A Few Certainties About Covid-19

There is plenty that we do not know about the coronavirus. But let us take stock of the things that we do know for sure, and of some other things that we will soon know.Screen Shot 2020-01-27 at 2.29.15 PM

Real-world Exponentiality

By now, a child understands exponential growth. If you start with one apple on March 1st and double every three days, you will have a thousand apples on March 31st and a million on April 30th.

But in the real world, not the abstract world of math, there are constraints on that growth. Doubling your apples every three days is feasible for a month or so because you can probably find a thousand apples and also find a place to store them. But it would be more difficult to find, transport and store a million apples, unless you are willing to pack a six car garage with apples from floor to ceiling (accurate math). If you did, most of them would rot and your neighbors would call for psychiatric help, two other constraints on unbridled exponentiality.

Similarly, the real-world constraints on the exponential growth of coronavirus cases are either 1) a saturation of the population, when all the people susceptible of being infected are infected, or 2) human intervention that stalls the spread long before we reach that saturation.

The first case, saturation, means that say 50 to 80% of the population would get the virus, or 165 to 265 million people for the United States. Assuming a 1% fatality rate, this would result in 1.7 to 2.7 million deaths, an unacceptable death toll. Further, the fatality rate may be higher than 1% when hospitals are overwhelmed.

If you approach the question with more granularity and apply a different fatality rate for each age group, you also end up with an unacceptably high death count. It is true that coronavirus deaths have occurred mainly among elderly people who had comorbidities, but these lives were a long way from ending. The average age of the dead in Italy is about 80 years, but life expectancy in Italy at age 80 is nine years.

Any death is “unacceptable” but here, we are measuring this total against the incremental number of deaths that would result from a continued shutdown or slowdown of the economy. We are in the midst of another pandemic of what Princeton Professor Angus Deaton has called ‘deaths of despair’, deaths from suicide and substance abuse. We would not shut down the economy for a peak of 10,000 coronavirus deaths, and indeed we do not shut it down every year for the estimated 25,000 to 60,000 fatalities that result from the seasonal flu.

Because “letting it burn” would exact a high death toll, governments all over the world have opted for the second constraint on exponential growth, human intervention to kill the spread. Home confinement has delivered the wanted results in China and looks on track to stall the spread in Europe. Below is the daily growth in confirmed cases in Italy and France over the past three weeks (smoothed over five days).

Screen Shot 2020-03-27 at 1.47.32 PM
Sources: populyst, WHO, Worldometers

Though nominally encouraging, these numbers should be treated with caution. First, if the number of daily tests is constant or growing very slowly, then the declining percentage would be a normal result and not an indication of improvement. Second, these are only confirmed cases. The number of actual cases may be significantly greater but with home confinement, it is  probable that it has a similar declining growth profile.

Nonetheless, the daily change in the number of deaths also shows an encouraging trend. The early March data for the US is not significant as numbers were relatively small.

Screen Shot 2020-03-30 at 1.45.40 PM
Sources: populyst, Worldometers

Finally, the other side of human intervention is to find a cure or vaccine. Many people are working on it but there is no visibility near-term on when these efforts will bear fruit.

Incomplete numbers

We know that the numbers given by the World Health Organization (WHO) and by other sources such as Worldometers represent only confirmed cases, not the actual total of cases in each country. I tried to address this issue in a recent podcast by estimating the likely gap between the reported number of confirmed cases and the unknown number of actual cases. I took the number of dead + critical in each country and divided it by a fatality rate of 1 to 3% depending on the country.

I cannot be entirely pleased with the analysis and recognize that the numbers that I estimated could be not just off, but off by several orders of magnitude. We simply do not know enough about the likely number of asymptomatic people out there. In the next few weeks, we will have more clarity and I expect that we will be able to estimate actual cases more accurately. As testing expands, that number will converge with the number of confirmed cases.

Insufficient testing

We all know that testing needs to expand significantly. As of today, March 27, it was expected that the US would be able to run 100,000 tests daily. How will we know when we are running enough tests to identify all or most of the cases? We will know when the percentage of positives/all tests drops below 10% and then below 5%, or when the percentage of dead/confirmed cases falls below 1%.

Although the number of positives nationwide is now 15.5% of all tests, New Jersey and New York still have over 30% positives/total tests. New York state already has run 23% of all US tests, but we still need to expand testing significantly in New York, probably by a factor of 3 or 4. That suggests that we need total testing capacity nationwide of at least 200,000 per day, or possibly much higher if other hot zones develop after New York around the country.

*  *  *

Here are some things that we do not know for certain but that we will know soon.

Effect of lockdowns

Italy, France and Spain have been in lockdown mode for several weeks. The numbers of confirmed cases in all of these countries are growing at a declining rate. It will take another few weeks to evaluate more confidently whether the lockdowns were effective to stall the spread for good, or only to slow it down temporarily.

Seasonality

As temperatures warm in the northern hemisphere, we will find out whether spring and summer will cause the virus to recede. Some research is fully supportive of this hypothesis but not everyone is convinced. Other research shows that warm weather will slow down the virus but will not cause it to disappear. There is some opinion that it would live on in the southern hemisphere and that it could strike north again in the fall. If such is the case, we should be better prepared by then.

Chloroquine  and other treatments and vaccines

The chloroquine and hydroxychloroquine + azythromycin treatments advocated, among others, by Professor Didier Raoult in France will be found to be effective or not effective in coming weeks.

There are other drugs and vaccines in the works, from Regeneron, Sanofi and others. I am not going to pretend to know them or to give you an exhaustive list. You can find better sources in science or popular science publications.

Although we live in a difficult period and the headlines will be unnerving for a while longer, some uncertainty is about to lift. With human ingenuity, seasonality, and a bit of luck, we will skirt the more extreme scenarios.

Related: The Pandemic as a ‘Gray Rhino’ Event, with Michele Wucker

One thought on “A Few Certainties About Covid-19

  • Amazingly, Early Results from Europe as reported weekly by the Euro MOMO project (European MOnitoring of excess MOrtality) this flu season is seen to be surprisingly mild compared to previous flu seasons in the EU in 2016/17, 2017/18, 2018/19. See their website https://www.euromomo.eu/index.html

Leave a Comment:

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s