The official coronavirus epidemic numbers don't add up
February 14, 2020 8:53 AM   Subscribe

Body Count Viet Nam War casualty reporting as metaphor for the official numbers coming out of China concerning the coronavirus epidemic.

The official numbers being provided on the coronavirus epidemic appear to correlate to a quadratic growth model; since all epidemics take the form of an exponential function, we can assume the official numbers are completely made up.

Links to the original Reddit thread, and an article from The Lancet concerning the nCoV2019 propagation model, are in the linked blog post.
posted by Bron (58 comments total) 11 users marked this as a favorite
 
Isn’t it equally possible that the numbers reflect the limits of health officials’ testing ability, rather than being purely made up? My understanding is that the statistics reflect “confirmed” cases, and don’t even purport to measure the true extent of infections. That would still make the numbers meaningless for modeling purposes, but it doesn’t necessarily reflect dishonesty.
posted by Emera Gratia at 9:17 AM on February 14 [43 favorites]


+1 to Emera, I would assume that test capacity has a good chance of following a quadratic model.
posted by allegedly at 9:21 AM on February 14 [5 favorites]


Just today China reported a big increase in reported cases. This increase is primarily due to reclassification of previous cases using chest images, which supports Emera's theory:

https://www.bbc.com/news/world-asia-china-51482994
posted by justkevin at 9:29 AM on February 14 [4 favorites]


I've been reading some stuff from Marc Lipsitch, who has a fairly pessimistic take about current tracking of the virus and the possibility of containing it in the long term, but doesn't have a terribly high opinion of the finance guys doing epidemiology either.
posted by atoxyl at 9:30 AM on February 14 [1 favorite]


I like Emera's suggestion, but this would require an odd relationship between capacity and time.

It might occur, for example, if the capacity (in terms of people tested/day) were increasing linearly, and the population being tested had the same % of infected people every day, but the odds that the % of infected people is constant over time is nil.

The fact that this curve fits a quadratic so well strikes me as pretty implausible. Exponential curves don't behave like that.
posted by thatnerd at 9:30 AM on February 14 [2 favorites]


This write-up contains no serious statistical analysis. Once cannot compare models without some metric of significance. The only metric of significance deployed here is, literally, "I mean... c'mon man."

This reads as conspiratorial fear-mongering by someone who knows a tiny bit of math.
posted by mr_roboto at 9:32 AM on February 14 [32 favorites]


I would buy that if the published numbers were accompanied by a clear and explicit admission that they were limited by testing capacity.

I think they were explicitly leaning on their limited testing throughput as a plausible reason to publish blatant false numbers. But those numbers were becoming so blatantly false, that I believe they fell back to a different cooked-up number of “clinically diagnosed cases” to try to make it a little more believable.

But as long as they continue to claim quadratic growth of something that is fundamentally exponential, they will keep needing to periodically rebase onto some new line of bull, as it repeatedly becomes untenable to defend whatever the current lie is.

I mean, look at the Diamond Princess. Patient zero was on the ship for 6 days. It’s about 3 weeks after he left, and almost 6% of the ship has tested positive, despite the entire ship being being under room isolation quarantine for about half of that time. That’s what 10 days uncontrolled spread from a single case can do.

Imagine how many uncontrolled cases are in China right now.
posted by notoriety public at 9:32 AM on February 14 [7 favorites]


No one knows anything about China, including China. (Sl Foreign Policy) We don’t know China because, in ways that have generally not been acknowledged, virtually every piece of information issued from or about the country is unreliable, partial, or distorted. - This is from 2018, but I find it always relevant when we're discussing China. I don't think things have changed.
posted by toastyk at 9:33 AM on February 14 [19 favorites]


Kremlinology with Chinese characteristics
posted by Freelance Demiurge at 9:38 AM on February 14 [12 favorites]


I suspect a more-useful metric of what is going on in China at the moment can be gleaned from seeing What They Do instead of What They Say.

About 6.5% of the entire world's population is affected by China's current lockdown and travel limitations. Source: That's basically 500 million people.

500,000,000 people.

They are claiming (as of today) to have 64,466 total confirmed cases of COVID-19 in the entire world.

China has restricted the everyday lives, schools, jobs, and recreation of 500,000,000 people on the grounds of 64,466 sick people?

