In the wake of  Omicron-related restrictions, some commentators have been keen to draw a link between population density and covid rates. The population spread of the UK has been cited as a reason why it is a mistake “to compare coronavirus rates”. And New Zealand, for example, is routinely accused of a comparatively easy ride in their efforts to achieve a ‘net zero’ pandemic due to the sparsity of its residents.

But, this is all a lazy way of interpreting the data. When we compare population density to the severity of the pandemic internationally, an entirely different picture of density and Covid appears. It becomes clear that national population density is meaningless when seeking to understand the average citizen’s proximity to their neighbour

Initially, let’s take this graph above: population density is on the x-axis, and deaths per 1,000,000 is on the y-axis. If you’re struggling to make out a correlation in the data, it’s because none emerges. The result is surprising: it appears a country’s population density has no impact on its Covid outcomes.

It’s worth noting that both Covid and population data are imperfect. Some countries, such as the UK, are reasonably effective at recording and reporting deaths, while nations with effective and funded censuses will be well-placed to understand distributions of their population.

Regardless, we should expect that the data would allude to a pattern on a broader scale if national population density did lead to higher incidence of Covid cases.

But notions of population density on a national level do not reflect the way most people live within countries. Take the four nations which make up the UK as an example. Notionally, the population density of England is 281 people per square kilometre. In reality just 30% of the population lives at or below this density.

High-density cities and conurbations like London, Manchester and the West Midlands house people at densities of up to 13,000 people per square kilometre, but these regions are set against places like Devon, where sheep outnumber people two to one.

Even in Scotland and Wales, while the majority of people live in rural environments, a significant portion of the population lives in major urbanised areas. In empty and rural Scotland (32% of the UK landmass; 8.2% of the UK’s population) just four cities – Glasgow, Edinburgh, Dundee and Aberdeen – make up almost one third of the total population. Cardiff alone represents 12% of Wales’ population; and one in five residents of Northern Ireland lives in Belfast. All of these cities maintain populations at densities greater than 1,600 people per square kilometre.

Still, Wales’ overall population density is just 150, Northern Ireland’s is 133, and Scotland’s is yet sparser at 65 people per square kilometre.

All of these results highlight that population density, read as a national metric, is pointless for drawing conclusions about the average citizen’s proximity to their neighbour.

The disparity between the apparent population densities of these countries and the realities for many of the people living in them is even more striking when mapped.

The central belt of Edinburgh and the region of South Wales carry the bulk of the population, while areas like the highlands of Scotland or Mid Wales are scantily populated.

Vast areas of extremely low population skew overall population densities down, distorting national figures.

These distortions, inhomogeneity in the distribution of a country’s population, make comparisons on a national level impossible. Monaco, for example, is a relatively consistent population of 40,000 in an area smaller than the square mile. As a consequence its population density is 400 times higher than Scotland, despite Scotland claiming multiple cities of a density comparable with Monaco.

To understand better the relationship between our physical proximity and our realistic likelihood of catching Covid then we must take a more granular view of population.

In this graph the picture is much clearer. Comparing cases in the recent omicron wave at a lower level, a pattern emerges between our physical proximity and our likelihood of catching Covid.

National population density is too blunt an instrument to interpret the spread of Covid-19

A study published in the national library of medicine found a similar positive correlation in Sergipe State, in Northeast Brazil. The study also notes that population density correlates strongly with other factors which can also influence the spread of Covid-19, such as socioeconomic inequality.

Pharmaceutical interventions, too, play an important role. But within the UK the distribution of vaccines — and booster jabs in particular — is fairly consistent between nations.

It is perhaps telling then that England, which faced political criticism for failing to implement Covid restrictions from December 27th, seems to broadly follow the trend set by Scotland and Wales, where restrictions were in place, during the Omicron wave.

If there were variations in distribution of pharmaceutical interventions or some other socioeconomic factors which would uniquely promote the transmission of Covid in Scotland and Wales, then these restrictions may have been necessary to align Scotland and Wales’ outcomes with those in England. However, this does not seem to have been the case.

Most apparent is that national population density alone is too blunt an instrument to interpret the impact of measures and responses on the spread of a virus like Covid-19 on a country-wide scale. Only with a more nuanced understanding of how people are distributed within a country can we begin to appreciate the implications of physical distance on stemming the spread.