DATE: 20200715

ONTARIO

SUPERIOR COURT OF JUSTICE

BETWEEN:

LILIANA DI CIENZO

Applicant

– and –

ATTORNEY GENERAL OF ONTARIO

Respondent

Neil G. Wilson and Yolanda Song, for the Applicant

Hart Schwartz and Daniel Guttman, for the Respondent

HEARD: March 9, 10 and 11, 2020 (written submissions on remedy received April 6 and April 20, 2020)

Kimmel J.

Table of Contents

REASONS FOR JUDGMENT. 4

I. OVERVIEW... 4

II. THE APPLICANT. 5

III. THE IMPUGNED REGULATIONS AND SCOPE OF THE CONSTITUTIONAL CHALLENGE 6

1. Regulations at Issue. 6

2. Notice of Constitutional Question(s) 7

2.1) Does the Constitutional Challenge Apply Only to s. 18(3) of the Regulations or Can it be Extended to s. 18(4)?. 7

2.2) Does the Constitutional Challenge Apply to All Commercial Licence Holders covered by Sections 18(3)(a) and (b), or is it Limited to Class C drivers?. 9

2.3) The Effect of the Restrictions on the Applicant’s Constitutional Challenge. 9

IV. THE EXPERTS. 10

1. The Relevance of the Expert Evidence. 10

2. What Do the Experts Say?. 11

2.1) Dr. Arshinoff. 11

2.2) Dr. Owsley. 12

2.3) Dr. Peli 13

2.4) Dr. Smiley. 15

2.5) Chris Janusz. 16

3. The Applicant’s Challenges to the Admissibility of Two of Ontario’s Experts. 16

3.1) The Applicant asserts that Dr. Owsley is not impartial; Ontario disputes this assertion. 16

3.2) The Applicant asserts that Mr. Janusz is not impartial and not a qualified expert; Ontario disputes these assertions. 17

4. The Applicable Legal Principles Regarding Challenges to Experts. 20

4.1) The test for admissibility, generally. 20

4.2) Factoring independence and impartiality into the analysis. 20

5. Analysis of the Applicant’s Challenges to Ontario’s Experts. 21

5.1) Concerns about Dr. Owsley’s impartiality or independence. 21

5.2) Concerns About the Impartiality, Independence and Qualifications of Mr. Janusz. 21

6. The Studies Reviewed and Considered by the Experts. 23

6.1) Ontario’s Literature Reviews in 1995 and 1998. 23

The 1995 MTO Report 23

The 1998 Engel-Townsend Report 23

6.2) The Underlying Studies. 24

Motor Vehicle Collision Rate Studies. 24

The Rogers Study. 24

The US-DOT Study. 25

The Keeney Study. 26

The Laberge-Nadeau Study. 26

Driver Performance Studies. 27

6.3) The Ontario Studies: 1995, 2006 and 2017. 27

The 1995 MTO Report 28

The 2006 Review and 2017 Report Regarding the Waiver Program.. 28

7. Findings Arising From the Expert Evidence and Reviews. 29

7.1) Impacts of Monocularity. 29

Visual Field. 29

Loss of Stereopsis and Binocular Summation. 30

7.2) Driver Collision Rates vs. Driver Performance Studies. 30

Driver Collision Studies: MVC Rates. 31

Driving Performance Studies. 31

7.3) The Applicant 32

V. SECTION 15 ANALYSIS. 33

1. Preliminary Questions. 33

1.1) What is the impact of the Grismer case?. 33

1.2) Does the reasoned apprehension of harm question imbue the analysis at the s. 15 Charter breach stage as well as the section 1 stage?. 35

2. Reviewing section 15 jurisprudence: Kapp, Withler, Quebec v. A, Taypotat, and the Pay Equity cases. 36

2.1) The Jurisprudential Background. 36

2.2) Quebec v. A and Taypotat 37

2.3) The 2018 pay equity companion cases. 38

3. The Section 15 Issues and Analysis. 39

3.1) Characterizing the Applicant’s Claims. 39

Section 18(3)(a) 39

Section 18(3)(b) 39

3.2) Does s. 18 of the Regulations Establish a Distinction Based on a s. 15 Ground?. 40

3.3) Is the Distinction in the Regulations Substantively Discriminatory?. 41

The Essence of Substantive Discrimination. 41

Overview of the parties’ arguments: 41

The Applicant’s position. 42

The Respondent’s position. 43

Correspondence of the Regulations with the characteristics of the group. 44

Correspondence of the Regulations with the characteristics of the individuals. 45

Do the Regulations infringe substantive equality because the lines they draw are themselves arbitrary? 47

Correlation of the Vision Standards with Driver Safety or Performance Statistics. 47

Vision Standards Fail to Capture Other Groups for Whom the Same Safety Concerns Should Apply. 48

3.4) Conclusion: Do the Vision Standards of the Regulations Violate s. 15 of the Charter?. 48

VI. SECTION 1 ANALYSIS. 50

1. Pressing and Substantial Objective. 50

2. The Appropriate Standard of Deference: The Precautionary Principle and a “reasoned apprehension of harm” 51

3. Proportionality of the Infringing Statutory Provision. 53

3.1) Rational Connection. 53

3.2) Minimal Impairment 54

Suggested alternatives to a blanket prohibition. 55

The approach of other Canadian jurisdictions. 57

Conclusion on Minimal Impairment 59

3.3) Balancing salutary and deleterious effects. 59

4. Conclusion on s. 1 of the Charter. 60

VII. REMEDY.. 60

1. What remedy does the Applicant seek?. 60

2. Should the remedy by suspended to allow Ontario to amend the Regulations to allow for Individual Assessment?. 61

3. Can the Applicant seek an individual remedy in the case of a suspended declaration of constitutional invalidity?. 61

VIII. CONCLUSION.. 64

REASONS FOR JUDGMENT

I. OVERVIEW

[1] The Applicant, Liliana Di Cienzo, lost her right eye to cancer when she was 41 years old. She was a bus driver with Oakville Transit at the time. She has not been able to return to work as a bus driver. Her Class C commercial driver’s licence, required for her to drive a public transit bus, was revoked due to her inability to meet the requirements of s. 18(3) of the Ontario Drivers’ Licences Regulations, O. Reg. 340/94 (“the Regulations”) under the Highway Traffic Act, R.S.O. 1990, c. H.8. A person having only one eye (“monocular”) cannot meet the requirements of s. 18(3) of the Regulations. This means that Ms. Di Cienzo is not eligible to be considered for the Class C commercial driver’s licence required to drive public transit vehicles.

[2] Ms. Di Cienzo challenges s. 18(3) of the Regulations under s. 15 of the Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982 c. 11, on the grounds that it discriminates against her based on her physical disability. Section 15 of the Charter guarantees that all individuals will be treated with equality under the law. Ms. Di Cienzo claims to be a safe driver, is not restricted from driving passenger vehicles and wants the opportunity to demonstrate that her safe driving abilities extend to driving public transit vehicles. She argues that the Regulations substantively discriminate against her based on her physical disability as a monocular person and that this discrimination is not justified under s. 1 of the Charter. She seeks a declaration that this section of the Regulations is unconstitutional and of no force and effect.

[3] The Attorney General of Ontario (“Ontario”) counters that the minimum vision standard contained in the Regulations is not discriminatory on the basis of disability and does not offend s. 15 of the Charter; rather, it corresponds with a reasonable assessment of driver safety and safety requirements for drivers of commercial vehicles. Further, Ontario contends that maintaining its high standards for driving safety in the Province is a pressing and substantial objective that provides a justification under s. 1 of the Charter for the vision standards imposed by the Regulations.

[4] For the reasons that follow, I find that the vision standards contained in s. 18(3) of the Regulations substantively discriminate against the Applicant based on her physical disability. The Regulations create a blanket prohibition against the grant of Class C commercial driver’s licences to monocular persons. This violates s. 15 of the Charter. These visions standards are based on and perpetuate the stereotype that monocular drivers are not able to safely drive public transit vehicles. They create or perpetuate a discriminatory disadvantage without allowing for individual exemption based on the actual driving capabilities and characteristics of individual monocular drivers such as the Applicant.

[5] I further find that the restrictions contained in s. 18(3) of the Regulations, while rationally connected to Ontario’s pressing and substantive objective of maintaining the high safety standards of Ontario highways, are not minimally impairing in the absence of any opportunity for individuals to seek exemption based on their driving abilities such as is afforded to similarly situated individuals in other jurisdictions, including in most other Canadian provinces. In these circumstances, the salutary benefits of these regulatory restrictions do not outweigh their deleterious effects on the Applicant’s ability to earn her livelihood as a public transit bus driver. The s. 15 violation is thus not justified under s. 1 of the Charter.

[6] Section 18(3) of the Regulations is declared to be unconstitutional under s. 52 of the Constitution Act, 1982. I have accepted the request by Ontario for a temporary twelve (12) month suspension of this declaration of unconstitutionality. I have denied the Applicant’s request for an exemption from that suspension.

II. THE APPLICANT

[7] Ms. Di Cienzo was formerly employed as a bus driver with Oakville Transit. She began working for Oakville Transit in 2008. Ms. Di Cienzo had a Class C commercial licence that enabled her to drive a public transit bus. Ms. Di Cienzo enjoyed her work, and it provided her with economic stability. While working for Oakville Transit, Ms. Di Cienzo had a good driving record and was not involved in any accidents.

[8] In 2013, Ms. Di Cienzo was diagnosed with cancer in her right lacrimal tear duct gland. She was 41 years old. Treatment of the cancer required removal of Ms. Di Cienzo’s right eye. Ms. Di Cienzo underwent surgery on October 23, 2013. As a result of the surgery, Ms. Di Cienzo is now monocular. Ms. Di Cienzo continues to have full vision in her healthy left eye.

[9] After her surgery, Ms. Di Cienzo went on short-term disability to begin acclimatizing to her change in vision. She began driving again in a passenger vehicle. Around September 2014, Ms. Di Cienzo felt ready to return to work. On or around September 30, 2014, Ms. Di Cienzo received a letter from the Ministry of Transportation informing her that they had downgraded her licence from a commercial Class C licenc to a general Class G licence.[^1]

[10] As a result of the loss of her right eye, Ms. Di Cienzo no longer met the required vision standards set out in the Regulations for a Class C commercial driver’s licence. Consequently, Ms. Di Cienzo was no longer able to work as a bus driver for Oakville Transit. Ms. Di Cienzo continues to drive under a regular Class G driver’s licence. She has had no accidents since the loss of her right eye.

[11] Ms. Di Cienzo continued her employment with Oakville Transit and the Town of Oakville for a period of time in various other roles after losing her Class C licence, but she is no longer employed by them.

III. THE IMPUGNED REGULATIONS AND SCOPE OF THE CONSTITUTIONAL CHALLENGE

1. Regulations at Issue

[12] Ms. Di Cienzo challenges the constitutional validity of s. 18(3)(a) and s. 18(3)(b) of the Regulations.

[13] Section 18(3) sets the minimum standards of visual acuity and visual field required of commercial licence holders (Classes A to F[^2]) in Ontario. Visual acuity and visual field are the two metrics by which vision is measured. Acuity measures how “sharp” vision is and how clearly one sees objects at a given distance. Field measures in degrees the area over which a person can see. Functionally, the minimum standards for both acuity and field impact monocular drivers.

[14] Section 18(3)(a) requires a minimum visual acuity “as measured by Snellen Rating that is not poorer than 20/30 with both eyes open and examined together and not poorer than 20/100 in the weaker eye, with or without the aid of corrective lenses.” The visual acuity standard under s. 18(3)(a) of the Regulations prohibits monocular drivers as they must have a minimum of 20/100 vision in their weaker eye. This, of course, would be impossible for a monocular driver.

[15] For visual field, s. 18(3)(b) requires “horizontal visual field of at least 150 continuous degrees along the horizontal meridian and at least 20 continuous degrees above and below fixation, with both eyes open and examined together.”

[16] The visual field standard under s. 18(3)(b) also prohibits monocular drivers. According to the expert consensus in this case, a monocular person’s visual field might be 150 degrees or greater but there is no accepted way to accurately measure their full visual field under accepted testing conditions. Further, s. 18(4)(d) of the Regulations requires that “no less than half of the continuous degrees of the horizontal visual field that are required along the horizontal meridian shall be found on either side of the vertical meridian.” This impacts the way in which the visual field test is administered which, combined with the limitations of the accepted test methods, creates another obstacle to a monocular driver’s ability to meet visual field standards. If the vertical meridian (point of fixation) remains centered as prescribed under the accepted test methods, a monocular person is not able to achieve a measurement of 75 degrees on each side of the vertical meridian (point of fixation).

[17] The current Regulations do not provide discretionary exceptions to the vision standards required for commercial drivers’ licences. Ms. Di Cienzo seeks an opportunity to demonstrate that she can safely drive a bus even though she has a visual impairment that prevents her from meeting the vision standards in the Regulations.

2. Notice of Constitutional Question(s)

2.1) Does the Constitutional Challenge Apply Only to s. 18(3) of the Regulations or Can it be Extended to s. 18(4)?

[18] The parties agree that s. 18(3) is properly in issue, having been referenced in the Notice of Constitutional Question that was delivered on behalf of Ms. Di Cienzo. Where they part company is on the matter of s. 18(4)(d), which provides further criteria for the assessment of horizontal visual field.

[19] When issuing a Charter application, the applicant must give both the Attorney General of Canada and the Attorney General of Ontario a notice of constitutional question (Courts of Justice Act, R.S.O. 1990, c. C.43, s. 109).

[20] As part of the challenge to the 150 degree minimum visual field requirement under s. 18(3)(b), the Applicant has identified a further issue with the elaboration of the visual field requirements under s. 18(4)(d), that “no less than half of the continuous degrees of the horizontal visual field that are required along the horizontal meridian shall be found on either side of the vertical meridian.”

[21] While the parties do not agree on this point (as discussed below), the Applicant asserts that Ms. Di Cienzo, and indeed most monocular people, do have a visual field greater than 150 degrees in their healthy eye. The Applicant argues that by requiring half of that visual field on each side of the vertical meridian, s. 18(4)(d) has the effect of imposing an impossible and arbitrary barrier on monocular individuals. A monocular person’s visual field, while exceeding 150 degrees, may not be situated with 75 degrees of visual field on each side of the vertical meridian.

[22] Ontario contends that it did not receive a Notice of Constitutional Question regarding s. 18(4)(d). Therefore, while the 150 degree visual field (under s. 18(3)(b)) is properly being challenged, the requirement that the visual field be split half and half on each side (under s. 18(4)(d)) is not properly before the court. Ontario further responds that constructive notice or implied notice is not appropriate for constitutional questions.

[23] The Applicant responds in two related ways. First, that there was no confusion about what was in issue and no prejudice to Ontario. Further, the Applicant argues that she is not challenging s. 18(4)(d) in isolation. She is challenging s. 18(3)(b). To challenge a statutory provision, it must be properly interpreted. Anything in the statutory scheme which assists in interpreting s. 18(3)(b) is necessarily and properly to be considered.

[24] The requirement for notice of the specific constitutional question is mandatory and entrenched. There are but two exceptions: de facto notice and the Attorney General’s consent. See Guindon v. Canada (Attorney General), 2015 SCC 41, [2015] 3 S.C.R. 3, at para. 120. Ontario objects here to the Applicant’s late challenge to s. 18(4) of the Regulations.

[25] In Eaton v. Brant County Board of Education, 1997 366 (SCC), [1997] 1 S.C.R. 241, Sopinka J. observed that there were two “strands” of case law (at para. 49). In one strand, the absence of constitutional notice necessarily nullified a decision. In the other, the absence of notice could nullify a decision if prejudice was established. Justice Sopinka held that the absence of notice itself was prejudicial to the public interest, without specific evidence of prejudice from the Attorney General (at para. 53), although he also held that it was not necessary to “express a final opinion on these questions.” He also held, at para. 54, that there was “room for interpretation” of s. 109 of the Courts of Justice Act, and some cases might involve “de facto notice which is the equivalent of a written notice.” In understanding de facto notice, the purpose of constitutional notice is important.

[26] Earlier in Eaton, Sopinka J. reviewed the purpose of constitutional notice at para. 48:

The purpose of s. 109 is obvious. In our constitutional democracy, it is the elected representatives of the people who enact legislation. While the courts have been given the power to declare invalid laws that contravene the Charter and are not saved under s. 1, this is a power not to be exercised except after the fullest opportunity has been accorded to the government to support its validity. To strike down by default a law passed by and pursuant to the act of Parliament or the legislature would work a serious injustice not only to the elected representatives who enacted it but to the people. Moreover, in this Court, which has the ultimate responsibility of determining whether an impugned law is constitutionally infirm, it is important that in making that decision, we have the benefit of a record that is the result of thorough examination of the constitutional issues in the courts or tribunal from which the appeals arise.

[27] This understanding of constitutional notice’s purpose was followed in Guindon. At para. 19, the majority cited para. 48 and stated: “Notice requirements serve a vital purpose in ensuring that courts have a full evidentiary record before invalidating legislation and that governments are given the fullest opportunity to support the validity of legislation.”

[28] While the absence of notice can itself be prejudicial, the focus of the analysis should be on the purpose of requiring the notice and whether it has been achieved. In this case, the evidentiary record before the court is extensive as it relates to monocular commercial drivers and the Applicant’s challenge to the constitutional validity of the vision standards in the Regulations. This includes s. 18(4), which provides specificity around the measurement of the visual field that is prescribed under s. 18(3)(b). The requirement of a central vertical meridian (fixation point) and the limitations on how visual field can be measured for monocular persons has been addressed by the experts on both sides. I find that Ontario had de facto knowledge and notice of a challenge that was, in substance, directed to the requirements of both s. 18(3) and (4) of the Regulations. See Eaton, at paras. 53-54.