Yeah, I'm not buying that.
posted by which_chick at 9:54 AM on February 14 [10 favorites]


Yeah, I'm not buying that.

Why not? What are best practices in this case and how is China not following them? What course of action would you recommend were the reported numbers accurate? Can you put this in the context of the regularity of intercity travel in China and the state of Chinese public transit infrastructure?

Just wondering.
posted by mr_roboto at 10:06 AM on February 14 [15 favorites]


They are claiming (as of today) to have 64,466 total confirmed cases of COVID-19 in the entire world.

China has restricted the everyday lives, schools, jobs, and recreation of 500,000,000 people on the grounds of 64,466 sick people?
It is important to note that no one, not in China nor anywhere else, believes that the 64,466 confirmed cases is the number of infected persons. It is the number of people who have confirmed cases of the virus primarily confirmed by testing and in a limited number of cases by reworking previous lab data. These are primarily the sickest of the sick, people who presented at hospital and braved ridiculous crowds and dangers to do so. The number of people actually infected are presumed by everyone concerned to be much higher, but those numbers cannot be reported because we won't even have plausible estimates for months.
posted by Lame_username at 10:08 AM on February 14 [30 favorites]


I've been following this relatively closely and it seems like people aren't sure whether the quarantine is effective or not - nothing like it has been attempted before. People are starting to go back to work in many parts of China after the extension of the holidays, so we're going to see what happens on a pretty large scale in the next couple of weeks.

My one thought is that I would have expected a lot more people to be a lot sicker outside of China. I don't expect "accurate" numbers from China because they have, whoa, a huge and overwhelming pandemic going on, but I feel like if we don't see big self-sustaining clusters of infection outside of China in the next couple of weeks, I will start to wonder whether the assumptions about transmissibility and acuteness of disease are correct.

Like for instance, someone just showed up at a UK hospital in an Uber, having just come from China and gotten sick. She didn't follow the protocol that had been in the news, she put the Uber driver, everyone in all-purpose intake at the hospital and god knows how many other people at risk....if it's a super transmissible virus, there will be cases in the next couple of weeks from her actions, and the people those people infect, etc etc etc. They say that it can take up to 3 weeks for symptoms to show up, but it's usually much less than that - a few days or a week, so I would expect that somewhere outside of China there will be a lot more cases and a lot more deaths if it is in fact universally an acute, highly transmissible disease and there isn't some oddball factor about what's happening in Wuhan.
posted by Frowner at 10:15 AM on February 14 [18 favorites]


They say that it can take up to 3 weeks for symptoms to show up, but it's usually much less than that - a few days or a week, so I would expect that somewhere outside of China there will be a lot more cases and a lot more deaths if it is in fact universally an acute, highly transmissible disease and there isn't some oddball factor about what's happening in Wuhan.

The Diamond Princess is one such example. I think it’s likely we will see others soon.
posted by notoriety public at 10:24 AM on February 14


The Diamond Princess is also a cruise ship. Consider how norovirus can spread on those things - I'm not taking a cruise ship as typical of how the virus spreads anywhere else.
posted by Frowner at 10:27 AM on February 14 [17 favorites]


LOL. The official government numbers for everything are suspect, but Zerohedge has the correct ones.

Also the author is taking it as a given that epidemics grow exponentially, but none of the charts I've found show exponential growth rates for 'epidemics'. Ebola 2014 is one example, and the others (SARs, bird flu) seem to be pretty similar.

Even if they do grow exponentially, the official reporting numbers can be off because there can be time lags when reporting, there can be a lack of personnel qualified to deal with the epidemic, and there can be people who don't report for whatever reason, and the technology of quarantine, disease mapping, and technical advances in diagnosis, which are all legit reasons why the numbers are off. They may also be using models due to the work of consolidating a bunch of regional reports into a single official report takes time.
posted by The_Vegetables at 10:33 AM on February 14 [3 favorites]


The Diamond Princess is also a cruise ship. Consider how norovirus can spread on those things - I'm not taking a cruise ship as typical of how the virus spreads anywhere else.

Yeah, cruise ships are designed to be communal spaces with just enough personal space for the guests to sleep. The whole point is to be out and about doing stuff. It's like a college dorm with even more drinking and much much better food.

