Bruff-Murphy v. Gunawardena
[Indexed as: Bruff-Murphy v. Gunawardena]
Ontario Reports
Court of Appeal for Ontario
Lauwers, Hourigan and Benotto JJ.A.
June 16, 2017
138 O.R. (3d) 584 | 2017 ONCA 502
Case Summary
Evidence — Expert evidence — Trial judge erring in qualifying defence psychiatrist as expert in personal injury action despite having serious reservations about witness' methodology and independence — Trial judge failing to conduct cost-benefit analysis with respect to witness' evidence as he erroneously believed that he was obliged to qualify witness as expert if witness met Mohan threshold — Trial judge also erring in failing to exclude witness' evidence or to alert jury to problems with witness' testimony after witness clearly became partisan advocate for defence — Trial judge's gatekeeping role not ending when he qualified witness as expert — Admission of witness' testimony resulting in miscarriage of justice.
The plaintiff claimed to have suffered soft tissue damages when her vehicle was rear-ended by the defendant's vehicle. She also alleged that the accident had left her with a chronic pain condition with attendant anxiety and depression. One of the two expert witnesses called by the defence was B, a psychiatrist who conducted an independent medical examination of the plaintiff. Despite having serious reservations about B's methodology and independence, the trial judge qualified him as an expert. When B testified, it became apparent that he had crossed the line and become a partisan advocate for the defendant. His report and his testimony essentially amounted to an attack on the plaintiff's credibility. In keeping with his usual methodology, he had examined the plaintiff before reviewing her medical records, and had then gone through the medical records looking for discrepancies between what she told him and what was in the records. Those discrepancies formed the largest portion of his report. The trial judge did nothing to exclude B's opinion evidence or alert the jury to the problems with his testimony. The jury assessed general damages at $23,500 and rejected all other heads of damages. The plaintiff appealed.
Held, the appeal should be allowed.
The trial judge did not err in ruling that B could not be cross-examined regarding prior court and arbitral findings made against him.
The trial judge failed to properly discharge his gatekeeper duty at the qualification stage. He did not perform a cost-benefit analysis with respect to B's evidence because of his apparent erroneous belief that he was obliged to qualify B as an expert if B met the Mohan threshold. Had he done so, he would have concluded that the risks of permitting B to testify far outweighed any potential benefit from the proposed testimony. It was evident from a review of B's report that there was a high probability that he would prove to be a troublesome expert witness who was intent on advocating for the defence and unwilling to properly fulfill his duties to the court.
It became obvious during B's testimony that he understood his primary role to be to expose inconsistencies and not to provide a truly independent assessment of the plaintiff's psychiatric condition. Where, as here, an expert's eventual testimony removes any doubt about his independence, the trial judge must not act as if he were functus, but must continue to exercise his gatekeeper function. The court has residual discretion to exclude expert evidence even after admitting it, if later in the trial prejudice emerges that was not apparent at the time of admission. In this case, the trial judge could have advised counsel that he was going to give either a mid-trial or final instruction that B's testimony would be excluded in whole or in part, received submissions from counsel in the absence of the jury, and proceeded as he saw fit. Alternatively, he could have asked for submissions from counsel on a mistrial and ruled accordingly. The admission of B's testimony resulted in a miscarriage of justice.
Other Cases Referred To
White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23, [2015] 2 S.C.R. 182; Briscoe Estate v. Canadian Premiere Life Insurance Co., 2012 ONCA 854, 113 O.R. (3d) 161; Browne v. Dunn (1893), 6 R. 67 (H.L.); Bruff-Murphy (Litigation guardian of) v. Gunawardena, 2016 ONSC 7, [2016] O.J. No. 6; Desbiens v. Mordini, [2004] O.J. No. 4735; Gabremichael v. Zurich Insurance Co., [1999] O.F.S.C.I.D. No. 198; Morrison v. Greig, [2007] O.J. No. 225, 46 C.C.L.T. (3d) 212; Pietkiewicz v. Sault Ste. Marie District Roman Catholic Separate School Board, 71 O.R. (3d) 803; R. v. Bingley, 2017 SCC 12, [2017] 1 S.C.R. 170; R. v. Boyne, 2012 SKCA 124, [2012] S.J. No. 795; R. v. Ghorvei, 46 O.R. (3d) 63; R. v. J. (J.-L.), 2000 SCC 51, [2000] 2 S.C.R. 600; R. v. K. (A.), 45 O.R. (3d) 641; R. v. Karaibrahimovic, 2002 ABCA 102, [2002] A.J. No. 527; R. v. Mohan, [1994] 2 S.C.R. 9; R. v. Ranger, 67 O.R. (3d) 1; R. v. Shafia, 2016 ONCA 812, [2016] O.J. No. 5627; R. v. White, 2011 SCC 13, [2011] 1 S.C.R. 433; Sohi v. ING Insurance Co. of Canada, [2004] O.F.S.C.D. No. 106.
