Lockdowns Have Negligible Effect on Covid Deaths

The coronavirus plannedemic, part of a dastardly plan for a police state, relied upon the false claim that various restrictions on freedoms were necessary to save hundreds of thousands or even millions of people from dying from the virus. In fact, the daily growth in deaths in a Covid epidemic always approximates a Gompertz function. Although various forms of lockdown certainly have devastating effect on the economy, on people's mental health, on non-Covid illnesses and fatalities, on quality of life and general wellbeing, etc., they have minimal impact on the problem they are supposed to solve.

Proponents of lockdown policies were already refuted by data from non-lockdown nations like Sweden and Belarus, showing that the doom-mongers' claims never materialised, and yet these nations were not silly enough to destroy their economies and cause thousands of excess non-Covid deaths because of mass media-driven mass hysteria over a flu-like virus. As of June 4, 2020, Sweden had 452 Covid-19 deaths per million of population and Belarus had 26 (twenty-six), which compares with 589 in lockdown UK.

Appendix A shows the daily accumulated total of Covid-19 deaths for the UK, along with the percentage daily increase on the previous day, and the percentage daily increase averaged over the previous seven days. (That's calculated from the 7th root of the ratio of current day's total deaths to total deaths seven days ago.)

The 7-day averaged growth smooths out some noise in the data, such as the smaller increases on a Sunday and Monday when fewer deaths are reported over the weekend. Or such as the extra 445 deaths slipped into the total on Monday June 1 that had occurred over the previous five weeks, where 'proof' of Covid-19 had supposedly been performed at private labs.

The scaremongers tried to imply that if measures were not taken, then total infections, followed by deaths, would increase exponentially up to frightening numbers. However, the virus never manages an exponential growth, even towards the start of an outbreak. The UK figures have a peak of almost 61% average daily increase in deaths for the seven days up to day 10, March 14 as the total jumps up to 28, and then this rate declines.

Here we consider the number of deaths rather than infections. The total infections would be expected to follow a similar curve to the total deaths, albeit displaced in time. But infections data is dependent on number of tests conducted, and therefore less reliable.

The UK's lockdown was introduced on March 24, day 20. However, any immediate effect of this could only have been upon the number of fresh infections. Reported infections would lag this, and deaths would considerably lag reported infections.

For the coronavirus "Covid-19", the median duration from exposure to symptoms onset (incubation period) is 5.1 days, and the median time from illness onset to death is 18.5 days. Thus, there is a lag of about 24 days between the introduction of a "social distancing" or lockdown policy and any possible effect upon the daily death rate, and the UK's lockdown measures commencing on Tuesday March 24 should have had their greatest effect around Friday April 17.

It's actually a little longer than that, since there is also the time taken for the deaths to be reported.

Yet the peak of 60.97% daily or a 28-fold increase over the previous week declines after March 14, and even if that is discounted as noise in the data, then in the first few days of April, within one or two weeks of introduction of the lockdown, there is already a clear slowing of the increase in fatalities. The 7-day averaged daily growth is already down from well over 30% to 20% and below. And it continues to decline over the next weeks, with no pronounced effect apparent around April 17.

Curve fitting of epidemics is explained by the famous scientist and Nobel laureate Professor Michael Levitt who goes on to show how the infections data from South Korea and New Zealand fits the Gompertz function.

The function is:

G(t) = N exp (-exp (-K(t-T) ) )

For the UK cumulative fatalities data we use N = 42,500, K = 0.05509 and T = 46. (t is the x-axis, the day number starting with day 1 on March 5 when total deaths start at one.) So that looks like this:

(See below to see how the raw data fits the Gompertz function.)

So, note how day 20 is the Tuesday when the UK lockdown came into effect, and day 44 or later is when any effect of the lockdown policy on brand new infections would typically be translated into fatalities. Thus, about half of the effect would come in a little before day 44, and about half would show up over the following days.

The derivative of the Gompertz function is:

d/dt [G(t)] = K N exp (-exp (-K(t-T) ) - K(t-T) )

When we add this to the above curve and focus in on the early part for clarity, this is what we find:

Suppose you have exponential growth of something like 30% daily. The derivative of the function would also increase exponentially, in line with the function, except that it's also multiplied by ln (multiplying factor). That's ln(1.3) ~0.26236 in our example. But here the derivative (in green) is always increasing much less rapidly than the function, even well before day 20, let alone day 44.

An interesting feature of the Gompertz function is that, when displayed on a log scale, the derivative is a downward-sloping straight line.