As an aside: My sister is a senior customer service rep for Princess, and uh, it hasn't been a good couple weeks at work. If even someone does something stupid with their personal 20ft sailboat and gets on the news her dept gets a million calls about it, all boats on Earth are related to all others, apparently. So, much worst when it's not only a cruise ship but a Princess cruise ship in quarantine.

But, the absolutely best part is that the owners of the local news radio station (1220AM KHTS) in Santa Clarita, where Princess is headquartered, are on the Diamond Princess! As you can imagine, there is currently lots of negative press
coming from their hometown station. It's like Jonathan Gold (RIP) getting locked inside your restaurant for two weeks against his will!
posted by sideshow at 10:40 AM on February 14 [10 favorites]


So... we should ignore an example of it behaving like a highly transmissible virus, because it spreads exactly like another highly transmissible virus does in the same environment?

I agree that it’s pretty much the most favorable environment for such a spread, but that’s also why it’s such an easily visible early example. It’s also happening in a country that is a little less likely to cover it up, and the nature of the event would make a coverup extremely difficult in any case.
posted by notoriety public at 10:41 AM on February 14


The number of people actually infected are presumed by everyone concerned to be much higher, but those numbers cannot be reported because we won't even have plausible estimates for months.

Yeah this is a good point - everybody knows the numbers are an underestimate, and everybody knows that testing capacity isn't there globally.

I feel like if we don't see big self-sustaining clusters of infection outside of China in the next couple of weeks, I will start to wonder whether the assumptions about transmissibility and acuteness of disease are correct.

One of Marc Lipsitch's big concerns (the pessimistic Harvard Public Health guy I linked above) is countries like Indonesia, which have not reported any cases - but have very little testing capability, and would be expected to have cases based on models.

Of course an underestimate of mild cases is sort of good news in terms of severity, but bad news in terms of containability.
posted by atoxyl at 10:51 AM on February 14 [2 favorites]


So. . . the model the author is proposing is that a large cabal of Chinese authorities randomly chose to generate fake data fitting a physically un-motivated quadratic model, 'cause they thought it was a neat function?

I'm happy to believe that governments lie. But, they either do it better or worse than this. Compared to a complicated mix of detection efficiency and mitigation that makes for non-intuitive data, this sure seems unlikely.

Then again, I come from a field where every function is either approximated as a quadratic or a power law, or sometimes both. Fitting a complicated, poorly defined thing to a quadratic seems like a perfectly fine thing to do, assuming you don't extrapolate too far. That you get a pretty good fit to a curvy thing with increasing slope that starts at zero. . . doesn't seem all that surprising.
posted by eotvos at 10:56 AM on February 14 [5 favorites]



So... we should ignore an example of it behaving like a highly transmissible virus, because it spreads exactly like another highly transmissible virus does in the same environment?


Well, I haven't had norovirus since I was a young teenager, and I have lived in dorms and shared housing of one kind or another for most of that, plus I work at a large university around students and use bathrooms shared by many, many people. What's more, no one in my immediate family has had norovirus since then either. My partner and my partner's family haven't had it in the time I've known them. This all even though a number of cruise ships have had norovirus outbreaks, there have been norovirus cases in my community, etc. This leads me to believe that while norovirus is extremely contagious, it simply isn't contagious enough to account for the worst-case scenarios people are describing for this pandemic. Mark Lipsitch, linked upthread, is estimating that 40-70% of the world will get this virus during the current pandemic. I am confident that 40-70% of people don't get norovirus each year.
posted by Frowner at 11:00 AM on February 14 [12 favorites]


So, I mean, it could be worse than norovirus, or different, etc. But "spreads fast on a cruise ship" doesn't seem to me to prove anything in particular about anything that is spread, as we know this is, by oral-fecal transmission.
posted by Frowner at 11:06 AM on February 14 [2 favorites]


I am confident that 40-70% of people don't get norovirus each year.
Current CDC estimates are that around 6% of the US population is infected with norovirus each year. Most of them have no idea they had it. Of those, there will be around 1.8 million who seek medical treatment, 400,000 ER visits, 63,000 hospitalizations and 585 deaths.

