COURT OF APPEAL FOR ONTARIO DATE: 20240506 DOCKET: COA-22-CV-0387

Lauwers, van Rensburg and Thorburn JJ.A.

BETWEEN

Shoba Kolapully Plaintiff (Respondent)

and

Lynda Myles and Toronto Transit Commission Defendants (Appellants)

Counsel: Chad Townsend and Max Luburic, for the appellants Mireille Dahab and Avneet Kaur, for the respondent Joseph Obagi and Elizabeth Quigley, for the intervener Ontario Trial Lawyers Association James Tausendfreund and David Zuber, for the intervener Canadian Defence Lawyers

Heard: January 30, 2024

On appeal from the judgment of Justice P. Tamara Sugunasiri of the Superior Court of Justice, sitting with a jury, dated December 1, 2022, and from the costs endorsement dated July 17, 2023.

Lauwers J.A.:

A. Overview

[1] The respondent, Shoba Kolapully, was struck by a Toronto Transit Commission (“TTC”) bus while crossing the street at Ellesmere Road and Nielson Road in Toronto on March 6, 2012. The appellant, Lynda Myles, was driving the bus and is a TTC employee. Ms. Kolapully has sought relief in a number of forums over a lengthy period of time. All of her claims have been vigorously resisted by the TTC on its own behalf and for Ms. Myles. It appears that the TTC considers Ms. Kolapully to be a malingerer and considers all the decisions to the contrary in the long litigation litany to be mistaken in whole or in part, including the jury verdict, the trial motions, and the costs decisions under appeal.

[2] On December 3, 2013, Ms. Kolapully started this action for general and specific damages claiming that she suffered serious and permanent physical and psychological impairment.

[3] Ms. Kolapully also initiated a proceeding before the Licence Appeal Tribunal (“LAT”). In January 2018, the LAT found she was catastrophically impaired according to the Statutory Accident Benefits Schedule (the “SABS”): Kolapully v. Toronto Transit Commission. The TTC applied for reconsideration, which was denied: Kolapully v. Toronto Transit Commission. Ms. Kolapully applied for arbitration at the Financial Services Commission of Ontario. On March 9, 2018, Arbitrator Barrington found she was entitled to non-earner benefits under the SABS: [2018] O.F.S.C.D. No. 45.

[4] After a six-week trial in 2022, the jury awarded Ms. Kolapully $175,000 in non-pecuniary damages and $200,000 in damages for past loss of income. The jury apportioned the degree of negligence at 25% to Ms. Kolapully and 75% to the appellants. The trial judge determined that the amount Ms. Kolapully had received for non-earner benefits under the SABS – about $95,000 – was not to be deducted from the damages award for past income loss under s. 267.8 of the Insurance Act, R.S.O. 1990, c. I.8.

[5] The TTC appeals on the grounds set out below. For the reasons that follow, I would allow the TTC’s appeal in part.

B. Issues on Appeal

[6] The TTC takes two lines of attack. First, the TTC submits that the trial judge erred in three ways that give rise to the need for a new trial: 1) she erred in allowing into evidence an untried medical test, the Single Photon Emission Computed Tomography (“SPECT”) scan of Ms. Kolapully; 2) she erred in excluding evidence of attendant care claims forms Ms. Kolapully signed in blank as part of her alleged participation in a fraudulent scheme; and 3) she erred in her articulation of the test for causation, particularly in her jury instructions and in her instructions on the calculation of damages.

[7] The second line of the TTC’s attack is to argue that in her disposition after the jury verdict, the trial judge erred in refusing to deduct non-earner benefits from the tort award under s. 267.8 of the Insurance Act, and she erred in her costs award.

[8] I address each issue in turn.

C. Analysis

[9] The TTC argues that a new trial is required because the trial judge made three errors that undermine the justice of the jury’s verdict.

(1) Did the trial judge err in admitting the results of the Single Photon Emission Computed Tomography (“SPECT”) Scan?

[10] The trial judge ruled that a clinical neurologist, Dr. Manu Mehdiratta, would be permitted to testify about a SPECT scan of Ms. Kolapully as one of the tools that he used to support his diagnosis that she had suffered a mild traumatic brain injury in the accident. The trial judge found that the scan was properly used as a diagnostic aid and was not novel science: Kolapully v. Myles, 2022 ONSC 6674.

[11] The trial judge made an additional ruling on the admissibility of the report prepared by the doctor of nuclear medicine who performed the SPECT scan, Dr. Hui-Yin Siow, who was not available to testify. The report included a summary written by Dr. Siow. At issue was whether the SPECT scan report could be admitted under s. 35 of the Ontario Evidence Act, S.O. 1990, c. E.23 or as a common law exception to s. 52 of that Act. The trial judge ultimately admitted Dr. Siow’s report, on which Dr. Mehdiratta was permitted to rely, but ordered the summary portion to be fully redacted.

[12] Although the admissibility of the SPECT scan was challenged at trial, the TTC proffered no expert evidence to counter its admissibility. The TTC’s medical expert Dr. Robert Yufe was unfamiliar with the use of SPECT scans. The case presents as one in which Dr. Mehdiratta’s evidence was uncontested beyond cross-examination and legal argument.

(a) The Governing Principles on the Admission of Expert Evidence

[13] There are two steps in the test to admit expert evidence. The first is the threshold requirement, and the second engages the judge’s discretionary, gatekeeper function: R. v. Mohan, [1994] 2 S.C.R. 9. The threshold requirement has four elements for admissibility: the evidence must be relevant; it must be necessary in assisting the trier of fact; no other evidentiary rule should apply to exclude it; and the expert must be properly qualified. At the second, gatekeeping step, the trial judge must weigh the potential risks and benefits of admitting the evidence in order to decide whether the potential benefits justify the potential harm to the trial process that might flow from the admission of the expert evidence: White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23, [2015] 2 S.C.R. 182, at para. 24; Bruff-Murphy v. Gunawardena, 2017 ONCA 502, 7 C.P.C. (8th) 1, at paras. 35-36.

[14] The reliability of expert evidence that draws on novel or contested science, or science being used for a novel purpose, has drawn significant judicial comment.

