Dobson v North Cumbria Integrated Care NHS Foundation Trust (Sex Discrimination)

The claimant, Ms Dobson, was employed by North Cumbria Integrated Care NHS Foundation Trust (the "Trust") as a community nurse. Community nursing in the NHS is a profession whose workforce is overwhelmingly female. The claimant brought proceedings before the Employment Tribunal alleging that a provision, criterion, or practice ("PCP") applied by the Trust in the organisation of working hours and on-call rotas placed female employees at a particular disadvantage when compared with male employees, contrary to section 19 of the Equality Act 2010. The claim arose against a background of long-standing concern within NHS workforce management about the interaction between rostering practices and the disproportionate childcare and caring responsibilities borne by female community nurses.

The Trust had introduced a revised rostering framework requiring community nurses to commit to extended and unpredictable on-call availability across a seven-day operational window. The claimant contended that this requirement, while facially neutral, bore more heavily upon women because statistical and sociological evidence consistently demonstrated that women in the relevant occupational cohort shoulder a substantially greater share of domestic and caring responsibilities than their male counterparts. The claimant further argued that the Trust had failed to consider or implement less discriminatory alternatives, including part-time rotas, job-sharing arrangements, and capped on-call obligations.

At first instance, the Employment Tribunal dismissed the claim. The Tribunal accepted that the PCP existed and that the claimant had established group disadvantage by reference to workforce statistics and expert evidence. However, the Tribunal found that the Trust had objectively justified the PCP under section 19(2)(d) of the 2010 Act on the grounds that the operational demands of delivering safe, continuous community healthcare across a rural geography constituted a legitimate aim pursued by proportionate means. The Tribunal placed particular weight on the Trust's evidence that patient safety in remote areas of Cumbria was critically dependent upon an uninterrupted on-call capacity and that no less discriminatory alternative had been shown to be operationally viable.

The claimant appealed to the Employment Appeal Tribunal on two principal grounds: first, that the Employment Tribunal had applied an insufficiently rigorous proportionality analysis and had failed to engage adequately with the less discriminatory alternatives put forward by the claimant; and second, that the Tribunal had erred in its approach to individual disadvantage, conflating the question of group disadvantage with the separate question of whether the claimant herself had suffered a particular disadvantage as required by the statutory framework.

The Trust cross-appealed on a narrower point, contending that even if the Tribunal's proportionality analysis were found to be defective, the case should be remitted to the same Tribunal with guidance rather than to a freshly constituted panel, on the basis that the factual findings underpinning the legitimate aim assessment were not in dispute and required no fresh evaluation of oral evidence.

The primary issue before the Employment Appeal Tribunal was whether the Employment Tribunal had correctly applied the test for objective justification under section 19(2)(d) of the Equality Act 2010, and in particular whether the Tribunal had discharged its obligation to scrutinise rigorously the proportionality of the PCP by considering whether the discriminatory effect could have been avoided or reduced by less onerous means. The EAT was required to determine whether errors of law infected the proportionality analysis or whether the criticisms advanced by the claimant amounted to no more than a disagreement with permissible findings of fact.

The secondary issue concerned the correct analytical approach to individual disadvantage within an indirect discrimination claim. The EAT considered whether the Tribunal had erred in law by treating the establishment of group disadvantage as sufficient to satisfy the individual disadvantage limb of the test, or whether it had in substance addressed both elements, rendering any error immaterial.

The subsidiary cross-appeal raised the procedural question of the appropriate remedy in the event that a legal error was identified: whether remittal to the same or a differently constituted Tribunal was the correct order, having regard to the principles governing remittal in EAT proceedings and the overriding objective of dealing with cases justly and at proportionate cost.

The EAT commenced its analysis by reaffirming the structure of the indirect discrimination inquiry under section 19 of the Equality Act 2010. The Tribunal emphasised that the four-stage analysis — identification of the PCP, the application of the PCP to both groups, group disadvantage, and individual disadvantage — must be conducted sequentially and with precision. It noted that the conflation of any stage with another is an error of law capable of vitiating the overall assessment, drawing upon the analysis in Essop v Home Office (UK Border Agency) [2017] UKSC 27; [2017] 1 WLR 1343, in which the Supreme Court affirmed that indirect discrimination does not require an explanation for why a particular PCP disadvantages the protected group, only proof that it does.

On the question of group disadvantage, the EAT upheld the Employment Tribunal's finding that the rostering PCP placed women at a particular disadvantage. The EAT considered the approach mandated by Essop to be plainly satisfied: workforce data demonstrated that women comprised the overwhelming majority of the affected cohort and that, within that cohort, women were disproportionately likely to have primary caring responsibilities that were incompatible with unpredictable extended on-call working. The EAT declined to disturb this factual finding, confirming that statistical evidence of disparate impact, taken together with sociological evidence regarding the gendered distribution of caring obligations, was sufficient to discharge the claimant's burden at this stage.

Turning to individual disadvantage, the EAT identified a material legal error in the Employment Tribunal's reasoning. The Tribunal below had, in the EAT's analysis, moved directly from establishing group disadvantage to the justification stage without expressly determining whether the claimant herself had suffered a particular disadvantage by reason of the PCP. The EAT confirmed, consistent with Dobson being heard against the backdrop of Homer v Chief Constable of West Yorkshire Police [2012] UKSC 15; [2012] ICR 704, that the individual disadvantage inquiry is a distinct legal requirement. It is not satisfied merely by membership of the disadvantaged group; the claimant must demonstrate that the PCP operated to her personal detriment. The EAT found, however, that the evidence before the Tribunal was sufficient to support a finding of individual disadvantage as a matter of inference and that the error, while real, was not in this instance fatal to the outcome on that limb of the claim.

