Spire Healthcare Limited v Royal & Sun Alliance Insurance Limited

This appeal concerns the interpretation and application of an aggregation clause in a combined liability insurance policy, number YMM830842 (“the Policy”), underwritten by the Appellant, Royal & Sun Alliance Insurance Ltd (“the Insurers”), in favour of the Respondent insured, Spire Healthcare Ltd (“Spire”). The Policy was subject to an aggregate overall limit of £20 million.

Spire belongs to a substantial group, and is responsible for the operation of a number of private hospitals. Section 4 of the Policy provided cover in respect of Spire’s legal liability for accidental injuries arising out of medical negligence at those hospitals. Clause 5(a) of that section (“the aggregation clause”) provides that:

The underlying claims for personal injuries which gave rise to Spire’s claim for an indemnity under the Policy arose out of the serial misconduct over many years of a Mr Ian Paterson, a consultant breast surgeon employed by the Heart of England NHS Foundation Trust (“HEFT”). Mr Paterson carried out surgical procedures at two private hospitals near Birmingham operated by Spire. In 2011 he was suspended from practice by the General Medical Council (“GMC”).

It transpired that over a period of around 14 years, Mr Paterson had performed operations on patients, the vast majority of whom were female, without their informed consent. His conduct was both negligent (i.e. a breach of the duty of care he owed to his patients) and dishonest. That conduct is described in greater detail in the judgment of HH Judge Pelling QC (“the Judge”) [2020] EWHC 1299 (Comm), but for present purposes the following summary will suffice.

In patients where a mastectomy was clinically indicated, Mr Paterson failed to remove all breast tissue, thereby exposing the patient to an unnecessary risk of recurrence and/or metastasis. This practice was referred to as performing “sub-total mastectomies” or “STMs,” and the Judge found that the reason for it was never adequately explained. Officials at HEFT first found out about this practice in 2007. They sought and received an assurance from Mr Paterson that he would stop performing STMs. However, by 2011 it became apparent that he had not done so, and it was this discovery that led to his suspension by the GMC. The patients who underwent STMs are referred to as “Group 1” patients. They were both NHS and private patients.

After his suspension, it was discovered that Mr Paterson had also falsely reported pathology test results as indicative of the presence (or risk of the presence) of cancer, and then carried out unnecessary surgical procedures on the patients concerned, including mastectomies. The motivation behind this behaviour appears to have been financial gain, as he collected fees from the patients or their insurers for carrying out the procedures and follow-up consultations. Most of the affected individuals were private patients. They are referred to as “Group 2” patients. There are some unfortunate examples of patients falling into both groups, but by and large Groups 1 and 2 had different constituents.

Around 750 former patients of Mr Paterson made claims against Mr Paterson, Spire and HEFT. Master Particulars of Claim, addressing 7 lead cases in the ensuing group litigation, were served on 4 December 2015.

In April 2017, in Nottingham Crown Court, Mr Paterson was convicted of 17 offences under section 18 of the Offences Against the Person Act 1861, and 3 counts under section 20 of that Act, though those offences only related to a small sample of the patients affected by his malpractice (9 women and one man). He is currently serving a sentence of 20 years’ imprisonment.

The group litigation was settled in October 2017, six months after Mr Paterson’s conviction, by the setting up of a substantial compensation fund for the victims. Spire contributed around £27 million to the fund. Its overall outlay (including on its own defence costs) was a little over £37 million. On any view, therefore, the aggregate Policy limit of £20 million was exceeded.

The evidence at trial established that Spire’s outlay in respect of Group 2 claims alone exceeded £10 million.

The Insurers accept that they are liable to indemnify Spire under Section 4 of the Policy; the sole issue on appeal is whether that liability is capped at £10 million, as the Insurers contend, or £20 million, as the Judge held.

For the reasons that follow, I consider that the Judge fell into error. On the correct application of the aggregation clause, the claims all arose out of the same source or original cause, namely, Mr Paterson’s conduct in disregarding the welfare of his patients and performing operations on them without their informed consent. It is immaterial that for the Group 1 patients a mastectomy was clinically indicated, whereas for the Group 2 patients it was not.

Section 4 of the Policy provides that the Insurers will indemnify “any Person Entitled to an Indemnity” against their legal liability for damages in respect of accidental injury of any person arising out of Medical Negligence, where the claims are made against them and notified to the Insurers in the prescribed manner during, or within 30 days after the relevant period of insurance. There is no issue about timely notification in this case.

