Stirrett v. Cheema et al.

[Indexed as: Stirrett v. Cheema]

Ontario Reports Court of Appeal for Ontario, Hoy A.C.J.O., van Rensburg and L.B. Roberts JJ.A. May 6, 2020 150 O.R. (3d) 561 | 2020 ONCA 288

Counsel: William D. Black and Christine Wadsworth, for appellant. Gavin MacKenzie, Brooke MacKenzie, Richard Bogoroch and Toby Samson, for respondent. Barbara Legate, for intervenor Ontario Trial Lawyers Association. Jeremy Opolsky and Alexandra Shelley, for intervenor Canadian Cancer Trials Group. John Adair and Jordan V. Katz, for intervenor Canadian Cardiovascular Research Collaboratory.

BY THE COURT:

I Overview

[1] Tragically, David Stirrett died on February 12, 2005 as a result of complications that arose during an angiogram he underwent as part of his participation in a clinical research trial. Karen Stirrett, the respondent, sued the two doctors who performed the angiogram on her husband. She also sued the appellant, Dr. Bradley Strauss, the medical researcher who headed up the clinical research trial at the hospital where the angiogram was performed. 1

[2] This appeal raises two main issues: first, whether the trial judge erred in finding that the appellant, as a medical researcher, owed and breached an ad hoc fiduciary duty to Mr. Stirrett, as a participant in a clinical research trial; and, second, whether the trial judge erred in accepting that causation was "remove[d] . . . from the analysis" in determining the respondent's entitlement to compensation for breach of fiduciary duty.

[3] We decline to decide the first issue. The research intervenors' submissions had the effect of framing the ad hoc fiduciary duty issue differently on appeal than how it was argued before the trial judge. While we acknowledge that the purpose of the intervenors' submissions was to present a different perspective, their arguments would require us to make determinations with respect to an issue not squarely argued at trial, namely whether the trial judge erred in failing to consider whether the appellant undertook to act in Mr. Stirrett's best interests. Given our conclusion on the causation issue, it is not necessary for us to determine this issue.

[4] Even if the appellant owed Mr. Stirrett an ad hoc fiduciary duty and breached that duty in the manner found by the trial judge, the trial judge erred by not taking into account causation and awarding damages for the breach. As we shall explain, the respondent was not entitled to compensation because the breach did not cause Mr. Stirrett to undergo the February 2005 angiogram that led to his death. Accordingly, we allow the appeal.

II The Background

(1) The Stent Restenosis and Metabolism Study

(a) The recruitment of Mr. Stirrett

[5] Mr. Stirrett was a 54-year-old man who was a non-insulin dependent Type II diabetic. He began to have chest pain and shortness of breath in early 2004 and was referred for an angiogram. An angiogram uses x-rays and contrast dye to view blood vessels supplying the heart. The angiogram revealed a 90 per cent blockage, or stenosis, of his left circumflex artery and he was scheduled to undergo an angioplasty at St. Michael's Hospital (the "Hospital") in Toronto four days later, on June 11, 2004.

[6] That day, before the successful angioplasty, Mr. Stirrett was recruited to participate in the Stent Restenosis and Metabolism Study (the "Study").

(b) An overview of the Study

[7] The Study was a clinical trial examining whether the use of insulin to control glucose levels in non-insulin dependent diabetics with coronary artery disease would reduce restenosis (the recurrence of stenosis -- the narrowing of a blood vessel) after balloon angioplasty or stenting. Restenosis results in restricted blood flow, occurs at a higher rate in diabetics and is particularly difficult to treat.

[8] Study participants were randomized into either the control group or the experimental group. The control group, of which Mr. Stirrett was a part, received "best medical care", including exercise and dietary recommendations, and the experimental group received insulin. For medical research purposes, all participants would have a follow-up angiogram six months after the initial procedure to see how much restenosis had occurred.

[9] The Study operated at seven hospitals throughout Canada. The appellant, an interventional cardiologist, was the Study's principal investigator at the Hospital.

(c) The Tri-Council Policy Statement

[10] The medical researchers conducting the Study were required to comply with the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (the "Policy"), a policy adopted by the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council of Canada and the Social Sciences and Humanities Research Council of Canada. The Policy describes standards and procedures governing research involving human subjects. Among other things, the Policy requires that research be overseen by a Research Ethics Board ("REB") and imposes disclosure obligations on researchers. It stipulates that "[r]esearchers shall provide, to prospective subjects . . . full and frank disclosure of all information relevant to free and informed consent" and that participants shall be given "continuing and meaningful opportunities for deciding whether or not to continue to participate".

(d) Changes to the on-going Study

[11] The REB at each participating hospital was required to approve the Study and, on an annual basis, to approve its continuation. The REB at the Hospital first approved the Study in February 2002 and annually approved its continuation during the relevant period. The protocol for the Study, provided to the REB, described a "planned recruitment of 240 patients". While Data Safety Monitoring Boards ("DSMB") are not mandatory, the protocol also provided that a "Data and Safety Monitoring Committee (to be named) . . . will oversee the progress of the trial and monitor safety of the intervention".

[12] For various reasons, the Study had difficulty in recruiting the planned number of participants, which had an effect on the Study's funding. The Heart and Stroke Foundation of Ontario (the "Foundation"), which had granted funding to the Study for a three-year period, advised in September 2003 that it would not release new moneys corresponding to the third year of the Study (2003-2004) due to the Study's shortfall in recruiting participants. The Foundation permitted the Study to retain the unspent funds, totalling $171,942, that were already released to it and to use them for the duration of the grant year ending June 30, 2004. Mr. Stirrett was not advised of this change in funding.

[13] The appellant and other researchers involved in the Study were of the view that the Study should continue until it reached 100 participants. The appellant disagreed with two research colleagues who felt the REBs at their respective hospitals should be advised of both the change in the intended number of participants and the Foundation's decision not to release additional funds. In the end, the appellant did not advise the REB at the Hospital of those changes. However, the Hospital's REB was aware of the slower than hoped for enrolment. In the annual report to the Hospital's REB, dated February 9, 2004, the appellant reported that there were 55 participants enrolled in the Study. He explained that "[e]nrollment has been slower than anticipated at the start of the study however the sponsor has included more sites and enrollment is now steadily increasing", and that "[r]ecruitment has been slower than anticipated due to the fact that some patients live too far away to participate in follow-up or are reluctant to inject themselves with insulin". All of the hospitals continued to participate in the Study.