The EAT then turned to the central and determinative issue: the proportionality analysis within objective justification. The EAT set out the applicable legal framework, confirming that an employer must demonstrate, first, that the aim pursued is legitimate and, second, that the means adopted are proportionate in the sense of being both appropriate to and reasonably necessary for the achievement of that aim. The EAT cited the formulation adopted in Bilka-Kaufhaus GmbH v Weber von Hartz (Case 170/84) [1987] ICR 110 as the foundational authority and noted that domestic courts and tribunals are required to apply a structured proportionality test that is neither a light-touch review nor a substitution of the tribunal's own judgment for that of the employer.

The EAT accepted that the Trust's aim — the maintenance of safe, continuous community healthcare across a geographically dispersed rural population — was plainly legitimate. Patient safety and the continuity of NHS services are recognised legitimate aims in employment law, and the EAT declined to revisit the Tribunal's finding on this point. It confirmed that operational necessity in healthcare settings can in principle justify rostering practices that might otherwise constitute indirect discrimination, provided that the necessity is genuine and the measures deployed are no more onerous than the achievement of the aim requires.

However, the EAT found that the Employment Tribunal had committed a legal error in its proportionality analysis by failing to engage adequately with the less discriminatory alternatives identified by the claimant. The EAT drew upon the guidance in Hardys and Hansons plc v Lax [2005] EWCA Civ 846; [2005] ICR 1565, in which the Court of Appeal made clear that the proportionality assessment requires the decision-maker to consider whether the aim could have been achieved by means having a lesser discriminatory effect. The Tribunal below had, in substance, accepted the Trust's evidence that no viable alternative existed without subjecting that evidence to critical scrutiny. The EAT held that this fell short of the structured analysis required by law.

In particular, the EAT noted that the claimant had adduced evidence of rostering models operating in comparable NHS community trusts in other regions which achieved comparable levels of patient safety coverage while accommodating capped on-call obligations and enhanced flexibility for part-time workers. The Employment Tribunal had dismissed this evidence in a single paragraph without engaging with the specific operational comparisons drawn. The EAT held that this constituted an error of law: a Tribunal is not entitled to dismiss evidence of alternative measures without providing reasoned findings as to why those alternatives were not viable in the specific operational context of the respondent employer. The EAT drew support from Allonby v Accrington and Rossendale College [2001] EWCA Civ 529; [2001] ICR 1189 for the principle that the proportionality inquiry must be substantive and evidence-based, not conclusory.

The Trust had argued before the EAT that its geographical and operational context was so distinctive — combining a sparse population density, extended travel times between patients, and a historically depleted nursing workforce — that comparisons with urban or semi-urban trusts were of limited probative value. The EAT acknowledged the force of this submission as a matter of fact, noting that Tribunal findings as to operational context are entitled to respect. However, the EAT concluded that the proper course was not for the Tribunal simply to dismiss comparative evidence without engaging with it, but rather to make explicit findings about why the distinctive features of the Trust's operating environment rendered the alternatives unworkable. The absence of such findings was a legal error going to the adequacy of the reasons given, pursuant to the duty articulated in Meek v City of Birmingham District Council [1987] IRLR 250.

The EAT also addressed an argument advanced by the Trust that the claimant's appeal amounted to impermissible fact-finding by an appellate tribunal. The EAT reaffirmed the well-established principle, most recently stated in Volpi v Volpi [2022] EWCA Civ 464 in the civil appellate context and applied by analogy in employment appeals, that an appellate tribunal does not disturb findings of fact merely because it would have weighed the evidence differently. However, the EAT distinguished between the impermissible substitution of a different factual conclusion and the legitimate identification of a legal error in the reasoning process by which a factual conclusion was reached. The failure to engage with material evidence going to proportionality fell into the latter category.

On the cross-appeal concerning the appropriate form of remittal, the EAT applied the principles set out in Sinclair Roche and Temperley v Heard [2004] IRLR 763, under which remittal to the same Tribunal is appropriate where the error is confined to a discrete issue which can be addressed without reopening the entirety of the factual matrix, and where there is no risk that the original Tribunal will be unable to approach the remitted issue with an open mind. The EAT concluded that remittal to a differently constituted Tribunal was nonetheless warranted in the present case given the extent to which the proportionality analysis was interwoven with the Tribunal's overall assessment of the Trust's evidence and the risk that the same Tribunal's conclusions on related questions might subconsciously influence its fresh approach to the remitted issue.

The Employment Appeal Tribunal allowed the appeal. It held that the Employment Tribunal had erred in law in its application of the proportionality element of the objective justification defence under section 19(2)(d) of the Equality Act 2010 by failing to engage with material evidence of less discriminatory alternative rostering arrangements and by failing to provide adequate reasons for rejecting that evidence. The finding of group disadvantage was upheld. The finding on individual disadvantage, while technically containing a legal error, was treated as non-fatal given the sufficiency of the inferential evidence.

The EAT set aside the Tribunal's finding on objective justification and remitted the case to a freshly constituted Employment Tribunal to conduct a properly structured proportionality analysis, with specific direction that the Tribunal must make explicit findings as to whether the alternatives proposed by the claimant were operationally viable having regard to the Trust's specific geographic and workforce circumstances, and must provide reasoned conclusions supported by engagement with the evidence on each identified alternative.

The Trust's cross-appeal on the form of remittal was dismissed for the reasons given in the court's reasoning concerning the risk of subconscious influence on a re-hearing by the same panel.