By Peter J Pliszka of Fasken Martineau DuMoulin
During the relatively short history of class proceedings in Canada, developers and manufacturers of medical devices and pharmaceuticals (medical products) have found themselves at the receiving end of a significant number of class action claims. To date, the main battleground in respect of these claims has been the certification motion, a preliminary step in which the court assesses whether or not a given claim should proceed as a class action.
Overall, the certification case law weighs heavily in favour of certifying proposed class proceedings involving medical products. However, a recent series of court decisions dismissing certification motions in medical products class actions signals that a new trend may be afoot in Canada.
This article will discuss these developments in Canada and consider whether the recent series of dismissals of certification motions represents a trend in the law toward raising the threshold that plaintiffs must meet to obtain certification of a proposed class action.
Until recently, Canadian courts had set what appeared to be an exceedingly low bar or threshold to be met for each element of the certification test, particularly in the context of medical products claims. This trend made it very challenging for defendants in class proceedings to defeat certification.
However, several court dismissals of certification motions in medical products class actions over the past three years suggest that Canadian courts may now be questioning that liberal approach to certification, and might be inclined to scrutinise more discriminately each proposed class action case to determine whether it really is reasonable and fair to allow it to proceed as a class action.
This development began with the May 2012 decision in Martin v AstraZeneca Pharmaceuticals PLC. There, the Ontario Superior Court refused to certify a class action relating to the antipsychotic drug Seroquel and the side-effect health risks it is alleged to cause. Remarkably, this 2012 decision was the first Ontario Superior Court decision refusing to certify a proposed class action involving a pharmaceutical product.
Seroquel is an antipsychotic medication approved by Health Canada for use in the treatment of schizophrenia, the acute management of manic episodes, bipolar disorder and depression associated with bipolar I and II disorder. The drug was also being prescribed for certain off-label uses, including treatment of anxiety, sleep disorders, and dementia. The plaintiffs made a broad range of allegations against AstraZeneca, including that the drug caused significant health risks to those who used it and that the defendants had failed to warn class members of these risks, which included significant weight gain, problems with balance, elevated blood sugars, hyperglycaemia, loss of energy, numbness in the extremities, pancreatitis, blindness, et al. The plaintiffs also alleged, among other things, that the defendants had been negligent in the design, manufacture and distribution of the drug, and that the defendants had conspired to conceal information from Health Canada and to promote the drug for certain off-label uses for which it had not been approved. The proposed class was “all persons in Canada who were prescribed, and who consumed, Seroquel”.
In denying certification, Justice Horkins commenced with a recitation of the well-established legal principle that, on a certification motion, the evidentiary burden on plaintiffs is low: all that the plaintiffs are required to do is to adduce evidence that shows “some basis in fact” to meet the requirements of s. 5(1) (b) to (e) of the test for certification as a class action. Justice Horkins also noted that a defendant is entitled to deliver evidence in rebuttal, but the standard of proof on the defendant is inversely heavy: the defendant must show that there is no basis in the evidence for the facts asserted by the plaintiffs. Nevertheless, and notwithstanding the extensive evidence filed on the motion by both the plaintiffs and the defendants, Justice Horkins found that plaintiffs had failed to satisfy every element of the certification test.
In this respect, the court ruled as follows:
The plaintiffs’ pleading was “seriously deficient” and failed to disclose a cause of action. In the court’s view, the pleadings were inconsistent, lacked clarity, failed to provide particulars, lacked material facts, and failed to identify the specific acts undertaken by the defendant which supported their claim.
The plaintiffs failed to provide a sufficient evidentiary basis to establish that a class of two or more persons exists (it was not sufficient for plaintiffs’ counsel to simply point to potential class members who had contacted them, without any evidence about the nature of the contact) and the proposed class definition was over-broad (it should have, at a minimum, been bound by a start date when Seroquel was first introduced in Canada).
None of the common issues proposed by the plaintiffs were issues common to the class, the resolution of which would significantly advance the proceeding. This included the “general causation” question of whether Seroquel can cause weight gain and diabetes. The court held that, even if this common issue was resolved in favour of the plaintiffs, the plaintiffs did not provide any evidence to show that a methodology exists whereby general population data (or some other approach) could be used to assess this issue in common and arrive at an answer that would be of any use to the class. Each plaintiff would still have to prove that Seroquel caused his or her weight gain and/or diabetes.
Given that there was no single common issue that would significantly advance the litigation for the class, the court felt there was no reason to conclude that a class action would be a fair, efficient and manageable method for advancing the claim.
The proposed representative plaintiffs were not suitable candidates to represent the proposed class because they were not adequately informed about the action and did not have a real interest in the action.
Justice Horkins’ decision is instructive for both plaintiffs and defendants in medical products class proceedings. The judge’s extensive, methodical and detailed analysis of the claim in relation to the criteria for certification had the potential, if followed by other judges, to enliven the level of scrutiny applied to future certification motions involving medical products. The message sent to litigants by the ruling was clear: certification should not be assumed, even in product liability cases, and even where the claims and proposed common issues closely mirror those put forward in previous certified cases. Whether a given case is certified by the court should not turn on whether claims involving similar issues have previously been certified by the court. Rather, certification should depend on the evidence filed by the parties and the adequacy of the presentation of the plaintiffs’ case as it relates to the five criteria for certification set out above.
Subsequent to this case, on 30 January 2014 in Wakelam v Wyeth Consumer Health Care (2014 BCCA 36), the British Columbia Court of Appeal reversed a certification order in a proposed class action relating to children’s cough and cold medicine. In 2008, Health Canada had reversed a long-standing policy that had permitted the sale of certain non-prescription cough and cold medicines for use by children. Manufacturers of such medicines had already voluntarily withdrawn such products from the market for use in children under the age of two, but Health Canada now required them to relabel the medicines to instruct consumers that they should not be used for children under six. The plaintiff alleged that she had purchased this type of medicine over a number of years, but she did not allege that she had given the medicine to her child or that her child had suffered any injury.
The crux of the plaintiff’s claim was that in marketing the medicines for use in children under six, the manufacturers had engaged in “deceptive acts and/or practices” contrary to provisions of the Business Practices and Consumer Protection Act (BPA) and the federal Competition Act. As damages, the plaintiffs sought reimbursement for their “waste of money” on the medications, as well as disgorgement of any profits received by the defendants as a result of their alleged contraventions of those statutory duties. The motion judge granted certification, but this order was reversed by the British Columbia Court of Appeal (BCCA). In brief, the BCCA determined that the primary causes of action pleaded, and the relief claimed, by the plaintiffs did not disclose a cause of action. The court held that the BPA is an exhaustive code regulating consumer transactions, and that restitutionary remedies are not available under the BPA. Similarly, the Competition Act did not contemplate the restitutionary remedies sought by the plaintiffs in this action.
The BCCA regarded the only causes of action left in the pleading as effectively de minimis, and exercised its discretion to set aside the certification order, without prejudice to the plaintiff to reapply to seek re-certification of what remained of the action.
The BCCA issued another favourable ruling for drug and device manufacturers in January 2015. The BCCA’s decision in Charlton v Abbott Laboratories Ltd (2015 BCCA 26) reversed a motion court’s decision which had granted certification of a class action relating to sibutramine, the active ingredient in the weight-loss drug Meridia.
The plaintiff had brought a proposed class action on behalf of all persons who had used or purchased sibutramine. The plaintiffs alleged that the ingestion of sibutramine caused or contributed to an increased risk of adverse cardiovascular events, such as heart attacks and strokes, increased blood pressure and heart rate, and irregular heartbeat. The plaintiffs relied upon a study showing that sibutramine should not be used in patients with existing cardiovascular disease. However, the available statistical evidence suggested that there was no measurable increase in the risk of cardiovascular events for patients without a diagnosed history of cardiovascular disease, and the plaintiffs’ expert acknowledged that he was unable to quantify that risk.
Further, the contraindications in the product monograph for Meridia made it clear to doctors that Meridia should not be prescribed to patients who had a history of cardiovascular disease. Consistent with that, the defendant adduced expert evidence from cardiologists who testified that they, and other cardiologists, well understood that Meridia ought not to be prescribed to patients with a history of coronary artery disease, and the product monograph specifically referred doctors to that fact and recommended regular monitoring of blood pressure and heart rate because cardiovascular disease is a known complication of elevated blood pressure.
Despite that evidence, the motion judge had found that the plaintiffs had adduced sufficient evidence – ie, they had met the minimal and undefined “some basis in fact” evidentiary threshold – to establish a common issue of whether sibutramine causes or contributes to heart attacks, strokes and arrhythmia.
