Court File and Parties

COURT FILE NO.: CV-14-511333-00CP DATE: 20190903

SUPERIOR COURT OF JUSTICE - ONTARIO

RE: Anne Levac, Plaintiff

– AND –

Stephen Rose James, Sue-Ellen Solger, Izabella Gerbec, Erin Kostuch, Anita Takyi-Prah, Joana Nunes, Elizabeth Hicken, Marissa Allin, Rachel Schrijver, Annie Michaud, Anna Nudel, Elena Polyakova, Raymund Tanalgo, Jefferd Felix, Jason Foster, Paolo Galvez, Glenn Francesco, Peter Rothbart and Rothbart Centre For Pain Care Ltd., Defendants

BEFORE: Justice E.M. Morgan

COUNSEL: Paul Harte and Maria Damiano, for the Plaintiff Erica Baron and Patrick Gajos, for the Defendant, Stephen Rose James David Lee, for the Defendant, Marissa Allin

HEARD: July 11-12, 2019

CERTIFICATION AMMENDMENT AND DE-CERTIFICATION MOTION

[1] This case tests the extent to which statistical correlation and legal proof overlap.

I. Background

[2] The Defendant, Stephen Rose James (“Dr. James”), a physician who specializes in pain relief by administering epidural steroid injections, was found by Toronto Public Health (“TPH”) to have had his hands colonized by Methicillin-sensitive staphylococcus aureus (“MSSA”), a strain of bacteria that is relatively rare in North America. The investigation by TPH was prompted by an outbreak of infections that occurred during a two-year period allegedly corresponding with this bacterial colonization. From January 2010 to November 2012, Dr. James’ patients experienced a higher than average rate of infection – 7.5:1000 as opposed to the norm of 1:1000. Some, but not all of the 21 infected patients, tested positive for MSSA.

[3] The infections were severe and often debilitating for the effected patients. However, each infection having come and gone, it can be difficult to identify their source. Some of the infections resulted from the same bacteria that was colonized on Dr. James, others appear to have resulted from different strains of bacteria, and still others have been difficult to identify after the fact. Defendants say that negligence and factual causation must be established patient by patient. Under these circumstances, can statistical correlation replace specific proof in a tort claim?

[4] This class action was commenced as a proposed class action on September 2, 2014, and was certified by Order of Perell J. on December 9, 2016: Levac v James, 2016 ONSC 7727. That decision was reversed by the Court of Appeal on November 3, 2017 and sent back to Superior Court for reconsideration: Levac v James, 2017 ONCA 842. It was then certified by me on consent of the parties on February 23, 2018, and was set down for trial scheduled to commence on January 21, 2019. The trial did not proceed as scheduled due to the parties’ need for further discoveries.

[5] Extensive discoveries have now been completed. The Plaintiff seeks to amend her pleading to reflect the reality that she says the discovery evidence has revealed. In conjunction with that, the Plaintiff also seeks to amend the certification Order to revise and expand the common issues in accordance with that new information. Plaintiff’s counsel submits that these amendments will clarify and better focus the common issues trial to advance the case on behalf of all class members.

[6] For his part, Dr. James seeks to de-certify the action. His counsel submit that the discovery evidence has revealed that there is insufficient commonality among the claims of the class members. While he does not object to the amendment of the Plaintiff’s pleading or to the continuance of the action as an individual claim, Dr. James’ view is that each of the 21 known class members should sue as individuals. His counsel propose that this be done in a joint action so that while each individual Plaintiff would have to prove his or her case, there can be efficiencies in the trial process in the sense of having expert witnesses and the Defendants themselves testify only once.

[7] The Plaintiff’s claim arises from infections suffered by a group of some 21 patients who were administered epidural steroid injections by Dr. James. The class members were all infected patients at the Defendant, Rothbart Centre For Pain Care Ltd. (“Rothbart Centre”), where Dr. James practiced.

[8] The Defendants are Dr. James, the Rothbart Centre (which is now insolvent), the Rothbart Centre’s administrator, Dr. Peter Rothbart, and 16 nurses who worked at the Rothbart Centre during the relevant time. The central issues in this motion are whether there are sufficient common issues and whether a class proceeding is the preferable procedure for the purposes of ss. 5(1)(c) and (d) of the Class Proceedings Act, 1992, SO 1992, c 6 (“CPA”).

[9] More specifically, this case raises the question of whether statistical evidence with respect to rates of infection and evidence of Dr. James’ infection prevention and control (“IPAC”) practices can amount to proof of causation and standard of care for the purposes of tort liability, or whether the cause of each claimant’s injury must be determined individually. There is conflicting evidence as to whether or not Dr. James IPAC practices were uniformly implemented with respect to all of his patients. There is also some contentious evidence about likely sources of the infections.

[10] It is these causation and standard of care issues that gives rise to the Plaintiff’s wish to expand the common issues and Dr. James’ wish to decertify the proceeding as lacking commonality at all. It is therefore these questions that lie at the heart of the controversy over whether a class action is the preferable procedure here.

II. The proposed amendments

[11] The action is currently certified as a class action with three common issues relating to the negligence claim and one relating to punitive damages:

a) Whether the Defendants owed a duty of care to the Class to maintain infection prevention and control practices (“IPAC”);

b) Whether the Defendants breached the standard of care required for IPAC;

c) In what ways, if any, did the Defendant(s)’ routine invariable IPAC practice(s) breach the standard of care required for IPAC; and

d) Whether there is conduct sufficient to attract punitive damages, and if so, whether punitive damages should be awarded and in what amount.

[12] The Plaintiff seeks to both reformulate these questions and to expand them to add more subject areas to the common issues. The new questions are organized along lines that match the two substantive causes of action found in the Plaintiff’s Fresh as Amended Statement of Claim: negligence and related causation issues, together with fiduciary duty. The Plaintiff also seeks to add a question with respect to the limitation issues posed by some of the class members’ claims, as well as to restate the question regarding punitive damages.

[13] In addition to the revisions to the common issues, the Plaintiff proposes adding two new subclasses to the definition of the class contained in the February 23, 2018 certification Order. The class is currently defined without any subdivision among the members:

All persons who received an epidural injection administered by Stephen Rose James (“Dr. James”) at the Rothbart Centre For Pain Care Ltd. (the “Clinic”) between January 1, 2010 through November 30, 2012, and who subsequent to their treatment developed signs or symptoms clinically compatible with bacterial meningitis, epidural abscess or cellulitis, of a bacterial origin and/or bacteremia (“Clinical Infection”), or where such a person is deceased, the personal representative of the estate of the deceased person who developed a Clinical Infection; (the “Infected Patients”), and

All living parents, grandparents, children, grandchildren, siblings, spouses and same sex partners (within the meaning of s. 61 of the Family Law Act) of Infected Patients, or where such a family member died after his or her related Infected Patient developed signs or symptoms of Clinical Infection, the personal representative of the deceased family member (“Family Law Claimants”).

[14] The Plaintiff seeks to add a subclass of what Plaintiff’s counsel has termed Genetically Linked Patients. This subclass would consist of all class members whose infection can be genetically matched to MSSA, the strain of bacteria colonized on Dr. James. The evidence is that this is a small group of the overall class; but despite its size, it is significant in that the source of this group’s infection is far less controversial than for other class members.

[15] The Plaintiff also proposes a second subclass which Plaintiff’s counsel has termed Presumptively Untimely Claims. This group consists of class members whose claims fall outside of s. 5(2) of the Limitations Act as having received epidural injections more than 2 years before the commencement of the action. These class members raise issues surrounding discovery of their claim that need not be addressed by other class members.

III. The test for decertification

[16] The authority for decertifying a class proceeding is found in s. 10(1) of the CPA:

On the motion of a party or class member, where it appears to the court that the conditions mentioned in subsections 5(1) an (2) are not satisfied with respect to a class proceeding, the court may amend the certification order, may decertify the proceeding or may make any other order it considers appropriate.

[17] This court since early in its class action jurisprudence has observed that, “[t]he same conditions relate to certification and decertification”: Peppiatt v Royal Bank of Canada (1996), 1996 CanLII 7952 (ON SC), 27 OR (3d) 462, para 45 (Gen Div). There is little difference, therefore, between the Defendants’ motion here and the analysis to be undertaken in a certification case of first instance.

[18] As a caution, the courts have frequently reiterated that parties are not to approach a motion under s. 10(1) of the CPA as if it is an appeal of the certification order. “The moving party has…the burden of showing that the earlier decision would not have been made in the light of new evidence – including evidence of facts that…demonstrate that, contrary to the original finding, it is not manageable as a class action”: Pearson v Inco Limited, 2009 CanLII 9371, para 24 (SCJ). As Plaintiffs’ counsel notes in their factum, the Defendants previously consented to certification, which makes it difficult (although not impossible) for them to satisfy this burden.

[19] The court also has authority under s. 8(3) of the CPA to amend and expand the common issues. The key is for the amendments to make efficient use of judicial resources and contribute to furtherance of the goals of class actions: Ducharme v Solarium de Paris Inc., 2013 ONSC 2540, para 19. This includes the power to restate and clarify the common issues once discoveries have been complete and in advance of trial in an effort to better focus the common issues trial: Rumley v British Columbia, 2003 BCSC 234, para 90.

[20] It is understood that certification under the CPA is not a particularly high hurdle. “[A]n issue can be a common issue even if it makes up a very limited aspect of the liability question and even though many individual issues remain to be decided after its resolution”: Western Canadian Shopping Centres Inc. v Dutton, [2001] SCR 534, para 39. The essential principles with respect to common issues have been stated by Strathy J. in Singer v Schering-Plough Canada Inc., 2010 ONSC 42, para 140 and restated by the Court of Appeal in Fulawka v Bank of Nova Scotia, 2012 ONCA 443, para 81:

The proposed common issue must be a substantial ingredient of each class member's claim and its resolution must be necessary to the resolution of that claim: Hollick v Toronto (City), 2001 SCC 68, [2001] 3 SCR 158, para. 18.

A common issue need not dispose of the litigation; it is sufficient if it is an issue of fact or law common to all claims and its resolution will advance the litigation for (or against) the class: Harrington v Dow Corning Corp., 1996 CanLII 3118 (BC SC), [1996] BCJ No. 734, 48 CPC (3d) 28 (SC), aff'd 2000 BCCA 605, [2000] BCJ No. 2237, leave to appeal to SCC ref'd [2001] SCCA No 21.

