Bayer Inc., Bayer A.G., Bayer Corporation, v. Bayer Healthcare LLC, 2019 SKCA 64

Court of Appeal for Saskatchewan Docket: CACV3327 Citation: Bayer Inc. v Tluchak Estate, 2019 SKCA 64Date: 2019-07-25Between:Bayer Inc., Bayer A.G., Bayer Corporation, and Bayer Healthcare LLCApplicants/Prospective Appellants (Defendants)AndThe Estate of Tluchak and Verna TluchakRespondent/Prospective Respondent (Plaintiffs)Before: Whitmore J.A. (in Chambers)Disposition: Leave to appeal deniedOn Application From: 2018 SKQB 311, ReginaApplication Heard: June 12, 2019Counsel: Frank McLaughlin and Jason Mohrbutter for the Appellants E.F.

Anthony Merchant, Q.C., Anthony Tibbs, and Linh Pham for the Respondent Whitmore J.A. I. Introduction [1] This is an application by Bayer Inc., Bayer A.G., Bayer Corporation, and Bayer Healthcare LLC [Bayer] for leave to appealthe certification order made by the Chambers judge pursuant to s. 39(3)(

a) of The Class Actions Act, SS 2001, c C-12.01 [Act], and s. 8 ofThe Court of Appeal Act, 2000, SS 2000, c C-42.1: Tluchak Estate v Bayer Inc., 2018 SKQB 311 [Chambers Decision]. [2] The Chambers judge found that there was a basis to conclude there is a common issue which he characterized as follows: “DidBayer breach a duty of care in marketing Xarelto in Canada or in failing to provide a reasonable warning that Xarelto could cause seriousand irreversible bleeding?” (at para 110). [3] This alleged breach relates to the product monograph, which is essentially a description of the medication that providesinformation as to the medication’s indications, contraindications, adverse reactions, etc.

The monograph is required to be made availableto all consumers and doctors. [4] Bayer seeks leave to appeal on the following grounds: (

a) the learned judge erred in concluding that the proposed failure to warn issue was appropriately certifiable as a common issueunder s. 6(1)(

c) of the Act; (

b) the learned judge erred by finding that a class action was the preferable procedure in this case within the meaning of s. 6(1)(d)of the Act; and (

c) the learned judge erred in certifying a common issue on punitive damages when liability will not be determined at thecommon issues trial. [5] If leave is granted, Bayer’s proposed notice of appeal requests that this Court set aside the decision of the Chambers judge anddismiss the application to certify this action as a class action. II. Analysis [6] The test to be applied to determine whether leave should be granted is set out in Rothmans, Benson & Hedges Inc. vSaskatchewan, 2002 SKCA 119, 227 Sask R 121 [Rothmans].

It requires that the proposed appellant—in this case, Bayer—show that theproposed appeal is one of sufficient merit and of sufficient importance to weigh decisively in favour of granting leave (Rothmans): [6] The power to grant leave has been taken to be a discretionary power exercisable upon a set of criteria which, on balance, must beshown by the applicant to weigh decisively in favour of leave being granted: Steier v. University Hospital, (SK CA),[1988] 4 W.W.R. 303 (Sask. C.A., per Tallis J.A. in chambers).

The governing criteria may be reduced to two – each of which features asubset of considerations – provided it be understood that they constitute conventional considerations rather than fixed rules, that they arecase sensitive, and that their point by point reduction is not exhaustive.

Generally, leave is granted or withheld on considerations of meritand importance, as follows: First: Is the proposed appeal of sufficient merit to warrant the attention of the Court of Appeal? • Is it prima facie frivolous or vexatious? • Is it prima facie destined to fail in any event, having regard to the nature of the issue and the scope of the right of appeal, for

instance, or the nature of the adjudicative framework, such as that pertaining to the exercise of discretionary power? • Is it apt to unduly delay the proceedings or be overcome by them and rendered moot? • Is it apt to add unduly or disproportionately to the cost of the proceedings?