Statutes Referred To
Insurance Act, R.S.O. 1990, c. I.8, s. 267.5(12)
Rules and Regulations Referred To
Rules of Civil Procedure, R.R.O. 1990, Reg. 194, rule 4.1.01
APPEAL from the judgment of Kane J. of the Superior Court of Justice, sitting with a jury, dated August 22, 2016.
Geoffrey D.E. Adair, for appellants.
Daniel I. Reisler and Jessica L. Kuredjian, for respondent.
The judgment of the court was delivered by
HOURIGAN J.A.:
A. Introduction
[1] The law regarding expert witnesses has evolved considerably over the last 20 years. Gone are the days when an expert served as a hired gun or advocate for the party that retained her. Today, expert witnesses are required to be independent, and their function is to provide the trier of fact with expert opinion evidence that is fair, objective and non-partisan.
[2] The role of the trial judge in relation to expert witnesses has also evolved. Appellate courts have repeatedly instructed trial judges that they serve as gatekeepers when it comes to the admissibility of expert opinion evidence. They are required to carefully scrutinize, among other things, an expert witness' training and professional experience, along with the necessity of their testimony in assisting the trier of fact, before the expert is qualified to give evidence in our courts. This gatekeeper role is especially important in cases, such as this one, where there is a jury who may inappropriately defer to the expert's opinion rather than evaluate the expert evidence on their own.
[3] In the present case, the trial judge qualified an expert to testify on behalf of the defence despite some very serious reservations about the expert's methodology and independence. It became apparent to the trial judge during the expert's testimony that he crossed the line from an objective witness to an advocate for the defence. Despite his concerns, the trial judge did nothing to exclude the opinion evidence or alert the jury about the problems with the expert's testimony.
[4] On appeal, the appellants advance several arguments to the effect that trial fairness was breached, such that a new trial is necessitated. All of these arguments focus on the impugned expert.
[5] In my view, the appeal must be allowed and a new trial ordered. I reach this conclusion because the trial judge failed to properly discharge his gatekeeper duty at the qualification stage. Had he done so, he would have concluded that the risks of permitting the expert to testify far outweighed any potential benefit from the proposed testimony.
[6] In addition, the trial judge's concerns about the expert's testimony were substantially correct; the witness crossed the boundary of acceptable conduct and descended into the fray as a partisan advocate. In these circumstances, the trial judge was required to fulfill his ongoing gatekeeper function and exclude in whole or in part the expert's unacceptable testimony. Instead, the trial judge did nothing, resulting in trial fairness being irreparably compromised.
B. Background Facts
(1) The Trial
[7] The appellant Liese Bruff-McArthur was hit from behind by the respondent while stopped in her motor vehicle. She alleged that as a consequence of the accident she has suffered multiple soft tissue damages in her neck, lower back and right shoulder. Ms. Bruff-McArthur also alleged that the accident has left her in an apparent chronic pain condition with attendant anxiety and depression. She says that she is unable to work and that her enjoyment of life has been substantially diminished.
[8] Ms. Bruff-McArthur commenced an action against the respondent, who admitted liability. The sole issue in the 23-day jury trial was what damages, if any, she suffered.
[9] In support of her case, Ms. Bruff-McArthur called a number of physicians who had either treated or examined her, two of whom were retained by insurers to conduct independent medical examinations. The consensus among these witnesses was that she suffered in the manner complained of and that the cause of her suffering was the motor vehicle accident.
[10] The defence called two witnesses, both of them medical expert witnesses who had been retained by the defence to conduct independent medical examinations. The first, Dr. Gianni Maistrelli, an orthopedic specialist, testified that he found nothing wrong with Ms. Bruff-McArthur from a musculosketal standpoint. This conclusion was not surprising given that she was complaining of soft tissue injuries.
(2) Dr. Bail's Evidence
[11] The other defence expert witness was Dr. Monte Bail, a psychiatrist. Counsel for Ms. Bruff-McArthur objected to his testifying on two grounds.
[12] First, she argued that his report was essentially an attack on Ms. Bruff-McArthur's credibility. Counsel pointed to numerous instances in the report where Dr. Bail commented on discrepancies between the information Ms. Bruff-McArthur provided in her interview with him and what he later found in her medical records. Dr. Bail never put those alleged inconsistencies to Ms. Bruff-McArthur. Counsel sought an order that excluded the parts of Dr. Bail's report that did not meet the test in Browne v. Dunn (1893), 6 R. 67 (H.L.) and an order that Dr. Bail not be permitted to testify regarding his views on her credibility.
[13] Second, Ms. Bruff-McArthur argued that Dr. Bail was biased. In support of this argument, counsel submitted that she should be permitted to cross-examine Dr. Bail on findings made in another court case and two arbitrations to the effect that he was not an independent witness. The trial judge ruled, relying on R. v. Karaibrahimovic, 2002 ABCA 102, [2002] A.J. No. 527, R. v. Ghorvei, 46 O.R. (3d) 63 and Desbiens v. Mordini, [2004] O.J. No. 4735 that Dr. Bail could not be cross-examined on prior court rulings or arbitration decisions where his testimony was rejected or his objectivity as a witness had been questioned.
[14] The trial judge then put to counsel for Ms. Bruff-McArthur that there remained the issue of whether Dr. Bail had sufficient professional objectivity to provide independent evidence and he asked her if she wished to cross-examine Dr. Bail on this issue as part of a voir dire. Counsel declined that offer and elected instead to cross-examine Dr. Bail on the issue as part of her cross-examination in the trial proper.
[15] The trial judge then proceeded to rule that Dr. Bail could not testify on certain sections of his report. The relevant sections were primarily where Dr. Bail was critical of the reliability of the conclusions reached by other doctors examining Ms. Bruff-McArthur. The trial judge also made clear that he did not want Dr. Bail testifying about Ms. Bruff-McArthur's credibility.
[16] Dr. Bail testified in-chief that his methodology was not to review any of a subject's medical records before meeting with them. Consistent with this methodology, after the examination of Ms. Bruff-McArthur, which took just over an hour, Dr. Bail spent ten to 12 hours reviewing her medical records, looking for discrepancies between what she told him in the meeting and what was in the records. These discrepancies formed the largest portion of his report.
[17] In summary, Dr. Bail testified that in his opinion: Ms. Bruff-McArthur did not develop any psychiatric disorders or limitations as a result of the accident; required no psychotherapy or psychotropic medication in relation to the accident; her pre-accident psychiatric profile was not exacerbated by the accident; and she did not require housekeeping or attendant care as a result of any psychiatric condition.
(3) The Verdict
[18] Dr. Bail was the last witness to testify at trial. After closing submissions, the trial judge gave his charge to the jury. The charge was previously subject to a pre-charge conference and it was provided to the parties in advance of being presented to the juries. No objection was made to the charge and no special instruction regarding Dr. Bail's testimony was requested.
[19] As part of his charge, the trial judge reviewed very briefly Dr. Bail's testimony. He did not instruct the jury regarding the duty of expert witnesses. Nor did he raise any concerns with respect to the substance of Dr. Bail's testimony or his independence.
[20] After the jury retired to consider their verdict, defence counsel brought a threshold motion, arguing that Ms. Bruff-McArthur did not meet the threshold in s. 267.5(12) of the Insurance Act, R.S.O. 1990, c. I.8 of suffering a permanent serious impairment of an important physical, mental or psychological function.
[21] Following completion of the motion, the jury returned with a verdict assessing general damages at $23,500 and rejecting all other heads of damages, including special damages, future care costs and past and future income loss.
(4) Threshold Motion Ruling
[22] Approximately one month later, the trial judge released his reasons on the threshold motion: see Bruff-Murphy (Litigation guardian of) v. Gunawardena, 2016 ONSC 7, [2016] O.J. No. 6. He concluded that Ms. Bruff-McArthur's claim for general damages met the threshold in s. 267.5(12) of the Insurance Act. In reaching that conclusion, the trial judge analyzed the evidence adduced at trial, including the evidence of Dr. Bail.
[23] I note parenthetically the respondent's submission that this court should not rely on the ruling on the threshold motion because it was released after the jury's verdict. In what follows, I will only reference the ruling on the threshold motion to gain insight into the trial judge's concerns with Dr. Bail's testimony and to understand the trial judge's reasons for permitting Dr. Bail to testify. When it comes to determining whether the trial judge's concerns were justified or misplaced, I will conduct my own review of the record.
[24] In his ruling, the trial judge stated, at para. 58, that during the trial he permitted Dr. Bail to testify because of the "very high threshold before a court may exclude expert testimony for bias established by the Supreme Court in White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23, [2015] 2 S.C.R. 182, at paras. 48-49".
[25] The trial judge's analysis of Dr. Bail's evidence was highly critical and included the following observations:
The vast portion of his testimony in-chief consisted of Dr. Bail telling the jury about prior medical notations and how they contradict what Ms. Bruff-McArthur allegedly told him in his interview (para. 68).
The only semi-psychiatric element of Dr. Bail's report was entitled "Mental Status Examination", which consumed one half of a page of the 20-page report (para. 69).
In order to be fair and objective, Dr. Bail should have asked the plaintiff why her verbal reporting of her prior medical condition was so vastly different from her prior medical records. Dr. Bail could not do that because his methodology in conducting independent medical examinations was to not read such medical records before the interview (para. 70).
Dr. Bail testified that he discarded any notes he may have made during his interview of Ms. Bruff-McArthur as to what she allegedly told him. His only record of her comments was contained in his report dictated after he interviewed Ms. Bruff-McArthur and after his subsequent lengthy review of her medical records (para. 73).
Dr. Bail was making up evidence as he testified to support his conclusions adverse to Ms. Bruff-McArthur (para. 108).
The vast majority of Dr. Bail's report and testimony in-chief was not of a psychiatric nature but was presented under the guise of expert medical testimony and the common presumption that a member of the medical profession will be objective and tell the truth (para. 122).
