Methods
Random-effect meta-analysis was performed to produce a pooled prevalence for each symptom at four different follow-up time intervals. Between-study heterogeneity was evaluated using the I2 statistic and was explored via meta-regression, considering several a priori study-level variables. Risk of bias was assessed using the Joanna Briggs Institute tool and the Newcastle-Ottawa Scale for prevalence studies and comparative studies, respectively.
Results
After screening 3209 studies, a total of 63 studies were eligible, with a total COVID-19 population of 257 348. The most commonly reported symptoms were fatigue, dyspnea, sleep disorder, and difficulty concentrating (32%, 25%, 24%, and 22%, respectively, at 3- to <6-month follow-up); effort intolerance, fatigue, sleep disorder, and dyspnea (45%, 36%, 29%, and 25%, respectively, at 6- to <9-month follow-up); fatigue (37%) and dyspnea (21%) at 9 to <12 months; and fatigue, dyspnea, sleep disorder, and myalgia (41%, 31%, 30%, and 22%, respectively, at >12-month follow-up). There was substantial between-study heterogeneity for all reported symptom prevalences. Meta-regressions identified statistically significant effect modifiers: world region, male sex, diabetes mellitus, disease severity, and overall study quality score. Five of six studies including a comparator group consisting of COVID-19–negative cases observed significant adjusted associations between COVID-19 and several long-term symptoms.
Conclusions
This systematic review found that a large proportion of patients experience post-acute COVID-19 syndrome 3 to 12 months after recovery from the acute phase of COVID-19. However, available studies of post-acute COVID-19 syndrome are highly heterogeneous. Future studies need to have appropriate comparator groups, standardized symptom definitions and measurements, and longer follow-up.