>Increasing numbers of people previously deemed medically unfit to work are being taken off state benefits after assessments by a doctor.

Margaret McCartney investigates the ethics and fairness of the system

A medical degree can open all sorts of doors. One is the brown and tinted glass entrance of the Atos Healthcare offices in Glasgow, where I checked that my name was on the list held by the security guard at the desk. This recruitment evening was for doctors, nurses, and physiotherapists interested in working with this French information technology firm, which is subcontracted to the Department for Work and Pensions to provide work capability assessments.

In November last year Atos announced a three year extension to its contract with the department, worth £300m (€350m; $480m), to “support the UK government’s welfare reform agenda.” Atos is the sole contractor, and the medical reports it generates are used to make decisions about eligibility for employment and support allowance. This benefit, which has been replacing incapacity benefit and income support since 2008, is paid to people who are medically unfit to work because of illness or disability. The weekly allowance, once the claim has been verified with an assessment of capability, is worth up to £96.85. The government estimates that 2.5 million UK citizens receive sickness benefits at an annual cost of around £12.6bn to the taxpayer.

A quick glance at internet discussion forums suggests widespread dissatisfaction from people who have been assessed. The adverts for Atos, however, consist of smiling, badged professionals saying, “Getting home on time has become part of my daily routine.” The lack of on-call duties and the 9-5 office hours were also the major advantage plugged at the evening, where nurses and doctors working for Atos helped to promote joining the company.

Isolated View

But what are the ethical issues in performing disability assessments in this way, separate from the NHS and without access to patients’ full medical records? Atos was awarded the assessment contract in 2005 and claims that its reports are “evidence based, clearly presented, legible and fully justified.” Are medical assessments accurate enough to make major decisions about people’s ability to work? And is Atos the best company to do them?

The message from the recruitment evening was quite clear. We were told: “You are not in a typical caring role. This isn’t about diagnosing.” And: “We don’t call them patients . . . We call them claimants.” Training is provided for each type of benefit examination. Its length, we were told, depends on experience but is generally up to five days of classroom training, followed by sessions accompanied by a trainer that are audited afterwards.

Full time doctors can earn £54 000 as basic salary plus various benefits including private healthcare. Sessional doctors work a minimum of four sessions a week and are paid “per item”— £35.16 for an incapacity benefit examination and £51.37 for non-domiciliary disability living allowance (DLA) examination, for example. The application forms for sessional doctors state that “10 DLA domiciliary visits cases per week would earn £40 211.60 per annum. Five LCWRA/ LCW [limited capacity for work related activity/limited capacity for work] cases per session, for six sessions per week, would earn £62 883.60 per annum.”

Throughput is a clear focus. The average morning or afternoon session should consist of five assessments, and it was made clear at the recruitment evening that clinicians who did not achieve this regularly would be picked up quickly on audit trails and speed of work addressed.

Nurses and physiotherapists do effectively the same job as doctors in the centres, but do not see people with neurological conditions such as stroke or multiple sclerosis. Otherwise people are seen on a first come, first served basis. One nurse in the audience asked about training in mental health, as she had had little training in this area and would not feel competent to assess it in a fitness for work setting. The reply was that health professionals were “very thoroughly assessed” at interview for their abilities; however, general nurses were often taken on and given training. Is a relatively short training course thereafter enough to ensure the assessments are medically accurate and fair?

Duty of Care

Atos chose not to be interviewed by the BMJ, although the Department for Work and Pensions referred me to the organisation for questions about recruitment, training, and audit that it couldn’t answer.
However, from the recruitment evening it was clear that the medical examination consisted of a computerised form to be filled in by choosing drop down statements and justifying them. For example, you could say “able to walk with ease” if you witnessed this or the patient told you this.
The professional role of the doctor is very different from that in the typical NHS. Paul Nicholson, chair of the BMA Occupational Medicine Committee, says that working in this environment brings specific difficulties. “Notwithstanding a contractual obligation to provide a report to a government department, I still have a professional duty of care to the patient and to make the care of the patient my first concern.”

The Faculty of Occupational Medicine publication Good Occupational Medical Practice reinforces the General Medical Council’s position that, good medical care “must include adequately assessing the patient’s conditions, taking account of the history (including the symptoms, and psychological and social factors), the patient’s views, and where necessary examining the patient; providing or arranging advice, investigations or treatment where necessary; referring a patient to another practitioner, when this is in the patient’s best interests.”
In other words, it expects doctors to adhere to the same professional conduct as they would in any other role.