[26] The trial judge found that Dr. Bail was not a credible witness and did not honour his obligation and written undertaking to be fair, objective and non-partisan pursuant to rule 4.1.01 of the Rules of Civil Procedure, R.R.O. 1990, Reg. 194. He summarized Dr. Bail's evidence as follows, at paras. 123-25:
The vast majority of Dr. Bail's testimony to the jury amounted to nothing other than the following:
(a) The plaintiff did not tell me the truth in my interview;
(b) Here are all the instances I found in my 10 to 12 hour review of her medical records which prove that she did not tell me the truth;
(c) If I as a psychiatrist cannot believe her; how can you?
The primary purpose of R. 4.1.01 is to prohibit and prevent such testimony in the guise of an expert. Dr. Bail undertook and thereby promised to not do what he did in front of this jury.
I will not qualify witnesses as experts in the future whose reports present an approach similar to that of Dr. Bail in this case.
C. Issues
[27] This appeal raises the following issues:
Did the trial judge err in not permitting Ms. Bruff-McArthur to cross-examine Dr. Bail on prior court and arbitral findings made against him?
Did the trial judge err in qualifying Dr. Bail as an expert and/or in not intervening or taking steps to exclude Dr. Bail's testimony?
Did the respondent violate the rule in Browne v. Dunn?
[28] As I will discuss in the analysis section of my reasons, I have concluded that the trial judge did not err in ruling that Dr. Bail could not be cross-examined regarding prior court and arbitral findings made against him. However, the trial judge did err in permitting Dr. Bail to testify and in failing to exclude in whole or in part Dr. Bail's testimony and, consequently, a new trial is required. Given this finding, it is unnecessary to consider the Browne v. Dunn argument.
D. Analysis
(1) The Scope of Dr. Bail's Cross-Examination
[29] Counsel for Ms. Bruff-McArthur sought to cross-examine Dr. Bail on three previous comments regarding his testimony in other cases, indicating that he had:
become an advocate for the party calling him as a witness, which is not the role of an expert: see Morrison v. Greig, [2007] O.J. No. 225, 46 C.C.L.T. (3d) 212, at paras. 47-48;
appropriated the role of advocate of the insurer rather than an impartial witness, took a partisan approach and focused on inconsistencies in the information given by claimant, such that his credibility was seriously weakened and should be disregarded: see Gabremichael v. Zurich Insurance Co., [1999] O.F.S.C.I.D. No. 198, at paras. 31-33; and
presented as a notably partisan witness: see Sohi v. ING Insurance Co. of Canada, [2004] O.F.S.C.D. No. 106, at para. 38.
[30] Ms. Bruff-McArthur submits that the trial judge erred in denying her the right to cross-examine Dr. Bail on these findings because the trial judge failed to draw a distinction between prior comments rejecting the evidence of the witness and prior findings of discreditable conduct, namely, the failure of Dr. Bail to abide by his oath as an expert.
[31] I do not accept this argument. In my view, the prior comments made about Dr. Bail do not amount to a finding of discreditable conduct. Rather, they are the opinions of a judge and two arbitrators regarding the reliability of his testimony in particular cases. This is analogous to the situation in Ghorvei, where a witness' credibility had been attacked in previous proceedings. Charron J.A. (as she then was) held, at para. 31, that those credibility findings from the previous proceedings were not proper material for cross-examination:
In my view, it is not proper to cross-examine a witness on the fact that his or her testimony has been rejected or disbelieved in a prior case. That fact, in and of itself, does not constitute discreditable conduct. I do not think it would be useful to allow cross-examination of a witness on what is, in essence, no more than an opinion on the credibility of unrelated testimony given by this witness in the context of another case. The triers of fact who would witness this cross-examination would not be able to assess the value of that opinion and the effect, if any, on the witness's credibility without also being provided with the factual foundation for the opinion.
See, also, R. v. Boyne, 2012 SKCA 124, [2012] S.J. No. 795, at paras. 48-51.
[32] In the present case, the comments of the judge and arbitrators about Dr. Bail's testimony in the previous cases would have been of no assistance to the jury without an understanding of their factual foundation. That necessary context would only have served to divert the jury from the task at hand and convert the trial into an inquiry regarding the reliability of Dr. Bail's testimony in the three other proceedings. Thus, in my view, the trial judge did not err in prohibiting this line of cross-examination.
(2) The Trial Judge's Gatekeeper Role with Respect to Expert Opinion Evidence
(1) Qualification Stage
[33] Ms. Bruff-McArthur submits that the trial judge should have exercised his gatekeeper function to exclude Dr. Bail from testifying on the grounds that his methodology was unfair; he was biased; he was engaged in an exercise to destroy her credibility; and his prospective evidence would amount to a violation of the rule in Browne v. Dunn. In the alternative, she argues that the trial judge erred in not instructing the jury that they should disregard Dr. Bail's testimony.
