An effective response to the Covid-19 pandemic needs to be diligently scientific and well-coordinated. There is ample data logging the spread of the virus throughout our world this year. We can use this data to measure the US response against its peer countries to determine the relative effectiveness of this response. This analysis paints a picture of the wrong kind of American Exceptionalism: exceptionally bad.
This analysis focusses on Covid deaths, rather than confirmed cases because of the large variation in test capacities and test methodologies among regions and countries. Comparing deaths, although not perfectly consistent, is a much more uniform criterion. In developed countries, a majority of Covid deaths occur in hospitals after days of observation, and missing Covid as a cause in these cases is unlikely.
This analysis will also focus on comparing the US response to countries in the European Union as the best example of peers. There are many factors relevant to the Covid spread that correlate well between the US and the countries of the EU. These factors include:
- having similar climates
- having similar arrival time of first infection
- are in the same hemisphere and therefore having the same seasons
- having similar healthcare system quality
- having similar life expectancies
- having similar prevalence of contributing health risks
- having similar population density distributions
The rate of infection spread is a strong function of population density. The higher the population density, the higher the number of people an infectious person will encounter resulting in faster viral transmission. Visualizing the Covid prevalence in the US and the EU using the dimensions of Covid deaths and population density is an illuminating way to compare the effectiveness of their relative responses. Rather than comparing the US directly to EU countries, this analysis will instead compare individual US states to EU countries as they compare similarly in geographic area, total population, and overall cardinality. Population distributions are not particularly uniform, with populations significantly clustered in cities and surrounding areas. To account for this population clustering, the analysis uses a weighted population density metric, rather than simply the overall density of the individual states and countries. To arrive at this weighted density metric, the density of sub-regions (Counties in the US and NUT3 regions in the EU) were taken and combined weighted by population to a single metric for each state or country. EU countries from the contiguous European continent and close surrounding islands with a population greater than 200,000 were selected for comparison as were the 48 contiguous US states plus the District of Columbia. The geographic outliers of Alaska, Hawaii, and Iceland were not included because this isolation makes them poor peers for this comparison. The data used to create this analysis is taken from the following sources:
- US Population and Densities - census.gov
- EU Populations and Densities - ec.europa.eu
- Covid Deaths - worldometers.info
The scatter chart above illustrates this population density Covid infection trend with US states as red dots and EU countries as blue dots. The chart above shows the clear correlation between the density of the population and the prevalence of Covid-19 infection as indicated by the death rate(deaths/1M ppl). If you are having trouble interpreting this chart, the lower-right corner is the region of high population density and low infection prevalence indicating the most effective response, while the upper-left corner is the region of low population density and high infection prevalence indicating the least effective response. Also, as indicated by the included trendlines, it shows that the Covid-19 death rate trend among US states is more than 4 times higher than the trend among the EU peers. Though we cannot fully account for all differences, this significant factor of additional deaths points clearly to an inferior response to containing the virus in the US.
I do not want this analysis to morph into political opinion, so I will refrain from speculation on why the US response has been so inadequate compared to its world peers. I do urge readers in the US to think about why this is the case, and what they can do personally to improve the situation. I would also like to remind US readers that the biggest opportunity that most of us have to influence policies that could improve the situation is coming soon, Tuesday, November 3rd, 2020.