[29] However, the same cannot be said of the Attorney General of Canada, who did not have actual or de facto knowledge of the challenge to s. 18(4). While the Attorney General of Canada has taken no position on the s. 18(3) challenge and it is unlikely that a different position would have been taken if notice of a s. 18(4) challenge had been provided, there was no notice. Absent an exception, the Applicant’s failure to give notice to the Attorney General of Canada of a challenge to the constitutional validity of s. 18(4)(d) of the Regulations means that the Applicant cannot make a direct challenge nor to seek any remedy in respect of that section. See Paluska Jr. v. Cava (2002), 2002 41746 (ON CA), 59 O.R. (3d) 469, at paras. 22-24 (CA)

[30] I would have been prepared to give effect to the Applicant’s second response had it been necessary to do so. The specificity around the measurement of the visual field that s. 18(4)(d) adds to the requirements of s. 18(3)(b) could be considered in the context of the interpretation and effect of the visual field requirement contained in s. 18(3)(b) for which notice of a constitutional question was given. This point of statutory interpretation was not argued directly, but I do not believe it to be controversial that the determination of the constitutional validity of s. 18(3)(a) and (b) requires those sections to be read in their entire context and harmoniously with the scheme of the Regulations. In this case, that could involve consideration of the effects of s. 18(4) as well. See Rizzo & Rizzo Shoes Ltd. (Re), 1998 837 (SCC), [1998] 1 S.C.R. 27, at para. 21.

2.2) Does the Constitutional Challenge Apply to All Commercial Licence Holders covered by Sections 18(3)(a) and (b), or is it Limited to Class C drivers?

[31] The parties also disagree about whether the constitutional challenge is restricted to Class C bus drivers (which was the class of licence previously held by the Applicant), or whether it includes all classes of commercial drivers’ licences to which ss. 18(3)(a) and 18(3)(b) apply.

[32] The Respondent argues that the Applicant cannot challenge trucks or other classes of commercial licences. Evidence about trucks and their safety was not the focus of the record. For example, Dr. Smiley provided no human factors evidence for trucks. Further, the Applicant is challenging the constitutional validity of the Regulations as they apply to commercial licences for buses because she is a bus driver and has standing to do so.

[33] The Applicant’s challenge is, on its face, clearly specific to Class C bus drivers. Unlike the circumstances of s. 18(4), I am not able to say with confidence that the evidentiary record was fully developed in respect of the other classes of commercial licences and thus this would not qualify for a de facto notice exemption under the principles described in Eaton and Guindon in relation to the Attorney General of Ontario. Nor does it qualify for an exemption in relation to the Attorney General of Canada, for the same reasons indicated in relation to the s. 18(4) challenge.

2.3) The Effect of the Restrictions on the Applicant’s Constitutional Challenge

[34] The Applicant’s constitutional challenge is restricted to s. 18(3) of the Regulations and to Class C bus drivers. Practically speaking, that does not mean the analysis and outcome of this case will have no bearing on s. 18(4)(d) of the Regulations or to other classes of commercial drivers’ licences. However, the scope of the implications of this decision on those other aspects of the legislation is a matter for the legislature to address when it brings s. 18(3) into constitutional compliance for Class C drivers. Those broader implications are not a matter for this court to address on this application.

IV. THE EXPERTS

[35] There was an extensive written evidentiary record in this case, comprised of many volumes of witness affidavits, exhibits, and transcripts of cross-examinations of the various witnesses. The parties rely upon the reports of a combined total of five experts. The Applicant has tendered the evidence of Dr. Arshinoff and Dr. Peli. Ontario has tendered the evidence of Dr. Owsley, Dr. Smiley and Mr. Janusz. The Applicant challenges the qualifications of Dr. Owsley and Mr. Janusz on grounds of bias, impartiality, and in the case of Mr. Janusz, also on the grounds that he lacks relevant expertise.

1. The Relevance of the Expert Evidence

[36] The Applicant argues that the effect of the Regulations, which is to prohibit her from driving a bus, violates s. 15 of the Charter because it is a discriminatory distinction based on her physical disability. The Applicant seeks to establish discrimination in two main ways.

[37] First, she argues that there is limited evidence that monocular drivers, as a group, are actually unsafe commercial drivers. The Applicant argues that the evidence from past studies is mixed, and at most, demonstrates a slightly elevated risk of collision. If monocular drivers as a group are not unsafe, then the prohibition is discriminatory because it is based on stereotype.

[38] The Applicant’s second argument in the alternative is premised upon the court finding that monocular drivers, are more likely to be involved in motor vehicle collisions. Even if that were the case, she argues, to prohibit all monocular persons from driving commercially based on their membership in that group perpetuates an arbitrary disadvantage, because it is based on a stereotype. The stereotype – that they are unsafe or less safe drivers – does not correspond to the actual characteristics and circumstances of individuals within the group, such as the Applicant.

[39] The Applicant seeks a remedy of individual assessment to demonstrate that she is a safe commercial driver, even if she belongs to a group that is statistically less safe. The Applicant supports her position, favouring individual assessment, with reference to several studies and programs in other jurisdictions which allow monocular people to become commercial drivers through exceptions based on individual assessment.

[40] The crux of the debate is about whether the Applicant should be entitled to prove her ability to be a safe bus driver through individual assessment (driver performance) rather than be judged to be an unsafe driver based on statistics. Ontario maintains that there is a critical distinction between driving safety (which its expert Dr. Owsley insists can only be measured by motor vehicle collision rates or statistics) and an individual’s driving performance. This distinction underlies the disagreement between the parties in this case. Statistics about driving safety are what Ontario relies on to justify the restrictions in the Regulations that result in the across-the-board exclusion of monocular drivers from ever qualifying for a Class C commercial driver’s licence.

[41] Ontario agrees that the Regulations prohibit monocular drivers from driving commercially. Ontario argues that the evidence reviewed by its expert Dr. Owsley, and its own studies, demonstrate that monocular commercial drivers have an elevated risk of collision. Ontario argues that this evidence demonstrates that monocular drivers, as a group, are unsafe. If monocular drivers are unsafe, then the Regulations prohibiting them from driving are not discriminatory. Rather, the Regulations would accurately correspond to the group’s capacities. Regulatory line drawing may, at some level, be arbitrary but it is not discriminatory when it corresponds with measured predictors of safety such as motor vehicle collision rates.

[42] In the alternative, Ontario argues that even if the evidence is mixed on the driver safety question, the evidence is inconclusive at best when it comes to reliable predictors of safety based on an individual driver’s tested performance. Therefore, Ontario’s position prohibiting assessments based on driver performance is reasonable and Ontario should be entitled to deference. It argues that governments are entitled to deference when scientific evidence is unclear, particularly in the area of public safety.

2. What Do the Experts Say?

[43] The experts reviewed and synthesized the available scientific reports and studies, in addition to offering their own assessments and opinions about the safety of monocular drivers and how that can be assessed.

2.1) Dr. Arshinoff

[44] Dr. Steve Arshinoff testified on behalf of the Applicant. He is a medical doctor and a practicing ophthalmologist. He assessed Ms. Di Cienzo’s visual field. He provided evidence about the physiology of the eye and the impact of monocularity on vision.

[45] Dr. Arshinoff conducted an ophthalmic assessment of the Applicant on August 26, 2016. He assessed her visual acuity as 20/20 in her left eye. He assessed her horizontal visual field as 150 degrees. Dr. Arshinoff provided evidence that Ms. Di Cienzo met the requirements of s. 18(3)(b), but that she could not meet the requirements of s. 18(3)(a). She could not meet the visual acuity standard of 20/100 in her weaker eye.

[46] Using the Humphrey Visual Field Analyzer to apply the Esterman test in accordance with the manual and its customary usage, the Applicant measured as having a visual field of 126 degrees. Dr. Arshinoff posited, based on the generally accepted view in the literature, that as a monocular individual, the Applicant’s full field of vision from her healthy eye would be larger than what this test was able to measure and that it was at least 150 degrees.

[47] The Humphrey Visual Field Analyzer, which uses a variety of tests to measure visual field, has limitations which make it difficult to accurately measure a monocular individual’s visual field. The apparatus requires the test subject to centre their view. When performing the Esterman test, the apparatus does not have reference points for measurement beyond about 76 degrees to each side.

[48] Dr. Arshinoff subsequently performed the test in a different manner than is customary. Rather than centering her view, he asked the Applicant to tilt her head 15 degrees. He did not specifically measure how far she tilted her head.

[49] The theory behind Dr. Arshinoff’s use of this unconventional test methodology is that it accommodated the particular characteristics of a monocular person. Specifically, that a normal monocular field of vision would display a greater field of vision on one side and that the way the visual field test is typically administered does not allow for the full measurement of the field of vision in the healthy eye. The test is administered in a way that does not account for the full visual field on the stronger side.

[50] Dr. Arshinoff also provided evidence about the functional impairments that result from the loss of an eye. He stated the chief visual deficit of a monocular person is their loss of stereoscopic vision. He described stereoscopy and provided examples.

[51] Dr. Arshinoff also conducted a literature review of several academic articles on vision and driving. Dr. Arshinoff considered both collision rate studies and driving performance studies. Dr. Arshinoff applied this theory and his knowledge of the science and literature to his functional and clinical assessment of the Applicant. He concluded that the Applicant possessed the functional ability to meet the driving standards for a Class C licence, despite not meeting the specific requirements of s. 18 of the Regulations.

2.2) Dr. Owsley

[52] Dr. Cynthia Owsley testified on behalf of Ontario. She is a professor in the Department of Ophthalmology at the University of Alabama. She holds the Nathan E. Miles Chair of Ophthalmology. She holds a doctorate degree in experimental psychology and subsequently attained a master’s degree in public health. Her research focuses on aging-related eye disease and vision impairment, as well as the relationship between vision and driving. She provided evidence about statistical analysis versus individual performance analysis when assessing driver safety, the impacts of monocularity on vision, reviewed past academic studies on the safety of monocular drivers, and responded to Dr. Arshinoff’s report.

[53] Dr. Owsley testified that Ms. Di Cienzo did not meet Ontario’s vision standard for visual acuity under s. 18(3)(a), as she does not have a functioning right eye. Dr. Owsley also testified that the Applicant did not have a horizontal visual field of at least 150 degrees. Dr. Owsley based this on four visual field tests conducted by Drs. Gill and Yoo who tested the Applicant using the Humphrey Visual Field Analyzer.

[54] Dr. Owsley distinguished between driver safety and driver performance. She characterized driver safety as the metric measuring motor vehicle collisions (“MVC”) incurred by a population referenced against driving exposure, usually in terms of miles driven, kilometers driven, or years driven. Safety is therefore the MVC rate for a given population. MVC rates for different populations can be compared – such as for drivers with a certain type of vision problem.

[55] Conversely, Dr. Owsley defined driver performance as “behind the wheel” vehicle control. This could include lane keeping, speed, rapid acceleration or deceleration, movement through intersections, and obeying traffic control signals. Driver performance is typically measured in short periods of time, accompanied by an evaluator. The evaluator makes a subjective qualitative assessment. The driver knows that they are being evaluated and is therefore on their “best behavior.” According to Dr. Owsley, collision risk can only be measured by the MVC rate, not by driver performance.

[56] Dr. Owsley testified about the visual demands of bus drivers, and the impact of having one eye. Specifically, monocular individuals do not have the benefit of binocular summation,[^3] which leads to better light sensitivity, contrast vision, flicker sensitivity, acuity, color perception, and motion perception. Further, she opined that monocular disadvantages for visibility could be exacerbated under common visual circumstances for bus drivers.

[57] Dr. Owsley reviewed and summarized four studies on monocular drivers and road safety: the Rogers study (1992), the US Department of Transportation study (1996), the Keeney study (1981), and the McKnight study (1991). These studies are reviewed in greater detail below. Dr. Owsley concluded that only the Rogers study and the Keeney study were reliable predictors of driver safety, having regard to her definition of driver safety which requires a metric that measures MVCs as a rate referenced against driving exposure to compare multiple population groups. According to Dr. Owsley, research has shown that “behind the wheel” driving performance assessments are not predictive of MVC rates.

[58] Dr. Owsley concluded, based on the questions asked of her, that the Applicant met neither the visual acuity standard nor the visual field standard under the Regulations. Dr. Owsley concluded that monocular vision leads to a number of specific visual weaknesses. Relying on the two population-based studies, the Rogers study and the Keeney study, Dr. Owsley concluded that monocular drivers were 1.4 to 1.9 times more likely to be involved in MVCs than drivers who did not have that vision problem.

2.3) Dr. Peli

[59] Dr. Eli Peli testified on behalf of the Applicant. He is a professor of ophthalmology at Harvard Medical School. He holds a master’s degree in biomedical engineering and a post-graduate degree in optometry. His research includes the interface between engineering, ophthalmology, and vision research. He provided evidence about the consequences of the loss of one eye, including loss of visual field, stereoscopic vision and binocular summation. He also provided evidence about driving a commercial vehicle as a monocular person and provided responses to Dr. Owsley and Dr. Smiley’s reports.

[60] Dr. Peli concluded that there is no evidence that the loss of an eye affects the safety of drivers of any vehicle, including large commercial vehicles. Dr. Peli noted that there was very little evidence that driving with one eye limited the functionality of the driver of any vehicle. The loss of an eye causes impairment. Dr. Peli opined that this impairment is not linked to increased MVC rate, and that causality has not been established.

[61] Dr. Peli testified that the normal visual field for a monocular person was 155 degrees. Dr. Peli relied in part on “Dynamic Visual Fields of One-eyed Observers,” a 2005 paper by Good et al., to explain Ms. Di Cienzo’s visual field. In the Good paper, the authors evaluated subjects’ visual fields at different gaze positions and found they had over 50 degrees of visual field nasally (towards the nose) and over 100 degrees of visual field temporally (towards the outside of the head). However, they only measured as having 131 degrees of visual field when measured using the Esterman test, due to the machine’s limitations in assessing monocular individuals.

[62] Dr. Peli also disagreed with Dr. Owsley’s statement that Ms. Di Cienzo had a horizontal visual field of less than 150 degrees. He explained why the perimetry tests used were inappropriate for monocular people and did not adequately describe her visual field. Since the visual field of a healthy eye is typically over 150 degrees, that should be a given unless there was some indication otherwise. He noted that this point does not need to remain theoretical if it could be individually determined using a different testing apparatus.

[63] Dr. Peli also reviewed the physiological impacts of monocularity, including loss of some visual field, loss of stereoscopic vision, and lack of binocular summation.

[64] Dr. Peli noted that many jurisdictions allow monocular drivers to drive passenger vehicles but have a higher standard for commercial vehicles. Dr. Peli opined that many of these vision standards did not rationally correlate to actual capabilities of the individuals in question. In his view, the visual acuity standard of 20/100 made “little scientific or medical sense in terms of acuity.”

[65] Dr. Peli also reviewed the 2008 Expert Panel he sat on with Dr. Owsley. Drs. Peli, Owsley, and Frank Berson were appointed as Medical Expert Panel Members to advise the USA Federal Motor Carrier Safety Administration on the vision standards for commercial drivers. Two questions they considered were particularly relevant to this case. The Expert Panel concluded that the evidence available at that time did not support a recommendation to change the standard, which had the effect of prohibiting monocular drivers. Although the available evidence (the US-DOT study) showed that monocular drivers had fewer crashes than the binocular control group, the Panel was concerned about the methodology in selecting the control groups in the studies under consideration.

[66] The second question was whether visual field loss was associated with increased crash risk. The Panel (including Dr. Peli and Dr. Owsley) concluded that there was insufficient evidence to support this hypothesis and did not recommend increasing the visual field standard for commercial truck drivers in the United States from 70 degrees in each eye to 120 degrees in each eye.

[67] Dr. Peli agreed with Dr. Owsley that visual field loss is elevated with an increased rate of MVCs. However, he also pointed out that it matters where the visual field loss is. Visual field loss in the right field region was not associated with elevated collision rates in Dr. Owsley’s paper.

[68] Dr. Peli concluded there was no evidence that the loss of an eye affects the safety of a driver, including a driver of large commercial vehicles. Further, he did not consider the loss of an eye to be a scientific basis for the Regulations, as research has failed to link that driver performance or safety.

[69] Dr. Peli did not agree that driving performance studies have no value. To the contrary, he considers those that have been conducted to be an indication that tests could be developed to assess individual monocular driver safety. Put another way, the absence of a specific accepted performance test for assessing the safety of monocular drivers is not an indication that their safe driving abilities cannot be tested.

2.4) Dr. Smiley

[70] Dr. Allison Smiley testified on behalf of Ontario. She is a certified engineer and has been the president of the consulting firm Human Factors North since 1984. Dr. Smiley holds a master’s degree and a doctorate degree in systems design engineering. Dr. Smiley’s past experience includes working on human factors assessments of driving. She provided evidence about the human task of driving, the use of vision when driving, the task of driving a bus (specifically in Oakville), the differences between bus driving and passenger car driving, and the relationship between monocular vision and crash risk. Dr. Smiley visited Oakville Transit and rode on various bus routes on March 9 and March 21, 2017.

[71] As a human factors expert, Dr. Smiley provided evidence about the difference between driving transit buses versus passenger vehicles. Some notable differences include that:

a. Bus drivers must change lanes more frequently than drivers of passenger cars.

b. Bus drivers need longer rear gaps to change lanes, and consequently spend more time looking in their left mirror while merging with traffic.

c. When turning right, bus drivers must enter the lane left of the curb lane to create enough space to safely turn.

d. Bus drivers are regularly stopping to pick up and drop off passengers.

[72] Dr. Smiley provided evidence about the visual demands of different driving tasks. Dr. Smiley testified that the typical visual field of a monocular person is larger than 150 degrees, but she explained the concept of driver visual search, including the need for, and process of, drivers making multiple visual fixations.

[73] Dr. Smiley also provided evidence about the impact of monocularity on the task of driving. She testified that it may be possible for drivers to compensate for a reduced field of view by turning their heads further, or by scanning more frequently. However, she added that driving tasks are “time-pressured,” and increasing the amount of head-turning or visual fixations takes additional time.

[74] Some examples of “time-pressured” visual search tasks occur when drivers must detect unexpected path intrusions, such as pedestrians crossing against a light, or when determining whether to accept a gap, when changing lanes or entering a roundabout. Dr. Smiley noted that many crashes would have been preventable if the driver had responded 0.5 seconds to 1 second earlier. Dr. Smiley pointed out that the additional time spent scanning or turning the head during visual search tasks could potentially lead to more collisions.

2.5) Chris Janusz

[75] Chris Janusz testified on behalf of Ontario. He is a senior policy analyst with the Ontario Ministry of Transportation. He holds a master’s degree in transportation engineering and a master’s degree in transportation planning. He works in Ontario’s Road Safety Office.