[34] In White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23, [2015] 2 S.C.R. 182, a decision released shortly before the judgment under appeal, the Supreme Court of Canada provided clarity and guidance regarding challenges to experts on the basis of bias and lack of independence. Cromwell J., writing for the court, stated, at para. 19, that the basic structure for the law relating to the admissibility of expert evidence has two main components.
[35] The first component requires the court to consider the four traditional "threshold requirements" for the admissibility of the evidence established in R. v. Mohan, [1994] 2 S.C.R. 9: (i) relevance; (ii) necessity in assisting the trier of fact; (iii) absence of an exclusionary rule; and (iv) the need for the expert to be properly qualified.
[36] The second component is a "discretionary gatekeeping step" where "the judge balances the potential risks and benefits of admitting the evidence in order to decide whether the potential benefits justify the risks": para. 24. It is a cost-benefit analysis under which the court must determine whether the expert evidence should be admitted because its probative value outweighs its prejudicial effect.
[37] The analysis under the second component is best thought of as a specific application of the court's general residual discretion to exclude evidence whose prejudicial effect exceeds its probative value: R. v. Bingley, 2017 SCC 12, [2017] 1 S.C.R. 170, at para. 16. As Charron J.A. wrote in R. v. K. (A.), 45 O.R. (3d) 641, at para. 76:
The balancing process which lies at the core of the determination of the admissibility of this kind of evidence is not unique to expert opinion evidence. It essentially underlies all our rules of evidence.
In White Burgess, Cromwell J. referenced Mohan and made the same point, at paras. 19 and 20:
Mohan also underlined the important role of trial judges in assessing whether otherwise admissible expert evidence should be excluded because its probative value was overborne by its prejudicial effect -- a residual discretion to exclude evidence based on a cost-benefit analysis: p. 21.
The reasons in Mohan engaged in a cost-benefit analysis with respect to particular elements of the four threshold requirements, but they also noted that the cost-benefit analysis could be an aspect of exercising the overall discretion to exclude evidence whose probative value does not justify its admission in light of its potentially prejudicial effects: p. 21.
[38] Cromwell J. further explained that lack of independence or impartiality on the part of an expert witness goes to the admissibility of the witness' testimony, not just to its weight: para. 40. Specifically, in the governing framework for admissibility, the court should consider an expert's potential bias when determining whether the expert is properly qualified at the initial threshold inquiry: para. 53.
[39] However, he added that bias should also be considered when the court exercises its gatekeeping exclusionary discretion, writing, at para. 54:
Finding that expert evidence meets the basic threshold does not end the inquiry. Consistent with the structure of the analysis developed following Mohan which I have discussed earlier, the judge must still take concerns about the expert's independence and impartiality into account in weighing the evidence at the gatekeeping stage. At this point, relevance, necessity, reliability and absence of bias can helpfully be seen as part of a sliding scale where a basic level must first be achieved in order to meet the admissibility threshold and thereafter continue to play a role in weighing the overall competing considerations in admitting the evidence. At the end of the day, the judge must be satisfied that the potential helpfulness of the evidence is not outweighed by the risk of the dangers materializing that are associated with expert evidence.
In the overview of his discussion of the admissibility of expert opinion evidence, he instructed, at para. 34, that:
A proposed expert's independence and impartiality go to admissibility and not simply to weight and there is a threshold admissibility requirement in relation to this duty. Once that threshold is met, remaining concerns about the expert's compliance with his or her duty should be considered as part of the overall cost-benefit analysis which the judge conducts to carry out his or her gatekeeping role.
[40] In the present case, the trial judge cited White Burgess and appears to have relied upon Cromwell J.'s statement that in the threshold inquiry it would be quite rare for a proposed expert's evidence to be ruled inadmissible. As Cromwell J. noted, at para. 49, all that needs to be established at that stage is whether the expert is "able and willing to carry out his or her primary duty to the court". The trial judge concluded that Dr. Bail met this rather low threshold requirement.
[41] That was a discretionary decision, which is entitled to deference from this court: R. v. Shafia, 2016 ONCA 812, [2016] O.J. No. 5627, at para. 248. Another judge might well have concluded that Dr. Bail failed to meet even this low threshold test. I do not need to decide whether the trial judge erred on this point, however, because he clearly erred in principle in failing to proceed to the next step of the analysis -- consideration of the cost-benefit analysis in Dr. Bail's testimony. The trial judge did not reference this second component of his discretionary gatekeeper role. To the contrary, he appears to have believed that he was obliged to qualify Dr. Bail once he concluded that the witness met the initial Mohan threshold. There is, therefore, no decision to defer to and it falls to this court to conduct the second part of the analysis.
[42] In my view, on a proper balancing, the potential risks of admitting Dr. Bail's evidence far outweighed the potential benefit of the testimony. It was evident from a review of Dr. Bail's report that there was a high probability that he would prove to be a troublesome expert witness, one who was intent on advocating for the defence and unwilling to properly fulfill his duties to the court.
