The 2020 Coronovirus Crisis - Part 2 of 3

Lock-down Policies - Some Comparisons

Initially in March 2020 the UK government appeared to be intending to take what was called a mitigation strategy for COVID-19 relying on building up herd immunity. Opponents of this strategy claimed that recent “modelling by scientists” showed such a strategy would result in a huge number of deaths in the UK. Additionally there would be massive overloading of the NHS caused by the rapid spread of COVID-19 which would result in additional deaths due to lack of  both ICU capacity and of the medical staff and facilities to deal with the normal non-COVID medical care required.

As reported by the BBC on 17th March https://www.bbc.co.uk/news/health-51915302 the modelling projections were :

Do nothing - 81% of people would be infected 510,000 would die from COVID-19 by August.
Mitigation - 250,000 deaths and completely overwhelm intensive care in the NHS.

The experience of Italy's severe problems in dealing with the rapid spread of COVID-19 was used as supporting evidence for the projections - by March 17th the official Italian COVID-19 death toll was 2,185 and rising rapidly by over 300 a day and still rate accelerating. The UK government therefore  decided to change tack and on 23rd of March moved to suppression rather than mitigation and implemented a “lock down” i.e. closing all schools and non-essential businesses, banning all non-essential travel, all public events etc. This was similar to the policy implemented by Italy a couple of weeks earlier, by China in February and being adopted by many, but not all countries world wide. 

It was hoped that the lockdown could reduce the number of UK deaths caused by COVID-19 and “a very good result" would be limiting the deaths to 20,000. As it has turned out the impact of COVID-19 on the UK compared to on Italy has been worse, both in terms of deaths per million and total UK deaths now reported as over 42,000.

I should point out that for the comparisons I'm using,despite their limitations covered in Part 1, the figures for officially reported COVID-19 deaths. I consider that as they are consistent with the available excess mortality statistics they are suitable for drawing relative comparisons i.e countries with higher EM also have higher reported COVID-19 deaths. They also provide a more detailed picture of how events have unfolded over time (and also the data on excess mortality is simply not readily available for one of the countries I compare the UK  with).  

Note:    All graphs shown are from the excellent Oxford University Site, "Our World in Data", where you can select the specif data you want to plot.  Example as per link. 

https://ourworldindata.org/grapher/daily-covid-deaths-per-million-7-day-average?country=BLR~ITA~SWE~GBR

 COVID-19 Reported Deaths per Million - UK vs. Italy

However some EU countries have had vastly lower levels of deaths than those experienced in the most severely affected countries i.e UK, Italy, Spain, France, Belgium and the Netherlands. In particular Greece has fared remarkably well as shown below :

 COVID-19 Reported Deaths per Million - UK vs. Greece 

So why the huge difference? A possible often cited and certainly, given experience on how previous epidemics have been controlled, credible explanation is that it was because Greece had implemented a lock-down faster than the UK and Italy. Greece closed schools as soon as the 3rd death in Greece from COVID-19 was confirmed, non-essential shops the day after and banned non-essential movement 9 days later. The UK took two weeks from the 3rd confirmed UK death before banning non-essential movement and closing schools. For details of the various lock downs and when they were implemented see here:

https://www.politico.eu/article/europes-coronavirus-lockdown-measures-compared/

However, some European countries did not implement a lock-down. Sweden in particular came and still comes under heavy criticism for not doing so and it was widely claimed they were recklessly ignoring "the science". A comparison with the UK however shows that the outcome to date in terms of deaths per million has been lower in Sweden than the UK .

 COVID-19 Reported Deaths per Million - UK vs. Sweden

Even more surprising is a comparison with Belarus with a population of 9.45 Million similar to Sweden. Belarus did not implement any form of lockdown and has received wider and harsher media criticism and claims of virus "denialism". As it has turned out the reported situation in Belarus is actually much more comparable to that of Greece and vastly better than the UK and Sweden

 COVID-19 Reported Deaths per Million - UK vs. Belarus

Comparison with Modelling

If we simply scale, to allow for their smaller populations, the early March UK modelling projections of UK COVID-19 deaths to the strategies applied by Sweden and Belarus we get.

Projected Deaths - note 1	Sweden	Belarus	UK
Do Nothing	n/a	69,690	500,000
Mitigation only -note 2	37,000	n/a	250,000
Suppression (lock down)	n/a	n/a	20,000 - note 3
Total reported to date	5,053	331	42,000

note 1 The modelling projections always acknowledged they were based on a number of uncertainties
note 2 The Swedish strategy implemented was not as severe as the UK proposed mitigation strategy 
note 3 As above the 20,000 figure was essentially only ever a hope

We will never know what would have happened in UK if we had not implemented or only implemented a less severe lockdown strategy.

