Covid-19: Part 3 — The Covid Deniers (A Data Science Perspective)
“No Worse than the Common Flu”
Many home remedies allegedly offer protection against Covid-19.
It is claimed that citrus fruits, garlic, sunshine and fresh air improve immunity and so reduce the chance of Covid-19 infection. I am not sure to what extent the claims have been established through rigorous controlled experiments, but obviously these do no harm and might very well do good.
Some Indian villagers are covering their houses with mango and neem leaves and spraying a liquid made from turmeric, salt, neem and nochi leaves to ward off the corona virus. This does no harm either as long as they also practice social distancing. And the leaf covered houses do look charming!
Other touted remedies such as colloidal silver and cow urine are more dubious and may cause real harm.
And yet do the “experts” - the epidemiologist, the virologists and the statisticians — know any better about Covid-19 ? Many doubt.
There is a minority who continue to deny the seriousness of the Covid-19 outbreak and claim that the attention being paid to it is largely a result of media hype. I will call such people Covid Deniers.
In many cases the Covid Deniers’ arguments are based on faith and hope. But some of them do back up their claims with data, analysis and logic.
Contrarian views deserve a fair hearing. So let’s to try summarize the Covid Deniers’ arguments in a few bullet points.
(1) The death rate for Covid-19 is uncertain. It could be much lower than believed right now & quite possibly comparable with ordinary flu (~ 0.1 %) instead of the widely claimed 2–4%.
(2) Ordinary flu takes tens of thousands of lives each year in the US. Tuberculosis and malaria routinely kill hundreds of thousands in India. Yet these diseases barely register on our consciousness while the dread of Covid-19 is all encompassing and drives government actions.
(3) This fear of Covid-19 is unreasonable. It is dreaded more than familiar diseases because it is unfamiliar. Fear is also fanned by the media hype around the disease; you would have been equally spooked by flu or TB if a constant parade of death and case numbers kept marching across your screens.
(4) Let us for the sake of argument — continues the Covid Denier with an appearance of great fairness — accept that Covid-19 has a fatality rate about 10 times higher than ordinary flu. Even so, the fatality rate differs greatly according to age and previous medical condition. Most of the dead are among the old and the sick. For healthy young adults Covid-19 death rates are no worse than that of flu. Why then should we lock down the entire population? Let the under 40’s go about their business as usual. We might, if necessary, consider taking some steps to protect the more vulnerable older segment of the population.
(5) And yet governments are clamping lockdowns across the world on the basis of this very dubious fear. The Coronavirus might not kill more than other routine diseases but lockdowns surely will, by destroying the economy.
The Covid Denier’s arguments seem hard-headed, fact-based, logical and persuasive. They look so right. So convincing!
But wait a moment. Let’s examine them in greater detail.
Let’s begin with the first bullet. As we have seen in Part 2 of this blog, there is considerable uncertainty surrounding the calculation of the true Covid-19 death rates, i.e. the proportion of all infected persons who die from the virus. This is necessarily lower than the observed death rate based only on confirmed cases because of reasons that we have discussed.
A paper published in the Lancet Journal of Infectious Diseases on 30th March. suggests that Covid-19 true death rate is around ~0.66% and hospitalization rate around 3.3%. The numbers are indeed lower than what had been thought though still considerably higher than common flu. Of course any estimate made when the epidemic is in progress is questionable but this seems to be the best we have.
But in any case the death rate may not matter that much. Covid-19 could be a much bigger killer than common flu even if the death rates were not much different.
That possibly sounds counter-intuitive & paradoxical. So, let’s work through a numerical example to see why.
Take India with its vast 1.4 billion population. Assume that just 1% of that population will become infected by Covid-19. Further assume that 1% of the infected cases will be hospitalized and 10% of the hospitalized (i.e. 0.1% of the infected) will die. Then we are looking at “only” 14,000 deaths and 140,000 hospitalizations. Surely that’s a tiny number for India?
But wait. Suppose that all infections, hospitalizations and deaths happen in a month or two. What would be the impact?
It would lead to the complete collapse of India’s underfunded and overstretched healthcare system. India is estimated to have less than 100,000 ICU beds and around 1.4 million hospital beds in all. And obviously these resources cannot be reserved just for Covid-19 patients since many others also require urgent care.
Hospital Beds in India
As per the provisional data compiled by the Central Bureau of Health Intelligence (CBHI), there are 13,76,013 beds…
What India can learn from China and South Korea to ward off coronavirus
by Soham D Bhaduri Covid-19 cases have well entered the triple-digit territory in India. Our approach has grown from…
In fact the healthcare system in India would collapse much before 1% of the population is infected. If Jacob Steinhardt is right — though his analysis does seem a little questionable to me — then just about ~1% of the Spanish population were infected on 20th March when Spain’s health care system had already begun to collapse.
