Non-fiction‎ > ‎

Reforming the Tests for Illness and Disability in UK Benefits Law

This essay was my entry for the 2010 Bar Council Law Reform Essay Prize.  It did not win.  I hope nonetheless that someone may find it interesting.

Social security law reform is once again at the forefront of the political agenda. The government hopes that by simplifying its structure and introducing a universal credit, it will be able to produce long term savings and make the system fairer. This discussion will focus on how this might be achieved in relation to sickness benefits, and in particular the metrics by which sickness is measured within the context of UK benefits law. It will do so with particular focus on Disability Living Allowance (DLA), Employment and Support Allowance (ESA), and its predecessor Incapacity Benefit (IB), although the issues discussed here may also affect other benefits.

According to a 2009 study by the Institute for Fiscal Studies, expenditure on sickness benefits per claimant in 2007/8 was eight times that for unemployed claimants.[1] Some of this variation may be explained by difference in the length of a claim or the level of benefit paid, but the need to assess the health of a claimant is nonetheless significant in rendering sickness benefits more expensive than unemployment benefits. It is relatively easy to determine whether someone is in work or not. Determining whether they meet a certain threshold of sickness is a much harder proposition.

Clearly a threshold of sickness is necessary. A finite amount of money must be spent in such a way as to make sure that those in most need are cared for. Claimants of DLA are often disappointed to discover that it offers no financial assistance in respect of their inability to do housework, gardening, and shopping - such care and attention needs are not in relation to a bodily function or necessarily performed in the presence of a claimant.[2] Although these needs are often genuine, public policy dictates that the numbers eligible for assistance should be kept to an affordable level. This is one of the ways in which the scope of benefits is limited in order to achieve this aim.

The current practice is therefore that the government collects evidence from the claimant, its own medical examiners, and (sometimes) the claimant’s own clinicians. On the basis of this evidence, a government employee makes a decision on entitlement, which can then be appealed to the independent Tribunals Service. Appeals from the First Tier Tribunal on points are law are made to the Upper Tribunal (formerly the Social Security Commissioners) whence the majority of case law in this field originates.

How then does the state measure need? Two broad categories of approach are present in UK social security legislation. These can be categorised as broad functional and deconstructive analytical approaches.

Disability Living Allowance was introduced by the Social Security Contributions and Benefits Act 1992, and adopted a broad test related to two factors; the need for care delivered by another person, and the need for supervision or assistance when walking out of doors. This functional approach proved problematic in that, while it was not necessary for a specific medical diagnosis to exist in order for a disability to exist,[3] in practice the best evidence of disability available was often medical in nature and the tribunals rightly took this evidence into account when resolving disputes. In practice eligibility for DLA is determined largely on the basis of consistency of the medical evidence available with the “witness statement” provided by the claimant on the claim form.

When the government introduced Incapacity Benefit in the Social Security (Incapacity for Work) Act 1994, they adopted a different approach to the measurement of capability. The test for unfitness to work broke down the question into a list of physical and mental descriptors, each reflecting the ability to perform specific tasks related to a particular category, such as standing up, or coping with change. Each descriptor had a score associated with it and depending upon the total points scored, the claimant was declared fit or unfit for work. These descriptors are considered independently of each other and are then aggregated to give an overall impression of the severity of the claimant’s dysfunction.[4]

In 2008, following the Welfare Reform Act 2007, Incapacity Benefit was replaced by Employment and Support Allowance. This introduced different descriptors under a broadly similar structure. Entitlement to ESA is initially determined on the basis of a GP provided sick note. After thirteen weeks, the DWP makes its own medical assessment of the claimant and then applies the Limited Capability for Work Assessment (LCWA) to determine continuing eligibility.

Case law has established several important principles regarding the use of both general and specific descriptors. Of particular importance is Moyna v Secretary of State for Work and Pensions, which lays down the principle of health descriptors in legislation as “thought experiments”, considering the claimant’s typical capabilities in the abstract rather than whether they have an actual need to perform the activities listed in the descriptors. [5]

The law does not permit the interpretation of these descriptors to be overly prescriptive. It was at one point thought that the ability to walk more than fifty metres was sufficient to demonstrate that a claimant was not “virtually unable to walk” (and thus ineligible for the high rate mobility component of DLA), but case law has indicated that this naïve approach is insufficient, and any decision should be made taking into account gait, speed, distance and pain.[6]

The descriptors for DLA are general in nature, and have been clarified by case law. By contrast, the ESA descriptors are highly specific and intended to be used in concert to give an overall arithmetical descriptor of incapacity. This approach has some fundamental shortcomings. The points allocated to each descriptor are defined in legislation and if there is a rationale for the scores given to each activity, then it is well hidden. Consider descriptor 9 of the LCWA, where a limitation to the visual field of between 25% and 50% and inability to see well enough to recognise a friend are regarded as having an equal effect on the claimant’s ability to work (each scoring 6 points). Are these two things really equally debilitating? Are they more or less debilitating than the inability to stand up for more than 30 minutes are a time (which also scores 6 points under descriptor 2)? The principle of parliamentary sovereignty dictates that questions of this nature are ultra vires for tribunals.

