The extent of the duty to make reasonable adjustments to avoid placing a disabled person at a substantial disadvantage, including taking steps to get them back to work, is highly fact sensitive, and has been looked at by the EAT in a number of recent cases without giving a definitive answer about how effective an adjustment needs to be to qualify as a "reasonable" one to expect the employer to take.
The Equality Human Rights Commission (EHRC) Code of Practice on Employment Chapter 6: Guidance on Reasonable Adjustments indicates that the effectiveness of a measure is a factor which may be take into account in deciding whether a measure is reasonable, but just how effective does a measure have to be to be reasonable?
For example, where a proposed adjustment to reduce emphasis on communication skills in a set of redundancy selection criteria would nonetheless still not have prevented an employee with a social anxiety disorder being selected for redundancy, that adjustment was held to be "not reasonable" (Lancaster v TBWA Manchester). Likewise, from the case of Salford NHS Primary Care Trust v Smith, it seems that although consultations, trials, and exploratory investigations may lead to the making of a reasonable adjustment, because they do not directly alleviate the disadvantage the disabled person suffers they are not in themselves "reasonable adjustments" as defined in the legislation. Therefore a failure to undertake them will not, apparently, be a breach of the duty.
However, before every employer breathes a sigh of relief and thinks that they needn't bother, this conclusion - albeit undoubtedly correct as a matter of legal interpretation - should in practice be approached with extreme caution. Consultation and investigation are excellent - and, indeed, very often the only - ways of identifying whether there is an adjustment which could reasonably be made. The employer who fails to spot that may well find themselves in difficulty before a tribunal.
Given that it is accepted that it is not reasonable to expect an employer to make an adjustment which would definitely not be effective, how likely does it have to be that it will work for it to then become capable of being reasonable? Leeds Teaching Hospital NHS Trust v Foster suggests that all there need be is "a mere prospect" - which sounds like little more than a possibility that the adjustment (in this case putting an employee on a redeployment list - which did not guarantee that he would find a new job) will remove the disadvantage. An earlier (2007) case, Romec v Rudham, set the bar a little higher, suggesting that there should be a "real prospect" of the adjustment being effective.
It seems the question to ask is this: having identified a potential adjustment, would it remove the disadvantage created by a provision, requirement or practice? If this is possible, the adjustment may be reasonable, but if not, it will not be a reasonable adjustment. With all these "if's, but's and maybe's" the current state of guidance from the courts can at best be described as only moderately helpful.