Second : Is the proposed appeal of sufficient importance to the proceedings before the court, or to the field of practice or the state of the law, or to the administration of justice generally, to warrant determination by the Court of Appeal? • does the decision bear heavily and potentially prejudicially upon the course or outcome of the particular proceedings? • does it raise a new or controversial or unusual issue of practice? • does it raise a new or uncertain or unsettled point of law? • does it transcend the particular in its implications? (All emphasis in original) [ 7 ] Bayer submits that the proposed appeal meets the Rothmans test in both sufficient merit and importance. [ 8 ] The purpose of the certification stage in a class action is to determine whether it is appropriate for the claim to be prosecuted as a class action.

It is not meant to be a test of the merits of the action. The Rothmans test, therefore, is to be considered in the context of that purpose: see Hollick v Toronto (City) , 2001 SCC 68 at para 16 , [2001] 3 SCR 158 [ Hollick ]; and Pro-Sys Consultants Ltd. v Microsoft Corporation , 2013 SCC 57 at para 102 , [2013] 3 SCR 477 [ Pro-Sys ]. In Pro-Sys , Rothstein J. noted the following: [105] Finally, I would note that Canadian courts have resisted the U.S. approach of engaging in a robust analysis of the merits at the certification stage.

Consequently, the outcome of a certification application will not be predictive of the success of the action at the trial of the common issues. I think it important to emphasize that the Canadian approach at the certification stage does not allow for an extensive assessment of the complexities and challenges that a plaintiff may face in establishing its case at trial. After an action has been certified, additional information may come to light calling into question whether the requirements of s. 4(1) continue to be met.

It is for this reason that enshrined in the CPA is the power of the court to decertify the action if at any time it is found that the conditions for certification are no longer met (s. 10(1)). [ 9 ] Accordingly, the applicant for a class action must show some basis in fact for each of the certification requirements but is not required to show whether the claim itself is likely to succeed ( Hollick at paras 16 and 25 ; and Pro-Sys at paras 99–105 ). [ 10 ] I will consider Bayer’s proposed grounds of appeal in the application of the Rothmans test. III. Merit of the Proposed Appeal A.

Is the proposed appeal frivolous or vexatious or is it predestined to fail? 1. Did the learned Chambers judge err in concluding that the proposed failure to warn issue was appropriately certifiable as a common issue under s. 6(1)(

c) of the Act ? [ 11 ] The Chambers judge certified a failure to warn common issue but did not certify a general causation common issue. Bayer submits that the certification of a common issue of general causation is a necessary precondition to the certification of the failure to warn question, and the Chambers judge erred in failing to certify a general causation common issue.

Bayer notes that while the Chambers judge did find that there is some basis in fact to conclude that Xarelto has a propensity to injure in its ordinary use, the failure to warn issue is dependent on whether the general causation question is a common one. [ 12 ] The prospective respondents submit that it is not necessary to certify a common issue for general causation as the Chambers judge had adequate evidence to consider general causation and decide that it was not an issue.

The prospective respondents argue that Bayer’s submissions regarding the necessity of a certification of a common issue for general causation are based on the strength of the evidence surrounding the general causation issue and would require a more robust analysis of the merits of the action than is appropriate at the certification stage. [ 13 ] The analytical framework that typically applies to product liability claims is outlined in Harrington v Dow Corning Corp. , 2000 BCCA 605 at paras 42–46 , 193 DLR (4th) 67 [ Harrington ].

In Batten v Boehringer Ingelheim (Canada) Ltd. , 2017 ONSC 53 at para 38 [ Batten ] (aff’d 2017 ONSC 6098 (Div Ct), 20 CPC (8th) 414), Perell J. summarized the framework as the following: [38] … The first step, known as the general causation step, determines whether the product is capable of causing harm. The second step is part of determining whether the manufacturer had a duty of care not to sell the product or to sell it only with an appropriate warning. The third step focuses on the adequacy of the warning.

The fourth step will determine individual causation and the quantification of the compensation for the consequent harm. [ 14 ] The first step was described by Huddart J.A. in Harrington :

[42] At the risk of oversimplifying a complex decision-path, I venture to suggest the first step in every products liability case allegingnegligent design, manufacture, or marketing is the determination of whether the product is defective under ordinary use or, although non-defective, has a propensity to injure.

Some American authorities refer to this step as “general causation”, whether a product is capable ofcausing the harm alleged in its ordinary use. [15] Pharmaceutical class actions often include common issues relating to general causation where there is a dispute as to whetherthe product is capable of causing harm. The reason the question of common causation may be required is that it informs the need for, andcontent of, the duty to warn.

