top of page

Evidence-based review of the WCA

The government has been carrying out a study of suitable descriptors to use for assessments for ESA. The results were supposed to be released in spring of this year, then summer, and then finally were released earlier this month.

The superficial results suggest that the current descriptors are better than the alternative descriptors tested in the EBR. But there are number of provisos that indicate that the current descriptors and assessment process are still very flawed. Atos assessors and members of the expert panels self-assessed as carrying out good assessments, but the data suggests poor information-gathering and decision making; the expert panels reported that the majority of those whom they considered ‘fit for work’ still faced potentially substantial barriers to work; and independent reviews of the ‘expert panel’ decision making by experts in relevant health conditions suggest that even the ‘experts’ on the panels were minimising the extent of people’s barriers to work.

Most assessors considered it very or fairly easy to choose which descriptor applies to a claimant. But when 42% are overturned at appeal, are assessors easily choosing the right descriptors? The expert panels were confident of their decision in 94% of cases, but an independent reviewer disagreed with them 38% of the time.

The assessors generally reported that they didn’t have difficulty with ‘eliciting information from claimant,’ but the majority of appeals are overturned on new ORAL evidence – evidence that SHOULD have been elicited in the original assessment. Similarly, the expert panels were largely confident that they had enough information – but if you are not presented with a piece of information then you do not know that it exists and therefore cannot know to miss it.

Expert panels said 70% were fit for work, but of these 83% were likely to need adjustments such as flexible working hours, disability leave, aids or appliances, working from home, or a support worker – so they may struggle to find a job that meets access needs, and then to convince the employer to take them on. Furthermore, the 50% who need flexible or altered hours and the 24% who need periods of disability leave may struggle to support themselves with paid work due to an inability to work sufficient hours (note the two groups are not mutually exclusive – some people may need both flexible or altered hours, and periods of disability leave).

One panel said that a need for counselling and anger management was separate from an ability to work; employees may question whether a person with anger-management issues is a safe and reliable colleague. Another panel who (presumably, based on the context) found a person fit for work then described that person as needing “significant adjustments and support,” part-time work, pacing, support and possibly also disability leave.

It is therefore questionable whether all or even many of those rated as fit for work by the expert panel have any realistic chance of finding, getting and maintaining employment at a high enough level to provide a stable liveable income. Many of these people may legitimately merit and need incapacity benefits, and more intensive work support than is available on Jobseeker’s Allowance.

The WCA missed 39% of the people identified by the expert panels as unable to work, whereas the alternative descriptors missed 19%. The WCA didn’t identify any other people as unable to work; the alternative descriptors found an additional 53% not identified by the expert panels.[i] From the point of view of those affected, the alternative system is better because people are not incorrectly told they are fit for work. From the perspective of the DWP, the WCA is better in that it saves money by not giving benefit to those who may not need it – but with the concerns given here, it may be that the alternative system is right to identify the extra 53% as unfit.

The WCA was significantly more consistent than the alternative descriptors. However, this would mean that the WCA consistently misses 38% of those unfit for work. The alternative may be better if it were improved to make it more consistent.

28 of cases were given further review; this equated to 91 appraisals by expert panels. The reviewers gave a score of 3 or more to 62% of the appraisals, where 3 represents ‘reasonably or somewhat’ appropriate. 38% were therefore inadequate, as considered by the independent reviewer.

In one case study provided, a claimant with Crohn’s Disease was found by two panels to be capable of work. The independent reviewer, who had expertise in bowel conditions, disagreed on the basis of the claimant’s difficulty with getting to a toilet on time. The case study noted that the claimant “can struggle to reach the toilet even in her own home,” which suggests that the claimant should have been considered incapable of work on grounds of incontinence. The reviewer commented that “both panels have underestimated the extreme urgency being described.”

All this evidence suggests that the WCA continues to consistently under-estimate the true extent of work incapacity. Furthermore, Atos assessors and members of the expert panel may lack insight into the extent of their under-estimation; and those with expertise in relevant conditions may be better able to make an accurate assessment of a person’s capability for work.

[i] This is based on estimates where I have re-classified some of the people identified as fit under the expert panel as unfit. This is because the DWP reports that the expert panel found 30% unfit, and 11% as borderline unfit (the text doesn’t say if this is an additional 11% or part of the 30%), but the data table shows the expert panel as finding just 20% unfit. I therefore moved 11% of the people assessed as fit by the expert panel to unfit, to account for the 11% border-line and to bring this in line with the 30% cited in the text.

Recent Posts

See All

Unsupported tropes used to cut disability benefits

One of the government’s most common tropes when it is discussing welfare is a desire to focus support on ‘the most needy’. Superficially positive, this trope actually allows governments to cut support


bottom of page