These kind of ratios are part of what epidemiologist call the iceberg principle and help explain why subtle shifts in the transmission rates or severity of any particular disease result in vastly different outcomes.
posted by Lame_username at 11:14 AM on February 14 [9 favorites]


Well, I haven't had norovirus since I was a young teenager..

Worth noting that every "wow, I had a bit of a stomach flu the other day!" was somewhat likely somebody describing Norovirus. Just like many bouts of the sniffles was someone dealing with Influenza. A highly contagious virus doesn't mean the CDC is going to show up with tanks and soldiers in bunny suits every time a single case is diagnosed.

So, all your statements of "this never happened!" are extremely likely to be false.
posted by sideshow at 11:28 AM on February 14


I don't mean to bore you with my gastro stuff, but I really, really haven't had "stomach flu" in many years. And in any case, if 6% of the population gets norovirus each year, it's still pretty darn unlikely that I've had it many times without noticing.

Again, they're talking about 40-70% of the world getting this coronavirus if the worst case scenario pans out. That is in no way remotely comparable to 6% of Americans getting norovirus, even if most people don't develop serious symptoms.
posted by Frowner at 11:36 AM on February 14


Again, they're talking about 40-70% of the world getting this coronavirus if the worst case scenario pans out. That is in no way remotely comparable to 6% of Americans getting norovirus, even if most people don't develop serious symptoms.

I am a bit lost here: are you saying that the 40-70% estimate is way too high/implausible, or how does coronavirus transmission relate to norovirus transmission (other than the cruise ship comparison)?
posted by witchen at 11:59 AM on February 14 [1 favorite]


This article drew parallels between US reporting of North Vietnamese casualties (to prove they were winning the war) and Chinese reporting of Wuhan virus casualty figures (to prove...). I found it well written and thought provoking, I found the near instant knee jerk accusations of "conspiratorial fear mongering" in this thread to be trollish. Of course there are limits on testing but are people here suggesting this applies to death figures? Are the Chinese under reporting deaths because of limitations of test kits?

The reddit thread contained a table with predictions for the total number of casualties and fatalities that would be reported for the forthcoming next week if those numbers were derived from a simple quadratic formula. These predictions were eerily close to the numbers actually reported over the next week.
posted by epo at 12:12 PM on February 14 [2 favorites]


how does coronavirus transmission relate to norovirus transmission (other than the cruise ship comparison)?

I think the cruise ship example suggests that COVID-19 is very, very transmissible--even more so than an ordinary corona/cold virus or even norovirus. Yes, cruise ships are petri dishes, but as I understand it in this case there was one infected person who was on board for four days. When the company realized a few days later that he had tested positive, the entire ship was put into passenger lockdown/quarantine. So after that point, there are no big gatherings, no buffet lines, none of the normal vectors that we're used to with, say, norovirus transmission. And even with all the passengers in lockdown, 217 people on board (and one Japanese quarantine officer) come down with the virus during the two-week quarantine period. (And the numbers are possibly higher--here, they are only testing people with symptoms and known close contacts. So we don't know if there are additional asymptomatic cases, though it seems likely). But luckily the vast majority of those seem (at least so far) to be very minor cases.

There's a ton we don't know, but I think the evidence is building that the virus is both more transmissible and less deadly than we thought. The biggest danger might not be the risk of death from the virus itself, but rather the overwhelming of our health care systems and daily life when a huge proportion of the population is all sick at once.

Also, if the stats are right that 40-70% of the population are likely to get infected, then we are in for a very bad year, but on the bright side a much higher chance that the virus will burn out quickly--if that level of the population becomes immune (or even partially immune, since the virus could mutate), then the herd immunity effects start to kick in.
posted by Emera Gratia at 12:24 PM on February 14 [6 favorites]


I am a bit lost here: are you saying that the 40-70% estimate is way too high/implausible, or how does coronavirus transmission relate to norovirus transmission (other than the cruise ship comparison)?

My comment was totally unclear!

What I was trying to say was, "just because both norovirus and novel cornonavirus thing spread rapidly on a cruise ship, this does not prove that they will spread similarly off a cruise ship, so the cruise ship does not give us the kind of information we'd have if there were a self-sustaining cluster of infections in, eg, Burbank" and "even if it looks comparable to norovirus, that wouldn't be the kind of worst case scenario that people are talking about".

