I think there is now enough data to form a very preliminary assessment of the situation.
Clearly Wuhan is a massive outlier when it comes to fatalities, and personally, I think the current 2-3% mortality rate is massively inflated if you remove Wuhan.
The reasoning for me making such a significant dataset from the whole is because of the very special circumstances of the outbreak in Wuhan.
During the initial phases of the outbreak, there is a critically important short period where the correct medical information and procedures were not yet widely circulated.
That resulted in huge overcrowding in Wuhan hospitals, with patients taking none of the recommended protective measures. So you ended op having large numbers of people all crammed into enclosed spaces for extended periods of time, usually together with several infected.
That not only helped the virus to spread like wildfire, it also meant many of the earlier sufferers where exposed to heavy doses of the virus for prolonged periods of hours or even days.
As any half decent medical professional will tell you, the concentration and duration of the viral exposure has a direct impact on the severity of the infection, and thus also the mortality rate.
As such, I think the initial phase of the infection had a much higher mortality rate than is typical of this virus because of the heavy viral load those sufferers where initially exposed to.
Subsequent sufferers who were exposed to less concerted doses of the virus for shorter periods had much better survival rates as a direct consequence. But in a wider outbreak, now that we know what this thing is; odds are people are going to be exposed to low concentration doses of the virus; and as such, the mortality rates outside of Wuhan’s initial wave of sufferers may be a much better predictive model of the likely wider mortality rates.