[14] The DSMB contemplated by the protocol was not established. The appellant's evidence at trial was that this was because the concern about hypoglycemic episodes that motivated the provision for the DSMB never materialized. The REB at the Hospital was not advised of the failure to establish a DSMB.

(e) The Consent Form

[15] Mr. Stirrett signed a consent form when he enrolled in the Study on June 11, 2004 (the "Consent Form"). The Consent Form played a significant role at trial.

[16] The Consent Form indicated that it contained all the relevant information for Mr. Stirrett to decide whether to participate in the Study. The Consent Form also explained that "[y]our doctor will inform you of any new information about the study that might develop during the course of this research and might influence your willingness to participate in the study" and that "you may discontinue participation at any time during the study without penalty".

[17] The Consent Form identified the sponsor of the Study as the Foundation. It described the Study as a study "of about 240 patients similar to yourself", although at the time that Mr. Stirrett signed the Consent Form, the number of planned participants had already been reduced.

[18] The Consent Form explained that the follow-up angiogram at six months was not part of a regular clinical practice, but rather was part of the Study. It also stated that there was a one in one thousand risk of a serious complication, such as heart attack, stroke, or death, from an angiogram. Mr. Stirrett was also informed of the risks associated with the follow-up angiogram on other occasions, including when he signed a further consent form prior to undergoing the procedure.

[19] The Consent Form stated that "[t]he treatment may or may not be of personal benefit to you but the information gathered from the study will be very important in discovering new treatments in people like yourself", and "[i]n addition, you may benefit from the follow-up angiogram and ultrasound performed at six months post-angioplasty which is not part of the routine examination".

(f) The follow-up angiogram

[20] The follow-up angiogram contemplated by the Study was performed on February 10, 2005.

[21] The respondent testified that her husband was again experiencing chest pain in the months before the follow-up angiogram. He had begun carrying nitroglycerin spray with him at all times, using it when he had chest pain. She also testified that both he and she would have wanted to find out the cause of the pain. It was common ground that the angiogram was a diagnostic tool to discover the cause of Mr. Stirrett's chest pain.

[22] Tragically, Mr. Stirrett died on February 12, 2005 from complications which arose during the follow-up angiogram. The post mortem examination confirmed that Mr. Stirrett had experienced very significant restenosis.

(g) The publication of the Study

[23] The Study continued following the death of Mr. Stirrett, who was the 73rd participant out of a total of 78 participants in the Study. The Study results were inconclusive. A paper based on the Study was published in 2012 and has been subsequently cited in other published studies. The paper identified the Foundation, among other organizations, as having funded the Study.

(2) The litigation

[24] The respondent sued the two doctors who performed the follow-up angiogram and the appellant, both in negligence and for breach of fiduciary duty. The parties agreed on damages, but not on the issue of liability. The action against the doctors who performed the angiogram was dismissed in its entirety. Below, we outline how the litigation proceeded.

(a) The respondent's negligence claim against the appellant

[25] In her Fresh as Amended Statement of Claim of March 19, 2018, the respondent alleged that her husband's death was caused by negligence and/or breach of contract. In the case of the appellant, the pleaded particulars of the negligence and/or breach of contract included: failing to ensure that the Study was monitored by a DSMB; failing to provide Mr. Stirrett with all of the relevant information regarding the Study that he required in order to decide whether he should participate and undergo the February 10, 2005 angiogram; and failing to halt the Study when required in the circumstances.

[26] The negligence claim proceeded to trial before a jury in March 2018. The respondent put four main arguments to the jury in advancing her negligence claim against the appellant.

[27] First, she alleged that the appellant breached the standard of care of a reasonable and prudent principal investigator of the Study by failing to

• establish a DSMB in accordance with the research protocol;
• advise the REB that a DSMB had not been established;
• amend the protocol to indicate how the goals of the Study had changed;
• obtain subsequent REB approval for the redesigned Study;
• amend the Consent Form for the Study to reflect accurate information, including that the Study planned to recruit 100 participants, not 240.

[28] Second, she alleged that, but for the breach, Mr. Stirrett would not have undergone the follow-up angiogram, since had the appellant acted in accordance with his standard of care, the Study likely would have been terminated before the date of the angiogram.

[29] Third, the respondent argued that the appellant had breached the standard of care of a reasonable and prudent principal investigator in obtaining Mr. Stirrett's informed consent. Among other things, the respondent relied on the fact that the intended number of participants on the Consent Form was incorrect, that the impact this had on the Study's ability to achieve its primary objective was not communicated to Mr. Stirrett, and that the sponsorship information on the Consent Form was misleading.

[30] Finally, the respondent argued that, but for the failure to obtain Mr. Stirrett's informed consent, a reasonable person in Mr. Stirrett's circumstances would not have undergone the angiogram on February 10, 2005.

[31] As is described below, the jury partially accepted the respondent's first argument but rejected the other three.

(b) The respondent's fiduciary duty claim against the appellant

[32] The respondent also pleaded breach of fiduciary duty. She claimed the appellant breached the standard of care and his fiduciary duties to Mr. Stirrett by failing to establish a DSMB when the protocol required it and failing to halt the Study when it became apparent due to the low number of participants that the Study was unlikely to achieve its primary objective. She further alleged that the appellant owed a "special duty of care" to Mr. Stirrett, which included a duty to take special care to ensure Mr. Stirrett did not feel pressured to consent to the follow-up angiogram. She also alleged that, "[i]n breaching this duty of care", the appellant and the two other doctors breached their fiduciary duty to Mr. Stirrett.

[33] Claims for "other equitable relief", such as breach of a fiduciary duty, must be tried without a jury: Courts of Justice Act, R.S.O. 1990, c. C.43, s. 108(2). Accordingly, following the jury's verdict, the trial judge heard submissions and decided the fiduciary duty claim.

[34] The respondent contended that the appellant owed Mr. Stirrett a fiduciary duty and advanced three broad arguments as to how this duty was breached. First, she alleged that the appellant breached a fiduciary duty owed to Mr. Stirrett by not telling him (1) the new sample size; (2) the primary objective of the Study had changed as a result of the change in sample size; and (3) the Foundation was not providing additional funding for the third year of the Study. Second, the respondent argued that the failure to set up a DSMB was a breach of fiduciary duty because it was an important oversight mechanism. Finally, the respondent argued that the appellant breached his fiduciary duty by putting his interests as a researcher ahead of those of Mr. Stirrett by failing to stop the Study once its primary objective had changed.

(c) The jury's verdict on the negligence claim against the appellant

[35] There is no appeal from the jury's verdict. The appeal arises only from the trial judge's decision on the fiduciary duty claim. However, we outline the jury's verdict, given that it has a bearing on our causation analysis.

[36] In the case of the appellant, the jury was asked four questions, corresponding to the four arguments the respondent put to the jury, which are outlined above. On the first question, the jury found the appellant had breached the standard of care of a reasonable and prudent principal investigator in the following four ways:

• aample size should have been changed in the protocol from 240 patients to 100 patients;
• the DSMB was never set up as set out in the protocol;
• the Consent Form was never updated with the new sample size from 240 patients to 100; and
• no protocol deviation was submitted to the REB based on the information above.

[37] On the second question, however, the jury found that the respondent had not proven on a balance of probabilities that, but for the particulars of the breach of the standard of care, Mr. Stirrett would not have undergone the angiogram that resulted in his death.

[38] The jury also found, on the third question, that the respondent had not proven on a balance of probabilities that the appellant had breached the standard of care of a reasonable and prudent principal investigator in obtaining Mr. Stirrett's informed consent. As a result, the jury did not proceed to answer the fourth question of whether, but for a breach in the standard of care in obtaining Mr. Stirrett's informed consent, Mr. Stirrett would have undergone the follow-up angiogram on February 10, 2005.

[39] Consequently, the appellant was not liable in negligence.

(d) The trial judge's reasons on breach of fiduciary duty

[40] The trial judge began by noting that a fiduciary relationship can arise in a doctor-patient relationship but it is not limited, as the defendants argued, to cases such as Norberg v. Wynrib, [1992] 2 S.C.R. 226, where the patient is exploited by the doctor. For what gives rise to a fiduciary duty, he relied on Frame v. Smith, [1987] 2 S.C.R. 99 and Hodgkinson v. Simms, [1994] 3 S.C.R. 377.

[41] The trial judge found that the scope of the relationship between the appellant and Mr. Stirrett was one of researcher to patient, in addition to doctor and patient. He stated that the "obligation of a researcher to the participant when it involves humans is more strict than a doctor to patient relationship": at para. 47. While he accepted that not all aspects of the participant to researcher relationship, where it involves research on humans, will rise to the level of a fiduciary duty, he concluded that the appellant owed a fiduciary duty to Mr. Stirrett. He described the duty owed, at paras. 49-50, as a duty to comply with the Consent Form, drafted in accordance with the Policy, and to advise the REB:

The fiduciary duty was to comply with the terms set out in the consent form as drafted and agreed by David Stirrett. The consent form was drafted in accordance with the principles set out in the Tri-Counsel [sic] Policy Statement. The consent form required Dr. Strauss to inform David Stirrett of new information "about the study that might develop during the course of this research and might influence your willingness to participate in the study" . . .. Dr. Strauss failed to do so as stated in the particulars of negligence given by the jury. In addition, it was apparent or at least contemplated by more than one other source that the study should not continue in its existing format without resubmitting the revised protocol to each participating hospital's Research Ethics Board. The letter from the Heart & Stroke Foundation of September 23, 2003 contemplated the "closing" of the study. The emails of Dr. Cohen and Dr. Seidelin expressly set out their view that the changes required notice and approval by their respective Research Ethics Boards.

While the changes made in the STREAM study from when the consent form was drafted may not have been significant or changed the risk of harm to David Stirrett, it was not something for Dr. Strauss to decide. His obligation, or duty, was to pass on these changes to David Stirrett (and to the Research Ethics Board) in order to permit them to re-evaluate their previous decision. This would have protected Dr. Strauss from liability.

[42] He also concluded, at para. 51, that the appellant had breached that duty:

Dr. Strauss, by his actions failed to give David Stirrett the opportunity to consider, reflect and determine if he should "discontinue participation at any time during the study without penalty". . . . By not providing the information about the STREAM study which varied from the content of the consent form that was explained to David Stirrett on June 11, 2004 and to which Dr. Strauss, as principal investigator at St. Michael's Hospital for the STREAM study agreed, Dr. Strauss breached his fiduciary duty.

[43] The trial judge agreed with the respondent's submission that a finding of a fiduciary duty and a breach of that duty "removes causation from the analysis on whether there will be recovery as occurs in the determination of negligence followed by causation": at para. 52.

[44] The trial judge granted judgment against the appellant in the agreed amount of damages.

III The Issues on Appeal

[45] The two main issues raised by this appeal can be articulated as follows:

(i) Did the trial judge err in finding that the appellant owed and breached a fiduciary duty to Mr. Stirrett in the circumstances of this case? (ii) If the appellant did owe a fiduciary duty, and breached that duty, did the trial judge err on the issue of causation?

[46] The appellant also argues that the trial judge provided insufficient reasons for his decision. However, given our disposition of this appeal, it is unnecessary to address this further argument.

IV The Fiduciary Duty Issue

[47] It is important to begin by noting what is and what is not at issue on this appeal.

[48] The Canadian Cancer Trials Group and the Canadian Cardiovascular Research Collaboratory (the "research intervenors") submit that the trial judge's decision potentially imposes a novel fiduciary obligation on all research doctors to advance the best interests of individual human research subjects and that he erred in finding that physicians who are researchers owe fiduciary duties to voluntary participants in clinical studies. According to the research intervenors, research doctors should not owe such duties because they would have a chilling effect on the operation of clinical trials in which a researcher seeks broader knowledge and a societal benefit greater than the individual participant's interests. They say that researchers are required to balance the interests of participants in clinical trials with their duty to maximize the benefits of a research study for the advancement of knowledge and that this balancing of interests is inconsistent with the recognition of a fiduciary duty. The research intervenors are concerned about the precedential impact of this case and its practical implications for medical researchers in Canada.

[49] By contrast, the Ontario Trial Lawyers Association ("OTLA") argues that, not only did the trial judge correctly find that the appellant owed and breached his fiduciary duty to Mr. Stirrett, but that this court should go further and generally find that there is a breach of fiduciary duty where a researcher has failed to obtain the approval of a REB. The OTLA says that the recognition of a fiduciary duty owed by a researcher to a human subject will enhance the community's trust in an important institution.

[50] The trial judge did not find, nor did any of the parties to the appeal argue, that the relationship between a medical researcher and study participant is a new category of per se fiduciary relationship, such as solicitor-client or trustee-beneficiary. Further, the parties to this appeal did not dispute that there could be circumstances in which a medical researcher may owe an ad hoc fiduciary duty to a participant in a clinical trial.

[51] As such, the question in this appeal is not whether medical researchers owe a per se fiduciary duty to research participants, or even in general terms when an ad hoc fiduciary duty may arise. It is unnecessary and beyond the proper scope of this appeal for us to resolve any such issue in the context of this appeal.

[52] The narrow question raised by this appeal is whether the trial judge erred in finding that the appellant owed an ad hoc fiduciary duty to Mr. Stirrett in the particular circumstances of this case.

[53] In Alberta v. Elder Advocates of Alberta Society, [2011] 2 S.C.R. 261, 2011 SCC 24, at para. 36, McLachlin C.J.C., writing for the court, stated what a claimant must establish before a court will impose a fiduciary duty outside of the traditionally recognized categories of per se fiduciary relationships:

In summary, for an ad hoc fiduciary duty to arise, the claimant must show, in addition to the vulnerability arising from the relationship described by Wilson J. in Frame: (1) an undertaking by the alleged fiduciary to act in the best interests of the alleged beneficiary or beneficiaries; (2) a defined person or class of persons vulnerable to a fiduciary's control (the beneficiary or beneficiaries); and (3) a legal or substantial practical interest of the beneficiary or beneficiaries that stands to be adversely affected by the alleged fiduciary's exercise of discretion or control.

[54] Here, the trial judge recognized that the issue was whether the factual matrix before him raised a fiduciary duty. While he did not cite Elder Advocates for the factors giving rise to a fiduciary duty, he relied on Frame and Hodgkinson, two other leading cases from the Supreme Court on fiduciary duty.

[55] The appellant repeats arguments that he made before the trial judge, that is, that the requisite vulnerability and discretion to impose a fiduciary duty were not present in this case. We reject those arguments as the elements of vulnerability and discretion are fully supported by the evidence. The appellant, who is an accomplished researcher, had all the information about the Study that might influence Mr. Stirrett's willingness to participate and could control the disclosure of that information; the information was complex and extremely important; and Mr. Stirrett, who had no expertise and relied on the appellant to provide that information, could not make an informed decision without that information.

[56] With respect to the additional requirement emphasized in Elder Advocates for establishing an ad hoc fiduciary duty -- an undertaking by the alleged fiduciary to act in the best interests of the alleged beneficiary -- the research intervenors argue that the trial judge erred by failing to expressly determine whether there was an implicit or explicit undertaking by the appellant to act in the best interests of Mr. Stirrett. As a result, the research intervenors argue, the trial judge's finding that the appellant owed an ad hoc fiduciary duty to Mr. Stirrett is not entitled to the normal deference it would be entitled to on appeal: see Waxman v. Waxman, 2004 ONCA 39040, at paras. 501, 716, leave to appeal to S.C.C. refused [2004] S.C.C.A. No. 291.

[57] In this case, the trial judge cited [at para. 41] the following passage from pp. 409-10 S.C.R. of Hodgkinson: "[W]hat is required is evidence of a mutual understanding that one party has relinquished its own self-interest and agreed to act solely on behalf of the other party." While Elder Advocates does not talk about a "mutual understanding", this passage from the trial judge's reasons includes the requirement in Elder Advocates of an undertaking by the alleged fiduciary to act in the best interests of the alleged beneficiary or beneficiaries.

[58] We acknowledge, however, that the trial judge did not explicitly analyze whether the appellant undertook to act in Mr. Stirrett's best interests in relation to the interest in question. He did not do so for the good reason that it was not the subject of argument before him. No party referenced Elder Advocates or discussed the undertaking requirement.

[59] The respondent argued at trial that all the defendant physicians owed fiduciary duties to Mr. Stirrett and that those duties required them to put Mr. Stirrett's best interests before their own. Counsel referred to the "hallmarks" of a fiduciary relationship, as set out in Frame, at p. 136 S.C.R., which are: (1) the fiduciary has scope for the exercise of some discretion or power; (2) the fiduciary can unilaterally exercise that power or discretion so as to affect the beneficiary's legal or practical interests; and (3) the beneficiary is peculiarly vulnerable to or at the mercy of the fiduciary holding the discretion or power. It was submitted that physicians, as a class, owe fiduciary duties to their patients.

[60] In response, the appellant argued that not every relationship or interaction between doctors and patients is fiduciary in nature. He made no reference to the undertaking requirement in Elder Advocates. He confined his trial submissions to the other elements of an ad hoc duty, framing them in accordance with the three hallmarks of a fiduciary relationship. The appellant maintained that Mr. Stirrett was not vulnerable and that the defendant physicians did not engage in any conduct that would come near to the kind of egregious abuse of discretionary power, such as in Norberg, that amounted to a breach of fiduciary duty. It was also the appellant's position that the jury's findings on negligence foreclosed any finding of liability for breach of fiduciary duty that was grounded on the same facts and that the action failed on the issue of causation.

[61] Even assuming, without deciding, that the trial judge erred by failing to analyze the "undertaking" element of an ad hoc fiduciary duty, we are not persuaded that we should undertake that analytical exercise. It would require us to make a determination of mixed fact and law on an issue not argued at trial. Given our conclusion on the causation issue, it is not necessary for us to do so.

[62] In declining to deal with this issue, we should not be taken as expressing any opinion on whether the trial judge was correct in accepting that a fiduciary duty was owed to Mr. Stirrett, or that, if a duty were owed, it was the duty articulated by the trial judge. Nor should these reasons be read as foreclosing or requiring a finding of a fiduciary relationship between a medical researcher and study participant in another case. The presence or absence of a fiduciary relationship will depend on the evidence in each particular situation. Any time it is alleged that one party owes an ad hoc fiduciary duty to another party, the court must engage in an examination of the particulars of the relationship to determine whether the Elder Advocates test is satisfied in the circumstances.

V The Causation Issue

[63] The causation issue is dispositive of the appeal. For the purposes of this discussion, we accept, without deciding, that the appellant breached his fiduciary duty to Mr. Stirrett in the manner described by the trial judge.

[64] As indicated above, the trial judge concluded that a finding of fiduciary duty and a breach of that duty obviated the need to consider the issue of causation. The trial judge stated the following, at paras. 52-53 of his reasons:

Having found a fiduciary duty existed and that Dr. Strauss breached that duty, the third issue is to determine the legal outcome. Counsel for the plaintiff submitted the finding of a fiduciary duty and the breach of that duty removes causation from the analysis on whether there will be recovery as occurs in the determination of negligence followed by causation. I agree.

I rely on the statement of Justice McLachlin in Norberg v. Wynrib, [1992] 2 S.C.R. 226 at paragraph 95, "Equity has always held trustees strictly accountable in a way the tort of negligence and contract have not". In addition, Justice McLachlin goes on to state, at paragraph 98, "The physician is pledged by the nature of his calling to use the power the patient cedes to him exclusively for her benefit. If he breaks that pledge, he is liable".

[65] As we will explain, the trial judge misstated the law regarding the role of causation in cases involving a breach of fiduciary duty. Simply put, for compensation to be awarded for breach of fiduciary duty, the plaintiff must establish that the defendant's breach caused the plaintiff's loss. As a result, the trial judge erred in failing to consider and determine the issue of causation.

[66] Considering and applying the correct principles with respect to the issue of causation afresh, assuming that the appellant owed an ad hoc fiduciary duty to Mr. Stirrett and breached that duty, we conclude that there was no causal link between the appellant's breach and the angiogram Mr. Stirrett underwent that led to his death. Moreover, the jury's determination with respect to "but for" causation in the negligence claim is determinative. In the result, the respondent is not entitled to damages.

[67] Before turning to an analysis of the trial judge's reasons, it is first necessary to review the law on causation in the fiduciary context.

(1) Compensation for breach of fiduciary duty

[68] Compensation for breach of fiduciary duty is typically determined according to restitutionary principles, where the plaintiff is entitled to be put in as good a position as he or she would have been in had the breach not occurred: Hodgkinson, at p. 440 S.C.R., per La Forest J. In M. (K.) v. M. (H.), [1992] 3 S.C.R. 6, La Forest J., writing for the majority, pointed out that in equity there is no capacity to award damages and that the distinction between damages and compensation is often slight, with the courts tending to merge the principles of law and equity when necessary to achieve a just remedy: at pp. 80-81 S.C.R. Over time, courts have used the term "damages" to denote monetary compensation for breach of fiduciary duty. Remedies in cases of breach of fiduciary duty (such as disgorgement of profits and exemplary compensation) can also have a prophylactic or deterrent purpose: Strother v. 3464920 Canada Inc., [2007] 2 S.C.R. 177, 2007 SCC 24, at paras. 74-77. Irrespective of the purpose, there must be a causal link between the breach of fiduciary duty and the compensation sought.

[69] As we will explain, when considering equitable compensation, or damages, the fiduciary breach must have been the cause in fact -- the effective cause -- of the loss in respect of which compensation is sought. There is of course a difference between the right to a remedy, and the assessment of damages. Causation in fact is relevant to the first issue. Legal causation, which incorporates limiting factors such as remoteness, proximity, foreseeability and intervening act, is part of the second issue.

[70] We acknowledge that a source of confusion over the role of causation is in the use of the word "causation" in some of the cases both to describe causation in fact and as part of the test for applying common law limiting factors to limit the extent of a damages claim. These two uses should not be confounded.

[71] In the tort context, Philip H. Osborne cautions that causation in fact, which focuses on the factual issue of the sufficiency of the connection between the defendant's wrongful act and the plaintiff's loss, should not be confused with the "control device" of remoteness of damages, sometimes known as proximate cause, which may excuse a defendant from liability for loss caused to the plaintiff on the ground of fairness: Philip. H. Osborne, The Law of Torts, 5th ed. (Toronto: Irwin Law, 2015), at p. 54.

[72] Similarly, Sir Andrew Tipping, in "Causation at Law and in Equity: Do We Have Fusion?" (2000), 7:3 Canterbury L. Rev. 433, at p. 433, emphasizes the conceptual difference between the need to demonstrate a causal relationship that is "separate from and precedes the further controls provided by the concepts of foreseeability and remoteness".

[73] Writing on fiduciary law, Leonard I. Rotman distinguishes between legal and factual causation: Leonard I. Rotman, Fiduciary Law, (Toronto: Carswell, 2005). At p. 634, he writes:

Both the common law and Equity require that there be some connection between the harm or loss caused and the actions of the person who is alleged to be liable for it.

Each starts with the idea of "but for", "cause-in-fact", or "sine qua non" causation. This generally satisfies Equity, but the common law requires more; it demands a finding of materiality or substantial cause to link the impugned activity with the harm to the plaintiff. Further, the common law imports ideas of foreseeability (or reasonable contemplation) and remoteness into its assessment of causality. Mitigation of losses is another relevant consideration under the common law's assessment of damages for harm or loss, as is contributory negligence. These other considerations do not readily enter into Equity's assessment of fiduciary accountability.

(Footnotes omitted)

[74] As this passage suggests, and as we will discuss below with reference to the case law, cause in fact is required in the fiduciary context. This case turns on the cause in fact requirement: did the respondent prove that the appellant's breach of fiduciary duty caused the loss in respect of which compensation is sought? That loss is Mr. Stirrett's death from the February 2005 angiogram. Because damages were agreed, it is unnecessary to go further and consider the extent of recoverable losses or whether a different measure should apply where the claim is for breach of fiduciary duty rather than negligence.

[75] We turn now to the case law on causation.

(2) Factual causation

[76] While the appropriate approach to the assessment -- that is, the measure and extent -- of damages for breach of fiduciary duty was the subject of debate and discussion in the Supreme Court during the 1990s, namely, in Canson Enterprises Ltd. v. Boughton & Co., [1991] 3 S.C.R. 534, and Hodgkinson, the Supreme Court was unanimous that whether dealing with a common law cause of action, or a claim sounding in equity, the plaintiff must establish that the defendant's wrong was the cause in fact of some injury or loss.

[77] Canson involved a claim against a solicitor who handled a real estate transaction and who failed to disclose to his clients, the purchasers, a secret profit made by a third party. It was claimed that the solicitor was not only liable for the secret profit but also for the losses flowing from the negligence of engineers and pile-drivers who performed work on the purchased property.

[78] All eight justices who heard the case agreed that the defendant solicitor should be liable for the secret profit but not for the construction losses that were caused by the engineers. And, while the justices expressed differing opinions on whether and when the common law limiting factors would apply to compensation for breach of fiduciary duty, importantly, they agreed that, in order to award compensation for breach of a fiduciary duty, there must be a loss or injury that "flows from" or "results from" the breach.

[79] La Forest J., writing for the majority, stated, at pp. 578-79 S.C.R., that, "[i]n the case of a mere breach of duty [by contrast to a breach of trust], the concern of equity is to ascertain the loss resulting from the breach of the particular duty", and that it was imperative "to ascertain the loss resulting from the breach of the relevant equitable duty" (emphasis added).

[80] While generally agreeing with La Forest J., Stevenson J. wrote brief separate reasons in which he noted that the "losses [were] too remote, not in the sense of failing the 'but for' test, but in being so unrelated and independent that they should not, in fairness, be attributed to the defendant's breach of duty": at p. 590 S.C.R. (emphasis added).

[81] McLachlin J., writing for herself, Lamer C.J.C., and L'Heureux-Dubé J., agreed that the court was engaged in determining the loss resulting from the breach of the relevant equitable duty: at p. 551 S.C.R. She distinguished causation in fact from legal causation, stating, at p. 552 S.C.R., that "[t]he requirement that the loss must result from the breach of the relevant equitable duty does not negate the fact that 'causality' in the legal sense as limited by foreseeability at the time of the breach does not apply in equity" (emphasis added). Similarly, she stated that "while the loss must flow from the breach of fiduciary duty, it need not be reasonably foreseeable at the time of the breach": at p. 552 S.C.R. (emphasis added).

[82] The Supreme Court next addressed compensation for breach of fiduciary duty in Hodgkinson, a case involving alleged breaches of fiduciary duty and contract in the performance of a contract for investment advice and other tax-related financial services.

[83] La Forest J., writing for the majority, found that Mr. Simms, an investment advisor, owed, and had breached, a fiduciary duty to Mr. Hodgkinson. He found that the damages owed for breach of fiduciary duty were the same as for breach of contract, taking into account the impact of market fluctuations that occurred after the breach.

[84] La Forest J. referred to the task of determining the damages "flowing from" the breach of fiduciary duty. The investment advisor induced Mr. Hodgkinson to make investments that he would not have otherwise made by deliberately concealing his own financial interest, thus "initiat[ing] the chain of events leading to the investor's loss": at p. 443 S.C.R.

[85] In summary, in Canson and Hodgkinson there was a causal link between the breach of fiduciary duty of the defendant and the harm to the plaintiff. The plaintiffs in Canson would not have entered into the transaction if the defendant solicitor had disclosed the secret profit. And, in Hodgkinson, the plaintiff would not have made the investments if he had known of the true relationship between the defendant and the developers. The point of contention in these cases was not whether a loss was caused by the breach, but the extent to which compensation for consequential losses could be recovered.

[86] The need for cause in fact to be established before compensation or damages are awarded for breach of fiduciary duty has also been consistently recognized by this court.

[87] For example, in Martin v. Goldfarb, [1997] O.J. No. 1918, the plaintiff claimed damages as a result of losses suffered in commercial dealings with a disbarred lawyer who had been convicted of fraud. Following a first trial awarding $5.95 million in damages to the plaintiff, a successful appeal by the defendant, and a new trial which resulted in the dismissal of his claim, the plaintiff appealed. The issue on the appeal of the new trial was whether the trial judge erred in requiring that the plaintiff's personal losses be direct: see Martin v. Goldfarb (2003), 68 O.R. (3d) 70, 2003 ONCA 28757. This court stated, at para. 8, that "[d]amages cannot be awarded absent evidence of a causal connection". This court held that the trial judge was justified in dismissing the claim because the plaintiff had not established a causal connection in fact between the losses he sustained in a bankruptcy and the breach of fiduciary duty.

[88] Further, in Waxman, this court accepted that "[t]he basic rule of equitable compensation is that the injured party will be reimbursed for all losses flowing directly from the breach": at para. 651 (emphasis added).

[89] Also, in Standard Trust Co. (Liquidator of) v. Metropolitan Trust Co. of Canada, 2007 ONCA 897, MacFarland J.A. confirmed, in a breach of fiduciary duty case, that the trial judge was required "on a common sense and reasonable consideration of the evidence, [to] conclude what the losses were that flowed from the breach": at para. 49 (emphasis added).

[90] To put it succinctly, a plaintiff seeking compensation for breach of fiduciary duty must establish that the losses flowed from the breach.

[91] We add this. While legal causation is not at issue in this appeal, we note that the Supreme Court and other appellate courts have accepted that common law limiting principles may apply to limit equitable compensation in order to treat similar wrongs similarly, but only where (1) it is necessary to achieve a just and fair result; and (2) doing so does not raise any policy concerns: Canson, at pp. 581, 586-87 S.C.R., per La Forest J.; Hodgkinson, at p. 443 S.C.R., per La Forest J.; Waxman, at para. 662; and Dhillon v. Jaffer, 2016 BCCA 119, at paras. 26-28.

[92] For example, in M. (K.), La Forest J. declined to award any additional compensation for a parent's breach of fiduciary duty after concluding that the underlying policy objectives for compensation were the same as those animating the jury's award of damages for sexual assault and battery in a case of incest: at pp. 81-82 S.C.R.