However, the necessity of including a general causation common issue is on a case specific basis. [16] Here, the Chambers judge explained why he did not find it necessary at this stage to certify a common causation issue. Theissue before him was whether Xarelto had a propensity to injure. As the Chambers judge noted, Bayer’s own expert, Dr. Crowther,pointed out that Xarelto, like all anticoagulants, causes blood thinning, which increases the risk of bleeding.

Thus, the first Harringtonquestion had been answered. [17] In Peter v Medtronic, Inc. (2007), 50 CPC (6th) 133 (WL) (Ont Sup Ct) [Peter], the plaintiffs alleged that the defendantsbreached their duty to warn in relation to the defects in the batteries that operate implantable defibrillators. The certification judge did notconsider Harrington, but did certify the following two questions relating to the duty to warn: [87] The plaintiffs propose the following nine common issues, which reflect various refinements made during the course of the hearing: …

(3) Did the Defendants, or either of them, owe a duty to the Class to warn of the potential battery shorting defect associated with theDefibrillators, and if so, when did such duty arise?

(4) If so, did the Defendants, or either of them, fail to warn the Class of the existence of the potential battery shorting defect associatedwith the Defibrillators? [18] The defendant did not object to these common issues and the certification judge certified the common issues without engagingin a discussion over the elements of each issue (at para 89).

The seven other proposed common issues did not include a general causationcommon issue. [19] Similar common issues were raised in Robinson v Medtronic, Inc. (2009), (ON SC), 80 CPC (6th) 87 (WL)(Ont Sup Ct) [Robinson], (aff’d 2010 ONSC 3777 (Div Ct), 267 OAC 126): [153] After several adjustments made during the course of the argument, the Plaintiffs propose the following common issues: … (iii) Did the Defendants, or either of them, owe a duty to the Class to warn of the risk of fracture associated with the Leads, and if so,when did such duty arise? (iv) If so, did the Defendants, or either of them, fail to warn the Class of the existence of the risk of fracture associated with the Leads? [20] The certification judge in Robinson certified these common issues without an in-depth discussion on the basis of the commonissues and no concern was raised about general causation (at paras 155–156). [21] In Price v H.

Lundbeck A/S, 2018 ONSC 4333 (Ont Sup Ct) [Price], the plaintiff, similar to the proposed respondents in thematter before me, withdrew their proposed general causation common issue shortly before the certification trial (at paras 124–125). JusticePerell, the certification judge, found that the proposed duty to warn existed, but failed to certify it as a common issue because there was alack of commonality in the proposed class (at para 131).

Additionally, Perell J. found that the resolution of the duty to warn common issuewould not form a substantial part of each class members claims: [134] … Put bluntly, the duty to warn issue does not connect the dots for a common issues trial that has any utility for a class proceedingthat inevitably end with individual issues trials with very significant causation issues associated with the breach of the duty to warn. [135] These points can be demonstrated by using the examples of the causes of action of Mesdames Bews, Gardner, Lovett, Price, andStymiest.

Each of their cases is built on the issue that Lundbeck breached a duty to warn that citalopram may cause congenitalmalformations.

Notwithstanding Lundbeck’s arguments to the contrary, the issue of whether Lundbeck breached a duty to warn exists andhas an air of reality to it, but the issue is not a common issue because the breach of the duty would be inconsequential across the Class andeach individual Class Member would be unharmed by the breach of the duty to warn unless they were actually harmed as a consequenceof their ingestion of citalopram. … [140] But as I have already said, all of the parties’ arguments about whether there is some basis in fact for a common issue based ongeneral causation is of no moment because there is no such common issue being proposed.

And while there is some basis in fact for anissue about a duty to warn, it is not a common issue for the reasons set out above. [22] Justice Perell did not use the Harrington analysis for product liability but rather focused on whether the proposed duty to warnissue was a common issue that could form a substantial part of each class member’s case. While the decision to remove the general

causation issue was not viewed favourably by Perell J., it ultimately was not the reason the duty to warn common issue was not certified. [23] On the face of it, I see no error on the part of the Chambers judge in certifying a duty to warn common issue and in notcertifying a separate general causation common issue.