Basically, I am neither worried nor reassured by the cruise ship situation, and I still feel like it's weird that there are so few reported cases in Zhejiang and Guandong and so many in Hubei. Zhejiang and Guandong had around 1000 cases each a week ago, and Zhejiang has ~1200 and Guandong has ~1300 currently. That isn't in line with the rapid transmission that people are fearing, and I don't feel like we're getting really bad rumors out of those places (at least none I've seen) so I'm assuming that there's not too much sheer lying going on.

So this makes me wonder - is the virus containable if you lock down when you have a few cases? Is there something about Hubei that made transmission there really intense that's not true in Zhejiang or Guandong? Or are we going to see a huge spike in the next week or so?
posted by Frowner at 12:26 PM on February 14 [5 favorites]


Of course there are limits on testing but are people here suggesting this applies to death figures? Are the Chinese under reporting deaths because of limitations of test kits?

They very much are. If someone dies and has not been confirmed for the coronavirus, their death is listed as being from pneumonia (which is true, but often incomplete information), and is not included in the confirmed COVID-19 death statistics.
posted by Emera Gratia at 12:28 PM on February 14 [3 favorites]


I've been following the outbreak relatively closely since it started, and even early on I thought there was a good chance that the number of infected was closer to 40x the official Chinese estimate (based on reading the same model that's linked in the blog post), and I still almost flagged this post for the article's lazy analysis and fear-mongering. I didn't flag it, because I think the discussion can be worthwhile.

"...this should be impossible. This is not how epidemics work."

There are many potential reasons that the reported statistics are not ground truth, and may not even trend like the ground truth numbers. They're not even obscure reasons--some have been mentioned in the thread, and the reality is that there is probably a combination of factors at play. Using language like "this should be impossible" is not the kind of thing that serious analysts usually say.

"...willing stooges at WHO..." "If you’d like to see how professionals who are not toadies of the CCP might model the spread..."

Again, this is not the language of someone trying to do a serious analysis. This sounds like reddit-driven econometrics bloggers trying to get clicks. The author might even be right, but this article doesn't have any evidence or persuasive content for his theory. I'm generally conspiracy-sympathetic, but this is a crappy article.
posted by jjwiseman at 1:46 PM on February 14 [21 favorites]


One thing I'm looking forward to is reading an analysis of why World Dream's quarantine seemed to be effective (it lasted less than a week before passengers were released, and the number of infected was low) while Diamond Princess is still in quarantine, on day 10, with at least 218 people infected. I can imagine that even something like "Movies on the in-room TVs are now free!" could be a factor.
posted by jjwiseman at 2:08 PM on February 14 [4 favorites]


It would be interesting to know if the later cases on the Diamond Princess are from the inside cabins where the only ventilation is from the air conditioning.
posted by Botanizer at 2:47 PM on February 14 [1 favorite]


Hint: casualty numbers coming out of totalitarian regimes are seldom reliable.
posted by Twang at 3:00 PM on February 14 [2 favorites]


Mark Lipsitch, linked upthread, is estimating that 40-70% of the world will get this virus during the current pandemic. I am confident that 40-70% of people don't get norovirus each year.

To be clear (as he made a point of clarifying and has been quoted out of context) he meant that to include asymptomatic infections. I think it's a simple extrapolation from the current case reproduction estimates. I'm not completely sure how those estimates are arrived at if one is also assuming a significant number of people will be infected but asymptomatic, though? And I feel like there must be more factors that differentiate disease transmission patterns that I don't fully understand. Like - I'm pretty sure the accepted R0 number for norovirus is at least as high as anybody has proposed for 2019- nCoV, so if it only infects 6 percent of people every year is that because outbreaks tend to burn out/not spread between populations as effectively?
posted by atoxyl at 3:35 PM on February 14 [1 favorite]


The 1918 flu pandemic is estimated to have infected 20-30 percent of the population, so it's not hard to see where one gets 40-70 percent if this becomes pandemic if one assumes it spreads like flu but a bit more so. But I wish I knew a little more about the exact methodology.
posted by atoxyl at 3:39 PM on February 14 [1 favorite]


if it only infects 6 percent of people every year is that because outbreaks tend to burn out/not spread between populations as effectively?

