ATOS Kills… or does it?

I suspect I’m going to make myself rather unpopular in some quarters with this, but I’m afraid that may commitment to good science trumps any political opinions I might have every time, so in for a penny, in for a pound.

Political Scrapbook is running yet another ‘man/woman dies after being turned down for ESA‘ story this morning, one that follows the usual tabloid narrative:

Blinded, half-paralysed man dies day after ATOS stop his benefits.

Yet another disabled person has died after being declared “fit for work” by DWP contractor Atos, despite being struck down by a crippling illness. Brian McArdle died the day after his disability benefits were stopped, suffering a fatal heart attack which relatives insist was caused by stress related to the decision.

The 57 year old Scot, was paralysed down one side of his body, blind in one eye, and couldn’t speak — yet in a bizarre but not uncommon decision Atos thought he wasn’t deserving of Employment and Support Allowance.

Figures from the Daily Mirror have revealed that 32 claimants each week — nearly 1,700 per year — have died after they were branded “fit for work”.

Criticising the previous incapacity benefits regime in March, then employment minister Chris Grayling said:

“To have such a high percentage who are fit for work just emphasises what a complete waste of human lives the current system has been.”

Grayling has moved from a metaphorical “waste of lives” to a literal one

You’ll forgive me, I’m sure, for removing Scrapbook’s rather idiosyncratic formatting, which I find jarring at the best of times, but otherwise that’s the complete story. An obviously very sick man has died only a day after receiving a letter knocking back his claim for ESA and this letter is, in the eyes of his family, at least indirectly responsible for causing his death.

But is that necessarily true?

There is an important detail missing from PS’s account. We told that this unfortunate man was disabled but not why, a detail that is included in the original story:

Kieran said that Atos had to take responsibility for his 
father’s death.

“My dad was severely disabled after he took a massive stroke which caused a blood clot in his brain,” he said.

So he’d previously had a ‘massive stroke’ which left him disabled and, also in the same story we discover that:

Despite his condition, the schoolboy said, Mr McArdle, 57, who lived in Larkhall, was asked by Atos to attend a work capability assessment – part of the Westminster Government’s drive to cut the amount of money spent on welfare benefits. Days before the meeting, the former security guard had another stroke.

So that’s at least two prior strokes one of which, at least, was major, so he was indeed a very sick man, so ill, in fact, that one would hardly be surprised to hear the sad news of his passing even if he hadn’t been subject to the not-so-tender ministrations of ATOS and a work capability assessment.

And that, of course, is the problem – the proximity of these two events, the letter relating to his benefits and his unfortunate death, is not proof of a causal relationship between the two. I don’t doubt that he found undergoing the assessment and the contents of the decision letter stressful, and stress is never a good thing when you’re in that condition, but the temporal sequence of these event is not, of itself, proof of causation and the son’s assertion that ATOS has to take responsibility for his father’s death my easily be nothing more than post hoc fallacy.

As for the ‘news’ that “32 claimants each week — nearly 1,700 per year — have died after they were branded “fit for work”.” this is misleading on at least two levels – Sorry, Laurence – one of which should be fairly evidence from the scare quotes around the words ‘fit for work’.

The ‘evidence’ for this claim comes from Penman and Sommerlad’s ‘Investigate’ column in the Daily Mirror and is set in their article as follows:

We’ve used the Freedom of Information Act to discover that, between January and August last year, 1,100 claimants died after they were put in the “work-related activity group”.

This group – which accounted for 21% of all claimants at the last count – get a lower rate of benefit for one year and are expected to go out and find work.

This compares to 5,300 deaths of people who were put in the “support group” – which accounts for 22% of claimants – for the most unwell, who get the full, no-strings benefit of up to £99.85 a week.

We don’t know how many people died after being found “fit to work”, the third group, as that information was “not available”.

But we have also found that 1,600 people died before their assessment had been completed.