[76] Mr. Janusz jointly swore an affidavit with his superior, Yoassry Elzohairy. The affidavit described Ontario’s Commercial Vision Waiver Pilot Program (“Waiver Program”). The Waiver Program allowed for the licensing of commercial truck drivers who otherwise did not meet the vision requirements to acquire Class A or Class D licences. The program existed for ten years, from 1996 to 2006.

[77] Waiver Program participants were required to renew their waivers annually. Ontario collected data during this pilot project. A review of the program conducted in 2004, using the “relative risk” method of statistical analysis, indicated the risk of collision for Waiver Program drivers was 2.3–3.5 times higher than for drivers not in the Waiver Program. The program was not renewed after 2006. No new drivers were given waivers, but participants were allowed to continue renewing their waivers on a going-forward basis.

[78] In 2017, Mr. Janusz and his colleague Mr. Elzohairy reviewed the cumulative data of the Waiver Program and drivers who had maintained their waivers after the Waiver Program’s end in 2006. The 2017 data showed that drivers in the Waiver Program had between 1.40-1.64 times more collisions per 10,000 drivers than other Class A and Class D drivers in Ontario. This analysis was based on a raw numerical comparison, rather than the relative risk method.

3. The Applicant’s Challenges to the Admissibility of Two of Ontario’s Experts

[79] The Applicant asks the court to rule that two of three of Ontario’s witnesses who were proffered as experts, Dr. Owsley and Mr. Janusz, should not be qualified as such and their opinion evidence should not be admitted.

3.1) The Applicant asserts that Dr. Owsley is not impartial; Ontario disputes this assertion

[80] The Applicant primarily criticizes Dr. Owsley for being an advocate and not providing a balanced perspective or opinion. Among the criticisms of her lack of objectivity was her failure to disclose her participation in the 2008 Expert Panel.

[81] That panel of American experts (that included both Dr. Owsley and Dr. Peli), among other things, declined to recommend an increase in the minimum required visual field of a commercial interstate truck driver from 70 degrees in each eye to 120 degrees in each eye on the basis that there was insufficient evidence to support the hypothesis that visual field loss was associated with increased crash risk. Further, the Expert Panel expressed some criticisms of the Rogers and Keeney studies that Dr. Owsley preferred to rely upon in her expert report. When questioned about the 2008 Expert Panel in cross-examination, the Applicant asserts that Dr. Owsley was defensive and argumentative in her responses.

[82] The Applicant also critiques Dr. Owsley for stating in her initial report that the normal visual field for a monocular person with a healthy eye is 140 degrees. This was contradicted by the other experts, including Dr. Smiley, and an authoritative text. Dr. Owsley eventually conceded that the normal expected monocular horizontal visual field is 155 degrees. Further, she selectively chose to prefer the two of four studies that say the monocular drivers have increased risk without discussing their limitations (including the limitations noted in by the 2008 Expert Panel that she participated in). Finally, she was not given Ontario’s 1995 and 1998 internal studies to review before drafting her report. Nor did the Applicant have an opportunity to cross-examine her on these studies, due to their late disclosure.

[83] Consequently, the Applicant argues Dr. Owsley should not be qualified as an expert, or that her expert evidence should be given little weight.

[84] Ontario says Dr. Owsley’s expert report was a preponderance of evidence report and it was reasonable for her to look at all of the prior studies, including studies with weaknesses.

[85] Ontario argues that exclusion based on partiality is rare. The Respondent says that “the Applicant’s essential position appears to be that Dr. Owsley was under an obligation to highlight and explain any publication, or limitations in the publications, that might be favourable to the Applicant’s case.” Further, Ontario argues that because Dr. Owsley has listed all of the studies in which she has participated in her CV, she was not required to expressly disclose her participation in the 2008 Expert Panel. The Applicant was free to cross-examine her about the 2008 Expert Panel and did so.

[86] Ontario also points out that Dr. Owsley’s expertise is acknowledged by the Applicant’s own expert, Dr. Peli. Ontario maintains that any criticism of her evidence should go to its weight, not its admissibility.

3.2) The Applicant asserts that Mr. Janusz is not impartial and not a qualified expert; Ontario disputes these assertions

[87] The Applicant argues that Mr. Janusz is not independent, impartial, or otherwise qualified to give the opinion evidence he has proffered. Mr. Janusz was not presented as an expert until Ontario served an “Acknowledgment of Expert’s Duty” on the Applicant one week before his scheduled crossexamination, two years after Mr. Janusz swore his affidavit.

[88] The Applicant relies on a case in which the expert was not apprised of the expert’s duty at the outset of the retainer and was also the source of the information relied upon for the opinion evidence given (see: M.M. v R.M., 2016 ONSC 7003, 94 R.F.L. (7th) 204, at para. 16). In that case, as the expert had a prior retainer as an accountant to one of the litigants, it was impossible for him to be informed of the expert’s role and duty at the outset of his engagement – which would have predated the litigation.

[89] The Applicant analogizes to Mr. Janusz. He was an employee of Ontario during the period in which the Waiver Program was in place and being studied. He is the source of the information on which he is opining. He was presented to the Applicant as a fact witness. He was later presented as an expert witness. He could not have been aware of the expert’s duty of independence and impartiality if and when, in the course of performing his employment duties, he learned about the result of Ontario’s Waiver Program between 1996-2006, nor when he was asked by the Respondent in 2017 to provide an updated analysis of the persons who continued in the Waiver Program after 2006.

[90] The Applicant also argues that Mr. Janusz lacks the necessary expertise to be qualified as an expert. He does not have a background in statistics. Nor can he be qualified as a participant expert witness, as the Court of Appeal allowed for in Westerhof v. Gee Estate, 2015 ONCA 206, 310 O.A.C. 335. He did not prepare the 2017 Waiver Report as part of his normal course of duties, nor was he the individual who prepared the 2006 Waiver Report as part of his duties for the Ministry of Transportation. He cannot claim the status of a participant expert for either of those reports about which he testifies.

[91] Mr. Janusz’ lack of expertise in statistics is said by the Applicant to be borne out in several criticisms of his analysis. The Applicant argues that Mr. Janusz made a number of statistical errors, all of which compound in the same direction: skewing the numbers to indicate that monocular drivers in the Waiver Program have higher collision risk rates, including:

a. There was no calculation of statistical significance.

b. The study used “number of years with a licence” to measure driver exposure, rather than “kilometers driven”, even though Appendix A to the 2006 Review and the Stewart paper upon which the study relies state that “exposure to risk,” the denominator for a risk measurement rate, should be measured in terms of distance travelled (if not kilometers driven then a proxy such as “number of trips” or “time spent on the road”). If not, assumptions must be made. The 2006 Review states in Appendix A that it assumes that all drivers are driving the same distances, with the same travel patterns, and maintaining an active status each year.

c. As the Waiver Program drivers had to renew their licences more regularly (every year compared to every three to five years for non-waiver drivers), the non-waiver pool likely had a greater proportion of inactive (and therefore less accident-prone) drivers than the waiver pool, which Mr. Janusz’ report did not account for;

d. Non-waiver drivers’ years licensed were rounded up, while waiver drivers’ years licensed were rounded down, resulting in an inflated calculation of the waiver drivers’ collision risk. The 2006 Review and 2017 Report counted the number of collisions by the number of drivers in each year. The 2006 Review counted Waiver Program drivers and non-Waiver Program drivers differently. A Waiver Program driver who received a licence in February 2000, for example, would not be counted for the Waiver Program’s statistics until the end of year analysis for 2001. Conversely, a non-Waiver Program driver who received a licence in July 2000 would be counted as one of the drivers for 2000. Mr. Janusz agreed in cross-examination that when dividing the number of collisions by the number of drivers, this would overestimate collisions for the Waiver Program drivers and underestimate collisions for the non-Waiver Program drivers.

e. Time spent serving a suspension was subtracted from the Waiver Program drivers in the 2017 Report, but not from the non-waiver drivers. This, too, would inflate the collision rates for Waiver Program drivers (by providing a lower denominator).

[92] Mr. Janusz acknowledged that these biases existed in the analysis but did not agree that they were intentional or that they materially affected the conclusions presented in the 2017 Report. Ontario maintains that:

a. Calculations of statistical significance or confidence intervals were not necessary, as Mr. Janusz used the entire population of Waiver Program drivers for the 2017 Report’s calculations.

b. There were challenges inherent in getting kilometers-driven data. Mr. Janusz did not disagree that this would have been a better denominator if that data had been available, nor did he attempt to justify the assumptions made in the absence of that data. It was sufficient that they were disclosed.

c. The non-waiver group likely included inactive drivers, and Mr. Janusz agreed that it could potentially influence the results by deflating the collision rates of the non-Waiver Program drivers but did not agree that it materially impacted his conclusions.

d. Ontario conceded that “years licensed” was rounded differently for the Waiver Program drivers and the non-waiver drivers. In the 2017 Report, Mr. Janusz was able to partially correct the discrepancy in the way in which the number of drivers were counted for establishing MVC rates. In the 2017 Report, he measured the Waiver Program drivers by their specific entry and exit date. This improved the reliability of the 2017 Report but the number of drivers in the non-Waiver Program, calculated annually, was still artificially inflated because the 500,000 Class A and Class D drivers were too many to sort through to determine the date of registration. However, this discrepancy would only affect a small number of newly registering drivers, limiting its effect on the data.

e. Identifying and calculating each of the suspensions for the over 500,000 individual Class A and Class D drivers would have been too time-consuming to sort through. Further, Ontario did not believe that eliminating this discrepancy would have a significant impact on the data.

[93] Ontario argues that for some of the statistical critiques raised by the Applicant, the impact is calculably negligible. For the other challenges, they would likely have made little or no difference to the final result. Notably, the Applicant seeks to rely upon Mr. Janusz’ validation of the statistical observations of the Applicant’s counsel Ms. Song that suggest that 3% of the waiver drivers were responsible for 20% of the collisions, and 50% of the waiver drivers were not involved in any collisions.

[94] Further, even though a notice of acknowledgment of expert’s duty was not originally served, Mr. Janusz has since signed one. An internal employee can be independent and impartial. Challenging his expertise based on the fact that he does not have a Ph.D., or because he has not published academic articles ignores his legitimate qualifications. He is a civil engineer with expertise in traffic safety who does this type of statistical analysis of road safety data all the time.

4. The Applicable Legal Principles Regarding Challenges to Experts

4.1) The test for admissibility, generally

[95] In White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23, [2015] 2 S.C.R. 182, the Supreme Court of Canada confirmed that the legal test for admissibility of expert evidence has two components. First, the party tendering the evidence must persuade the trial judge that the proposed expert opinion satisfies the four criteria set out in R. v. Mohan, 1994 80 (SCC), [1994] 2 S.C.R. 9. The evidence must be: (i) relevant; (ii) necessary; (iii) not barred by any other exclusionary rule; and (iv) given by a properly qualified expert (White Burgess, at para. 23).

[96] Second, the trial judge must engage in a gatekeeping function and balance the potential risks and benefits of admitting the evidence to determine whether the potential benefits justify the risks (White Burgess, at para. 24).

4.2) Factoring independence and impartiality into the analysis.

[97] Experts have a duty to the Court (White Burgess, at paras. 26-32). As part of this duty, expert witnesses must be independent and impartial (at para. 32). The impartiality and independence of an expert can impact the admissibility of that expert’s evidence (at paras. 34, 40, 45). The court considers these factors at two separate points during the admissibility analysis (at paras. 34, 52-54).

[98] First, the court considers independence and impartiality at the “threshold stage” of the analysis, or the first step of the admissibility analysis (White Burgess, at para. 53). The threshold stage relies on the four Mohan criteria to determine threshold admissibility. The court assesses independence and impartiality when considering if an expert is “properly qualified,” the fourth Mohan criterion (at paras. 52-53). For an unanimous Supreme Court, Justice Cromwell stated at para. 53, “A proposed expert witness who is unable or unwilling to fulfill this duty to the court is not properly qualified to perform the role of an expert.”

[99] Second, the court also considers independence and impartiality at the “gatekeeper stage” of the analysis, or the second stage of the analysis (White Burgess, at para. 54). This is distinct from the analysis at the first stage. For example, the court could have “some concern” about an expert’s independence and impartiality, but find that overall, the expert was properly qualified. The expert would pass the “threshold stage”. However, at the gatekeeper stage, in conjunction with other factors, the court could rule that in the exercise of its gatekeeper function, the expert’s evidence nonetheless would not be admissible.

[100] If an expert passes the threshold stage and the gatekeeper stage of the analysis, that expert’s evidence will be admissible. Determining admissibility does not end consideration of independence and impartiality. Even if the expert’s evidence is admissible, any possible concerns about independence and impartiality that arose earlier can still be considered when the trier of fact weighs the evidence (White Burgess, at para. 45).

5. Analysis of the Applicant’s Challenges to Ontario’s Experts

[101] The Acknowledgement of Expert’s Duty signed by both of the challenged experts is an important consideration in the assessment of their independence and impartiality. They have, by that signature, acknowledged in writing that they understand it is their duty in this proceeding, prevailing over any obligation which they may owe to any party on whose behalf they were engaged, to provide:

a. evidence that is fair, objective and non-partisan; and

b. opinion evidence only related to matters within their area of expertise.

5.1) Concerns about Dr. Owsley’s impartiality or independence

[102] It is in the face of such a written acknowledgment from a professional that Ontario quite properly observes that it is a rare occasion that an expert’s proffered testimony will not be admitted because it is nonetheless found to be partial and lacking in objectivity. On the threshold question under the White Burgess test, of whether Dr. Owsley’s alleged impartiality is of a nature and extent that it obfuscates her qualification on matters within her areas of expertise (which are admitted by the Applicant’s own expert Dr. Peli), the evidence of partiality would have to be more than what is alleged.

[103] A finding of partiality and lack of objectivity would require more than her not drawing explicit attention to her role in the 2008 Expert Panel and the panel’s view that conclusive evidence of an association between visual field loss and increased crash risk was lacking. She did disclose in her report that she had participated in the 2008 Expert Panel and explained the consensus views when she was asked about them in cross-examination. Similarly, while she did not disclose the limitations that had been noted by the 2008 Expert Panel about the Rogers and Keeney studies that she prefers to rely on, she was clearly aware of those limitations and has explained that they did not change her views about the reliability of those studies.

[104] I find Dr. Owsley to be properly qualified as an expert in the field of the impact of visual impairment on driver safety and performance under the first part of the White Burgess test. I do not find the concerns about the timing and manner of her disclosure of certain evidence that the Applicant seeks to characterize as “unfavourable” to the position of Ontario to be of a nature or quality that disqualify her from giving her evidence at the second, gatekeeper stage of the White Burgess test. Dr. Owsley’s expert evidence is admitted. The concerns raised by the Applicant may, however, go to the weight afforded to Dr. Owsley’s evidence, when considered against the evidence of the experts who have disagreed with her.

5.2) Concerns About the Impartiality, Independence and Qualifications of Mr. Janusz

[105] I agree with the Applicant that Mr. Janusz is not a “participating” expert in that the work he did in relation to Ontario’s Waiver Project was all done as part of Ontario’s response to this application. He did not directly participate in the 2006 Review of the Waiver Project. He summarized it for purposes of this proceeding and the 2017 Report of the Waiver Project was specifically done for purposes of Ontario’s response to this proceeding. See Westerhof, at para. 60.

[106] To be qualified to give provide evidence about statistics, one does not require a Ph.D. or other graduate level degree in statistics. The fact that Mr. Janusz’ job requires him to apply statistical analyses to data within the Ministry of Transportation affords him sufficient credentials to make a statistical presentation about data that was generated within Ontario’s Waiver Program.

[107] An expert can be employed by a party. See Wise v. Abbott Laboratories, 2016 ONSC 7275, 273 A.C.W.S. (3d) 252, at para. 71.[^4] Having considered the explanations provided by Mr. Janusz and Ontario, I find that any statistical biases in the methods used to analyze and present the data that might have tended to inflate the MVC risk for monocular drivers were not intentional. Mr. Janusz did concede some “directional” patterns in the examples provided. His testimony was fair and balanced on cross-examination. He was careful not to tread outside of his expertise (to the point that the Applicant was critical of the limitations he placed on the scope of his testimony) and he conceded points that the Applicant relies upon as favourable to her position in this proceeding.

[108] There is a risk of an unconscious confirmation bias for someone like Mr. Janusz whose evidence supports decisions that have been made by his employer. This theoretical concern is balanced by the fact that his 2017 Report dilutes the results of the 2006 Review; it does not seek to uphold or defend it. Further, the unconscious trends that he acknowledged might have tended to inflate the MVC risk for monocular drivers have not been demonstrated to have materially impacted the results of either the 2006 Review or the 2017 Report. With the limitations that he himself placed on the scope of his expertise in mind, I find Mr. Janusz to be properly qualified as an expert in the analysis of road safety data under the first part of the White Burgess test.

[109] I do not find the concerns about the timing of his acknowledgment of expert’s duty or the competing duties that he may also have to the respondent who employs him to disqualify him from giving evidence at either the threshold stage or at the second, “gatekeeper” stage of the White Burgess test. Albeit after he completed his review and analysis, he has acknowledged that his duty to the court takes precedence over any duties he may owe to Ontario. While it cannot be assumed that he undertook his work having regard to the expert duties that he was only asked to acknowledge afterwards, he signed the acknowledgement and responded to the challenges to that work that were raised on cross-examination after having accepted those duties. The concerns about his independence are not as salient as in the case of M.M. v. R.M. relied upon by the Applicant, and do not amount to a reason to exclude the evidence of Mr. Janusz in this case. Mr. Janusz’ expert testimony presenting and analyzing the data about drivers who participated in Ontario’s Waiver Project is admitted.

6. The Studies Reviewed and Considered by the Experts

6.1) Ontario’s Literature Reviews in 1995 and 1998

[110] The experts relied on many secondary sources that have studied monocular drivers. Dr. Owsley conducted a literature review as part of her work, which is summarized earlier in these reasons.

[111] Prior to engaging Dr. Owsley, Ontario had undertaken its own literature reviews in 1995 and 1998.

The 1995 MTO Report

[112] In 1995, Ontario’s Ministry of Transportation conducted the “Monocular Vision and Commercial Motor Vehicle Safety” report (“the 1995 MTO Report”). The 1995 MTO Report included, among other things, information about vision, an analysis of collision rate data for Ontario drivers, and a literature review of scientific literature on motor vehicle operation and visual impairment.