[15] In R. v. J.(J.-L.), 2000 SCC 51, [2000] 2 S.C.R. 600, following Mohan, Binnie J. rejected the “general acceptance” test formulated in the United States in Frye v. United States, 293 F. 1013 (U.S. D.C. Ct. App., 1923), opting instead to follow the “reliable foundation” test laid down by the U.S. Supreme Court in Daubert v. Merrell Dow Pharmaceuticals Inc., 509 U.S. 579 (U.S. Cal., 1993). At para. 33, Binnie J. endorsed the factors that the U.S. Supreme Court set out at pp. 593-94 of Daubert for evaluating the reliability of novel science, which have become the criteria for reliability:

(1) whether the theory or technique can be and has been tested:

Scientific methodology today is based on generating hypotheses and testing them to see if they can be falsified; indeed, this methodology is what distinguishes science from other fields of human inquiry.

(2) whether the theory or technique has been subjected to peer review and publication:

[S]ubmission to the scrutiny of the scientific community is a component of “good science,” in part because it increases the likelihood that substantive flaws in methodology will be detected.

(3) the known or potential rate of error or the existence of standards; and,

(4) whether the theory or technique used has been generally accepted:

A “reliability assessment does not require, although it does permit, explicit identification of a relevant scientific community and an express determination of a particular degree of acceptance within that community.”

Widespread acceptance can be an important factor in ruling particular evidence admissible, and “a known technique which has been able to attract only minimal support within the community,” ... may properly be viewed with skepticism.

[16] Daubert holds that the reliability of expert opinion premised on novel or contested science depends on the reliability of the underlying scientific methodology. As Alan Gold put it:

[C]redibility is the product of methodology. Methodology — the logic of research design, measures, and procedures — is the engine that generates knowledge that is real. Observations (or a hunch or a guess) is a starting point in suggesting areas of inquiry, but the whole point of real science is to test observations (or hunches or guesses) systematically. The focus must be solely on principles and methodology, not on the conclusions that they generate. [1]

[17] In his celebrated Report on the Inquiry into Pediatric Forensic Pathology in Ontario, [2] Justice Stephen T. Goudge, after reviewing the relevant case law, listed a number of factors that can be taken as a form of checklist, at p. 495:

1. the reliability of the witness, including whether the witness is testifying outside his or her expertise;

2. the reliability of the scientific theory or technique on which the opinion draws, including whether it is generally accepted and whether there are meaningful peer review, professional standards, and quality assurance processes;

3. whether the expert can relate his or her particular opinion in the case to a theory or technique that has been or can be tested, including substitutes for testing that are tailored to the particular discipline;

4. whether there is serious dispute or uncertainty about the science and, if so, whether the trier of fact will be reliably informed about the existence of that dispute or uncertainty;

5. whether the expert has adequately considered alternative explanations or interpretation of the data and whether the underlying evidence is available for others to challenge the expert’s interpretation;

6. whether the language that the expert proposes to use to express his or her conclusions is appropriate, given the degree of controversy or certainty in the underlying science; and

7. whether the expert can express the opinion in a manner such that the trier of fact will be able to reach an independent opinion as to the reliability of the expert’s opinion.

[18] Justice Goudge’s factors have been referred to in several Superior Court cases: see e.g., R. v. St. Amand, 2014 ONCJ 800 and R. v. Munoz Hernandez, 2013 ONSC 4257.

[19] In Mole and Mole v. Manwell, 2017 ONSC 3357, the expert proposed to give an opinion based on an admittedly novel surgical technique. Conlan J. considered the Goudge factors, noting that the evidence at hand did not satisfy many of them: at paras. 74-89. However, he stated that these factors “were not meant to be exhaustive”, and considered three other factors: First, the evidence is “comprehensible”; second, “it is not overly complex”; and third, the evidence “is not to be assessed in a vacuum”: at paras. 81-83. Conlan J. concluded that, although it was a close call, the evidence in the case met the reliability threshold.

[20] I would adopt the Goudge factors as a useful framework for trial judges to use in assessing the reliability of expert opinions based on novel or contested science. However, scientific techniques that are commonly used and reliable enough for clinical purposes might nonetheless be treated as novel when used for forensic or evidentiary purposes. As Binnie J. wrote in J.(J.-L.), at para. 35:

While the techniques [the expert] employed are not novel, he is using them for a novel purpose. A level of reliability that is quite useful in therapy because it yields some information about a course of treatment is not necessarily sufficiently reliable to be used in a court of law to identify or exclude the accused as a potential perpetrator of an offence.

[21] The lists of various factors in J.(J.-L.) and in the Goudge Report are not exhaustive, nor do they constitute mandatory codes, nor is there any need for them to be listed in a decision in rote form. That said, trial judges should consider these factors and may consider other relevant factors in the exercise of their discretion as evidentiary gatekeepers.

(b) The Principles Applied

[22] For reasons that are not clear, as noted, the TTC chose not to call an expert witness conversant in SPECT scan analysis or the underpinning technology. The TTC’s expert, Dr. Robert Yufe, admitted his lack of expertise and knowledge during cross-examination, leading the trial judge to find that his evidence should be given little weight: Kolapully v. Myles, 2022 ONSC 6674, at para. 3. This would have affected both the threshold decision on admissibility and the weight the jury would have given to the SPECT scan evidence.

[23] The trial judge was clearly alive to her gatekeeping role and the applicable legal principles. She ultimately admitted Dr. Siow’s report into evidence and permitted Dr. Mehdiratta to speak to it but ordered the summary portion to be fully redacted. The trial judge cited and distinguished Meade v. Hussein, 2021 ONSC 7850, in which Bale J. declined to admit similar evidence, finding, at para. 41 of that case in which Dr. Siow was also a proposed expert: “that Dr. Siow's brain SPECT evidence is inadmissible as it fails to satisfy the reliable foundation test for novel scientific evidence.” Bale J. alluded to the governing principles and the cases in his decision.