This should take 13 weeks, while the claimant gets a reduced payment of up to £67.50 a week, but delays have led to claims the system is in “meltdown”.

The first thing that leaps out is this statement

We don’t know how many people died after being found “fit to work”, the third group, as that information was “not available”.

So, if the information on mortality amongst people found ‘fit for work’ is not available, how has PS come with the ‘fact’ that nearly 1,700 people per year are dying after being ‘branded’ as ‘fit for work’.

The answer, of course, is that PS has extrapolated its figures from the 1,100 deaths amongst people referred to the ‘work related activity group’ which, strictly speaking, is the ‘could be fit to do some sort of work with the right help and support’ group and not a group that the system treats as actually being fit for work.

What’s much more interesting, however, are the mortality figures for the ‘support group’ and ‘work related activity group’. We’re told that both groups are approximately the same size – 22% of claimants currently go into the support group compared to 21% into the work related activity group and yet mortality in the support group is getting on for five times higher.

This is, of course, only to be expected as the support group is for claimants with the greatest degree of disability and/or the poorest health which is exactly where you’d expect to see the highest mortality rates.

Whether one can read anything at all into the scale of the difference in mortality in these two groups is a very different matter because what we don’t have are any baseline mortality rates for the kinds of people who are being placed into these groups that pre-dates the introduction of the system and, therefore, no way of determining whether mortality rates for these groups have increased, decreased or stayed very much the same since the introduction of work capability assessments and employment and support allowance.

The ‘statistics’ quoted by the Mirror, and by Political Scrapbook are entirely devoid of meaning. All the figures actually tell us is that sick people die – which is unlikely to come as news to anyone – and nothing whatsoever about ATOS, the system as a whole, or its impact on sick and disabled people.


‘ATOS kills’ might be a good slogan but without any baseline figures to back up that claim the truth, or otherwise, cannot be established, which I personally find rather frustrating as its reduce the argument to the level of the pot calling the kettle black despite the fact that the Minister who was, until recently, responsible for system – Chris Grayling – is well known as a serial statistics abuser.

I’m not trying to defend this system, or ATOS, just pointing out the obvious – campaigning on the back of a bunch of meaningless ‘statistics’ is no way to claim the moral high ground, even when the campaign relates to a system as repellent as this one.

37 thoughts on “ATOS Kills… or does it?

  1. Pingback: Unity
  2. Pingback: BendyGirl
  3. Pingback: Virginia Moffatt
  4. Pingback: BendyGirl
  5. Pingback: Not Jane Russell
  6. Pingback: Catherine Brunton
  7. Pingback: Catherine Brunton
  8. “I doubt that he found undergoing the assessment and the contents of the decision letter stressful” You base this statement on what facts?

  9. Pingback: linda kirk
  10. Pingback: fibromyalgiaman
  11. Pingback: Woodo
  12. I am just copying here a link to a comment in the Guardian which answers your question, or add to the argument.
    Look at the 23rd of October 8,27pm.

  13. People placed in WRAG die at 2.5 times the expected rate for people of working age.  So on average, people expected to be able to return to work in the near-ish future are very unwell compared to the working population.  The issue isn’t just, if at all, about causality.  It’s about very sick people – people who die from their illnesses – being told that they can work, or are close enough to being able to work to be able to attend interviews, work placements and experience, training, skills assessments and so on.

  14. Pingback: Christos Palmer
  15. Stress is a killer, and extreme stress is one of the major consequences of finding yourself in this Kafkaesque system. Stress is also a major trigger of strokes, moreso if youve already had one. So it seems unfair to dismiss the fact the guy had a stroke just days before he knew he was due to be poked and prodded by these callous people, have to desperately try and justify his very existence yet again, and then be essentially told that he must be a scrounger and needs to get up off his backside. Its too cooincidental to dismiss out of hand, and its not the only example. Then you have the terribly desperate people who end their own lives as a direct result of interactions with this system, perhaps hard figures for that are difficult to come by but it is known that it does happen. Would you dismiss them so easily too?

