Covid-19: UK Nations Compared

per 100,000 population and smoothed to a seven day running average.

England Scotland Wales Northern Ireland
🧍 55.98 million 🧍 5.454 million 🧍 3.136 million 🧍 1.885 million
Conservative logo
Conservative and Unionist Party
(80 seat majority UK)
SNP logo
Scottish National Party
(minority admin)
Labour logo
Labour-led
Administration
DUP logo
It's complicated*
NHS England logo
NHS England
NHS Scotland logo
NHS Scotland
NHS Wales logo
NHS Wales
Health and Social Care Northern Ireland logo
Health and Social Care
Northern Ireland

* I used the DUP logo for NI because Arlene Foster is First Minister but the power sharing agreements mean this is an over simplification. It was a pain choosing a colour for NI without offending anyone. Interesting politics.

Motivation

This web page was inspired by a BBC Newsnight presentation of the same data recently that was reported in this article in The National newspaper. There are loads Covid dashboards out there that allow you to visualize data in different ways and the media have been reporting graphs of every shape and size since the pandemic began. But I have been frustrated by how rarely there is a simple comparison made of the UK nations corrected for population percentage.

In many respects that might influence a pandemic the nations are very similar: culture, climate, mean age, level of urbanization (Scotland is as urbanized of England despite what you might see as a tourist) and macroeconomics. But they are different in two key ways. They are each run by very different political parties with significant local power and have their own quite separate health services. (NHS is just an umbrella term for four separate systems which sometimes pool resources). The nations differ in the key things that are important for managing a pandemic. It is therefore very useful to compare the four home nations as they form a natural experiment. Most variables are held constant. Only the mechanisms of delivery of pandemic response on the ground differ. Importantly the nations have agreed on some common way of keeping count which gives rise to the data presented here.

There are reasons you won't see these comparisons in the media very often. Some of them are good. Here are a few.

Articles like Why Scotland isn't faring much better than England despite Nicola Sturgeon's 'zero-Covid' approach are written by partisan sources and they have to work pretty hard at it. Their efforts date quickly. Another example is Scotland’s Covid death rate prompts questions over Sturgeon’s strategy. I've not seen articles about why England is doing so badly compared to the other nations of the UK but looking at the actual data that is definitely the question that springs to mind.

An article in the British Medical Journal Why Scotland’s slow and steady approach to covid-19 is working was supportive of Scotland's approach. But the BMJ doesn't get the readership of the Telegraph or FT. One of the two responses to the article (from a named, working GP in Glasgow) says: "This article seems to be part of a nationalist attempt to use the pandemic to differentiate Scotland from the rest of the UK and increase support for separatism." So presenting a comparison of data between devolved regions of the UK must be approaching sedition! There is nothing so dangerous as unmediated facts. Also of interest is that both responses to that article say Scotland had one of the highest per capita death rates in the world but fail to point out it was lower than the UK average.

Note that deaths by actual date of death are most accurate but not for the last two weeks as it can take quite a while for registrations of deaths to make it through the pipeline, especially at holiday periods.

Data & Tech

All the data here comes from the GOV.UK official data site apart from the population numbers that I got by Googling (they seem reasonable). I do little manipulation. This is not an analysis but a presentation of the data. The data is pulled fresh from gov.uk every time the page is loaded. This means it is always up to date but if they go down or change their API this page will stop working. The code consists of this single page written in HTML & Javascript with the ChartJS library. I wrote it in a Saturday evening (rather than watch a movie with the family) and Sunday morning (before they got up). If you want a copy of the code just do "view source". I'd welcome constructive criticism by email: Roger Hyam. I don't do social media.