My revised best guess credible scenario with caveats and uncertainties - this is nowhere near the 2-3MM deaths in the US scenarios some have produced, but it's not "just the flu" either. Take this with a large grain of salt, as I'm certainly not an epidemiologist.
Wuhan had a 0.5% infection rate according to the Chinese with maybe 50K cases, but I don't think anyone trusts the Chinese numbers, which I think are 10X greater as there's no way only 0.5% were infected. On the other hand, I've also never felt that the 50-70% infection rates for the population were realistic (those feed the 2-3MM deaths per year numbers).
However, I do think we need to at least consider the possibility that what was observed on the Diamond Princess could be a credible worst case, especially for densely populated areas that do not do a good job of testing, tracking, quarantining (of positives and their contacts), which can let the infection take off exponentially, as we've seen almost everywhere, at least for awhile, overwhelming health care systems, as we saw in Northern Italy and which we're at risk of in NYC.
The DP was a perfect "floating virus transmission laboratory" with 3711 people, who were exposed to the virus for 14 days (before the quarantine) in a location as densely populated as NYC with conditions ripe for transmission, given communal activities and meals and close quarters. About 712 people tested positive (19%) and of those about half had symptoms and half did not and we have no idea how many of the rest were infected but tested negative, due to low virus levels not detectable by the test - we'd need to test all of them via the serological antibody assay to know for sure (can't believe that hasn't been done yet). And of those 705, 9 died for a case fatality rate of 1.3%, while 37 (5.2%) required intensive care.
To me the 19% infection rate is the absolute worst case scenario, practically speaking, for larger populations (there could always be more for isolated case, like we might be seeing in nursing homes and individual families), assuming we did no interventions. So, for example if 20% of the US were infected and 1.3% died, we'd have about 830K deaths (vs 35K flu deaths/year in the US). Of course, 20% of the US is pretty unlikely due to much lower overall density but it's possible infection rates could reach 5-10% in some cities, like NYC and especially Manhattan with the highest population density in the US and heavy reliance on mass transit, plus a very high rate of people going in and out of the area.
Even at just 5% infected (and we're at 0.3% in NYC now, but projected to reach at least 300K infections or more which would be 3.3% vs 9MM in NYC and that's with some social distancing) and if we had a 1.3% mortality rate, that's 5900 deaths and at 5.2% in the ICU that's 23,600 ICU cases. And if we extrapolated those numbers over the ~80% of the US that's considered urban, then 5% would be 13MM infections and at a 1.3% mortality rate that would be 170K deaths and 4X that many ICU cases, assuming no interventions.
Having said that, all of the people who died on the DP were over 65 and the cruise was older than the US on average, so some have done calculations suggesting the "true" DP death rate would be more like 0.7% which is about half of the 1.3% from the DP (see the 2nd link, which was published when only 6 had died, but that eventually became 9, so their 0.5% adjusted mortality becomes 0.7%). Even at half the numbers I just calculated above (using a 0.7% mortality rate), that would be 85K deaths and 340K ICU cases in the 80% of the US which is urban if we implemented few interventions, both of which are far in excess of an annual flu year.
We'd need a case rate of about 1-2% to bring the CV2 deaths down near the annual flu and that simply doesn't seem possible to achieve in densely populated areas (and hotspots outside of urban areas) without significant interventions (testing, tracing/quarantining, social distancing, no crowds, closures, etc.), like we're seeing now. Also, keep in mind that the numbers in this scenario are totals for a year or so, comparing to the flu - the other issue with this virus is the very sharp peaks we've seen almost everywhere, which can potentially overwhelm anyone's health care system, which is another reason for "flattening the curve" via interventions.
https://www.cdc.gov/mmwr/volumes/69/wr/mm6912e3.htm?s_cid=mm6912e3_w
https://www.sciencenews.org/article/coronavirus-outbreak-diamond-princess-cruise-ship-death-rate
And that's why most experts feel we need to intervene with testing, tracking, quarantining and social distancing especially in our cities, as we don't want to "hope" for spring reducing transmissions (which I think it will, but we don't know or by how much). How long we have to do this for is obviously the big question. My guess is 6 more weeks, since we got such a horribly late start on our interventions, especially on testing - we could have easily followed South Korea's model and peaked at 45K infections and be on the decline right now, but we didn't follow their playbook. The other wild card is if any of the clinical trials with repurposed older drugs help us, which is unlikely, from everything I've read (HCQ did not work in China much).
And the last, but best wild card was just announced today, as NY got FDA approval in record time (one place Trump/Hahn deserve full credit for temporarily suspending many regs on new treatments) to start testing antibody-laden blood plasma collected from recovered infected patients with antibodies to the virus in very sick patients and eventually as possible preventative for elderly/high risk people and health care workers. It's low tech and cheap, but it should work to at least some extent and we should know in a few weeks (see the link).
https://rutgers.forums.rivals.com/t...social-distancing.191275/page-34#post-4470810
