The method governments are using to measure deaths caused by Coronavirus is only now beginning to emerge. It is fundamentally flawed, and it suggests that the disease is a lot less dangerous than had been imagined.
The early twentieth century philosopher Bertrand Russell cited a famous example of the fallacy of causation. Just because two events take place at the same time, it does not follow that one causes the other. He observed (somewhat anachronistically by contemporary standards) that the factory whistle blows in London at 5.00pm, and at 5.00pm all the workers go home in Manchester. These two events occur invariantly every day, but neither causes the other. Since much of the science of statistics involves a hidden inference from correlation to causation, we must be eternally aware of the fallacy Russell exposed. Statistical analyses that make this unwarranted inference are bunk: or perhaps, using contemporary nomenclature, we should label them "fake science".
The National Health Service recently released a revealing statistic, namely that 90% of the recorded deaths from coronavirus were persons in hospital with an average of 2.7 "pre-existing conditions". To parse this, we need to understand what the NHS means by the phrase "pre-existing conditions". They means life-threatening illness or other medical condition (advanced cancer, heart attacks, critical illnesses such as advanced AIDS, etcetera). In other words, 90% of the figures recorded for coronavirus deaths are for people who are terminally ill in an average of 2.7 different ways.
It has recently also emerged that for many deaths in hospital of people with coronavirus, the cause of death has not been marked on the death certificate as coronavirus but instead as something else. Given that 90% of hospital deaths for people who have tested positive for coronavirus are people who are multiply terminally ill, this perhaps is not surprising. Nevertheless these deaths are counted in the coronavirus statistics that we are fed with on a daily basis.
Each day, we are given updates for the number of people who (a) have contracted coronavirus; and (b) have died of it. The figures go up only, of course; there is then debate as to whether the rates by which these figures are increasing is itself increasing or decreasing: in other words, we measure the acceleration (or deceleration) of these figures. But measuring acceleration and deceleration can only be revealing if the underlying figures are reliable. And the figures we are given each day (another 500 people died in the past 24 hours; infections are up by another 10,000) are unreliable and indeed worthless. Let me explain why.
The fact that we receive daily updates of both figures (a) and (b) is revealing of what we already know anecdotally to be true: both these figures are data collected by hospitals. Then the data from different hospitals is added up, and you have a daily infection rate and death rate. That is why there is often a delay in obtaining reliable figures from Fridays; the weekend is coming and presumably the data collators are not working on the weekend.
The practice adopted by hospitals is as follows. For all new hospital admissions (whatever the reason), it is routine now to test the patient for coronavirus. Because the symptoms of coronavirus are similar to or overlap with the symptoms for all sorts of other diseases, patients' assertions of symptoms are not used by hospitals as measures of the virus. Therefore all figures for (a) are far too low, because the vast majority of people with coronavirus symptoms (many of whom will have coronavirus, because it is so contagious) never enter hospital. Indeed if you telephone the National Health Service complaining of symptoms of coronavirus, you are instructed not to see a doctor or come into a hospital for seven days, the period by the end of which the symptoms of the disease have typically finished.
In other words, the rate of infection - the true number for figure (a) is vastly higher than any official statistics admit, and we do not know what it is. Some scientists have speculated that it could be 50% or higher of the population - and that was several weeks ago. It could be nearer 100%. We just don't know, but it is very important that we understand just how high it may be.
One a patient is admitted to hospital and tested positive for coronavirus, a record of their other "pre-existing conditions" (i.e. terminal illnesses) is taken: zero or more than zero. If they have terminal illnesses, they are treated for their terminal illnesses, and inevitably a lot of those people die. If they have no terminal illnesses, they are typically sent home. That is obvious from the fact that 90% of the people dying have terminal illnesses. The people who are healthy but for the coronavirus are sent home. Coronavirus is not on its own a life-threatening illness requiring hospitalisation. The author has had it, and it is much milder than seasonable influenza. You can still go to work. You can still take care of yourself. You can still exercise, albeit not as hard as you might do. You need more sleep than usual, but that might be said about modern life in general.
When a person who is kept in hospital because they have coronavirus and life-threatening illnesses, and they die, they are counted as a coronavirus statistic. And this is where Russell's insight comes in. Just because a person infected with coronavirus dies, it does not mean that the coronavirus caused their death. A person with an average of 2.7 life-threatening conditions who enters hospital might be thought highly probable to die anyway. People with multiple life-threatening conditions do die. Death is a typically tragic fact of life. It is common for people with chronic life-threatening conditions to acquire new such conditions over time, as the body gradually degrades in its ability to fight disease and infection. People with these degenerative medical problems often die in hospital. Coronavirus might hasten their deaths, or it might not.
Anecdotal evidence from medical professionals this author has spoken to are varied and inconclusive. But anecdotal evidence, no matter how individually compelling, is insufficient a causal relationship. The data you would need to examine whether there is a causal relationship between contracting coronavirus and dying is to undertake cross-sectional testing samples of the population and compare the mortality rates of people with equivalent (or no) underlying conditions of similar ages and similar levels of health, and compare the mortality of rates of people infected with the virus as opposed to those without the virus, controlling for the other variables that might affect mortality.
In statistics, this is called a regression analysis. Without a regression analysis you cannot investigate causation; only correlation. We currently have no evidence that coronavirus causes death, because we have not been collecting the right evidence that would enable us to test the hypothesis that it kills people. It is no good taking doctors' words for it. Causation is a question of statistical science; it is a relationship that emerges when a change in one factor moves in a predictable pattern to a change in another factor, such that the causative factor is identified important in making the world a better or a worse place. We cannot test whether that is true of coronavirus at the current time, but thinking about the contents of this essay, one may wonder whether there is any strong evidence that coronavirus causes deaths. The statistics we have collected might be taken to indicate that it doesn't; the people who are dying were, for at least the most part, going to die anyway.
If coronavirus infection is only correlated with death (remember that 1% of a country's population - some 680,000 people a year - die anyway), and does not cause it (or does not do so substantially as to justify the damage to society caused by the preventative lockdown measures), then the science being used to justify social isolation measures begins to look absurd. It has about as much logic as the government banning red underwear, because people who wear red underwear when they are admitted to hospital are more likely to die. It is so absurd as to defy belief.
The twentieth century philosopher Karl Popper taught us that no scientific hypothesis has any value unless there is a way of testing it using data, such that one data would result would indicate that the hypothesis may be true and another data result would prove it false. There is no data we are currently collecting that, whatever its content, could go to demonstrate that coronavirus causes death. The net result of all of this is that government is currently basing radically oppressive measures in social, healthcare, economic and law enforcement policy upon a scientific hypothesis that not only is unproven but that no effort has been made to test.
We are in the midst of a collective hysteria, in which nobody is asking the right questions and there is no scientific basis for governments to destroy our societies and economies through forcing us to stay apart from one-another. This is madness. We are in the grip of collective psychosis. When the history books come to be written about the twenty-first century coronavirus pandemic, they will surely focus not upon the lethal nature of the disease (coronavirus is likely to turn out to count as the least lethal of the approximately 20 global pandemics recorded in history) and the effect mass deaths had upon society. Instead historians will focus upon the wanton self-damage citizens permitted their governments to do them in the name of scientific hallucination. For now we may only speculate how much of the economic successes of the late twentieth and early twenty-first centuries we may undo in the name of this fake science. It is surely the ultimate irony of our times, that modern successes in creating free, prosperous societies with broad enjoyment of luxury and consumer goods and widespread material comforts are being undermined by a proliferation of fake news.