Hacker News new | past | comments | ask | show | jobs | submit login
Preliminary assessment of the International Spreading Risk of Coronavirus [pdf] (mobs-lab.org)
86 points by Anon84 31 days ago | hide | past | web | favorite | 28 comments

This is the paper summary: https://www.mobs-lab.org/2019ncov.html

And this is the Google Data Studio Visualization Dashboard:


And data is provided by official source from DXY.cn, http://3g.dxy.cn/newh5/view/pneumonia

The least expensive and safest approach of preventing a spreading of 2019 nCorv outbreak globally is still isolation, isolation and isolation.

If the human contact is absolutely unavoidable, then you follow the guide in the link.

This estimate is broadly consistent with that published last week by a Imai et. al. [1] That previous work looked at foreign cases up to 18th January, and concluded the number of contagious cases in Wuhan as of that date is likely in the range of 1000 to 9700.

This paper gives estimates of Wuhan-area cases ranging from 7600 to 43600 for January 26.

A doubling time of 6 days was estimated for the pre-lockdown phase of the Wuhan epidemic. [2] That gives a factor of 2.5 increase in cases between Jan. 18 and 26.

Unfortunately the newer estimated case numbers are somewhat higher than 2.5 times as many; probably the older number were too low.

Note the estimated rate of increase is much LOWER than one would naively get from looking at the officially confirmed case numbers. Confirmed cases have been more than doubling every two days. This growth rate of confirmed cases is really the rate of increase of China's capability of carrying out confirmation labwork.

There exported case research shows there is still a huge backlog of unconfirmed cases to be identified in the Wuhan area.

[1] See "Report 2" at https://www.imperial.ac.uk/mrc-global-infectious-disease-ana...

[2] https://news.rthk.hk/rthk/en/component/k2/1505090-20200127.h...

2.5 indeed looks too low now.

The median estimate of the size of the outbreak by Imai et al as of Jan 18 is 4,000; the new median estimate by Chinazzi et al as of Jan 26 is 21,300 (assuming a catchment population of 20M, their middle case). If both median estimates happen to be spot-on, the size of the outbreak would have grown by a factor of 5.3 between those two dates.

I don't think you've got the right conclusion. The earlier estimate of the total cases was too low.

Probably the newer estimate of total cases is ALSO biased too low. The reason for underestimate is they are assuming international surveillance found all exported cases. The paper describes this shortcoming:

The estimates contained in this report have been constantly growing with respect to older versions compiled with the data available at previous dates. This is because the number of detected cases at International locations and a travel history from Wuhan city has nearly tripled in the last week. It is worth stressing that this is not implying that the epidemic is growing at the same rate. The estimated size of the outbreak refers to all cases occurred in the area since the beginning of the outbreak, and notification and detection delays may play an important role that, at the moment, do not allow the evaluation of the epidemic growth rate.

Thank you. Yes, you're right. It seems more likely that earlier estimates were low.

This is interesting. China can't cover up the true scale of the outbreak because the rate of infection overseas quickly gives the game away if they are publishing very low estimates of domestic infection.

It’s pretty hard to cover up the dimension of an outbreak after it gets outside your borders. There are many things you can use to infer what the real dimension is likely to be.

That being said, it’s also entirely likely that even the Chinese known exactly how many cases there are:

1. Cases aren’t detected unless you feel sick enough to go to the doctor

2. Even if you go to the doctor, the symptoms have to be strong enough for the doctor to order the specific test. This is a bigger factor in the early days. After the alert is given they likely start testing everyone, at least until the outbreak is large enough that it becomes moot

3. The big question (that I haven’t seen answered yet) if how long the assymptomatic incubation period is. In other words how long can you spread it before realizing you’re sick.

In the end, the factor that will determine how large of a global outbreak were actually dealing with is now transmissible it is from person to person. From what I’ve seen all foreign (outside China) cases have been traced back to China, meaning that person to person transmission isn’t a large factor yet. If/when it does become a factor that’s when the real containment “game” starts...