Estimating Coronavirus Prevalence by Cross-Checking Countries
As we scale up efforts to combat Covid-19, a big unknown is the infection prevalence in different countries. Since…
And please remember that Spain boasts an extremely advanced healthcare system ranked by WHO as the seventh best in the world. India’s is beginning to creak with just around 3,000–4,000 confirmed cases; shortage of PPE (Personal Protective Equipment) is already being reported at many hospitals.
So what would happen to the number of deaths as India’s healthcare system collapses?
In our example we had assumed that 0.1% of the infected population will die. This is about 6 times lower than the best currently available estimate. But of course the death rate is not a constant of nature like the velocity of light. It depends on several factors, one of the most important of which is the quality of care. Recall from Part 2 of this blog, that Covid-19 death rates vary by an order of magnitude between countries.
So the number of deaths in India will keep going up as hospitals go under. The death count will be in hundreds of thousands rather than tens of thousands.
And we have supposed that just 1% of India’s population would get infected? What if exponential growth takes the number to 5% before the epidemic peaks? 50%? India could potentially be looking at millions of deaths — a holocaust on a scale that has not occurred since its independence in 1947.
The Covid Deniers are wrong. Dead wrong!
Covid-19 is much more serious than the common flu because:
a. The disease has a fatality rate much higher than the flu — perhaps six times higher;
b. The disease is more contagious than the flu. An influenza patient on the average infects 1.3 others while according to current estimates that figure is about 2.5 for Covid-19. So Covid-19 can infect more people in a short term and overwhelm health care systems.
The Covid Deniers
The Covid Denier’s case that I presented is not a straw man argument. It is based on articles like the following:
Why Flattening the Curve is Overrated | Industry News | Pensford
I'm about to upset a lot of people this week and I apologize in advance. I actually deleted a lot of this from last…
(Cass Sunstein — a top lawyer who played a significant role in the Obama administration — has apparently now changed his mind about the seriousness of Covid-19. But all deniers haven’t.)
What would be the policy consequences of Covid Denial? Probably something like what Ray & Subramanian (referenced in Part 2) propose and what UK was trying out before the Imperial College epidemiological model convinced them that it could lead to a quarter million deaths.
As yet, very few Covid deniers are in a positions of power and influence though there are exceptions.
In Tanzania all schools have been closed, but President Magufuli refused (at least till 23rd March) to shut places of worship. He claimed that the coronavirus, being satanic, could not thrive in churches.
The premier football (soccer) league has been suspended in all countries of Europe except one. The exception is Belarus. Alexander Lukashenko, the President of Belarus advises his countrymen not only to play and watch football but also to drink vodka, go to work, and visit saunas, though not necessarily in that order.
Jair Bolsonaro the President of Brazil has famously (or infamously) referred to Covid-19 as a “little flu”. Donald Trump, the President of the most powerful country in the world, has waffled and often seemed to come down on the side of the Covid Deniers. However, both seem to be taking the crisis more seriously in recent days. But when the lockdowns begin to bite both Trump and Bolsonaro may flip-flop again.
Taming Your Models
Covid-19 modeling has become something of a cottage industry now.
Google search of “Covid-19 modelling” threw up more than 3 billion results. “Covid-19 modelling in R” threw up 318 million results. “Covid-19 modeling in Python” threw up around 17 million results.
At least 2 calculators which use the relatively simple SIR (Susceptible, Infected, Removed) method to model epidemic progress are freely available online.
Country specific models are now available for the US, UK, Netherlands, Kenya, India … in fact more than one model for some countries including India.
The models differ widely in their predictions which is of course grist to the Covid Denier’s mill.
So how should policy makers use models?
Can they be combined with other statistical techniques to help avoid a complete economic collapse while half the world goes into lockdown?
Please stay tuned for my next post
Covid-19: Part 4 — How to Tame Your Models (A Data Science Perspective) — Forthcoming
Disclosure: I am the Director of Smart Consulting Solutions Pte Ltd, incorporated in Singapore and its subsidiary Radix Analytics Pvt Ltd, incorporated in India. I am also a Visiting Faculty Member at the Indian Institute of Management, Udaipur. However, the opinions expressed in this post are solely mine and not necessarily shared by any company or institution with which I am affiliated.
Tomas Pueyo’s articles still remain the best data-based analysis of the Covid-19 crisis.
I am, as always, grateful to my friend Dr. Ashish Kumar Dawn for many insightful comments and sharing some documents.
Navin Suri and Rajat Poddar referred me to some of the articles discussed in this post.
Do please share the post if you think it may inform or interest others. I invite comments on the post. Please feel free to write to me at firstname.lastname@example.org
I have spent over 30 years in academia and industry exploring how to use mathematical methods to solve real world problems
Stay up to date on coronavirus (Covid-19)
- You tested positive for Covid-19. Now what?
- How to clean your hands during the coronavirus outbreak.
- The 3 most promising coronavirus treatments, explained.
- Staying sane amid the coronavirus crises.
Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage — with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade