In Scotland, the GMB union have been vocal in calling for the proposed increase in the MUP threshold from 50p to 65p to be scrapped, on the basis that this is a 30% rise, while wages and disposable incomes have risen by barely 1% since 2018.

But wait...

gmbscotland.org.uk/newsroom/gmb-s… ...The 1% figure that GMB are citing is *after adjusting for inflation*. The latest data shows that household disposable incomes are 1.3% higher than they were at the start of 2018 in real terms.

But the MUP threshold hasn't been adjusted for inflation over the same period...

OK, I read this and while there are some sensible points there is an awful lot of nonsense in here.

So here's a few things I'd like to address:

https://twitter.com/LicensingLaws/status/1761353228753473733

1) As I've already had a little rant, the attempt to dismiss the use of counterfactuals as "speculation" (but only in studies that show a benefit from MUP, they are fine in studies that find no benefit) is unbelievably dumb.

https://x.com/VictimOfMaths/status/1762768863559606700

Quite a few takes today on the evaluation of MUP which talk about it 'failing' or having 'mixed impacts', in spite of headline evaluation results showing it reduced alcohol consumption and deaths.

To some extent this reflects one challenge of such a comprehensive evaluation... If the only studies that had been done were those on population level alcohol consumption and harm, you might reasonably claim on that basis that MUP had been an unqualified success.

Our new paper looking at alcohol, drug and suicide mortality during the first 2 years of the pandemic in the USA and the UK nations has just been published in @RSPH_PUHE

Lots of attention on these deaths early in the pandemic, so what actually happened?

doi.org/10.1016/j.puhe… We used publicly available mortality data to calculate age-standardised mortality rates for each cause and country from 2001-2021.

The shaded grey area represents the pandemic preiod.

Americans, look away now.

Ugh. I have a lot of time for @IHME_UW but their consistent refusal to engage with reasonable criticism is disgraceful (@TheLancet don’t come out of this too well either, mind).

Full credit to @WHO for taking a more open and receptive approach.

https://twitter.com/jschoeley/status/1625056164969840642

@IHME_UW @TheLancet @WHO This refusal to engage with criticism is a depressingly recurring feature of IHME. Their COVID infection modelling repeatedly gave completely implausible results. They never publicly responded to people pointing this out, just quietly made (opaque) changes to their models...

In 2020, we saw a sharp increase in alcohol-specific deaths.

ONS haven't yet released their alcohol-specific deaths data for 2021, but I realised you can reconstruct the figures from data they have already published, and it's not good news, I'm afraid, a further 7.6% increase. This represents the second largest single year increase in at least the past two decades, behind only last year.

Both men and women have seen alcohol-specific mortality rise sharply since the start of the pandemic.

Last week's data from @NISRA on alcohol-specific deaths in Northern Ireland in 2021 continued the recent trend, with NI separating from the lower rates of England & Wales and heading up towards Scotland.

We should get 2021 data for E&W in December I think. Interesting that there is still a gap for alcohol-specific death among men between Scotland & Northern Ireland, but the rates for women are almost identical.

Since @NISRA published new data on alcohol-specific deaths in Northern Ireland in 2021 yesterday, here's an updated map showing how alcohol-specific death rates vary across the UK. Here's the same data on a conventional map, if that's more the way you roll.

Lots of people rightly pointed to the marked rise over the course of the pandemic in people aged 16-64 who are out of work but not actively seeking a job due to long-term sickness when this data was published a few days ago.

I have some other questions though... 1/ The rise in long-term sickness started *before* the pandemic - back in early 2019 (so too early to be COVID before we knew what COVID was before anyone suggests that) - what gives?

Hey Twitter, we've got a shiny new preprint:

Led by @drjenndowd, we looked at deaths from alcohol, drugs and suicide in midlife in the US, Canada and the UK nations and how these have changed since 2001. With graphs.

doi.org/10.1101/2022.1… There has been a lot of (very interesting) research focused on these so-called 'deaths of despair' in the US, but, driven by the opioid crisis, this work often talks about how unique the US is in having an epidemic of midlife deaths of despair.

