Here& #39;s a nice little case study in what used to be called & #39;fake news& #39;. It may be worth reading if you& #39;ve seen that & #39;200,000 lockdown deaths& #39; number.

This is a short thread on how the tweet below is wrong, and how the Telegraph headline it misquotes was wrong.
Here& #39;s how the editor of the Telegraph article summarises it.

The article uncovered predictions from an official report put together back in April.
Here& #39;s how The Metro summarised The Telegraph& #39;s article.

Various other articles also summarised it in a similar way.
Pretty clear from those, right? "Coronavirus lockdown could cause & #39;200,000& #39; extra deaths& #39;.

Here is the actual Telegraph headline itself:
"Lockdown may cost 200,000 lives, government report shows".

But below is how the Telegraph summarised it on Twitter. Note the difference: "Lockdown *and* protecting the NHS"
And here& #39;s the actual figure in the article of the predicted impact in the first 6 months. Note that the figure is & #39;up to 25,000& #39;, not 200,000.

Below that it says & #39;a further 185,000& #39; in the medium to long term& #39;...
So that& #39;s 200,000 from lockdown, right?

Not really: Read what it actually says. "25,000 could die from delays to treatment".
Does & #39;lockdown& #39; mean the same as & #39;delays to treatment& #39;?

No: delays to treatment happened at the same time as lockdown, but they are not the same thing.

Delays to treatment were a policy designed to prioritise covid cases and, if you think about it, lockdown slowed covid cases.
And the same for the 200,000 number: It& #39;s a prediction on what would happen if & #39;75% of elective care& #39; was cancelled for the & #39;medium to long term& #39;.

And why would elective care be cancelled? Through prioritising covid cases. And covid cases were *reduced* through lockdown.
So nowhere here does the article fairly attribute any deaths to & #39;lockdown& #39; so far.

Here we get on to where the Telegraph summarise the numbers in a table.

The line here is the & #39;up to 25,000& #39; figure - ie, they& #39;ve taken the maximum amount.
If you work through that table, there *are* some actual death predictions & #39;within one year& #39;, which could be attributable to lockdown. Let& #39;s count them:

Recession: 600 to 12,000
Suicide: 500
Domestic violence: 20
Accidents at home: low tens

So absolute maximum estimate: 12,550.
Let& #39;s see if there are any other numbers in the article.

Here& #39;s the very last paragraph, which says:

& #39;The report points out that nearly 500,000 people would have died from coronavirus if the virus had been allowed to run through the population unchecked& #39;...
It ends by saying that there would have been & #39;more than a million non-Covid deaths resulting from missed treatment if the health service had been overwhelmed& #39;.

In other words: The report implies lockdown could save hundreds of thousands of lives - the opposite of the headline.
You can follow @danbarker.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: