David Platt KC and John Williams successful in Court of Appeal on low dose asbestos claims and what amounts to “de minimis” in mesothelioma cases
On the 15th April 2025 the Court of Appeal handed down judgment in the “de minimis” mesothelioma case of Alexander Johnstone (Personal Representative of the Estate of Elaine Johnstone, Deceased) v Fawcett’s Garage (Newbury) Limited [2025] EWCA 467. The judgment can be found here.
In its joint judgment the Court of Appeal (Coulson LJ, Nicola Davies LJ and Zacaroli LJ), dismissed the Claimant’s appeal against the judgment of His Honour Judge Simon (sitting as a Deputy High Court Judge) in which he dismissed the claim because the Claimant had failed to prove causation. The Court of Appeal also dismissed the Respondent’s Notice which sought to uphold the Judge’s finding on causation on additional grounds.
BACKGROUND
Following Fairchild v Glenhaven Funeral Services Limited [2003] 1 AC 32, causation in mesothelioma claims is the subject of a special rule which requires the claimant to prove the tortious exposure to asbestos materially increased the risk of the claimant/deceased contracting mesothelioma. In Sienkiewicz v Greif (UK) Limited [2011] 2 AC 229, the Supreme Court held that the special rule applies also to cases involving a single exposer. Sienkiewicz did not define what amounted to a “material increase in risk” but expressed the obiter view that it was a question of fact for a trial judge. How a material increase in risk may be proved in claims involving very low dose exposure to asbestos has been considered in a number of subsequent cases.
In Bannister v Freemans Public Limited Company [2020] EWHC 1256 (QB), the defendant adduced expert medical evidence at trial which drew on the important epidemiological study by Hodgson & Darnton (“The Quantitative Risks of Mesothelioma and Lung Cancer in Relation to Asbestos Exposure”: Ann Occ Hyg (2000) 44 (8); 561-601) (“H & D”) to estimate the risk of mesothelioma attributable to low level exposure to different fibre types and to support an “absolute risk” argument that the special rule of causation was not satisfied. This approach was accepted by the trial judge in Bannister but, because he found the claimant had failed to prove any exposure to asbestos, this part of the judgment was obiter. The basis for the judgment in Bannister was recently approved by the Scottish Outer House in Kerr v. Midlothian Council [2024] CSOH 112.
In Johnstone, the Defendant again pleaded a de minimis defence to a low dose mesothelioma claim and advanced arguments on causation based on expert medical evidence referenced to the H & D paper and other research by Peto & Rake (“P & R”) which addresses the risks of idiopathic mesothelioma and mesothelioma from environmental exposure to asbestos. This time however, the Claimant sought permission to adduce expert epidemiological evidence from Professor Norrie (Professor of Biomedical Statistics, Edinburgh University) to address these risks by way of critique of the relevant epidemiological studies. The Claimant’s basic argument was that the available epidemiology was unreliable and could not be properly used to calculate any sort of statistical risk estimate. The Defendant relied on epidemiological evidence from Professor Steve Jones.
FINDINGS AT TRIAL
The Deceased had been employed in the office of the Defendant’s Newbury garage between 1982-1989. During this time, she was exposed to small quantities of asbestos dust from work on asbestos brake & clutch linings (containing predominantly chrysotile asbestos) undertaken by a mechanic in an adjacent workshop. The office was separated by a door which was ordinarily closed.
The first key issue at trial was to estimate the Deceased’s occupational exposure to asbestos. The parties’ respective OH experts arrived at markedly different estimates but, having found that the Claimant’s expert (Mr Chambers) had misinterpreted the factual evidence and thereby used inappropriate data, the Trial Judge preferred the much lower estimate of the Defendant’s expert (Mr Stear) of between 0.001 to 0.002 f/ml-y. In reaching his findings, the Trial Judge rejected a late submission from the Claimant (not addressed in any of the written or oral evidence) that the Court should draw an adverse inference against the Defendant because it was in breach of the statutory duty under regulation 15 (1) of the Control of Asbestos at Work Regulations 1987 (in force from 01/03/88) to undertake air monitoring of work with asbestos containing materials.
