Thursday 19 September 2019

Minimum pricing: media fall for blatant cherry-picking

A few months ago, I lamented the media giving wall-to-wall coverage of the news that alcohol sales fell in Scotland in 2018 while ignoring the news that alcohol-related deaths rose in Scotland in 2018. For the purposes of the 'public health' lobby, 2018 was 'after minimum pricing', although minimum pricing wasn't introduced until May of that year.

As I said in June when the National Records of Scotland released the alcohol-related mortality figures to a seemingly uninterested world...

All I want here is a bit of consistency. If we're going to pretend that the calendar year of 2018 was the post-MUP period and use post hoc rationale, then minimum pricing drove down alcohol consumption while failing to reduce alcohol-related deaths. Today's statistics should have received at least as much attention as the sales figures did last week ('the whole policy here is about reducing alcohol related harm', after all).

Instead, today's figures have been released to the sound of crickets and tumbleweed.

Ever get the feeling you're being played? 

Today we have seen that the media are interested in what happened to alcohol-related deaths in Scotland in 2018 - so long as the figures are spun in a way that is helpful to minimum pricing campaigners.

The Times:

Fewer deaths after minimum alcohol pricing, Glasgow doctors say

Drink-related deaths plunged in Glasgow after Scotland set a minimum price for alcohol, researchers have revealed.

ITV:

Alcohol-related deaths ‘cut by more than 20% with minimum unit pricing’

Minimum unit pricing may have contributed to the number of alcohol-related deaths in Glasgow falling by more than a fifth, according to new research.

And, of course, the BBC:

Charity calls for alcohol minimum pricing to be extended across UK

A charity has called for Scotland's minimum unit pricing policy for alcohol (MUP) to be rolled out across the UK. 

It followed the publication of evidence suggesting MUP has had a significant impact on drinking patterns.

Data presented at a conference in Glasgow suggested alcohol-related deaths in the city had fallen by 21.5%.

The policy was introduced in May 2018, but organisers said there was already an indication it was working - and should be more widely applied.

None of these reporters showed any interest when the National Records of Scotland released data showing that the number of alcohol-related deaths rose from 1,120 in 2017 to 1,136 in 2018. But they have become very interested now that a longtime supporter of minimum pricing has announced the results of some unpublished research at a conference.

'Research' might be too strong a word. All he done is go to the data released in June, found the spreadsheet that shows the number of deaths by local authority area, and cherry-picked one where the number of deaths went down - against the national trend. (See Excel spreadsheet, tab 5.)


If he had picked Aberdeen City, he could have announced that the number of alcohol-related deaths rose by 51 per cent 'after minimum pricing'. If he'd picked Perth and Kinross, he could have claimed that they rose by 76 per cent.

None of this would be untrue per se, but it would be lying by omission. If we want to know what happened to alcohol-related deaths in Scotland, all we need to do is look at the number of alcohol-related deaths in Scotland, which is right there on the same spreadsheet. It rose by 1.4 per cent.

As junk science goes, this doesn't even have the merit of being sophisticated. It is straightforward cherry-picking. It could be checked and debunked by any responsible journalist in a matter of minutes.  

It is difficult to resist the suspicion that the media, especially in Scotland, are simply not interested in the truth of the matter. They only want to hear that minimum pricing has worked.

UPDATE

Here's the conference that started all this. The speaker is Ewan Forrest, the guy who carried out the 'research'.


The claim is that off-trade sales fell in Scotland but rose in England. From the figures I've seen, this is not true. Sales rose in both countries.

Be that as it may, the key claims relate to alcohol-related deaths. Forrest admits that there was 'no immediate effect upon overall figures for alcohol-related deaths in Scotland in 2018'. If he were to be more precise, he would say that they rose.

His slide then mentions 'possible early evidence of MUP effect in Glasgow' (the cherry-picked and unrepresentative decline) but says 'it is still too early to determine definite long-term clinical benefit'.

Hardly sufficient to justify the claims from the press today - which were further exaggerated by campaigners - but, as I say, few people now care about the facts.

No comments: