Class actions (Vioxx): causation & epidemiological evidence

Merck Sharp & Dohme (Australia) Pty Ltd v Peterson [2011] FCAFC 128

I refer to my earlier posting on the trial decision together with Amaca Pty Ltd v Ellis; Tabet v Gett and Seltsam Pty Ltd v McGuiness [2000] NSWCA 29; (2000) 49 NSWLR 262, which were applied. In so doing the “but for” test was affirmed and the “material increase in risk” test was disavowed. It also highlights the risk in such class actions where there are variable claimants with different conditions and responses.

Keane CJ, Bennett and Gordon JJ:

Causation: The Test


In Chappel v Hart (1998) 195 CLR 232 at [27] McHugh J said:

…If a wrongful act or omission results in an increased risk of injury to the plaintiff and that risk eventuates, the defendant’s conduct has materially contributed to the injury that the plaintiff suffers whether or not other factors also contributed to that injury occurring. …

This statement was approved in Naxakis v Western General Hospital [1999] HCA 22; (1999) 197 CLR 269 at [31] and [127].


In Seltsam v McGuiness at [107]-[108] and [119]-[120], Spigelman CJ emphasised that proof that “the risk eventuated” in the specific injury suffered by the plaintiff is part of the plaintiff’s burden, and that this burden is not discharged merely by showing an increased risk of injury by reason of the defendant’s conduct: “Whether or not the increased risk ‘eventuated’, is the issue which must be determined”. It must be shown that “Y had happened because of X” (emphasis in original).


That is the effect of the authorities on the test for causation under the common law in Australia. The position was summarised recently in Tabet v Gett (2010) 240 CLR 537 at [111]-[113]. Kiefel J, with whom Hayne, Crennan and Bell JJ agreed, said:


The common law requires proof, by the person seeking compensation, that the negligent act or omission caused the loss or injury constituting the damage. All that is necessary is that, according to the course of common experience, the more probable inference appearing from the evidence is that a defendant’s negligence caused the injury or harm. “More probable” means no more than that, upon a balance of probabilities, such an inference might reasonably be considered to have some greater degree of likelihood; it does not require certainty.

The “but for” test is regarded as having an important role in the resolution of the issue of causation, although more as a negative criterion than as a comprehensive test. The resolution of the question of causation has been said to involve the common sense idea of one matter being the cause of another. But it is also necessary to understand the purpose for making an inquiry about causation and that may require value judgments and policy choices.

Once causation is proved to the general standard, the common law treats what is shown to have occurred as certain. The purpose of proof at law, unlike science or philosophy, is to apportion legal responsibility. That requires the courts, by a judgment, to “reduce to legal certainty questions to which no other conclusive answer can be given”. The result of this approach is that when loss or damage is proved to have been caused by a defendant’s act or omission, a plaintiff recovers the entire loss (the “all or nothing” rule).

[Footnotes omitted].


The “but for” test serves, in this field of discourse, as a negative criterion. That is to say, unless the defendant’s actionable conduct is shown to be a necessary condition of the plaintiff’s injury, the plaintiff’s claim will not succeed. Thus, in Amaca v Ellis at [11]-[12], it was accepted that a plaintiff must show on the balance of probabilities that the actionable conduct of the defendant was a necessary condition of the occurrence of the harm in respect of which the plaintiff claims damages. It is true, as Counsel for Mr Peterson pointed out, that this rule was not the subject of argument in Amaca v Ellis; but it is also true that this rule represents the law in Australia binding on all courts below the High Court.


In some countries which share Australia’s common law inheritance this rule has been relaxed in recent years. In Evans v Queanbeyan City Council [2011] NSWCA 230 at [23][26] Allsop P explained how, in the jurisprudence of Canada and the United Kingdom, the concept of “material contribution” to injury may mean something different in some contexts from its meaning in Australian law. In some cases, a material contribution to an event may be shown by evidence of an increase in the risk of the event even though it is not possible to say that the event would not have occurred but for the conduct which gave rise to the increase in risk. Allsop P said (at [23]-[24]):

It can be accepted that in Resurfice Corp v Hanke at 342-343 [24]-[25], the Supreme Court of Canada expressed the framework for “material contribution” in the absence of a positive conclusion from the “but-for” test as follows:

However, in special circumstances, the law has recognized exceptions, to the basic ‘but for’ test, and applied a ‘material contribution’ test. Broadly speaking, the cases in which the ‘material contribution’ test is properly applied involve two requirements.

First, it must be impossible for the plaintiff to prove that the defendant’s negligence caused the plaintiff’s injury using the ‘but for’ test. The impossibility must be due to factors that are outside of the plaintiff’s control; for example, current limits of scientific knowledge. Second, it must be clear that the defendant breached a duty of care owed to the plaintiff, thereby exposing the plaintiff to an unreasonable risk of injury, and the plaintiff must have suffered that form of injury. In other words, the plaintiff’s injury must fall within the ambit of the risk created by the defendant’s breach. In those exceptional cases where these two requirements are satisfied, liability may be imposed, even though the ‘but for’ test is not satisfied, because it would offend basic notions of fairness and justice to deny liability by applying a ‘but for’ approach.

Further, it can be accepted that the House of Lords and now the United Kingdom Supreme Court has modified the common law in the United Kingdom by accepting that, at least as a special rule in mesothelioma claims, fairness and justice demand that the factual link between the wrongful act and the harm to the plaintiff need only be a wrongful exposure of the plaintiff by the defendant to an amount of asbestos that would materially increase the risk of mesothelioma: Fairchild; Barker v Corus; and Sienkiewicz. In Sienkiewicz, the statements of principle may also be seen not to be limited to mesothelioma.


Allsop P went on to note the policy considerations which led to the adoption overseas of a rule that increased risk may itself be sufficient to establish a material contribution for the purposes of causation. His Honour said at [27]-[31]:

…[W]hat was common to the several views of their Lordships in Fairchild was the necessity to depart from the common law’s usual rule of the need for proof on the balance of probabilities in circumstances where it could be unjust for there to be no recovery. See in particular Fairchild at 66 [32] per Lord Bingham of Cornhill, 69 [40] per Lord Nicholls of Birkenhead, 73 [56] per Lord Hoffmann and 112 [155] per Lord Rodger of Earlsferry. These views reflected the expression of the matter by Justice (now Chief Justice) McLachlin at the commencement of an article written by her Ladyship in 1998, “Negligence Law – Proving the Connection” in N J Mullany and A M Linden (Eds) Torts Tomorrow: A Tribute to John Fleming (LBC Information Services, Sydney 1998). The foundational legal and moral premise that can be seen in Fairchild and in the introductory comments in the above article is that tort law, as an aspect of the rule of law, is concerned with righting wrongful conduct. If self-evident wrongs (so characterised by legitimate human perceptions) are not recognised by the law’s rules, and thus go unremedied, people who legitimately feel themselves victims will be left with a sense of injustice. A legitimate sense of injustice should not be the product of the rule of law.

The avowed change to causal principle by the House of Lords that materially increasing the risk of injury was sufficient factual tortious involvement for causation (or attribution of responsibility) to be established was narrowly confined by all their Lordships in Fairchild: at 40 [2] and 55 [21] per Lord Bingham, 70 [43] per Lord Nicholls, 74 [61] per Lord Hoffmann, 91 [108] per Lord Hutton and 118 [170] per Lord Rodger. Crucial to that confinement were factors such as the causal element being singular (only exposure to asbestos) and the inability of medical science to explain cause in terms of a balance of probabilities.

Further illumination of the policy-based change to the common law in Fairchild can be seen from the lively debate in the judgments in Barker v Corus as to what was, in fact, decided in Fairchild. At issue in Barker v Corus was the extent of several liability of the wrongdoers who had exposed the plaintiff to the asbestos, causing a material increase in risk of contracting mesothelioma. Though all their Lordships (with a caveat by Lord Rodger: Barker v Corus at 610 [100]-[102]) came to the view that the exceptional approach in Fairchild should extend to circumstances where not all the exposure was tortiously caused by the defendants, a restriction was placed on the principle being that there must be one causative agent: Barker v Corus at 587 [24] per Lord Hoffmann (his Lordship recanting his view in Fairchild that this limitation was unprincipled), 599 [64] per Lord Scott of Foscote, 611 [104] per Lord Walker of Gestingthorpe and 615 [121] per Baroness Hale of Richmond. Save for Lord Rodger, all were agreed that the extent of responsibility of individual defendants should be measured by reference to the extent to which they had increased the risk, based on the respective length and intensity of exposure: Barker v Corus at 589-590 [35]-[36] per Lord Hoffmann, 599 [62] per Lord Scott, 612 [109] per Lord Walker and 616 [126] per Baroness Hale. The majority, recognising that it was justice and fairness that had given rise to the exceptional rule, thought that justice and fairness should limit the defendants’ respective liabilities by reference to the wrongful exposure. Lord Rodger dissented on this point – all defendants were liable in full, the change in rule in Fairchild being a relaxation of the rules of causation – thus each caused the damage.

At this point, the Parliament intervened with the Compensation Act 2006 (UK), which reflected Lord Rodger’s view in Barker v Corus as to the responsibility of defendants in mesothelioma cases, and not the majority’s view of fairness.

The above discussion of Fairchild and Barker v Corus reveals, at once, the policy questions involved in any conclusion that increasing risk is sufficient for a conclusion of causation or causal responsibility or legal responsibility. Such policy questions are a matter for the High Court, not this Court.