    1. Well said Mc (applause)   stress is indeed a killer leads to allsorts of fatalities…  Having gone through this system myself and been found “fit to work” 3 times (appealed x2) and lost!!!   3rd time put in WRAG!!!!  I know the severity of the stress involved.. most days I could not get out of bed and just wanted to die because of the day in day out levels of stress over and over again.  Having people not believe you even though you have Specialists reports, GP reports, independent GP reports and reports from your employer as to why you took early retirement because of your failing health it all has an effect and makes your conditions a whole lot harder to cope with.
      But if we all die we won’t be a “burden” on society will we..I worked for 35 years fulltime around bringing up my 5 kids and looking after my disabled husband, ive paid my dues and now i need help.. 

  16. Pingback: Flo Fflach
  17. Pingback: Simon
  18. The 1,700 figure is a result of a FOI request by the Mirror newspaper. A further request has raised this to 10,600 – 76 people a week – wrongly found ‘fit to work’ and dying within 6 weeks of being knocked off benefits. I find the tone of this article very patronising, in that you are using simplistic arguments to question whether ATOS have any link to these deaths. Of course any cause of death would be the respective conditions on a death certificate – however there is good evidence to suggest that a significant number have had their conditions worsened to the point their deaths have been sooner than expected. The stress caused by these assessments is horrendous, the stress caused by having your entitlement refused worse – what effect do you think this has on very ill people do you think?

  19. this is horrendous, why are they getting away with it, NO gvt have ever attacked the ill & disabled like this before, CAMERON SHOULD BE HUNG,   I HOPE HE CONTRACTS THE WORSE ILLNESS ANY HUMAN COULD EVER GET, AND MAY HE ROTT IN HELL.

    1. To the person ranting about David Cameron rotting in hell………….
      ESA was introduced in 2008 – 2 years before David Cameron was elected.

      1.  However IDS has introduced a target driven culture, a stated aim of knocking at least 20% of claimants off benefits – despite fraud being 0.3%, so at least 19.7% of people legally entitled will lose out. Also there’s a ‘little’ thing called the Welfare Reform Act, which has tightened the rules and stands to decimate the ‘safety net’.

      2. boring…….labour introduced ESA but on very different terms, scameron renewed atos contract and that is when atos became the nasty piece of work they are nhow

  20. Pingback: Sam Bentley
  21. Pingback: Sam Bentley
  22. This is not ‘good science’ as good science would gather evidence from those who have survived this process and relatives of those who hadn’t, not rely on the obviously qualitative nature of real world figures and causation. It is absolutely clear if you took the time to research this that the WCA system gives wrong results and increase major stressors on claimants, perhaps you have little experience of science when carried out in the field by anthropologists, as field work does not give you neat empirical simple causations. Your idea of science is a reductive and flawed in its application and somewhat falls on the side of  privilege telling people who have survived this abusive experience to stop complaining & stop being hysterical.
    Any physical or particularly mental condition will be negatively impacted by stressors, therefore it is reasonable to say the system is bad for health, Atos kills is a short slogan putting that in emotive terms, not ideal but no one is claiming in long form that atos are committing homicide. You are using a certain degree of straw mans to make a rather prosaic point about causation which does not in any way detract from the damage the WCA is doing yet you are comfortable for your piece to bring comfort to an organisation that is involved in destroying peoples lives and health. That is not just bad science but bad writing.

  23. Pingback: Rick B
  24. Pingback: Rick B
  25. The stress of being questioned intensely and under public scrutiny while suffering from a severe illness is very real all while politicians rinse the system beyond most peoples comprehension is sick and wrong.  But those in the middle find it easier to bully the weak than deal with the real problem.  It’s an animal mentality and not a very nice one at that.  I hope you get the world you deserve and the illness to go with it.


