From the first slide, "15 - 30 days from infection to death or discharge" so the mid-point is closer to 3 than 4 weeks assuming that death / discharge have the same even distribution throughout the range.
Would a 3 week lapse suggest that lockdown had a more significant impact?