To show our UK fatalities Gompertz function on a (natural) log scale, we take:

 ln(G(t)) = ln(N exp (-exp (-K(t-T) ) ) ) = ln(42500 exp (-exp (-0.05509(t-46) ) ) )

The derivative of that is:

d/dt [ln(G(t))] = d/dt [ln(N exp (-exp (-K(t-T) ) ) ) ]

= K exp(-K(t-T)) = -0.05509 exp(0.05509(t-46) )

Now show this on a log scale which nicely simplifies the expression:

ln(d/dt [ln(G(t))]) = ln(K exp(-K(t-T))) = ln(K) - K(t-T)

= ln(0.05509) - 0.05509(t-46) = 2.5341 -2.8988 - 0.05509t

= -0.36466 - 0.05509t

...and we have:

A comparison of the actual data with the Gompertz function (Appendix B) shows that the data matches the function within three or four days. For example, on day 20 the actual data is about one day behind the function. It overtakes it on day 28, and by day 44 is about three-and-a-half days ahead. At day 59 it is still a little more than three days ahead of the function, and this has included a couple of weeks when the lockdown policy should certainly have been curtailing deaths if the lockdown advocates were correct.

It takes until day 82 before the actual data has declined to just below the Gompertz function, and then it follows it quite closely.

Now compare with Sweden, which as of early June 2020 has lower Covid fatalities per capita compared to the UK, yet has no lockdown and has a voluntary "social distancing" of 1m rather than 2m. Sweden's total fatalities follows a similar Gompertz function to the UK's:

For Sweden we use the same function:

G(t) = N exp (-exp (-K(t-T) ) )

...where N = 5,400, K =  0.045 and T = 48.

The derivative is K N exp (-exp (-K(t-T) ) - K(t-T) )

...and when added in we obtain:

Now we display the Swedish data on a log scale: N = 5,400, K =  0.045 and T = 48

ln(G(t)) = ln(N exp (-exp (-K(t-T) ) ) ) = ln(5400 exp (-exp (-0.045(t-48) ) ) )

The derivative of that is:

d/dt [ln(G(t))] = d/dt [ln(N exp (-exp (-K(t-T) ) ) ) ]

= K exp(-K(t-T)) = -0.045 exp(0.045(t-48) )

Now also show this on a log scale which nicely simplifies the expression:

ln(d/dt [ln(G(t))]) = ln(K exp(-K(t-T))) = ln(K) - K(t-T)

ln(0.045 exp(-0.045(t-48))) = ln(0.045) + (-0.045(t-48))

= -3.1011 + -0.045t + 2.16 = -0.9411 - 0.045t

For the Swedish data, see Appendix D. A comparison of the actual fatality figures with the Gompertz function shows that the series fits to within a few days, as per the corresponding results for the UK.

Now let's compare the raw data with the Gompertz functions, for both UK and Sweden.

The lockdown advocate, desperately clutching at straws, might try to suggest that the UK curve was speeding up a little ahead of the Gompertz function, before the effects of the lockdown became apparent. But non-lockdown Sweden follows the very same pattern!

And a comparison of UK growth vs. the Gompertz function and exponential growth of just 40% daily (which compares with the 61% daily growth seen over the week to March 14, day 10 in the series) shows how the UK data is far from exponential long before April 17, day 44, when any infection-reducing effects from the lockdown policy would have been starting to show up as reduced fatalities.

Now let's do a direct comparison of lockdown UK vs. non-lockdown Sweden. The chart below shows Covid-19 fatalities per million of population, and the series ends at day 92, June 4, 2020, when the UK figure was at 589 and Sweden was at 452.

In conclusion, the growth in fatalities approximates a Gompertz function, is never anywhere near exponential, and any effect of lockdown policies are minimal, to say the least. Intuition suggests a lockdown might "flatten the curve" somewhat, but there is no evidence for that, and it would be merely to the extent that deaths are postponed for about three days after a period of more than a month, during which time tremendous collateral devastation is wrought upon society.

Given that medical staff had ample spare time to make cringeworthy dance videos, there was never any risk of hospitals becoming overwhelmed. Even if a couple of thousand Covid deaths were postponed for three or four days, and the evidence suggests otherwise, this does not justify the twelve thousand excess non-Covid deaths over the first five weeks of the lockdown in the UK (see Appendix C), and all the other suffering inflicted from a disastrous policy.

Appendix A

UK Accumulated Covid Fatalities data

Appendix B

Gompertz Curve Fit vs. Actual Fatalities Data, UK

Appendix C

Week 13 is the week ending Friday March 27. The UK's lockdown began on Tuesday March 24, so its full effect does not show up until at least week 14. These are figures for England and Wales only.

Source: Office for National Statistics (ONS)

Appendix D

Sweden Accumulated Covid Fatalities data