Abstract
The driver licensing standards and process for drivers with visual field impairment (VFI) have changed significantly over the past 20 yr. The visual field is the extent of the area that a person can see with his or her eyes held in a fixed position, usually measured in degrees. The normal binocular (using both eyes) visual field is 135° vertically and 180° horizontally from the fixed point. VFI refers to a loss of part of the normal visual field (Canadian Council of Motor Transport Administrators [CCMTA], 2015).
The Canadian provincial standards vary from the U.S. standards for drivers with VFI. Within the United States, the visual field standards for driving vary from state to state, and not all states will grant a license to a driver with VFI. The Canadian standards were established after a Supreme Court of Canada decision ( British Columbia v. British Columbia, 1999) regarding a driver with homonymous hemianopsia who was denied a driver’s license because of his medical condition. It was established that driver licensing decisions could not be made solely on the basis of a diagnosis, and the driver had the right to an individualized assessment.
The licensing authority, RoadSafetyBC, subsequently shifted from a diagnostic to a functional approach (Ministry of Public Safety & Solicitor General Office of the Superintendent of Motor Vehicles, 2010). The Canadian Medical Association’s (2012) guidelines have recommended a functional evaluation for people with VFI of less than 120° horizontal vision. RoadSafetyBC will issue a driver’s license to drivers with VFI if they successfully complete a Functional Driver Evaluation with an occupational therapist.
Previous studies (Bowers et al., 2005; Elgin et al., 2010; Papageorgiou et al., 2012; Parker et al., 2011; Racette & Casson, 2005; Tant et al., 2002; Wood et al., 2009, 2011) supported individual, functional assessment because there was considerable variation in driving performance for people with VFI. Some drivers with VFI successfully completed a simulator or on-road assessment with a specialized examiner, whereas other drivers were deemed unsafe to drive.
Research has focused on assessments that identify which drivers with VFI are more likely to be successful with an on-road evaluation. Diagnostic information, such as the location and the extent of VFI, was inconsistent in predicting on-road performance (Elgin et al., 2010; Racette & Casson, 2005; Wood et al., 2009). Self-evaluation was not consistent with observed driving and was not a reliable measure of driving skills (Winfred et al., 2005). Among drivers with VFI, visual or cognitive skills that are assessed by occupational therapy practitioners, such as contrast sensitivity, processing speed, and attention, were correlated with on-road driving performance (Bowers et al., 2005; Elgin et al., 2010; Wood et al., 2009).
These studies appear to indicate that the comprehensive, functional evaluations completed by occupational therapy practitioners and driver rehabilitation specialists provide the most predictive information regarding on-road driving performance. However, research has not supported the use of a single tool in making definitive decisions pertaining to driving competence (Bédard et al., 2008; CCMTA, 2015; Crizzle et al., 2012; Diegelman et al., 2004; Kwok et al., 2015; Stav et al., 2008). On-road assessment is considered the gold standard for determining fitness to drive.
Research regarding on-road assessments has identified specific driving behaviors present in people with VFI. Poor lane position (Bowers et al., 2005; Elgin et al., 2010; Tant et al., 2002; Wood et al., 2011) and traffic gap judgment (Bowers et al., 2005; Wood et al., 2009) were evident in the drivers with VFI who were deemed unsafe to drive. Evidence on speed control was not consistent. One study showed no difference in drivers with and without VFI in high-speed (interstate) driving (Wood et al., 2009), and other studies showed that drivers with VFI had difficulty maintaining speed during lane changes (Bowers et al., 2005) or during general driving (Wood et al., 2011). Drivers who were determined safe to drive had increased head and shoulder movements to the side of visual loss compared with drivers who were determined unsafe to drive (Wood et al., 2011).
Only one study (Dow, 2011) evaluated the safety of drivers with VFI over time. Over an 8- to 12-mo period, among 103 drivers with VFI, two minor injury accidents and five “fender benders” (a crash between two vehicles resulting in minor material damage) were reported. The crash rate was 6.79 crashes per 100 drivers with VFI, which exceeded the Quebec provincial crash rate of 5.5 crashes per 100 drivers. There was no analysis to determine whether this difference was statistically significant.
