Coronavirus Math: OMG Look What's Skyrocketing!

First off, I know you haven't weaned yourself off checking the infection numbers hourly like you're a panicked day trader.

To be honest, I'm still grappling with this mild form of mental illness as well, but I'm trying to kick the habit ("Hi my name is Michael Wester and I'm a Coronavirus Infection Count addict," I imagine myself telling a crowd in group therapy one day).

But given I'm powerless to stop that train (in fact I take a certain amount of blame for perpetuating it), I'd like to call your attention to a really critical number that is skyrocketing, perhaps even crawling up the elbow of the proverbial hockey-stick growth curve: recoveries.

Yes, recoveries are going way up, as they should — because we're only about 10 days into being constantly glued to these updates and all signs point to this virus being one of the weaker of humanity's recent plagues.

Which brings up another issue: whither the "death rate" for the Wuhan Coronavirus?

Several people have asked me (to which I immediately reply, "I am not an expert in any way, shape or form, I'm just bored and obsessed"): what is the actual death rate for this new virus?

The short answer is we don't know, because as of 11.15am this morning, only 4.4% of confirmed infections have come to a measurable conclusion (either recovery or death).

According to our go-to source, Dr. Lilac, here are the (relatively current) counts:

— 14,441 confirmed cases (I assume this means hospitalized, though I am not 100% sure all confirmed cases are being treated as in-patients)
— 331 recoveries
— 304 deaths

(There are also 19,544 suspected but unconfirmed cases, but let's ignore this for now)

OK so, what is the "death rate"?

Some will say:

304 deaths / 14,441 confirmed cases = 2.1%

However, others (including my 12-year-old) say it should be:

304 deaths / [total known outcomes, aka 304 deaths + 331 recoveries, or 635] = 47.8%

The answer is: both of them are wrong, for two reasons (1) ongoing cases and (2) imprecise language.

(1) Ongoing Cases: Until we have more data, there are just too many unresolved cases to make firm conclusions. It will likely be years (not days, weeks or even months) before all the data is sifted through and a true rate is determined.

(2) Imprecise language: "Death rate" does not mean anything until you specifically qualify what it is you are measuring.

"Deaths / confirmed hospitalizations" is a much more accurate term (though it does not roll off the tongue quite as well).

It's also important to note that even this is not likely the "actual" death rate of the virus, because as with all relatively mild viruses, the vast majority of those who contract one will not seek medical treatment nor be tested for the virus. When's the last time you went to the hospital and were admitted as the result of the flu?

Many have reached for data from other countries, and the US "death rate" for seasonal influenza is often being cited for comparison. This too is clouded by people looking at two different data points:

Deaths / confirmed hospitalizations
vs
Deaths / estimated number of infections

The problem here again is that the vast majority of people who contract the flu do not end up going to the hospital at all (let alone get admitted).

So ... what does this all mean? I don't know, but from the tea leaves I've been reading (aka CDC, WHO, and a number of trusted media sources), this is not an extremely deadly virus.

Reasons to be hopeful include the fact that the vast majority of doctors reporting from the front lines are saying most confirmed patients are currently stable rather than in critical or life-threatening condition, which implies the deaths expressed as a percentage of confirmed cases is likely to remain similar.

Also, the (admittedly very early) trend we have been seeing (as we have been obsessed with these numbers pretty much full time for 10 days now) is that the overall deaths/confirmed cases rate is moving ever so slightly downward, as it was around 2.9% when we started tracking these numbers.

Now, if you'll excuse me, I gotta go check the numbers again.

Comments

New comments are displayed first.
Giovanni Martini wrote:
Lord_Snooty wrote:

A good summary.

There are indications that the mortality rate could be considerably lower than 2%, a recent study of the cluster of cases in one German company indicates that two of the first four patients only displayed symptoms such as a mild sore throat and slight cough. It's not as if the overwhelmed Chinese health system has the time or resources to go searching for asymtomatic/ mildly symptomatic patients right now.

Although these cases in Germany also suggest that this new coronavirus is transmissible when symptomless, which makes me pessimistic that this virus can be eradicated swiftly on a national or global level, and has the potential to overwhelm the health systems in developing countries, particularly SE Asia.

Something that I’ve yet to see mentioned in any news report is that China is estimated to have 100 million people suffering from COPD, hardly surprising when factoring in the air pollution and smoking here. The link between COPD and pneumonia cases/mortality is well-established.

There are so many unknowns and it still feels very much like the beginning, rather than the end of this crisis. It's probably best to reserve any further judgement until further statistics become available in the coming weeks/months.

"There are so many unknowns and it still feels very much like the beginning, rather than the end of this crisis."

By all means, rely on your feelings. I'll stick with ol' Bobby Lee, "Do not take counsel of your fears." Heck, when I was a kid, there was Hong Kong flu. A million people died '68-'69. A couple of generations before that, Spanish Flu killed more foks than WW1. Those were crises. This is highly inconvenient and it sucks (sucks hard, with pointy little weasel teeth.) Armageddon it ain't.

Let's pull out all the stops, eh? What are you going to do if an asteroid hits Earth BEFORE the authorities have corona in hand?

I wasn’t suggesting that armageddon is imminent, simply that there are a lot of ‘known unknowns,’ which will become clearer once the bigger picture emerges in the coming weeks.

You reference the Hong Kong flu, which only had a 0.5% death rate, yet still killed a million, which demonstrates the danger of a novel contagion. According to the official data, the death rate of this new virus is around 2% (although I suspect it’s considerably lower due to the likely number of unrecorded mild cases) and appears highly infectious, so people are understandably concerned. It’s necessary to be cautious and take all the sensible precautions you can during an epidemic, especially one caused by a novel disease, whilst avoiding panic and disregarding media hysteria.

Just like everyone else, I can’t estimate with any degree of certainty when and how this will end, although as coronaviruses are generally known to be more virulent in colder climates, warmer temperatures and higher humidity in May should bring some respite, if things don’t improve before then. Let’s hope they do.

Happy to share this graph from Dr Lilac at 2:19pm Feb 6:

Books by current and former Beijinger staffers

http://astore.amazon.com/truerunmedia-20

Lol my favourite article yet.

A good summary.

There are indications that the mortality rate could be considerably lower than 2%, a recent study of the cluster of cases in one German company indicates that two of the first four patients only displayed symptoms such as a mild sore throat and slight cough. It's not as if the overwhelmed Chinese health system has the time or resources to go searching for asymtomatic/ mildly symptomatic patients right now.

Although these cases in Germany also suggest that this new coronavirus is transmissible when symptomless, which makes me pessimistic that this virus can be eradicated swiftly on a national or global level, and has the potential to overwhelm the health systems in developing countries, particularly SE Asia.

Something that I’ve yet to see mentioned in any news report is that China is estimated to have 100 million people suffering from COPD, hardly surprising when factoring in the air pollution and smoking here. The link between COPD and pneumonia cases/mortality is well-established.

There are so many unknowns and it still feels very much like the beginning, rather than the end of this crisis. It's probably best to reserve any further judgement until further statistics become available in the coming weeks/months.