[24] The trial judge distinguished Meade on several bases, most particularly on the academic support Dr. Mehdiratta provided for the use of SPECT scans as a diagnostic tool to confirm or to add information to a provisional diagnosis. She noted, at para. 6:

Meade involved a nuclear medicine specialist diagnosing a TBI [traumatic brain injury] based solely on the SPECT scan. Had that been the request here, I would have agreed with the TTC and not permitted it. Indeed in this case I only permitted the scan to be shown to the jury with the nuclear medicine specialist’s opinions redacted, so as to not unduly influence the jury with his untested opinions and the unsupported use of SPECT scans as a singular diagnostic tool for TBIs.

[25] The trial judge found that “the proposed SPECT scan evidence as a diagnostic aid for a clinical neurologist such as Dr. Mehdiratta is not novel science”: at para. 5. She added: “While there may be disagreement on its accuracy, Dr. Mehdiratta was clear in his evidence that he had already made a TBI diagnosis based on this clinical evaluation before sending Ms. Kolapully for a scan to obtain any further information on abnormalities it might offer.” The trial judge also cited several cases in which courts had accepted SPECT scan evidence as a diagnostic tool: see e.g., Legree v. Origlieri, 2021 ONSC 7650; Marcoccia v. Gill; and Hornick v. Kochinsky.

[26] This is a discretionary decision by the trial judge that was adequately explained. It is plain from her reasons that the trial judge kept in mind the governing principles for the admission of contested science and the need to balance the prejudicial effect and probative value of the evidence. The TTC has not demonstrated any extricable error of law or any palpable and overriding error of fact in that exercise of discretion, and this court must defer. However, I should not be interpreted as finding that SPECT scans are reliable in detecting traumatic brain injury; the admissibility and use of the evidence in this case was determined by the trial judge applying the proper principles.

[27] On appeal, the TTC advances the result in two more recent decisions of the British Columbia Supreme Court: Gutfriend v. Case, 2022 BCSC 2055 and Bolduc v. Stratton, 2022 BCSC 1168. In Gutfriend, Thomas J. reviewed the studies and guidelines provided by the parties and concluded, at para. 97: “SPECT remains a very sensitive test for TBI.” He then added: “Unfortunately, the specificity [ability to distinguish TBI from other conditions] is not sufficient to provide legally reliable information for diagnostic purposes in a setting where the issue is to distinguish brain injury from other psychiatric disorders and/or conditions.”

[28] The TTC also relied on Gutfriend for a comment by the trial judge questioning Dr. Mehdiratta’s impartiality, at para. 74, to which counsel drew the panel’s attention:

I was troubled that there was no conflict of interest disclosure requirements associated with the CANM Guidelines. Dr. Cohen has provided at least 200 medical-legal reports for Dr. Mehdiratta using SPECT for identifying TBI over the last three years, a period which spans the creation and publication of the guidelines. Dr. Cohen has also invested close to $3,000,000 in SPECT technology at his hospital. In my view, these are interests that should have been noted in a disclosure statement.

[29] In Bolduc, Iyer J. excluded the evidence at the necessity stage, finding that the plaintiff had already been diagnosed with a mild traumatic brain injury based on clinical examinations before the SPECT scan was performed, and neither her prognosis nor her treatment plan rested on the results of the SPECT. He concluded that “admitting the SPECT evidence in these circumstances would trigger the concerns expressed in the jurisprudence about excessive deference to experts and add expense and time to trials.”

[30] I would make three observations about these more recent cases. First, the science appears to still be contestable and each case will turn on the evidence considered by the trial judge. A party must put its best foot forward. Here the TTC did not do so. Second, the cases turn on their own facts and evidence. Later cases that address a common factual or evidentiary issue are of no assistance on appeal. Third, the TTC’s reliance on para. 74 of Gutfriend was an improper attempt to influence this court against Dr. Mehdiratta. The TTC was entitled to challenge the impartiality of the CANM guidelines at Dr. Mehdiratta’s voir dire, or to attempt to bring a motion to adduce fresh evidence on appeal. It did not do so. Without the benefit of the evidence that the trial judge in Gutfriend relied on, this court has no basis to assess, let alone accept, the concern that he raised. [3]

(2) Did the trial judge err in refusing to allow the respondents to introduce the attendant care claims forms into evidence?

[31] The TTC sought to introduce into evidence attendant care forms signed in blank by Ms. Kolapully. It argued that the forms were both substantively relevant to her claim of disability, and would support its allegation that she was not credible because she was involved in a fraudulent scheme to recover monies for attendant care that she did not receive. The forms were blank when she signed them and were completed and submitted by others.

[32] Ms. Kolapully’s counsel moved to exclude the attendant care forms. The trial judge found the forms to be irrelevant and excluded them. Her ruling was brief:

The Defendants seek to introduce Attendant Care forms signed by the Plaintiff and part of what the TTC alleges is a fraudulent scheme to recover monies for care that was never received. The Plaintiff is not claiming attendant care in this action. Further, she signed blank forms and was not aware of, nor part of any alleged fraudulent scheme. The TTC has pursued its fraud theory in other legal actions in which the Plaintiff is not named. There is no relevance to these documents other than to raise suspicion in the minds of the jurors that Ms. Kolapully is not trustworthy. This is unduly prejudicial to Ms. Kollapuly. The defence will have ample opportunity to test Ms. Kolapully’s evidence and credibility on relevant issues during the course of this trial.

[33] On appeal, the appellant renews its arguments that the attendant care forms were both relevant to a fact in issue – namely, Ms. Kolapully’s level of functioning – and relevant for “questioning on a pure credibility basis” which is “an exception to the collateral fact rule.” The appellant also takes issue with the trial judge’s jury instruction on credibility. I find that the trial judge made no error in her conclusions on either basis for admissibility and that her jury charge was appropriate. I would therefore dismiss this ground of appeal.

(i) The attendant care forms were not relevant to a fact in issue

[34] The TTC complains that Ms. Kolapully’s counsel at trial wavered on whether she was advancing the claim for attendant care benefits, but argues that the forms were relevant to the claim of disability either way.

[35] For the purposes of this appeal, what is important is whether the attendant care claim was ultimately put to the jury. Whatever frustration may have been caused by the elusiveness of Ms. Kolapully’s trial counsel on this issue has no bearing on whether the attendant forms were admissible as relevant to a fact in issue. Had the claim been advanced, the way the blank forms were completed and processed would have been relevant. But the claim was not pursued and did not go to the jury.

