Predictions are fun, they can even be useful. Misrepresenting what has already happened is less fun, and can be dangerous, and that article appears to. Italy already has more deaths due to this than any other country in the world, including China. It is continuing to accelerate, not slow down in the past few days, regardless of when this was written for that paper. Use the
same Johns Hopkins data that was quoted there - there are already an additional 1000 deaths since that was printed. (EDIT2: Italy just announced 627 deaths on Friday, up from 427 the prior day, now up over 4,000) A slightly more usable website is
right up here, but you can always compare the numbers with Johns Hopkins to verify over time. Johns Hopkins appears to split out only the US by state by default, while defaulting the rest as countries, I imagine because they were asked to. (EDIT: if you click on US on the left side of that page, it does show totals for US and then for each individual state). Yesterday, US just crossed 100 deaths. Today, we crossed over 200. If we check back in 10 days, I think it's most likely we will consider those numbers very small.
I agree with some of your points, calculating a mortality rate can vary based on the unknown denominator of cases if the symptoms are such that many are unreported - but extrapolating it from the curve of the death rate alone has been more than enough to frighten governments all over the world. Our capability to change the death rate is also partially separate from the academic lethality of the disease itself - once it hits the threshold where medical care needs to be throttled due to capacity, it will clearly be higher than it would have been otherwise. Lack of preparedness in the governmental and healthcare space coupled with populations not actively and successfully slowing the transmission rate will kill people that wouldn't have died otherwise.