private law,

Darnley v Croydon Health Service NHS Trust: delineating duty and breach


By William Wong Mar 13, 2019

Case Note: Darnley v Croydon Health Service NHS Trust [2018] UKSC 50, [2018] 3 WLR 1153

A. Introduction1

Darnley v Croydon Health Service NHS Trust2 (‘Darnley’) deals with the deceptively simple trio of negligence – duty of care, breach, and causation. In recent years, some judges have conflated them in a composite assessment by over-specifying the content of duty of care. In this refreshingly succinct judgment, the Supreme Court (‘UKSC’) gave a principled analysis of the nature of these components and preferred a more systematic and compartmentalised approach.

B. Brief Facts

The Appellant attended the Accident and Emergency Department at Mayday Hospital, Croydon.3  Despite having indicated to the receptionist that he had a head injury required urgent attention, he was told of a waiting time of 4 to 5 hours. In fact, the hospital’s policy would require a nurse to attend to the Appellant within 30 minutes. 4 After waiting for 19 minutes, he decided to leave. Later, he felt increasingly unwell and was taken back to the same hospital by an ambulance. It was revealed that he suffered from a extradural haematoma with a marked midline shift.5 The Appellant eventually suffered permanent brain damage.6

The Appellant claimed against the Respondent for a negligence, that the reception staff gave wrong information about the time he would have to wait, and its failure to assess the Appellant for priority triage. The claim was dismissed by the trial judge. The Court of Appeal (‘CA’) dismissed the appeal (McCombe LJ dissenting), holding that none of the duty, breach, and causation requirements were met. The UKSC allowed the appeal.

C. Analysis
i. Incremental approach affirmed

Darnley is consistent with the UKSC’s recent restatement in Robinson v Chief Constable of West Yorkshire Police7 (‘Robinson’), that it is unnecessary to consider the Caparo8 ‘fair, just, and reasonable’ (‘FJR’) test so long as the case falls under a pre-existing category where a duty of care could arise. Lord Lloyd-Jones did not even feel the need to review fully the existing authorities, since relying on Lord Reed’s exposition in Robinson seems to suffice. Common to both cases is a reluctance, almost distaste, for the FJR test. This reaffirms that the UKSC is honouring its promise to abandon the search of an one-size-fits-all test based on a unifying principle. With a category established in the case law, it is effectively presumed to be ‘fair, just, and reasonable’. The courts will no longer ‘re-open’ the category for scrutiny. The problem, however, is its presupposition that the previous cases are correctly decided and devoid of problems. Such presumption might deny the Courts the freedom to revisit some anomalous categories should the opportunity arise. In fact, Caparo’s ‘incremental approach’,9 which was approved by the UKSC in Darnley, is nothing novel nor commendable - the very mode of common law reasoning embraces analogy and incrementalism. Nevertheless, certainty triumphs over the search for principle here.

As a matter of precedent, no case law has specifically touched upon whether a receptionist in a hospital owns a duty of care towards prospective patients. With a note of pragmatism, the Court acknowledged that not very precise factual situation will have been the subject of a reported judicial decision. The ‘previously established category’ that the Court had in mind is that ‘those who provide and run a casualty department to persons presenting themselves complaining of illness or injury and before they are treated or received into care in the hospital’s wards.’10 This means that before resorting to the Caparo test. it is necessary to analyse the case law at a high level of conceptual abstraction.11 But at what level of specification should it be conceptualised is not expressly dealt with.

However, the Court’s finding that it is unnecessary to distinguish between medical and administrative staff is more dubious.  Lord Lloyd-Jones asserted that the receptionist is tasked with the responsibility to ‘provide accurate information as to [medical assistance]’s availability’.12 He feels that the distinction is only relevant to the question of breach. This is going arguably too far because in ascertaining duty of care, the focus is commonlly placed on factors such as an individual’s status and expertise relatively to that of the claimant. A more convincing way of putting it is to adopt the analysis in Macaulay v Karim and Croydon Health Services NHS Trust,13 which similarly concerned misinformation given by a receptionist. Foskett J classified the events as a ‘system failure’, instead of attributing the breach to a particular individual.14 Whilst this has the virtue of differentiating different members of staff in painstaking details, difficult questions of attribution can arise in determining which acts or omissions constitute the system failure, smacking of artificiality.

