Hot on the heels of its ‘correction’ item that represents the mainstream media’s first genuine acknowledgement that the hysterical claims of ‘needless’ deaths were as wildly incorrect concerning Stafford hospital as they were for the 14 ‘Keogh’ hospitals falsely accused by the right-wing press of causing 13,000 avoidable deaths, comes a second, more high-profile and emphatic article on the same topic,
The Guardian appears to be now fully out of the ‘mortality’ closet.
For a long time, the newspaper repeated the mortality claims about Mid Staffs as emphatically as most other media sources, but deserves great credit now for being prepared to put its error on the record and for beginning to allow some of the voices that have been exposing the ‘avoidable deaths’ myth for some time, though of course it would be even better to see the headlines as large and prominent as those that have, in various media, propagated the falsehoods.
Yesterday’s article by Guardian health editor Sarah Boseley quotes expert statistician David Spiegelhalter at length on the subject of HSMRs, SHMIs and statistical mortality measurement generally, as well as putting a major question-mark against the integrity of Professor Brian Jarman, the supposedly-independent ‘expert’ whose pronouncements have so often been at the heart of falsely-damning media headlines.
Here are some of the key quotes. On the media’s treatment of the death claims:
Dr David Spiegelhalter, who is Winton professor for the public understanding of risk at the University of Cambridge, said newspapers and politicians are wrong in their widespread use of the number.
He gave as an example the Sunday Telegraph stating that “13,000 died needlessly at 14 worst NHS trusts”, but other papers ran similar claims.
On the blatant misrepresentation that an above-average HSMR or SHMI (hospital standardised mortality ratio/summary hospital-level mortality indicator) means any kind of avoidable-death phenomenon:
The crucial fact is that both the SHMI and HSMR are standardised to recent national performance, and so we would expect at any time that around half of all trusts would have ‘higher than expected’ mortality, just by chance variability around an average. Indeed, for the SHMI between January 2012 and December 2012, 56% of trusts (80/142) had above expected mortality.
On the phenomenon of hysterical NHS mortality claims in general, and on the ludicrous Mid Staffs claims in particular:
“So what about the ‘1,200 needless deaths’ at Mid-Staffs?” he writes. “A recent BBC news story claims: ‘Data shows there were between 400 and 1,200 more deaths than would have been expected between 2005 and 2008.’ But there are no published data that show this, as fully discussed in the first Francis report. Like the ‘1,200’ at Mid-Staffs, ‘13,000’ threatens to become a ‘zombie statistic’ – one that will not die in spite of repeated demolition.“
The first couple of statements are relatively muted and scholarly – as you’d expect, but more so than the issue deserves – but the last is incredibly forceful for an academic, and leaves no room for misunderstanding:
The claims about Mid Staffs – and NHS mortality in general – are zombies. Dead, rotten, lifeless, stinking and malevolent – and continuing to move even though they’ve been absolutely demolished.
And by the SKWAWKBOX first of all. It’s good to be shown to be right, and even better that the word is now starting to get out – but the fact that most of the media continue to use these stats as if they’re not thoroughly debunked means that there’s no room to ease up in pressing home the point.
The article also gives a fascinating insight into the mentality and motives of Professor Jarman, who has relentlessly promoted his HSMR statistical system in spite of its known and massive flaws.
Ms Boseley had an email exchange with Prof Jarman on the topic of the media abuse of his statistics in which he responded:
I am very grateful to the media, of all political persuasions, for helping us to get people to take our data seriously – it has been a long struggle for over a decade … … we have been doing no more than ask that people use the data as a trigger to look further – to use SMRs, mortality alerts, individual patients, staff and patient surveys etc. We would like to take it out of the political arena and look at the data rationally. So, no, I don’t regret the use of our data by the press and media although I would like them to be more precise – it may be my fault that I am not persistent enough in explaining a rather complicated subject
Can you imagine another academic being happy at his statistics being wrongly used, let alone grossly misrepresented? Yet the Professor is – and has done a lot more than ‘ask people to use the data as a trigger to look further’.
And more besides. The Guardian follows Jarman’s comment with Spiegelhalter’s opinion, which is already partially shown above:
But Spiegelhalter said: “The crucial fact is that both the SHMI and HSMR are standardised to recent national performance, and so we would expect at any time that around half of all trusts would have ‘higher than expected’ mortality, just by chance variability around an average. Indeed, for the SHMI between January 2012 and December 2012, 56% of trusts (80/142) had above expected mortality.
“It would be absurd to label all these as outliers, and yet a BBC news item claims that: ‘Outliers are trusts which have a higher than expected number of deaths.’ It is enough to make a statistician sob.”
A serious statistician, at least. Yet Professor Jarman is happy about it – because it’s got people taking notice of his statistical system again.
Which is extremely telling about the whole sorry mess of distortion and ulterior motivation that has led to a hospital, its staff and even its town being demonised while those who do the demonising posture as heroes.
No, no room for letting up on the pressure at all – not until this issue is finally brought out into the sunshine to be disinfected of its poison.