[36] There was also no error in the trial judge’s conclusion that the forms were not relevant generally to Ms. Kolapully’s disability claim. The forms were not filled out by Ms. Kolapully. The TTC inconsistently argues that the claims are relevant to credibility as evidence of a fraud, and that they disclose reliable facts about Ms. Kolapully’s condition.

(ii) The trial judge did not err in excluding the attendant care forms for questioning on a pure credibility basis

[37] Having concluded that the attendant care forms were not substantively relevant to a fact in issue, the next question is whether they were admissible for the purpose of attacking Ms. Kolapully’s credibility.

[38] According to Justice Paciocco and his co-authors, there is an unresolved debate in the law between the view that credibility and reliability are not collateral where they matter in a case, on the one hand, and on the other hand, that credibility and reliability are always only collateral. [4] The Paciocco text takes the position that the “prevailing view is that credibility and reliability are collateral”, citing R. v. Krause (1986), 54 C.R. (3d) 294 (S.C.C.), at p. 301, per McIntyre J.: a matter is collateral where it is “not determinative of an issue arising in the pleadings or indictment or not relevant to the matters which must be proved for the determination of the case.”

[39] Previous versions of the Paciocco text advocated for a modification of the collateral facts rule “so that it reflects the principled approach that has come to be accepted in almost all areas of the law of evidence: if the contradiction is probative enough to outweigh the prejudice it may cause, admit it.” [5] The text cited this court’s decision in R. v. F.(C)., 2017 ONCA 480, 349 C.C.C. (3d) 521, at para. 58.

[40] In my view, the trial judge’s decision not to allow the attendant care forms to be put to Ms. Kolapully and explored further in her evidence was a reasonable exercise of her discretion under the collateral facts rule. I say this for four reasons.

[41] First, the evidence about how the forms were signed by Ms. Kolapully and then used by someone in a fraud would have diverted the jury down a difficult and time-consuming forensic rabbit hole quite far from the task before it. This was precisely the mischief that the collateral facts rule was designed to prevent.

[42] Second, the trial judge was well placed to assess the balance of probative value and prejudicial effect in view of her awareness of the other evidence that was to be adduced.

[43] Third, this evaluation was fair and accurate, since the TTC’s case included a frontal attack on Ms. Kolapully’s credibility on the substantive evidence before the jury. The trial judge’s statement that the “defence will have ample opportunity to test Ms. Kolapully’s evidence and credibility on relevant issues during the course of this trial” was amply borne out.

[44] Fourth, the exclusion of the attendant care forms did not deprive the TTC of its line of attack on Ms. Kolapully’s credibility or leave the jury with a distorted perception. Counsel for the TTC’s jury address had Ms. Kolapully’s lack of credibility as its central theme. Early in his address, he stated: “Clearly much of the plaintiff’s evidence is simply wrong. She admits that she lied to three assessors about having hired help.” He pointed out her inconsistencies and used terms such as “untrustworthy”, “willing to embellish”, and noted her “exaggeration”, all in the execution of what he called her “carefully constructed narrative.” The jury had plainly before it the TTC’s invitation to find against Ms. Kolapully on the basis of her credibility and many instances in the substantive evidence on which they might base such an assessment.

(iii) The trial judge’s jury instruction on credibility was appropriate

[45] The appellant adds this complaint: while the trial judge agreed to instruct the jury that a “a falsehood or lie is always serious”, she omitted the words “or lie” in her reading of the charge. The written charge included the standard instruction: “Discrepancies on trivial details may be unimportant; however, a falsehood or lie is always serious.” In her reading of the charge, though, the trial judge stated: “And so discrepancies on trivial details may be unimportant; but if it is a falsehood then that is something you should always take seriously.” I would make three observations. First, I see no material difference between a lie and a falsehood. Second, the jury had a copy of the written charge, which appears to have included the word “lie” even if the trial judge did not read it. Third, it does not appear that the appellant asked for a correcting instruction. This omission is inconsequential.

[46] The jury was well equipped to make the credibility findings and knew its task. The trial judge did not err in the exercise of her discretion in refusing to admit the blank attendant care forms into evidence.

(3) Did the trial judge err in her jury instructions on causation and the calculation of damages?

[47] The appellants argue that the trial judge made two errors in her jury instructions:

a) The trial judge did not provide an adequate instruction correcting Ms. Kolapully’s counsel’s error in referring to “material contribution” as the applicable test for causation; and,

(b) The trial judge did not provide an adequate instruction correcting Ms. Kolapully’s counsel’s closing statement that the calculation of non-pecuniary damages should be done on a per diem basis.

[48] I would dismiss this ground of appeal. The trial judge included appropriate correcting instructions in relation to both sets of comments in her jury charge.

(i) The “material contribution” error in the test for causation

[49] In her closing address to the jury, Ms. Kolapully’s counsel erroneously stated that the proper test for causation is the “material contribution” test, and that a material contribution is anything more than a trivial cause:

And the law says, if the accident materially contributed to her losses and damages — again, it’s not the sole cause. You may find it is, but I’m saying that that threshold is not that high. It materially contributed to her losses, and the compensation is at a hundred per cent, a hundred per cent for her pain and suffering and losses. Better — Her Honour will explain the law a lot better than I will. And what is considered material? It’s a material contributing factor. Is anything more than trivial; that’s what material is. Anything more than trivial.

[50] Counsel for the TTC rightly objected, and the trial judge agreed that a correcting instruction was warranted. She consulted TTC’s counsel, who initially requested that the trial judge mention Ms. Kolapully’s counsel by name in the correcting instruction to avoid any confusion on the part of the jury. The trial judge determined that singling out counsel was unnecessary and potentially prejudicial. TTC’s counsel did not press the issue further and consented to the instruction that was ultimately included in the jury charge:

You should disregard any submissions made by counsel that the test is whether the accident materially caused the injuries with the word “material” understood as more than trivial. This is not the test and I ask that you strike it from any notes that you may have made.

The jury charge continued on to comprehensively explain the “but-for” test and its application to the facts.