Does it Work?

So is the current method of assessment fit for purpose? There is a queue of people who think not. The Citizens Advice Bureau Scotland, in a report last year, noted “the vast majority of complaints and advice queries stem from the Work Capability Assessment.” This includes both the result of the assessment as well as the manner in which it was carried out, it said. “Citizens Advice Scotland (CAS) is extremely concerned that many clients are being found fit for work in their Work Capability Assessment despite often having severe illnesses and/or disabilities. Our evidence has highlighted the cases of many clients with serious health conditions who have been found fit for work, including those with Parkinson’s disease, multiple sclerosis, terminal cancer, bipolar disorder, heart failure, strokes, severe depression, and agoraphobia.”

The report noted that under the previous incapacity benefit system, 37% of claimants were being found fit for work, whereas work capability assessments are finding 66% fit to work. It also gave examples where advice from people’s general practitioners that they were not fit to work was disregarded. The report found that clients often “felt hurried in their assessment and that the healthcare professional was ignoring the answers they were providing to the questions in the assessment. There was a general feeling that the assessor made little eye contact with the claimant and spent most of the assessment entering information into their laptop.” This tallies with the recruitment evening, when it was made clear that efficiency with entering details into the computer system was a stipulation of employment.
The Department for Work and Pensions says, in response, that “It’s unfair to suggest that the system isn’t working.” However, published statistics of the 600 000 new claims for employment and support allowance from October 2008 to May 2010, show that 39% were assessed as fit for work. Around a third of these people appealed this decision, 40% of whom won. The department says, “If a decision is overturned at appeal, it does not necessarily mean that the original decision was inaccurate—often, customers produce new evidence to their appeal.” However, this doesn’t really deal with the problem that the healthcare professionals doing the assessments are not, therefore, forwarding sufficient evidence to enable reliable decisions.
At the meeting, I asked how it was possible to know the variation in symptoms that a patient may have during a one-off assessment. I was told that this could be “difficult” but this “wasn’t an occupational health service.” Instead, it was a “functional assessment.” Despite being told that the decision to award benefits or not rested with a layperson within the work and pensions department, the reports were often referred to as “judgments” where making decisions may be “not always clear cut . . . but as long as you can say, on balance.” For example, you could watch someone walking into the room and ask how he or she got to the assessment centre in order to judge mobility. One assessment, which initially reported a woman was fit to work, reported as evidence that her mental health was reasonable that she “did not appear to be trembling . . . sweating . . . or make rocking movements.”

Other countries rely more on general practitioner assessments of fitness to work. For example, in the Republic of Ireland, general practitioners mainly certify patients and a doctor acting for the Department of Social Protection reviews around 12% of cases as a control mechanism. In Sweden, a certificate from your general practitioner does not automatically entitle you to sickness pay. However, the certificate asks for a description of how illness affects work capacity and a time frame. Just 3.5% of claims are turned down, and about 10% of these are overturned at appeal.

So what is the optimal system? Malcolm Harrington, emeritus professor of occupational health at the University of Birmingham, reviewed the work capability assessment system last year and is currently writing a second review. He wrote: “The pathway for the claimant through Jobcentre Plus is impersonal, mechanistic and lacking in clarity. The assessment of work capability undertaken for the DWP by Atos Healthcare suffers from similar procedural problems. In addition, some conditions are more subjective and evidently more difficult to assess. As a result some of the descriptors may not adequately reflect the full impact of such conditions on the individual’s capability for work. The final decision on assigning the claimant to one of the three categories theoretically rests with the Decision Maker at Jobcentre Plus but, in practice, the Atos assessment dominates the whole procedure. This imbalance needs correcting.”

His view is that “the Jobcentre Plus Decision Makers do not in practice make decisions, but instead they typically ‘rubber stamp’ the advice provided through the Atos assessment.” The Citizens Advice Bureau shares this view and says it wants “better accuracy” in reports. But how can this be achieved when funding is devolved to Atos with no routine access to detailed specialist or general practice based information and opinion? Professor Harrington says this “results in the Atos assessment driving the whole process, rather than being seen in its proper context as part of the process.” And so, as things stand, how sure can Atos doctors be that they doing their professional duty?

This article, written by Margaret McCartney, a Glasgow based GP, first appeared in the British Medical Journal on 2 February 2011.