While many pharmaceutical class actions will certify a general causation commonissue separate from the failure to warn common issue, the certification of a general causation common issue does not seem to be necessaryin every duty to warn analysis, especially when the general propensity of the product to injure is acknowledged by the manufacturer and itis only the specific details of this which are in dispute. [24] This case is unlike Harrington, where the issue was whether the breast implants were fit for their intended use.

In Harrington,the applicant for the class action submitted that silicone gel breast implants had a high rate of failure, leading to ruptures, and associatedconnective tissue disease and localized complications (at paras 1–2).

This case is also distinguishable from Miller v Merck Frosst CanadaLtd., 2015 BCCA 353, 81 BCLR (5th) 33 (leave to appeal to the SCC dismissed, ), where the issue was whetheringesting finasteride, a male pattern baldness medication, could cause persistent sexual dysfunction in some users even afterdiscontinuation of treatment. [25] The matter before me is more in line with Batten, which was a class action certification appeal that involved a failure to warncommon issue regarding the anticoagulant Pradaxa.

The court in Batten disallowed certification for the general causation common issuebecause there was no common class wide potential injury or risk of injury, and there was no methodology to prove general causation.However, the court analyzed the certification of the general causation common issue after deciding there was no basis in fact for finding afailure to warn common issue.

In determining that there was no failure to warn common issue, the court did not consider whether it wasnecessary to certify a general causation issue, but rather looked at whether there was any basis in fact to conclude that the failure to warncreated an undisclosed hazard to class members. [26] In my view, whether the certification of a general causation common issue is a necessary precondition to the failure to warncommon issue depends on whether the Chambers judge encompassed the general causation step in his certification of the failure to warncommon issue, or whether he did not and thus precluded the common issues trial from determining whether Xarelto has the propensity toinjure.

It appears to me that the Chambers judge operated under the presumption he was certifying a common issue that included theformer; that is, at the common issues trial the details of the general causation question would have to be proven in order to determinewhether there was a failure to warn. [27] As stated above, the purpose of the certification stage is not to decide the merits of the action but rather to examine whetherthere is any basis in fact for each requirement for certification.

The Chambers judge found that there was “some basis in fact to concludethat Xarelto has a propensity to injure in ordinary use” (Chambers Decision at para 125), satisfying the first step of the Harringtonanalysis. Later, the Chambers judge said: “resolving this general causation issue is important and that its resolution will materiallyadvance the litigation” (at para 171).

Based on this, I view the Chambers judge as embedding any question of common causation into themore general question of breach of duty to warn. [28] Bayer also submits that the Chambers judge erred by certifying the failure to warn common issue without requiring a classwide methodology for determining general causation, and that the Chambers judge further erred in finding that variations betweenbleeding types, indications, and comparative anticoagulants are irrelevant at the certification stage. [29] Bayer’s submissions in this regard are dependent on its allegation that it was an error to not certify a general causation commonissue.

In my opinion, these submissions are largely irrelevant at this stage because the Chambers judge did not certify a separate generalcausation common issue, he only certified a failure to warn common issue for which he found there was evidence of a workablemethodology (Chambers Decision at para 169). There is no need to have a workable methodology for a common issue that is not certified.As I have found it is not necessary in this case to certify a failure to warn common issue without a separate general causation commonissue, the class wide methodology for determining general causation is not needed.

This proposed ground of appeal is not likely tosucceed. 2. Did the learned Chambers judge err by finding that a class action was the preferable procedurein this case within the meaning of s. 6(1)(

d) of the Act? [30] Bayer submits that because general causation would not be addressed at the common issues trial and is there a lack of aworkable methodology for resolving it on a class wide basis, the Chambers judge erred in finding a class action was the preferableprocedure. [31] In Price, Perell J. stated that the preferable procedure criterion was not satisfied because of the lack of a general causationcommon issue: [154] I, however, would have come to the same conclusion even if Ms. Price had established a common issue about the breach of the dutyto warn.

In considering the preferable procedure criterion, the court should consider the nature of the proposed common issue and itsimportance in relation to the claim as a whole and it should consider the individual issues which would remain after determination of thecommon issue.

