Case Background

In this note Charles Feeny and Professor Damien McElvenny of the Institute of Occupational Medicine discuss the legal and epidemiological reasoning behind synergy.

The synergistic relationship between tobacco and asbestos was considered in the Australian case of Amaca Pty Ltd v Ellis 

[2010] HCA 5. The plaintiff was the executrix of her husband, Mr Cotton, who died of lung cancer. Mr Cotton had been a heavy smoker and was exposed to low levels of asbestos during his working life. The evidence at trial did not show that this was a cancer peculiarly associated with exposure to asbestos. The court was asked to determine whether it was more probable than not that the asbestos caused the cancer. Key to the plaintiff’s case was the argument that there was a synergistic relationship between tobacco smoke and asbestos.

At first instance the defendants were held liable. By a majority the appeal to the Court of Appeal of Western Australia was dismissed however, the appeal to the High Court of Australia was upheld.

Contributor Comments

Prof. McElvenny’s note (below) confirms the potent causative effect of synergy when there are significantly elevated risks, both from asbestos exposure and cigarette smoking. In these circumstances, the contribution to risk from exposure to asbestos when the synergy is taken into account is so strong that it is reasonable to infer that the victim would not have suffered from carcinoma in the absence of asbestos exposure. It may be that this perception of the effects of synergy has resulted in a belief that the increased risk of causation from synergy will result in causation by both factors, where both are known to be present. This erroneous belief appears to have been reflected in Lord Phillips’ comments in Sienkiewicz v Grief [2011] UKSC 10,

Epidemiological evidence indicated that, had the victim not been a smoker, his exposure to asbestos would have more than doubled the risk that he would get lung cancer. I do not, however, see that it was essential for the Claimant to prove this” (paragraph 76).

And,

For reasons that I have already explained, I see no reason for the application of the “doubles the risk” test in cases where two agents have operated cumulatively and simultaneously causing the onset of a disease. In such a case the ruling in Bonnington applies” (paragraph 90).

The Supreme Court in Sienkiewicz did not have the advantage of detailed evidence on relative risk and synergy in cases of carcinoma of the lung. However, in such cases the crucial issue is likely to be whether there was sufficient evidence of exposure to asbestos to make synergy a relevant factor which, in practical terms, involves proving at least a doubling of risk from asbestos exposure.

These issues were well illustrated and subject to detailed evidence in the Australian authority of Ellis (Cotton) v Amica Property Limited [2010] HCA 5. The case involved a man with slight asbestos exposure and a heavy smoking history. There were a number of estimations of relative risk from these two sources but the one specifically quoted in the decision of the High Court of Australia was that of Professor Berry, as follows:

  1. Due to smoking alone 92%
  2. Due to asbestos alone 0.1%
  3. Due to smoking-asbestos combination 0.9%
  4. Background risk 7%

The High Court in Australia, allowing the Defendant’s appeal, rejected an argument which was essentially to the effect that synergy must implicate asbestos in the causation of asbestos related carcinoma of the lung, where asbestos exposure was known to have occurred. This was clearly set out in the dissenting judgment of Martin C.J. in the first appeal to the Court of Appeal of Western Australia ([2006] WASC 270):

When the analysis undertaken by Mr Rogers (and others) is properly understood there is, with respect to the trial judge, no unspecified fallacy in the calculations. That is because, it does not follow from the synergistic effect that the instances of cancer due to a combination of asbestos exposure and smoking, must be greater than the incidence attributable to smoking alone in a group of cancer sufferers who have been exposed to both tobacco smoke and asbestos. Rather, the proportion within the group of cancer sufferers (as opposed to a group of community members generally) who would have suffered their disease irrespective of exposure to one or other of the carcinogens depends critically upon the assumptions made with respect to the levels of exposure. If the assumption is of a high level of exposure to tobacco smoke and a low level of exposure to asbestos, the evidence of all the experts (with the possible exception of Dr Leigh to whom I will refer below), was to the effect that the largest cohort within the group of cancer sufferers will be those who would have suffered lung cancer irrespective of exposure to asbestos. Conversely, if the assumption is of a group of cancer sufferers who have a slight history of smoking and a history of heavy exposure to asbestos, the largest group will be those who would have suffered lung cancer irrespective of their exposure to tobacco smoke. In either of these assumed cases, the group who would only have contracted lung cancer because of their exposure to both carcinogens will be much smaller than the group who would have contracted lung cancer because of their exposure to the predominant carcinogen, irrespective of their minor exposure to the other carcinogen.” (Paragraph 207)

The High Court of Australia made similar comments (at paragraph 57):

It may be accepted (at least for the purposes of debate) that the synergistic or multiplicative effect suggests that in some cases the two carcinogens will have contributed to the development of an individual patient’s cancer. But the proposition which the plaintiff advanced was an absolute proposition of universal application:  smoking and asbestos must work together and they must have worked together in this case. That proposition was not established.”

This point was reinforced by the court at paragraph 60:

If the description of exposure to smoking and asbestos as “more dangerous” than exposure to one or the other was intended to reflect a quantitative comparison of risk, it is a description that did not accurately reflect the evidence given by the witnesses about the relative risks of smoking compared with the relative risks of exposure to asbestos. And if the description “more dangerous” was intended to convey no more than that those who were exposed to both smoking and asbestos were at greater risk of developing cancer than those who were exposed to only one of those carcinogens, it is necessary to bear steadily in mind that the evidence did not establish that smoking and asbestos must work together.”