[51] On appeal, the TTC argues that the charge was inadequate to repair the damage done and should have mentioned counsel’s name. I would not accept this argument. The charge directly addressed and corrected the erroneous statements that were put to the jury and clearly set out the proper law on causation. The fact that the TTC’s counsel accepted the proposed correction and did not request a re-charge reinforces its adequacy.

(ii) The per diem calculation of non-pecuniary damages

[52] Ms. Kolapully’s counsel concluded her closing remarks by suggesting to the jury, incorrectly, that they ought to calculate non-pecuniary damages on a per diem basis:

And I’ll quickly tell you, those pain-and suffering numbers, nobody seems to understand them, but a senior lawyer told me, you have to look at it how many days does — from the time of the accident to the day of the trial, and how many more days does she have to live? And you look, how much money would you give this individual a day to compensate them for everything that they’re looking through each day? And the ranges that I gave you, basically, it compensates her anywhere between twenty to thirty dollars a day. And that’s how he taught me to understand how pain and suffering damages work. We’re just paying a compensation per day for this individual who has to undergo these impairments and injuries in function.

[53] Counsel for the TTC objected, and the trial judge included a correction in the jury charge:

First, you heard in closing submissions that some counsel conceptualize damages for pain and suffering by thinking about what amount should be awarded to the plaintiff per day for the rest of her life. Counsel noted that with the range she suggests it amounts to $20 to $30 per day. And let me be clear, calculating damages for pain and suffering is not mathematical and you should not reduce your assessment to such a calculation.

[54] The TTC argues that the correction was inadequate. The trial judge’s correction clearly identified the offending comments, explained why they were incorrect, and instructed the jury not to follow them. There is no merit to the TTC’s complaint.

(4) Are non-earner benefits deductible from an award for loss of income under s. 267.8 of the Insurance Act?

[55] Before trial, Ms. Kolapully received more than $95,000 in non-earner benefit statutory benefit payments as the result of earlier proceedings before the Financial Services Commission. Section 267.8 of the Insurance Act mandates that, in an action for loss or damage from bodily injury in an automobile accident, the trial judge must reduce the pecuniary damages tort award by the amount received by the plaintiff in corresponding statutory accident benefits. The TTC moved to have Ms. Kolapully’s non-earner benefits deducted from the $200,000 damages award for past income loss that she received at trial. The trial judge dismissed the TTC’s motion.

[56] Relying on this court’s decision of Walker v. Ritchie (2005), 197 O.A.C. 81 (C.A.), at paras. 84, 87, the trial judge found that non-earner benefits are not related to loss of income, and are therefore not deductible from a tort award for loss of income under s. 267.8(1). The TTC argues that this conclusion conflicts with this court’s decision in Cadieux v. Cloutier, 2018 ONCA 903, 143 O.R. (3d) 545, at para. 12, which lists non-earner benefits in the category of “income replacement benefits.” If the trial judge had given effect to this inclusion in Cadieux, the non-earner benefits would have been deducted from the jury award for past income loss. However, because Cadieux was not a case about accounting for non-earner benefits, the trial judge found that the court’s inclusion of these benefits in the category of income replacement benefits was obiter and not binding on her. She did not deduct the non-earner benefits from the pecuniary damages award for income loss. This was an error, as I will explain.

(a) The Governing Principles

[57] Before addressing the particulars of this issue, I set out some of the context and the governing principles. The system of compensation for injuries suffered in motor vehicle accidents in Ontario is a hybrid of no-fault insurance coverage and traditional tort law. In exchange for limits on the ability to sue for injuries, insured parties have access to no-fault benefits: Meyer v. Bright (1993), 15 O.R. (3d) 129 (C.A.), at para. 6. The statutory scheme aims at full compensation for an injured party whose injuries exceed the statutory threshold, like Ms. Kolapully. Such injured parties can sue for damages.

[58] The statutory scheme also aims to prevent double compensation. This is a bedrock governing principle that has been repeated throughout the case law, from Brown v. Bouwkamp (1976), 12 O.R., (2d) 33, at para. 10, to Cadieux, at para. 17, where the court stated:

These two forms of compensation – SABs and tort damages – are independent of one another. It is inevitable, however, that there will be overlap between the compensation provided to an accident victim by no-fault SABs and the award of damages to that person in a civil tort action. Section 267.8 of the Insurance Act contains provisions designed to address this overlap and to prevent double recovery. It reflects the principle that victims should be fairly compensated, but not over-compensated. Automobile insurers, who provide first-party benefits through SABs insurance, should not be required, when wearing their fault-based liability insurer hats, to compensate an accident victim twice for the same losses.

[59] To determine the proper interpretation of the Act as it relates to the deductibility of non-earner benefits, it is necessary to review the history of the legislation and how its evolution sought to prevent double recovery. For accidents that occurred between 1971 and October 1989, a provision of the Insurance Act, R.S.O. 1980, c. 218, which originated in S.O. 1971, c. 84, s. 17, applied. Section s. 239(2) provided:

(2) Where a claimant is entitled to the benefit of insurance as provided in [the relevant Schedule], this, to the extent of payments made or available to the claimant thereunder, constitutes a release by the claimant against the person liable to the claimant or his insurer.

[60] For accidents occurring between October 1989 and January 1994, a revised provision applied in the Insurance Act, R.S.O. 1990, c. I.8, which originated in S.O. 1990, c. 2, s. 57. Subsection 267(1) provided:

(1) The damages awarded to a person in a proceeding for loss or damage arising directly or indirectly from the use or operation of an automobile shall be reduced by:

(a) all payments that the person has received or that were or are available for no-fault benefits and by the present value of any no-fault benefits to which the person is entitled;

[61] In Bannon v. McNeely, [1998] 38 O.R. (3d) 659 (C.A.), dealing with an accident that occurred in 1991, this court interpreted the 1989-1994 statutory scheme, and determined that the proper method for deducting statutory accident benefits from tort damages was the “apples to apples” or “strict matching” approach. Under that approach, “apples should be deducted from apples, and oranges from oranges”, meaning statutory accident benefits were only deductible based on a “precise matching of individual benefits within those against the identical heads of damages”: Bannon, at p. 679; Cadieux, at para. 39.