With regard to the common issues, ‘success for one member must mean success for all. All members of the class must benefit from the successful prosecution of the action, although not necessarily to the same extent.’ That is, the answer to a question raised by a common issue for the plaintiff must be capable of extrapolation, in the same manner, to each member of the class’: . . . Dutton, at para. 40, [Ernewein v General Motors of Canada Ltd., 2005 BCCA 540, 46 BCLR (4th) 234], at para 32; Merck Frosst Canada Ltd. v Wuttunee, 2009 SKCA 43, [2009] SJ No 179 (SC), at paras. 145-146 and 160.

A common issue cannot be dependent upon individual findings of fact that have to be made with respect to each individual claimant: Williams v Mutual Life Assurance Co. of Canada (2000), 2000 CanLII 22704 (ON SC), 51 OR (3d) 54, at para 39, aff'd 2001 CanLII 62770 (ON SCDC), [2001] OJ No 4952, 17 CPC (5th) 103 (Div Ct), aff'd 2003 CanLII 48334 (ON CA), [2003] OJ No 1160 and 1161 (CA); Fehringer v Sun Media Corp., [2002] OJ No 4110, 27 CPC (5th) 155 (SCJ), aff'd [2003] OJ No 3918, 39 CPC (5th) 151 (Div Ct).

Where questions relating to causation or damages are proposed as common issues, the plaintiff must demonstrate (with supporting evidence) that there is a workable methodology for determining such issues on a class-wide basis: Chadha v Bayer Inc., [2003] OJ No 27, 2003 CanLII 35843 (CA), at para. 52, leave to appeal dismissed [2003] SCCA No 106, and Pro-Sys Consultants Ltd. v Infineon Technologies AG, 2008 BCSC 575, [2008] BCJ No 831 (SC) at para 139.

IV. Common issues

[21] With these principles in mind, the list of common issue questions proposed by the Plaintiff can be addressed topic by topic.

a) Negligence

[22] Plaintiff’s counsel propose the following rephrased common issues relating to the negligence claim:

a) Whether the Defendants owed a duty of care to the Class to take reasonable precautions to prevent the transmission of health care associated infections (“Duty”)?

b) What was the standard of care applicable to each Defendant relating to their Duty (the “Applicable Standards”)?

c) Whether the Defendants breached the Applicable Standards (“Breach/es”)?

[23] The first question raises no special difficulty. The defendants are all health care professionals in a medical clinic and the class are all patients of the clinic. There is every reason to conclude that the duty of care owed to one of the class will define the duty owed to them all. The relationship at this basic level was uniform for the entire class.

[24] Counsel for Dr. James take issue with the second and third questions. They submit that an analysis of the standard of care, and whether that standard was breached, requires a case by case analysis. In Dr. James’ view, it is not enough that, for example, an expert witness testify to the fact that properly sterilizing the patient’s back at the point of injection is a necessary IPAC measure. There must be evidence as to whether Dr. James did or did not properly sterilize each patient’s back, and this evidence inevitably entails an individualized assessment.

[25] Dr. James’ counsel point out that this is particularly the case since the Plaintiff takes the position that Dr. James did not have an invariable practice in this regard. If his practice varied from patient to patient, they contend, then so will the evidence with respect to each class member.

[26] Plaintiff’s counsel submits that this puts the Plaintiff and other class members in an impossible position. To continue with the same example, Dr. James’ IPAC measures, including sterilization at the point of injection, were done literally behind the Plaintiff’s back. In effect, they contend that it cannot be the case that a physician can be held liable for unsterile injections to the arm or other visible part of the body, but that a physician is immune from liability where the injection has taken place beyond the patient’s line of vision. Some means of proof other than direct visual evidence must be relied on in such a case.

[27] What Plaintiff’s counsel propose is reliance on indirect rather than direct evidence. For the most part, this evidence will be drawn from evidence about what Plaintiff’s counsel contends is the general state of Dr. James’ IPAC measures during the class period. That evidence will apparently come from Dr. James himself, the clinic nurses, the observations of TPH investigators, and observations from some of the patients themselves. Counsel for the Plaintiff submits that this is sufficient to meet the relatively low “some basis in fact” standard for the existence of the common issues with respect to standard of care: Pro-Sys Consultants Ltd. V Microsoft Corp., 2013 SCC 57, [2013] 3 SCR 477.

[28] Dr. James’ counsel argue that the evidence proposed by Plaintiff’s counsel is insufficiently cogent to constitute proof, and that in any case much of it is inadmissible. Given that this is a certification motion rather than the common issues trial itself, I do not agree with that submission. That is, there is certainly admissible evidence – the observations by patients and nurses of Dr. James’ actual practices are certainly admissible. And while these may or may not ultimately prove that Dr. James fell below the requisite standard of care for each of the procedures that he performed, that determination is for the common issues trial.

[29] Likewise, the evidence of TPH inspectors, which Dr. James’ counsel characterize as inadmissible, may be admissible for some purposes and not for other purposes, and it may form only part of an entire picture; that does not mean, however, that at this stage it must be dismissed in its entirety. The TPH personnel did not make any direct observations about Dr. James’ IPAC at the time of his actual treatment of any class members, but they did make direct observations about the state of the facility in which he worked and other related matters. This evidence, when combined with the observations by nurses or other patients, may or may not add up to proof at trial. But it is certainly “some evidence” at this stage of the action.