The Claimant’s arguments, based upon the evidence of Dr James Leigh, ultimately try to make the risk from asbestos exposure in carcinoma of the lung equivalent to the stochastic risk known to exist in mesothelioma claims. However, there are clear distinctions between the two. In a mesothelioma claim, unless there is some argument that the mesothelioma is idiopathic, the causation issue is that it could not be proved which of any particular exposures had been causative, as any one on its own could have been the cause. The Fairchild exception was essentially created to avoid the undesirable result flowing from the logic of this argument. However, in a carcinoma of the lung case, as in Cotton (Ellis) the issue is not which exposure was causative but whether the asbestos exposure could be considered causative at all. In this context, Professor Berry’s figures indicated that, even with the synergistic effect taken into account, the chance of the victim’s carcinoma being related to asbestos was at the 1% level, whereas there was a 92% chance that it was related to cigarette smoking alone. This, therefore, is an argument that can be disposed of on a conventional basis.

Clearly, within a cohort of 100 persons with similar histories and similar diagnosis, there will be one who in fact suffered carcinoma of the lung as a result of asbestos exposure, but as Patrick Walsh of Pannone pointed out, in the discussion in the webinar (Causation in asbestos: Minimal or material? Risk or probability), there is no way in the absence of any medical or pathological evidence that it can be established whether a particular victim was this one person, as opposed to being one of the 92 who suffered from carcinoma of the lung as a result of their cigarette smoking habit.

Synergy (asbestos and tobacco smoking and the risk of lung cancer)- Damien McElvenny

Using data from Hammond et al (Hammond EC, Selikoff IJ, Seidman H (1979). Asbestos exposure, cigarette smoking and death rates. Ann N Y Acad Sci 1979; 330: 473-490), the relative risk (this can be thought of as approximately the ratio of the probability of getting the disease if exposed divided by the probability of getting the disease if unexposed) of lung cancer from tobacco smoking versus not tobacco smoking was approximately 10, and in non-smokers, the relative risk from exposure to asbestos was approximately 5.  These risks may be higher or lower in other working populations depending on the level of asbestos exposure and the level of tobacco smoking, and so are used for illustrative purposes only.  These relative risks can be represented by the following table, where the relative risk from asbestos 5 (in non-smokers) is the value in bold:

Relative risk Relative risk of tobacco smoking
Smoker Non-Smoker
Relative risk of asbestos   Exposed 5.0
Unexposed 10.0 1.0

If the effects on lung cancer from tobacco smoking and asbestos exposure were completely independent of each other (i.e. didn’t interact in any way), then the excess relative risks (the amount of relative risk above 1.0) would be additive and the table of relative risks would be as shown below (as before the relative risks for asbestos are in bold):

Relative risk Relative risk of tobacco smoking
Smoker Non-Smoker
Relative risk of asbestos   Exposed 14.0 5.0
Unexposed 10.0 1.0

In this (non-interaction) scenario, for every lung cancer case in a non-smoker unexposed to asbestos, there will be an additional 13 cases of lung cancer, 9 of which are due to tobacco smoking and 4 of which are due to asbestos.  However, because the cases in practice will be clinically indistinguishable from each other, we cannot say which cases are due to which exposure, nor say which cases are the ones that would have happened anyway.  If we knew a case didn’t have either exposure (i.e. didn’t smoke and weren’t exposed to asbestos), they are most likely to be one of the background cases (i.e. a case caused by some other lung carcinogen).  If they had exposure to asbestos in the absence of tobacco smoking, then asbestos would be the most likely cause, and ditto for tobacco smoking in the absence of asbestos exposure.  Where it is a little more difficult, is in the situation where a lung cancer case has had both exposures.  Here, we would have to say on the balance of probabilities that the case is most likely due to tobacco smoking (9 cases due to tobacco smoking versus 4 due to asbestos exposure).  (This could be further refined if levels of exposure were taken into account).

If the effects are synergistic in a fully multiplicative way, which is believed to be approximately the actual situation, then we have the following table (as before the relative risks for asbestos are in bold):

Relative risk Relative risk of tobacco smoking
Smoker Non-Smoker
Relative risk of asbestos    Exposed 50.0 5.0
Unexposed 10.0 1.0

In this scenario, the individual effects of asbestos and tobacco smoking in the absence of each other (or both) are the same, but now we have 49 excess cases for every case unexposed to either causal agent, 36 (49-13) of which are due to the synergy or multiplicative effect of asbestos and tobacco smoke.  Under the synergy model, the majority of the lung cancer cases (36 out of 50) are due to the synergy.  Unlike for the additive model, for those cases with both exposures, we’re not able to reach a judgment on whether asbestos or tobacco smoking is the most likely cause.  Here (because they are clinically indistinguishable from each other), we can only say that they are jointly responsible (the “but for” test), since the removal of either cause, would prevent these 36 excess cases due to the synergistic effect from occurring.

Download the Legal Report