NOTE: This post has been updated to address potential concerns of critics who may frown upon using estimates based on low event counts.
In this post, I accidentally forgot to extend the count of M-pox deaths by several days past the count of M-pox cases. The reason to do that with cumulative numbers is because cases show up “before” the deaths do. After reworking it, M-pox is no more deadly than first estimated. Here is the computer readout of the IFR for M-pox:
The histogram at right shows the most likely IFR was about 0.03%, though the 99% Credible Interval at bottom-left shows that IFR values up to 0.089% are consistent with the delayed-death data for the 19,915 known infections. A bad flu season is worse than that upper bound value. Here are the cases:
And, looking one week forward in time, here are the deaths:
That works out to a point estimate of 1 M-pox death for every 2,845 cases of M-pox, but due to a low number of deaths, it is more important to hone in the focus on the upper bound of a 99% Credible Interval (which incorporates the uncertainy involved). At that upper bound value of 0.089%, there is one death for every 1,124 cases.
With severe flu seasons though, you can get one death for every 700 cases of flu. Evidence using M-pox deaths out to 25 Jul 2022 indicates that M-pox is not more dangerous than a severe flu — in spite of the morbid rash that often comes with it.
Reference
Edouard Mathieu, Fiona Spooner, Saloni Dattani, Hannah Ritchie and Max Roser (2022) - “Mpox” Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/mpox' [Online Resource]