[113] The 1995 MTO Report reviewed some of Dr. Keeney’s scientific literature, including two 1981 papers, a 1994 paper, and Dr. Keeney’s 1993 expert report in Martin v. The Queen, the previous Charter challenge to the Regulations which settled prior to adjudication. The 1995 MTO Report also reviewed the 1987/1992 Rogers studies and the 1991 McKnight study. These studies were also reviewed by Dr. Owsley.

[114] In s. 3.1.2, “Conclusions from the Scientific and Clinical Literature”, the 1995 MTO Report focused on the Rogers studies and the McKnight study. The Report noted that McKnight and Rogers drew different conclusions about the safety of monocular commercial vehicle drivers. The 1995 MTO Report also was prescient in its observation that determining the appropriate measure of safety – whether performance-based studies or statistical analyses of collision rates – would greatly assist in determining the safety of monocular commercial vehicle drivers. Specifically, it was observed that the McKnight study and Rogers study used different approaches (McKnight being performance based and Rogers being statistics based) and that this could be the reason for their different results.

The 1998 Engel-Townsend Report

[115] In 1998, Engel and Townsend Inc. produced a report for Ontario’s Ministry of Transportation titled “Monocular Vision and Commercial Vehicle Driving” (“the Engel-Townsend Report”). The report included an analysis of the visual task of driving, and a robust literature review.

[116] The Engel-Townsend Report identified three categories of scientific literature: (a) collision rate analyses, (b) expert opinion reports, and (c) experimental tests of monocular drivers.

[117] In the collision rate analyses, the Engel-Townsend Report reviewed the Rogers studies, the Johnson and Keltner study, the 1995 MTO Report analysis, and the United States Department of Transportation study of its own waiver program.

[118] In the expert opinion section, the Engel-Townsend Report reviewed research by Dr. Keeney and the Decina study, which is not relevant for this application.

[119] In the experimental test section, the Engel-Townsend Report reviewed the McKnight study and the Troutbeck and Wood study. These are also described as “driver performance” studies.

6.2) The Underlying Studies

[120] The scientific literature and studies that the experts were most focussed on for purposes of this case are summarized in more detail below. Monocularity was defined differently in the different studies. Some of the studies talk about monocularity without defining it at all. All of the studies are clearly referring to persons who have a severe visual deficit in one of their eyes. However, the extent of the deficit varied in some studies and was unclear in some others.

Motor Vehicle Collision Rate Studies

The Rogers Study

[121] The Rogers study was a two-year study of heavy vehicle operators. The study was published in 1987 and a subsequent article followed in 1992 (hereafter collectively referred to as “the Rogers study”). The study evaluated 16,465 male drivers, including 1,202 visually impaired drivers. Drivers with visual acuity worse than 20/40 but better than 20/200 in the weaker eye were grouped as “moderately impaired,” while drivers with visual acuity worse than 20/200 in the weaker eye were grouped as “severely impaired.”

[122] There were 660 drivers with severe visual impairment in one eye. 533, or 81% of the severely impaired subgroup, had no measurable vision in one eye or lacked an eye, for example, they were monocular. The severely visually impaired drivers were found to be 1.46 times more likely to have an increased MVC rate than non-impaired drivers. However, the moderately visually impaired drivers also had a similar increased MVC risk rate.

[123] The study concluded that heavy vehicle commercial drivers who failed to meet federal visual acuity standards had, on average, more accidents. The Rogers study found stronger evidence of this link for the severely impaired group, most of whom were monocular. However, the authors also noted that the evidence should not be considered compelling as there was insufficient data on driver exposure (distance or time driving). Dr. Owsley relies on this study and considers it to be credible.

[124] The Rogers study was assessed in both Ontario’s 1995 MTO Report and the 1998 Engel-Townsend Report. Both literature reviews found the Rogers study to be credible.

[125] The 1995 MTO Report assessed weaknesses admitted by Rogers and the other authors. One significant issue was that the visually impaired drivers were not supposed to drive out of California. This created two cascading issues. First, exposure data would be markedly different if visually impaired drivers were only driving intrastate, while non-impaired drivers were driving interstate. Second, drivers were not required to report out-of-state collisions. If visually impaired drivers were only driving in California, this would artificially inflate their collision numbers.

[126] However, this issue may have been resolved. Rogers later surveyed the participant drivers, both non-impaired and visually impaired, to measure their driving mileage. Reviewing the survey responses, Rogers realized that the visually impaired drivers were actually doing just as much interstate driving as the non-impaired drivers. Because the restriction only appeared on medical certificates, Rogers concluded that many of the visually impaired drivers actually did not realize they were prohibited from interstate driving. The authors of the MTO Report corresponded with Rogers and included responses to some other critiques in the 1995 MTO Report.

[127] The Engel-Townsend Report also noted some weaknesses of the Rogers study. They repeated the critiques of the 1995 MTO Report. However, the Engel-Townsend Report also noted “[E]ven if one accepts all Chipman’s criticisms, they do not seem sufficient to reject Rogers et al.’s basic finding that the monocular drivers in the study had more collisions and violations than the unimpaired drivers did.” In the conclusion on collision rate studies, the Engel-Townsend Report accepted that the Rogers study suggested monocular commercial drivers had a materially greater safety risk, although it was contrasted against other studies with divergent conclusions.

[128] The Applicant argues that the Rogers study provided “only conditional evidence” that monocular drivers had an increased collision rate. Further, she argues that the fact that moderately visually impaired drivers also had an increased collision rate in this study is relevant. The moderately visually impaired group would include some drivers who are entitled to commercial licences in Ontario.

The US-DOT Study

[129] The United States Department of Transportation conducted an extensive study of truck drivers involved in the United States federal vision-standard waiver program between 1993-1995 (“US-DOT study”). Drivers who fell below the vision standards received a waiver entitling them to drive commercially. There were several criteria to participate in the program, including having visual acuity below 20/40 in the weaker eye and three years of experience driving a heavy vehicle. There were 2,656 drivers in the program in August 1993 and 2,275 drivers in August 1995.

[130] Drivers in the US-DOT waiver program had markedly lower collision rates than the national collision rate. The waiver group’s crash rate was not higher, nor were the crashes more severe.

[131] Dr. Owsley criticized the study as being difficult to interpret because the severity of visual impairment was not available. In her view, the US-DOT study was not sufficiently specific in its definition of monocular to be a useful indicator of monocular driver safety. The 2008 Expert Panel also had concerns about this study, as noted by Dr. Peli.

[132] The Engel-Townsend Report echoes this critique, that it is not clear how many of the drivers were monocular. For example, some of them may have had significantly worse vision than 20/40 in their weaker eye, while some may have had vision only slightly worse than 20/40 in their weaker eye.

The Keeney Study

[133] The Keeney study (1981) examined the collision records of 52 monocular drivers in Kentucky’s Driver Limitation Program between 1976-1980 and compared them with the collision records of Kentucky drivers in general. Monocular was defined as individuals with 20/200 visual acuity or less in the weaker eye. Keeney found monocular drivers were 1.9 times more likely to have been involved in a MVC. Keeney concluded that monocular drivers should not be allowed to drive school buses, public transit vehicles, or emergency vehicles.

[134] Dr. Owsley attached Keeney’s 1981 study to her expert report and relied upon it. Dr. Owsley assumed that monocular drivers in the study were included because of their monocularity and not for any of the other grounds for inclusion in the study.

[135] Dr. Peli expressed concerns about the Keeney study. Namely, the monocular drivers were drawn from the “Driver Limitation Program” and were not representative of the population. The Driver Limitation Program had eight criteria that could result in an individual’s inclusion. One of the criteria was on the basis of a physician’s recommendation that an individual’s medical status would make them an unsafe driver. Some of the other criteria related to poor driving behavior, including inclusion on the basis of three collisions within the past twenty-four months, three convictions for driving violations while intoxicated within the last five years, or being reported by law enforcement as driving in an erratic manner because of a medical condition.

[136] The study does not distinguish between drivers in the different categories for inclusion. Although the study included 52 monocular drivers, it is not clear under which criteria they were included – whether some drivers were included because they were monocular, or whether they were included for other risky driving behaviours and also happened to be monocular.

[137] The monocular drivers within the Driver Limitation Program could have been there solely because they were monocular, or because they had otherwise qualified for the Driver Limitation Program due to their unsafe driving. If the study measured particularly unsafe monocular drivers, this would artificially inflate the collision rates of the monocular drivers by focusing on a subset of monocular drivers who were unsafe drivers for other reasons. The 1995 MTO Report included the 1981 Keeney study, but did not comment extensively on it. The Engel-Townsend Report was critical of Keeney’s research generally.

The Laberge-Nadeau Study

[138] The Laberge-Nadeau study (1995) was conducted in Quebec on visually impaired commercial drivers. 55% of the subjects were monocular. The study divided the group in eight categories: four sub-groups of taxi drivers and four subgroups of truck drivers, each subgroup representing a different age category. The study found that in seven of the eight subgroups, there was no statistically significant difference in accident rate. There was a statistically significant increase in accident rate for truck drivers age 35 and younger.

[139] This last conclusion is in contrast with the 2006 Review, which found that drivers in its youngest age cohort were the only drivers who did not have a significant increase in relative risk.

Driver Performance Studies

[140] The McKnight study compared driving performance between monocular and binocular commercial “heavy truck” drivers on a particular driving route. The study assessed the drivers’ visual search, lane placement, clearance judgment, and hazard detection, among other things. However, it did not define what it meant by “monocular.” Dr. Smiley noted that the McKnight study did not find a significant difference in the monocular drivers’ number of mirror checks. However, she also commented that “it is unlikely that the route required as many lane changes as is the case for bus drivers.”

[141] The McKnight study did not examine the collision rate over time for the two groups of drivers. It therefore provided little or no information about the statistical safety of monocular drivers. The study was designed to measure their performance at a given task. As it did not compare the collision rate for the two groups (a statistical comparison), Dr. Owsley opined that it therefore provided no information on the safety of monocular drivers.

[142] Both the Troutbeck and Wood study (1992) and the Racette study (2005) came to similar conclusions as McKnight. The Troutbeck and Wood study had individuals with healthy binocular vision obscure their vision to simulate being monocular, and to simulate having visual fields of 40 and 20 degrees. Visual fields of 40 and 20 degrees were associated with poor performance on some of the driving tasks, but a monocular field was not.

[143] The Racette study collaborated with a medical rehabilitation centre which assessed driving after an illness or injury. The study reviewed the rehabilitation files for 1350 patients who had recently undergone a driving assessment, 131 of whom had some visual field loss, and 25 of whom were monocular. The study grouped assessment results into “unsafe, unknown, and safe.” Based on the review of the driving assessments, the Racette study concluded that visual field loss caused no significant impact on driving performance, and that “a large proportion of the monocular drivers were safe drivers.”

6.3) The Ontario Studies: 1995, 2006 and 2017

[144] Ontario has undertaken its own reviews of available data on driver safety. All of the Ontario studies that were the subject of comment and review by the experts in this case were conducted after the introduction of the Regulations in 1994.[^5] Ontario maintains that because this is not a judicial review of Ontario’s decision to make the impugned Regulations (Di Cienzo v. Ontario (Attorney General, 2017 ONSC 1351, at para. 30) but is rather an application for a declaration that those Regulations are unconstitutional, it is entitled to rely on the best evidence currently available to defend its laws. See Irwin Toy v. Quebec, 1989 87 (SCC), [1989] 1 S.C.R. 927, at p. 984; R. v. Butler, 1992 124 (SCC), [1992] 1 S.C.R. 452, at pp. 494-496. Conversely, the Applicant is also entitled to rely on the best evidence currently available in her challenge of these laws.[^6]

The 1995 MTO Report

[145] The 1995 MTO Report included more than the literature review. It also included an assessment of individuals who were “legacied” into commercial driving licences, even though they fell below the minimum medical standards. The 1995 MTO Report reviewed data in 1990, 1991, and 1992. The 880 monocular drivers had 31.4 collisions per 100 drivers, the other binocular drivers with a medical condition had 31.1 collisions per 100 drivers, and all other Class A drivers had 31.3 collisions per 100 drivers. There was very little difference between the groups.

[146] The 1995 MTO Report acknowledged that only drivers with two years of a clean driving record were waived in when the medical and vision requirements changed in 1978, which could bias the group in favour of safer drivers. The Engel-Townsend Report critiqued the data, as the collisions were attached to the driver, rather than by the vehicle type, so it was not clear how many of the collisions happened in heavy vehicles. The Engel-Townsend Report also critiqued the 1995 data as it had limited exposure data, relying strictly on the number of drivers, rather than kilometers driven. This was the case in Ontario’s later studies as well.

The 2006 Review and 2017 Report Regarding the Waiver Program

[147] The 2006 Review and the 2017 Report are analyses of drivers who participated in Ontario’s commercial driver vision waiver program (“Waiver Program”). The analyses were conducted by employees of the Ministry of Transportation. Leo Tasca conducted the 2006 Review, and Chris Janusz conducted the 2017 Report. Janusz testified about the 2006 Review based on information and belief from Tasca, but he also familiarized himself with the 2006 Review in order to prepare the 2017 Report.

[148] The 2006 Review analyzed data for Waiver Program participants from 1999-2004. At the time of the report, there were 401 Class A and Class D drivers in the program. The 2006 Review used the “Relative Risk Analysis Method” to determine risk. The 2006 Review found that the road travel risk of collision for the Waiver Program participants was 2.3–3.5 times higher than that of drivers not in the Waiver Program. It is important to note that this is not a 2.3–3.5 times higher rate of past collisions. Rather, it is supposed to be a predictive metric of risk.

[149] The 2006 Review explained that Relative Risk is more than a bare numerical comparison. Appendix A of the 2006 Review points the reader to a 1998 paper by D.E. Stewart to explain how Relative Risk is calculated. The Stewart paper explains that calculating Relative Risk involves the “human, vehicle, road/infrastructure, environment, trip and temporal factors and their relative characteristics.” The combination of these factors, input into Stewart’s formula, is said to demonstrate Relative Risk. The 2006 Review lists these factors but does not describe or explain any of them for the data in question. It is not clear how the factors interacted to calculate Relative Risk in the 2006 Review. It is a multi-variable calculation, but the 2006 Review does not provide those variables – only the result.

[150] The 2017 Report involved a simple numerical rate of collisions per 10,000 drivers. The numerical data from the 2017 Report (which includes 1999-2004, the years of the 2006 Review) does not show any year in which the numerical collision rate approaches the Relative Risk ratio from the 2006 Review. In cross-examination, Mr. Janusz stated that he could have repeated the Relative Risk analysis, but did not do so because it was no longer the common practice in his office.

[151] In the 2017 Report, Mr. Janusz calculated the number of collisions per 10,000 drivers for each year. As the Waiver Program had fewer than 10,000 drivers, Mr. Janusz extrapolated upwards to establish comparable rates. He took the total number of collisions divided by the total number of driver years across the entire period of 1997-2015 and divided them into distinct age categories.

[152] For comparison, Mr. Janusz took the total number of Class A and Class D drivers in 2000 and 2005, added them together, considered the total number of collisions in 2000 and 2005, and divided number of collisions by number of drivers. This was expressed as MVC involvement rate per 10,000 licensed drivers. Mr. Janusz compared the collision rate of that group against the collision rate of the Waiver Program drivers. Both groups were divided into categories by age.

[153] Although the data sets for the comparator groups against which the Waiver Program drivers were being measured are not directly comparable because the statistics for the comparator group in the 2017 Report were averaged, the linear measurement of collisions per 10,000 drivers does not appear to replicate the results of the 2006 Review that suggested that the drivers in the Waiver Program were 2.3–3.5 times riskier. The 2017 Report indicates a smaller increased risk factor for the drivers in the Waiver Program, of 1.53 (drivers aged 16-29), 1.40 (drivers aged 30-39), 1.51 (drivers aged 40-49) and 1.64 (drivers aged 50+) times as many crashes.

7. Findings Arising From the Expert Evidence and Reviews

[154] Having considered the expert evidence and the studies they considered, there are some relevant findings that are important to the analysis that follows.

7.1) Impacts of Monocularity

[155] Monocularity in this context is referring to a person who has only one eye.

Visual Field

[156] Dr. Owsley was the only outlier, among the experts who addressed this topic, who would not agree that there is a standard expected monocular visual field of 150 degrees or higher. According to Drs. Peli, Smiley and Arshinoff, it is generally accepted that the normal binocular visual field is approximately 200-220 degrees and that the normal monocular field is about 155-170 degrees. This is supported by the earlier editions of the authoritative text in this area: David O. Harrington, The Visual Fields: A Textbook and Atlas of Clinical Perimetry, 4th ed. (C.V. Mosby Company, 1976). A more recent edition of this Harrington text, 6th ed. (C.V. Mosby Company; 1990) gives a nuanced and less determinative standardized monocular and binocular visual field projection but still depicts a normal monocular visual field of at least 150 degrees.

[157] There is consensus among the experts that there is no recognized or readily available test to measure the full visual field of a monocular person.[^7] The limitations inherent in the standard tests produce monocular visual field measurements of less than 150 degrees.

Loss of Stereopsis and Binocular Summation

[158] Dr. Peli reviewed other physiological impacts of monocularity, beyond the loss of some visual field, including loss of stereoscopic vision and lack of binocular summation. There is no evidence that either of these impacts are associated with driving safety. There are other medical conditions known to cause stereoscopic vision and lack of binocular summation and drivers with those conditions are not restricted from obtaining commercial drivers’ licences under the Regulations.

7.2) Driver Collision Rates vs. Driver Performance Studies

[159] Monocular drivers referred to in the studies include persons with only one eye, but these studies might also include in the monocular driver groups persons with two eyes, one of which is severely impaired. The degree of impairment in the relevant studies is variable.

[160] The driver collision rate studies indicate an increased MVC rate for the population of monocular drivers, whereas the driver performance studies suggest no material difference in the ability of monocular drivers to perform driving tasks safely. It is not clear that one or the other approach is the right one to assess driver safety, to the exclusion of the other. Dr. Owsley’s evidence that driver collision rates are the only established predictive measure of risk is challenged by her peer, Dr. Peli, who says that there is no evidence that monocularity affects the ability of individual drivers to perform driving tasks safely and also no evidence that the vision standards in the Regulations correlate to the capabilities of individual monocular drivers or to the increased MVC rates in the population of monocular drivers.