On the issue of causation, the Trial Judge preferred the expert evidence of Professor Jones and Dr Moore-Gillon that the application of the H & D risk model to the Deceased’s estimated occupational exposure together with an assessment of the Deceased’s baseline risk of mesothelioma as identified in the P & R research led to a lifetime increase in relative risk “in the order of 0.1% or less” which was not a “material” increase. The Trial Judge did not discretely address the Defendant’s further argument based upon medical opinion as to the “absolute” risk arising from the occupational exposure (i.e. that any such overall risk was not “material” in medical terms).
THE JUDGMENT OF THE COURT OF APPEAL
Adverse Inference
On this issue & although “not entirely clear from the grounds of appeal”, the Claimant argued that the Trial Judge erred in law in failing to use the adverse inference principle to dismiss the expert evidence of Mr Stear and to find as proved the expert evidence of Mr Chambers. The Court of Appeal noted that the basis upon which it was argued an adverse inference should be drawn and the nature of that inference varied during both the trial and the appeal – with the Claimant ultimately settling upon the absence of air monitoring records for the mechanic’s work.
The Court of Appeal reviewed the modern authorities on adverse inference (Keefe v Isle of Man Steam Packet Co Limited [2010] EWCA Civ 683; Shawe-Lincoln v Neelakandan [2012] EHC 1150 and Mackenzie v Alcoa Manufacturing (GB) Ltd [2019] EWCA Civ 2110. The Court observed that in Keefe, the principle had been deployed in a case where there were two competing versions of the facts and where one side had been hampered by the other’s failure to adduce written or oral evidence. Here however, there was no conflicting evidence of fact on which the Trial Judge could look critically (#38). The Court (#37) did “not suggest that adverse inferences can never be applied to expert evidence” but “the adverse inference submission in this case makes a leap of logic which is unsustainable” for the multiple reasons given by the Court (#26-36). Ultimately, Mr Chambers’ calculations of dose were “far removed” from the factual findings made by the Trial Judge and “in those circumstances the adverse inference principle cannot extend to an effective deletion of those findings of fact and an uncritical acceptance of Mr Chambers’ calculations.”
Causation
The Court of Appeal described (& re-named) the different approaches taken at trial to assessing material increase in risk; viz-
- The “exposure/risk” approach argued by the Claimant in which a material increase in risk is equated to a material increase in exposure, but with account being taken for the Deceased’s “background exposure” to asbestos in assessing the statistical increase in risk: see #56-61.
- The “direct risk assessment” approach adopted by the Trial Judge based on a comparison of risk from occupational exposure (estimated from H & D) and the “background risk” (estimated from P & R): see #62-76 where this approach is explained.
- The “absolute risk” approach in which the risk from occupational exposure is estimated (again using H & D) and is then the subject of medical opinion as to whether this risk is “medically significant”. This was the “Bannister” approach: see #77-81.
The Court of Appeal noted that it was not for it to assess the evidence heard at trial on these different approaches and thereby decide determinatively whether the “direct risk” approach adopted by the Trial Judge is, generally, a valid and appropriate method. Instead, the Court was required to address the 4 Grounds of Appeal.
Ground 1 related to the Judge’s treatment of Professor Norrie’s evidence. This was in essence an appeal against a finding of fact. Having reviewed the various passages in the evidence relied upon by the Claimant as evidence that Professor Norrie considered the H & D study to be worthless, the Court of Appeal concluded that, properly analysed, Professor Norrie was “not as dismissive of using the model in the H & D Study in relation to much lower rates of exposure as the Appellant contends” (#99). In any event, where there was a difference between the experts, the Trial Judge had accepted the evidence of Professor Jones that the H & D risk model could be relied upon with appropriate caution.
Ground 2 related to the Trial Judge’s treatment of Professor Jones’ evidence regarding the incidence of idiopathic mesothelioma reported in the P & R paper. However, whilst it was correct that the Trial Judge had failed to consider Professor Jones’ revised opinion on this issue in his oral evidence, the Court Appeal accepted the Defendant’s argument that this would have made no difference to the Judge’s overall finding on materiality as the increase in risk was still less than 0.1%. In dealing with this issue, the Court of Appeal questioned the Claimant’s argument that any calculation of increase in risk should default to the most favourable figures (for the Claimant) and further noted that neither it, nor the Trial Judge, had heard submissions as to where on the spectrum the point of materiality would be met.