[30] Counsel for Dr. James further contends that an issue such as this, which relies on evidence regarding individual infections, is insufficiently common to meet the test for certification. They particularly stress that Dr. James did not have a universal practice in terms of IPAC measures, and so indirect evidence that explains his measures with respect to one patient may not be applicable to any other patient.

[31] While of course it is true that the issues for a common issues trial must, first and foremost, be common, they do not have to be proved on a balance of probabilities to be common at the certification motion. After all, it may turn out that the current evidence by Dr. James is incorrect and that he did in fact have a universal, or partly universal practice (i.e. one that he engaged in for all instances of a given procedure, if not for all procedures). That will have to be determined at trial. It is to be kept in mind that s. 5(1) of the CPA set out criteria for what is a procedural motion, not a determination on the merits. There is before me some evidence – although perhaps shy of outright proof – of commonality. That is all that is required under s. 5(1) of the CPA.

[32] Furthermore, the standard of care that the Plaintiff has put in issue is not limited to Dr. James’ IPAC measures at the time of administering the pain relief injections. It also encompasses the claim that Dr. James failed to remediate his practice once he became aware of an outbreak of infections among his patients. That allegation will rely on expert evidence regarding proper medical practice in respect of remediating specific practices, much of which is already in the record and was explored in discoveries. It will also rely on evidence by various patients in terms of when they reported their symptoms to Dr. James as well as the evidence of Dr. James himself.

[33] Again, it would be premature at this stage to opine on the strength or weakness of the evidence in this regard. That said, I have little hesitation in saying that there is “some evidence” that Dr. James failed to change his practices prior to the last of the class members experiencing infections. The question for trial will be what Dr. James knew, when he knew it, and what he ought to have done in the circumstances. All of those issues can be addressed once by Dr. James and need not be addressed in a series of repetitive, one-off trials.

b) Causation

[34] The Plaintiff’s counsel also propose three new common issues relating to causation:

d) Was any Breach sufficient to have caused or contributed to Clinical Infection in the Infected Patients?

e) Should an inference be drawn that any Breach, in the absence of evidence to the contrary, caused or contributed to Clinical infection in the Infected Patients?

f) Did any Breach cause or contribute to Clinical Infection in the Genetically Linked Patients?

[35] The causation issues represent the central controversy of the present motion. The problem, as explained by Dr. James’ counsel, is that “questions of causation must be determined on an individual basis before any liability of the defendants to pay damages in respect of infection contracted by an infected patient will arise”: Rose v Pettle, [2004] OJ No 739, para 33.

[36] Dr. James’ view is that “the Plaintiff ha[s] not demonstrated, with supporting evidence, that there is a workable methodology for determining on a class-wide basis that [Dr. James’ IPAC measures] caused [infections]…for each individual member of the class”: Norman v Thunder Bay Regional Health Sciences Centre, 2015 ONSC 3252, para 94. He submits that in the circumstances of this case (as with virtually all negligence cases), causation requires individual analysis of each patient’s situation.

[37] The Plaintiff, not surprisingly, is of the opposite view. While Plaintiff’s counsel concede that that the answer to question (d) above may be negative – that is, it seems likely that no breach of the standard of care can be directly identified as the cause of a particular patient’s infection. The Plaintiff’s case therefore turns on inferences of causation to be drawn from an identifiable breach. To return to the example discussed above, if it is established that Dr. James failed to engage in proper sterilization procedures at the point of injection, is there “some evidence” on which to infer that this breach caused class members’ infection?

[38] It is worth noting that there is already a body of case law indicating that infectious outbreaks are well suited to certification: see Anderson v Wilson, 1999 CanLII 3753 (Ont CA); Healey v Lakeridge Health Corp., 2011 ONCA 55. The overall commonality of cause and effect among the health care professionals and their patients in these situations often allows issues of duty and standard of care, their breach, factual causation and proximate cause to be addressed on a collective basis: see, e.g. Rizzi v Handa, 2019 ONSC 1802 (SCJ). Plaintiff’s counsel take the position that such is the case here, where even Dr. James’ expert, Dr. Nel Rau, testified in cross-examination that the Clinic was an “environmental disaster” [Q. 663], and that he “had enough information from the environmental sampling and from the nurse testimony about the nature of the cleaning to come to that conclusion. That did not require individual case assessment at all” [Q. 665].

[39] The Supreme Court of Canada pointed out R v Villaroman, 2016 SCC 33, [2016] 1 SCR 1000, para 42, that circumstantial evidence must be analyzed in the light of human experience. The Court went on to draw from R v. Dipnarine, 2014 ABCA 328, paras 22, 24-25, to the effect that this kind of indirect evidence “‘does not have to totally exclude other conceivable inferences’; that the trier of fact should not act on alternative interpretations of the circumstances that it considers to be unreasonable; and that alternative inferences must be reasonable, not just possible.”

[40] It is thus no longer the case that “the degree of probability that the occurrence of the facts proved to be proved is so high that the contrary cannot reasonably be supposed”: Martin v Osborne (1936), 55 CLR 367, 375 (Ont HC), at p. 375. Rather, the Supreme Court of Canada has made it clear that, “[A]n inference of causation may be drawn although positive or scientific proof of causation has not been adduced”: Snell v Farrell, 1990 CanLII 70 (SCC), [1990] 2 SCR 311.

[41] The standard articulated by the Supreme Court in Villaroman makes it possible, although difficult, to establish causation without having definitive proof in individual cases. To do so, the Plaintiff must provide some basis to conclude that the epidemiological evidence might be sufficient to establish individual causation, or at least advance the case in that respect: Merck Frosst Canada Ltd. v Wutunee, 2009 SKCA 43, leave to appeal refused, [2008] SCCA No 512.

[42] In their factum, counsel for the Plaintiff proposes that “the number of relatively rare infections is sufficient evidence to prove causation on a balance of probabilities”. The numerical basis for this submission will be supported in the evidentiary record through expert testimony. From this data, Plaintiff’s counsel proposes a “robust and pragmatic approach to the…facts”: Blatch v Archer (1774), 1 Cowp 63, 98 ER 969, 970. Plaintiff’s counsel is confident that at least “some evidence” – indeed, they submit that there is substantially more than just “some” – exists to establish causation on a class-wide basis. Plaintiff’s expert epidemiologist, Dr. Freeman, has provided a taste of that evidence when he explained in cross-examination:

The greatest risk factor that affect [class members] is their exposure at the clinic and then everything else is so minimal and tiny in comparison that being part of the cluster is what defines them as the most likely cause of their illness.

[43] As Plaintiff’s counsel explain it, Dr. Freeman’s opinion, on which they primarily rely for the statistical causation argument, is not based on the assumption that the 20 or so known infections were caused by what they put forward as evidence of Dr. James’ substandard IPAC. Rather, Dr. Freeman’s opinion is that the number of associated infections is sufficient to draw the inference. The rate is too high, and clusters such as the one found in the Rothbart Centre should not ordinarily occur.

[44] It is the Plaintiff’s submission that Dr. Freeman’s expert evidence demonstrates that there is a valid methodology to infer negligence throughout the class period on the basis of epidemiology. Counsel for the Plaintiff point out that Dr. James’ own expert witness, Dr. John Conly, essentially agrees that this methodology is a valid one. In cross-examination, Dr. Conly conceded that if an epidemiologist followed the required epidemiologic and clinical assessments to draw a conclusion, “one could use that conclusion to at least draw an inference with respect to the likely determinants of the outbreak”.

[45] Dr. James’ counsel finds this the most problematic part of the Plaintiff’s case. In their view, factual causation is an essential ingredient in establishing tort liability and the Plaintiff’s proposed methodology is effectively designed not to address but to skirt the issue. They submit that no breach of IPAC or any other act for which Dr. James is responsible can be imputed to every class member based on statistics alone. They further point out that courts have previously held that, “Causation is an individual issue with respect to every infected member of the class”: Anderson v Wilson, 1999 CanLII 3753 (ON CA), [1999] OJ No 2494, para 28.

[46] Counsel for Dr. James further note that Dr. Freeman did not review any of the underlying medical records or interview any of the actual patients to assess their medical outcomes. He did, however, acknowledge in cross-examination that the facts of each individual patient’s case are relevant to any conclusion as to what caused their medical complication. As Dr. James’ counsel point out, had Dr. Freeman engaged in a patient-by-patient examination, his inference from the statistical data that all of the infections were caused by Dr. James may have changed. In making this observation, they point to a number of infected patients that suffered from different forms of complications or infections in different anatomical locations that reduce or potentially eliminate the possibility that Dr. James was the source. As in Norman, supra, para 95, Dr. James’ counsel contends that, “An expert opinion would be needed to prove causation in each case.”

[47] I agree with Dr. James’ counsel that the Plaintiff’s approach is novel. In their factum, Plaintiff’s counsel assert, without citing any authority, that “the Plaintiff is not required to prove any specific breach to establish liability.” For this reason, they have not only added some proposed questions to the negligence/causation common issues, but seek to eliminate the previous question premised on Dr. James having routine, invariable IPAC practices. In saying that it does not matter what Dr. James did in treating each individual class member, Plaintiff’s counsel are effectively saying that the class, and the issues of breach and individual causation, can be treated collectively and statistically as would an epidemiologist conducting a study of the clinic.

[48] In my view, the Plaintiff’s proposed methodology is rather close to the now largely rejected, or much diluted doctrine of res ipsa loquitur. Dr. Freeman’s analysis that the rate and type of infection is rare enough that there is no other likely cause than Dr. James’ negligent IPAC is much like the classic ‘scalpel left behind’; no direct proof is needed because any other explanation other than the doctor’s negligence and causation is virtually inconceivable. Plaintiff’s counsel submits that when circumstantial evidence of this nature permits an inference of fault, “the plaintiff may succeed unless the defendant offers an explanation to negate that inference”: Hassen v Anvari, [2001] OJ No 6085, para 10, aff’d 2003 CanLII 1005 (Ont CA).

[49] As Dr. James’ counsel point out, however, epidemiological statistics are not so clear cut. Moreover, they submit that the shifting burden of proof described by Plaintiff’s counsel in these circumstances is overstated. The Supreme Court of Canada stated in Fontaine v British Columbia (Official Administrator), 1998 CanLII 814 (SCC), [1998] 1 SCR 424, para 23, that, “As in any negligence case, the plaintiff bears the burden of proving on a balance of probabilities that negligence on the part of the defendant caused the plaintiff’s injuries. The invocation of res ipsa loquitur does not shift the burden of proof to the defendant.” It is Dr. James’ position that the Plaintiff cannot use the class action certification as a means of replacing causation analysis with collective, epidemiological-like analysis, and thereby avoid the actual proof of negligence and causation that the law requires.

[50] It is still the case that, “There are situations where the facts merely whisper negligence, but there are other circumstances where they shout it aloud”: Allen M. Linden, Canadian Tort Law (5^th ed., 1993), p. 2. But the Plaintiff cannot use this common sense notion to displace actual proof that Dr. James’ negligent conduct caused the Plaintiff’s (or any other class member’s) injury. In employing an epidemiological approach to the infections at issue here, the Plaintiff attempts to walk a fine line between inferred proof and an educated form of conjecture that does not amount to proof. Dr. Freeman’s methodology may work, but it may not.

[51] In my view, the real test for this methodology will come at trial. The proposed methodology for approaching questions of standard of care and causation could potentially work to prove the Plaintiff’s case, depending on how the evidence comes out. After all, it is not clear at this stage whether Dr. James will turn out to have a universal approach to IPAC, or whether the infections incurred by the 21 known patients are more uniform in nature, or whether and when Dr. James was aware of the rate of infection and what he did about it.

[52] As with a novel cause of action, a certification motion does not strike me as the right context to dispense with a novel methodology of proof. The methodology should be tested in the laboratory of a trial and in light of all of the admissible evidence. The Supreme Court has said in the well-known case of Hunt v Carey, 1990 CanLII 90 (SCC), [1990] 2 SCR 959, it must be “plain and obvious” that the novel approach cannot succeed to dispense with the possibility at a preliminary stage; “if there is a chance that the Plaintiff might succeed, then the Plaintiff should not be ‘driven from the judgment seat’”. This court has applied the ‘plain and obvious’ test in the context of class actions since the early days of its jurisprudence under the CPA: see Chippewas of Sarnia Band v Canada (Attorney General) (1996), 1996 CanLII 8015 (ON SC), 29 OR (3d) 549, 564 (Gen Div).

[53] Accordingly, I am prepared to certify questions (d) and (e) as common issues, to be answered at trial. There is enough evidence and a proposed methodology for adducing that evidence to pass the certification hurdle, although this by no means guarantees that the proposed methodology will work the way the Plaintiff wants it to work. The questions may be answered in the affirmative based on the evidence of Dr. Freeman and the other circumstantial evidence brought to bear by Plaintiff’s counsel; conversely, they may be answered in the negative based on the ultimate failure of Dr. Freeman’s evidence and other circumstantial evidence to satisfy the law’s requirement that the Plaintiff actually prove her case.

[54] The proposed causation question with respect to the Genetically Linked Patients asks about those class members who were infected by MSSA bacteria. The methodology is not particularly different for these patients, but it is the Plaintiff’s position that proof of negligence and causation will be far easier to ultimately achieve. The expert evidence is expected to show that the rarity of MSSA is such as to establish a link to Dr. James that is all but impossible to refute. Again, whether or not that pans out in the evidence, and the inference suggested by the Plaintiff can be made by the trier of fact, will have to wait until trial. But there is good reason to distinguish this subclass of patients since the factual basis of their claim may result in a different answer for them than for the rest of the class.

c) Fiduciary duty

[55] Counsel for the Plaintiff further propose four new common issues relating to fiduciary duty:

g) Whether the Putative Fiduciary Defendants owed a fiduciary duty to the Class?

h) For those Putative Fiduciary Defendants found to owe a Fiduciary Duty, what was the nature of the fiduciary duty owed to the Infected Patients (the “Fiduciary Duty”);

i) For those Putative Fiduciary Defendants found to owe a Fiduciary Duty, whether these Defendants, or any of them, breached their Fiduciary Duty (the “Fiduciary Breaches”)?

j) Whether the Fiduciary Breaches, or any of them, caused or contributed to Clinical Infection in the Infected Patients?

[56] Counsel for Dr. James object to the proposed questions dealing with fiduciary duties on similar grounds as they do with respect to the negligence question. That is, they contend that the issue of fiduciary duties was inextricably engaged with the specific relationship between the alleged fiduciary and the beneficiary, and, accordingly it necessitates an analysis of each specific relationship. Dr. James’ counsel submit that since each class member may have had a different relationship with the Defendants, there is no commonality embedded in the fiduciary duty question.

[57] The position taken by Dr. James is premised on other cases in which doctor-patient relationships were scrutinized by courts. As an illustration, Dr. James’ counsel cite Norman, supra, for the proposition that although relationships between medical professionals and institutions on one hand and patients on the other can give rise to fiduciary duties owed to the patients, this will turn on the specifics of each situation. As the court put it in Norman, para 96:

The Hospital and the various doctors, nurses, pharmacists and technicians owe duties to patients in different fact situations. In some situations there would be no relationship between a particular doctor and a particular patient, or between a particular nurse, pharmacist or technician and a particular patient.

[58] One can understand the logic of the Norman case given that the setting of the claim was a regional hospital with a large staff of physicians and different types of supporting professionals and staff. It stands to reason that the treating physician is in a different relationship to the patient than an emergency room doctor who examines a patient on admission to the hospital and then never encounters her again or a radiologist who interprets the patient’s X-rays without ever actually meeting the patient in person. Each relationship in a large medical institution must be evaluated on its own terms.

