More on Excess Mortality and COVID19’s Hidden Toll

I’m returning to this topic of calculating the full death toll of the COVID19 Crisis. As I noted yesterday, samplings of data from New York, Spain and Italy suggest a pattern in which the true scale of mortality tied to the COVID19 Crisis is roughly twice that recorded in the COVID19 death tolls we see each day, and sometimes much higher. Here are a few more data points which add weight to this emerging pattern.

First, the United Kingdom has released provisional data through the 14 week of the year – March 30 through April 5th. For England and Wales, the excess mortality over the five year average was 6,082 whereas the total COVID19 death toll was 3,475. In this case, the official toll accounts for only 57% of the recorded excess mortality. A bit higher than the rule of thumb but close. (Oddly the UK dataset refers to this as the week ending April 3rd, despite Week 14 ending on April 5th. Regardless, the numbers and percentages are the same whichever seven days are being counted.)

Second, the Netherlands’ National Institute for Public Health and the Environment has produced this report which notes that excess mortality in the country from March 26th through April 1st was “approximately twice as high as the COVID-19 deaths reported to RIVM in the same week.”

Third, I’ve previously noted some data from Spain. Here I noted data from Madrid in the second half of March, that was roughly double. But there’s also data from March 10th-16th. In that case the 192 official COVID19 fatalities account for only 37% of the full excess mortality during the period (524). A comparable look at the Castilla la Mancha from March 15th-24th shows official COVID19 deaths accounting for 51% of the total excess mortality. (See this March 28th article in El Pais.)

Finally, The Economist notes more extensive data which L’Eco di Bergamo compiled of 82 localities in Bergamo province. In March, they logged a total excess mortality of 2,420 deaths of which only 1,140 were officially attributed to COVID19. Here the percentage captured is 47%.

These are different countries, different epidemics, different record keeping systems, all taking place in one of the most turbulent run of weeks since the Second World War. For all that chaos and variety, the consistency of this simple pattern is remarkable.