This article contains visuals regarding the global impact of the COVID-19 virus. It will be updated as new data is retrieved, analysed and visualised.
The global pandemic that is the COVID-19 coronavirus is the first to occur in an era of open source data and tools to learn about its impact. This article contains a series of visuals analysing the number of deaths attributed to COVID-19 worldwide. These statistics may be upsetting. If you do not want to view such information, please leave the page now.
Two groups visualisations are published here. The first group plots countries per continent, showing the changing cumulative deaths daily from the day the first death attributed to COVID-19 was reported. The second group compares countries worldwide with at least 200 deaths reported. Each chart starts day 1 from a different count: 1 death, 10 deaths, 100 deaths and 1,000 deaths respectively. It indicates that countries may not be correctly reporting the first death attributed to COVID-19.
In all visualisations posted here, the data are actual counts. They have not been scaled to account for differences in population sizes between countries. That will be the subject of a separate post (link will be added here when it goes online).
The primary data source is the time series of reported deaths attributed to COVID-19 being published openly on Github by the Centre for Systems and Science Engineering (CSSE) at John Hopkins University.
Cumulative deaths daily by continent
Data as of: 28 March 2020
Data source: https://github.com/CSSEGISandData/COVID-19
The following visuals show the cumulative number of deaths daily from the day the first death was recorded, per country grouped by continent. For comparison, the 28th day is shown as a dotted vertical line. The y-axis scale varies by continent. Labels are not included for countries with low counts relative to others in the continent (for readability).
Comparing countries worldwide from different starting points
The following visuals show the cumulative number of deaths daily for countries worldwide with at least 200 deaths attributed to COVID-19. Each visualisation starts day 1 from a different count: 1 death, 10 deaths, 100 deaths and 1,000 deaths respectively. They indicate that countries may not be correctly reporting the first death attributed to COVID-19. France appears to have a long period of low counts before increasing exponentially. Looking at these plots, it suggests that France detected COVID-19 earlier than other countries in Europe. The same may also be true of other countries…
When starting day 1 from when 100 deaths have been recorded, the curves more closely align between countries and two different curves become evident. One for China and Iran, one for all other countries. Within the UK, the official reporting has been based on the UK following the China path. This data suggests otherwise. However, it could be that only looking at total deaths, rather than scaling for population size, may be having an effect. This is discussed in a separate post, to be published shortly (link will be added here when it is live).
a) day since 1st death recorded
b) day since at least 10 deaths recorded
c) day since at least 100 deaths recorded
d) day since at least 1,000 deaths recorded
Note: for these visuals, the maximum number of days plotted is capped at 56 and the data is as of 29 March 2020 (deaths recorded up to 23:59 on 28 March 2020).
The data is retrieved from https://github.com/CSSEGISandData/COVID-19. Locations are tidied according to notes in the related Daily Updates post. Counts are actual values. They have not been scaled to account for different population sizes. The accuracy of the reporting is unknown and may vary by country. For notes about preparation and the daily change log, visit COVID-19: Daily Updates
All analysis and visualisation completed in Python 3.