  26. RickB:

    First, you’re not making any kind of anthropological claims here, so the qualitative methods of field anthropologists are hardly relevant to this debate and now is not the time to get into a discussion of the flaws with the standard social model, which has been a plague on social anthropology for the better part of a century – so don;t go trying to teach me to suck eggs.

    It is, I agree, well established that the WCA generates far to many inaccurate outcomes and that the system is poorly designed and fails to adequate reflect the reality of disability.

    That this is clearly stressful in not a matter of legitimate dispute nor can one dispute that stress, generally, has an adverse effect on people’s health and wellbeing – this too is well-evidenced.

    However the leap from ‘the system is stressful’ to ‘the system caused or contributed to this individual’s death’ is a leap too far in the absence of supporting clinical evidence. Temporal proximity is not sufficient to establish causality as one cannot rule chance solely on that basis and so, in order substantiate the claim that ‘Atos kills’ even as mere slogan, one needs to put in a little more effort to establish a factual basis for that claim. That’s a point I addressed at somewhat greater length in this follow-up post, which I suspect you may not have read.

    ‘Real world figures’ are not qualitative, they’re quantitative – the clue’s in the word ‘figures’ – and causation cannot be established on the basis of qualitative evidence alone, hence the aphorism ‘the plural of anecdote is not data’.

    As for physical and psychological conditions being negatively affected by stressors, your assertion is generic to point of being meaningless. The degree to which a particular stressor will adversely impact a particular, physical or mental, depends entirely on the type of stressor and the nature of condition you’re talking about.

    And, to be absolutely clear. I am neither arguing from a position of privilege – and frankly I’m getting a bit sick of people throwing in ‘privilege’ as an excuse for their inability or unwillingness to put forward a reasoned argument – nor am I suggesting that anyone should stop complaining about the system

    What I am pointing out is that campaigners will be able put forward a much more credible argument if they take the time to put a bit of solid evidence under their campaign rather than relying purely on rhetoric and anecdotes. It will take a bit of additional effort and maybe ever the assistance of a medical statistician but if, as is claimed, the system itself is having an adverse impact on claimants health this should be evident in the mortality and morbidity statistics for relevant conditions, so your starting point should be that of comparing the relevant mortality and morbidity rates for the year immediately proceeding the introduction of WCA with rates for the most recent year for which data is available to see if there any statistically significant differences and which direction those difference lie.

    That would, involve, quite a bit of number crunching were you to go for national data, however you may be able to keep the amount of data you’re dealing with down to a manageable level by using local public health data from several individual areas.

    One possibly fruitful approach would be to take a list of primary care trusts, order the list by position in the Index of multiple deprivation, and extract a list of sample areas by choosing every 10th or 15th PCT on the list. This will give a decent mix of areas ranging from the most well off to the least well off.

    Next choose a subset of medical condition, preferably ones where its known that there have been issues with the accuracy of assessments, and obtain the relevant data for those conditions for each area for the years you’re looking to compare. At the very least you should obtain the mortality data but I’d also go for data on hospital admissions and on re-admiisions on mental health grounds. I’d also try and get the monthly or quarterly data, in additional to the annual data. to control for any seasonal variations and related sources of confounding.

    I would also make a point of targeting the areas in which WCA was piloted as these may provide additional trend data,

    Put all that together and, if any significant differences are found, and particularly if there is evidence of increase in mortality or hospital admissions then you’ll a damn sight stronger argument that you will ever manage to generate from mere sloganeering.

    Now, do you understand where I’m coming from on this?

  27. No I don’t understand where you are coming from, except you are preaching in a at the least a very patronising tone, and on the other side you are talking like a machine, no use when dealing with real people.  I would advise other people to not keep bothering replying/responding to this person as I think they are coming from a very different place and there will no agreement or reasoning, talks like an Aspergers Syndrome person, a person with little or no empathy and just keeps repeating the same “facts/words” over and over again(yes, I do have experience and knowledge of Aspergers)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.