In Reply We thank Hu for his comments on the statistical issues of our study.1 The first comment is about the left truncation of the exposure window due to the unobserved exposure that occurred before the imported cases arrived at Taiwan and the right truncation due to case detection and isolation. The truncation might cause a bias if it affects the outcome of interest (secondary infection) directly. In our study, the truncation affected the exposure window, which was the exposure of interest, resulting in a different exposure window distribution compared with what would have been observed otherwise (ie, no importation of cases and no case detection and isolation). Under the assumption that the exposure windows were correctly measured for all close contacts included in the study and the outcome (secondary infection) was correctly identified, the truncation itself would not introduce biases when analyzing the association between different exposure windows and the secondary attack rate in the original study.
Cheng H, Ng T, Lin H. COVID-19 Transmission Conclusions Justified by the Analysis Results?—Reply. JAMA Intern Med. 2020;180(9):1262–1264. doi:10.1001/jamainternmed.2020.4100
Artificial Intelligence Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.