From: David Cheifetz <david.cheifetz@rogers.com>
To: Stephen George Alexander Pitel <spitel@uwo.ca>
obligations@uwo.ca
Date: 10/07/2012 03:20:20 UTC
Subject: Re: Class Action Trial Decision - Medical Devices - Causation
There's another unusual aspect of the decision which might mean it's not going to be the trial judge's last word. The Anderson reasons were released on June 26/12.
 She wrote at paras. 569-70:
 
[569] In the present circumstances, I believe the British Columbia Court of Appeal’s words in Moore v. Castlegar and District Hospital are apposite.[132] In that case, the Court held that it is not open to a trial judge to draw a common sense inference of the cause of the medical complication where both parties have led expert medical evidence of causation. Moore was cited with approval in Sam v. Wilson, a case in which Snell was distinguished for similar reasons.[133]
 
[570] In the present case, the two sides have adduced conflicting expert testimony. Further, there is simply no reliable evidence, other than the epidemiological evidence, upon which I could base an inference of causation. Thus, I cannot apply the robust and pragmatic approach as it was outlined in Aristorenas v. Comcare Health Services to draw an inference of causation. In that case, the court stated that “a series of facts and circumstances established by the evidence led at trial may enable the trial judge to draw an inference even though medical and scientific expertise cannot arrive at a definitive conclusion”.[134] In the present case, the “series of facts and circumstances” upon which I could base such an inference is absent. The only reliable evidence of causation is epidemiological evidence, and I have interpreted that evidence consistently with how it is treated by qualified experts in the medical and scientific communities
But, on June 29, the SCC, in Clements v. Clements, 2012 SCC 32 at para. 9 (an appeal from a BCCA decision) overruled (in my view) that line of authority on the use of the robust common sense approach, albeit without mentioning the BCCA cases. However, the admonition is clear enough:"The 'but for' causation test must be applied in a robust common sense fashion. There is no need for scientific evidence of the precise contribution the defendant’s negligence made to the injury." British Columbia, I believe, is the only jurisdiction in Canada that has rule that the common sense approach to but-for (whatever it means) can't be used when expert evidence is called by both sides on factual causation.
Anderson is the first time that rule has been applied - perhaps the first time those cases have been cited - in an Ontario decision. This trial judge would know, and would have mentioned, the rule's use in Ontario if it had been used before.
 
The BC rule may or may not, before Clements, have extended to instances where only one side calls the evidence. In Clements, the only expert who testified on causation was called by the defence. The plaintiff cross-examined. Does that mean that both sides "led" expert evidence. (The trial judge used some of the cross-examination answers to reject the expert's opinion.)
 
Will this make the trial judge reconsider her but-for analysis, or would the trial judge say her decision would have been the same regardless? Or decline to reconsider at all since the causation discussion is obiter? I believe it is still open to her to withdraw her reasons and reconsider since she isn't yet functus.
 
David
 
 
 

 
From: David Cheifetz <david.cheifetz@rogers.com>
To: Stephen George Alexander Pitel <spitel@uwo.ca>; "obligations@uwo.ca" <obligations@uwo.ca>
Sent: Sunday, July 8, 2012 1:40:22 PM
Subject: Re: Class Action Trial Decision - Medical Devices - Causation
Stephen,
 
The technically correct answer for why common issue 3 was stated as it was is the phrasing of common issue 3 wasn't defined by the trial judge. It was formulated in the certification motion 9 years ago, in 2003: see http://www.canlii.org/en/on/onsc/doc/2003/2003canlii5686/2003canlii5686.html at [63]. The certification motion was argued on the basis that causation law was as set out in Athey as Athey was understood then.
 
But, since we, here, get to discuss the merits, and if we assume that it were open to the trial judge with or without the parties' consent (even assuming that were plausible - I've no idea) to restate this common issue because the law had changed since 2003 -
 
then the reference to "materially increase the risk" is valid only if the trial judge was prepared to treat statistically doubling the risk as a sufficient basis, in some cases, from which to draw a valid inference of specific causation on the balance of probability. My impression from my initial glance through the reasons is that that was her approach. The trial judge could have looked at cases such as Sienkiewicz on the use of statistical evidence and how that relates to the problem of drawing inferences under but-for, but that's different from the increase in risk test offered as an alternative to the but-for test.
 
