Thursday, 18 August 2011

Once, twice, three times a headline

Hello, is it an accurate headline you're looking for?

On the afternoon of Friday 12 August, the BBC tweeted the below:

 BBC News 

I could pretend that it made me think, wow - that's shocking news! But because it is just way, way, way too far fetched to be remotely believable, it immediately made me think, oh dear, somebody's made a major gaffe.

The previous day (Thursday 11 August), the Department of Health had released the latest A&E performance data, covering Quarter 1 of the 2011/12 financial year (1 April 2011 to 30 June 2011).

The previous Government's target was for a four-hour maximum wait in A&E from arrival to admission, transfer or discharge. In practice, this used to be measured by using a 98% threshold - in other words, an NHS trust was deemed to have achieved the target if 98% or more of people attending A&E waited less than four hours.

With the change in Government, the 'target' remained. Ish. 'Target' is now a dirty word and a big no,no - it smacks of the previous regime. So the 'target' became a 'standard', and the threshold for achievement was relaxed from 98% to 95% in June 2010.

Now that we're in 2011, and a whole new financial year - the first full financial year since the change in Government - we have the Department of Health further distancing themselves from the culture of targets.... by.... well...... by using much more creative language.

We don't have targets any more.

We don't even have standards.

What we have is IPMfNOs.

Ok, so I made up the acronym, but I can only dream of coming up with the satirical masterpiece that is.... ready for it.

Integrated performance measures for national oversight.

So targets then? No, no, no - integrated performance measures for national oversight.

Standards? Nope, you're not listening. Pay attention at the back - integrated performance measures for national oversight.

Riiiiight. You say integrated performance measure for national oversight. I say target. And I'll be in Scotland afore ye.

The reason why the BBC's tweet is too far fetched to be remotely believable is that 3.6 million people attended major A&E units between April and June 2011. For the time that they had to wait to almost double would be a mind-blowingly catastrophic deterioration in service.

What the latest figures from the Department of Health actually show is that 95.5% of the 3.6 million people attending major A&E units between 1 April 2011 and 30 June 2011 waited less than four hours.

During the same period last year (1 April 2010 to 30 June 2010), 97.7% of the 3.6 million people attending major A&E units waited less than four hours.

A deterioration in performance? Yes.

A doubling in waiting times? NO.

In Quarter 1 last year, 84,439 people waited longer than four hours. In Quarter 1 this year, 161,422 people waited longer than four hours.

Ahhhh! So that's the almost doubling that the shockingly inaccurate headline is trying to address! We've got almost twice as many people waiting more than four hours this year than last year.

Interestingly, the current official figures (much more detailed data is now available, but it's very much still in its experimental infancy) don't capture how long individual people actually wait - they only look at the four hour barrier, and report those waiting less than four hours and those waiting more than four hours. So actually the 161,422 people waiting more than four hours this year might (incredibly unlikely, but might) all have only waited four and a half hours, while the 84,439 people waiting more than four hours last year might (again, incredibly unlikely, but might) have waited seven hours.

Unrealistic example, granted. But the point is that we just don't know. All we know is the volume under four hours and the volume over four hours.

I replied to the BBC after their tweet:

 Chris Mason 

They didn't acknowledge me, but the power of the Goldacre is great, particularly when he re-tweets you:

 ben goldacre 

A dozen or so re-tweets later, the BBC seemed to get the message (although still no acknowledgement... and they still haven't deleted their original tweet).

Their story headline went from the diabolical:

To the rubbish and cryptic:

To the still misleading and far too generic:

And that's what's currently up there.

Three swings, three strikes.

And I used to take note of the 'Last updated' tag, believing it to be a trustworthy way of knowing whether and when stories had been amended. The fact that all three versions of the story purport to be 'Last updated at 15:11' is clearly misleading nonsense.

All in all, a pretty poor show, and a totally unfounded and easily avoidable slur on the performance of the NHS.

Always dangerous to to be sanctimonious, but you'd have to think that a 10 second reality check would have prevented the original error. You don't need to be a health expert. You just need to think for a second and say, "So does this mean that people are waiting twice as long as they used to?".

If the answer is a resounding no, then tailor your headline appropriately. And if you revise it (repeatedly), try to remember to mention A&E, and the crucial four hour barrier.

It, you know, helps.

Wordle: Once, twice, three times a headline