Oft-referred to as "quintessential class actions", the majority of product liability actions that have sought certification as class proceedings in the last 10 years have been granted certification. However, two recent decisions denying certification may signal a shift towards greater scrutiny of proposed product liability class actions. Judges have become more willing to refuse certification where it is not clear there is sufficient commonality amongst the class members or that a class proceeding will be the preferable procedure.

In Warner v Smith & Nephew Inc., a recent Alberta certification decision, the product at issue was Smith & Nephew's hip resurfacing system, known as "the Birmingham system". After being implanted with the Birmingham system, the representative plaintiff was found to have toxic levels of metal ions in her blood (the Birmingham system is a metal product) and had to have the device removed. Certification failed on the issues of identifiable class and preferable procedure.

Regarding identifiable class, the judge found that a class of all people who had the Birmingham system implanted and their dependants would be appropriate since compensable harm could arise as soon as a defective device is implanted. However, Justice Poelman held that the plaintiff had not established some basis in fact for the existence of at least two class members who had experienced complaints similar to those of the representative plaintiff. The evidence of the representative plaintiff on this issue was rejected as double hearsay, "being passed from the anonymous nine people through an unnamed person at the law firm to the plaintiff."

Certification was also denied for failure to meet the preferable procedure requirement. In finding that a class action was not the preferable procedure, Justice Poelman pointed out one of the significant problems in many medical device class actions: that complaints about the device can be caused by a host of patient-specific and surgeon-specific factors unrelated to the device. Here, those factors included the individual nature of the patient's hip problem, the patient's relevant personal characteristics and medical background, the recommendation of the surgeon on the appropriate device, whether the patient accepted that recommendation, the difficult surgical technique involved with properly implanting the device, and subsequent monitoring and general health. Since all of these issues require individual assessment, they will likely not be common across class members and are therefore likely more suitable for individual actions.

The decision in Warner follows another denial of certification in Charlton v Abbott Laboratories, Ltd. by the British Columbia Court of Appeal. The proposed class in Charlton was composed of those who took the drug sibutramine, a prescription weight-loss medication that allegedly increased the risk of cardiovascular events, such as heart attack and stroke. The evidence adduced on the certification motion suggested there might be an important distinction between class members who had pre-existing cardiovascular conditions, and those who did not. The defendant manufacturers had warned against prescribing the drug to patients with pre-existing conditions. The B.C. Supreme Court originally granted certification, but the Court of Appeal reversed that decision because the plaintiffs had failed to establish evidence of a "methodology for establishing that the class as a whole, as opposed to those who were wrongly prescribed sibutramine despite a history of disease, was affected or put at risk by its use of sibutramine".

The requirement of a "methodology" for establishing harm on a class-wide basis is well-established in indirect purchaser class actions where price-fixing is alleged, but new to product liability class actions. Interestingly, the plaintiffs' expert, a cardiologist, had opined that all class members (both with and without pre-existing conditions) were put at risk by using sibutramine. However, the Court of Appeal scrutinized the evidentiary foundation of the expert's opinion, and found no support for the proposition that sibutramine users without pre-existing conditions were at increased risk. There was no prospect of such evidence being obtained, since the drug had been pulled off the market in 2010 and was no longer being studied.