A living systematic review and meta-analysis of symptoms
Contributors
Rachel Geyer, Grace Kim, Matthew Thompson (Department of Family Medicine), Monisha Sharma (Department of Global Health) and Sophie Morse, Kelli O’Laughlin (Department of Emergency Medicine).
Objective
We aimed to evaluate which symptoms were most predictive of positive COVID-19 cases and assessed variations by sub-groups (i.e. age, disease severity). This analysis may be useful for clinicians and researchers using predictive symptoms to prioritize testing and resource allocation.
We will update results and share findings on this webpage as new studies are published to keep these data up-to-date.
Methods
We conducted a systematic literature review and meta-analysis following Cochrane and PRISMA guidelines. We searched EMBASE and PUBMED for studies reporting symptoms among persons diagnosed with COVID-19 published in English between January 1 and April 28th, 2020.
Search terms included variations of “clinical characteristics,” “symptomatology,” and MeSH terms for COVID-19. All articles were screened based on predefined exclusion criteria and assessing quality through Boyle’s framework for prevalence studies.
We included 37 studies, which included ten studies on disease severity, four specific to children populations, and four specific to elderly populations. Overall, a majority of the studies took place in China and hospital settings. A total of 6,909 individuals were sampled, and a half (50.3%) of the participants included were male.