Although at an earlier stage of the litigation there was a dispute about this, it was common ground before us that the phrase “accidental injury of any person arising out of Medical Negligence” must be judged from the perspective of Spire, as the assured, and so it would not affect cover if the injuries were inflicted deliberately by Mr Paterson: see Hawley v Luminar Leisure [2006] EWCA Civ 18, [2006] Lloyd’s Rep IR 307.

“Medical Negligence” is defined in the general definition section of the Policy as an “actual or alleged negligent act, negligent error or negligent omission committed by any Person Entitled to an Indemnity which arises from the conduct of the Insured’s Business as defined in the Schedule.” (Emphasis added). “Person Entitled to an Indemnity” is defined as including, so far as is relevant, the Insured and any “Person Employed” by the Insured. That expression is in turn defined as meaning an employee or a “self-employed individual (not being in partnership with the Insured) while under the direct control or supervision of the Insured.”

Mr Paterson was an independent contractor, who contracted directly with his private patients. He was not under the direct control or supervision of Spire. In any event, even if he had been, a “General Memorandum” at the start of Section 4 makes it plain that the indemnity provided to any person employed by Spire “will not apply to any doctor, surgeon [or] consultant…” Medical professionals are expected to, and usually do, take out their own professional liability insurance. Therefore, in order to make good the claim against the Insurers under the Policy, Spire had to demonstrate that it was legally liable for damages in respect of accidental injury to patients operated on at the two hospitals concerned, arising out of its own negligence, or the negligence of persons other than Mr Paterson falling within the definition of “Persons Employed”.

Spire asserted in Para 67 of its Particulars of Claim in the proceedings against the Insurers that it was liable to the patients on 4 separate bases, namely:

It was liable for the acts and omissions of a named employee who facilitated Mr Paterson’s course of conduct and who failed to report his behaviour when she should have done;

It was liable for the failure of management at the hospitals to carry out adequate investigations into Mr Paterson’s conduct and take appropriate action;

It was liable for breach of an implied term that the services the patients would receive at the hospitals would be carried out with reasonable skill and care; and

There was a risk that the principles of vicarious liability would be developed in such a way as to make Spire liable for Mr Paterson’s conduct despite the fact that he was employed by HEFT.

It was no part of Spire’s case that the fact or nature of its liability depended on whether the patient making the claim fell within Group 1 or Group 2.

There was very little dispute between the parties as to the correct approach to the construction of the aggregation clause. As the Judge stated at [20] and [21], such clauses are to be construed in a balanced fashion without a predisposition towards a narrow or broad interpretation: see Lloyds TSB General Insurance Holdings v Lloyds Bank Group Insurance Co Ltd [2003] UKHL 48, [2003] Lloyd’s Rep IR 623 per Lord Hobhouse at [30]; AIG Europe Ltd v Woodman [2017] UKSC 18, [2017] 1 WLR 1168 per Lord Toulson JSC at [14]. The usual principles of contractual construction apply.

However, because this is standard wording used in an insurance policy, it is appropriate to follow the construction of identical or materially similar provisions in earlier cases, unless there is a clear contextual distinction or other strong reason that suggests it would be inappropriate to do so: see Hooley Hill Rubber & Chemical Company Ltd v Royal Insurance Company [1920] 1 KB 257 per Scrutton LJ at 272. As the Judge recorded at [21], neither party contended that there was a reason for such a departure in this case.

Aggregation clauses like this one, which refer to claims or occurrences “consequent on or attributable to one source or original cause”, use a traditional and well-known formula to achieve the widest possible effect, as Longmore LJ (delivering the judgment of the court) stated in AIG Europe Ltd v OC320301 LLP and others [2016] EWCA Civ 367, [2017] 1 All ER 143 at [21]. Whilst the decision of the Court of Appeal in that case was overturned by the Supreme Court in AIG v Woodman (above), those observations were not disapproved. Indeed, they are consistent with a long line of earlier authorities.

