Coffee House

NHS review: Where did the “13,000 deaths” figure come from?

16 July 2013

5:21 PM

16 July 2013

5:21 PM

There is a lot of rage in Westminster today (beyond the everyday anger exhibited by some of its inhabitants that Parliament contains other people who disagree with them) about the 13,000 deaths figure that has been bandied around ahead of the publication of the Keogh review. It’s worth noting firstly that Jeremy Hunt did not refer to this 13,000 in his statement to the Commons, but the figure made its way into the newspapers before the report’s publication.  He did say that ‘no statistics are perfect, but mortality rates suggest that since 2005, thousands more people may have died than would normally be expected at the 14 trusts reviewed by Sir Bruce’.

It isn’t a made-up figure, but neither is it a list of 13,000 people who have definitely ended up in a mortuary as a result of the failings at the 14 hospitals examined by Keogh. It is in fact a calculation by Professor Sir Brian Jarman of Imperial College. Jarman calculates that had the hospitals had average death rates, 13,000 deaths could have been avoided in that period (you can download his working here). That is very different to 13,000 avoidable deaths that definitely occurred. The Keogh report sounds a note of caution about hospital standardised mortality ratios and summary hospital-level mortality indicators:

‘However tempting it may be, it is clinically meaningless and academically reckless to use such statistical measures to quantify actual numbers of avoidable deaths. Robert Francis himself said, “it is in my view misleading and a potential misuse of the figures to extrapolate from them a conclusion that any particular number, or range of numbers of deaths were caused or contributed to by inadequate care”.’


That’s not to say that the Keogh report doesn’t make extremely difficult reading for the trusts involved. Whatever you can take from mortality figures, the hospitals were failing to do so: Keogh said few hospitals had a good understanding of why their rates were high. ‘This contributed to them having weak or incomplete strategies for improving performance,’ the report said.

More clinically relevant, perhaps, is the number of ‘never events’, which are serious incidents such as surgery on the wrong body part, the wrong drugs being administered, or tools being left inside a patient after an operation. Keogh found that all but two of the trusts had ‘never events’ which the report said was ‘extremely concerning’. If a never event should never have happened, it is reasonable to assume that it should at least never happen again in a hospital. But the review said:

‘Of even more concern is that a number had multiple never events relating to similar themes, such as retained foreign objects post-operation, where we were not assure that lessons had previously been learnt in response.’

You can also read the individual reports on the trusts here.

Subscribe to The Spectator today for a quality of argument not found in any other publication. Get more Spectator for less – just £12 for 12 issues.

Show comments
  • The_greyhound

    The disastrous mess that is the NHS has really been THE news story of the last year, and its not finished yet.

    What fascinates me is that the Grauniad almost never mentions the health service, but quite recently colluded with a traitor to disclose US and UK intelligence material to, among others, China and Russia. The Grauniad then bombards its readers, all wearing specially knitted tinfoil hats, with hysterical coverage of the ‘story’ including live coverage of matters as portentous as Snowden being visited by Amnesty International, and devotes acreages of self-serving crap to its shocking discovery that Intelligence Services collect intelligence.

    It couldn’t be that this is just more of the idiot comic’s diversionary tactics could it?

  • Mynydd

    I can say without a shadow of doubt that,two months before the next General Election, using the same methods as this report,there will be 14 trusts with above average death rates. Average mean some will be above and some below. The question is: will Mr Cameron/Hunt publish the figures or cover them up?

  • DavidL

    It’s really very simple. The SMRs measure the difference between the predicted and actual mortality rates. If the SMR is high, it signals that further investigation is needed. It doesn’t, in this case, mean that 13,000 extra people died, but the fact that the SMR is so high indicates that there may well have been excess deaths, and there are problems that need investigating. Without doing an exhaustive case review, it is impossible to say with any accuracy how many “unnecessary” deaths actually occurred.
    Indeed the real story here is why DH officials have for so long refused to use SMRs as a lead indicator. Using SMRs as an indicator in general practice in the 1990s would have highlighted Dr Shipman’s practice. And using SMRs would have triggered earlier intervention to investigate these 14 Trusts. In this context it was disturbing to see Professor Black, one of the Keogh team, apparently still rubbishing SMRs in interviews he gave yesterday.

  • Davidh

    Obviously, the truth is more complex than the 13,000 figure and newspapers that report that number as “excess” or “needless” deaths are sensationalizing the story to push their own agenda.