David
 
From: Stephen George Alexander Pitel <spitel@uwo.ca>
To: David Cheifetz <david.cheifetz@rogers.com>; "obligations@uwo.ca" <obligations@uwo.ca>
Sent: Sunday, July 8, 2012 12:03:07 PM
Subject: Re: Class Action Trial Decision - Medical Devices - Causation
David,
 
Fair enough in terms of the cases.  But less fair given how common issue 3 was phrased.  So is there a reason why common issue 3 was phrased as it was?  Or why it was even a relevant inquiry?
 
Stephen
 
 
On 07/08/12, David Cheifetz <david.cheifetz@rogers.com> wrote:
Stephen,
 
The reasons were released on June 26, 2012. I agree the judge's phrasing of the causation question is odd for a but-for analysis. But, as of June 26, regardless of her phrasing of the question, the trial judge would have had to ignore Ontario Court of Appeal decisions binding on her to even suggest there could be a separate causation test based on negligence increasing the risk of the occurrence of the injury which occurred, rather than negligence being a cause of that injury. Clements wasn't released until June 29. The trial judge would have known that, over the past 5 years, the ONCA had shown absolutely no interest in engaging with any of the decisions on an increase in risk test, not even the Canadian. Two ONCA decisions (Barker v. Montfort Hospital; Monks v. ING) asserted Resurfice hadn't changed the law at all. One ONCA decision (Fraser v. Haukioja) asserted that the Resurfice version of an increase in risk test produces a decision that factual causation is established on the balance of probability. No ONCA decision even formally conceded that there could be a separate causation test based on increase in risk. I'd have been astonished if the trial judge had any devoted time to an obiter discussion of a doctrine which didn't officially exist in Ontario.
 
Regards,
 
David
 
 
From: Stephen Pitel <spitel@uwo.ca>
To: "obligations@uwo.ca" <obligations@uwo.ca>
Sent: Sunday, July 8, 2012 8:45:23 AM
Subject: Class Action Trial Decision - Medical Devices - Causation
The Ontario Superior Court of Justice has released Andersen v St Jude Medical Inc, 2012 ONSC 3660 (all 595 paragraphs of it).  The case is a rare trial on the merits of a class claim in respect of prosthetic heart valves and annuloplasty rings.  The decision is available at

http://www.canlii.org/en/on/onsc/doc/2012/2012onsc3660/2012onsc3660.html

Much of the decision deals with whether the defendant breached the standard of care.  The court concludes it did not.  In terms of the outcome, one could stop reading at that point.

There is then a considerable discussion of issues of causation, had the court found differently on the breach issue.  Somewhat oddly, common issue three is phrased as:  "Does a Silzone coating on heart valves, or annuloplasty rings, materially increase the risk of various medical complications including, but not limited to, paravalvular leakage, thrombosis, thromboembolism, stroke, heart attacks, endocarditis or death?"  While this is clearly put in terms of material increase in risk, the court's causation analysis seems entirely on the basis that the relevant test is, and was intended in stating the common issue to be, but-for on the balance of probabilities.  The court does not engage with the leading Canadian, much less British, decisions on using an increase in risk test.

There is also an interesting discussion of "waiver of tort".  I usually take this to mean that the plaintiffs are seeking disgorgement damages as their remedy in tort rather than compensatory damages, which is commonly raised in class claims so each plaintiff does not have to establish individually caused damage and the appropriate level of compensatory damages.  This is tricky in negligence cases since damage is an element of the cause of action.  The law on this in Canadian class actions is a mess, as noted at paras 578-594, and the judge here does not clean it up, despite acknowledging hopes in the legal community that she would do so (para 577).  But some of the policy comments are likely to stir the debate further.

The decision also contains much discussion of the role of scientific evidence in this sort of litigation.

Stephen
-- Western Law Dr. Stephen G.A. Pitel Associate ProfessorGoodmans LLP Faculty Fellow in Legal Ethics 2012-13 Faculty of Law, Western University (519) 661-2111 ext 88433