[62] In 1996, the legislature amended the statutory scheme to prescribe three categories of statutory accident benefits, each to be deducted from a corresponding category of tort damages:

Income loss and loss of earning capacity

267.8 (1) In an action for loss or damage from bodily injury or death arising directly or indirectly from the use or operation of an automobile, the damages to which a plaintiff is entitled for income loss and loss of earning capacity shall be reduced by the following amounts:

1. All payments in respect of the incident that the plaintiff has received or that were available before the trial of the action for statutory accident benefits in respect of the income loss and loss of earning capacity.

1996, c. 21, s. 29.

Health care expenses

(4) In an action for loss or damage from bodily injury or death arising directly or indirectly from the use or operation of an automobile, the damages to which a plaintiff is entitled for expenses that have been incurred or will be incurred for health care shall be reduced by the following amounts:

1. All payments in respect of the incident that the plaintiff has received or that were available before the trial of the action for statutory accident benefits in respect of the expenses for health care.

Other pecuniary loss

(6) In an action for loss or damage from bodily injury or death arising directly or indirectly from the use or operation of an automobile, the damages to which a plaintiff is entitled for pecuniary loss, other than the damages for income loss or loss of earning capacity and the damages for expenses that have been incurred or will be incurred for health care, shall be reduced by all payments in respect of the incident that the plaintiff has received or that were available before the trial of the action for statutory accident benefits in respect of pecuniary loss, other than income loss, loss of earning capacity and expenses for health care. 1996, c. 21, s. 29.

[63] Two lines of case law emerged after the introduction of the new statutory language in 1996. The first line continued using the “apples to apples” approach formulated in Bannon and applied it to cases governed by the new provision: see e.g., Gilbert v. South (2015), 127 O.R. (3d) 526, 2015 ONCA 712. In Walker v. Ritchie (2005), [2005] 197 O.A.C. 81 (C.A.), this court also applied the “apples to apples” approach in concluding that non-earner benefits are not akin to loss of income and therefore are not deductible from damages awards for loss of income.

[64] The second line of cases rejected the “apples to apples” approach, opting instead for a more flexible “silo” approach: see e.g., Basandra v. Sforza (2016), 2016 ONCA 251, 130 O.R. (3d) 466 (C.A.), at para. 21, and El-Khodr v. Lackie (2017), 2017 ONCA 716, 139 O.R. (3d) 659, at paras. 35, 60, 70, and 71. Under the silo approach, all statutory accident benefits falling within the same broad category are deductible from all damages awards in the corresponding broad category. This line of cases followed the Supreme Court’s decision in Gurniak v. Nordquist, 2003 SCC 59, [2003] 2 S.C.R. 652, which rejected the use of the strict matching approach by the British Columbia Court of Appeal dealing with similar statutory language. MacFarland J.A observed the emergence of this approach in El-Khodr, at para. 70: “As appears in this court’s decision in Basandra, courts are moving towards a more relaxed approach that considers whether the pre-trial benefit received generally fits within one of the broad statutory categories of damages.” The case law shows that the effect of this new approach was to make statutory accident benefits more frequently deductible from tort damages awards.

[65] In Cadieux, a five-judge panel of this court resolved the conflict in favour of the silo approach: at paras. 4, 12-14, 39, 56, and 57-120. They observed:

[T]he statutory provisions in force when Bannon was decided did not group benefits to be deducted into categories or silos. Instead, it lumped all accident benefits together and required their deduction from the tort award. In contrast, the current statute sets out broad categories of SABs from which the tort award must be deducted.

[66] In light of the statutory changes and the Supreme Court’s rejection of the strict matching approach in Gurniak, the Cadieux court held that the proper interpretation of the current version of s. 267.8 “does not support matching at a more particular level than the three silos of income loss, health care expenses and other pecuniary loss”: at para. 85. The silos are described at paras. 12-14:

There are three broad categories of SABs under the Insurance Act and the Statutory Accident Benefits Schedule, O. Reg. 34/10. These were referred to in El-Khodr as silos. The first category provides income replacement benefits or, if the person was not employed at the time of the accident, “non-earner” benefits, or “caregiver benefits”, if they provided caregiver services to another person at the time of the accident. [Emphasis added.]

The second category is health care benefits. “Health care” is a defined term in s. 224(1) of the Insurance Act. It “includes all goods and services for which payment is provided by the medical, rehabilitation and attendant care benefits provided for in the Statutory Accident Benefits Schedule” …

The third category of benefits addresses “other pecuniary loss”, which includes lost educational benefits, expenses of visitors and housekeeping, and home maintenance expenses.

Note the underlined words, which put non-earner benefits in the silo for income replacement benefits.

[67] The key principle is that the deduction from the tort pecuniary damages award is confined to the apposite silo. If the claimant gets no damages award for future income loss, for example, the amount the claimant received for income loss under the SABS is not deducted from any award under the other two silos – “health care benefits” or “other pecuniary loss”. An additional principle is that statutory accident benefits are never to be deducted from general damages for pain and suffering: Insurance Act, s. 267.8(7).

(b) The Trial Judge’s Ruling

[68] The trial judge considered herself bound by the result in Walker and, as noted, found that the inclusion of non-earner benefits in the income loss silo in para. 12 of Cadieux was obiter. In reaching that result, she compared the SABS in place at the time Walker was decided to the SABS in place at the time of the accident and this court’s decision in Cadieux. The purpose of this exercise was to identify whether any interim changes to the SABS could justify a departure from Walker.

[69] The trial judge concluded that provisions in the current regime were “ambiguous at best” as to whether non-earner benefits were to be considered “income replacement benefits”; she found that if the legislature had wanted non-earner benefits to be considered to be income replacement benefits, the legislation would have used express language to so provide. The trial judge considered the interplay between s. 35 of the SABS, which permits a recipient of non-earner benefits to elect instead to receive an income replacement benefit or a caregiver benefit, and s. 12(4)(c), which relieves the insurer of the duty to pay the non-earner benefit if the recipient has elected under s. 35 to receive income-replacement or caregiver benefits instead. In her view, these provisions demonstrate the distinction between non-earner benefits and income replacement benefits, and their mutual exclusivity. She held that the amount received by Ms. Kolapully for non-earner benefits was not deductible from the tort damages award for past income loss.

