- Greg Wood, former naval doctor and Atos disability analyst, London
Greg Wood went to the media with concerns about the ethics surrounding tests for fitness to work—and eligibility for benefits—that the UK government outsourced to Atos
Actually, two whistleblowers went public before me, and several other doctors have raised concerns anonymously.
I am a former general practitioner in the Royal Navy, where work related assessments are bread and butter stuff. The UK Department for Work and Pensions (DWP) devised the work capability assessment (WCA) to judge whether people who receive out of work sickness benefits could, in fact, cope with most forms of work.
A more stringent test came into use in 2011, and the government made no secret of the fact that it hoped this would boost the labour market, improve people’s self esteem and personal income, and, of course, reduce government debt.
For many years the information technology and “business process outsourcing” company Atos has had a contract, now worth £100m (€116m; $155.4m) a year, to carry out several social security benefit assessments, including the WCA, for the Department for Work and Pensions. In my view this risks tension between doctors’ professional concerns on the one hand and business imperatives on the other.
The WCA had a troubled childhood. From early on, claimants and disability groups were reporting problems. They felt the assessment was a box ticking process, where medical assessors spent most of their time punching superfluous lifestyle data into the computer. And the likely outcome as they saw it? Computer says no.
In fact, the test, on paper at least, isn’t too bad, though it isn’t going to win anyone a Nobel prize. But it cannot adequately take into account health conditions that fluctuate unpredictably, and it tries to include too broad a range of jobs. Driving, call handling, shelf stacking, data entry, and cleaning, for example, are all theoretically covered. And although the test is nominally a pre-employment medical test of sorts, it is really still about measuring the person’s level of disability.
In early 2013 the WCA was still causing a rumpus in public, despite a series of external reviews.
One problem that dawned on me over time was the widespread use of five ill conceived so called rules of thumb that were promulgated during the training of new assessors.
On one, manual dexterity, the guidance was just plain wrong. The training said that this all boiled down to an inability to press a button, whereas the regulations allow points to be awarded when there are difficulties forming a pinch grip, holding a pen, or operating a computer.
The other “rules of thumb” showed a combination of discrepancies and questionable interpretations of medical knowledge—for example, moving from one room to another at home was supposed to be equivalent to moving 200 metres. The effect was to reduce a claimant’s likelihood of entitlement to financial help.
Another concern was the absence of documentary evidence, which, in my experience, occurred in about a fifth of assessments. This was a simple failure to move important pieces of paper from one building to another but the assessment was expected to go ahead regardless.
And my third concern was that there was an implicit assumption that the most likely outcome of an individual face to face assessment was that the person would be found fit for work. I have no reason to believe that this was deliberate; it was probably more a question of wishful thinking and a misunderstanding of basic statistical principles. You can’t expect the proportions of claimants who are fit to work who are seen by an individual doctor to correspond to national trends. The general culture was one where, at the point when their file was being opened for the first time by the assessor, it was broadly assumed that an individual claimant was more likely than not to be found fit for work.
My fourth concern was that Atos auditors, for quality assurance purposes, were in the habit of demanding that healthcare professionals change their reports without seeing the patients themselves. This seemed fairly reasonable if the amendment could be justified, but not so reasonable when the doctor who had seen the patient thought otherwise. For instance, auditors supposed that they could tell that a patient with a chronic and only part treated psychotic illness had adequate mental focus, despite not assessing the patient for themselves, and using solely a report.
The position of the General Medical Council is that doctors should not alter such reports if they think that it would make a report less accurate, or would render it misleading to the body commissioning it—that is, the DWP. I resigned from Atos primarily over this widespread interference with reports, which I felt encroached on my professional autonomy and crossed ethical boundaries.
So I blew the whistle and found myself talking to parliamentarians and journalists, and then making an appearance on BBC news. It was nerve wracking trying to choose my words carefully while keeping the message clear and simple. Obviously I worried about the repercussions, but what had tipped it for me was that the DWP had stonewalled on this for more than two years; medical knowledge was being twisted; misery was being heaped on people with real disabilities; and the cost to the taxpayer of these flawed assessments and the subsequent successful tribunal appeals was going up and up.
Three months after I blew the whistle, the DWP announced that all Atos assessors were to be retrained and that external auditors had been called in to improve the quality of the WCA.
To others considering blowing the whistle, I would say this: if it is important enough to you and you do not believe that the problem can be fixed by more conventional means; if you can back up your assertions with evidence; if you are prepared to risk alienating your colleagues; and if you are robust enough to deal with the slings and arrows that might come your way; then blow your whistle loud and blow it proud.
Cite this as: BMJ 2013;347:f5009
Competing interests: I have read and understood the BMJ Group policy on declaration of interests and have no relevant interests to declare.
Provenance and peer review: Not commissioned; not externally peer reviewed.