We estimate that on 22 January, in Wuhan there were currently 14464 infected individuals (prediction interval, 6510–25095), and a total 21022 infections (prediction interval, 11090–33490) since the start of the year. We also estimate t 24 currently infected individuals (prediction interval, 19–30) in other locations of China on this date. For comparative purposes, we estimate the total number of infections in Wuhan from 1January to 18 January inclusive to have been 6733 (prediction interval, 3500–10914). This estimate of the total infections is comparable to other published estimates based on travel data and reported cases identified outside of China (estimated between 1,700 and 7,800)(Imai et al., 2020a), and highlights our estimated low ascertainment rate, the rapid growth of the epidemic, and uncertainty in model predictions.
Should the transmission continue at the same rate in Wuhan, with no control or change in the behaviour of individuals (such as spontaneous social distancing) our model predicts that on 29 January the epidemic in Wuhan will be substantially larger, with 594 cases expected to be detected on that day in Wuhan (prediction interval, 446–788) and 105077 currently infected (prediction interval, 46635–185412); see Figure 3 and Table 1. If transmission has reduced, either through control or spontaneous public response to the epidemic, this will be a gross overestimate, though it may be useful to help gauge the effectiveness of interventions.