On appeal, the defendant argued that the certification judge had erred by failing to consider whether there was any evidence of an available methodology to effectively determine that common question that was certified. The BCCA adopted the principle from some recent antitrust class action decisions that the plaintiff has an onus to provide some evidence of a workable methodology that will enable the plaintiff to prove causation on a class-wide basis, and thereby advance the litigation for the entire class. Given that the hallmark of a common issue is that success for one member of the class must mean success for all, the representative plaintiff must adduce some evidence to show how the answer to a question raised by a common issue for one plaintiff will be capable of extrapolation to each member of the class.
In essence, each common issue certified required a finding that sibutramine increases the risk of cardiovascular events for all class members – ie, including those who had no prior cardiovascular condition. The BCCA looked at the evidence adduced by the plaintiff on the certification motion, and observed that there was no evidence of a methodology for establishing that the class as a whole, as opposed to those members who had a pre-existing history of cardiovascular condition, were affected or put at risk by using of sibutramine. The expert witnesses were uncertain about whether there even was a risk to the class members who did not have a pre-existing cardiovascular condition, and they could not describe a methodology for addressing that question. In other words, the plaintiff had failed to present any evidence to establish that this question of general causation for the class as a whole was capable of resolution. Accordingly, the BCCA reversed the order of the motion judge and denied certification of the class.
One month later, in February 2015, the Court of Queen’s Bench of Alberta dismissed a motion for certification of a proposed class action for damages allegedly resulting from a hip resurfacing system in Warner v Smith & Nephew Inc (2015 ABQB 139). In Warner, the court found that the plaintiff:
failed to establish that a class action was the preferable procedure for resolving the claims because of the absence of a common issue that could materially advance the class-wide claim; and
failed to establish the existence of an identifiable class of two or more persons.
The plaintiff had been implanted with the Birmingham metal-on-metal hip resurfacing system in 2005. About three months after the procedure, the plaintiff noticed the system beginning to “pop and click”. A blood test showed elevated metal ion levels, and her doctor then began to monitor her blood. By 2012, the plaintiff was told by her doctor that the levels of cobalt ions in her system were considered toxic. In April, 2012, the plaintiff had a revision surgery to remove the Birmingham hip system and replace it with a total hip implant system.
The plaintiff commenced a class action on behalf all persons in Canada who had been implanted with the defendant’s Birmingham metal-on-metal hip resurfacing system. The second element of the certification test in Canada is that the plaintiff must show that there is an identifiable class of two or more persons. Specifically, there must be evidence of at least one person other than the representative plaintiff who shares the same complaint and wants that complaint prosecuted in a class proceeding. In her supporting affidavit for certification, the plaintiff testified that she had been informed by someone in the law firm that was representing her that the law firm had been retained by nine other class members, and that these class members had experienced problems with the Birmingham hip system similar to those which the plaintiff had experienced. However, the plaintiff had not communicated with any of those individuals, and she did not know the identities of those individuals, the nature of their complaints or why they had retained counsel.
The motion judge held that the undetailed hearsay information provided to the plaintiff by her law firm was insufficient and inadmissible evidence to meet this element of the certification test. There was no identification of any other individual member of the class nor any evidence that such person wished to pursue his or her claim in a class action. Similarly, the description that the other nine individuals had “similar problems” was considered so brief and generic as to defeat any attempt to evaluate whether there was some basis in fact of this allegation. In essence, it was double hearsay – information being passed from the anonymous nine people, through an unnamed person at the law firm, to the plaintiff. For this reason, the judge found that the plaintiff had failed to establish that there was an identifiable class of two or more persons.
Beyond that, the judge held that the plaintiff also failed to demonstrate that a class action was the preferable procedure for this litigation, because the plaintiff did not establish that the root common issue could materially advance the litigation on a class-wide basis. At the hearing, the plaintiff proposed that the common issue at the root of all claims was whether the Birmingham resurfacing system should have been made available for implantation in patients at all. Yet the plaintiff failed to adduce even minimal evidence to establish that this common issue had any realistic prospect of succeeding. Among other things, the Birmingham system is federally licensed and regulated, and had never been the subject of a recall in Canada. The judge noted that while not definitive, that fact makes it less likely that the system is so deficient that it should not be used for any application a surgeon might recommend. Further, the Birmingham resurfacing system had been the subject of an extensive, ongoing study by the Alberta Hip Improvement Project. The preliminary report of that project indicated that current information supports the Birmingham system remaining available as recommended by surgeons, within existing practice guidelines. Moreover, evidence showed that many qualified surgeons continued to recommend and use the Birmingham system for patients considered appropriate; the revision rate experienced by patients with the Birmingham system was comparable with, and in some cases better than, that experienced by patients with total hip replacements; the metal-on-metal aspect of the Birmingham system was an improvement on the materials that were used for other resurfacing systems; and not even the plaintiff’s expert suggested that the Birmingham system was so deficient that it should be removed from the list of available hip replacement products. Consequently, the court denied certification.
