COVID-19: Europe Report, Omnibus Edition

In recent weeks, I have been developing the “magic spreadsheets” which help me to follow the statistics of the COVID epidemic, with the aim of significantly increasing the number of countries I am able to look at. This is the first report based on the new technology. It covers the whole of Europe, a total of 46 countries divided into four groups. Here are the groups:

Europe 14Rest of Western EuropeEastern Europe (North)Eastern Europe (South)
BelgiumFinlandCzechiaBosnia and Herzegovina
LuxembourgSan MarinoPolandMontenegro
NetherlandsVaticanRomaniaNorth Macedonia
Portugal RussiaSerbia
Spain SlovakiaSlovenia
Sweden Ukraine 

I’ll end this essay with an assessment of the UK’s performance against the virus to date. I think it’s fair to say that to call my assessment “scathing” would be an understatement.

Looking ahead, I have divided the 189 countries which have reported COVID cases into a total of 20 groups, which I then aggregate together into six “supergroups” as follows:

  1. Europe (Europe 14, Rest of Western Europe, Eastern Europe (North), Eastern Europe (South)).
  2. Americas (North America Mainland, South America Mainland, West Indies (North), West Indies (South)).
  3. Middle East/North Africa (Middle East North, Middle East South, North Africa).
  4. Sub-Saharan Africa (West Africa, Central Africa, East Africa, Southern Africa).
  5. Rest of Asia (North East Asia, East Asia, South Asia, South East Asia).
  6. Australasia and Oceania.

Once again, the data sources are (for epidemic data) Our World in Data and (for lockdown regulations) the Blavatnik School of Government, both at Oxford University. The data I used included figures up to and including February 19th.

It’s worth noting that there are a number of places for which I cannot show any data. This is because the Our World in Data feed, which I use, now excludes dependencies, such as Gibraltar and the Faeroe Islands. (I am not certain whether or not their statistics will have been folded in to the parent country’s data.) This is a pity, because Gibraltar has the very worst record in the world in deaths per million, and the Faeroes one of the very best!


This report introduces some scatterplots. These are able to plot any of ten columns of country data (Hospital beds per 1,000, ICU beds per 100,000, Cases per Million, Deaths per Million, Deaths per Case %, Tests per 100,000, Cases per Test %, People Vaccinated %, People Fully Vaccinated % and Average Lockdown Stringency %) against an index column. The two index columns I have chosen to use are:

  1. The UN’s Human Development Index (HDI) percentage rating.
  2. Population densities (in people per square kilometre).

According to Wikipedia: “The United Nations Development Programme (UNDP) compiles the Human Development Index (HDI) of 189 countries in the annual Human Development Report. The index considers the health, education and income in the country to provide a measure of human development which is comparable between countries and over time.”

The countries

I thought I’d start with some bar charts of how the different European countries measure up on the UN’s HDI rating. I’ll also show their populations per square kilometre, as I’m wondering if population density may perhaps be a factor in the transmission of the disease.

Now, here are the population densities:

For population density, Monaco dwarfs the rest. Indeed, the second, third and fourth in its group, Malta, the Vatican and San Marino, are all more densely populated than the Netherlands which comes in fifth! The “density divide” between Western and Eastern Europe is also apparent.


To the dynamic data. As usual, I’ll start with cases. Here are the spaghetti graphs of total cases per million for each of the four groups:

What comes through here is the contrast between Western and Eastern Europe. In Western Europe, there was a clear first wave, which by the summer had (somehow) been controlled. Then a second wave began in Luxembourg in July and Spain in August. Initially, it spread slowly, with most starting to feel its effects in late September or October. In Eastern Europe, on the other hand, with the (surprising?) exception of Belarus, the first wave was never controlled. Despite a few peaks and troughs, they are all in effect still in the first wave.

After that, the virus in each country seems to be proceeding at more or less its own pace. This can be seen even more clearly by looking at the graphs of daily new cases per million:

It isn’t just Belarus that shows up differently from the others here. Moldova, for example, shows five or perhaps even six peaks, and is on the way to another one:

That suggests that the dynamics of this virus are a whole lot more complicated than just “first wave” and “second wave.” It will be interesting to see what the rest of the world has to show.

Let’s have a look at a scatterplot of cumulative cases per million against HDI rating:

The countries with the very highest HDIs – Finland, Norway, Iceland – have relatively low case counts. Maybe in these places it’s more a matter of population density? But these countries apart, I was expecting a larger positive trend, on the grounds that countries with higher HDIs will tend to have more international travel, and so be more vulnerable to re-seeding of the virus.

Let’s have a look at cases per million versus population density. I capped the densities at 1500 people per square kilometre, so Monaco and Malta are represented by those two points very close together at the far right:

This looks almost like two disparate sets of data. There is a low density set on the left, of countries with less than about 200 people per square kilometre. To which set, I might be tempted to fit a trend line from the origin towards that point at the top (Andorra). Then there is a higher density set of countries, in which the cases per million don’t seem to depend much if at all on the population density.