[43] The first red flag was Dr. Bail's methodology. There is a real risk of unfairness in engaging in a hunt for discrepancies between what a plaintiff says during a short interview and what medical records dating back several years reveal. This unfairness is exacerbated when the expert denies the plaintiff the opportunity to explain the apparent discrepancies. As anyone with the slightest experience with litigation would attest to, oftentimes what appears to be an inconsistency in witness' evidence is not an inconsistency at all. Oftentimes all that is required is a simple explanation to resolve what appears to be a conflict in what a witness said on two different occasions. Ms. Bruff-McArthur was not given an opportunity to offer such an explanation.
[44] A related concern is that the vast bulk of the content in Dr. Bail's report was the recitation of perceived inconsistencies between what Ms. Bruff-McArthur said in the independent medical examination and what the medical records revealed. In conducting that analysis, Dr. Bail was not bringing to bear any medical expertise. This was work that is routinely done by trial lawyers and law students or clerks in preparation for a cross-examination. Thus, the benefit of the evidence was very low, while the potential mischief was very high, especially given that none of these inconsistencies were put to Ms. Bruff-McArthur.
[45] It was also clear from the report that Dr. Bail was coming dangerously close to usurping the role of the jury in assessing Ms. Bruff-McArthur's credibility. In the "Summary and Conclusions" section of his report, he opines:
It is my opinion that if Ms. Bruff-McArthur was being forthright, this pattern of discrepancies and inconsistencies should not exist. I am therefore of the opinion that Ms. Bruff-McArthur has not been forthright with respect to her accident related claims and her provided medical and psychological history, and that the history which she has been providing over time since the accident cannot be relied upon. It is evident that Ms. Bruff-McArthur has serious credibility issues regarding her accident related claims.
In the penultimate paragraph of his report, he states: "lack of reliability, credibility and validity are factors in this case".
[46] Next, the whole tone of the report was a reliable predictor of Dr. Bail's testimony. He goes out of his way to make points that are meant to damage Ms. Bruff-McArthur's case. For example, he opines on the views of several physicians who examined Ms. Bruff-McArthur, concluding that she misled them. Dr. Bail speculates that one of her therapists may have been improperly holding herself out as a qualified psychologist. He criticizes a psychiatrist who treated Ms. Bruff-McArthur, Dr. Arora, because they discussed "personal family things, such as her daughters' potty training and her son's school problems" when "psychotherapy was requested and paid solely in relation to treating accident related claims". Dr. Bail notes that Ms. Bruff-McArthur and Dr. Arora discussed the notions of karma and reincarnation. He chastises Dr. Arora for introducing personal religious beliefs in a therapy session. I note that there is no evidence that these topics reflect Dr. Arora's personal beliefs.
[47] I could go on with further examples, but the point is that in his report Dr. Bail goes beyond a mere lack of independence and appears to have adopted the role of advocate for the defence. Given the paucity of psychiatric analysis in the report versus the high degree of potential prejudice in wrongly swaying the jury, a cost-benefit analysis would have invariably lead to the conclusion that Dr. Bail should have been excluded from testifying.
[48] To be fair to the trial judge, he attempted to ameliorate these concerns by specifically instructing the witness not to testify regarding certain issues, such as his criticism of other doctors. However, as the trial judge essentially acknowledged in his threshold motion ruling, had he undertaken the cost-benefit analysis he would not have permitted Dr. Bail to testify.
(2) During the Expert's Testimony
[49] As we know, the trial judge permitted Dr. Bail to testify and determined that Dr. Bail crossed the line of acceptable expert evidence. In order to analyze his response to this situation, it is first necessary to consider whether the trial judge's concerns regarding Dr. Bail's testimony were well founded. Assuming that they were, the next issue is what the trial judge should have done in the circumstances.
(1) Did Dr. Bail's Testimony Indicate Lack of Impartiality?
[50] I have had the opportunity to consider in detail Dr. Bail's evidence and I concur with the trial judge that it is most troubling. For present purposes, it is unnecessary to recount his testimony in full. Instead, I will focus on some of the more concerning aspects of his testimony.
[51] First, I repeat my concern regarding his methodology. It was fundamentally unfair to Ms. Bruff-McArthur not to give her an opportunity to explain the alleged inconsistencies in the information she provided. As mentioned above, there is a real concern that Dr. Bail was usurping the role of the trier of fact in determining the issue of Ms. Bruff-McArthur's credibility. Despite that concern, I am willing to acknowledge that in a case such as this, where the existence and extent of the alleged injuries are not easily determined, consideration of the plaintiff's veracity is a necessary part of an independent medical examination. However, if Dr. Bail were serious about probing this issue, he would not have adopted this methodology. He would have reviewed the inconsistencies with Ms. Bruff-McArthur.
[52] Second, and equally troubling, is that to the extent that Dr. Bail referred to the scientific testing conducted, he torqued the results so that they produced results that supported his conclusion. For example, he testified that Ms. Bruff-McArthur was administered a test where she was instructed to count backwards from 100 by sevens. He noted that she provided a few incorrect answers in her count. Dr. Bail considered this to be an inconsistency because she was able to get some of the count right but also made mistakes. For Dr. Bail, inconsistencies meant that the subject was not being truthful about her condition.