Driver Collision Studies: MVC Rates

[161] None of the studies are perfect. Each has potential limitations, but the Rogers study has been most commonly accepted and its limitations have been justified or explained. It indicates that monocular drivers have an increased MVC rate of 1.46 times that of non-monocular drivers. This is consistent with the low end of the range of statistics presented by Mr. Janusz from Ontario’s 2017 Report on the Waiver Program, which indicated an increased MVC rate ranging from 1.4 to 1.64. I accept the low end of the range from the 2017 Report, having regard to Mr. Janusz’ testimony that some of the statistical biases present in the results of Ontario’s 2006 Review (indicating higher MVC rates) were addressed in the 2017 Report, and also in order to account for other possible numerical trends (towards higher MVC rates) that were acknowledged by Mr. Janusz in the statistics presented, even though he did not consider their effects to be material.

[162] There are other studies indicating MVC rates on either end of the spectrum that I have not accepted. The acknowledged biases and other criticisms of Ontario’s 2006 Review of the Waiver Program call into question its conclusion indicating a higher MVC rate for monocular drivers, of 2.3–3-5 times greater than the general population. On the other end, the Laberge-Nadeau study out of Quebec and US DOT study that suggest no statistically different MVC rate for monocular drivers are also open to criticism, for the reasons indicated by Dr. Owsley and others who reviewed those studies.

[163] I find, based on the evidence of Dr. Owsley and Mr. Janusz, that it was reasonable for Ontario to believe that monocular drivers as a group are more likely to be involved in a MVC than the general population of drivers, with an increased MVC rate in the range of 1.4 times. Dr. Peli does not offer a different opinion about the higher MVC rates for monocular drivers, but rather suggests that driver performance should also be considered in assessing whether monocular drivers are unsafe.

Driving Performance Studies

[164] Just because monocular drivers as a population may have a higher MVC rate, that does not mean that all monocular drivers are unsafe. According to Dr. Peli, there is very little evidence that driving with one eye limited the functionality of the driver of any vehicle. While the loss of an eye causes impairment, Dr. Peli opined that this impairment has not been causally linked to increased crash rates of monocular drivers.

[165] The driving performance studies on the whole are more task specific and subjective which make them harder to draw generalizations from. None specifically involved bus drivers. As Dr. Smiley testified, the driving tasks of a bus driver are different, and in some instances more demanding, because of the constant stopping and starting, lane changes and presence of passengers and pedestrians. The potential for a monocular bus driving having to turn their head rather than simply change their visual fixation point (for example, to check the side mirror on their restricted side, and then look forward again) could also lead to slower reaction times for time-pressured tasks.

[166] That said, the performance studies (McKnight, Troutbeck and Wood, and Racette) universally conclude that monocularity did not materially impact the ability of drivers to safely complete the driving tasks that they assessed. Dr. Smiley corroborates that it is possible to individually assess monocular drivers at certain tasks to determine their ability to drive safely. However, there is no universally accepted road test, simulator, occupational assessment, or functional assessment that has yet been validated as a reliable predictor of road safety, for bus drivers or for drivers generally.

7.3) The Applicant

[167] Based on Dr. Arshinoff’s test of her visual field, the Applicant argues that she has the 150 degrees of visual field required by s. 18(3)(b).

[168] Neither Dr. Owsley nor Ontario accept that the Applicant’s visual field is 150 degrees as measured by Dr. Arshinoff’s testing method. Even if the literature and general view is that monocular visual field is greater than what the Humphrey Visual Field Analyzer measures on each side, Ontario argued that the test he conducted was not a reliable measure of Ms. Di Cienzo’s field of vision. According to Ontario’s expert, Dr. Owsley, none of the three most commonly used perimetry machines could be used to functionally administer the Esterman test with the subject’s head also on an angle. Further, Ontario argues that not only was there no evidentiary basis for Dr. Arshinoff to change the way visual field is measured, but that Dr. Arshinoff did not even measure whether the Applicant actually did move her head 10-15 degrees as he had asked her to do.

[169] Dr. Arshinoff conceded that his test was unconventional and did not measure the Applicant’s visual field in a manner consistent with how visual field tests are typically conducted.

[170] Dr. Owsley acknowledged that the Applicant’s visual field is likely larger than the 126 degrees that the Humphrey Visual Field Analyzer measured based on a centered vertical meridian (fixation point) and its limited reference points. However, she was not satisfied that there is evidence that Ms. Di Cienzo’s visual field is in fact 150 degrees or more. Further, while she acknowledges that the Applicant’s field of vision is likely more than 75 degrees on the left (temporal) side, she did not agree with the Applicant’s predictions of visual field on the nasal side.

[171] Absent a scientific basis for the method by which Dr. Arshinoff conducted the non-conventional visual field test and given that he did not control for the circumstances under which the measurements were taken, I agree with Dr. Owsley that his test results are not a sufficient evidentiary basis upon which to make a finding about the measurement of Ms. Di Cienzo’s visual field. The inability to accurately measure a monocular person’s visual field fully and accurately is a function of the lack of readily available and accepted scientific techniques for doing so.

[172] To find that Ms. Di Cienzo’s visual field is 150 degrees or greater I would have to infer that it was based on the literature that says that the normal monocular visual field is 155-170 degrees. That is not specific to her and I am not prepared to draw that inference. I do not foreclose the possibility that Ms. Di Cienzo has a visual field of 150 degrees or more, but that is not something that can be determined on the record before me.

[173] Even if I had been persuaded that the Applicant’s visual field was 150 degrees or greater, she would face two further obstacles. First, she cannot meet s. 18(3)(a) because her weaker eye does not have visual acuity of 20/100. Second, a schematic reading of s. 18(3)(b) together with s. 18(4)(d) requires that the 150 degrees of visual field be evenly divided with at least 75 degrees on each side.

[174] Ms. Di Cienzo’s monocularity (having only one eye) has the effect of making it impossible for her to meet either of the requirements of visual field or acuity under s. 18(3) of the Regulations.

V. SECTION 15 ANALYSIS

1. Preliminary Questions

1.1) What is the impact of the Grismer case?

[175] The Applicant argues that the case ETFO et al. v. Her Majesty the Queen, 2019 ONSC 1308, 144 O.R. (3d) 347, unified the test for discrimination under human rights law and under s. 15 of the Charter. Therefore, British Columbia (Superintendent of Motor Vehicles) v. British Columbia (Council of Human Rights), 1999 646 (SCC), [1999] 3 S.C.R. 868 (“Grismer”) is apposite and determinative of the s. 15 issue in this case, even if not for s. 1. Prior to the full s. 15 analysis, I discuss the impacts of Grismer on this case.

[176] Terry Grismer was a mining truck driver with homonymous hemianopia (“H.H.”). H.H. is a serious visual condition. British Columbia required drivers to have a visual field of at least 120 degrees horizontally. British Columbia allowed some exceptions. However, individuals with H.H. always have a visual field less than 120 degrees, and they were always precluded from accessing the exception available to some drivers with less than 120 degrees of visual field. In effect, British Columbia had a blanket prohibition against drivers with H.H.

[177] In Grismer, the applicant established that some people with H.H. might be able to drive safely, and that was enough to overcome the alleged risk for the group of drivers as a whole. Although Mr. Grismer died shortly after the first instance decision, the Supreme Court of Canada ruled that British Columbia should have afforded him an opportunity to demonstrate his ability to be a safe driver despite his disability.

[178] The Applicant says that the evidence in her case is even better. Monocularity is not as severe a condition as H.H., the studies about monocularity are not consistent in their indications of risk, and there is now better evidence about alternatives to a blanket prohibition than the evidence of such that was available when the Grismer case was argued. Therefore, if refusing to allow Mr. Grismer an opportunity for an individualized assessment was discriminatory, not allowing the Applicant an opportunity for an individualized assessment must also be discriminatory.

[179] The Respondent argues that Grismer is not determinative, for three reasons. First, ETFO did not lower the threshold of s. 15 to that of the human rights law, but rather the opposite. Second, other, more recent cases have since warned against applying human rights law to s. 15 cases. Finally, Grismer was essentially a judicial review of the superintendent’s exercise of discretion not to issue a driver’s licence. It was decided on its own facts, and on a sparse evidentiary record. The Supreme Court of Canada specifically contemplated at para. 45 that “In another case, on other evidence, that burden [of proving that a blanket refusal without reasonable accommodation was reasonably necessary] might be met.”

[180] Under human rights law, once a distinction based on a ground is established, it is prima facie discriminatory. The burden then shifts to the alleged discriminator to establish that the distinction is not discriminatory. That is not the case in s. 15 Charter jurisprudence. The applicant must establish that the distinction is discriminatory. I agree with Ontario’s interpretation of ETFO. When challenging a law under the Ontario Human Rights Code, R.S.O. 1990, c. H.19, the Charter standard applies.

[181] Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567, very clearly stated that human rights cases do not assist the consideration of justifications of an otherwise discriminatory law. The “reasonable accommodation” standard in human rights law is conceptually distinct from the Oakes test (Hutterian Brethren, at paras. 65-71). For this reason alone, Grismer cannot be wholly dispositive of this case.

[182] Finally, as Ontario states, Grismer was decided on a relatively sparse record. The court has more recent evidence, and a greater amount of it, available on this record.

[183] Nonetheless, human rights cases are not without value. Andrews v. Law Society of British Columbia, 1989 2 (SCC), [1989] 1 S.C.R. 143 relied on previous discussion of discrimination in human rights cases, at pp. 173-175. More recently, in Quebec v. A, 2013 SCC 5, [2013] 1 S.C.R. 61, Abella J. referred to human rights cases in assessing what discrimination is and how it functions. At para. 355, Abella J. referred to the case of Nova Scotia (Workers’ Compensation Board) v. Martin, 2003 SCC 54, [2003] 2 S.C.R. 504, noting that Gonthier J. had relied on two human rights cases in understanding discrimination. Indeed, when discussing the meaning of arbitrary disadvantage in its factum, Ontario relies on McGill University Health Care (Montreal General Hospital) v. Syndicat des employés de l'Hôpital général de Montréal, 2007 SCC 4, [2007] 1 S.C.R. 161, itself a human rights case.

[184] Grismer is not dispositive of the issues in this case. Fundamentally, the evidentiary record in this case is significantly different and more expansive. However, Grismer can assist the court in understanding the meaning of discrimination and it can contribute to the s. 15 analysis in that way. It is not relevant to the s. 1 stage of the analysis.

1.2) Does the reasoned apprehension of harm question imbue the analysis at the s. 15 Charter breach stage as well as the section 1 stage?

[185] In oral argument, Ontario suggests that the “reasoned apprehension of harm” test should be applied at the s. 15 breach stage, not only when considering justification under s. 1 of the Charter.

[186] In support of this argument, Ontario put forth the case of Cochrane v. Ontario (Attorney General), 2008 ONCA 718, 92 O.R. (3d) 321, at paras. 25-30 and suggested that its reasoning could be applied in this case. An examination of why that analysis took place at the breach stage in Cochrane is instructive.

[187] Cochrane was a s. 7 case. This is a s. 15 case. This is an important distinction.

[188] Under s. 7 of the Charter, a law cannot deprive an individual of life, liberty or security of the person except in accordance with the principles of fundamental justice. Three of the principles of fundamental justice, namely arbitrariness, overbreadth, and gross disproportionality, are analogs to the different components of the Oakes test, being rational connection, minimal impairment, and proportionality. Thus, in a s. 7 Charter analysis, the court is necessarily applying Oakes considerations at the breach stage and at the s. 1 stage. See Cochrane at para. 30:

I disagree with the appellant's submission that this reasoning applies only at the minimal impairment stage of s. 1 and that the application judge erred by applying it to determine whether there had been a violation of s. 7 on account of overbreadth. First, as a matter of authority, the Supreme Court of Canada appears to have assimilated the minimal impairment analysis under s. 1 with the overbreadth analysis under s. 7: see, e.g., R. v. Clay, 2003 SCC 75, [2003] 3 S.C.R. 735, [2003] S.C.J. No. 80, at para. 35; R. v. Nova Scotia Pharmaceutical Society, 1992 72 (SCC), [1992] 2 S.C.R. 606, [1992] S.C.J. No. 67, at p. 629 S.C.R. Second, as a matter of principle, I fail to see why the determination of whether legislation is overbroad and therefore in violation of s. 7 should involve a more stringent test than the test to determine whether minimal impairment has been satisfied.

[189] Cochrane relied on a number of other s. 7 cases, to which the same consideration will apply, and two cases from the s. 2(b) context: Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827, and Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927. Harper considered the reasoned apprehension of harm in the s. 1 analysis. Irwin Toy, while not considering that particular phrase, likewise considered deference to legislature’s goals of protecting society in the s. 1 analysis.

[190] In Charter cases that are not s. 7 cases, the principles of fundamental justice, such as overbreadth, do not arise at the breach stage. In this s. 15 case, I find that the question of reasonable apprehension of harm is appropriately analyzed when the court considers the question of minimal impairment under s. 1, but not as part of the s. 15 analysis.

2. Reviewing section 15 jurisprudence: Kapp, Withler, Quebec v. A, Taypotat, and the Pay Equity cases.

2.1) The Jurisprudential Background

[191] The Supreme Court of Canada altered the structure of the test for a violation of s. 15(1) of the Charter in R. v. Kapp, 2008 SCC 41, [2008] 2 S.C.R. 483. In Withler v. Canada (Attorney General), 2011 SCC 12, [2011] 1 S.C.R. 396, the Supreme Court reaffirmed the two-part test formulated in Kapp:

a. Does the law create a distinction based on an enumerated or analogous ground?

b. Does the distinction create a disadvantage by perpetuating prejudice or stereotyping?

[192] To apply the law, first I must try to consolidate and reconcile the various iterations of the s. 15 test that the parties have referred to, various parts of which they each selectively rely upon.

[193] In the previous jurisprudence, the Supreme Court identified contextual guiding factors to be considered in deciding whether a distinction was arbitrary. These included:

a. Any pre-existing disadvantage of the claimant group;

b. The degree of correspondence between the differential treatment and the claimant group’s actual needs, capacities and circumstances;

c. The ameliorative impact or purpose of the law; and

d. The nature of the interests affected.

Kapp, at para. 19 and Law v. Canada (Minister of Employment and Immigration), 1999 675 (SCC), [1999] 1 S.C.R. 497, at paras. 62-75, 88.

[194] In the discussion of how to establish a s. 15 claim, Withler grouped or aligned these contextual factors based on the type of s. 15 claim being advanced. Withler discussed how to establish a s. 15 claim at para. 38:

Without attempting to limit the factors that may be useful in assessing a claim of discrimination, it can be said that where the discriminatory effect is said to be the perpetuation of disadvantage or prejudice, evidence that goes to establishing a claimant’s historical position of disadvantage or to demonstrating existing prejudice against the claimant group, as well as the nature of the interest that is affected, will be considered. Where the claim is that a law is based on stereotyped views of the claimant group, the issue will be whether there is correspondence with the claimants’ actual characteristics or circumstances. Where the impugned law is part of a larger benefits scheme, as it is here, the ameliorative effect of the law on others and the multiplicity of interests it attempts to balance will also colour the discrimination analysis.

[195] Withler added that those factors did not need to all be assessed in every case, and that other contextual factors beyond these ones could also be helpful in a given case (at para. 66). Withler also eliminated the requirement of a comparator group (at paras. 40, 55-60). At paras. 71-79, the Supreme Court approvingly reviewed the contextual analysis of the trial judge that considered the larger benefits scheme that was at issue.

2.2) Quebec v. A and Taypotat

[196] Quebec v. A is a complicated decision. In terms of the decision’s effect, upholding the law, Lebel J. is the “majority”. In terms of the s. 15 analysis, Abella J. is the majority writer.

a. Lebel J. (with Moldaver, Fish and Rothstein J.J.) found no violation of s. 15.

b. Abella J. found a violation of s. 15 not justified by s. 1.

c. Deschamps J. (with Karakatsanis and Cromwell J.J.) wrote brief, slightly narrower concurring reasons agreeing with Abella J.’s s. 15 analysis.

d. McLachlin C.J. was the swing vote. She found a s. 15 violation and agreed with Abella J.’s s. 15 analysis (Quebec v. A, at para. 416, McLachlin C.J. reasons). McLachlin C.J. alone found that the violation was justified by s. 1.

[197] Abella J. reviewed the history of s. 15, including Kapp and the introduction of stereotype and prejudice (Quebec v. A, at paras. 319-332, Abella J. reasons). Abella J. did not eliminate prejudice and stereotype from the s. 15 test. However, she was clear that they were not requirements that must be established by the claimant (at para. 327, Abella J. reasons). Requiring prejudice or stereotype “improperly focuses attention on whether a discriminatory attitude exists, not a discriminatory impact, contrary to Andrews, Kapp and Withler.”

[198] The test, as stated in Kapp and Withler (and relied on by both parties), must be considered with regard to the broadened analysis articulated by Abella J.: “If the state conduct widens the gap between the historically disadvantaged group and the rest of society rather than narrowing it, then it is discriminatory.” (Quebec v. A, at para. 332). Justice Abella harkened back to Andrews, where the basic goal is “substantive equality.” (at para. 333). She specifically reiterated that there was “only one question: Does the challenged law violate the norm of substantive equality in s. 15(1) of the Charter?” (at para. 325) The breadth of this question is perhaps unhelpful from an analytical perspective, but it prevents an unduly formal “test” which can too easily reject claims at the s. 15 stage.

[199] Justice Abella referred to “contextual factors” but cited the paragraph from Withler which referred to “contextual factors” in a general sense (Quebec v. A, at para. 331). The citation from Withler indicated that s. 15 is generally a contextual question, that the four factors from Law were not required in every case, and that other factors could arise (see para. 331, citing Withler, at para. 66). Chief Justice McLachlin, meanwhile, in reiterating that s. 15 is a “contextual analysis”, specifically did refer to the four factors from Law (see para. 418, citing Kapp and Withler).

[200] In Kahkewistahaw First Nation v. Taypotat, 2015 SCC 30, [2015] 2 S.C.R. 548, Abella J. wrote for the unanimous Court. Consistent with her approach in Quebec v. A, Abella J. focused s. 15 on “laws that draw discriminatory distinctions – that is, distinctions that have the effect of perpetuating arbitrary disadvantage based on an individual’s membership in an enumerated or analogous group.” (See Taypotat, at para. 18, emphasis in the original.)