It looks as if, above a certain population density, the national density is not a big factor in case numbers; other factors like testing régime and containment policies are more important. Even though observations, both from my own local area and from the Netherlands where I used to live more than 40 years ago, suggest that local population density can significantly spur case activity. In particular, high-rise living is not conducive to avoiding COVID.

Lastly for cases, here is a list of all the countries, ordered by cumulative cases per million:

A high number of cases per million isn’t necessarily either a good or a bad thing. It may represent a strong program of testing (as in Luxembourg), or alternatively a high rate of transmission (as in Andorra). A low number, on the other hand, means (probably) that the virus has been contained, for now. But that doesn’t necessarily mean that it will stay contained – as witness what happened in Belgium in October.

Case Growth and Lockdowns

I don’t usually show the graphs of weekly case growth right from the beginning of the epidemic. This is because much of the data is highly confused. The clear-sighted may find themselves dazzled by bright spaghetti, and the colour-blind will have difficulty picking out any salient features at all. But I’ll break my rule on this occasion, because of the interesting things that happen towards the right of the graphs:

The Eastern European graphs show a decline in the size of the peaks of weekly case growth, starting in October or November. The Western European graphs would show something similar, if you took out the wild excursions in Ireland, Spain and Monaco. What could have caused this? Lockdowns? Let’s look at the record:

These graphs show the series of “copycat” lockdowns imposed across Europe around late October and early November. But in Eastern Europe particularly, the decline in the peaks of weekly case growth had already started prior to that time. In my thinking, at least, the jury is still out on that one. I think lockdown efficacy can only be addressed on a local basis.

Meanwhile, we can look at the R-rate, the number of infections each infected person passes on to others. While this does give a far clearer picture than weekly case growth, I’m reluctant to set too much store by R-rates, as they are modelled, not based directly on measured data. And they are modelled in different ways by different countries, as shown by some (for example the UK) being very jagged, and others (for example Sweden) far smoother.

Again, that looks inconclusive to me.

Now for a histogram the UK politicians won’t like. I’ve added to my spreadsheets a calculation of the average level of lockdown in each country (on a per-day basis), since the beginning of the epidemic:

You can see most of the expected suspects near the top! Ireland, in particular, has been trigger-happy on the lockdowns all along. But at this point, I’m more interested in the bottom. Finland is dead last in Europe in cumulative cases per million, Belarus is seventh from bottom, and even Estonia is in the bottom half. In the cases of Finland and Estonia, it looks as if the first wave was controlled without much of a harsh lockdown, but the second wave is proving larger. Belarus, I’ll take with a pinch of salt; but it looks from the Blavatnik data that the only mandatory lockdown measure they have taken is complete closure of the borders since October 29th. If only all of them had done that back last March… Sigh.


Here, I’ll simply show the graphs of tests per hundred thousand for each of the countries. You’ll see that some countries set great store by testing (such as Luxembourg, Denmark, Slovakia and Cyprus), and others do not (such as the Netherlands, Ukraine and Albania).


The further you get from the core of Western Europe, the patchier the hospital data gets. Here, first, are the numbers of hospital beds per million for each of the countries:

Here are the numbers of hospital beds (per thousand, this time) plotted against the UN’s HDI rating:

That downward trend is interesting. In Europe at least, the more developed a country is, the less hospital beds it tends to have, relative to its population!

Here are the numbers of hospitalized COVID patients per million in each of the countries:

Again, the difference between the epidemic patterns in Western and Eastern Europe becomes obvious.

Here are the percentages of hospital beds currently occupied by COVID patients in the Europe 14 region (blank bars are those countries that do not report data):

In the rest of Western Europe, only Iceland, Finland and Norway report this data, at 0.8%, 0.6% and 0.4% respectively. Here’s the data for Eastern Europe:

So, the six worst affected countries in terms of hospital bed occupancy by COVID patients are Spain, Portugal, the UK, Slovakia, Italy and Latvia.

Intensive Care

If hospital data is patchy away from the centre of Europe, data on Intensive Care Units (ICUs) is even more so. Here are the numbers of ICU beds per million in each country:

This time, the trend between HDI rating and number of ICU beds goes the other way:

Here are the numbers of COVID patients occupying ICU beds, per million population:

As to the occupancy percentage of ICU beds by COVID patients, here are the numbers from the Europe 14:

The Portuguese must have been doing some emergency hospital building!

In the rest of Western Europe, only Finland is reporting, and the figure is just over 5%. In Eastern Europe, Czechia is reporting 93%, Estonia 19% and Slovenia a staggering 104%. That’s all the data I have on ICU occupancy. So, the worst hit ICUs right now are in Portugal, Slovenia, Spain, Czechia and the UK, in that order.


To the final curtain. Total deaths per million in each region are as follows:

It looks very much as if there is a “race to the bottom” going on here. The UK is closing in on the Belgian world record holder (among countries with populations bigger than a few tens of thousands). Slovenia and Czechia are catching him up, too. And watch out for the “dark horses” of North Macedonia and, coming up on the wide outside, Slovakia.