[53] Dr. Bail then testified that in cases where a subject mathematically "just doesn't have it together", he asks them to recite the months of the year in reverse order. Apparently, Ms. Bruff-McArthur did very well on this test, answering correctly and quickly. Dr. Bail testified that this result was also an inconsistency because she did so well on that test and so poorly on the sevens test. So, despite the fact that Dr. Bail testified that he administers the month test as a check for those who are not mathematically inclined, he calls into question her credibility for doing well on the month test and faring poorly on the sevens test.
[54] Dr. Bail went on to administer another mathematical test, requiring her to calculate how many $1.50 magazines could be purchased with $10. Ms. Bruff-McArthur did not do well on this test and Dr. Bail considered this to be an inconsistency. The other logical conclusion, that Ms. Bruff-McArthur was consistently weak in performing math exercises, seems not to have crossed his mind.
[55] In short, the tests were deliberately interpreted to fit a theory of mendacity. Unless she got every question on every test correct, she was inconsistent and, in Dr. Bail's opinion, inconsistency equated to an untruthful subject.
[56] A third concern relates to a subtle point that demonstrates Dr. Bail's fundamental misconception of his role. He questioned Ms. Bruff-McArthur regarding her physical limitations. It is, of course, perfectly appropriate for a psychiatrist conducting an independent medical examination to ask questions about a subject's physical injuries and resultant limitations. That information could provide useful context for the examination. However, Dr. Bail was quite open about the fact that he asked the questions for an entirely different purpose. He testified that he asked about physical limitations so that he could compare those answers to any future surveillance evidence he may receive. This is consistent with how Dr. Bail regarded the purpose of his review of the medical records. There is a troubling pattern that suggests that he understands his primary role to be to expose inconsistencies and not to provide a truly independent assessment of Ms. Bruff-McArthur's psychiatric condition.
[57] Fourth, when Dr. Bail was cross-examined about his emphasis on perceived inconsistencies, he denied ignoring those parts of the medial records that did not fit his diagnosis. He explained their absence from his report on the basis that "you can't put everything in your report". Later in his cross-examination, Dr. Bail stated: "I'm interested in the things that don't corroborate, not the things that do corroborate." Again, this testimony makes plain Dr. Bail's lack of awareness of the need to be impartial as an expert witness.
[58] Before turning to what the trial judge should have done in face of this testimony, I wish to correct one of his findings. The trial judge stated in his reasons on the threshold motion that Dr. Bail did not have any notes of his examination of Ms. Bruff-McArthur. Based on this observation, he concluded that Dr. Bail was making his testimony up as he went along to support his position.
[59] That is not accurate. Dr. Bail did have notes. Indeed, the trial judge ruled that he could refer to them as he testified. It is not a fair conclusion that Dr. Bail was making up his testimony. Having reviewed his evidence carefully, I am of the view that there is no basis to conclude that Dr. Bail was anything but truthful in his testimony. I have concerns regarding Dr. Bail's independence and his methodology; I do not have any concerns about his veracity.
(2) What Should the Trial Judge Have Done in This Case?
[60] Under the White Burgess framework, and in most other leading cases on the admissibility of expert evidence, the issue of admissibility is decided at the time the evidence is proffered and the expert witness' qualification is requested by a party. To the extent that this is possible, it should be the norm: R. v. J. (J.-L.), 2000 SCC 51, [2000] 2 S.C.R. 600, at para. 28.
[61] In the present case, however, the trial judge appears to have assumed that, once Dr. Bail was qualified as an expert, his gatekeeper role was at an end. The trial judge erred in law in reaching that conclusion.
[62] A trial judge in a civil jury case qualifying an expert has a difficult task. She must make a decision based on an expert report that will, in most cases, never be seen by the jury. While the report provides a roadmap of the anticipated testimony and specific limits may be placed on certain areas of testimony, the trial judge obviously cannot predict with certainty the nature or content of the expert's testimony.
[63] Where, as here, the expert's eventual testimony removes any doubt about her independence, the trial judge must not act as if she were functus. The trial judge must continue to exercise her gatekeeper function. After all, the concerns about the impact of a non-independent expert witness on the jury have not been eliminated. To the contrary, they have come to fruition. At that stage, when the trial judge recognizes the acute risk to trial fairness, she must take action.
[64] Charron J.A. made this point in K. (A.), writing as follows, at para. 73:
In some cases it may be possible to rule on the admissibility of the proposed evidence on the basis of counsel's submissions alone. However it may at times prove necessary to hold a voir dire in order to properly consider all relevant factors. Where the trial is before a jury and the question of admissibility cannot be clearly determined in a summary fashion, it may indeed be prudent to scrutinize the evidence during the course of a voir dire before admitting it. While in some cases the ruling can be made early in the proceedings, in other cases, it may be only later in the trial that the value of the proposed evidence can be properly assessed. For example, in this case, it was only after the main Crown witnesses had testified and the defence strategy became apparent that the determination of the admissibility of the expert evidence could properly be made.