What is the purpose of such an artificial construction for measuring ability to work? The answer to this question may lie in the necessity of implementing a procedure to assess disability. Although the criteria for DLA are much broader in nature than those for ESA, the application form filled in by the claimant is not significantly shorter. As a case in point, in order to determine whether a person’s personal care needs are to be considered significant or frequent (often the discriminating factor in respect of low rate versus mid rate care component DLA) it is necessary to determine precisely how much care the claimant needs.

The approach adopted by the DWP is to ask about many of the common activities that have been demonstrated to amount to the legislatively defined criteria, along with questions about how often and for how long each of these needs occurred. Claimants generally find this confusing as their needs tend to fluctuate, rather than observing routine patterns or consistent duration.

Even when the DWP are demonstrated to be taking an overly analytical approach to the plain English words of the DLA legislation their response may merely be to refine their analysis. After R(DLA) 4/03,[7] their response to having interpreted “virtually unable to walk” too rigidly in terms of distance was to introduce new questions on the application form to measure pain, distance, gait and walking speed. These questions are often problematic for claimants who may know that they are in pain when they walk, but do not normally think about distance in terms other than the distance from their home to the end of the road.

By contrast the approach in ESA law is based on factors that are ostensibly more measurable by their specificity. This enabled the DWP to create medical reports for claimants that purported to assess these factors directly. However, eliciting many of these pieces of information still requires asking the claimants rather than direct observation. If a claimant is observed sitting apparently comfortably during a twenty minute examination, it does not follow logically that the same claimant is able to sit comfortably for half an hour.

Increasingly the DWP has come to rely on such examinations, which are currently outsourced and created using a computerised system consisting of multiple choice descriptors selected from a list by the examining professional. In practice claimants are frequently confused by the reports that they receive, which appear to bear little resemblance to their condition as they perceive or have described it.[8] It is of particular concern that reports in relation to one sickness benefit are being used as evidence in relation to a benefit with completely different eligibility criteria – an approach which has been criticised by the Upper Tribunal.[9]

The descriptors create a test of capability that attempts to keep costs down by reducing the claimant’s condition to what is in effect a list of tick boxes. Unfortunately such a test fails to consider the truly relevant question for any claimant - is the claimant’s health status sufficiently severe as to present a practical barrier to their ability to work? This assertion is borne out both by the 128% quarter on quarter increase in ESA and IB appeal caseload (to 45,400 appeals) for Q1 2010/11[10] and by 42% of ESA and IB appeals in Q4 2009/10 being decided in favour of the claimants (as compared to 17% of JSA appeals).[11] Similar quality medical evidence has been ruled insufficient when provided by the claimant.[12]

It appears that the cost of administering benefits is merely shifting from the DWP to the Tribunals Service. The additional cost in human terms is in the considerable distress and uncertainty experienced those being denied benefit and appealing - a lengthy process that is costly to the State. Not all appeals will involve misapplication of a descriptor, but there are doubtless many cases where an incorrect decision in a medical report will remain unchallenged by a claimant out of fear or lack of understanding.

The legislation as it stands can also fail to produce equitable outcomes in the case of conditions of fluctuating seriousness. R(IB) 2/99 indicates that where a claimant’s health varies from day to day, the law favours a numerical approach based not on fitness for work on any given day, but on the frequency of bad days, and length of periods of ill health, and of intervening periods of good health.[13] This is indicative of the dangers of taking an overly literal interpretation of incapacity in line with the descriptor based approach laid down by legislation, rather than the underlying issue, namely whether the claimant is capable of work.

Consider the example of a claimant with fluctuating back pain who is laid up in bed for an average of two days a week, although he is relatively able to cope on the remaining days. He is not, however, able to predict when in the week these two days will occur. Very few jobs would allow this person the flexibility to work on those days where he felt well, and yet the law leads us to the conclusion that as he is not incapacitated for most of the time, he cannot be legally regarded as incapable of work.

The DWP has increasingly come to rely on the medical evidence provided by its own practitioners over that provided by the claimant’s own clinicians. Often a decision will be made on their own evidence without any attempt to elicit an opinion from the claimant’s own GP or specialist. There is however evidence to suggest that physicians’ own medical assessments correlate well with entitlement to health benefits when claimants are helped to apply by experts familiar with the legal basis for entitlement.[14]

This trend is concerning in that evidence from a single examination, sometimes in relation to a different benefit, is being preferred to that from an on-going therapeutic relationship. It is usually the case that chronic or long term conditions are being assessed in this manner. An examiner can only obtain a snapshot picture and cannot gain the same longitudinal insight into a claimant’s condition as the claimant’s own physicians. Case law supports the idea that evidence should be assessed on its individual merits and no blanket assumptions made about the quality of a particular type of evidence at tribunal,[15] but the DWP is not obliged to follow this ruling within its internal procedures, suggesting one possible reason for the high numbers of appeals . The DWP appears to prefer its own evidence, perhaps due to concerns as to the patient’s own clinicians being biased or unwilling to challenge insistent patients. There does not appear to be any evidence to support such a conclusion, and if the DWP is in fact doing this it shows little respect for the integrity of highly trained and dedicated professionals.

