What’s Unique about US COVID Deaths?
How Are We Doing?
How has the United States fared in the COVID-19 pandemic compared to other countries? As the country passed an estimated 200,000 deaths from the virus this week, news reports repeated the claim that the US has the most COVID deaths in the world. Coverage has also implied that the dire statistics reflect how America’s political leadership bungled the response.
Global comparisons are often misleading. The “most COVID deaths” one includes developing countries whose medical surveillance is not as good as the US and where the demographics skews towards younger people who are less vulnerable to the disease. When people compare the US homicide rate to Central American countries, experts rightly cry foul and they should in this case as well.
European countries with a similar economic level as the US have had death rates ranging from much lower to the US (Germany) to somewhat higher (United Kingdom). Yet the news was filled with reports of Europe returning to something more like normalcy this summer while the US faced a mounting death toll and renewed lockdowns.
So is the US doing badly or not?
Comparing Death Counts
Comparisons tend to be among total casualties of COVID or the week to week numbers. Here we look at both. Excess deaths – the difference between the number of deaths in each week of 2020 compared to average values for the same week in previous years – gives us a measure of deaths from COVID. In countries hit hard by the virus, it is the main cause of excess deaths.
Comparing raw number of deaths is misleading, since different countries have different populations. Here, I plot the fractional change in excess deaths: (2020 deaths) – (previous year deaths) divided by (previous year deaths). This automatically adjusts for population. It also partly adjusts for demographics. For instance, a population with a bigger fraction of older people will have a higher death rate in a normal year, and a higher death rate from COVID (which kills older people at a higher rate).
The top panel shows much higher peaks in weekly excess death in Italy, Spain, and England/Wales (over 3% of total annual deaths) than in the US (about 1.4%). However those countries, as well as others with less excess death, all returned to near-zero weekly excess by week 26, and all but England/Wales before week 20.
US on the other hand had a smaller peak than most, but only a mild decline in late spring, an increase starting around week 23, and a second peak in the summer. The US surge in deaths started after week 13, several weeks after all the European countries except England/Wales started surging.
The cumulative totals (bottom panel) show the total deaths by the end of the summer (value of each curve for the last week shown). Like other comparisons, this shows that US totals were about the same as Italy, Spain, and England/Wales. and higher than France and the Netherlands. As the top panel shows, Germany had hardly any surge in deaths from COVID, and because 2020 started with fewer weekly deaths than previous years, it ended up with virtually no excess deaths for the first 30+ weeks of the year.
What the Numbers Tell Us
The story I see in these numbers is as follows: among Western countries, Italy and Spain got hit first. It is not clear why they got hit so intensely, but part of the reason may be that even the week or two extra warning time that other countries got was enough to help them respond better. European countries, except for Britain to some extent, responded quickly and strongly to the rising infections and therefore severely reduced infections. Britain followed suit after Prime Minister Boris Johnson, an early skeptic of lockdown measures, came down with the disease himself.
The US was lucky to get a slow initial onslaught. This may have been due in part to geography. The US has several different population centers, but at first it only grew in a few places, probably carried by people traveling from Europe. New York City got hit hard in the beginning, but the greater New York metropolitan area (about 20 million people) is only 6% of the US population. In comparison, metro Paris (13 million people) is 19% of France.
As the virus slowly rolled through the sprawling territory of the United States, we had a unique opportunity to stop the virus, an opportunity we squandered. It is clear that much of blame for this lost opportunity goes to Donald Trump and Republican governors who followed his lead in undermining efforts to stop the virus through social distancing and use of masks.
About the Data
Data is from the Our World in Data website (see figure above for web address) . Thanks to Mark Herdeg for pointing out this set of statistics.
I used all the data for Western European countries of more than 10 million people and the United States. Canada is perhaps the most similar country to the US but was not included in the dataset, and I couldn’t find excess death data for it anywhere else.
All the time series ended between week 33 and week 36 except Italy, which ended at week 26. I don’t know why that is.