Ground 3 involved broader challenges to the “direct risk assessment” approach used by the Trial Judge based upon: (1) alleged improper use of statistics and (2) the Claimant’s contention (as a matter of legal principle) that the creator of the risk must bear the consequences of any and all evidential difficulties of uncertainties. As to (1), the Court of Appeal concluded that there was nothing in the passages in Sienkiewicz relied on by the Claimant to support the proposition contended for. As to (2), the Court of Appeal agreed that the “creator of risk” principle derived from McGhee had informed the development of the special rule in Fairchild but it did not follow that, in applying the special rule, the same principle again applied to evidential uncertainties a claimant might encounter in proving a material increase in risk. In any event, the Court of Appeal found no fault with the Trial Judge’s approach to the epidemiological evidence. First, issues of risk necessarily involve statistics, epidemiological research and studies. Second, uncertainties within such studies do not devalue the principle that such evidence may be relied upon, but it can affect the weight to be ascribed to any epidemiological figure. Third, in this case, the assessment of material increase in risk was not simply a matter of statistics. Fourth, the specific criticism that the conclusions of the P & R research regarding background risk could not be personalised to the Deceased were misplaced as this was an observational study of a population cohort in which the Deceased sat & which provided a “depth and relevant detail” not provided in the alternative data sets of “background exposure” levels relied upon by the Claimant in the “exposure/risk” approach. Ultimately, it was not a misuse of statistics to consider the fact that chrysotile alone (albeit contaminated with tremolite) was present in the garage while leaving unresolved the uncertainty as to the forms of asbestos to which the Deceased could otherwise have been exposed.
Ground 4 complained that the Trial Judge had not explained why he preferred the “direct risk assessment” approach to the “exposure/risk” approach. However, the Court of Appeal noted that the Trial Judge had preferred the evidence of Professor Jones to that of Professor Norrie and had given reasons why he preferred the approach of Professor Jones. That was sufficient and there was no merit in this ground of appeal.
The Respondent’s Notice
In considering the “absolute risk” approach raised at trial (but not relied upon by the Trial Judge to dismiss the claim), the Court of Appeal concluded that this approach “did not assist in proving the special rule of causation” as it did not (unlike the “direct risk assessment” approach) assist in attribution and would, if adopted, make the word “increase” within the special rule otiose. Further, the approach seeks to “take advantage of the difficulty in proving causation in industrial disease cases” such that even a large material increase in risk of developing mesothelioma caused by a defendant could mean a claimant would not have a cause of action if their risk of developing the disease was deemed medically insignificant. This, according to the Court of Appeal, “flies in the face of the reasons for introducing the Fairchild exception” and also involves delegating what is an issue for the trial judge to the medical experts. Finally, the absolute risk approach taken in Bannister was not binding on the Court of Appeal- it being noted that epidemiological evidence was not available to the Court in Bannister.
DISCUSSION
- The approach taken by Mr Chambers on behalf of asbestos claimants has been the subject of previous comment in News items on this website: see the note on Lawton v Boots plc (11/06/24). In this case, Mr Chambers based his dose calculation on a misinterpretation of the factual evidence. Reports by Mr Chambers on behalf of asbestos claimants require close scrutiny.
- Professor Norrie has been instructed by claimants in several low dose mesothelioma claims to challenge/undermine the H & D risk model. This case was his “first foray into the world of asbestos-related disease” with (as the Trial Judge noted) some misunderstanding and over-simplification of the principles and practice of occupational epidemiology. As the Court of Appeal’s analysis of Ground 1 confirms, Professor Norrie did not come up to proof as regards any opinion that the H & D risk model is worthless.
- The Court of Appeal’s treatment of the adverse inference principle highlights its use in cases involving disputed factual evidence where one party has been placed at a disadvantage but emphasises the difficulties of applying the principle to disputed expert evidence – at least where the dispute arises from one expert’s misinterpretation of the factual evidence. The judgment also highlights the need to decide precisely what adverse inference one is inviting the Court to draw and, given the need to consider all the circumstances of the case, to “sense-check” that inference against both logic and other evidence in the case. It also requires that the inference be properly identified and the implications be canvassed at trial and in evidence.