In an oft-quoted extract from his speech in Axa Reinsurance UK Ltd v Field [1996] 1 WLR 1026 at 1025, Lord Mustill said this about a clause which referred to an “originating cause”:

It is well established that in this context there is no distinction to be drawn between an “originating cause” and an “original cause.” As Morison J put it in Countrywide Assured Group Plc & Others v Marshall and others [2002] EWHC 2082 (Comm), [2003] Lloyd’s Rep IR 195 at [15]:

“Original cause” in this context does not mean “proximate cause”, but instead connotes what Christopher Clarke J described as a “considerably looser causal connection”: Beazley Underwriting Ltd v The Travelers Companies Incorporated [2011] EWHC 1520 (Comm) at [27]. It follows that the “original cause” need not be the sole cause of the insured’s liability. However, as Moore-Bick J observed in American Centennial Insurance Co v INSCO Ltd [1996] LRLR 407 at 414, it is still necessary for there to be some causative link between the originating cause and the loss, and there must also be some limit to the degree of remoteness that is acceptable. As Mr Shapiro QC, for Spire, put it, not every “but for” cause is sufficient to amount to an “original cause.” In searching for the unifying factor, one must not go back so far in the causal chain that one enters the realm of remote or coincidental causes which provide no meaningful explanation for what has happened.

Thus, for example, in The Cultural Foundation v Beazley Furlonge Ltd [2018] EWHC 1083 (Comm), [2019] 1 Lloyd’s Rep 12, Andrew Henshaw QC, sitting then as a deputy High Court Judge, took the view at [204] that to encompass any claims arising from bad design on a particular project by the insured architect, RMJM, would give too vague a meaning to the words “original cause”. In that case there were claims for negligently defective structural engineering which made the building unsafe, and the insurers sought to aggregate these with claims for failure by RMJM to include in the designs certain details and provision for co-ordination that was necessary for the efficient execution of the project (which impacted only on cost). Mr Henshaw QC decided that to describe these claims as all emanating from “poor initial design” set the test at so generalised a level as not to be useful in the context of a search for an effective original cause. On the facts of that case, the distinction he drew between the different types of claim made by RMJM’s clients was understandable, but he was not seeking to lay down any principle of wider application.

It is convenient at this stage to dispose of the argument advanced by the Insurers in their Grounds of Appeal, though pursued less enthusiastically by Mr Elkington QC in his oral submissions, that there is a distinction to be drawn between “source” on the one hand and “original cause” on the other. That is not so; as Eder J pointed out in Standard Life Assurance Ltd v ACE European Group [2012] Lloyd’s Rep IR 655, at [259], the word “source”, when added as an alternative to “original cause” simply serves to emphasise the intention that the doctrine of proximate cause should not apply, and that the losses should be traced back to wherever a common origin can reasonably be found. The Judge was right to treat the two expressions as interchangeable.

Mr Elkington submitted that, as Spire’s liability to patients and the claim for an indemnity was the same across both Groups, it would be strange to draw a distinction between them. He pointed out that Spire (and the Judge) had accepted aggregation within each of the two Groups. The Judge should have asked himself whether there was a unifying factor in the history of the claims with which Spire was faced, and for which it was liable. Had he taken the same approach as Morison J did in the Countrywide case and asked whether there was such a factor behind the whole problem, he would have identified that there was, namely, a single individual, a “rogue consultant” who habitually acted in breach of his duties to his patients. All the patients’ claims were based on Mr Paterson’s improper, indeed, dishonest, conduct. That conduct, in all cases, involved operating on the patients without their informed consent and with disregard for their welfare. However, instead of looking for a unifying factor, the Judge concentrated on identifying differences between the patients’ claims falling within Group 1 and Group 2, and embarked upon an unnecessary and inappropriate analysis of Mr Paterson’s motivation for his conduct.

In response to that criticism, Mr Shapiro contended that the conduct of Mr Paterson in respect of the Group 2 patients was fundamentally different in nature from his conduct in respect of the Group 1 patients, and the Judge was right so to find. By reference to the approach taken by Mr Henshaw QC in The Cultural Foundation v Beazley (above) Mr Shapiro contended that the proposition that the original cause of Spire’s liability was Mr Paterson, or Mr Paterson’s misconduct, was too vague and too remote.

The difficulty for Mr Shapiro lay in articulating precisely what the distinction was, in terms of identifying two separate sources or original causes of Spire’s liability to the patients which triggered its right to indemnity under the Policy. He initially suggested that different types of negligence by one surgeon would each be separate originating causes. The fallacy of that approach, however, was demonstrated by the example put to Mr Shapiro of the orthopaedic surgeon who negligently fails to diagnose an incomplete fracture in Patient A, then negligently performs an operation on Patient B, and finally gives the wrong aftercare instructions to the physiotherapist in respect of Patient C. In all three cases the reason why the operator of the private hospital would be liable to those patients for their ensuing personal injuries is because it engaged (or failed to take adequate steps to supervise or manage) an orthopaedic surgeon who was incompetent. The fact that the pattern of incompetence manifested itself in different ways did not alter the fact that the problem could be traced back to one incompetent/negligent surgeon who failed to treat his patients with adequate skill and care.