(c) The Principles Applied

[70] As noted, the jury awarded Ms. Kolapully $200,000 in damages for her past loss of income, which is in the income benefit silo. The trial judge did not deduct the amount that Ms. Kolapully had received for non-earner benefits under the SABS, about $95,000, from this award. This was an error.

[71] My reasoning proceeds in two steps. First, I conclude that the result in Walker was overruled by the court in Cadieux and was therefore not binding on the trial judge. Next, I find that the inclusion of non-earner benefits in the income loss silo in para. 12 of Cadieux was authoritative and not obiter, was the correct interpretation of the provision’s current text, and was therefore binding on the trial judge.

(i) The result in Walker was not binding on the trial judge

[72] The decision in Walker rested on the now defunct approach to deductibility from Bannon. In para. 79 of Walker, the court cited the trial judge’s decision in that case, specifically his quotation from Bannon, that: “any no fault benefit deducted from a tort award under s. 267(1)(a) must be deducted from the head of damages or type of loss akin to that for which the no fault benefits were intended to compensate ... if at all possible, apples should be deducted from apples, and oranges from oranges ...” : Walker v. Ritchie, [2003] O.J. No. 5596 at para. 42. Following the “apples to apples” approach, the court concluded that non-earner benefits are not “in any way related to loss of income”: at para. 84.

[73] In Cadieux, this court demolished the underpinnings on which Walker rested. In line with the 1996 statutory changes noted earlier, the Cadieux court replaced the “apples to apples” approach with the silo approach, which groups specific statutory accident benefits and heads of damages together under three general categories. The conclusion in Walker that non-earner benefits did not relate to the income loss head of damages does not imply that non-earner benefits would not come within the income replacement benefits silo, which is broader by design. The shift from the “apples to apples” methodology to the silo methodology in Cadieux had the effect of displacing the result in Walker and inviting a new assessment of the deductibility of non-earner benefits.

[74] Although the trial judge considered the many amendments made to the SABS between the Walker decision and the Cadieux decision, which I return to below, she did not consider the important changes to s. 267.8 of the Insurance Act. The legislature’s introduction of silos into the text of s. 267.8 provides critical context for the interpretive exercise, and the trial judge’s failure to consider this context led her into error.

[75] Further, the Walker court found that non-earner benefits were not only unrelated to income loss but were in fact most akin to general damages, but this reasoning was in error, was rejected by the Cadieux court, and should not be followed. As the trial judge in Walker noted, “the new statutory provisions specifically prohibit a deduction of statutory benefits amounts from non-pecuniary damages.” There is, with respect, no support for the proposition that non-earner benefits are a surrogate for general damages. Indeed, such a characterization of non-earner benefits as general damages is inconsistent with the statutory scheme, under which a claimant does not get access to general damages unless the significance of his or her injuries exceeds the statutory threshold. Indeed, such a reading would leave double recovery in place for those who got non-earner benefits and also got general damages in tort.

(ii) The inclusion of non-earner benefits in the income loss silo in Cadieux was binding on the trial judge

[76] The trial judge’s decision fails to pay due respect to the nature and purpose of this court’s decision in Cadieux. The court’s intention was plainly to reset on firm and predictable grounds the way in which trial courts and lawyers are to approach the intersection of the no-fault benefits regime under the Insurance Act and tort damages. I would infer this intention from the tone and text of Cadieux and the fact that it was written by a panel of five judges as a decision “by the Court”, purposively in order to overcome a longstanding tendency among trial judges “to apply the strict matching approach based on Bannon, in spite of that approach having been rejected by the Supreme Court in Gurniak”: at para. 61.

[77] In my view, the trial judge erred in characterizing as mere obiter the court’s inclusion of non-earner benefits in the silo of income replacement benefits which are to be deducted from a tort award for loss of income. She noted correctly that Cadieux made no reference to Walker. However, the court accepted the inclusion of non-earner benefits in that silo as proposed in the appellant’s factum in Cadieux. Doing so could not be characterized as a thoughtless oversight.

[78] Further, the Cadieux court’s treatment of non-earner benefits aligns with the intended effect of the legislation and the silos in recapturing SABS. All of the statutory accident benefits are required to be accounted for in those cases in which there is a tort damages award for pecuniary losses. It would be odd indeed if the only statutory accident benefits to be excluded from any silo were the non-earner benefits. There would appear to be no other benefits that would be treated as similarly exceptional.

[79] I noted above the trial judge’s view that s. 35 of the SABS, which permits a recipient of non-earner benefits to elect instead to receive an income replacement benefit or a caregiver benefit, demonstrates the distinction between non-earner benefits and income replacement benefits, and their mutual exclusivity. With respect, I would take the opposite view. Non-earner benefits belong in the income replacement silo precisely because the legislature has signaled that they are an alternative to other benefits in that silo. Far from revealing dissimilarity, the substitutability of the other benefits for non-earner benefits reinforces the view that they are interrelated and therefore belong in the same income replacement silo. [6]

[80] Finally, the trial judge referred to but sidestepped important statutory language changes that explain the differing results in Walker and Cadieux. The trial judge said, at paras. 7 and 8 of the non-earner benefits deductibility decision:

The earlier version of SABS allows the insurer to deduct those amounts that a plaintiff has received or is eligible to receive for loss of income at subsection 12(4) from non-earner benefits. In the newer version, subsections 12(2) and 12(3) deduct “all other income replacement assistance” from the non-earner benefits. While the concept of deducting payments for loss of income remains the same in both regimes, the use of the word “other” in subsections 12(3) (“less the total of all other income replacement assistance, if any”) imply that non-earner benefits also relate to income replacement. On its face, this is a compelling argument. However, I find that the inclusion of one word leaves the treatment of non-earner benefits ambiguous at best and does not have the express language one would expect to see if the Legislature intended to group non-earner benefits with other income replacement benefits.

The 2012 regulation also adds subsection 12(4)(c), which sets out that the insurer is not required to pay the non-earner benefit if the recipient has elected under s. 35 to receive an income replacement benefit or a caregiver benefit. This provision, while new, captures a long-standing principle of mutual exclusivity of non-earner benefits and income replacement benefits. [Emphasis added.]