And in a later decision in R. v. Ranger, 67 O.R. (3d) 1, at para. 63, Charron J.A. stated:
[The dangers of expert opinion evidence] must be considered in the balancing process that forms part of the test for admissibility. Further, the trial judge's gatekeeper function does not end with the ruling on admissibility. The expert evidence must be carefully constrained in its presentation with a view to minimizing the associated dangers so that, in the end result, the judge is still satisfied that the probative value of the evidence exceeds its prejudicial effect and is properly admissible.
[65] As mentioned above, the cost-benefit analysis under the second component of the framework for admitting expert evidence is a specific application of the court's general residual discretion to exclude evidence whose prejudicial effect is greater than its probative value. This general residual discretion is always available to the court, not just when determining whether to admit an item of evidence, but after the admission stage if the evidence's prejudicial effect is only revealed in the course of its presentation to the trier of fact.
[66] An instructive discussion is found in R. v. White, 2011 SCC 13, [2011] 1 S.C.R. 433, a case that dealt with the admissibility of post-offence conduct in criminal matters. A majority of the Supreme Court stated, at para. 50:
Otherwise admissible evidence may still be removed from consideration by the jury on the basis that it is more prejudicial than probative. This may be achieved by refusing to admit the evidence at trial. It can also happen that the disproportionately prejudicial nature of a certain item of evidence only becomes apparent in light of the evidence as a whole. The trial judge may then instruct the jury in his charge that they may not consider a certain item of evidence in their deliberations.
The discussion from White makes clear that the court's residual discretion to exclude prejudicial evidence is an ongoing one that continues throughout a trial. It may be invoked if prejudice manifests after initially admitting the evidence. Thus, because the second component of the framework for admitting expert evidence is an application of this residual discretion, the court has residual discretion to exclude expert evidence even after admitting it, if later in the trial prejudice emerges that was not apparent at the time of admission.
[67] Given this ongoing gatekeeper discretion, the question remains of what, as a practical matter, the trial judge could or should have done in this case. His first option would have been to advise counsel that he was going to give either a mid-trial or final instruction that Dr. Bail's testimony would be excluded in whole or in part from the evidence. Had he taken that route, he would have received submissions from counsel in the absence of the jury and proceeded as he saw fit. Alternately, he could have asked for submissions from counsel on a mistrial, again in the absence of the jury, and ruled accordingly. In the event that he had to interrupt Dr. Bail's testimony mid-trial, he would have had to consider carefully how best to minimize the potential prejudicial effect of the interruption from the respondent's perspective.
[68] The point is that the trial judge was not powerless and should have taken action. The dangers of admitting expert evidence suggest a need for a trial judge to exercise prudence in excluding the testimony of an expert who lacks impartiality before those dangers manifest.
[69] I am mindful that counsel for Ms. Bruff-McArthur did not seek an instruction regarding Dr. Bail's evidence. The law is generally that the failure to object to a civil jury charge is fatal to a request for a retrial on appeal based on misdirection or non-direction. However, this rule is subject to the exception that where the misdirection or non-direction resulted in a substantial wrong or miscarriage of justice, it may warrant a new trial: Pietkiewicz v. Sault Ste. Marie District Roman Catholic Separate School Board, 71 O.R. (3d) 803, at paras. 22-28; and Briscoe Estate v. Canadian Premiere Life Insurance Co., 2012 ONCA 854, 113 O.R. (3d) 161, at paras. 70-71. In my view, the admission of Dr. Bail's testimony resulted in a miscarriage of justice.
[70] I would go further and state that, given the importance of a trial judge's ongoing gatekeeper role, the absence of an objection or the lack of a request for a specific instruction does not impair a trial judge's ability to exercise her residual discretion to exclude evidence whose probative value is outweighed by its prejudicial effect.
[71] The respondent submits that even if this court concludes that Dr. Bail's testimony should have been excluded, there is no basis to order a new trial because he was just one of many witnesses and his testimony likely did not have a significant impact on the jury's verdict.
[72] It is impossible to gauge with any certainty the impact of Dr. Bail's testimony. The fact that he was one of only two witnesses to testify for the defence suggests that his testimony may well have been an important factor in the jury's analysis of the case. In any event, a focus on the inability to measure the precise prejudice caused by the testimony misses the point entirely, which is that there has been a miscarriage of justice in this case. This court has a responsibility to protect the integrity of the justice system. This is not a "no harm, no foul" situation. No doubt, another trial will be costly and time consuming, but it is necessary because the defence proffered the evidence of a wholly unsuitable expert witness.
E. Disposition
[73] I would grant the appeal, set aside the judgment below and order a new trial. I would award the appellants their costs of the appeal in the amount of $22,000, inclusive of fees, disbursements and applicable taxes.
[74] The parties may make written submissions on the issue of the costs of the first trial.
Appeal allowed.
Notes
1 On appeal, the appellants are represented by different counsel than the counsel they were represented by at trial.
