Fig. 4

Bias adjusted cumulative incidence in India, 2021. The adjusted cumulative incidence is shown as a function of the proportion of exit screening in India that cannot be trusted. Discarding the biased risk of infection among travelers (i.e., q = 1), thick black lines show the estimate from dataset 2 among healthcare workers in the United Kingdom [30], while thick grey lines show estimates from dataset 3 derived from American prisoners [31]. Dashed lines represent the 95% confidence intervals that were obtained during the original estimation with q = 1. Thin black and grey continuous lines show estimates from datasets 2 and 3, respectively, when we assume that the risk of infection among travelers was 20% smaller than the general population in India (i.e., q = 0.8)