Thread by @VictimOfMaths

Today @NatRecordsScot have released their first set of detailed mortality data for Scotland for a month.

Lots to unpick in here, but here's a few things I've noticed.

Firstly, deaths from all causes have been at the higher end of normal for a few weeks.

It *could be that this is just random noise - Scotland is a smaller country than England & Wales, so we'd expect the line to 'bounce around' a bit more. But 3 weeks is a bit of a coincidence, albeit it's not a long way above normal.

Some interesting age patterns - the 'bump' in recent weeks seems to be largely among 45-74 year olds.

It's also not likely to be anything to do with the heatwave-related mortality bump we've seen in England & Wales, since it's a good few weeks later. Also, Scotland is cold.

If we look at excess deaths by cause, it's clear that the recent bump isn't being driven by COVID-19 deaths. There doesn't seem to be any obvious pattern, though deaths from 'other' causes have been consistently above 'usual' levels for a few weeks.

If we look at *where* people in Scotland are dying, we see, in common with England & Wales, that deaths at home are well above historical levels.

It *looks* as though these are mostly displaced deaths from hospitals, but we can't be at all sure of that.

Finally, the really nice thing about this Scottish data is that we can look at deaths by cause *and* location.

So many interesting stories here. Hospital excess deaths were all COVID. Care homes excess deaths mostly COVID and demential/Alzhemiers.

It *looks* as though a key part of the story is a displacement of cancer deaths from hospital to home (although we don't know if they are actually the *same* deaths). For a long while it looked the same thing was happening with circulatory deaths, but seemingly not any longer.

It's also perhaps interesting that there have been a lot fewer respiratory deaths in hospital than usual, and these seemingly haven't shown up in other settings. But this 'deficit' has shrunk to almost nothing in recent weeks.

I'm not sure what this all tells us. The data probably invites at least as many questions as it answers, although those questions are at least more targeted. It's also not clear how much we'd expect the same things to be happening in the rest of the UK.

But it's fascinating

So all credit to @NatRecordsScot for publishing it, and I can't wait until we have some similar insight for the rest of the country.

Code for these various plots is here (1st plot):
github.com/VictimOfMaths/…
here (deaths by age): github.com/VictimOfMaths/…
and here (deaths by cause/location):
github.com/VictimOfMaths/…