[81] These language changes support the conclusion that non-earner benefits are in the silo of income replacement benefits and are therefore deductible. The trial judge simply repeated the result-selective approach to statutory construction taken in earlier trial decisions that this court criticized in Cadieux. Although she purported to accept the reversal of the matching principle in Bannon, the trial judge allowed the concept to slip in by the back door on the basis that there was no express inclusion of non-earner benefits in the statutory language. That is not how the silos are designed to operate according to Cadieux.

[82] The trial judge was required to deduct the amount that Ms. Kolapully had received for non-earner benefits under the SABS, about $95,000, from the pecuniary damages award.

(5) Did the trial judge make errors in her costs award?

[83] The trial judge awarded Ms. Kolapully all legal fees on a partial indemnity scale, including her costs in the action in the total amount of $654,993.20 (HST included); her legal costs for the LAT proceeding in the amount of $64,359.15 (HST included); and disbursements for both the action and LAT hearing in the total amount of $215,055.72. For the following reasons, I would dismiss TTC’s motion for leave to appeal the costs award.

[84] The governing test is that leave is exceptional and will not be granted “save in obvious cases where the party seeking leave convinces the court there are ‘strong grounds upon which the appellate court could find that the judge erred in exercising his discretion’”: Brad-Jay Investments Ltd. v. Szijjarto, [2006] O.J. No. 5078 (C.A), at para. 21, leave to appeal refused, [2007] S.C.C.A. No. 92. None of the TTC’s three arguments provide such strong grounds.

[85] First, the TTC argues that the trial judge failed to consider the factors listed in r. 57.01 of the Rules of Civil Procedure, R.R.O. 1990, Reg. 194 that would have warranted a reduction of the costs award. Instead, she held that the TTC’s conduct, specifically the “fraud lens” through which it defended the case, disentitled it to any reduction. The TTC submits that this reasoning resulted in “a form of elevated costs” despite no formal finding of “reprehensible conduct”, which is itself an error in principle: Young v. Young, [1993] 4 S.C.R. 3, at p. 134.

[86] This argument has no merit. Contrary to the TTC’s characterization, the trial judge did not order elevated costs. She ordered partial indemnity costs, as is standard: Akagi v. Synergy Group (2000) Inc., 2015 ONCA 771, 128 O.R. 64, at para. 31. The trial judge expressly considered the relevant r. 57.01 factors but held that they did not call for a reduction in the costs award. She was entitled to exercise her discretion not to reduce the award, in part because of the impact of the “fraud lens” on the length and complexity of the trial.

[87] Second, the TTC argues that the trial judge made a palpable and overriding error in finding that the TTC viewed the case through a “fraud lens”. There was ample basis to find that the TTC took that view. The TTC’s counsel cross-examined Ms. Kolapully on her allegation that her former lawyer told her to lie in making claims for attendant care benefits; moreover, in costs submissions before the trial judge, the TTC’s counsel persisted in asserting that the trial was needed to stop Ms. Kolapully’s “attempt to defraud the self-insured TTC.”

[88] Third, the TTC argues that the trial judge erred in awarding Ms. Kolapully her costs of the LAT hearing because there was no deduction of her statutory accident benefits from the damages award against the TTC. This argument misconstrues the prerequisite for awarding costs of the LAT hearing: what is required is a reduction of damages flowing from the LAT disposition, not a deduction. It was open to the trial judge under s. 131(1) of the Courts of Justice Act, R.S.O. 1990, c. C.43 to award “costs of and incidental to a proceeding”. This court stated in Cadieux, at para. 129: “Legal fees and disbursements in pursuing SABs can reasonably be considered incidental to the proceeding where the SABs have reduced the damages payable by the tortfeasor.”

[89] The trial judge made no error in concluding that Ms. Kolapully’s successful pursuit of the catastrophic impairment designation, which gave her access to funds for medical, rehabilitation, attendant care, and housekeeping costs, narrowed the scope of her tort action and reduced the quantum of damages recoverable against the TTC. The deductibility of the non-earner benefits adds to the force of this argument. It was open to the trial judge to order that the TTC pay Ms. Kolapully’s costs of the LAT hearing.

D. Disposition

[90] I would allow the TTC’s appeal in part, and require the deduction of the amount that Ms. Kolapully received for non-earner benefits under the SABS from the pecuniary damages award. Ms. Kolapully has substantially prevailed in this appeal, and I would award her $40,000 as the all-inclusive costs for the appeal as agreed.

Released: May 6, 2024 “P.D.L.” “P. Lauwers J.A.” “I agree. K van Rensburg J.A.” “I agree. Thorburn J.A.”

[1] “‘What if they made a revolution and nobody came’: Expert Evidence in Canadian Courts,” Alan D. Gold Collection of Criminal Law Articles, available on the QuickLaw database (November 20, 2016).

[2] Justice Stephen T. Goudge, Report on the Inquiry into Pediatric Forensic Pathology in Ontario (Toronto: Ontario Ministry of the Attorney General, 2008).

[3] It might be time to revisit the not-quite-blanket rule in R. v. Ghorvei (1999), 46 O.R. (3d) 63 (C.A.), per Charron J.A., at para. 31, which bars the use, in a witness’s cross-examination, of negative judicial comment in other proceedings about the witness’s credibility, reliability, bias, or misconduct. The law is set out comprehensively in R. v. Holloway, 2021 ONSC 6136, per D.E. Harris J. There are several situations in which relaxing the rule would enhance the search for the truth: the testimony of expert witnesses – the most egregious example of shielding a biased expert witness is Dr. Charles Smith, whose bad forensic evidence escaped scrutiny for years and led to many wrongful convictions, as detailed in the Goudge Report referred to above; and the testimony of police officers, as in Holloway.

[4] David M. Paciocco, Lee Stuesser & Palma Paciocco, The Law of Evidence, 8th ed. (Toronto: Irwin Law, 2020)–, at pp. 600-605.

[5] Paciocco, The Law of Evidence, at p. 604.

[6] The intervener, Ontario Trial Lawyers Association, argued at some length that placing caregiver benefits in the first silo would create anomalous effects. This case is not about caregiver benefits.