If we reverse the logic and scale the Belarus experience to the UK then doing nothing hypothetically should have resulted in only 2375 UK deaths. Despite my significant misgivings about the accuracy of officially reported COVID-19 deaths (see Part 1) a death toll of only 2375 in no way aligns with the reality of the huge increase in UK excess mortality that has occurred and there is no credible argument available to explain how the suppression tactics adopted in the UK could have resulted in a much higher COVID-19 death toll relative to doing nothing  

However see Part 3 for further discussion on this point since excess mortality that is not directly due to COVID has increased substantially - 12,900 above the the 5 year UK average by 1 May 2020 - as reported and discussed in the official UK statistics published by the ONS where it is clear, albeit not stated as such, that this increase may be due to both direct and indirect consequences of the "lock down" itself rather than COVID-19.  

https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/articles/analysisofdeathregistrationsnotinvolvingcoronaviruscovid19englandandwales28december2019to1may2020/technicalannex

Conclusions 

The vastly differing experiences of countries in dealing with COVID-19 are still not properly understood and no convincing (see Part 3) explanations have been put forward as to why the lack/limited level of lock-downs in Belarus and Sweden did not result in a much higher death toll. 

note:  I hope it goes without saying that i don't consider it credible that it's due to "drinking vodka having saunas and driving tractors"  https://www.politico.eu/article/belarus-lukashenko-is-defying-the-coronavirus-and-putin/

We still don’t properly understand the new virus nor how COVID-19 will affect a given population.

Part 3  .....

The 2020 Coronovirus Crisis - Part 1 of 3

Why this post?

This set of three posts aim to give a different and more balanced perspective to that being presented in the mainstream media on the current crisis resulting from the the emergence of the new virus, SARS-CoV-2 and the new disease it causes, COVID-19. At the time of writing, June 2020, there are still major gaps in our understanding of how the virus spreads - how many people have been infected by it and how many have died from COVID-19. The various policies implemented by Governments around the world to deal with the impact of this new virus were and continue to be made in a fog of uncertainty and confusion resulting in some very poor decision making that has and will continue to have grave and lasting consequences for huge numbers of people.

You don't know what you don't know

Attempting to understand what's happening using any analysis based solely on official published numbers of COVID-19 related deaths or cases is beset with difficulties since where a death has been recorded as a COVID-19 death i.e. the person who died had tested positive it is uncertain whether COVID-19 was:

• the primary cause of death
• a contributory factor
• not a factor

Also the limited testing and varying approaches to reporting of COVID-19 deaths across countries means we don't know how many people actually have or have had COVID-19, nor how many deaths that were due or partially due to COVID-19 have not been accounted for. However, there is a well understood and established measurement that is not influenced by the uncertainty associated with published COVID-19 statistics. That measure is Excess Mortality (EM) which is simply the number of recorded deaths above the average for a particular time of year. This measure has been gathered by an independent European body of epidemiologists for many across 20 European Union countries. It is also monitored at a UK level as published by the Office of National Statistics (ONS). The statistics and details of the 20 countries involved and methods used are freely available at :

https://www.euromomo.eu/graphs-and-maps

Excess Mortality - Overall Statistics across twenty European Countries 

The graph below how shows how excess mortality EM varies year to year and week to week due primarily to seasonal flu and excess winter mortality i.e. more people die in winter than summer. The impact of COVID-19 is shown by the sharp spike in EM starting March.

Excess Mortality - 2016 to June 2020

 
You can see from the graph that currently (week 23 - June) there is now NO significant overall excess mortality albeit there still is in some individual countries notably England.  So, to get these numbers into context we need to consider the total number of excess deaths in 2020 to date, the cumulative EM, relative to the total population of the twenty countries involved . This 2020 cumulative excess mortality up till week 23 June is 194,284. 

Annual Cumulative Excess Mortality 2018,2019,2020 

The population of the twenty countries is 435 Million. So, making for the moment the incorrect assumption (see later posts) that COVID-19 was the sole cause of all excess mortality in 2020:

EM (all ages)           = 194,284

Total Population      = 435,010,967

As a percentage 

        Excess deaths             = 194,284 / 435,010,967 
                                           = 0.045% or  1 in 2239

However, COVID-19 affects elderly people much more severely than younger people with 91% of all excess deaths being in the 65 and over age group. Breaking down the figures for excess mortality by age group, in percentage terms they are:

Age 0 to 14   = 0.00001% or 1 in 7,131,327

Age 15 to 64 = 0.00399% or 1 in 25,063

Over 65         = 0.04046% or 1 in 2,472

A comparison that might help to get these numbers into context is to compare these figures with the risk of dying in a car accident. Assuming you travel in a car regularly your annual risk in the UK is roughly 1 in 25,000 or over a lifetime 1 in 500.

Comparison with Seasonal Flu / Previous years

The cumulative EM graph for the twenty EU countries shows clearly the higher number of excess deaths in 2020 than in 2018 and 2019. So again, making for now the (false) assumption that all 2020 excess mortality is due solely to COVID-19 then in percentage terms the increase in EM across the 20 countries that can be attributed to COVID-19 is at most. 

85% above 2018  and 239% above 2019

These are of course very significant increases.

 At a world wide level however the total COVID death figures published on Worldometer are approximately 450,000 (still rising). By comparison the number of deaths worldwide due to seasonal flu is typically as described below: 

“According to new estimates published today, between 291,000 and 646,000 people worldwide die from seasonal influenza-related respiratory illnesses each year, higher than a previous estimate of 250,000 to 500,000 and based on a robust, multinational survey.”  

https://www.cdc.gov/media/releases/2017/p1213-flu-death-estimate.html 

So at a worldwide level the current figure of 450,000 deaths officially recorded as COVID-19 deaths, albeit still rising, is pretty much in the middle of the range. So in terms of number of deaths worldwide COVID-19 so far is directly comparable to seasonal flu.

Conclusions /Comment

See COVID-19 Parts 2 and 3