[201] However, while focusing the essence of the s. 15 analysis on that fundamental inquiry, Abella J. did continue to uphold the two-part analysis in Taypotat at paras. 19-20. Part one asked if “on its face or in its impact, a law creates a distinction on the basis of an enumerated or analogous ground.” The second part “focuses on arbitrary – or discriminatory – disadvantage, that is, whether the impugned law fails to respond to the actual capacities and needs of the members of the group…” This is similar language to one of the contextual factors from Law. Prejudice and stereotype are not considered in Taypotat – simply whether a law causes arbitrary, or discriminatory disadvantage.

[202] In the right context, the arbitrariness of a disadvantage remains a key consideration. See Taypotat, at paras. 16-18 and Christian Medical and Dental Society of Canada v. College of Physicians and Surgeons of Ontario, 2019 ONCA 393, 147 O.R. (3d) 444, at paras. 87-88. In this case, where the Applicant makes arguments of stereotype and the Respondent makes arguments of legitimate correspondence, the arbitrariness is a relevant consideration. However, as the Supreme Court of Canada’s most recent jurisprudence indicates, that is not a necessary component.

2.3) The 2018 pay equity companion cases

[203] Justice Abella authored two more s. 15 cases in 2018, writing for the majority both times.

[204] In the first case, Quebec (Attorney General) v. Alliance du personnel professionnel et technique de la santé et des services sociaux, 2018 SCC 17, [2018] 1 S.C.R. 464, (Pay Equity 1), Abella J. briefly conducted step 1 of the s. 15(1) analysis, a distinction based on a ground, at paras. 26-27. At para. 28, she briefly reviewed the previous case law, reiterated the Supreme Court’s movement away from a step-by-step analysis of Law’s four contextual factors, and focused the inquiry on the impact of the distinction.

[205] At paras. 33-42, Abella J. performed step 2 of the s. 15(1) analysis. She did not consider any specific contextual factor. She did not mention the words prejudice or stereotype. She did not mention the claimant’s “needs and capacities.” Justice Abella simply considered the distinction based on a ground, its impact on the claimant, and asked if it perpetuates or exacerbates their disadvantage. In this case, Abella J. did not even ask about the arbitrariness of the disadvantage. In Pay Equity 1, Abella J. found a s. 15 breach that could not be justified under s. 1.

[206] In the second case, Centrale des syndicats du Québec v. Quebec (Attorney General), 2018 SCC 18, [2018] 1 S.C.R. 522, (Pay Equity 2), the analysis took a similar structure. Justice Abella found that step 1 of the s. 15(1) test was satisfied: there was a distinction based on a ground (paras. 24-29). Abella J. asked only one question: “whether the distinction is a discriminatory one, that is, whether it imposes burdens or denies benefits in a way that reinforces, perpetuates, or exacerbates disadvantage.” At paras. 30-36, she found that it did. Justice Abella did not examine Law’s contextual factors, did not mention prejudice or stereotype, and did not consider arbitrariness. In Pay Equity 2, the breach was justified under s. 1.

3. The Section 15 Issues and Analysis

[207] The test to demonstrate a violation of s. 15 has two parts. First, the law must draw a distinction based on an enumerated or analogous ground. Second, that distinction must be discriminatory. The distinction is discriminatory if it “imposes burdens or benefits in a way that reinforces, perpetuates, or exacerbates disadvantage” (Pay Equity 2, at para. 30). The onus is on the Applicant to demonstrate a violation of s. 15.

3.1) Characterizing the Applicant’s Claims

[208] The Applicant’s claims must be characterized before applying the s. 15 test. Does the visual acuity standard in s. 18(3)(a) discriminate against monocular drivers? How should this court characterize the Applicant’s challenge to s. 18(3)(b)? Does the visual field standard in s. 18(3)(b) discriminate against monocular drivers?

Section 18(3)(a)

[209] The Applicant argues that s. 18(3)(a) violates s. 15 of the Charter, as it directly prohibits monocular drivers since they can never meet the minimum acuity standard in their weaker eye.

Section 18(3)(b)

[210] The Applicant’s arguments about the visual field requirements of s. 18(3)(b) are not as straightforward because of the theoretical probability that a monocular person has 150 degrees of visual field. The challenge arises from the manner in which the visual field is measured:

a. If Ms. Di Cienzo’s visual field had been established to meet the 150-degree visual field requirement of s. 18(3)(b), her challenge to s. 18(3)(b) would have to be considered in the broader context of the requirement under s. 18(4) that she demonstrate at least 75 degrees of horizontal visual field on each side of the vertical meridian (the central fixation point). This ties back to an earlier issue about the scope of the Applicant’s constitutional challenge. For the court to consider s. 18(4) when it was not referenced in her Notice of Constitutional Question, it must be on the basis that s. 18(4) is a necessary interpretive aid to understand the meaning and impact of s. 18(3)(b).

b. However, since Ms. Di Cienzo’s visual field has not been demonstrated to meet the 150-degree visual field requirements, the court must consider whether a monocular person ever could meet that visual field requirement if there is no accepted methodology that permits a monocular person’s field of vision to be measured in a way that might allow them to demonstrate the full capacity of their visual field.

[211] Dr. Arshinoff’s unconventional administration of the Esterman test showed the Applicant to have 150 degrees of visual field. However, I do not accept the results of Dr. Arshinoff’s unconventional test. While it is open to the court to accept that an adaptation to a customary medical measurement to accommodate an individual could be accurate, Dr. Arshinoff did not carry out his unconventional test with any precision, on his own evidence. If I had accepted Dr. Arshinoff’s unconventional method, it would have been open to the court to engage in a contextual interpretation of s. 18(3)(b) by reading that section in light of how the visual field must be measured under s. 18(4). However, I do not need to go there.

[212] I accept the views of the preponderance of the experts that Ms. Di Cienzo’s visual field is very likely greater than the 126 degrees that was measured using the Humphrey Visual Field Analyzer. However, the evidentiary record does not support a finding that Ms. Di Cienzo’s visual field is, in fact, 150 degrees or greater. The experts agree that there is no accepted way to actually measure a monocular person’s true visual field, even though their consensus is that it theoretically should be at least 150 degrees if it could be measured. Based on the expert evidence, there does not appear to be any way for the Applicant, or any other monocular person, to demonstrate that they can pass the standard in s. 18(3)(b). The requirement of 150 degrees of visual field effectively excludes monocular drivers from ever satisfying the visual field requirement.

[213] The Applicant’s challenge under s. 18(3)(b) is to this bare requirement of a demonstrated 150 degrees of visual field. Framing this constitutional challenge of s. 18(3)(b) as a direct prohibition against monocular drivers brings it in line with the Applicant’s challenge to s. 18(3)(a).

3.2) Does s. 18 of the Regulations Establish a Distinction Based on a s. 15 Ground?

[214] I find that Ms. Di Cienzo’s monocularity has the effect of making it impossible for her to meet either of the requirements of visual acuity or visual field under s. 18(3) of the Regulations.

[215] Sections 18(3)(a) and (b) establish distinctions. Section 18(3)(a) distinguishes between individuals with visual acuity of 20/100 or better in their weaker eye, and individuals with worse visual acuity. Section 18(3)(b) distinguishes between individuals with a horizontal visual field of 150 degrees or more, and those who with a lesser horizontal visual field.

[216] This distinction is based on a ground. Facially, both sections establish minimum vision standards for individuals seeking to acquire commercial drivers’ licences. Functionally, the two sections establish a direct, blanket prohibition on monocular individuals. A person who is monocular has a physical disability. Physical disability is an enumerated ground under s. 15 of the Charter.

[217] Ontario concedes that s. 18(3) of the Regulations makes a distinction based on visual impairment, a physical disability that is a s. 15 Charter ground. Ontario concedes that the Regulations denies monocular persons a benefit or privilege available to others because of their disability. Ontario defends this on the basis that the Regulations do not single out monocularity, since persons with two functioning eyes but who have other conditions (such as glaucoma, scotomas, or poor visual acuity) may also not meet the vision standards. The fact that binocular persons also might not meet the vision requirements in the Regulations does not detract from its functionally wholesale prohibition against any monocular individuals.

[218] The Applicant has satisfied the first part of the s. 15 Charter test by demonstrating that the vision standards under the Regulations create a distinction on the basis of a protected enumerated ground.

3.3) Is the Distinction in the Regulations Substantively Discriminatory?

[219] Under the second part of the s. 15 Charter test, the Applicant still must satisfy the court that the distinction created by the Regulations is substantively discriminatory.

The Essence of Substantive Discrimination

[220] The broadened analysis articulated by Abella J. requires the court to consider “only one question: Does the challenged law violate the norm of substantive equality in s. 15(1) of the Charter?” (See Quebec v. A, at para. 325.) Discriminatory distinctions violate that norm.

[221] Three quotes provide some understanding of what discrimination means in this context:

I would say then that discrimination may be described as a distinction, whether intentional or not but based on grounds relating to personal characteristics of the individual or group, which has the effect of imposing burdens, obligations, or disadvantages on such individual or group not imposed upon others, or which withholds or limits access to opportunities, benefits, and advantages available to other members of society. (Andrews, at pp. 173-175.)

The focus of s. 15 is therefore on laws that draw discriminatory distinctions — that is, distinctions that have the effect of perpetuating arbitrary disadvantage based on an individual’s membership in an enumerated or analogous group: Andrews, at pp. 174-75; Quebec v. A, at para. 331. The s. 15(1) analysis is accordingly concerned with the social and economic context in which a claim of inequality arises, and with the effects of the challenged law or action on the claimant group. (Taypotat, at para. 18.)

This case is not about whether unsafe drivers must be allowed to drive. There is no suggestion that a visually impaired driver should be licensed unless she or he can compensate for the impairment and drive safely. Rather, this case is about whether, on the evidence before the Member, Mr. Grismer should have been given a chance to prove through an individual assessment that he could drive. It is also about combatting false assumptions regarding the effects of disabilities on individual capacities. All too often, persons with disabilities are assumed to be unable to accomplish certain tasks based on the experience of able-bodied individuals. (Grismer, at para. 2.)

Overview of the parties’ arguments:

[222] The parties’ arguments for the most part apply to both ss. 18(3)(a) and (b). Unless an argument specifically pertains to visual field or to visual acuity, the arguments about the constitutionality should be read as applying to both sections.

[223] Some of their arguments, particularly those based on correspondence and arbitrariness, overlap between arguments made in s. 15 and in s. 1. On occasion the parties both have referred to these considerations under both ss. 1 and 15. Where the arguments assist with assessing arbitrary disadvantage, I have included them in the s. 15 analysis. Where they relate solely to s. 1 issues (such as the precautionary principle and reasoned apprehension of harm), they are confined to that analysis.

The Applicant’s position

[224] Applicant makes several arguments about why the direct prohibition on monocular drivers is discriminatory. Although the Applicant has argued that she is a member of a historically disadvantaged group, based on Canada’s history of exclusion and marginalization of disabled persons (See Eldridge v. British Columbia (Attorney General), [1997] 3. S.C.R. 624, at para. 56), the Applicant’s submissions were focussed on demonstrating that the vision standards in the Regulations perpetuate an arbitrary disadvantage and stereotype that lacks correspondence to the actual capacities of individual monocular drivers.

[225] The Applicant’s arguments refer to various evidentiary contentions, including:

a. The Applicant challenges the premise that the studies identified by the experts establish that monocular drivers as a group are less safe. Some scientific studies showed little, if any, increase in crash rate from monocular commercial drivers. She argues that Ontario’s studies and the Rogers study, which do show an increase in MVC rates, are unreliable.

b. The focus should be on the individual drivers and their functional abilities, and not on the group of monocular drivers. The scientific evidence is widely variable regarding the safety of monocular drivers, including within Ontario’s own Waiver Program.

c. The vision standards in the Regulations are themselves arbitrary and do not correlate to the safety of commercial drivers:

i. Dr. Owsley and Dr. Peli sat on a panel in 2008 which declined to recommend raising the minimum visual field for commercial truck drivers to even 120 degrees due to lack of evidentiary support, yet Ontario’s minimum is 150 degrees (for all commercial drivers’ licences).

ii. The Regulations simply prohibit monocular drivers and do not target or test for the actual visual impairment problems which accompany monocularity, which can also occur in binocular drivers with other conditions.

iii. Drivers whose vision is just slightly above the visual acuity standard could have worse overall visual acuity than monocular drivers due to binocular inhibition. These drivers were rated as similarly risky to monocular drivers in the Rogers study, yet are not precluded from obtaining a Class C licence by the Regulations.

d. Other provinces have regulations involving individual exceptions based on functional assessments.

[226] The Applicant argues that the legislation is based on and perpetuates a stereotype. “A stereotype may be described as a misconception whereby a person or, more often, a group is unfairly portrayed as possessing undesirable traits, or traits which the group, or at least some of its members, do not possess.” See Law, at para. 64. If monocular drivers are portrayed or viewed as having the undesirable trait of unsafe driving, but that does not accord with reality, then it is a discriminatory distinction.

[227] The Applicant argues that the focus should be on the capacities of the individuals. Here, there is denial of benefit based on membership in a group. The Applicant asserts this denial is based on an inaccurate perception of not only that group’s characteristics, but the characteristics and capabilities of the individual members of the group: a classic case of what s. 15 is intended to cover. This involves stereotype, as the judgment about the Applicant is not based on her actual characteristics. Focusing on the group is also arbitrary, as it ignores the significant variance within the group, per Andrews, at p. 174.

[228] The Applicant argues that the use of MVC rate studies and Ontario’s reliance on the “correspondence” contextual factor is misplaced. The Law case says, at para. 106, that correspondence needs to be more precise for an individual or group being excluded if they are already disadvantaged in society. Quebec v. A, at para. 333 explains the importance of the distinction between s. 15 and s. 1. By relying heavily on the correspondence factor, Ontario is using justificatory arguments in the s. 15 analysis.

The Respondent’s position

[229] Ontario puts forward several arguments in response, contending that correspondence need not be perfect and that it has behaved responsibly based on sufficient evidence and a reasonable belief that monocular drivers have an increased MVC rate and pose a safety risk. These arguments refer to various evidentiary contentions, including:

a. Previous scientific studies showed a significant increase in MVC rates of monocular commercial drivers.

b. Although not initially produced by Ontario, its two internal reports produced at the request of the Applicant – the 1995 MTO Report and the Engel-Townsend Report – demonstrate that Ontario took the issue seriously after settling a lawsuit in the early 1990s with a driver who was restricted by the vision standards. These reports recommended further study in the area.

c. That is exactly what Ontario did. It ran a Waiver Program for monocular commercial drivers to produce Ontario-specific data. Drivers in the Waiver Program had higher MVC rates and the program was discontinued.[^8]

d. Ontario is proceeding cautiously after cancelling its Waiver Program for monocular commercial drivers and is currently collecting information about whether functional assessments for monocular Class G drivers are effective.[^9]

e. There is currently no evidence to suggest that an in-clinic functional impairment assessment, or a road test is an effective or reliable predictor of safety. Ontario is not taking any chances with Class C bus drivers in its ongoing assessments, where there is an enhanced concern for public safety, until there is adequate data.

f. Driving a bus is much harder than driving a car.[^10]

[230] Ontario maintains that drawing a line at some point when creating regulatory schemes is necessary. Regulatory line-drawing which cuts off some individuals is not necessarily discriminatory.

[231] Ontario maintains that its approach to the Class C licence has been consistently driven by public safety concerns. While its effect may disadvantage monocular drivers, this disadvantage is not arbitrary.

Correspondence of the Regulations with the characteristics of the group

[232] Ontario argues that the prohibition on monocular drivers is consistent with studies demonstrating they are unsafe drivers, and therefore it accurately corresponds with their abilities as a group. It argues that its vision standards in the Regulations were based on informed generalizations that led to the drawing of appropriate lines. It says that correspondence between those lines and the affected group’s capacity does not need to be exact or perfect.

[233] Ontario maintains that studies, such as the Rogers study relied upon by Dr. Owsley and the 2017 Report authored by Mr. Janusz, demonstrate an elevated collision risk for monocular commercial drivers and provide a basis for its belief that persons who do not meet the vision standards in the Regulations will have a higher MVC rate. I have found that these studies do establish that it was reasonable for Ontario to believe that monocular drivers as a group are approximately 1.4 times more likely to be involved in a MVC than the general population of drivers.

[234] However, the reasonableness of Ontario’s belief regarding the increased MVC rate of monocular drivers does not end this inquiry. The fact that there may be some scientific basis for a belief that monocular commercial drivers have increased MVC rates is, at best, support for a finding of imperfect correspondence between the vision standards in the Regulations and the notion that monocular drivers are not safe drivers of commercial vehicles.

[235] The Applicant argues that Ontario is using monocularity as a proxy for road safety. None of the experts go so far as to say that monocular drivers, as a group, are not able to drive a commercial vehicle safely. They only speak to the increased MVC rate of monocular drivers of commercial vehicles and the enhanced challenges of driving a bus.

[236] The Applicant points out that some studies (including Ontario’s own 1995 MTO Report) showed no statistically significant difference in the MVC rates of monocular commercial drivers and others (such as the 1998 Engel-Townsend Report) only go so far as to suggest that there is “some evidence” that monocular drivers have a higher rate of collision than binocular drivers do and that “monocular drivers may have a performance disadvantage that warrants some degree of special treatment.” That special treatment could be established through the introduction of an exemption based on driver performance evaluations.

[237] Dr. Peli opined that MVC rates do not translate into a correlation with individual monocular driver safety and that driver performance should also be considered in assessing whether monocular drivers are unsafe. Dr. Owsley’s perspective is too narrow in her insistence that driver performance cannot contribute to the assessment of driver safety and, to that extent, having regard to the challenges to her evidence raised by the Applicant (discussed earlier in these reasons) I prefer and accept the broader perspective offered by Dr. Peli.

[238] This dovetails with the remedy the Applicant seeks. She seeks an opportunity to individually demonstrate she is a safe driver and can drive a commercial vehicle, specifically, a public transit bus, safely.