In the immediate case, the deferral of the general causation issue to individual issues trials meant that there was little utilityto a common issues trial in making progress to access to justice, and thus the preferable procedure criterion is not satisfied. [32] Price can be distinguished from the case before me because, in this case, the Chambers judge did find that the resolution of theduty to warn was a common issue that could advance the claim of every class member (Chambers Decision at paras 137–138).

I interpretthe Chambers judge as intending to include the general causation issue within the duty to warn issue to be decided at the common issuestrial. [33] There would be no reason to certify a duty to warn common issue that did not include the opportunity for both sides to bringevidence on the level of danger entailed by the ordinary use of Xarelto. The certification stage is not meant to be dispositive of the meritsof the common issues and all that the Chambers judge found was that “There is clearly some basis in fact to conclude that Xarelto has apropensity to injure in the ordinary use.” (at para 125).

I interpret the Chambers judge’s decision as including the general causation

common issue in certifying the duty to warn common issue. [ 34 ] Bayer also submits that the failure to warn common issue should not have been certified due to its vague and abstract phraseology, focusing on the words “serious and irreversible bleeding”.

Bayer acknowledges that the Chambers judge does state in regard to the duty to warn: [138] I note as well the alleged breach includes the failure to provide adequate instructions as to how to assess individual risk both at the outset and in the course of treatment, how to reduce that risk in the course of treatment, and how to respond to and manage bleeding in light of the risk. Those allegations respond to the complexity identified by the defendants as part of the common issue.

They enhance the potential value of an answer to the common question to class members. [ 35 ] Bayer contends that it is unclear whether these questions would be part of the failure to warn common issue at trial. [ 36 ] The Chambers judge did reject certification for the marketing common issue on the grounds that it was too broad (at paras 151– 153).

The failure to warn common issue, with the inclusion of “reasonable warning that Xarelto could cause serious and irreversible bleeding” (at para 110) provides more guidance. [ 37 ] In Batten , the proposed failure to warn common issue was dismissed because the motions judge did not find any basis in fact to conclude that a common issue about a duty to warn existed. In dismissing the certification application, the court did not mention whether the proposed certification question was too broadly worded.

The proposed wording in Batten was similar to the Chambers Decision question: “Did any of the Defendants have a duty to warn the Class Members of the risks of harm, namely from the lack of an antidote for Pradaxa?” ( Batten at para 182 (2)) versus “Did Bayer breach a duty of care in failing to provide a reasonable warning that Xarelto could cause serious and irreversible bleeding?” ( Chambers Decision at para 110), wherein the Chambers judge substituted irreversible for lack of an antidote . [ 38 ] The following are examples of where a common issue has been dismissed or amended for being too broadly worded: (

a) R v Brooks , 2009 SKQB 509 , [2010] 6 WWR 81 (leave to appeal to Sask CA refused, 2010 SKCA 55 ): “‘Adverse health effects’ is vague and infinite and not proscribed in any way.” (at para 163) (dismissed); (

b) Martin v Astrazeneca Pharmaceuticals Plc , 2012 ONSC 2744 , 27 CPC (7th) 32 (aff’d 2013 ONSC 1169 (Div Ct)): “related metabolic disturbances as well as secondary injuries flowing therefrom.” (at para 221) (dismissed); (

c) Schroeder v DJO Canada Inc. , 2010 SKQB 125 , [2010] 10 WWR 324 (aff’d 2011 SKCA 106 ): “serious adverse effects” (at para 93) amended to whether the device in question caused “chrondrolysis when placed in the synovial cavity of a knee or shoulder following surgery” (at para 94); (

d) Sweetland v GlaxoSmithKline Inc. , 2016 NSSC 18 , 369 NSR (2d) 229: “adverse cardiovascular events” (at para 60) was amended to limit the common issue to “heart failure, heart attack and stroke” (at para 62). [ 39 ] The proposed respondents submit that the terminology used by the Chambers judge is appropriate because Bayer has used the terms “serious” and “life-threatening” in relation to bleeding to warn patients about the potential side effects of Xarelto.

Additionally, the proposed respondents submit that “irreversible” does not lack precision because, given the context of the Chambers Decision , it clearly refers to the lack of a pharmaceutical antidote. Bayer concedes that the proposed respondents’ position as to the Chambers judge’s use of the term “irreversible” is likely accurate. [ 40 ] Given the context of the Chambers Decision , I am of the view that the Chambers judge was referring to a lack of an antidote when referring to “irreversible” and that part of the common issue is not too vaguely worded.

