Are covid-19 deaths around the world being under-counted? To find out my colleague @J_CD_T is tracking (with brilliant viz by @martgnz ) excess mortality around the world. Check out the tracker here: (1/n) https://www.economist.com/graphic-detail/2020/04/16/tracking-covid-19-excess-deaths-across-countries

If you've being living under a rock and somehow missed this, see @J_CD_T's fantastic thread, too: (2/n) https://twitter.com/J_CD_T/status/1250815341044924425

In UK, ONS figures on excess mortality released last week produced quite a shock: covid-19 deaths only accounted for some 58% of the total excess mortality observed over the previous three weeks. (3/n) https://twitter.com/d_spiegel/status/1249986522692096003

Were these excess deaths down to misclassification issues or because of more people dying of causes unrelated to covid (due to strains on NHS)? Last week I tried to look for clues in the release... (4/n) https://twitter.com/JamesFransham/status/1250081883834257409

The @ONS has released new figures today (covering the period to April 10th). Some good news: the excess mortality accounted for by covid-19 increased from 57% (in week to April 3rd) to 78% (in week to April 10th). That suggests that misclassification was indeed the primary cause.

Here's a screen shot of our excess mortality tracker, showing covid-19's increasing share of excess deaths. There were still some 1,783 excess deaths in week to April 10th unexplained by covid-19: (6/n)

That misclassification, though less, clearly continues. This chart shows weekly deaths from respiratory diseases (black line) versus avg. over past 5-years (dotted blue line); + covid-19 deaths (red columns). There are still some ~400 "excess" deaths from respiratory illnesses.

That misclassification was most apparent in South West; & Yorkshire & Humber regions. This chart shows excess mortality by region, and by day of week, with covid-19 deaths as black dots. Chart shows covid-19 accounts for higher share of that excess mortality in those regions now.

And the same chart showing excess mortality by age group, gender and by week. Demonstrates that excess deaths unaccounted for by covid-19 have reduced, by still present in older age groups. People aged 75+ account for 78% of unexplained excess. (9/n)

I expect next week's data will show a lower share of unexplained excess as covid-19 awareness increased across the country. Do keep tabs on @TheEconomist's excellent excess mortality tracker for a global view on the true toll of the covid-19 virus: https://econ.st/2Km7b3d  (10/10)