Tracking Coronavirus at a subnational level

The SARS-COV-2 virus has affected billions of people globally either directly, through testing, quarantining, cancelled meetings and conferences, production and supply difficulties and a run on various products. The economic effects of the event will likely be above $1 trillion given the large economic implications in China, USA, and production activity globally.

It is important however to look at the scale of the outbreak and the number of affected countries. As of today, 106 countries had at least 1 infection with China, Iran, Italy and Korea being those most affected. With over 4000 dead, the absolute death toll is worse than the 2012 MERS-CoV (Middle East respiratory syndrome coronavirus) with 912 deaths (at 37% lethality) and the 2002-2003 SARS (Severe Acute Respiratory Syndrome) with 774 deaths (at 9.6% lethality). However, SARS-COV-2 (coronavirus) currently has a lethality in the order of 2% once detected, however this will likely decrease once more mild detections are now being found. 116,000 infections have been found globally, with 65,000 of these having already recovered.

The highest rates of infections per day are now in Iran and Italy (upwards of 1000 infections per day) versus China (around 50 infections per day).

http://www.risklayer.com/media/blog/img/World_Subnational Infected 09032020.png

What is interesting however is that 550+ provinces have been affected. Of these only three have over 0.1% (1 in 1000 people) of the population infected (Daegu in Korea, San Marino and Vatican City). The provinces differ markedly however with some of the worst affected also having very large populations. The data has been sourced from only official websites of public health ministries, subnational government entities, centers for disease outbreak prevention and national government ministries. 

The following are the top provinces across the globe with more than 1 in 10,000 people being infected:

Province Country Infected Cumulative Dead Recovered Relative Infected
Daegu South Korea 5533 5663 38 92 0,236%
  San Marino 49 51 2 0 0,144%
  Vatican City 1 1 0 0 0,125%
Qom Iran 673 751 78 0 0,049%
Lombardia Italy 4490 5469 333 646 0,047%
Alava Spain 150 153 3   0,045%
Gyeongsangbuk-do South Korea 1043 1117 14 60 0,036%
La Rioja Spain 103 104 1 0 0,033%
Semnan Iran 219 221 2 0 0,029%
Hubei China 16995 67760 3024 47741 0,029%
Markazi Iran 413 416 3 0 0,027%
Mazandaran Iran 880 886 6 0 0,026%
Marche Italy 313 323 10 0 0,026%
Hoefuoborgarsvaeoi Iceland 57 57 0 0 0,025%
Emilia-Romagna Italy 1286 1386 70 30 0,023%
Qazvin Iran 254 257 3 0 0,018%
Gilan Iran 490 524 34 0 0,018%
Ticino Switzerland 65 66 0 1 0,017%
Basel-Stadt Switzerland 29 29 0 0 0,015%
Tehran Iran 1999 2114 115 0 0,014%
Alborz Iran 338 339 1 0 0,014%
Veneto Italy 694 744 20 30 0,014%
Corse France 37 38 1 0 0,013%
Madrid Spain 761 782 21 0 0,011%
Valle d'Aosta Italy 15 15 0 0 0,011%
Esfahan Iran 615 618 3 0 0,011%
Saint Barthelemy Saint-Barthelemy 1 1 0 0 0,010%

*from the analysis it can be expected that the Iranian estimates are somewhere between 2-6 times higher due to lack of testing/difficulty in reporting etc.


It is important to note that of these, only 1 of these provinces is in China, with Korea, Spain, Iceland, Italy and Iran having the highest percentages. In Nordrhein-Westfalen in Germany for instance, although 545 infections have been seen, this over a population of over 18 million, translates to roughly only 1 person in every 30,000 being infected.

At a municipality (or Level 2 administrative) level, these infected percentages increase significantly depending on the scale examined. An example for Germany is shown below, where we see 4 Landkreise (Districts) with over 1 in 10000 people infected. Kreis Heinsberg in Germany has 12.7 people per 10,000 infected (0.127%) - enough to put it in the top 3 on a province level. As an example from Italy, the province of Lodi with over 900 cases in Italy, has a higher percentage of the population infected than Daegu on the state level (0.47%). This details the importance of scale, but also likely the need for better definition of locations for travel advisories as well as alert systems across the globe to be able to reduce the spread.

For now, the SARS-COV-2 virus continues to spread around the world, and unfortunately it is likely that most countries will not escape cases of it before a vaccine is found. 

Age-mortality ratios, subnational level statistics and more will continue to be updated. Pandemic modelling is part of the Hotel Resilient Standards. For information, contact Risklayer at www.risklayer.com and info@risklayer.com.