Sometimes, statistics can be scary, or at least, made to look more menacing than what these unbiased, risk-adjusted numbers really say.
Conversely, threatening, inconvenient data can be slighted, ignored, or perhaps misrepresented as something harmless, or even beneficial.
Both the above cases apply to the reporting of data regarding the spread of SARS-CoV-2 and the resultant pandemic of Covid-19.
Looking from a statistical point of view, one thing stands out starkly.
The data available about Covid-19 is currently incomplete, hence media speculation about the mortality (death) rate and prevalence of infection is mostly that: Raw, uninformed speculation – and much of it is about as useful as the lies that drinking silver solutions or eating garlic can prevent Covid-19 (note that silver is not even a metal involved in any human bodily processes).
That is not to say the current data available about Covid-19 is useless.
In statistics, where data is incomplete, several techniques can be applied to “clean up” the numbers and extrapolate the likely scenarios needed to manage the scope, extent and damage done by SARS-CoV-2 in various countries.
And we need to know this information as soon as possible, so that the world can know what measures are the most effective in defeating this coronavirus pandemic NOW, and also importantly, to take measures that can prevent recurrences in the future.The immediate thought about Covid-19 is that there are only two outcomes that are of primary interest:
- Recovery from the disease
However, looking at the World Health Organization (WHO) and US Centres for Disease Control and Prevention (CDC) statistics, it would appear that there are very different ways people can survive or die from Covid-19, and it is important to understand why different countries have different rates for both events.
For example, is it demographics (e.g. more older people in some countries) or is it related to better healthcare?
Or is it something else, or more usually, a combination of different factors?
Also, how does statistics help end this SARS-CoV-2 pandemic?
The following explains the data that the media should be providing and why they are needed.
One immediate problem is the media focus on the mortality rate of Covid-19.
The US news agency CNN recently reported that the death rate is “less than 1%”, and hence, Covid-19 may not necessarily be a huge problem.
The US government also likes to talk down the death rate, often quoting the higher mortality numbers from normal influenza.
Note that these statements are made before knowing the true Covid-19 CFR applicable in the United States.
CNN got its “less than 1%” death rate from South Korea, where the current number of infections (as of Mac 9,2020) is 7,513 and the number of deaths is 58, giving a coarse death rate of 0.77%. This number is not statistically relevant.
The reason is because the number for “T” is not known, but if we assume a low parameter of 10, then based on the number of infections on Feb 29 (2020) of 3,150, the CFR needs to be adjusted to 58/3,150 = 1.84%.
So CNN is probably very wrong about reporting a death rate of less than 1%. That is just how statistics work when processed dispassionately.
The true CFR for normal influenza is 0.1%, and this number is derived from the known statistics and distribution of influenza sufferers over several years.
At present, the contagion rate known as “R0” (or average number of new infections generated from a single infected person) is not known precisely, but we have a good guess.
The pattern of how Covid-19 spreads appears to be initially in small clusters, which immediately become widespread within a short space of time.
That was the pattern in Wuhan where it began.
A more recent example is Italy where only 323 infected cases on Feb 25 (2020) exploded to 9,172 cases two weeks later on Mac 9 (2020), an over 28-fold increase.
Although the Italians claim surprise at the infectiousness of SARS-CoV-2, they have no right to do so.
The contagion rate from Wuhan had been earlier investigated by Imperial College London in Britain.
To match the observed spread rate of the disease there, they calculated a transmissibility rate of 2.6, i.e, every infected person would infect 2.6 other people unless quarantined.
The paper Transmissibility of 2019-nCoV was published on Jan 25 (2020) on the institution's website after previously being shared with WHO, governments and academic networks. It and was obviously not taken seriously by many governments at the time.
The statistic is stark. As long as the value of R0 remains significantly above one, SARS-CoV-2 will spread relentlessly unless it is contained by prolonged quarantines.
As a comparison, normal influenza has a R0 of 1.5 and managed to infect an estimated 35,520,883 people in the US last year (this was an improvement from 45,000,000 the year before).
Importantly, there were only 491,000 flu cases that required hospitalisation in the US in 2019, or 1.38% of the total infected with influenza.
Without a vaccine or standardised treatments, the only known options for containment are via:
- Stringent quarantines
- Early detection
- Contact tracing of all persons associated with any newly-detected infection
- Scrupulous personal hygiene, especially regarding washing hands with soap, and
- Social distancing (e.g. cancellation of large-scale public events, closing of schools, etc).
In statistical terms, China has achieved the impressive feat of reducing R0 from (an estimated) 2.6 to practically 0.
Whether it can remain at this level remains to be seen. However, this is the target R0 that all countries are required to achieve to end this pandemic.
Note that several vaccines have been developed against influenza and it still managed to infect over 35 million people in the US alone last year (2019).
As such the Covid-19 situation remains perilous without any such preventive options.
Chris Chan works in advanced statistical and mathematical modelling of risks in large investment banks and often applies his analytical expertise to other fields such as chemistry and biochemistry. He also writes the Curious Cook column for StarLifestyle. For more information, email email@example.com. The information provided is for educational purposes only and should not be considered as medical advice. The Star does not give any warranty on accuracy, completeness, functionality, usefulness or other assurances as to the content appearing in this column. The Star disclaims all responsibility for any losses, damage to property or personal injury suffered directly or indirectly from reliance on such information.