In Cox v Bankside [1995] 2 Lloyd’s Rep 437, at 455, Phillips J rejected a submission by counsel that the single originating cause of all the claims was the negligent approach to underwriting of three underwriters for the same Lloyd’s syndicate, because they each laboured under an identical misapprehension when they wrote the business. He described the reasoning as fallacious. Whilst a mis-appreciation in an individual which leads him to commit a number of negligent acts can arguably be said to constitute the single event or originating cause responsible for all the negligent acts and their consequences, “the same is not true when a number of individuals each act under an individual mis-appreciation, even if the nature of that mis-appreciation is the same”.

Tellingly, the multiple instances of negligence by each individual were not treated by Phillips J as separate originating causes. For example, one of the underwriters failed to monitor aggregates, failed to competently estimate his exposure, and had no proper appreciation of the nature of the business he was underwriting. Nevertheless, the single unifying factor in his case was identified by the judge as his negligent approach to underwriting risks for the particular Lloyd’s syndicate. Cox v Bankside therefore illustrates that the negligence of one individual can be an originating cause for the purpose of an aggregation clause of this type, even though his negligence may take different or multiple forms.

Mr Shapiro then sought to draw a distinction between conduct that was merely negligent, and conduct which was also dishonest. However, the Judge found (at [31]) that all Mr Paterson’s conduct was dishonest. Moreover, it was plain that the dishonesty was inextricably bound up with the negligence. The Judge himself attempted to distinguish between different types of dishonesty, instead of treating the dishonest conduct as a unifying factor.

Mr Shapiro submitted that what distinguished the Group 2 cases was the deliberate nature of the misconduct, but that did not provide a bright line either, because most of the Group 1 operations were deliberately performed as STMs (although some may have been intended to be full mastectomies that were inexpertly carried out). Ultimately, Mr Shapiro fell back on the distinction which the Judge drew between the cases in which surgical procedures were clinically indicated, but were carried out in a sub-optimal fashion, and cases in which the patient did not require surgery. However, the question whether the patient did or did not require surgery had no bearing on Spire’s liability for the injuries sustained in consequence of the surgery that was carried out.

In my judgment, Mr Elkington’s criticisms were sound. Having correctly adumbrated the applicable principles, the Judge failed to conduct the wide search for a unifying factor in the history of the claims that the authorities required him to carry out. Indeed, he appears to have noted the factors that were common to all the claims but then disregarded them, in the course of searching for what he termed a “single effective cause,” which is not the correct test.

It appears that the Judge’s approach was based to a large extent on the passage from Phillips J’s judgment in Cox v Bankside to which I have already referred (see paras [23]-[24] and [39] of the judgment). He reasoned that if identical misapprehensions by different individuals could be treated as separate originating causes, then so too could different misapprehensions leading to negligent acts by the same individual. It appears that it was this which drew the Judge into considerations of motivation, which he then elided with causation in the passage in the judgment at [43]–[46]. However, Mr Paterson’s motivations (whatever they may have been) were irrelevant to Spire’s liability.

The passage in Cox v Bankside provides no justification for the approach adopted by the Judge, which introduced unnecessary complication into what should have been a relatively simple and straightforward exercise. The claims against Spire were not based upon a negligent misunderstanding, such as a surgeon’s mis-appreciation of the correct way in which to carry out a particular type of operation. They were based upon a pattern of deliberate (and dishonest) behaviour by one individual who operated on hundreds of patients over 14 years in the two private hospitals run by Spire, with cavalier disregard for their welfare. There may be cases in which, on the facts, the behaviour of one individual will be too remote or too vague a concept to provide a meaningful explanation for the claims, but this is not one of them.

As a matter of ordinary language, and applying the principles applicable to aggregation clauses expressed in these wide terms, it seems to me to be plain that any or all of (i) Mr Paterson, (ii) his dishonesty, (iii) his practice of operating on patients without their informed consent, and (iv) his disregard for his patients’ welfare can be identified either singly or collectively as a unifying factor in the history of the claims for which Spire were liable in negligence, irrespective of whether the patients concerned fell into Group 1 or Group 2 (or both). None of the factors which I have identified could be described as falling within the realm of remote or coincidental causes which provided no meaningful explanation for what happened.

The Judge therefore ought to have concluded that all the claims within Group 1 and Group 2 aggregated with one another, and the £10 million limit applied. For those reasons, I would allow this appeal.

I agree.

I also agree.