I’ve been looking for the weekly or monthly Italian death stats (all deaths, not Covid-19 deaths). Looks like they are not publicly available, which is rather remarkable. Even the Italian National Statistics Institute doesn’t publish them, although I expect they have them.
However, this site has been mentioned by a few people on the internet (and also by one of my commentators): https://www.euromomo.eu/
It’s a group called the European Monitoring of Excess Mortality for Public Health Action.
It has official death stats for a large number of European countries, including Italy. One of its partners is the Istituto Superiore di Sanità, which according to Wikipedia is “the leading technical-scientific body of the Italian National Health Service“, so we can presume it has the real Italian stats (in as much as, cynics might say, there can be real stats for a place like Italy).
Euromomo don’t publish a lot of stuff on their site, but they do have some interesting graphs.
Before I go, let me note that no excess deaths are visible yet at a European-wide level, in any age group. In fact, current deaths are at very low levels.
But, you may say, we wouldn’t expect that at this stage. So let’s look at the countries first.
The graphs for individual countries are quite small and it’s hard to see what’s going on, so I’ve blown them up for you. Here, for example, is the Netherlands (click to enlarge):
I cut and pasted the footer at the bottom of the page to this graph, so you can see what’s going on. The ‘2019-16’ type numbers right under the graph refer to the year and the week of that year (some of these numbers are a bit hard to make out). The graphs go back about four years.
What the graph shows is z-scores. What’s a z-score? Well, first of all you work out the average number of deaths in that country for a day, or a week, or whatever. This is called the ‘baseline’, and is represented by the red line. (Euromomo base their baseline on the last 4 1/2 years of data, and it looks like they base it on certain parts of spring and autumn, rather than the whole year.)
Then you graph the real daily figures as a line around the baseline. When the daily (or weekly figures, depending on what they’re using) are over the baseline, the line gets higher. When the real figures are lower than the baseline, the line goes under the red line. In effect, you see when things diverge from the norm (as often happens in in winter, when we normally see more deaths).
(Apologies to those for whom this is very basic.)
So with the Netherlands we can see that generally things have been pretty normal for the last four years, but there were some spikes in winter 2016-17 and 2017-18, which is quite typical of winter, although there weren’t excess winter deaths in winter 2018-19. But look now at the last right-hand bit of the graph, winter 2019-20. Hardly any spikes in the data for that period. No sign at all of the scariest soil-your-pants epidemic in a century there. In fact, the last few months have seen the deaths falling to pretty low levels.
Big deal, you might say. We know Covid-19 is not doing much in the Netherlands so far. But it might later on. (In actual fact, though, you’d have to see an awful lot of deaths for that graph to get anywhere remotely scary levels.)
Anyway, let’s look now at Italy, because this is where the biggest horror story is supposed to be right now (click to enlarge):
Look at the last right-hand bit again. This time, we have something. There’s a pronounced uptick in the last few weeks (it looks like the stats go up to the middle of last week). This is presumably a sign of Covid-19 killing people. But far from being an unprecedented slaughterhouse direct from Stephen King’s imagination, we’re still not seeing anything out of the ordinary. Winter 2016-17 and 2017-18 were worse. Of course, Italy may get worse still, and probably will for a few weeks, but as we can see from all these European graphs, a high number of extra deaths due to communicable disease after mid-spring and through summer is basically unknown. Summer heatwaves sometimes kill extra people, but communicable diseases very rarely do.
That doesn’t mean it can’t happen this time, but it does mean that we would need very good evidence to the contrary to justify shutting down the world’s economies. What’s more likely is that we”ll have a recurrence in winter 2020-21 in some parts of Italy, perhaps a somewhat bad one, for which the Italians should ready their health service for. And perhaps it makes sense to have some mitigation strategies in place in in vulnerable parts of northern Italy now. But I’m still not seeing anything resembling World War 3. (Remember that Italy is supposed to be where everything is happening right now, not in two weeks time.)
It can also be seen that most of this increase in the death rates in Italy is due to increases in deaths of over 65s. Click on the >=65 years button at the top of the page at euromomo to see this, and compare it with the other age groups.
Let’s now take a look at the UK, which is divided up into its constituent countries (click to enlarge):
Nothing whatsoever to see here. All totally normal. Not even any recent upticks. This all accords with the recent UK graphs two of my readers have done. Wales is currently even lower than normal.
There have, of course, been some UK Covid-19 deaths since the middle of last week (which this graph goes up to), but even if they really all are excess deaths, the numbers are way too small to make much difference to these graphs.
You might still think, despite everything I’ve said, that the apocalypse is defintely coming for us this summer. Here’s one more Euromomo graph to look at in that case (click to enlarge):
(Note that these are death numbers, not z-scores. I don’t know how they’ve calculated the baseline here, but they’ve done it for each age group.)
What do you notice about this? Well, for one thing, the numbers for the kids are fairly straight lines (on average), whereas for the adults, it’s a series of waves. The deaths go up in winter, and down in summer. It’s this wave-like effect that is important, especially with Covid-19, which kills almost exclusively adults, not kids. I expect you would see something similar pretty much every year with Western countries (apart from the World War I and II years). (If anyone has a link to such graphs, please post it in comments.)
A communicable disease epidemic in summer is really, really unusual. So unusual that I don’t expect it to happen, and in the absence of rock-solid evidence that it will, we should expect things to die down in a few weeks or a month. That’s one reason — not the only reason, but one important reason — why you shouldn’t freak out over claims of exponential growth that are just around the corner.
Update: We have more revealing graphs coming in to put up for you tomorrow. Keep ’em coming.