Or am I also confusing the number affected with the number infected there? I also saw an estimate that 20 percent of children under 5 get sick from norovirus every year.
posted by atoxyl at 3:41 PM on February 14


The R0 of Covid-19 was originally believed to be 2.2 to 2.7, but the last findings suggest that the "R0 value is likely to be between 4.7 and 6.6 ." The R0 of norovirus is between 1.6 and 3.7.
posted by Emera Gratia at 3:44 PM on February 14 [5 favorites]


The best source I've found for contextual reporting on Covid19 / coronavirus numbers is journalist William Yang.
Highly infectious given the rising number of group infections, whether via church, workplace, or home.
posted by spamandkimchi at 5:49 PM on February 14 [2 favorites]



The reddit thread contained a table


A table, you say?? It's never ceases to amaze me how liberal minded people that hold themselves above the idiots on Facebook that share fake news etc will credulously line up to swallow arrant bullshit from reddit, ffs, if it happens to align to their prejudices.

The post, and article, are horseshit. Why are we even discussing this? It's no better than Facebook posts estolling turmeric to fight coronavirus.

There are many epidemiologists, public health experts, pandemic experts,writing and opining on this. How in the ever loving fuck do we think some random dipshit on reddit would have access to secret knowledge and reasoning that the experts don't???

Goodness me, this exactly how nonsense about climate change, vaccinations, and racism spreads. Don't buy into it. You, and Joe Reddit, lack the expertise to evaluate any claims about coronavirus. So let the public health experts do it.
posted by smoke at 6:22 PM on February 14 [27 favorites]


Worldometer graph, and explanation of mortality rate and its relative difficulty to calculate (which basically needs an incubation rate unless the virus ends first).
posted by Brian B. at 7:38 PM on February 14 [1 favorite]


I found the near instant knee jerk accusations of "conspiratorial fear mongering" in this thread to be trollish.

Trollish my ass. I called it conspiratorial fear mongering because it's conspiratorial fear mongering. It's hot garbage and has all the intellectual value of one of Gwyneth Paltrow's vagina eggs.

That's my "knee jerk" scientific opinion. If I get some free time this weekend, I'll use my considerable scientific training to tear it the fuck apart.
posted by mr_roboto at 8:12 PM on February 14 [9 favorites]


BTW, I knew it was snot-dripping ignorant nonsense when I read this, in bold italics:

All epidemics take the form of an exponential function, not a quadratic function.

Uh-huh.

Jesus Christ.
posted by mr_roboto at 8:16 PM on February 14 [4 favorites]


It's a respiratory illness of a class we know about. Is the authoritarian government lying about the numbers? Almost certainly. But It's still honestly not that bad of a syndrome. I highly recommend anyone freaking out about this (and I understand this, disease is scary) to read a book I read recently- The Pandemic Century. It shows the real danger in novel illnesses is rarely the disease itself but the reaction to it- and as with it's fellow coronavirus SARS, covid-19 is most certainly not going to be a planet altering disaster, PRC official's mendacity notwithstanding.
posted by Homo neanderthalensis at 8:30 PM on February 14 [6 favorites]


All epidemics take the form of an exponential function, not a quadratic function.
Jesus Christ.


mr_roboto has it. There is an entire field of epidemiology literature that demonstrates that most epidemics are not exponential but "sub-exponential" or polynomial. Research by real experts in epidemiology that have spent their lives studying these phenomena.

This is a classic example of "engineers disease", where techbros spend 10 minutes of analysis in which they presume to overturn 100 years of research in a field they know nothing about. Like those Freakonomics idiots.
posted by JackFlash at 8:47 PM on February 14 [32 favorites]


I also saw an estimate that 20 percent of children under 5 get sick from norovirus every year.
That doesn’t seem likely to be accurate. The CDC says 1 in 6 children will seek medical care for Norovirus infection by age 5, which is probably the basis for the figure you saw reported.
posted by Lame_username at 7:34 AM on February 15 [2 favorites]


Whoops sorry I think I probably misread a version of that stat.
posted by atoxyl at 9:07 AM on February 15


If someone dies and has not been confirmed for the coronavirus, their death is listed as being from pneumonia

So has there been a massive uptick in reported pneumonia deaths in the quarantined area? Or deaths from any cause?
posted by saucysault at 9:23 AM on February 15


So has there been a massive uptick in reported pneumonia deaths in the quarantined area? Or deaths from any cause?