[93] We now turn to the trial judge's decision in this case.

(3) The trial judge's decision

[94] With respect, an examination of the trial judge's reasons demonstrates that he fell into error on the question of causation through his reliance on the comments by McLachlin J. in Norberg. As noted, at para. 52, he agreed with plaintiff's counsel that "the finding of a fiduciary duty and the breach of that duty removes causation from the analysis on whether there will be recovery as occurs in the determination of negligence followed by causation".

[95] Decided a year after Canson, Norberg dealt with allegations of sexual assault in the context of a doctor-patient relationship. McLachlin J., who was joined in her reasons by only L'Heureux-Dubé J., found there was a breach of fiduciary duty. McLachlin J. made the following comments, at pp. 290 and 293 S.C.R. of Norberg, which were relied on by the trial judge in this case on the issue of causation:

Equity has always held trustees strictly accountable in a way the tort of negligence and contract have not.

The physician is pledged by the nature of his calling to use the power the patient cedes to him exclusively for her benefit. If he breaks that pledge, he is liable.

[96] The trial judge's reliance on these comments was misplaced.

[97] We do not read McLachlin J.'s comments as supporting the proposition that "causation is removed" from the analysis in the fiduciary context. She made these observations in her discussion of whether there was a breach of fiduciary duty. Later in her reasons, in addressing damages, McLachlin J. noted that "Dr. Wynrib's breach of his duty to Ms. Norberg caused the following losses or injuries to her: (1) prolongation of her addiction; and (2) sexual violation": at p. 295 S.C.R. (emphasis added). After reviewing the evidence, she indicated she "would award an additional $20,000 for suffering and loss during the period of prolonged addiction for which Dr. Wynrib was responsible": at p. 296 S.C.R. Thus, causation in fact was present.

[98] Moreover, as we have explained, the trial judge's reading of McLachlin J.'s comments is inconsistent with Canson, Hodgkinson and case law from this court, which affirm the place of cause in fact in the fiduciary context.

[99] The trial judge failed to consider cause in fact, as he was required to do. We turn finally to explain why it is not made out in this case.

(4) Failure to establish causation

[100] As we will explain, applying the findings of the jury to this case, the claim for breach of fiduciary duty inevitably founders on causation. The claims against the appellant for breach of fiduciary duty and breach of the standard of care of a principal investigator in this case were based on substantially the same conduct.

[101] There was an agreement in this case on the amount of damages, which were premised on the loss of Mr. Stirrett's life following the February 2005 angiogram.

[102] While the jury did not need to assess damages, it was still necessary for them to determine whether the wrong that they found the appellant to have committed -- his breach of the standard of care of a principal investigator -- had caused the loss in question. After concluding that he had breached the standard of care in four ways, the jury responded in the negative when asked whether the plaintiff had proven, on a balance of probabilities, that but for the breaches of the standard of care they had described, Mr. Stirrett would not have undergone the February 2005 angiogram.

[103] The causation issue, as it developed in the evidence, addressed two questions. The first was whether, if the appellant had not breached his standard of care in the ways the jury had identified, the Study would have continued, or whether it would have been terminated once it was known that its stated purpose could not be achieved and that the Foundation had stopped its funding. The second question was whether, if the appellant had not breached his standard of care in the ways the jury had identified, Mr. Stirrett would have refused to participate in the Study or withdrawn from it. The argument was that, in either case, Mr. Stirrett would not have had the February 2005 angiogram.

[104] The jury's conclusion that there was no "but for" causation between the appellant's breach of the standard of care as a principal investigator and Mr. Stirrett's decision to undergo the February 2005 angiogram can only mean that the jury was not satisfied: (1) that the Study would have been terminated once the proper information had been provided to the REB; and (2) that Mr. Stirrett would not have participated in the Study, or would have withdrawn from it, if the proper information had been provided to him. In essence, the jury must have concluded that, notwithstanding the breach of the appellant's standard of care, in the ways they had identified, the Study would have continued, and Mr. Stirrett would not have ceased his participation and would have still undergone the February 2005 angiogram.

[105] The same issues are part of the causation analysis when the cause of action is breach of fiduciary duty. In determining whether the breach of fiduciary duty was the cause in fact of the respondent's loss, the trial judge would have had to answer the following question: would Mr. Stirrett have remained in the Study and undergone the February 2005 angiogram as part of the Study, if the appellant had not breached his fiduciary duty?

[106] The issue is whether there is a causal link between the breach of fiduciary duty and Mr. Stirrett's decision to undergo the February 2005 angiogram. The fact that Mr. Stirrett would have undergone the angiogram despite the appellant's breach of fiduciary duty breaks the chain of factual causation in the fiduciary duty claim, just as it did in the negligence claim. The result would have been the same.

[107] This is the case whether or not causation in a breach of fiduciary duty case is described as "but for" causation. That said, causation in the context of a breach of fiduciary duty is properly characterized as "but for" causation. "But for" causation is not simply a common law concept. It means that the defendant's breach of duty was necessary to bring about the plaintiff's loss. The defendant's wrong need not be the sole cause of the loss, but it must be part of the cause. "But for" causation raises the counterfactual question: what would likely have happened if the defendant had discharged his or her duty? Properly understood, "but for" causation simply means causation in fact.

[108] As we have seen, a number of the fiduciary duty cases speak of causation in fact in what are in substance "but for" terms: see Canson, at pp. 578-79 S.C.R., per La Forest J., at p. 552 S.C.R., per McLachlin J.; Hodgkinson, at pp. 393-94 S.C.R., per La Forest J.; and SFC Litigation Trust (Trustee of) v. Chan (2019), 147 O.R. (3d) 145, 2019 ONCA 525, at para. 117, leave to appeal to S.C.C. refused [2019] S.C.C.A. No. 314, 75 C.B.R. (6th) 1. Similarly, in Cadbury Schweppes Inc. v. FBI Foods Ltd., [1999] 1 S.C.R. 142, at para. 74, the Supreme Court considered the economic advantage the plaintiffs would have enjoyed "but for" the defendant's breach of confidence.

[109] Returning to this appeal, the respondent makes several arguments on causation.

