Study population and setting
This evidence synthesis assessed questions regarding to what extent children (<18 years) are the source of infections within households (index cases), and whether children are more or less likely than adults to get infected when there is a case in the household (comparison of secondary attack rates). This was a meta-analysis including 57 articles published from December 1, 2019 to August 24, 2020, which together included data from 213 household transmission clusters (2+ members in a household serially infected) across 12 countries.
Summary of Main Findings
Overall, children were rarely (8/213 clusters, 3.8%) the source of infection within households; this number increased to 18.5% in the subset of clusters in which the pediatric case was asymptomatic and assumed to be the index case. Children were also less likely to be infected by the index case as compared to adults within their households, with an estimated 38% lower secondary attack rate among children than adults (relative risk [RR] 0.62, 95% CI: 0.42-0.91) within households with a known case. Overall, the meta-analysis found that among children and adults, asymptomatic cases were 83% less likely (RR 0.17, 95% CI: 0.09-0.29) to infect other household contacts as compared to cases experiencing COVID-19 symptoms.
Household level data were leveraged across multiple studies and settings.
Lockdowns, school, and daycare closures were common across many settings during the time periods covered by these data and as such, children may have had limited opportunities for exposure to SARS-CoV-2; the proportion of household clusters in which the index case was a child may increase as children return to school and other activities. Newer data suggesting children may have longer incubation periods before symptom onset and separately may develop rapid immune responses and quickly clear infections may also result in misclassification of the index case within households, particularly if there was a common exposure and the child took longer to develop symptoms. Further, not every study tested all household members with an index case, so if children are more likely to be asymptomatic there may have been more missed index and secondary attack cases among children.
Through synthesizing and meta-analysis of the available evidence, the authors demonstrate compelling trends in the role to date of children in bringing infection from outside into the home, as well as information pertaining to consistency in secondary attack rates experienced among children as compared to adults across settings. These data can inform school reopening and vaccine prioritization.
This review was posted on: 4 February 2021