Before the first quarter of 2015 had ended, a judge of the Ontario Superior Court of Justice added a third medical products defence certification victory to the 2015 scoresheet in O’Brien v Bard Canada Inc, 2015 ONSC 2470. There, the plaintiff had moved for certification of a class action on behalf of women who were implanted with 19 different pelvic mesh medical devices manufactured by the defendant, 10 of which were used to treat pelvic organ prolapse (POP) and the other nine of which were used for treating stress urinary incontinence (SUI).
POP and SUI are medically distinct phenomena, and both have a broad range of severity. The defendant began making and selling products to treat POP and SUI in 2002, and produced a variety of pelvic mesh products with different designs and intended surgical uses. The products are designed to provide different approaches to the treatment of POP and SUI respectively. While all the products used surgical mesh, there are distinct design features which differ among the 19 products.
The plaintiff filed a report of a medical doctor as an expert witness. The plaintiff’s expert opined that there are very serious issues regarding the safety of pelvic mesh products, and that the warnings and information of the dangers of using the products that were provided to surgeons, patients and the general public were inadequate. The expert acknowledged that there are differences among the defendant’s 19 pelvic mesh products, but in his view the similarities made the pelvic mesh products collectively unfit for their intended purpose. In this regard, the plaintiff’s expert did not go so far as to say that surgical mesh products should never be used. Rather, he acknowledged that there is a role for device-assisted pelvic repair surgeries, but the plaintiff’s expert did not have the requisite expertise to opine on whether design changes could be made to pelvic mesh products to make them safer.
The judge found that the plaintiff failed to establish the common issue element of the certification test. In brief, the judge found that the plaintiff failed to adduce evidence to show a singular or common design defect in the defendant’s 19 pelvic mesh products. The judge observed that the commonality of surgical mesh in the 19 products conveyed only a false impression of commonality among these 19 products because:
the evidence showed that the 19 medical devices for the treatment of two very different medical ailments were different in materials, shape, size, weight, density, weave, porosity, flexibility, configuration, fixation methodology, design purposes and product warnings;
patients diagnosed with these conditions will have idiosyncratic symptoms, side effects and responses to treatment; and
the choice of the type of surgical intervention will be guided by patient-specific factors; surgeon-specific factors; and factors related to the mechanical and functional design and intended use of the various products available for surgical intervention.
Further, the judge found that the plaintiff had failed to show that any significant element of her claim is capable of proof on a class-wide basis. Citing the above-noted BCCA decision of Charlton v Abbott Laboratories Ltd, the judge noted that where a plaintiff seeks to address questions of general causation as the foundation of a class action, there must be some evidence of a methodology that will enable the plaintiff to prove causation on a class-wide basis. Where no such methodology is put forward by the plaintiff, there is not sufficient evidence before the court to show that the resolution of the proposed general causation common issue will efficiently advance the claim of the class as a whole. Here, the judge found that the plaintiff had failed to propose any methodology to show that a finding of causation of the multitude of injuries, conditions and complications from the use of the 19 different mesh products for POP and SUI surgery can be extrapolated across the class. Consequently, the judge held that the plaintiff had failed to meet the certification test, and he dismissed her motion.
Given the low evidentiary threshold established by Canadian law for certification motions, coupled with the financial incentive that class actions present for entrepreneurial plaintiffs’ counsel, the number of product liability class action cases launched in Canada has proliferated, particularly against manufacturers of medical products that are ingested into the body. However, the spate of medical product liability cases in which certification has been denied over the past-three years demonstrates that, contrary to popular belief among many plaintiffs’ counsel, certification of a medical product class action is no longer a foregone conclusion. It appears that, at long last, some Canadian courts are now showing an inclination to take a more disciplined approach toward the evidentiary thresholds of the certification test, and to conduct a more analytical assessment of the evidentiary record on these motions. If the judicial approach reflected by these recent decisions is followed by subsequent courts, the legal fortunes between plaintiffs and defendants in certification motions will shift toward a more reasonable and balanced middle ground.