We don't know, because China isn't publishing those statistics. The best we have is anecdotal accounts. The Washington Post interviewed some whose relatives likely died of COVID-19 (but were uncounted in the stats) and individuals in China have reported seeing unusual numbers of dead bodies, and have posted videos on Twitter. Reporting suggests that "[t]he number of deaths appears to be severely understated, with many reports of victims dying at home and being cremated before being counted in the official total." So in the words of the Magic 8 Ball, "signs point to yes," but there is no definitive answer.
posted by Emera Gratia at 10:00 AM on February 15 [3 favorites]


Is there something about Hubei that made transmission there really intense that's not true in Zhejiang or Guandong?

Yes.

Wuhan and/or Hubei is a major transportation hub, and news of the virus was suppressed (by, among other things, arresting the doctors who diagnosed early cases) during the pre-New-Year travel period when some large fraction of China is on the move. Also, Wuhan city officials held a massive tourist event and passed out free tickets to travel in the early weeks of the outbreak, while they were still trying to pretend it wasn't happening.

The early coverup and timing came together into a perfect storm to let the virus rage out of control. Other provinces were locked down ASAP and so there hasn't been the chance for unchecked spread. I'm not surprised that the massive and draconian quarantine is having a noticeable effect in those places.
posted by Ahniya at 11:24 AM on February 15


And as for the thread topic...does anyone really, seriously expect that China is giving accurate numbers? Even if they mean to publicize accurate facts about a massive screwup that's turned into an international loss of face (LOL no), they're in the middle of a massive outbreak that's overwhelming their medical capabilities and has panicked the whole country, which flared up during culturally important occasions where lots of people were on vacation. Do they have accurate numbers to publish?

I doubt that many of the relevant officials want to provide accurate numbers, and I don't think they have them to publish even if they did.
posted by Ahniya at 11:42 AM on February 15


That whole reddit forum is nonsense, people winding each other up. It's disaster porn, the same people are probably the ones having internet freak outs over hurricanes and thinking all of California is going to burn down..

If only we could point them at climate change. Because that's the global catastrophe they've all been waiting for.
posted by fshgrl at 11:29 PM on February 15 [3 favorites]


Also this is a zoonotic virus and we have no idea how it'll mutate but it might just disappear like SARS or rapidly lose virulence. Most of the European cases have been very mild and of short duration.
posted by fshgrl at 2:14 AM on February 16 [1 favorite]


I might be too easily persuaded by rudimentary design skills, but just to winnow things out I have a rule of thumb that if a web page looks a bit TimeCube, it probably is a bit TimeCube. And this web page looks a bit TimeCube.
posted by Grangousier at 3:26 AM on February 16 [2 favorites]


Wow there sure are a lot of epidemiologists around here. And experts on China. And statisticians. We're a smart and educated bunch, but let's have some humility: this is a topic that the most experienced specialists are struggling to grasp right now. Maybe we should be asking more questions than we are positing theories.

I've been having a hard time finding responsible (not full of wild extrapolations) and accessible analysis of this epidemic, partially because data is still limited... but that's what I'm interested in right now.
posted by latkes at 4:20 AM on February 16 [6 favorites]


Latkes, this is one of the most in depth articles I’ve seen that includes the perspectives of multiple researchers and experts..
posted by overglow at 9:58 AM on February 16 [1 favorite]


We were talking about this at work the other day and I learned about the existence of MERS, which is the really scary virus. If it ever becomes more transmissible, it has a 30% mortality rate.
posted by fshgrl at 10:38 AM on February 16


So, my son, Puerto Rican, asked Siri how many people have died from the coronavirus.

However, if you add a Latino accent the question sounds like "How many people have died from Corona beer?" Siri answered, "This is what I found about how many people have died from Corona beer."

This confusion is a thing.

Siri is trying to kill us.
posted by dances_with_sneetches at 7:07 AM on February 20


« Older Even graduate students think poop is funny ...   |   Margaret Atwood on a scooter. Newer »


You are not currently logged in. Log in or create a new account to post comments.