Here’s the league table of deaths per million among my 46 European countries. UK politicians may wish to look away at this point:

I won’t bother to include the daily deaths per million graphs, because they look very much like the cases per million graphs, just displaced to the right by three weeks or so. Thus, it’s time for the scatterplot of deaths per million against HDI rating:

That’s encouraging. While European countries with higher HDIs tend to get more cases per million, they also tend to get less deaths per million. There must be at least something good about this thing called “development,” of which the UN speaks.

So, I’ll pass to deaths per case. Now, if there is one COVID metric on which to judge a country’s health care system, this one is it. Poor testing, poor hospital care, insufficient (or poor) intensive care when needed; all these will increase this metric. I’ll leave aside all considerations of deaths per case at different stages of the epidemic, and simply go for the jugular. That is, deaths per confirmed case, as a percentage, over the whole course of the epidemic. As a European league table. I can almost hear the UK politicians saying “Ouch!”

I’m nearly done now; but I see that I haven’t yet discussed an important subject. That is…


After the final curtain, the encore. I left vaccinations until last, because they are new on the scene, and there is no way as yet to tell by observation whether or not they are having an effect. Another month, hopefully, will tell.

I will not graph the total vaccinations; which, to me, is a virtue signalling number rather than a significant statistic. Rather, I decided to show “people vaccinated” (one or two jabs) and “people fully vaccinated” (two jabs). As with hospitalizations and ICUs, the data becomes a bit patchy once you go beyond the core of Western Europe.

The UK, Malta and Serbia (and perhaps Latvia) seem to have gone out of the blocks like sprinters. So, let’s look at the numbers of people fully vaccinated with both jabs:

So, it looks as if the UK and Latvia have taken the strategy of getting the first jab out to as many as possible as soon as possible. Whereas everyone else is concentrating on getting as many as possible as fully immune as possible. This latter strategy, I think, is saner; for it gives the best protection as quickly as possible to the most vulnerable.

Here’s the league table of full vaccinations:

Amazing! The UK is third from last, among those countries that report data, in delivering the second dose! Latvia is only one place higher. Oh, and Denmark is up near the top. Again.

How well has the UK done?

As an Englishman, currently resident in England, I feel a need to assess the performance of the UK over the COVID virus so far.

  • Fourth out of 45 European countries in deaths per million from COVID. Second only to Belgium among countries with comparable populations (over about 10 million). And on the way to catching up.
  • Seventh out of these 45 countries in cumulative deaths per confirmed case. Second only to Italy among countries with comparable populations.
  • Seventh out of 42 European countries in average level of lockdown through the epidemic. Second only to Italy among countries with comparable populations.
  • Only 16th out of 46 European countries in cases per million, even though the UK is one of just seven of those countries which have done more tests than their populations.
  • Third in Europe in current percentage of hospital beds occupied by COVID patients.
  • Fifth in Europe in current percentage of ICU beds occupied by COVID patients.
  • While the UK is top out of 33 European countries in people vaccinated per million, it is third from bottom in people who have had both vaccinations.

And that doesn’t count things which don’t show up on the graphs. Like health secretary Matt Hancock misrepresenting the results of new cases versus tests back in May, on which he was caught out by Sir David Norgrove, chairman of the UK Statistics Authority. The political skulduggery and persistent alarmism of SAGE, the (supposedly) advisory group whose remit is to provide “scientific and technical advice to support government decision makers during emergencies,” but which seems to have made itself into the driver of government policy on COVID. Leading to the fiasco of the (well-intentioned) “tiered lockdown” system being abandoned after only a few weeks in operation, and replaced by a harsh national lockdown, which SAGE seem to have every intention of keeping going as long as they can. And beyond even these, the government have granted immunity (as opposed to indemnity) to vaccine manufacturers against prosecution for harmful side-effects of their vaccines.

As the Guardian reports (, the “road map” out of this fiasco looks like a Churchillian combination of “blood, toil, tears and sweat.” Small shops still to be closed until some time in April. The middle of May, at least, before hairdressers can re-open and those of us with beards can get them trimmed again. And there isn’t even a date given for gatherings like my brass band rehearsing (which we could do back in October, while still in Tier 1).

We want our money back! Along with our social lives, our economy, and our rights and freedoms. Overall Rating: F-.

Oh, there is talk of “reforming the NHS.” But I say, reforms will not be enough. A revolution is needed. The NHS in its present, politicized form must be scrapped, and a proper health care system built in its place. The sacred cow has reached the end of its useful life.

The problems are not with the doctors and nurses out in the field. The problems are with the bureaucrats and politicians, and the academics and others that advise them. We need to de-politicize health care. To separate its financing and its provision; perhaps on the German model, or something like it. To hire some trustworthy public health advisors from countries which have done relatively well against the virus, like Denmark. And to sack SAGE en bloc.

2 thoughts on “COVID-19: Europe Report, Omnibus Edition

  1. Pingback: Neil Lock’s Analysis of the UK’s Covid strategy! – Opher's World

Leave a Reply

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

You are commenting using your 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