From the proceeding discussion, it is clear that the process of making eligibility decisions in respect of sickness benefits is essentially an inquisitorial and sometimes adversarial one. There is evidence from three potential sources: the claimant, the claimant’s clinicians, and the state appointed clinicians. This evidence is considered and a verdict on capability delivered. A decision that is clearly legal in nature is being made by the DWP, whose impartiality is questionable in light of their responsibility for paying for the awards.

This state of affairs may be in violation of Article 6 of the European Convention on Human Rights (the right to a fair trial). [16] Justice may arguably be available via an appeal, but expecting the sick, many of whom are suffering from anxiety and stress related conditions, to wait many months for resolution is hardly reasonable.

The assessment systems as currently laid down by law have two principal issues. In general they are designed to measure something other than the circumstances which the benefits are intended to compensate them for, and the increasingly deconstructive approach to assessing eligibility is leading to increased error. This in return renders these benefits more costly and prevents them accomplishing their goal of aiding the needy.

There are two changes that may improve this situation. Firstly the DWP should elicit brief factual summaries of medical information from claimants GPs indicating the nature and severity of their medical problems. These should be used in accordance with the claimants own description of their difficulties to make a decision regarding consistency. If diagnosis is consistent with the level of disability reported by the claimant, an award should be made. This would make the decision process simpler and cheaper, as many cases will be sufficiently straightforward that an award can be made without the need to obtain further evidence, or the expense of obtaining extra medical advice.

In less simple cases, the DWP can commission their own medical opinion, and refer the case to an independent decision making body. Decisions can still be made by a single decision maker with appropriate legal and/or medical training, but such a decision maker would have no reason to prefer the evidence of one party over another other than the quality of the evidence itself. The cost per case of such an arrangement may be higher than the current internal process. However fewer cases would be referred and such a body would be likely to make fewer incorrect assessments, meaning administrative savings would follow both within the DWP and the Tribunals Service. Referring some randomly sampled non contentious cases may also provide valuable feedback to the DWP on the quality of their decision making.

Secondly eligibility should be based upon broad categories related to the purpose of the benefit, rather than artificial constructions created to make assessment easier. Such constructions may be necessary during the evidence gathering process, but such criteria would allow case law to refine decision making in light of the goal of the benefit, and minimise the possibility of injustices such as those with fluctuating conditions being declared capable of work despite the practical impossibility of working.

Adopting these two principles in social security legislation would result in a system that was both more affordable, and less prone to miscarriages of justice.



[1] P Levell and Others, A Survey of the UK Benefits System (Institute for Fiscal Studies, London, 2009) 7 <>.

[2] R v National Insurance Commissioner, ex p Secretary of State for Social Services [1981] 1 WLR 1017 (AC) 1022

[3] R(DLA) 3/06 43

[4] R(IB) 2/03 7, but this also holds true for ESA (GS v Secretary of State for Work and Pensions [2010] UKUT 244 (AAC) 11).

[5] [2003] UKHL 44 [2003] 1 WLR 1929 (HL) 17, which has wider applicability to both DLA and ESA (R(DLA) 5/05 2 and DW v Secretary of State for Work and Pensions [2010] UKUT 245 (AAC) 5 respectively).

[6] R(DLA) 4/03 24

[7] Ibid.

[8] Anecdotally, the author has seen one report which indicated that a claimant reported experiencing frequent loss of bladder control, stated that no medical examination in respect of continence was carried out, yet nonetheless concluded that the claimant’s incontinence was mild and infrequent. This is sadly not atypical.

[9] LD v Secretary of State for Work and Pensions [2009] UKUT 208 (AAC) 10

[10] Ministry of Justice Quarterly Statistics for the Tribunals Service 1st quarter 2010-11 (Ministry of Justice, London, 2010) 3 <>.

[11] Ministry of Justice, Quarterly Statistics for the Tribunals Service, 4th quarter 2009-10 (Ministry of Justice, London, 2010) 7 <>.

[12] IM v Secretary of State for Work and Pensions [2010] UKUT 110 (AAC) 6.

[13] At paragraph 16.

[14] D Memel and others ‘Prediction of successful application for disability benefits using the Health Assessment Questionnaire for people with arthritis’ (2002) 41 Rheumatology 100-102.

[15] R(DLA) 3/99 11

[16] The ECHR was determined to have jurisdiction over social security law in Deumeland v Germany (App no 9384/81) (1986) 8 EHRR 462.