[59] The Rothbart Centre is not that kind of institution. Of course, there is a different relationship between doctors and patients on one hand and nurses and patients on the other hand. But all of the class members were patients of Dr. James. It is in that capacity that they fall within the definition of class members. Dr. James’ relationship with his patients – all of whom came to him for pain relief symptoms and all of whom received one form or another of epidural injection – is not being compared with that of another doctor and her patients. Rather, Dr. James’ doctor-patient relationships are being evaluated on their own terms.

[60] Needless to say, the Defendants who are nurses are situated differently than Dr. James and will have to be considered separately. Likewise, the Defendant, Peter Rothbart, as the administrator of the Rothbart Centre, may fall into a different category and will have to be analyzed separately. But the physician at the core of this case, Dr. James, either had a fiduciary relationship with his patients or not. Whether he gave them an injection in the lumbar spine or the cervical spine, the ingredients of fiduciary duties can be evaluated on a common basis for all of Dr. James’ pain relief patients.

[61] As Plaintiff’s counsel points out that the healthcare provider-patient relationship is a previously recognized category in which fiduciary duties have been found to exist: McInerney v MacDonald, 1992 CanLII 57 (SCC), [1992] 2 SCR 138. Indeed, McLachlin J. (as she then was) observed in Norberg v Wynrib, 1992 CanLII 65 (SCC), [1992] 2 SCR 226, para 63, that “…the relationship of physician to patient also falls into that special category of relationships which the law calls fiduciary.” There is no reason not to subject the Defendants – and certainly Dr. James – to the analysis that the Supreme Court has suggested is an appropriate approach to the duties owed by a healthcare practitioner.

[62] The record already contains some basis in fact for concluding that not only was Dr. James in a fiduciary relationship with his patients, but that he breached that duty by his failure to take appropriate steps when he became aware of the infections suffered by his patients. Dr. Catherine Smyth, the Plaintiff’s pain management expert, has opined that, “Dr. James did not honour his duty to patients in that he failed to act in a timely way or with appropriate steps when he first became aware of the serious complications associated with his ESIs (epidural spinal injections) in 2011.”

[63] Of course, the timing of Dr. James’ knowledge of the problems will have to be explored in evidence at trial, as will his response. As with anything else that turns on trial evidence, it may or may not turn out that Dr. Smyth is correct in her opinion. In fact, Dr. James’ infectious diseases expert, Dr. Neil Rau, has indicated in his affidavit that there are a number of factual matters that need to be explored – the timing of the symptom onset, timing of the diagnosis, timing of the patient’s notification of the Rothbart Centre, incubation period between the procedure and the onset of symptoms, etc. – that need to be explored prior to any firm conclusion being drawn about breach of duty.

[64] Plaintiff’s counsel concedes that since the fiduciary claim is predicated on the knowledge of one or more infections among the class members, only a subset of the class is likely to benefit from a finding of breach. But it is not possible at this point to determine on what date a court would find there was a breach, if any. That said, there is sufficient commonality, and “some basis in fact” to explore these questions at a common issues trial.

d) Limitation period

[65] Plaintiff’s counsel also seek to add a new common issue relating to the limitation period:

k) Could the claims of the Presumptively Untimely Claims subclass have been discovered within the meaning of section 5 of the Limitations Act more than 2 years prior to September 9, 2014?

[66] Counsel for Dr. James submit that the running of a limitation period turns on discoverability, and by definition discoverability requires an assessment of each individual’s stare of knowledge. The Court of Appeal acknowledged this in Smith v Inco, 2011 ONCA 628, para 164, where it indicated that, “It is an error to treat the limitation period as running from the date when a majority, even an overwhelming majority, of the class members knew or ought to have known the material facts in issue.”

[67] On the other hand, Plaintiff’s counsel states that a significant amount of the evidence points to the limitation issue being uniform among the class members. As an example, they indicate that the evidence thus far shows that Dr. James did not tell any patients about the outbreak of infections, leading to the conclusion that none of his patients would have had the requisite knowledge to discover the claim that he failed to remediate once becoming aware of the outbreak.

[68] Accordingly, depending on the findings at trial the applicable limitation period may or may not be an individual issue. I see no reason at this stage not to certify the single limitation question as a common issue, on the understanding that there may be subclasses of patients to whom the answer applies and others to whom the answer does not apply. I note that the limitation question is indeed one of the issues on which Plaintiff’s counsel has proposed that subclasses be identified.

e) Punitive damages

[69] Finally, Plaintiff’s counsel seek to rephrase the question relating to punitive damages:

l) For the Putative Fiduciary Defendants, whether there is conduct sufficient to attract punitive damages, and if so, whether punitive damages should be awarded and in what amount?

[70] As phrased by Plaintiff’s counsel, the punitive damages question is wed to the fiduciary duty question. Punitive damages are understandably not being sought with respect to the negligence claim but only with respect to the breach of fiduciary duty claim. With the fiduciary duty questions being certified, I see no reason not to certify the punitive damages question as well.

[71] The Plaintiff has advanced a claim for punitive damages only against Dr. James, Dr. Rothbart, and the Rothbart Centre. If none of those Defendants are found to have breached a fiduciary duty to class members, then there will be no basis for awarding punitive damages.

