I caution against reading too much into the South Korean fatality rates because many of the patients are younger and generally healthy (<a href="http://www.koreaherald.com/view.php?ud=20200303000714" rel="nofollow">http://www.koreaherald.com/view.php?ud=20200303000714</a>).<p>Whatever the global mortality rate shakes out to be in the long-run, it’s impossible to ignore the fact that this virus killed nearly a dozen nursing home residents in the span of 2 weeks. There’s a lot we don’t understand about COVID-19, certainly more than a single metric can tell us.<p>Based on what we’re seeing in the state of Washington and around the world, it’s safe to say that older patients are at higher risk for developing respiratory failure. Beyond that, it’s simply way too early to draw conclusions.
Isn't it that there are two main strains, one (30% of cases) is less lethal (and this one is predominant in South Korea) and the other (70% cases) is e.g. in Italy?<p>Two strains source: <a href="https://www.newscientist.com/article/2236544-coronavirus-are-there-two-strains-and-is-one-more-deadly/" rel="nofollow">https://www.newscientist.com/article/2236544-coronavirus-are...</a>
Does anyone know of any good, publicly accessible case data (for any country)?<p>E.g. for each case,<p><pre><code> date of detection, age of patient, ongoing|recovery date|date of death</code></pre>
This corroborates Chinese numbers. China has also been very aggressively testing and mortality outside Wuhan is no more than 1% (according to Dr Aylward, following the WHO fact finding mission to China).<p>This implies very serious efforts, though, which I'm not sure we're seeing in the West.
Uh, just looked at map and Pyongyang/North Korea is holding great on stats - zero infections. Does anyone know if it's becase a) almost nobody travels there, or b) there's no outgoing information or both?
My theory on the coronavirus. I've managed to convince myself to have moderate levels of confidence in it; please tear it apart.<p>1) The Diamond Princess is our best source of data. Every person who got infected on it was identified (~700), and those who were infected got the best medical care possible. This led to a ~1% mortality rate (6/700) and a ~5% ventilator support rate (35/700). 6/700 is subject to significant error in both directions, but let's go with it.<p>2) Their average age was in the upper 40s, so assuming younger people don't suffer the most serious ill effects from it, a guess for the general population might be a 0.5% base mortality rate and 2.5% ventilator support. This is a rough order of magnitude estimate: obviously expected mortality should be adjusted for actual age distributions.<p>3) South Korea's mortality rate is 0.6%. Cases outside of Wuhan but within China have a reported mortality rate of 0.4%. Singapore (at ~100 cases) is 0%. These are all consistent with the estimated mortality rate I extrapolated from the DP.<p>4) All of the areas mentioned in 3) have very aggressive testing protocols. No evidence has come up suggesting widespread infection among the broader community. This includes China randomly sampling some areas and not finding meaningful unknown pockets of infection.<p>5) But what about Wuhan? It has a much higher mortality rate (>3%). The reports on the ground suggested that it was the medical system being overwhelmed that drove the mortality through the roof. If we assume everyone who needs a ventilator but doesn't get one dies, that explains in one fell swoop Wuhan's elevated mortality rate.<p>6) What about Italy? Its mortality rate is ~3.5%. But, compare the number of hospital beds in Italy and SK. It's ~3/1k vs ~10/1k. SK is already running into issues with hospital bed availability, so it stands to reason that the situation is much worse in Italy. Italy also has a significantly older population than SK. So it's starting to run into the same nightmare scenario as Wuhan. Of note, the US has a younger population than Italy, but fewer hospital beds.<p>7) What about Iran? It has ridiculous mortality rates. My best explanation is that testing constraints probably lead to more undercounting of infections in the hardest hit areas, so Hubei, Italy, and Iran probably all have artifically elevated mortalities, while the reality for all of them is much closer to 2.5%.<p>8) There's no need for weird hypotheses about Asians having different proteins in their lungs, there being multiple strains that act radically differently from each other, China running secret death camps for the infected, etc.<p>9) Policy-wise, we need to keep the rate of infection low enough such that the medical system is not overloaded. This is possible, as it's been done in multiple countries, and that's the difference between a severe-but-normal flu season and over a million dead in the USA. We also have to start taking decisive action yesterday, with today being a second best option.
South Korea already has huge experience with SARS <i>and</i> MERS. Most other countries don't, so "true fatality rate" would be higher. The demographics(age distribution) is also different in other countries.