- The judgment brings welcome clarification as to the application of the “creator of the risk” principle in mesothelioma claims. Notwithstanding the relaxed rule of causation in mesothelioma claims and statements from various senior Judges as to the need for caution in applying and extending the Fairchild rule, the Claimant in this case argued that he should not bear the burden of any difficulty in proving that the Deceased’s exposure led to a material in risk of mesothelioma. That argument was rejected as jurisprudentially unsound and, by implication at least, liable to distort the balance of competing interests reflected in the decision that mesothelioma claims should be the subject of a special rule of causation.
- The judgment is a helpful re-statement of the important role of epidemiology evidence in disease and other injury litigation. The issue raised by the Claimant on appeal was wrongly characterised as an “improper” use of statistics when, in fact, the real issues were the weight to be attached to epidemiological evidence and the relevance of that evidence to the circumstances of the claimant/deceased. Indeed, the very concept of an “increase in risk” almost always will require some use of epidemiology and statistics. Once the correct issues had been identified, the Court of Appeal adopted an entirely orthodox approach to epidemiological evidence – dismissing the Claimant’s appeal against the Trial Judge’s treatment of the epidemiological evidence.
- The nature of the appeal was such that the Court of Appeal did not rule definitively on which approach should be used to determine causation in mesothelioma cases. Much will depend on the evidence available to the Court. For example, in Ness v Carillion Capital Projects Limited [2023] EWHC 1219, the trial judge took an “exposure/risk” approach based on the limited evidence adduced at trial. In the present case (and as noted by the Court of Appeal), the Trial Judge had available to him a much wider portfolio of expert evidence which justified assessing causation by reference to the “direct risk assessment” approach. Of course, the “exposure/risk” approach advocated by the Claimant (and which would have produced an increase in risk figure of between 2%-4.3%) is itself dependent upon “back extrapolation” from studies of the risks associated with much higher occupational exposures – albeit on the assumption that the exposure/risk ratio is a linear one (an assumption challenged by the H & D risk model). Furthermore, in the absence of an established threshold, the “exposure/risk” approach also leaves open the question how much of an increase is required before it is deemed “material.”
- The Deceased’s dose in the present case was higher than the assumed dose in Bannister. Despite the Claimant’s arguments regarding tremolite contamination, it was properly regarded as exposure to chrysotile as the “chrysotile” studies are studies of “contaminated” chrysotile. Of course, the H & D risk model allows estimation of risks from low exposures to amphiboles as well such that the “direct risk assessment” method may also be used in such cases.
- Given the Court of Appeal’s decision to uphold the Trial Judge’s findings and Order, its comments on the “absolute risk” approach are obiter. This approach has the practical advantage of usually requiring expert medical opinion only (as per Bannister) and, therefore, can avoid the need for expensive epidemiological evidence. It also draws on expert medical evidence to express an opinion on causation – something that is entirely commonplace in personal injury claims which fall to be determined by reference to the “but for” and “material contribution” tests of causation. That is not to say that epidemiological evidence is necessary outside the Bannister approach – as the medical expression of opinion is itself based on epidemiology. That epidemiology can be disputed.
- As for the reasons advanced by the Cour of Appeal for rejecting the “absolute risk” method, causation is always a question for the trial judge assisted by expert medical evidence. In claims for noise induced hearing loss (where the “de minimis” issue relates to the contribution that any given tortious exposure to noise has made to overall hearing loss), it is commonplace for medical experts to opine on the questions of contribution to organic damage and “materiality.” As such, it is difficult to see why expert evidence on these issues amounts to a “delegation” to the medical expert or why the fact that medical opinions vary should matter.
- More than this, the obiter opinion of the Court of Appeal is somewhat difficult to reconcile with what is in fact necessary to understand proportionate conceptions of risk. The Court of Appeal appeared to favour a purely statistical exercise in assessing whether there had been a “material increase in risk” (i.e. a purely relative comparison with background risk). However, there is a serious danger that this process would become a dry theoretical and arithmetical exercise divorced from any medical or real world considerations. As such, and pursuant to such logic, an increase in risk from 0.000001 to 0.00001 in a million could be represented as a 10 fold increase in risk and therefore “material”. However, the practical reality must surely be that this risk increase never reaches the threshold of “materiality” because the overall impact is so tiny as to be legitimately ignored. This latter conclusion must be given by a medical expert and not simply by an epidemiologist.
- As such, we do not believe that this section of the Court of Appeal judgment will withstand detailed scrutiny in a future case.