In sum, the UKSC is largely reiterating the orthodoxy in applying the incremental test, while reducing the need to resort to the FJR scrutiny.

ii. Eliding duty and breach

To properly understand this one has to return to the very foundation of the notion of ‘duty’. Hohfeld sees ‘duty’ and ‘right’ as jural correlatives in a relationship between two given individuals.15 In the language of tort law, it is commonplace to describe that ‘A owed B a duty of care’. The ‘duty’ narrative at first sight focuses on the status of the individuals. Therefore we speak of solicitor-client, doctor-patient, and parent-child as ‘established categories of relationships where duty of care arises’. However, Lord Bridge sounded a note of caution in Caparo:

‘It is never sufficient simply to ask whether A owes B a duty of care. It is always necessary to determine the scope of the duty by reference to the kind of damage from which A must take care to hold B harmless.’16

Lord Bridge requires identification of the precise harm that A is required to undertake one’s duty to prevent B from being affected. It is the extent of specification that concerns us here. In this regard, Goudkamp saw the CA judgment in Darnley as eroding of the customary divide between the duty of care and breach elements in negligence.17 For instance, Sales LJ framed the ‘duty question’ as whether there is a duty to provide ‘precise and accurate information’ about the length of time before a patient might be seen by a triage nurse.18 This, in the UKSC’s view, goes to the question of seriousness of breach only.19

Conflating these two components is not an isolated mistake, but instead reflects a systematic problem ringing throughout tort law. In Sam v Atkins,20 May LJ made a particularly bold statement. He acknowledged that it is customary to examine in a compartmental manner duty of care, breach, causation and damage; but that is ‘technically wrong’ since negligence is a ‘composite’ concept.21 In S v Gloucestershire County Council, he further opined that the duty of care question ‘necessarily subsumes the question whether the acts or omissions of the defendant caused the damage relied on’.22 Even Lord Diplock in the seminal case of Home Office v Dorset Yacht Co Ltd23 has fallen prey to this. He framed the enquiry question in a detailed manner: whether ‘any duty of care to prevent the escape of a Borstal trainee from custody’ was owed ‘by the Home Office to persons whose property would be likely to be damaged by the tortious acts of the Borstal trainee if he escaped’.24 This level of graphic details detracts the court from first, analysing the nature of relationship between the Home Office and the property owners in the vicinity; and second, balancing the risk and consequence of escape with the costs of prevention for the ‘breach question’.

Another reason why the Court should not over-specify the duty of care element is that a varied standard can be adopted at the breach stage. Nolan, in a penetrating analysis, has argued that varying the standard is a more useful technique.25 It may well be said that Accident & Emergency departments operate in a high-pressure environment. Lord Lloyd-Jones is therefore perfectly entitled to say that this is an important factor when assessing whether there has been a negligent breach of duty. This is precisely where the law could make allowance for and depart cautiously from the stringent ‘objective’ standard.

iii. Conflation with other components - causation and defence

This wide understanding of duty of care not only bears the risk of confusion with the breach assessment, it also subsumes the discussion that should arise under causation and defence. In Darnley, the CA further conflated the duty question with that of causation. Jackson LJ held that the scope of the duty would not extend to ‘liability for the consequences of a patient walking out without telling the staff that he was about to leave’.26 The UKSC found that no chain of causation has been broken simply because the Appellant decided to leave. It was reasonably foreseeable and was made partly on the basis of the misleading information.27

Similarly, in Rees v Commissioner of Police for the Metropolis,28 police owed the deceased a duty to take reasonable care to prevent him from committing suicide. As Nolan observed, framing the duty of care in this manner effectively subsumed causation (novus actus interveniens) and defence (volenti non fit iniuria). Similar problems are observed in the statutory context for occupiers’ liability.29 Occupiers’ Liability Act 1984 enables the court to determine whether an occupier owes a duty to a person (other than a visitor) through considering knowledge, foresight, and obviousness of danger,30 which are traditionally the language used in ascertaining the standard of care. The CA’s error in Darnley is thus only emblematic of a greater structural difficulty.

iv. Public policy and social costs

Public policy arguments in negligence actions are commonplace, but the UKSC was blunt to dismiss the Respondent’s arguments on the social cost and potential expenses incurred as ‘misplaced’.31 Typical arguments employed by public authorities in negligence claims include:

  1. Perverse incentive and defensive behaviour: In Hill v Chief Constable of West Yorkshire,32 fear of defensive policing behaviour. Sales LJ in the CA in Darnley also alluded to this argument, expressing his fear that information will be withdrawal from the public.33 Rejecting this, Lord Lloyd-Jones simply said that the burden of proof lies on the claimant - so long as the staff testifies as to their standard practice, sufficient safeguard would be in place.34 The problem is that this justification surely applies with equal force to the police investigation context in Hill. In any event, A & E departments and serious crime investigation are both high-pressure environment.