If “serious” is too vague, this can be clarified in the common issues trial. Bayer makes the same submissions before me as it did before the Chambers judge but has not pointed to an error in principle. [ 41 ] Further, as stated above, the certification stage is not the place to perform an in-depth analysis of the merits of the plaintiffs’ case. The Chambers judge’s decision on certification is to be awarded deference by a reviewing court.

This is particularly so with respect to the Chamber judge’s decision with respect to the preferability procedure: see AIC Limited v Fischer , 2013 SCC 69 , [2013] 3 SCR 949. [ 42 ] I do not view this proposed ground of appeal as having sufficient merit to allow it to proceed. 3.

Did the learned Chambers judge err in certifying a common issue on punitive damages when liability will not be determined at the common issues trial? [ 43 ] The second common issue that was certified by the Chambers judge was as follows: “Does the manner in which Bayer obtained market authorization for Xarelto or the manner in which it marketed Xarelto justify an award of punitive damages?” ( Chambers Decision at para 110(b)). [ 44 ] The Chambers judge found that this was appropriate for class wide determination because it was a fact specific issue that was based on specific allegations in the plaintiffs’ statement of claim (at para 163). [ 45 ] Bayer did not submit any argument to support this ground of appeal.

[ 46 ] In

summary, Bayer submits that the proposed appeal is not prima facie destined to fail. While I cannot say that it is destined to fail, in my view, for the reasons stated, the appeal of the certification order would likely not succeed. Bayer’s submissions, while relevant at the common issues trial, are not relevant at the certification appeal stage. B. Delay and Costs [ 47 ] It has been 2.5 years since the certification application was first filed.

The respondents submit that there would be undue delay in the proceedings if this appeal were to proceed but make no submissions other than if the common issues trial does not proceed then the wrongful conduct of Bayer will continue. Bayer submits that the delay would not be inordinate and the costs of the proposed appeal would not be disproportionate compared to the large amount of damages at stake. [ 48 ] If this appeal were to proceed, I do not view the delay as being inordinate or that any additional costs to the proceedings would be disproportionate.

The appeal would not likely involve the introduction of fresh evidence as the parties would in all likelihood be limited to questions of law surrounding the requirement of a certified general causation question. The consideration of these issues does not weigh either in favour or against allowing the appeal to proceed. IV. Importance [ 49 ] In my view, this appeal of the certification order is not of overarching importance to these proceedings as it is not a determination of the issues on their merits.

Bayer has submitted arguments that might well be successfully advanced at the common issues trial to show that there is a lack of general causation, which would likely significantly impact the action.

It would no doubt be advantageous for Bayer to receive more information about the methodology and for the wording to be more precise in the failure to warn common issue, but Bayer has shown that it has a good understanding of what the terms “serious” and “irreversible” are intended to mean. [ 50 ] Bayer’s submission—that if, on appeal, it was determined that it was necessary to certify a general causation common issue the proceedings would be at an end—is not necessarily correct. As I stated, the Chambers judge implicitly recognized that there is a general causation common issue that can be determined.

Further, Bayer is able to continue to oppose the merits of the proposed class action at trial, and the appeal process is available to both parties after a trial. [ 51 ] This appeal could help to clarify whether it is necessary, in the certification of a failure to warn common issue, to also certify a general causation common issue. While most pharmaceutical cases do include both, it does not appear to be necessary as long as the failure to warn common issue encompasses a general causation element.

I do not see this proposed appeal as adding anything to class actions practice or to the administration of justice generally. [ 52 ] In my view, Bayer has not shown that the application of the Rothmans test weighs decisively in favour of granting leave to appeal. I come to this conclusion because Bayer’s arguments focus on the certification of a general causation common issue being a necessary precondition to the failure to warn common issue.

Its submissions rest on the argument that a general causation common issue should have been certified and dwell on the merits of the action, which is not appropriate at the certification stage. V. CONCLUSION [ 53 ] In the result, leave to appeal is denied. VI. COSTS [ 54 ] As the respondents have not sought costs, I make no order as to costs. “Whitmore J.A.” Whitmore J.A.