Forecasting Cases & Duration of Severe Acute Respiratory Syndrome (SARS) [A long but good read shoiwng that 'it's exponential, bro' is wrong; it's logistic]

http://condellpark.com/kd/sars.htm ^ | April 18, 2003 | Kevin Duffy

[DoodleBob]

Can simple mathematical models give an insight into the question of how many people will be infected by SARS and how well it is being controlled ?

In late 2002, a severe form of pneumonia called Severe Acute Respiratory Syndrome (SARS) broke out in Southern China. It was met by a government determined to deny its existence. Shortly after the Chinese New Year 1st February 2003, SARS spread across the world.

By mid April, over 1200 people in Hong Kong had been infected, including many health workers, and its 6 million people were in a state of virtual panic.

S-shaped Logistic equations have been shown to accurately model the February 2001 outbreak of Foot and Mouth Disease in the UK. What can they tell us about the 2003 SARS outbreak ? With hindsight, the outbreaks in Singapore and Canada were brought under control by early May, but logistic curve-fits made that apparent even from early April.

The results show here that while in the initial stages of an outbreak, the disease seems to spread exponentially, this situation can not be maintained and the outbreak tapers off logistically. While there is a need to respond agressively against the disease, even in April 2003 there was no need to panic - the sky was not falling in.

[DoodleBob]

Yes, I know...SARS isn't COVID19, but the idea is the same. This is a long but excellent read on why the "it's exponential, bro" people are just as wrong as the "it's just the flu, bro" people.

Here is a chart of the actual SARS case load and a fit of the logistic function to the data for Hong Kong...

...Taiwan...

...Singapore...

...and Canada (with it's quirks)...

Now, reasonable people can debate as to how to best 'flatten the curve.' I don't think these countries vaporized their economies to stop SARS, but they certainly imposed SOME controls. I also don't think they just sat there and fiddled. But as I've said elsewhere, I don't like Dilemmas: Give me Door Number 3!

[Jim W N]

Actually, according to the Mayo Clinic, COVID19 is a type of SARS.

“Coronaviruses are a family of viruses that can cause illnesses such as the common cold, severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS). In 2019, a new coronavirus was identified as the cause of a disease outbreak in China.

“The virus is now known as the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The disease it causes is called coronavirus disease 2019 (COVID-19).”

https://www.mayoclinic.org/diseases-conditions/coronavirus/symptoms-causes/syc-20479963

[caww]

....."COVID19 is a type of SARS"....

Once you realize that then it's understandable why the research people had their concerns at the start.

[bigbob]

According to “of one of the many Kevin Duffy’s on the ‘net.” who gives us nothing on which to evaluate his knowledge or credibility on the topic.

Hard pass.

[Jim W N]

If COVI 19 is a SARS issue, then the same type of logic is in play as is discussed here.

“FORECASTING CASES & DURATION OF
SEVERE ACUTE RESPIRATORY SYNDROME (SARS)

“Can simple mathematical models give an insight into the question of how many people will be infected by SARS and how well it is being controlled ?

In late 2002, a severe form of pneumonia called Severe Acute Respiratory Syndrome (SARS) broke out in Southern China. It was met by a government determined to deny its existence. Shortly after the Chinese New Year 1st February 2003, SARS spread across the world.

“By mid April, over 1200 people in Hong Kong had been infected, including many health workers, and its 6 million people were in a state of virtual panic.

“S-shaped Logistic equations have been shown to accurately model the February 2001 outbreak of Foot and Mouth Disease in the UK. What can they tell us about the 2003 SARS outbreak ? With hindsight, the outbreaks in Singapore and Canada were brought under control by early May, but logistic curve-fits made that apparent even from early April.

The results show here that while in the initial stages of an outbreak, the disease seems to spread exponentially, this situation can not be maintained and the outbreak tapers off logistically. While there is a need to respond agressively against the disease, even in April 2003 there was no need to panic - the sky was not falling in.”

Key point here - act quickly and aggressively but don’t panic and don’t overkill.

[NobleFree]

who gives us nothing on which to evaluate his knowledge or credibility on the topic.

That's why one needs to be mathematically and scientifically literate: then one can evaluate the results (which are what matter) rather than the credentials.

As a PhD physicist currently employed as a predictive modeler, I say Kevin Duffy nailed it.

[NobleFree]

Logistic fittings for COVID-19:

“Based on the data from 11. Mar 2020, a very rough estimate indicates that the number of cases [outside China] will be about 90000 (Fig 7); data from 13. Mar 2020 rise this number to 380 000. However, it is an early-stage epidemic, so these estimates are very questionable and will be changed with new data.” - https://www.researchgate.net/publication/339240777_Estimation_of_the_final_size_of_coronavirus_epidemic_by_the_logistic_model

[DoodleBob]

This is just what I was thinking should be modeled. Thank you..BFL.

I agree - Mr Duffy got it...and sadly it was 15 years ago and nobody (except us) is reading it.

[WildHighlander57]

Thanks for posting; bookmarking....

[FreedomForce]

I hope this sort of model for COVID-19 is how it goes. Even the possibility eases my mind.

[Trump Girl Kit Cat]

Difference is we did NOT shut the entire economy down because of SARS this panic has become ridiculous!!!

[tarator]

> I hope this sort of model for COVID-19 is how it goes. Even the possibility eases my mind.

Why? Just because it is an S-curve? Ebola growth is also governed by a logistics equation. All virus growth can be modeled by a logistics equation. That by itself is not particularly remarkable. It is about how steep and tall that S curve is. And how lethal the disease is.

[Arones]

According to “of one of the many Kevin Duffy’s on the ‘net.” who gives us nothing on which to evaluate his knowledge or credibility on the topic.

I've never heard the Kevin Duffy reference before? What does it mean? I ask because one of my professors in my data science MS was name Kevin Duffy-Deno.

[FreedomForce]

Because the numbers in the link NobleFree posted suggest this epidemic may not be as bad as feared. Still lots of unknowns at this point though.

[Arones]

Ah...I see. The author’s name.

[Hugh the Scot]

“Logistic”..

SMH