Last 5 posts

Archive

01 July 2020

Keeping Elephants off the Track



This article was written by Steven Spencer. All views and opinions expressed are strictly his own.


It’s been three months now and by general agreement I have not written about the dreaded coronavirus.

As the wonderful physician Hans Rosling once began one of his lectures (look them up, they are brilliant) “I am now going to talk about statistics. Don’t leave! Statistics can be very interesting!” So I am going to talk about statistics. Don’t log out!

It was rightly said, I think, at the early stages, that the spread of the virus depended on two things, population density and how dense is the population. Capitalising on the latter, mainstream media has delighted in publishing, without the slightest checks or examination, every dire prognostication emanating from the scientific advisers, such as the now infamous Professor Neil Ferguson. People believe what they read in newspapers, which then drives popular pressure for instant fixes from the government. But what if the predictions were wrong?

Now, mathematical biology is a respected science and Ferguson is a respected practitioner, but “rubbish in, rubbish out” as they say, and statistics 1.01 is “never average averages”.  The Imperial College model that has formed the bulk of the research underpinning the government’s handling of the crisis was developed over a number of years and is based on analysis of past virus epidemics, including the Spanish Flu of 1918-20, the  H2N2 virus of 1957, the SARS epidemic of 2003, and the H5N1 “Hong Kong” bird flu of 2005. Another epidemic that is included is the H1N1 Swine Flu of 2009. I need to include here the 1951 flu epidemic, the 2001 outbreak of foot and mouth disease and the 2002 outbreak of BSE.

The predictions generated by the models were wildly wrong in every case. In 1918-20, Spanish flu accounted for up to 50 million deaths worldwide. Relying on the transmission data of this virus compared to COVID19 is dangerous because of massive demographic shifts in post WW1 populations, weakened by 4 years of war and in the absence of modern medical intervention, plus the demobilisation of armies who returned from the Western Front. The plague probably, but not definitely in this case, originated in a Kansas military facility and was brought to Aldershot by US troops in 1917, then spread through the armies on both sides of the Western Front including Indian troops and Chinese labourers. Their return home was the principal spreading factor. These conditions do not arise today and in fact it was not known whether the Spanish Flu WAS a flu virus. That was not discovered until 1933. By the way, it didn’t originate in Spain: that was the name given because all western media was censored in wartime, but Spain was neutral and so was the only country to publicise the outbreak.

The COVID virus, having originated in China in late summer 2019, allowed Chinese from infected areas to spread the virus three times. Students going abroad to study in October, students returning to Western colleges in January and students returning home for Chinese New Year and bringing the virus back a third time in February. Add to that business travel in both directions and you have a perfect storm. As most of these travellers would be young and active, many would have carried the virus but been asymptomatic, or at worst not suffered more than mild symptoms. Totalitarian regimes have a long record of not admitting failures that could reflect on the Party. So the WHO were not informed of human to human transmission by the Chinese authorities until it was far too late. That resulted in governments imposing the lockdown later than would have been ideal, but the predictions of fatalities were still wildly wrong.

Let’s look at the Imperial predictions: in 2001, predictions of human deaths from foot and mouth were up to 150,000. There were fewer than 200 human deaths but millions of animals were culled costing the UK over £10 billion. In 2002 the prediction was 50,000 human deaths from BSE. The total was 177. In 2005 the prediction was that up to 150 million people worldwide could die from bird flu. The death toll? Just 282 over six years. In 2009 the prediction was that 65,000 would die in the UK from Swine Flu – the total was 457. And this year the prediction by the Imperial College model was up to 500,000 deaths and maybe 2.2 million in the USA.

Why were the predictions so dire? Statistical Modelling Columbia said “Last March, Ferguson admitted that his Imperial College model of the COVID-19 disease was based on an undocumented, 13-year-old computer code that was intended to be used for a feared influenza pandemic, rather than a coronavirus. Ferguson declined to release his original code so other scientists could check his results. He only released a heavily revised set of code last week, after a six-week delay.”

Julian Todd, a computer programmer and FoI activist wrote to Imperial under the FoI Act and asked:

“I have searched very hard for where your code might be published, but have been unable to find it. Therefore, under the Freedom of Information Act 2000, please can I be sent copies of:

(1) The source code for the simulation models being used in the above paper;

(2) The documentation to compile and run the simulation models;

(3) The data used as input to the simulation models that is either public or you have permission to publish, or the anonymized synthetic data derived from the unpublishable data;

(4) The scrapers used to download the data.”

To which he got the reply that “information intended for future publication is exempt from the FoI Act” and nothing more.

The problem now is that because the models were so poor and government policy was therefore dictated by faulty statistical data, resulting in hundreds of billions, perhaps trillions of pounds and dollars lost to the economies on both sides of the Atlantic, the post-crisis analysis will never prove what caused infections to cease.

I am reminded of the story of the passenger on a train who was tearing pages out of his newspaper, rolling them into balls and throwing them out of the window. The guard asked him what on earth he was doing. The passenger said “I’ve always done this. It keeps elephants off the track.”. The guard said “There are no elephants on the track”. The passenger said “I know that! It’s because of the newspaper”. 

comments powered by Disqus