  2. ‘Floodgate’ argument: Jackson LJ in the CA judgment was concerned that a fertile ground for litigation would ensue.35 Nevertheless, Lord Lloyd-Jones did not address this claim head-on. It is perhaps the specific factual circumstances of the case that led him to believe there would not be a large amount of ensuing claims.

Overall, the Court in Darnley seems hasty in dismissing these considerations. The policy arguments are not foreign – they have been extensively rehearsed as ‘stock arguments’ and are almost becoming cliché.  It is in the interest of judicial candour that the UKSC be forthright in addressing them. In Robinson, Lord Hughes admirably acknowledged policy concerns as the ‘ultimate reason’ why no duty is imposed on the police.36 Nevertheless, Lord Reed countered that policy should not be routinely considered when existing principles provide a clear basis.37 By adhering to Lord Reed’s policy-minimalist approach, Darnley managed to shield itself from allegations of judicial overreach and activism. This is understandable when the policy issues were not decisive in the present case.

D. Conclusion

Robinson, with Darnley as its latest reaffirmation, has for now largely settled the approach towards assessing public authority’s duty of care. Ultimately, it is the ‘malleability’ of duty enquiry that generates incoherence in negligence.38 It is plausible that such malleability was the culprit in the courts’ abandonment of the search for unifying principle, which in turn gives rise to rule-of-law concern due to its opaqueness. In sum, Darnley’s note of caution against conflating the two stages of assessment is reassuring and laudable.



[1] I am grateful to Ines Chu and Mythili Mishra for their research assistance.

[2] Darnley v Croydon Health Service NHS Trust [2018] UKSC 50, [2018] 3 WLR 1153.

[3] ibid [1].

[4] ibid [5].

[5] ibid [6].

[6] ibid [7].

[7] [2018] UKSC 4, [2018] 2 WLR 595.

[8] Caparo Industries plc v Dickman [1990] 2 AC 605, 617 (Lord Bridge).

[9] Adopted from the Australian case of Sutherland Shire Council v Heyman (1985) 60 ALR 1.

[10] Darnley (UKSC) (n 1) [16], citing Barnett v Chelsea and Kensington Hospital Management Committee [1969] 1 QB 428.

[11] Owain Thomas QC and Philip Havers QC, ‘Supreme Court rules that hospital receptionist owes a duty of care to a patient’ (11 October 2018, UK Human Rights Blog) https://ukhumanrightsblog.com/2018/10/11/supreme-court-rules-that-hospital-receptionist-owes-a-duty-of-care-to-a-patient-owain-thomas-qc/ accessed 24 February 2019.

[12] Darnley (UKSC) (n 1) [17].

[13] [2017] EWCH 1795 (QB).

[14] Ibid [166].

[15] W. N. Hohfeld, ‘Fundamental Legal Conceptions as Applied in Judicial Reasoning’ (1917) The Yale Law Journal Vol. 26 No. 8 710.

[16] [1990] 2 AC 605, 627.

[17] J. Goudkamp, ‘Case Comment Breach of duty: a disappearing element of the action in negligence?’ (2017) CLJ 481.

[18] Darnley v Croydon Health Services NHS Trust [2017] EWCA Civ 151, [2018] QB 783, [87].

[19] Darnley (UKSC) (n 1) [23].

[20] [2005] EWCA Civ 1452, [2006] RTR 14.

[21] Ibid [14].

[22] [2001] Fam 313, 337F.

[23] [1970] AC 1004.

[24] Ibid.

[25] D. Nolan, ‘Varying the standard of care in negligence’ (2013) CLJ 651.

[26] Darnley (CA) (n 18) [56].

[27] Darnley (UKSC) (n 1) [29]-[30].

[28] [2018] .

[29] K. Amirthalingam, ‘Occupiers’ liability in England: time for some housecleaning?’ (2017) PN 33(1), 46-50.

[30] Occupiers Liability Act 1984, s. 1(3).

[31] Darnley (UKSC) (n 1) [22].

[32] [1989] AC 53.

[33] Darnley (CA) (n 18) [88].

[34] Darnley (UKSC) (n 1) [22].

[35] Darnley (CA) (n 18) [55].

[36] Robinson (n 7) [103]-[120].

[37] Ibid [69].

[38] Nolan (n 25) .

Article by William Wong
LLB (London School of Economics and Political Science) '20 and Private Law Notes Editor of the LSE Law Review 2018-2019