Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Hospital trusts are 'fiddling' patient mortality data, claims report


The way hospitals record patient deaths could be covering up poor treatment and costing lives, claims a new report from data analysts.

Figures show a dramatic rise in the number of people recorded as needing “palliative care” at the end of their lives, with some hospitals saying more than 35% of their patients die this way.

Healthcare analysts Dr Foster claim in their report that hospitals could be hiding the fact patients were admitted for treatment which then failed.

Some hospitals may also be “fiddling” the figures to make their death rates appear better than they actually are, they said.

Data from across England in 2012-13 showed 36,425 deaths were coded as palliative – meaning people received some sort of care to relieve suffering at the end of their lives.

This was 17.3% of the total number of deaths and is almost double the 9.1% (2,1130) recorded as needing palliative care in 2008. In 2006, just 3.3% of deaths were palliative.

Some hospital trusts have massively outstripped this national rise. Out of 142 hospital trusts in the report, almost half (60) recorded the number of people needing palliative care in 2012 as higher than the national average.

Most of the 20 trusts with the biggest leaps in their palliative care coding have also significantly improved their performance on death rates between 2008 and 2012.

Palliative care deaths are not included in the hospital standardised mortality ratio (HSMR), which compares the expected rate of death in a hospital with the actual rate of death.

Trusts that code deaths as palliative effectively “remove” deaths from being included in the HSMR.

Professor Sir Brian Jarman, from Imperial College London, developed the HSMR. He told the public inquiry into Mid Staffordshire Foundation Trust that some trusts increased their coding of palliative care deaths to reduce overall death rates.

He said: “I am concerned that trusts know that increasing their palliative care rates assists their HSMR and am concerned that this could be the reason they are doing it.”

“The end result is that the variation in coding may disguise poor outcomes”

Roger Taylor

Roger Taylor, director of research at Dr Foster, said some trusts had changed the way they recorded deaths as a means of improving monitoring of palliative care.

But he said there were “real concerns around the gaming of indicators”, adding: “Whether or not you are doing it deliberately, the end result is that the variation in coding may disguise poor outcomes.”

He said “imprecise rules” governing how hospitals code deaths was harming patient care, may distort death rates and needed urgent review.

“The rules are too vague,” he said. “Poor quality data is harming patients because you can’t see where things are going wrong and you can’t see where there are issues.”

Tory MP Charlotte Leslie, a member of the House of Commons health committee, said: “This is very worrying data.

“We already know that at the time of the Mid Staffs scandals, hospitals were fiddling the figures to massage their mortality rates.”


Readers' comments (22)

  • Tip of the iceberg...

    Unsuitable or offensive? Report this comment

  • Well! I am surprised, not.

    Unsuitable or offensive? Report this comment

  • michael stone

    There are issues over EoL deaths - for example, the NHS has got an objective to identify 'final year of life patients', as part of ACP. But if you ask the statistics folks 'what are the figures, then' they currently tend to not exist - the statistics they do have, are very limited.

    If the 'rule' were that on admission (i.e. within 24 hours of admission) any 'dying phase' patients had to be recorded as such (and if not so recorded, the patient was considered as not 'dying phase' [using the terminology here, 'not palliative']), then subsequent fiddling wouldn't be easily possible, would it ?

    I have some issues around the word 'palliative' - it can lack clarity of meaning, too often.

    Unsuitable or offensive? Report this comment

  • tinkerbell

    no longer surprised by any of this sadly. It just keeps going on and on and ...................on.

    Unsuitable or offensive? Report this comment

  • Good logo, Tinkerbell, looks professional, businesslike and assertive. well done.

    Unsuitable or offensive? Report this comment

  • tinkerbell

    Anonymous | 28-Mar-2014 7:07 pm

    thanks, it's from the NHS rally.

    Unsuitable or offensive? Report this comment

  • tinkerbell | 28-Mar-2014 9:59 pm

    'impressive' would have been a better adjective than good. the best words and phrases do not always come to mind when posting. an edit facility would be useful as would voting and reply options.

    Unsuitable or offensive? Report this comment

  • tinkerbell

    the caption on the peanut bowl reads 'for well behaved monkeys' re: frozen paycut.

    Unsuitable or offensive? Report this comment

  • what is the point of meaningful data collection and statistics if it is not standardised across the NHS? yet another tragic story contributing to the failing health services.

    Unsuitable or offensive? Report this comment

  • tinkerbell

    Anonymous | 28-Mar-2014 10:54 pm

    sorry have now removed the SAVE OUR NHS avatar and replaced with the latest load of old tosh to garner our vote at the next election. Unbelievable!

    Unsuitable or offensive? Report this comment

Show 102050results per page

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.