV19-005/CIFMS 2021-I2M-1-048), the National Natural Science Foundation of China (82241056/82200114/82200009), the New Cornerstone Science Foundation.