Correspondence of the Regulations with the characteristics of the individuals

[239] Ontario accepts as a starting premise that: “The correspondence, or lack thereof, between the ground or grounds on which the claim is based and the actual need, capacity, or circumstances of the claimant or others”, referred to in Law, at paras. 69-71 and 88, remains an important contextual factor to consider. See also Quebec v. A, at paras. 165-166, 331, and 418. However, it argues that regulatory line-drawing which cuts off some individuals is not necessarily discriminatory. Legislation necessarily involves multiple considerations and competing interests. It does not need to perfectly correspond to each individual’s distinct, unique needs to be constitutionally valid. In R. v. T.M.B., 2013 ONSC 4019, 299 C.C.C. (3d) 493, at para. 62, Justice Code confirmed, relying on Canadian Foundation for Children, Youth and the Law v. Canada (Attorney General), 2004 SCC 4, [2004] 1 S.C.R. 76, that this principle – the adequacy of less than perfect correspondence – applied outside the contexts of benefits schemes.

[240] Ontario argues that performance assessments must be validated to be considered evidence-based for them to be relied upon a legitimate way to assess driver safety. Ontario maintains that there is no evidence-based approach to individualized assessment, and absent such, there is no safe way to licence drivers like the Applicant.

[241] The Applicant counters that the focus of the Regulations on the group, rather than on the individual, relies on and perpetuates stereotypes about drivers with a physical disability, and ignores specific evidence about the widely varying safety of monocular drivers, including in Ontario’s own Waiver Program.

[242] Individuals within a group who share the same characteristics will vary significantly. The Applicant relies on Andrews, at p. 174: it is discrimination if laws are made about a group relying on traits which the group or some of its members do not possess.

[243] As a general premise, it is uncontroversial that where there is statistical mean data about a group, members of that group will vary both above and below that statistical mean. Applying that premise, where there is an average collision rate for monocular drivers, it is likely that some members of the group had fewer collisions and some had more collisions. It is also likely that some of the group will continue to have fewer and some will continue to have more collisions.

[244] The Engel-Townsend Report observed that the same is true for driver performance data, which is known to have broad distributions with large overlaps between the distributions for different driver groups. The Engel-Townsend Report reviewed an analysis of a distribution of driving test scores for the G2 test, comparing novice drivers with less than one year of experience to professional drivers, each having ten or more years of professional driving experience. The professional drivers performed significantly better than the novice drivers, averaging a score 15 points lower than the novice drivers (the test is an error score; a lower score is desirable). However, the distribution data clearly shows that a substantial number of the professional drivers performed worse than a substantial number of the novice drivers.

[245] The Applicant analogizes this to the raw data produced by Ontario in this litigation, collected as a result of the Waiver Program. She argues that data demonstrates a similar trend: that some Waiver Program drivers were responsible for a disproportionate number of collisions. Mr. Janusz confirmed that this interpretation of the data was correct during cross-examination.

[246] The Applicant relies on driver performance studies, such as the McKnight study, to demonstrate that monocular commercial drivers are capable of driving safely and should be allowed to demonstrate their individual capabilities, consistent with the approach taken in other jurisdictions in Canada and in the United States, which allow for exemptions for individual commercial drivers to be licensed even if they do not meet the vision standards in place in those jurisdictions.

[247] I have already found earlier in these reasons that the driving performance studies on the whole are more task specific and subjective which make them harder to draw generalizations from, and none specifically involved bus drivers. As Dr. Smiley testified, the driving tasks of a bus driver are different, and in some instances more demanding, because of the constant stopping and starting, lane changes and presence of passengers and pedestrians.

[248] That said, the performance studies (McKnight, Troutbeck and Wood, and Racette) universally concluded that monocularity did not materially impact the ability of drivers to safely complete the driving tasks that they assessed. Dr. Smiley corroborated that it is possible to individually assess monocular drivers at certain tasks to determine their ability to drive safely, although reaction times must also be taken into consideration.

[249] These studies do not go so far as to establish, as the Applicant asserts, that monocular commercial drivers are safe drivers, but they do hold out the possibility that individual drivers within this group could be shown to be safe drivers on an individual basis and support Dr. Peli’s opinions to that effect, which I accept and prefer to the more narrow views expressed by Dr. Owsley (as noted previously).

Do the Regulations infringe substantive equality because the lines they draw are themselves arbitrary?

[250] The Applicant argues that the vision standards in the Regulations are themselves arbitrary and therefore infringe substantive equality. I have outlined some of the main arguments about arbitrariness below, most of which are based on an absence of evidence. Ultimately, while there may be some merit to these points, I do not need to decide the s. 15 Charter breach on the basis of these arguments and I decline to make findings based upon them as I am not satisfied that the record on these points has been fully developed.

Correlation of the Vision Standards with Driver Safety or Performance Statistics

[251] Dr. Peli also opines that the specific vision standards in the Regulations for visual acuity and visual field have not been shown scientifically to correlate in a causal way with the increased MVC rates for monocular drivers as a group, nor have these vision standards been shown to correlate with the capabilities of individual drivers within the group.

[252] The Applicant also points to the 2008 Expert Panel comprised of Dr. Owsley and Peli and one other panel member, Dr. Benson. One of the questions the panel considered was whether the standard for commercial truck drivers in the United States should be increased from 70 degrees of visual field in each eye to 120 degrees of visual field. The panel stated that there was insufficient evidence upon which to recommend making that change to increase the visual field requirement.

Vision Standards Fail to Capture Other Groups for Whom the Same Safety Concerns Should Apply

[253] The Applicant also contends that the vision standard is arbitrary because it does not test for two of the important differences in monocular vision, stereopsis and binocular inhibition. Stereopsis can assist with depth perception. As explained by Dr. Peli, with binocular inhibition (the converse of binocular summation), a person who has a little bit of vision (enough to meet the required 20/100 in the weaker eye) could actually have worse visual acuity overall than if the individual simply had no vision in the weaker eye.

[254] Poor depth perception and binocular inhibition can be caused by conditions other than monocular vision. Essentially, the Applicant argues that if Ontario was so concerned about the safety difference resulting from vision deficit, it would account for these vision deficits in binocular commercial drivers, which it does not. As the Regulations do not account for the actual vision deficits, but rather only ban monocular drivers, they are arbitrary.

[255] The Applicant also argues that the standard is arbitrary because it does not correlate with functional abilities. Some drivers who meet the visual field or visual acuity standards may have less functional driving ability than the Applicant, who does not meet it. This is a function of the driver safety vs. driver performance debate among the experts, noted above.

[256] The Applicant points to the Rogers study, relied upon to demonstrate that there is a statistical difference in their abilities to drive safely, as another example of arbitrariness. Based on the structure of the Rogers study’s groups, an individual could fall within the category of moderately visually impaired but still have sufficient vision to qualify as a commercial driver under s. 18(3). In the Rogers study, the moderately visually impaired individuals had similarly elevated collision rates to the severely impaired group (which included monocular drivers). The Applicant argues this demonstrates the arbitrariness of the lines drawn by the Regulations.

[257] As Ontario notes, pointing out that other individuals who are slightly above the vision standards may also have elevated collision rates, or that there are other groups who have elevated collision risks who are not excluded, does not render the vision standards in sections 18(3)(a) and (b) arbitrary. The fact that one group is not included within the ambit of a regulatory scheme does not mean that the inclusion of another group is arbitrary.

3.4) Conclusion: Do the Vision Standards of the Regulations Violate s. 15 of the Charter?

[258] Sections 18(3)(a) and (b) openly draw distinctions: that is their purpose, as components of a regulatory scheme. The distinction is based on a ground, physical disability. At the second stage of the s. 15 test, the Supreme Court of Canada’s most recent jurisprudence directs the court to scrutinize the legislation for whether it “imposes burdens or denies benefits in a way that reinforces, perpetuates, or exacerbates disadvantage” (Pay Equity 2, at para. 30) and to consider more broadly whether the challenged law violates the norm of substantive equality (see Quebec v. A, at para. 325). I find that s. 18(3) of the Regulations does violate the norm of substantive equality.

[259] The vision standards in s. 18(3) clearly deny a benefit or privilege to monocular individuals. I could infer that this, in turn, exacerbates the disadvantage that physically disabled individuals experience generally. This could be one ground upon which I might find there to be s. 15 Charter discrimination in the Regulations.

[260] Withler directs that the “nature of the interests affected” may be a relevant contextual factor to consider within this category of discrimination against a disadvantaged group. Ontario argues that the Applicant’s interests are less important because she is not entirely disqualified from driving (she has her G licence and can drive a passenger vehicle), suggesting that her interest in earning her livelihood as a bus driver is less important. Regardless of their relative importance, I find that the Applicant’s interest in preserving her ability to work and earn her livelihood in her chosen profession is worthy of Charter protection. As stated by Dickson C.J. (dissenting) in Reference Re Public Service Employee Relations Act (Alta.), 1987 88 (SCC), [1987] 1 S.C.R. 313: “Work is one of the most fundamental aspects in a person's life, providing the individual with a means of financial support and, as importantly, a contributory role in society. A person's employment is an essential component of his or her sense of identity, self‑worth and emotional well‑being.” Justice Binnie, writing for a unanimous court, brought this quote into the s. 15 context in Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381, at para. 40.

[261] I prefer to ground my analysis and finding of s. 15 Charter discrimination in the source of the Regulations, which are based on Ontario’s belief or view that monocular commercial drivers as a group are not safe. The Regulations are based on and perpetuate a stereotype about the abilities of monocular drivers as a group to safely drive commercial vehicles, including public transit buses. The Supreme Court’s jurisprudence is clear that where the stereotype (that monocular commercial drivers are unsafe) does not correspond with the actual characteristics and capacities of individuals within the group, that perpetuates a discriminatory disadvantage. (See Taypotat and Law.)

[262] There is some evidence found in the studies (Rogers, Keeney, and Ontario’s 2017 Review) to support Ontario’s belief that monocular commercial drivers as a group have a higher MVC rate. That does not translate into perfect, or even close to perfect correspondence with the capacity of the Applicant and other monocular individuals to drive a commercial transit bus safely.

[263] Ontario argues that the restrictions correspond, or at least could reasonably be believed to correspond, to the Applicant’s capacities. However, the expert evidence and studies reviewed overall do not support, and in some instances contradict, the supposition that there would be a reasonable level of correspondence with the safe-driving capacities of each of the individuals within the group. Ontario’s own data suggests that there is significant variance within that group. The Applicant herself has a safe-driving record, including as a driver of a passenger vehicle with her class G licence since she became monocular.

[264] This lack of correspondence to individual safe-driving capacities and characteristics of monocular commercial drivers leads me to conclude that s. 18(3) of the Regulations creates an arbitrary disadvantage by perpetuating a stereotype. This violates the norm of substantive equality.

[265] Ontario does not wish to use performance assessments. Ontario believes that there are no conclusive, evidence-based, scientific studies validating performance assessments as a legitimate way to assess and predict driver safety. Ontario argues this justifies the group-based approach to the lines drawn by Ontario that prohibit all monocular drivers without any allowance for performance-based exceptions. I have a difficulty with this position.

[266] As the Applicant points out, it is circular for Ontario to say that there needs to be evidence that these performance assessments are predictors of safe driving on an individual basis without allowing the exemptions for these assessments so that they can be assessed. The fact that Ontario is studying this for the Class G licence holders, to whom the s. 18(3) vision standards do not apply, is not a reason to continue to foreclose the possibility of individual driver performance assessments for Class C licence holders to whom these vision standards do apply. This is more appropriately considered in the s. 1 analysis, but I mention it here because Ontario also raised it as part of its response to the alleged s. 15 breach.

[267] The Applicant seeks recognition as an individual. She seeks the opportunity to demonstrate whether she can drive safely or not. Denying her this opportunity to demonstrate that, despite her physical disability, she is capable of driving a transit bus safely, violates the norm of substantive equality and her s. 15 equality rights under the Charter. I therefore find that the vision standards under s. 18(3) of the Regulations violate s. 15 of the Charter.

VI. SECTION 1 ANALYSIS

[268] The parties agree that the Oakes test is the appropriate analytical framework for the s. 1 analysis. In s. 1, the onus is on the government, in this case Ontario, to establish that an infringement of the Charter is justified. (See: R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103, at para. 66).

[269] The test for a justification of a Charter infringement has been refined, but not substantially changed, since the Supreme Court’s judgment in Oakes. As recently summarized in Carter v. Canada, 2015 SCC 5, [2015] 1 S.C.R. 331, to justify an infringement of a Charter right, Ontario must show:

[A] pressing and substantial object and that the means chosen are proportional to that object. A law is proportionate if (1) the means adopted are rationally connected to that objective; (2) it is minimally impairing of the right in question; and (3) there is proportionality between the deleterious and salutary effects of the law: R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103.

1. Pressing and Substantial Objective

[270] The parties agree that road safety is a pressing and substantial objective but disagree about how Ontario’s pressing and substantial objective should be characterized.

[271] Ontario states that its goal is to have the safest roads in North America, if not the world. Ontario argues that this pressing and substantial objective should animate the s. 1 analysis.

[272] The Applicant disagrees. The Applicant challenges the veracity of the stated goal of the “safest roads” as a basis for distinguishing what other provinces are doing. The Applicant makes two points. First, she argues that this is an aspirational goal. Second, she argues that Ontario is shifting the s. 1 test by using a superlative pressing and substantial objective.

[273] The Applicant points out that the evidence at best supports Ontario’s goal to be not “the safest” but to be “among the safest.” She also argues it is not meaningful to simply state the goal. Everyone wants safety. Yet compromises are made to road safety in other areas, such as allowing 16-year-olds and over 80-year-olds to drive despite statistics about their driving safety, and to allow individuals with multiple convictions for impaired driving back on the road after period of suspension.

[274] The comparisons that the Applicant seeks to draw with other groups of “risky” drivers are not apt and should not undermine Ontario’s claim to aspire to be have the best road safety in North America. Data is not before the court about the safety risks posed by those other groups. It is a judgment call for Ontario legislators to make regarding which risky groups may drive. There may also be other reasons for allowing some to keep their licences. For example, reliance would be a factor in allowing an elderly driver to keep their licence if they had a licence for their whole life, or for legacying an individual who was in a waiver program that was cancelled.

[275] Ontario’s annual reports showed that it ranked as having among the safest roads in North America (as measured by fatality rate), ranking either number one or number two over a number of years. Ontario submitted its 2013 and 2014 annual reports on road safety. In Ontario’s 2014 annual report, its fatality rate was 0.53 per 10,000 licensed drivers, second lowest in all North American jurisdictions behind only Prince Edward Island. In 2013, Ontario’s fatality rate was 0.54 per 10,000 licensed drivers, again second lowest in all North American jurisdictions, this time behind the District of Columbia. The 2014 annual report noted that Ontario had ranked lowest or second lowest in fatality rates across North America for 16 consecutive years. Ontario’s stated goal is to continue to achieve that ranking.

[276] Ontario might aspire to have the best road safety in North America. Ontario’s evidence does demonstrate a legislative objective of superior road safety, but the evidence does not demonstrate a legislative objective of superlative road safety. Ontario’s realistic legislative objective can be characterized as: having among the safest roads in North America.

2. The Appropriate Standard of Deference: The Precautionary Principle and a “reasoned apprehension of harm”

[277] To ground its arguments on rational connection and minimal impairment, Ontario seeks to rely on “the precautionary principle” and a “reasoned apprehension of harm.” The Applicant, in turn, argues that these principles are extrinsic to Charter litigation. The cases suggest otherwise, but these are part of, and not a replacement for, the s. 1 proportionality assessment. The cases relied on by the Respondent are essentially arguments for deference.

[278] I turn first to the precautionary principle.

[279] The Respondent relies on R. v. Michaud, 2015 ONCA 585, 127 O.R. (3d) 81. Michaud was also a case about the Highway Traffic Act. I reproduce the full text of the relevant quote below:

[102] There is good reason to favour ex ante rules where human life or safety is at stake and where there is scientific uncertainty as to the precise nature or magnitude of the possible harms. In such cases, regulators utilize a “precautionary principle,” which the authors of Risk Management note, “tackles the problem of an absence of scientific certainty in certain areas of risk, and directs that this absence of certainty should not bar the taking of precautionary measures in the face of possible irreversible harm” (1:40). The Supreme Court has recognized the precautionary principle in the context of environmental protection regulations: 114957 Canada Ltee v. Hudson (Town), 2001 SCC 40, [2001] 2 S.C.R. 241. [Emphasis added]

[280] The “precautionary principle” described here is relevant to this case. I do not read Lauwers J. as creating a new legal doctrine in Charter litigation. Rather, Justice Lauwers effectively describes the appropriate deference owed to legislators. When legislators act to protect human life or safety, the “precautionary principle” describes the use of foresight and caution in crafting legislation where the harm is not precisely known. This interpretation is consistent with both previous Charter jurisprudence and with further comments in Michaud.

[281] In Michaud, the Court of Appeal also noted the importance of judicial deference in the context of safety regulations. At para. 106, the Court of Appeal stated, “This court has noted that judicial deference to legislative choice is particularly appropriate where the legislation is concerned with public welfare or safety.”

[282] I turn next to the arguments on a “reasoned apprehension of harm.” As noted above, the Respondent relied on Cochrane for this principle. The Respondent also cites Harper, Irwin Toy, R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697, and Saskatchewan (Human Rights Commission) v. Whatcott, 2013 SCC 11, [2013] 1 S.C.R. 467, among others. The Respondent argues that courts have acknowledged the legislature’s role “in drawing a line” when there is no scientific certainty.

[283] To the extent that the Respondent relies on these cases for the general principle of judicial deference, I agree. However, the comments in these cases about scientific certainty require closer scrutiny.

[284] The cases listed – and others relied upon by the Respondent – are cases dealing with societal harms which are not amenable to proof by empirical evidence. Irwin Toy dealt with the impacts of advertising on children. Harper was about the effect of spending limits on election campaigns. Keegstra and Whatcott both considered the impacts of hate speech. None of these topics lend themselves well to statistical measurement or functional assessment. Some comments from Harper explain this further.

[285] At para. 77 of Harper, Bastarache J. noted: “Where the court is faced with inconclusive or competing social science evidence relating the harm to the legislature’s measures, the court may rely on a reasoned apprehension of that harm.” Justice Bastarache continued, at para. 78: “This Court has, in the absence of determinative scientific evidence, relied on logic, reason and some social science evidence in the course of the justification analysis in several cases[.]” He went on to cite a number of cases dealing with social issues where the potential harm was not easily measurable. Notably, the heading directly preceding these two paragraphs is titled “The Nature of the Harm and the Inability to Measure It”.