[110] First, the respondent argues that the appellant had the onus at trial of disproving causation. She submits that this court cannot give effect to the jury's finding on causation because it was a finding that she had not satisfied her burden to prove causation, whereas a fiduciary has the burden of proving, with concrete evidence, that the beneficiary would have suffered the same loss regardless of the fiduciary's breach.

[111] For this proposition, the respondent relies on Hodgkinson, at p. 441 S.C.R., where La Forest J. referred to "the long-standing equitable principle that where the plaintiff has made out a case of non-disclosure and the loss occasioned thereby is established, the onus is on the defendant to prove that the innocent victim would have suffered the same loss regardless of the breach".

[112] In the passage on which the respondent relies, La Forest J. continues to explain, citing Rainbow Industrial Caterers Ltd. v. Canadian National Railway Co., [1991] 3 S.C.R. 3, at pp. 14-17 S.C.R., that this principle has been affirmed with respect to damages at common law in the context of negligent misrepresentation, and that courts exercising both common law and equitable jurisdiction have approached this issue in the same manner.

[113] We do not accept the respondent's "reverse onus" approach to causation. This approach may apply where a plaintiff has first satisfied the burden of establishing cause in fact. At the stage of assessing damages, it may permit the defendant to argue that the injured party would have, in any event, assumed a position "other than the status quo ante" and that this should be taken into account in calculating damages: Rainbow, at pp. 15-16 S.C.R. This is the point made in Hodgkinson, where the defendant argued unsuccessfully that the plaintiff, who had relied on the defendant to make a particular investment, would have invested in tax shelters in any event, and suffered market losses.

[114] In this case, the reverse onus does not apply. The respondent did not first establish that the appellant's breach caused Mr. Stirrett to undergo the February 10, 2005 angiogram. In the present case, the "reverse onus" approach might arguably have applied to the assessment of damages, had damages not been agreed, but does not apply to the question of cause in fact.

[115] For these reasons, we do not accept the respondent's argument that the onus was on the appellant, as the breaching fiduciary, to demonstrate with concrete evidence that Mr. Stirrett would have undergone the February 2005 angiogram irrespective of the breach. The onus was on the respondent to establish that the appellant's breach was the cause in fact of the respondent's loss.

[116] Second, the respondent asserts that there was evidence to support a causal link between the appellant's breach of fiduciary duty and Mr. Stirrett's February 2005 angiogram. Essentially, the respondent argues that, based on the evidence, the Study would not have continued if the proper information had been provided to the REB and that, if Mr. Stirrett had been given the information in question, he would not have participated in the Study, or would have withdrawn from it.

[117] We agree with the appellant that this argument cannot succeed. It would require the court, in assessing the fiduciary duty claim, to make findings that are inconsistent with those made by the jury on the negligence claims.

[118] In this case, the respondent pleaded precisely the same omissions (failure to establish a DSMB and halt the Study) and relied on the same evidence to support her claims of breach of the standard of care and of the fiduciary duty. The respondent explicitly equated the alleged breach of the duty of care owed to Mr. Stirrett to the breach of fiduciary duty in the final paragraph of her pleadings. Thus, not only are the pleadings based on the same principles, but they relate to the same exact activity: see discussion in Dhillon v. Jaffer, 2016 BCCA 119, at para. 28. This is the factual matrix that was before the jury.

[119] The trial judge correctly noted, at para. 45, that he was obliged to accept the findings of the jury, as triers of the facts. Indeed, the trial judge observed that the specific breaches of the appellant's standard of care as a principal investigator identified in the particulars were consistent with the breaches relied upon in the fiduciary duty claim.

[120] In M. (M.) v. M. (P.), 2000 BCSC 1597, after a jury had determined a case of sexual assault, Bennett J. (as she then was) had to determine the fiduciary duty claims. Citing R. v. Brown, [1991] 2 S.C.R. 518, she noted that, in such circumstances, a trial judge may be required to make additional factual findings, but "is normally bound by the express and implied factual implications of the jury verdict": at para. 13.

[121] As noted above, the jury's finding in this case was that the respondent had not proven on a balance of probabilities, that, but for the appellant's failure to change the Study's sample size in the Consent Form and the research protocol, his failure to set up the DSMB, and his failure to submit a revised protocol to the REB, Mr. Stirrett would not have undergone the February 2005 angiogram that led to his death. This meant that the jury could not have been satisfied that, but for the breaches of the standard of care, either the Study would have been stopped, or Mr. Stirrett would have chosen not to participate. If either of those events had happened, the follow-up angiogram in February 2005 would not have taken place. No different result is open on the factual causation question in relation to the breach of fiduciary duty found by the trial judge.

[122] Indeed, at para. 3, the trial judge accepted that the jury found that the negligence of the appellant was not the cause of Mr. Stirrett undergoing the angiogram on February 10, 2005. At para. 42, he observed that the jury may have been influenced by the evidence that the risk of serious harm from the procedure was small, and the logic of it being better to know the extent of restenosis six to eight months following an angioplasty than not. In other words, even if he had received the correct material information, Mr. Stirrett may have decided to have the follow-up angiogram because of its expected advantages.

[123] To the extent that, in finding a breach of fiduciary duty, the trial judge pointed to failures that were not identified by the jury, such as the failure to inform Mr. Stirrett of the loss of the Foundation's sponsorship, these additional failures are similar in nature to the jury's findings on the particulars of the appellant's breach of his standard of care, and would not affect the outcome.

[124] In the end, the result is the same, whether the respondent's claim sounds in negligence or breach of fiduciary duty.

VI Disposition

[125] Accordingly, we allow the appeal and dismiss the action.

[126] If sought, the appellant is entitled to costs of the appeal in the agreed-upon, all-inclusive amount of $25,000. The order for costs in the court below is set aside and, unless otherwise agreed by the parties, the question of costs below shall be remitted to the Superior Court of Justice.

Appeal allowed.

Notes

1 Mr. Stirrett's estate was also initially a plaintiff in this action, but on March 20, 2018, the trial judge dismissed, on consent, the action brought by Mr. Stirrett's estate, leaving Ms. Stirrett's action to proceed alone. On appeal, the appellant does not challenge Ms. Stirrett's standing to bring the fiduciary claim.