[72] If all or some of Dr. James, Dr. Rothbart, and/or the Rothbart Centre are found to have breached a fiduciary duty to class members, then the question of punitive damages and their quantum can be addressed. There is no suggestion that the amount of punitive damages should vary as among class members, and so there is sufficient commonality to proceed with this question as part of the common issues trial.

V. Subclasses

[73] Subclasses can be designated at any stage of a class action: Bendall v McGhan Medical Corp. (1993), 1993 CanLII 5550 (ON SC), 14 OR (3d) 734. A division of the class into one or more subclass is considered appropriate where there are common issues applicable to some, but not all, class members: Caputo v Imperial Tobacco Ltd., 2004 CanLII 24753 (SCJ).

[74] As set out in paragraphs 15 and 16 above, the proposed subclass are Genetically Linked Patients and class members with Presumptively Untimely Claims.

[75] The Genetically Linked Patients are those whose infections can be matched to MSSA, the strain of bacteria colonized on Dr. James. For these individuals, the methodology of proof proposed by Plaintiff’s counsel is likely to be far more relevant and effective. If negligence is determined for the class as a whole, this subclass will have a relatively easier time establishing factual causation. In that case, the issue of damages will be all that remains to be determined.

[76] The Plaintiff is not a class member that would fit into this subclass. However, she has no opposing interest and there is no reason to doubt that she can fairly represent the interests of this subclass.

[77] The second proposed subclass, persons with Presumptively Untimely Claims, consists of all class members whose claims are presumed to fall outside of s. 5(2) of the Limitations Act, 2004, Schedule B. The timing of the Plaintiff’s epidural injection in August 2012 makes her a member of this subclass. Again, she appears to me to be perfectly capable of representing the interests of this subclass (and all other class members).

VI. Preferable procedure

[78] Given the novel methodology for proving the Plaintiff’s case, and the fact that it threatens to circumvent having to prove negligence and/or causation for each and every claimant, Dr. James’ counsel is of the view that a class action does not represent the preferable procedure for this case. Instead, they propose that the 21 or so class members each bring individual claims and that they all be tried together at one time by one trial judge. Counsel for Dr. James submit that in this way, some of the efficiencies of a class action will remain, but the trial can proceed in a way that does not dispense with crucial ingredients of legal proof on a claimant-by-claimant basis.

[79] Plaintiff’s counsel objects to this proposal, arguing that even if tried together, 21 individual cases, with 20 claimants testifying, does not represent a saving of judicial resources and goes only a short way toward facilitating access to justice. They point out that “the underlying question is whether allowing the suit to proceed as a [class action] will avoid duplication of fact-finding or legal analysis”: Western Canadian Shopping Centres, supra, para 39. Counsel for the Plaintiff submit that if certification on the basis of the new common issues and subclasses fulfills that goal, the class action procedure is the preferable one.

VII. Certification

[80] Law, Sartre might have said in his most aptly named work, is like a science. “You can demonstrate that you are right and that others are wrong”: Jean-Paul Sartre, Les Mains Sales (1948), act 5, scene 2. But when it comes to how one goes about that demonstration, he might equally have had a more innovative view: “How terrified you are of sullying your hands… What good will it do [to]…invoke purity as your rationalization for doing nothing”: Ibid.

[81] The question is, where direct proof is unattainable, can one innovate so that correlation equate to legal proof? Or, on the contrary, will the trier of fact dismiss evidence that does not conform to the most traditional methodology?

[82] The methodology offered by the Plaintiff is not foolproof, but it is sufficient at this stage to pass the relatively low certification hurdle. It is therefore equally sufficient to overcome the motion for decertification. As the Supreme Court has observed, the ‘some basis in fact’ test “does not require that the court resolve conflicting facts and evidence at the certification stage”: Pro-Sys, supra, para 102.

[83] Moreover, the common issues trial is an adaptable mechanism. Indeed, the CPA overall is imbued with wide latitude for case and trial management, such that it provides for “flexibility and adjustment at all stages of the proceeding… If it turns out later that there was a change in the standard of care at point in time X that could affect liability, it would be possible then to divide the class between those treated before point in time X and those treated later”: Anderson v Wilson (1999), 1999 CanLII 3753 (ON CA), 44 OR (3d) 673. The same flexible approach may apply in the present case, for example in respect of different categories of patients who received different types of epidural injections from Dr. James.

[84] This case presents novel issues of proof in the context of collective claims, which must be tested in the fire of a common issues trial with a full evidentiary record. Although it may challenge the boundaries of the law, it is not plain and obvious that the theory of liability offered by Plaintiff’s counsel must fail. Accordingly, I am prepared to amend the certification Order as requested by the Plaintiff.

VIII. Disposition

[85] The Plaintiff’s pleading is amended to reflect the changes contained in the Fresh As Amended Statement of Claim.

[86] The common issues are amended. They are now as set out in paragraphs 22, 34, 55, 65, and 69 above.

[87] Two subclasses are added to the class definition reflect the Genetically Linked Patients and the Presumptively Untimely Patients.

[88] Dr. James’ motion to decertify the action is dismissed.

[89] The parties may make written submissions as to costs. I would ask Plaintiff’s counsel to provide me with brief written submissions (3 pages maximum) within two weeks of the date of this judgment, and Dr. James’ counsel to provide me with equally brief submissions within two weeks thereafter.

Morgan J.

Date: September 3, 2019