Using secondary cases to characterize the severity of an emerging or re-emerging infection

Tim Tsang (1, 2) , Can Wang (1) , Bingyi Yang (1) , Simon Cauchemez (3) , Benjamin Cowling (1, 2)

1 Li Kashing Faculty of Medicine
2 HKSTP - Hong Kong Science and Technology Parks Corporation
3 Modélisation mathématique des maladies infectieuses - Mathematical modelling of Infectious Diseases

The methods to ascertain cases of an emerging infectious disease are typically biased toward cases with more severe disease, which can bias the average infection-severity profile. Here, we conducted a systematic review to extract information on disease severity among index cases and secondary cases identified by contact tracing of index cases for COVID-19. We identified 38 studies to extract information on measures of clinical severity. The proportion of index cases with fever was 43% higher than for secondary cases. The proportion of symptomatic, hospitalized, and fatal illnesses among index cases were 12%, 126%, and 179% higher than for secondary cases, respectively. We developed a statistical model to utilize the severity difference, and estimate 55% of index cases were missed in Wuhan, China. Information on disease severity in secondary cases should be less susceptible to ascertainment bias and could inform estimates of disease severity and the proportion of missed index cases.

Domaines: Mathématiques [math]

Tim Tsang, Can Wang, Bingyi Yang, Simon Cauchemez, Benjamin Cowling. Using secondary cases to characterize the severity of an emerging or re-emerging infection. Nature Communications, 2021, 12 (1), pp.6372. ⟨10.1038/s41467-021-26709-7⟩. ⟨pasteur-03512692⟩