[286] This case is distinct in that respect. According to Dr. Owsley, safe driving can be measured and predicted by MVC rate statistics. Dr. Peli opined that an individual’s capacity to safely perform driving tasks and their mastery of basic driving skills can also be assessed. Even if, as Ontario argues, there is no evidence-based, established method for doing so at this time, that does not mean that it cannot be measured.

[287] Thus, while the Regulations, which are in the nature of public safety legislation, are deserving of some deference, consistent with the Court of Appeal’s holding in Michaud, the level of deference is not as high as it might be in a different situation where the potential harm was not readily measurable.

[288] The Respondent seeks the following findings:

a. It was reasonable for Ontario to believe, based on the results of its own 10-year pilot program and studies in other jurisdictions, that persons falling below the vision standards have a higher crash/collision rate.

b. It was reasonable for Ontario to believe that the science demonstrates that there is higher safety with two eyes than with one eye.

c. Driving a bus involves greater risk to safety than other vehicles (for others on or outside of the bus).

d. It was reasonable for Ontario to take the position that no road test, simulator, occupational therapy assessment, or functional assessment has been validated as a reliable predictor of safety – certainly not for buses, but even generally.

[289] Ontario says all the court needs to find is that it was reasonable for Ontario to believe this, the court need not actually find these to be facts. That is what it says is its reasonable apprehension of harm, to which deference should be afforded. I disagree.

[290] I do not consider the level of deference to be afforded to Ontario in this case to be synonymous with immunity from constitutional scrutiny. The onus is on the government in s. 1. The Applicant has established a Charter violation. The government must do more to justify that violation than simply point to the fact that it has legislated and assert its reasonable belief, or its apprehension, of harm. Rational connection, minimal impairment, and the balancing of deleterious and salutary effects all still must be established.

3. Proportionality of the Infringing Statutory Provision

3.1) Rational Connection

[291] The Applicant argues that there is no rational connection between the stated objective and the Regulations. The Applicant says evidence of increased risk of monocular drivers as a group is so weak it fails the Oakes test here. The measures adopted “must not be arbitrary or unfair or based on irrational concern,” (Oakes, at para. 70).

[292] The Applicant acknowledges that this is a low threshold but argues it does require a causal relationship (Mounted Police Association of Ontario v. Canada, 2015 SCC 1, [2015] 1 S.C.R. 3, at paras. 143-144). Here, for reasons indicated above, the scientific evidence does not establish this causal link. The Applicant also argues the standard cannot have a rational connection because it is arbitrary and irrational. She asserts it is arbitrary and irrational to require her to have some vision in her weaker eye given that it is possible that an individual with very little vision in their weaker eye who meets the standard could suffer from binocular inhibition, actually giving them worse visual acuity than her.

[293] The Respondent argues that s. 18(3) of the Regulations is rationally connected to the legislative objectives. Ontario relies on Michaud at para. 117, saying that “the government need only demonstrate a reasonable prospect that the limit will further the objective to some extent, not that it will certainly do so.”

[294] Ontario’s goal is to have among the safest roads in North America. The threshold for establishing a rational connection is not high. I find that the evidence of the increased MVC rates (in the range of 1.4 times) for monocular commercial drivers as a group is enough to establish a rational connection between the restrictions in s. 18(3) of the Regulations and Ontario’s legislative objective.

3.2) Minimal Impairment

[295] The Applicant asserts that Ontario has not met its onus and has failed to demonstrate the Regulations to be a minimally impairing method of achieving those goals. The onus is on Ontario to demonstrate that the current Regulations are minimally impairing.

[296] Ontario’s position is that the Regulations fall within a range of reasonable alternatives to best protect Ontario drivers and uphold its legislative objective of having among the safest roads in North America. For the extent to which a government must go to find reasonable alternatives, Ontario relies on R. v. Chaulk, 1990 34 (SCC), [1990] 3 S.C.R. 1303:

Parliament is not required to search out and to adopt the absolutely least intrusive means of obtaining its objective. Furthermore, when assessing the alternative means which were available to Parliament, it is important to consider whether a less intrusive means would achieve the “same” objective or would achieve the same objective as effectively.

[297] Ontario says that its approach to the problem falls within the range of reasonable alternatives. Ontario asserts that there is scientific controversy regarding the safety of the statistical approach and the driver performance approach. Ontario prefers to rely on the statistical approach rather than an individualized one, and asserts it is within a reasonable range of alternatives for it to do so. Ontario does not believe that the alternatives suggested by the Applicant, including various methods of individual assessment, have been properly studied or validated. It is not only a question of whether Ontario’s vision standards fall within the range of reasonable alternatives, but also, whether choosing to rely on statistical evidence rather than unvalidated individual performance assessments falls within the range of reasonable alternatives.

[298] Ontario also relies on the precautionary principle to justify the Regulations as minimally impairing. Ontario relies on both Michaud, above, and Canadian Blood Services v. Freeman, [2010] O.J. No. 3811, at para. 27: “The precautionary principle is the justification for acting in the face of uncertainty. It is a tool for avoiding possible future harm associated with suspected, but not conclusive, risks.” Ontario says that the alternatives proposed by the Applicant are not validated and pose significant risk. Therefore, Ontario argues that the precautionary principle entitles it to a margin of appreciation in its choices about what legislative choices to make.

Suggested alternatives to a blanket prohibition

[299] Ontario argues that to show minimal impairment, the question is whether the goal of having among the safest roads in North America could be achieved using less impairing methods. Ontario says it cannot.

[300] The Applicant attempted to introduce some evidence late in the proceeding which pertained to the validity of some alternative means to demonstrate safe driving. The Applicant introduced this evidence without consent, after the cross-examinations had already concluded. The impugned evidence was contained in two December 2019 affidavits, one which had an academic article attached to the affidavit of a legal assistant working at the law firm of the Applicant’s counsel.[^11]

[301] Ontario objected to this late evidence. However, if the evidence were to be admitted, Ontario asked Dr. Owsley to review the evidence and produce a supplementary affidavit. The Respondent’s position is either that the evidence should not be admitted, or if it is admitted, then Dr. Owsley’s late affidavit should also be admitted.

[302] I rule that the Applicant’s late-filed new evidence should not be admitted. Not only was it late, but the aspect of this evidence that was most strenuously objected to (the Kristalovich and Mortenson article) could not be meaningfully considered by the court without some expert evidence about it. Merely appending an academic article such as this to a law clerk’s affidavit will not allow the court to properly consider the substance of it. I place no reliance upon the Kristalovich and Mortenson article that was specifically objected to. Therefore, Dr. Owsley’s response to that article also need not be admitted or considered.

[303] The Applicant proposes several other less impairing alternatives which could be implemented (no one necessarily exclusively, but some combination of them):

a. Functional driving assessments: For example, something similar to what Ontario already does for Class G licences. Ontario could take the Class C licence road test for commercial drivers, adapt it and test for use of compensatory strategies.

b. Functional driving evaluation with an occupational therapist, driving rehabilitation specialist, ophthalmologist or other appropriate professional. An ophthalmologist could give an opinion about the individual driver’s functional visual impairment. An occupational therapist could work with a monocular driver to assess their strengths and weaknesses.

c. An intensive training program, similar to the one administered by the City of Mississauga. The program trains bus drivers, with the intent to ensure they are prepared to operate a commercial vehicle safely. A specialized version could be developed for Class C applicants who don’t meet the vision standards.

d. Driving simulators: these could assess for particular, specific issues.

e. A period of adaptation, such as in British Columbia and Quebec, as a minimum qualification.

f. A period within which the monocular driver is collision free for a certain period of time, as a minimum qualification.

g. A clean driving record with no demerits as a minimal qualification.

h. A longer and more comprehensive road test.

[304] Ontario’s position is that these alternate means do not allow Ontario to achieve its same high safety goals, or achieve them as effectively, as that goal is achieved through the vision standards. Ontario has various responses to each of these suggestions.

[305] Ontario also rejects the period of adaptation and a clean driving record as useful criteria. This is exactly what Ontario did in the commercial vehicle Waiver Program. As Ontario has demonstrated, drivers in that program had a higher MVC rate. These criteria alone are not enough to ensure safety. However, it is possible that they could be combined with other criteria, such as a functional assessment.

[306] Ontario also takes the position that driving simulators cannot effectively demonstrate that a driver is capable and safe. This is consistent with the evidence given by the Applicant’s expert witnesses.

[307] Ontario rejects the idea of an extended road test would demonstrate sufficient competency and skill. In part, Ontario relies on Dr. Owsley’s position that only statistical assessments offer effective predictions of how safe a driver will be. Ontario says a road test is a snapshot and does not provide an assessment of their driving safety. It does not test real-world conditions, conducted as it is in an artificial environment, where the individual knows they are being tested. Ontario points out that Dr. Arshinoff agreed that road-testing is not sufficient to declare a driver safe.

[308] Ontario also rejects the idea of functional assessments by various specialists. Ontario says that occupational therapists and driving specialists will provide subjective assessments, which are not substantiated as a method for predicting driver safety. Ontario cites a survey which found that evaluators used subjective criteria or no criteria at all in their evaluations. Ontario currently carries out functional assessments for individuals involved in its Class G waiver program, but says that these assessments have not been corroborated as effectively predicting collision rates. Further, Ontario says the assessments are designed for Class G drivers, not Class C drivers.

[309] There is some circuity to Ontario’s argument. The fact that many evaluators use no criteria or only subjective criteria does not mean that Ontario must adopt this approach to driver evaluation. Indeed, Ontario has adopted an assessment approach for Class G drivers who fall below the vision standards. Presumably, Ontario has not intentionally implemented an assessment approach which relies on purely subjective criteria, or no criteria whatsoever.

[310] Ontario’s position is that these functional assessments have not been validated by statistics, or through the scientific method. This position appears to be a catch-22. These alternatives can only be validated if they are allowed and results can be measured. Ontario’s response is that it is currently assessing their viability through its Class G waiver program and is entitled to the deference to perform its analysis step-by-step. However, Ontario offers no time horizon by which it expects to have these results or a plan for implementing a further assessment for commercial class licence holders.

[311] It has been 22 years since Ontario’s internal Engel-Townsend Report concluded that there is “some evidence” that monocular drivers have a higher rate of collision than binocular drivers do and that “monocular drivers may have a performance disadvantage that warrants some degree of special treatment.” Ontario argues that it has been studying this problem responsibly since the 1995 MTO Report and Engel-Townsend Report were authored. However, its studies have not focussed on individual capacities. Deference only goes so far. At some point, Ontario needs to directly address this question of assessing the individual capabilities of commercial drivers. Ontario has not demonstrated that its goal of having among the safest roads in North America cannot be achieved using less impairing methods than the current approach of a blanket prohibition of monocolur commercial drivers. I am not satisfied that the blanket prohibition is the minimally impairing method of achieving that goal.

The approach of other Canadian jurisdictions

[312] Other provinces and states allow for individual assessment. The Applicant goes so far as to say that no other province absolutely prohibits a monocular person from obtaining a commercial class driver’s licence. The Canadian Council of Motor Transport Administrators (CCMTA) guidelines followed by some other provinces provide for individual functional assessment. The Federal Motor Carrier Safety Administration (FMCSA) in the US has a waiver program for intra-state monocular drivers. Ontario itself has a waiver program for G-class drivers who cannot meet the field of vision requirement of 120 degrees for that class of licence, under s. 21.2 of the Regulations.

[313] The Applicant refers the court to the CCMTA Medical Standards with B.C. Specific Guidelines, recently adopted in British Columbia. Chapter 22 deals with visual impairment, including impaired visual acuity, visual field loss, and monocularity. The Guidelines list functional assessment as an available tool to determine whether an individual with impaired vision can continue to drive commercially. Sections 22.6.2, 22.6.4, and 22.6.5 all offer the option of functional driving assessments to an individual who falls below the visual standards. There does not appear to be any indication that public transit bus drivers receive different treatment.

[314] Alberta, Manitoba and Saskatchewan all rely on the CCMTA Medical Standards. Section 22.5 covers compensation for visual impairment. That section says, inter alia:

The loss of some visual functions can be compensated for adequately, particularly in the case of long-standing or congenital impairments. When an individual becomes visually impaired, the capacity to drive safely varies with his/her compensatory abilities. As a result, there may be individuals with visual deficits who do not meet the vision standards for driving but who are able to drive safely. On the other hand, there may be individuals with milder deficits who do meet the vision standards but who cannot drive safely.

In these exceptional situations, it is recommended that the individual undergo a special assessment for the fitness to drive. The decision on fitness to drive can only be made by the appropriate licensing authorities. However, it is recommended the following information be taken into consideration: (1) favourable reports from the ophthalmologist or optometrist; (2) good driving record; (3) stability of the condition; (4) no other significant medical contraindications; (5) other references (e.g. professional, employment, etc.); (6) functional assessment.

[315] Section 22.6.5 covers monocular drivers. One of the eligibility criteria states: “Where required, a road test or other functional assessment indicates the driver is able to compensate for any loss of functional ability necessary for driving[.]” Section 5.3 of the Background indicates that individual assessment must be based on “personal abilities” rather than “presumed group characteristics.”

[316] New Brunswick requires a visual field of 120 degrees in each eye, including for public transit bus drivers, under s. 27(1)(a)(xi) of General Regulation, N.B. Reg. 83-42, pursuant to the Motor Vehicle Act, R.S.N.B. 1973, c. M-17. However, under s. 27(2)(b), the Minister may waive the vision requirements if the Applicant presents satisfactory evidence that he or she can safely operate that type of vehicle. The Minister has the authority to refer the applicant to a medical review board to assess that person’s ability to operate a motor vehicle, under s. 27(3).

[317] Nova Scotia requires a visual field of 120 degrees in each eye, including for public transit bus drivers, under s. 5(a)(viii) of the Classification of Drivers’ Licences Regulations, N.S. Reg. 174/82 pursuant to the Motor Vehicle Act, R.S.N.S. 1989, c. 293. However, the Registrar of Motor Vehicles may waive the vision requirements if the Medical Advisory Committee satisfies the Registrar that the person does not represent an unacceptable safety risk, under s. 15.

[318] Prince Edward Island has a similar exception to the ones described in New Brunswick and Nova Scotia. The Driver’s Licences Regulations, P.E.I. Reg. E.C. 550/02, pursuant to the Highway Traffic Act, R.S.P.E.I., 1988 Cap. H-5 define buses as Class 2 vehicles, under s. 1(1)(b)(ii). Under s. 3(3)(b), Class 2 drivers cannot receive a licence unless they meet the medical standards set out by the CCMTA. However, under s. 3(4)(c), the medical standard can be waived if the Registrar is satisfied “that the applicant can safely drive a motor vehicle that requires the class of driver’s licence for which the application has been made.”

[319] Ontario responds that it is not required to legislate to the “lowest common denominator” of other jurisdictions. The test for minimal impairment is informed by the values of federalism. Provinces must be entitled to make their own laws. Ontario quotes constitutional scholar, the late Peter Hogg in Constitutional Law of Canada, 5th ed (Supp. (loose leaf)), vol. 2, p. 38-39: “The uniformity of provincial laws that would be entailed by a stringent requirement of least drastic means is in conflict with the federal values of distinctiveness, diversity and experimentation.”

[320] This argument is an important consideration when making jurisdictional comparisons. However, the principle of federalism does not entitle Ontario to be out of step with other provinces when its divergent choice violates constitutional rights. There is no evidence to suggest that the other Canadian provinces do not have high standards of road safety.

[321] Interjurisdictional comparison alone is not what makes this law fail the minimal impairment branch of the Oakes test but it does call into question the level of deference that Ontario should be afforded in its decision not to consider exemptions based on individual driver performance assessments.

Conclusion on Minimal Impairment

[322] Ontario has not met its burden to establish that the vision standards under the Regulations are minimally impairing. There do appear to be less intrusive means by which its objective of maintaining its high standards of road safety could still be achieved. Suggesting that Ontario adopt approaches that have been adopted in several other provinces and nationally in both Canada and the United States is not suggesting that it reduce its standard to some lowest common denominator.

[323] Ontario is owed deference in matters of public safety and regulatory schemes. The precautionary principle allows Ontario to pre-empt harm with its legislation. However, the fact that Ontario is owed deference does not protect its laws absolutely from constitutional scrutiny. The Applicant seeks only the opportunity for monocular individuals to demonstrate they have the capacity to drive commercial vehicles safely. Ontario has a multitude of options, which are not limited to those presented in this case. Ontario is still fully entitled to determine what the broad rules will be. It is also entitled to shape the form of the assessment that it decides is applicable.

3.3) Balancing salutary and deleterious effects

[324] The third branch of the proportionality inquiry of the Oakes test is the final balancing of the salutary and deleterious effects. Here, the court must balance societal rights and benefits. Do the deleterious effects of the Regulations, prohibiting monocular drivers, outweigh its salutary effects? I conclude that they do.

[325] The Applicant argues Ontario’s prohibition is not reasonable when there are already some monocular Class G drivers on the road, some monocular drivers who have been legacied into having commercial licences, and other provinces allow monocular drivers to have commercial licences – drivers who could then drive into Ontario in the course of their work. The Applicant adds that the science and literature does not support the level of risk associated with monocular drivers that Ontario propounds.

[326] This is contrasted against the Applicant losing something she greatly valued: her ability to work in the job she enjoyed without being afforded the opportunity to establish that she can drive a bus safely. As noted above, the deleterious effects go beyond the simple economic impact. Employment is an important part of an individual’s life. Ms. Di Cienzo described the stress and frustration of trying to respond to the Ministry’s decision. She also described feeling as though she was “treated unfairly because of her disability.”

[327] Ontario argues the deleterious effects to the Applicant are not significant. Not every individual who falls below the vision standard for a commercial licence will lose their job or come into economic instability. More specifically, Ontario relies on Tadros v. Peel (Police Services), 2009 ONCA 442, 97 O.R. (3d) 212, at para. 51 for the holding that the Charter “does not protect a right to employment in any particular job or profession.” Respectfully, this appears to miss the point of Tadros and the Applicant’s argument. The Applicant is not seeking a freestanding right to employment, or to a particular job. Rather, she is seeking freedom from a blanket prohibition from a particular job, when that prohibition is based on a protected ground