UPDATE: Today @martgnz and I are publishing an update to @TheEconomist excess mortality tracker. We have new / revised data for all the countries, plus we are adding Indonesia, our first location outside of the West. (1/11)
economist.com/graphic-detail…
economist.com/graphic-detail…
There are some faint optimistic hints in the data: Italy, Spain, and the Netherlands all show signs of a plateau. But interpret that with caution. Many countries are retrospectively revising their numbers of all deaths upwards. (2/11) 
Starting with Italy: by April 4th there was a big drop in excess deaths in the Lombard municipality figures published by @istat_it. There may be some incomplete data there, but probably a real decline in deaths too. (3/11)
Spain looks to have levelled off, according to data published by @SaludISCIII. A key disclaimer though: the trend is largely explained by a big decline in Madrid, where total mortality data seems to be lagging quite a lot. (4/11)
For France’s national chart (but not yet for regions), we have added care-home covid fatalities, as published by @datagouvfr. These account for a large slice of the national excess deaths, which were still rising as of April 6th. (5/11)
In Britain, as recommended by @danielhowdon, we are now using @ONS covid deaths by “week registered”, since total deaths are also by week registered. This definition accounts for 57% of excess to April 3rd. (New data coming tomorrow.) (6/11)
In the Netherlands, new data from @statisticscbs suggest a plateau in excess deaths for the week ending April 12th. Because the official covid toll only includes people who tested positive, there is still a large under-count. (7/11)
In New York, we are now adding “probable” covid deaths, and have included @CDCgov total mortality data up to April 4th. There has been a huge spike in overall fatalities, but the confirmed + probable count picks up most of them. (8/11)
(Post-script for the New York data – I have seen some quite substantial upward revisions in the CDC weekly total mortality data before, so the excess in the most recent week might be much higher.) (9/11)
Finally, we’ve added numbers from Jakarta, which show 1,600 extra burials in March but only 84 official covid deaths. This data is far from perfect, but probably indicative of the sort of drastic under-counting we should expect to see in developing countries. (10/11)
Thanks to everyone who has sent tips for other countries to jamestozer@economist.com - please keep them coming. We will expand the sample in our tracker soon. Corrections such as the one by @danielhowdon are both very welcome and essential. (11/11)
