Manipulating the Cloud

Some thoughts on certainty

Nothing is certain. Things only look certain when their cloud of probabilities becomes narrow enough for us to ignore the rest. When we act to affect a certainty, what we are really doing is manipulating the cloud of probabilities around that idea.

This is true for medical diagnoses though we don’t like to think so. When a doctor tells us we have pneumonia and prescribes an antibiotic, we are hopeful that she is certain of the diagnosis as well as the causative organism and best treatment. In fact, that is rarely the case.

In a related way, when dealing with a new epidemic whether it’s the opioid overdose crisis or the possible pandemic rise of a novel virus, we are hopeful that public health experts know the exact cause and the best response.

But here too, because there is no certainty, there can only be actions to change the statistical cloud of possibilities around the phenomenon. The range and effectiveness of our attempts to manipulate this cloud is dependent both on intrinsic factors of the illness in question as well as our own assumptions regarding the cloud. We’ll take a look at both of these using the novel coronavirus most lately in the news.

The Illness

It should be clear to any observer that little is certain about the causative virus of SARS-CoV-2. We know it is a coronavirus. We know that coronaviruses can remain infectious on surfaces for a week or more. We aren’t certain how long this virus can live on surfaces. We know that this virus can make some people extremely ill, but we don’t know what determines who gets very ill, who has a mild case, and who wards it off completely. We know that the human protein ACE2 is the receptor for coronaviruses, but we aren’t sure that this virus is using that same receptor. We know this virus can cause epidemic spread when unchecked, but we are not clear on the measures of that spread. The R0 (the number infected by each case) ranges from 2 to 6 in some studies. We are sure that it can kill because it has, but we are not sure of the lethality. The latest estimate of the Case Fatality Rate (CFR) is around 2% on average. We are sure it is treatable but we aren’t sure which treatment is best at which stage and what treatments to use in cases where the medical system has been overwhelmed.

That’s a lot not to be sure about. And yet, given certain goals and assumptions, our logical course could be clear. So let’s check those goals and assumptions.

The Goals

In general, the goal in any public health endeavor is to save lives and prevent morbidity. I have no doubt that this is the goal of each individual healthcare worker on the front lines of response to this virus. However, large organizations such as WHO and CDC are part of the government or are quasi-government agencies. They have pressures on them from funding and supporting entities that individual doctors don’t experience.

To react too quickly and too forcefully with the goal of saving maximal lives, it may be that the epidemiological organizations will ‘overreact’ at times. This is almost certain because when they do react forcefully, they create a self-negating prophesy. By predicting danger and promoting a strong defense, the predicted risk is not experienced. Observers are left to experience the event as an overreaction.

When a country’s economy is highly levered and designed for no mistakes or emergencies, those planning economic activity for the country may feel that overreaction to a potential epidemic would spread fear and market disruption hurting more people than the illness would. The more financialized and levered the country’s general economy, the more this will be the case.

This idea of competing goals of government and public health workers leads us to look at the assumptions of these two groups.

The Assumptions

Any scientist knows the power of nature. Every scientist who has studied viruses knows that they have been at war with cellular life for hundreds of millions of years. Any gap in cellular defense or any novel tool the virus can gain can lead to rapid spread through a population. Such can happen when a virus jumps from an animal host to humans.

In general, this cross-species transmission of the virus results in a milder form of the illness than in the original host. A virus may be idealized for bats, and so, doesn’t do well at first in a human host. But viruses mutate rapidly, and those that mutate to become more virulent (cause more sneezing for instance) get spread more. So, with every generation of passage between humans, a new virus will tend to get more and more virulent until it reaches a stage where it starts killing the host so rapidly it cannot spread. Then evolution would push it in a new direction of lowered virulence with each passage.

This is the reigning assumption on how the 1918 influenza virus jumped to humans somewhere in Kansas as a mild form of flu. After multiple passages through humans around the Earth, it emerged the next season as a much more deadly strain. For perspective, its peak R0 was estimated to be around 2 and the CFR was around 2%.

That means we can’t be sure where or when the SARS-CoV-2 virus jumped to humans. We just know it came to our notice as a virulent infectious agent in China. We can’t be sure because currently testing for the virus is tightly controlled in several places like the US. We will know soon, hopefully.

So it is the working assumption of epidemiologists that this virus jumped to humans at some point and is now getting worse via passage or has peaked in its virulence and will decline with the passage. Not knowing is a good working assumption.

Governments, on the other hand, have other assumptions. They assume that they make the rules. They assume that they know the truth. And, like King Knute, they are most often disappointed that nature refuses to follow their assumptions.

Manipulating the Cloud

Rather than think we know exactly what we’re fighting and creating an impenetrable wall against it, we should keep our hubris in check. We could build a flexible system that can respond to an ever-changing reality. But the cost of this seems great.

First, the economic cost seems high because some profit each year must be given over to a lack of efficiency due to the need to be flexible. But this cost pales in what it costs us to stop a more efficient system every 10–20 years as China did this year and as the world did in 2003.

Seemingly more crushing is the psychological cost to each of us. Our brains are wired for certainty. When faced with a coherent narrative, our midbrain reward system sends us pleasure signals that tell us everything is fine. When faced with uncertainty, these signals go away and we get a distinctly unpleasurable feeling. We don’t like living with that so when someone in charge tells us they are certain, we will buy it.

But the truth is, there is no certainty. We can only push against and manipulate a cloud. And as hard as that sounds, it’s a lot easier when you recognize what you’re doing instead of pretending you’re doing something else.

*******************

Dr. Wetsman has retired from trying to fix the world, but if you want to know more about addiction, try his videos on Ending Addiction. He’s currently engaged in writing fiction which you can check out at his free Patreon. His first novel is The House on Constantinople Street.