Our new study, led by @FionaBeyer, is out now in @AddictionJrnl

We did a systematic review and meta-analysis to look at whether brief interventions for alcohol are more or less effective if delivered in person vs. digitally (e.g. via apps).

onlinelibrary.wiley.com/doi/10.1111/ad… We found *loads* of studies - 201 studies with a combined total of almost 95,000 participants.

The analysis showed that both practitioner-delivered and digital interventions led to a significant reduction in alcohol consumption at 6 months follow-up...

Another way of looking at the recent ONS data on inequalities in life expectancy and healthy life expectancy in England for 2018-20.

I think there are 4 really interesting and important things in this graph... 1) Overall life expectancy is unequal.

Life expectancy in the most deprived 10% of areas is 8 years shorter for women and 10 years shorter for men than in the least deprived 10%.

This isn't surprising, but it's still bloody depressing.

Yesterday ONS released new data on life expectancy by deprivation.

These figures are based on 3 years of pooled data, so the new 2018-20 figures are the first to include (some of) the impacts of the pandemic.

Bad news for everyone, but no surprise to see inequality growing. And this is only going to look even worse in 12 months time when the calculations will include 2020 *and* 2021.

I posted this graph yesterday as part of the #30DayChartChallenge and a few people have asked some reasonable questions about why I chose to present this data in this way.

So I thought I'd write a thread to explain my thought process... The first thing to say is that I don't think any dataset has some inherently 'right' type of graph. There are just different graph types which may be more or less effective at highlighting different aspects of the data.

Feels as though I should get in on the #30DayChartChallenge action at some point.

Spent far too long failing to make a custom legend for this, so you'll just have to work it out for yourself. R code here:
github.com/VictimOfMaths/…

It seems likely that COVID cases are now falling (although hard to be certain), but it's striking, and concerning, how much worse the impact on hospital admissions has been for older age groups with BA.2 than BA.1.

This would be a good time to get a lot of boosters in arms. I think this divergence is a combination of a) the waning impact of 3rd dose boosters in the Autumn, which will affect older age first as they were boosted first and b) possibly a difference in the way different age groups have reacted to the removal of restrictions...

A short(ish) thread on the COVID situation in English hospitals.

The majority of people in hospital with COVID are being treated for something else, but the number of people being treated *for* COVID has started to rise again as a result of the recent increase in prevalence. There remains a fair bit of regional variation in these two trends. In London, only around a quarter of patients with COVID are being treated for COVID, while that proportion compared to 55% of patients in the South West.

Down with this sort of thing.

Last week's data suggested a chunk of this rise is 'incidental' admissions where people aren't being admitted *for* COVID, but as @AdeleGroyer has shown, the number of people catching COVID in hospitals has also been increasing rather worryingly. Some perspective on those rises - admission rates are still very low, thankfully, for younger age groups.

NHS England have just published their latest COVID admissions breakdown by age. Which means I can update this plot of admission rates.

You can see the recent uptick in admissions, but I'm really struck by how slowly admissions have fallen since the Omicron peak. Compare and contrast with the speed with which things fell back last winter. Note the change in the colour scale, because things were *much* worse back then.

This also looks like good news - the recent increase in people in English hospitals with positive COVID tests is mostly driven by patients who are being treated for something else, *not* COVID. There are some big regional differences in this breakdown.

Reassuring in particular that the big rise in COVID hospital patients in the South West is almost entirely driven by these 'incidental' COVID diagnoses.

One constant throughout the pandemic has been people shouting "SEASONALITY" whenever I suggest uncertainty about shifts in the data.

I can certainly believe that seasonality has some influence, whether that is directly or indirectly through changes in behaviour... ...(e.g. people meeting more inside in the winter) but the idea that seasonality is some immutable, irresistable force that explains the vast majority of changes in the COVID data just doesn't seem to match with the basic facts as I see them...