Allsop P summarised the current position in Australia as follows, at [22]:

…Subject to the views of the High Court in respect of any development of the common law or to the operation of any legislation, it can be concluded that at common law, as a general proposition, the increasing of risk of harm by a tortious act is, alone, insufficient for a conclusion of causation by material contribution to that harm or for a conclusion of responsibility in law for that harm.


We respectfully agree with Allsop P.


The rule that a plaintiff must establish as a necessary condition of recovery that he or she would not have suffered loss but for the defendant’s actionable misconduct is deeply rooted in the policy of the common law that one person should not be liable for the loss suffered by another unless the plaintiff can establish that the defendant’s actionable conduct caused the plaintiff’s loss. It is not open to this Court to decide that we should no longer adhere to this rule and that a different and “better” rule should henceforth be applied. This Court must proceed on the footing that for Mr Peterson to show that the consumption of Vioxx materially contributed to his MI, in the sense relevant in Australian law, he is obliged to show that his consumption of Vioxx was a necessary condition for the occurrence of the heart attack on 8 December 2003. To say that the consumption of Vioxx was, for example, “in the mix” of possible causes is not enough in this regard. As Beazley JA said in Bendix Mintex Pty Ltd v Barnes (1997) 42 NSWLR 307 at 339, the onus of proof of causation “is not discharged by establishing that a particular matter cannot be excluded as a cause of the injury”.


We conclude that his Honour’s ultimate findings of fact are not sufficient, as a matter of Australian law, to sustain the determination of the issue of causation in Mr Peterson’s favour. We also consider that the conclusion of ultimate fact necessary to support such a finding is not open on his Honour’s findings of primary fact and we now turn to explain why we take that view.


Causation: Epidemiology as a strand in the proof of causation


A difficulty, as a matter of legal policy in the way of adopting the rule of thumb adopted in the United States cases, is that it is apt to mandate an award of compensation to applicants who have not, in truth, been injured by the respondent. That is because those applicants who were actually injured by causes other than the respondent’s actionable conduct will be able to recover compensation because, for them too, a relative risk of greater than 2 can be said to imply probability of greater than 50% that the respondent’s actionable conduct was the cause of their loss.


It should also be borne in mind that, while a relative risk of 2 might imply a 50% probability that the risk has come home in a typical case, a relative risk of less than 2 would imply a probability of less than 50%, that is to say less probable than not. The primary judge’s finding of relative risk was of “about 2”. And in any event the strength of that finding as a strand in Mr Peterson’s case is problematic because of the other candidates as causes of his injury. Further, the absence of a clear appreciation of the level of absolute risk to Mr Peterson from the actionable conduct of Merck and MSDA in comparison with those other candidates detracts from the force of the circumstantial case which he seeks to mount.


In Amaca v Ellis at [62], the High Court emphasised that the significance of an epidemiological study depends upon whether the plaintiff is a typical member of the population which is the subject of the study. Their Honours said:

As explained at the commencement of these reasons, there being no direct evidence about what actually caused Mr Cotton’s cancer, it was the plaintiff’s case that the epidemiological evidence established facts which “positively suggest[ed], that is to say provide[d] a reason … for thinking it likely” that, in exposing Mr Cotton to respirable asbestos fibres, the negligence of each defendant was a cause of his cancer. To draw an inference about causation from what was established by the epidemiological studies, it would be necessary to decide whether the particular case under consideration should be treated as conforming to the pattern described by the epidemiological studies. Absent evidence which suggests that the individual may stand apart from the ordinary, there may be sufficient reason to assume conformity, but whether or not that is so, it is important to recognise that the first step that must be taken, if an inference is to be drawn from epidemiological studies, is to relate the results of studies of populations to the particular case at hand. That step is not inevitable.


[Footnotes omitted].


In this case, as has been seen, there was a clear basis for concluding that Mr Peterson does indeed stand apart from the ordinary case. His personal circumstances were such that they afford a ready explanation for the occurrence of his injury independent of the possible effects of Vioxx. The strength of the epidemiological evidence as a strand in the cable of circumstantial proof is seriously diminished by this consideration. The epidemiological studies do not provide assistance in resolving the question whether it was the risk posed by Vioxx, either alone or in combination with the other candidates, which did eventuate in this case.


What then is the position here? The primary judge upheld Mr Peterson’s claim under s 74D on the basis that Vioxx was not as fit for the purpose for which goods of that kind were commonly bought as it was reasonable to expect because “they increased the risk of heart attack by a factor of about 2” in the population of the APPROVe study (Reasons at [975]). However, the considerations in s 74D(3) make it clear that an observation that there was an increase in risk “of about 2” does not, in this case, without more, demonstrate unfitness for purpose.


Brisbane Barrister – David Cormack

Related Posts

Recent Comments