    The important thing, as the report does try to do, would be to corrolate those above average moratily rates with other variables such as staffing levels and conditions, cleaning routines, funding, management structure, local health issues etc. Then you can start to target improvements. You should also look at historical mortality rates, if available and comparable, to see if the problem has got worse over time or if perhaps all hospitals are improving but the worst ones are just not improving as much as the best and the gap is increasing. That would be too complicated for a headline, though.

    If you just want to reduce that 13,000 figure to zero, then the quickest way would be to increase the mortality rates at the best hospitals to the levels of the worst. Kill thousands more at the best hospitals, all hospitals would be the same, none would be deviant from the average, nobody would have died needlessly. Stats are wonderful…

  • David Lindsay

    Who is this workfare person on Newsnight, purporting to be the Secretary of State for Health?

  • David B

    The old saying of “lies, dam lies and statistics” rings true once again.

    The real irony is that the Labour Government created a whole raft of statistics to provide dividing lines and gain political advantage while hiding bad news is now being hung on a statistical analysis.

  • HookesLaw

    The figure of 1200 did not occur in the Mid staffs report but a figure of between 400 and 1200 has been circulated.I think the 13000 is a product of statistics and should be treated with caution but it cannot be ignored, it is an indicator of lack of of compassion and of poor care and management, as you say.
    The words of the report are as important as its numbers.

    • 2trueblue

      Care is the prime thing. If properly cared for a greater number would probably have survived.

  • Span Ows

    Well the opposition broadcasting arm (that other ‘much loved best in the world rotten institution) was in full flow today, a few lines about the story then Burnham on allowed get away with lies, also blaming the government spin, guests saying this that and the other (as any fool says, working their way down to no problem at all, carry on) to selected excerpts from parliament giving the listener with the clear and definite impression that all the problems started with this government. Unbelievable despicable bias from the BBC.

    • 2trueblue

      Good old BBC. 13,000 more people died than one would have expected, especially during a time when there was huge investment in the NHS. All this time, money, and expertise and what have we actually got? What conclusions can these experts offer? Is there anyone out there that can speak plain English? Where are the journalists who should be doing their job?

      • Teddy Groves

        No in this context ‘expected’ means roughly ‘average’. The overall investment in the nhs is irrelevant.

  • pigou_a

    So I presume our delightful (and increasingly redundant) press corps will be issuing corrections?

    “Thousands of NHS patients may have died needlessly because the last Labour
    government” – The Telegraph.

    “A damning report by Sir Bruce Keogh, the NHS’s Medical Director, is tomorrow
    expected to spell out failings at 14 hospital trusts where there have been
    up to 13,000 needless deaths since 2005.” – The Telegraph

    “Health Secretary Jeremy Hunt announced NHS ‘hit squads’ will now take
    over failing trusts, where an estimated 13,000 people may have died
    since 2005.” – Daily Mail

    Newspapers – “Whopps, sorry readers, turns out what we wrote was not true.”

    Readers – “Oh never mind, we assumed you were just writing fiction long ago.”

    Seriously, what is the point of journalists who cannot even report basic facts?

  • anyfool

    So is this the start of a concerted effort to massage the extrapolated figures down to zero.
    Good now we can get back to singing the praises of the last Labour government who did not.
    1 Ruin the economy
    2 Flood the country with immigrants as voting fodder
    3 Abolished boom and bust
    4 Never went to war to satisfy the ego of its leader
    5 No one died needlessly at all during their time in office
    6 Never really lost the election and are just resting before they can bestow even more goodies on the population.

    • HookesLaw

      We should certainly not forget any of those things, but this is not the start of an effort to massage them away.

      • 2trueblue

        No, it is not the start of an effort to massage the figures away, it is a continuation of the efforts that the BBC perform every day on behalf of Liebore.

      • alabenn

        I am afraid that Cameron will run away from a full blown attack on these incompetent fools that created the conditions that led to this carnage.
        The likes of this columnist and others in various papers are implying that it is now about over reaction.
        Cameron should lay the blame full square on Labour even if it inflames the BBC and all the journalists whose sole objective is to become a BBC in house talking head.
        It also puts into the mind of the populace over the recess that it is Labour who are the problem.

    • Colonel Mustard

      I think it is the end not the start. The absence of vehement Labour rebuttal trolls in this thread suggests that they believe the crisis to be over and the threat to Labour’s reputation as the “caring” parteh no longer of concern. The BBC did an excellent job for them and once again their crimes against English humanity will go unpunished.