Employment Update by Tom Emslie Smith – Causation, Comparison and Disability Discrimination

Published: 06/11/2018 | Employment, Newsletters

By Tom Emslie-Smith

Sheikholeslami v The University of Edinburgh UKEATS/0014/17/JW

When preparing any disability discrimination case, it is important to think carefully about what the Claimant is – and isn’t – required to show in terms of causation. In Sheikholeslami v University of Edinburgh UKEATS/0014/17/JW, the Scottish EAT underlined the importance of avoiding a “one size fits all” approach to causation. Each section of the Equality Act 2010 has its own requirements. Some have a broader test than others. Some do not require causation to be considered at all.
Of particular significance in this case was the SEAT’s judgment on section 15 EqA, which affirmed a broader test of causation in cases of disability discrimination. However the judgment on reasonable adjustments also provided a salutary reminder of the error of importing questions of causation into areas where they do not belong.

The Facts

The Claimant, a professor in chemical process engineering, was offered employment on the basis that she would be given a suitable laboratory. There were problems with the refurbishment of the laboratory, which meant that Professor Sheikholeslami was without a working laboratory for nearly two and a half years. She raised grievances about her treatment, alleging sex discrimination. This culminated in her being signed off sick with work related stress. As a result of her complaints a diversity review was undertaken which highlighted cultural problems within the workplace.

Following the review, relationships broke down between Professor Sheikholeslami and other staff members in the school. Professor Sheikholeslami wrote a letter stating that she wished to return to work via gradual reintegration through another school but the department suggested instead that she return to work at her existing refurbished laboratory. Eventually, Professor Sheikholeslami was dismissed upon the asserted basis that her work visa was due to expire. She complained that the real reason for her dismissal was “to avoid the problem of her continued and sustained complaints of gender and disability discrimination.”

Section 15 Equality Act 2010

The Tribunal found that Professor Sheikholeslami was dismissed because she was not prepared to return to work at the Engineering School, and that this triggered the problems with her visa. The evidence showed that she was willing to return to work in another faculty, and there was no medical evidence that she was incapable of working in the Engineering school. The reason for her unwillingness to return to work appeared to be her perception that she would be treated badly by her colleagues (which triggered her disability in the first place). The Tribunal did not find that this amounted to disability discrimination because it did not have the evidence to say that her dismissal was because of her disability.

The EAT held that the Tribunal was wrong in its approach to causation. Section 15 Equality Act 2010 has a wider test, requiring the Tribunal to consider whether the unfavourable treatment was in consequence of the disability.

The two stage test to be followed under section 15 is:

(i) Did A treat B unfavourably because of an (identified) something? and
(ii) Did that something arise in consequence of B’s disability?

This causation test can encompass several links in a chain of facts and more than one consequence of the disability can be taken into account. The Tribunal erred in adopting too narrow a test for causation and failing to consider other relevant consequences of the disability.

On the SEAT’s analysis, the Tribunal concluded that the Respondent had treated the Claimant unfavourably because they thought that she was not prepared to return to work. However, the Tribunal failed to contemplate the next link in the chain, and consider why she would not return to her post. On the facts of the case, it was clear to the SEAT that the disability, the perceived discriminatory treatment that gave rise to it and the Claimant’s reluctance to return to work were interlinked in such a way that could satisfy the broad section 15 test.

Reasonable Adjustments

Along similar lines, the EAT also held that the Tribunal erred in its approach to the Claimant’s reasonable adjustments claim. The Tribunal found that there was a PCP that the Claimant should return to work at the Engineering school, but held that she had failed to show that she was at a substantial disadvantage because of her disability. The EAT held that the reasonable adjustments test was not a causation question. It required a comparison exercise with people who are not disabled, and so the Tribunal was wrong to require causation relating to the substantial disadvantage and medical evidence to prove it.


Requirements for causation vary according to which section of the EqA is being applied. This is clear from the particular wording of each section. Sheikholeslami is not the first case to emphasise this, and it illustrates the error into which the Tribunal can fall if it doesn’t stick closely to the questions required by the section at hand.

Section 15 differs from section 13 (direct discrimination) in that it prohibits individuals being treated unfavourably because of something resulting from their disability; the section is not restricted to treatment that is because of the disability itself. The test can be challenging to apply because it requires advisers to look at the situation with a wide lens. Which aspects of the Claimant’s situation have arisen in consequence of his or her disability? Was the alleged unfavourable treatment because of any of these circumstances? Too narrow a focus on factors that are because of the disability itself risks missing treatment that could found a successful section 15 claim.

The case was also a reminder of the distinction between the comparison exercise to be carried out in reasonable adjustments claims and the causation exercise that might come into play under e.g. section 13. As we see, the two can lead to different conclusions, and mustn’t be confused.