Some other unanswered questions:

- recovery time? E.g. How long for the virus to run its course for those that recover.

- total number of recovered people

- why no babies / kids appear to show up on fatality statistics at least?

- how useful are those surgical masks for preventing getting sick / infecting others?

- what kind of masks actually provide protection against virus?

> - how useful are those surgical masks for preventing getting sick / infecting others?

After SARS the CDC published updated guidance on masks in relation to Influenza in general. TL;DR is use regular paper masks for general population, hospital staff, and patients. N95 masks for staff doing intake and direct treatment. This addresses the primary problem with N95 masks: hospitals running out.

The "N95" rating means it filters 95% of particles 3 microns or larger. Research showed only Influenza particles 4 microns or larger carried enough RNA to replicate successfully.

Keep in mind that said masks don't protect the eyes, and people wearing said masks have gotten sick.

Any chance of sources for this? I could put a few minds at ease with this info

Here is a bit of an overview: https://www.cdc.gov/h1n1flu/masks.htm

You can also read Respiratory Protection for Healthcare Workers in the Workplace Against Novel H1N1 Influenza A: A Letter Report published by National Academies Press for all the details.

There seems to not be much solid info on this. But I have heard reports of medical workers with masks on getting sick.

The leading theory seems to be that eye protection is necessary as well.


If you consider YouTube comments reputable, you're in for a rough run.

I was a bit frustrated when I couldn't easily find an online guide to prevent the spread of coronavirus.. so I went ahead and compiled all the info I read and made this Coronavirus Prevention Guide http://stopcorona.org/

Hopefully it gets seen by enough people who will put the tips into practice and slow down the spread of the virus.

The theory is Wuhan Flu (there are actually a large number of "coronaviruses") is contagious before symptoms show, which is why it is so effective at spreading.

Preventing the spread basically comes down to avoiding people. I'm not sure that is great advice. The rest of this site is just things people should be doing anyway like washing their hands.

If you suspect that you were exposed to coronavirus or have it but are suffering from the milder symptoms, you should proactively self quarantine if you live with others, and wear a mask in public to avoid infecting others.

There is no information on the often-called "symptoms". What are the symptoms one should be aware of, one should check for?

Here's the best source I've found so far, it's quite reputable:


Symptoms and severity vary between individuals but most common is fever and coughing. Very similar to cold and flu symptoms.

To learn more please try googling this topic and look for reputable resources such as webmd or similar.


You made a guide homepage in the post I answerded to, which repeatedly mentions "if you have these symptoms". But the symptoms are not mentioned in the guide.

I think for a proper guide you should absolutely list the symptoms of corona virus at the introduction.

I searched in search engines. As far as I could find the corona virus symptoms are similar to pneumonia. The main symptoms are fever, cough and pain in the lungs.

This is not similar to flu symptoms which do not necessarily spread to the lungs.

I'm amazed Phnom Penh isn't on that list.

there are a lot more cases which aren't on the official figures.

What information do you have about that?

I think it's frankly disgusting that the international community did not learn from SARS. As has been said before, to stop the spread of a virus, you need containment and isolation. All trafic to and from China should have been stopped at the first mention of a new corona virus emerging there until the point where it could be determined either to be contained properly, or it it's lethality and infection rate were considered low enough to not be a potential international risk.

Furthermore, I think China should get some very severe sanctions from the international community for covering up the existence and how wide-spread the virus was.

I just watched HBO's Chernobyl last week, and the similarities are ringing in my ears.

....Under-report to avoid embarrassment.

...."We only gave them the propaganda number, so the aid they sent never could have helped."

....Then wait 'til the international community calls your bluff.

....which requires that the effects spread internationally before anyone uncensored starts getting solid data.

....by which time it's too late to contain.

China is a massive country with a billion people in it. You can’t just shut it down.

Why not? Don't allow ships coming from China to enter the harbour, don't allow planes coming from China, either directly or indirectly, to land at your airports, and for most countries outside of the China region of Asia that would already create a quite effective containment.

Applications are open for YC Summer 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact