Bannister v Freemans Plc [2020] EWHC 1256 (QB) (Part 1 of 3)

Today is the first instalment in a three-part series of blog posts on Bannister v Freemans Plc, a decision of Geoffrey Tattersall QC (sitting as a Judge of the High Court) handed down on 19 May 2020. The trial was heard over three days from 16 to 18 May 2020, probably the last asbestos trial to be heard before the country entered into lockdown.

Michael Rawlinson QC kindly agreed to write a piece on the case for this blog with the assistance of Samuel Cuthbert, a pupil at 12KBW. What they have produced is a little longer than our usual posts, so we are going to publish it in three parts starting today and continuing tomorrow and Thursday.

Today’s instalment sets out a full summary of the facts and considers the orthodox approach to evidence in low dose mesothelioma cases. In part 2, Mike introduces the concept of the “iron triangle” and the relationship between breach of duty, exposure and causation. In part 3, he considers the use of epidemiology.

First, a very brief introduction to the case for those who have yet to read it. It was alleged by the Claimant that her husband died of mesothelioma as a result of exposure to asbestos dust following asbestos removal carried out at his place of work in the early 1980s. It was conceded by the Defendant that if asbestos exposure was proved as alleged, breach of duty would be established. However causation was disputed, the Defendant arguing that the alleged exposure did not constitute a “material increase in risk”, i.e. it failed to meet the modified causation test which applies in mesothelioma cases following the House of Lords’ decision in Fairchild.

The Judge found on the facts that the Deceased had not been exposed to asbestos dust as alleged and accordingly the claim failed. However, the Judge went on to consider the causation question notwithstanding this finding. As Mike explains, this gave rise to some interesting evidential and legal analysis.

This analysis should be read with the following in mind:

  1. It is obiter, the Judge having found that the allegation of asbestos exposure was not made out.
  2. The Judge’s obiter finding on the extent of the asbestos exposure was that the cumulative dose would have been “in the region of no more than 0.0004 fibre/ml years”, i.e. extremely low.
  3. The Judge’s obiter finding in respect of causation was founded on the specific expert medical evidence in the case.

Over to Mike.


A:        Introduction

I should make clear at the outset:

(a) I am immensely grateful for the close collaboration and hard work of Samuel Cuthbert (one of our pupils at 12KBW) in the construction of this piece.

(b) Although this is being drafted within the period of time during which a notice of appeal could still be filed, I have not asked the Counsel team representing the family (namely Harry Steinberg QC and Gemma Scott) whether the same is contemplated and, if so, on what basis. This blog piece represents the views of Sam and me and, insofar as we detect heterodoxy within the Judgment (and, respectfully, we most certainly do) they should not be necessarily imputed to the trial Counsel team.

Low dose mesothelioma claims form an ever-greater proportion of the workload of any asbestos lawyer. When, therefore, a decision of the High Court is handed down dealing with issues of proof and causation and particularly where the Court had the benefit of full submissions from two silks highly experienced in the area, it commands careful consideration. The judgment of Mr Geoffrey Tattersall QC (sitting as a Judge of the High Court) in Bannister v Freemans plc [2020] EWHC 1256 (QB) is just such a decision. The need to analyse it with care is heightened further by some of the commentary it has attracted so far. One commentator has been moved to opine that “It should promote a more scepticalapproach” to the evidence of victims. This is, of course, wrong.

If brevity is the soul of wit, then, once again, I shall show myself to be witless. This piece is necessarily long. It is right therefore that I provide an executive summary at once.

(a) So far, at least, this is a first instance judgment which sits within the framework of the common law in respect of personal injury generally and mesothelioma claims in particular, as handed down by the appellate Courts. Any expressions of law contained within it must be construed accordingly

(b) The ratio of the decision is purely factual: on the facts, the family did not prove that the victim had been exposed to asbestos fibres at all during the course of his employment with the Defendant (“D”)[1]. All else is simply obiter.

(c) The judgment contains no warrant for a general approach whereby the evidence of victims (often given in the context of nil disclosure from D and always in respect of relative mundane matters occurring several decades prior) should be treated automatically with scepticism. The proper approach is to evaluate all the evidence before the Court neutrally, to test it for what may fairly be accepted either as to primary fact or reasonable inference and then apply to it the usual burden and then standard of proof.

(d) The Judge appears to have resolved the issue of what constitutes an exposure which materially increases the level of risk of developing mesothelioma by the adoption of a test suggested by a medic, namely a level of dose above that which  “… a medical practitioner who is aware of the medical risks would define as something that the average patient should not worry about” [168] (emphasis added). Perhaps unsurprisingly the medic’s formulation of the answer includes a frank error of law – namely the failure to take into account either the ‘eggshell skull’ rule (Smith v Leech Brain & Co Ltd) or the ‘crumbling skull’ rule (Environment Agency v Ellis/Athey v Leonati) (in the form of the known existence of genetic susceptibility of certain people to asbestos giving rise to mesothelioma). Perhaps more surprisingly, the test promulgated by the medic seems to not have been thought through as to how it would ever have been applied in practice.

(e) The Judge was led by D’s concession that if the facts alleged could be proven they constituted a breach of duty, into considering only (a) the issue of whether (and, if so, to what degree) exposure had occurred – proof of exposure and (b) whether such exposure (if proven) materially increased the risk of mesothelioma – proof of individual causation. In this, he concentrated on two sides of a triangle, but not the third, namely the test for breach. It is true that he briefly cites a short passage from Lord Phillips in Sienkiewicz on breach [107 of the Judgment in Sienkiewicz] and [27] of the Judgment in Bannister, but he rather skates over at [32] the role of material increase in creating a ‘foreseeable risk’ which is the essence of the proof of breach at common law in mesothelioma cases. Hence with respect, he did not fully express within the judgment an appreciation of how this third side of the triangle creates an iron logic in post 1965 cases which employers cannot evade by reference to some pseudo-objective test for de minimis. I return to the nature and effect of this ‘iron triangle’ in part 2.

In short therefore, the Judge’s obiter analysis of what constitutes material breach is contrary to appellate decisions eschewing objective dividing lines howsoever devised, ignores Hughes v Lord Advocate and the presence of genetic susceptibilities as a confounding factor when seeking to use epidemiology as a disproof of individual causation.

I approach the matter by summarising the case and then seeking to analyse where commentary on behalf of the insurance industry is not justified, and where the Judge’s obiter reasoning went awry.

B:        Summary of the Judgment

Dennis Bannister (“the Deceased”) died of malignant mesothelioma after a prolonged illness. The claim was brought by his widow, Valerie Bannister (“the Claimant”) and alleged that he had been exposed to asbestos whilst working as a manager in the accounts department under the employment of Freemans Plc (“D”). The allegation of exposure was that a partition wall had been removed from the Deceased’s office over the course of a weekend, following which the Deceased was exposed to a residue of asbestos dust in his office which was cleaned up over the course of a matter of days by the cleaners. As we have seen from the above summary it was admitted that, if proven to be asbestos, such exposure was tortious.

Evidence was given by the Deceased’s former colleague, Mr Ford. Mr Ford had met with the Deceased following the Deceased’s diagnosis, and reminded him both of the removal of the partitions and that a memo had been issued by the Defendant identifying and warning of the presence of asbestos containing infill panels. Expert engineering evidence was provided by Mr Raper for the Claimant and Mr Stear for D. Expert medical evidence was provided by two respiratory physicians; Dr Rudd for the Claimant and Dr Moore-Gillon for D. D denied that the dust contained asbestos and, in the alternative, that any exposure amounted to no more than non-actionable de minimis amounts.

The Judge outlined that, in making his findings of fact, there was need for particular care as regards the accuracy or reliability of the evidence of lay witnesses in historic disease claims. Such principles are derived from the dicta in Kimathi v Foreign and Commonwealth Office [2018] EWHC 2066, Gestmin SPGS SA v Credit Suisse (UK) Limited [2013] EWHC 3560 and Sloper v Lloyds Bank Plc [2016] EWHC 483. The Judge made three preliminary findings:

  • He drew an adverse inference as to the Deceased’s credibility (NB not honesty: credibility) from his denial of exposure to asbestos to treating medics on 3 separate occasions
  • Further, the Deceased’s evidence was largely prompted by Mr Ford’s recollection of the memo;
  • But for Mr Ford reminding the Deceased in early 2018 of the existence of the memo or the removal of the partition containing asbestos, the Deceased had no independent recollection of being exposed to asbestos.

The Judge found that, on balance, a memo was sent to the Deceased and that it referred to the infill panels in the partition being removed as having contained asbestos. He then went on to conclude that having identified the presence of asbestos and having sent a memo about the same it was probable that D also would have appreciated the need to engage a specialist contractor to undertake the removal of the asbestos with the appropriate precautions pertaining.

The Judge found that only the infill panels of the partition had been removed and not the central panels. He found that the infill panels were removed the weekend immediately after the memo was circulated to the Deceased. This would have created dust, and it is possible that the Deceased could then have been exposed to asbestos dust upon arriving back to work. However, the process of replacing the infill panels with non-asbestos material would also have created dust free from asbestos.

The Judge found overall that the Deceased was more likely to have been exposed to non-asbestos dust from a different but related process when he told Mr Ford that he could taste dust in his mouth (which was taken as an evidential token of having been exposed to some dust). Overall therefore on the facts the Deceased failed.

The Judge then turned to his obiter considerations. He went on to consider what the Deceased’s exposure would have been had he found that the Deceased was so exposed. He asked himself (i) how a court should assess the Deceased’s exposure and, (ii) what in law constitutes a material increase of the risk of the Deceased developing mesothelioma?

 The Judge recognised that the experts’ calculation of the cumulative dose has a limited value, albeit that it does have some value. Pursuant to the dicta of Aikens LJ in Williams, the Judge sought to make findings as to the Deceased’s actual level of exposure to asbestos, whilst accepting that this might be imprecise. The two experts, after cross examination, settled on very similar dose estimates such that they were effectively in agreement as to the Deceased’s cumulative exposure to asbestos. In any event the Judge preferred the evidence of the Mr Stear, given that Mr Raper’s estimates for both the lower and upper ends of exposure did not withstand close scrutiny. The Judge therefore found “in round terms”, in line with Mr Stear’s evidence that the Deceased’s exposure was no more than 0.0004fibre/ml years.

It was held that the test of causation is one of mixed fact and law because any assessment by the court had to reflect what test to apply in the assessment of the risk and what weight to give to epidemiological evidence. Following Sienkiewicz v Grief (UK) Ltd [2011] UKSC 10, what constitutes a material risk must be for a judge on the facts of the particular case. The Judge found the test set in Sloper v Lloyds Bank Plc [2016] EWHC 483 (QB) was an appropriate means of determining material increase in risk:

“a dose of asbestos which was properly capable of being neglected could be defined as a dose which a medical practitioner who is aware of the medical risks would define as something that the average patient should not worry about.”

The Judge preferred the evidence of Dr Moore-Gillon because he had attempted to evaluate the significance of the cumulative dose at 1 in 50 million, subsequently describing the increase in risk as “vanishingly low”. Dr Rudd’s evidence was criticised for not assessing the level of risk created by any exposure to asbestos whilst the Deceased was in the Defendant’s employment, and for “straining logic” regarding an annual risk of 1 in 50 million as a material increase.

The Judge rejected the submission that it would be an unusual situation where an exposure to asbestos which constituted a breach of duty was deemed not to be a material increase in risk. This was deemed to erroneously conflate breach and causation. Instead, the judge found that the burden was on the Claimant to show, on a balance of probabilities, that any exposure to asbestos suffered by the Deceased in the course of his employment by the Defendant gave rise to a material increase in the risk of the Deceased developing mesothelioma. Given the Judge’s preference for Dr Moore-Gillon’s evidence, it was found that such burden was not discharged and any exposure would have been de minimis.

 In sum, it was held on the balance of probabilities that the Deceased was not exposed to asbestos dust during his employment by the Defendant. In any event, had such exposure been proved, it did not represent a material increase in the risk of the Deceased developing mesothelioma.

C:        Comment

C1. The approach to evidence

 This requires consideration in respect of both lay and expert evidence separately.

Lay evidence:

(a) It is possible to spill a considerable amount of ink considering to what degree Gestmin principles should be applied outside of commercial litigation (in which usually there is a plentiful substrate of contemporaneous documentation) and personal injury claims. Doubt has been expressed in this regard in at least one decision of the High Court (CXB v North West Anglia NHS Trust [2019] EWHC 2053 QB) (but it is readily accepted that equally there are other Courts which have accepted its application in such cases).

(b) However, it is never more than an approach: to elevate into an evidential principle that lay evidence must always give way to written evidence, or that long term memory being relied upon by a witness is inherently unsafe owing to the passage of time takes matters too far. First, insofar as such a purported principle was to be justified on the grounds that there is a growing realisation that Judges as a body are no better interpreters of human fallibility and recall than any other interlocutor (which is often the justification) then I am afraid the medical, sociological and empirical underpinnings of such an assertion would become fair game and relevant issues of proof in any low level case. This cannot be right.

(c)  Second, it would create an exquisite trap for any victim: if they gave general evidence about what happened in a mundane factory 40 years ago, D would argue ‘ah, the lack of detail betokens want of real recall’. Conversely, if the victim gave an account in which there were elements of vivid detail, D would respond ‘ah, that detail is an after-acquired trick of the memory which betokens that the victim’s evidence is unreliable’.

(d)  The Judgment in Bannister itself records:

  1. Moreover, although Mr Steinberg stated that Lord Rodger`s dicta in Sienkiewicz were simply ‘a reminder that the relaxation of the causation test did not apply to or eliminate the other ingredients of tortious liability’, I am satisfied that such dicta and the other judicial observations in KimathiGestmin and Sloper , whilst in no way binding on me, are important as a helpful and cautionary general guide to evaluating oral evidence and the accuracy or reliability of memories and I do not propose to allow the Defendant, in Mr Steinberg’s words ‘to convert one of the inherent difficulties in asbestos litigation – the inevitably long latency periods of the disease – into its first line of defence.’

(e)  Thus there is no warrant from this Judgment for the D commentary promulgated online that victims’ evidence should be viewed sceptically as a species. D’s protection is that the burden lies on the victim to prove all elements of his case and to do so on the balance of probabilities. In looking at the evidence, the Court will take all relevant aspects into account. This may be other lay evidence presented by either side; this may be contemporaneous or subsequent documentation; this may include expert comment where such evidence is relevant and admissible. It will be a matter for the Judge to sift and give such weight to each element of the evidence as she deems proper. D should recall however, that where the only evidence is that of the victim themselves, the minatory words of the Court of Appeal in Brett v University of Reading per Sedley LJ:

where the evidence points neither way a straw in the wind may be decisive” (see also Hughes v Liverpool City Council, The Times 30   March 1988 per May LJ where he said “The onus was on the plaintiff to make out her case. If there had been only a scintilla of evidence called on her behalf tending to support the fourth inference to which I have referred, then in the absence of any contrary evidence, because no witness was called for the defendants, the judge would have been entitled to find even that scintilla sufficient to make out the plaintiff’s claim”.

Expert evidence:

(a)  The Judge effectively accepted a ‘back calculation’ regarding the level of exposure to asbestos which the Deceased experienced and did so by reference to figures suggested by D’s engineer. He did so:

  • Having determined as a matter of law that he was required to make findings of fact as to the level of exposure (citing and following Aikens LJ in Williams v University of Birmingham – who said that that was indeed the task of the Court; citing and not following Maurice Kay LJ in Cox v Rolls Royce to the contrary effect that no specific finding was necessary beyond a general finding that it was material [153-157]); and
  • Having acknowledged that the experts evidence had ‘limited’ value [151].

(b)  With respect to the Learned Judge, he appears not to have taken full consideration of Bussey.

  • First, he did not cite, nor apparently have in mind, the cautionary words of Underhill J about the exercise undertaken of ‘back calculation’ of dose by experts:

“[62] …Attempting to answer the issue in this case by comparing back-calculations (it might be fairer to say ‘back-guesstimations”) of Mr Bussey’s exposure against subsequently published figures of the kind appearing in TDN 13 is in my view unsound.”

It matters not whether the purpose of the Court was to test the level of exposure against a document such as TDN 13 or for the purpose of determining whether or not a ‘material’ risk had been created, the whole exercise is unsound and should not be undertaken to the point where experts are providing apparently detailed figures or even ranges.

  • Why? Because of the old computing acronym GIGO ‘Garbage In Garbage Out’. Such ranges/values are usually calculated using an evidence stock of an impressionistic history contained in lay witness evidence. At best the experts then seek to draw on analogies of exposure with the closest studies which ever measured such exposure. However, measurement did not become a precise science until instrumentation in the 1970s became available.
  • The preponderance of appellate authority (Bussey/Cox) is against the making of a precise finding of exposure because such precision is spurious when compared to the likely accuracy actually achieved. This is the practical reason why such exercises should not be undertaken. The deeper, structural reason why this should be so will be addressed in part 2, tomorrow.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.