For drivers with VFI, occupational therapy practitioners may be requested to assess their fitness to drive. Although there is research regarding the in-clinic and on-road assessment for fitness to drive, the relationship of these assessment results to driving performance over time has not been well studied. To address this gap, we examined the number of crashes and traffic violations during the year after a visual field test to determine whether drivers with VFI had a higher probability of crashes or traffic violations over time than drivers without VFI.
Method
The research protocol was approved by the University of British Columbia Behavioral Research Ethics Board and the Insurance Corporation of British Columbia (ICBC) Ethics and Privacy Board. Participants were identified through a retrospective review of neuro-ophthalmology records at the Department of Ophthalmology and Visual Sciences Eye Care Centre at the University of British Columbia. Potential participants were identified through ophthalmology billing codes for Goldmann visual field tests completed at least 1 yr before the study. For participants with more than one visual field test, the most current visual field test that met the inclusion criteria was used.
For study inclusion, participants were required to meet the visual acuity standard for driving, which is a binocular, corrected visual acuity of ≥20/50 (Ministry of Public Safety & Solicitor General Office of the Superintendent of Motor Vehicles, 2010). Only participants ages 26–74 yr were included in the study because drivers who are younger than age 25 yr or older than age 75 yr have an increased rate of accidents per kilometer driven (Cerelli, 1989; U.S. Department of Transportation, National Highway Traffic Safety Administration, 2010).
Participants who met the inclusion criteria were stratified into groups on the basis of Goldmann visual field test results. Group A had no VFI. Group B had VFI but met the driver licensing standards (Ministry of Public Safety & Solicitor General Office of the Superintendent of Motor Vehicles, 2010) of at least 120 continuous degrees along the horizontal meridian and 15 continuous degrees above and below fixation with both eyes open and examined together. Group C had VFI and did not meet the driver licensing standards. Diagnostic data, including the cause of VFI, were not collected. Groups B and C included a range of VFI, from hemianopia to central visual field loss to scotomas.
Participant identifiers and the individual date range for driving record review were provided to the research manager at the ICBC. The ICBC reviewed the driving history for each person for 1 yr after the visual field test and provided the primary researcher with anonymized data for each group. The research data included the number of participants with a valid license, the number of crashes, and the number of traffic violations in each group. Moreover, the number of unique people who had a crash or traffic violation was identified. A traffic violation was any incident for which a police officer issued a ticket. For the purpose of the study, participants were considered to have a valid license if they had a valid license for 6 mo or more in the year after the visual field test.
For confidentiality and protection of personal information, the ICBC research manager was not provided with visual field test results and was not advised which participant groups had VFI. All data were obtained retrospectively from existing records, and no treatment was provided.
Analysis
Given the nominal nature of the data, χ2 tests were used for the following analyses: The percentage of participants who held a driver’s license at any time, the percentage of participants who retained a valid license in the year after the visual field test, and the percentage of participants who held a driver’s license at any time and retained it after visual field testing were compared among the three groups (Group A, no VFI; Group B, VFI but met the driver licensing standards; Group C, VFI and did not meet driver licensing standards). We also used χ2 tests to compare the number of crashes and traffic violations among drivers with a valid license. The groups with VFI (Groups B and C) were compared against the group without VFI (Group A). A p of <.05 was considered significant. Given the small number of accidents, the Monte Carlo method was used to simulate a larger sample size.
Results
Four hundred forty-five participants were identified through ophthalmology chart review. One hundred seven participants (24%) were excluded from the study because they never had a driver’s license. Of the remaining 338 participants, 292 (66% of the original study group) were identified as having a valid driver’s license for at least 6 mo in the year after their most recent visual field test. The numbers of drivers and licensed drivers for each group are summarized in Table 1.
Number of Participants in Each Group With a Driver’s License
Note. VFI = visual field impairment.
We completed χ2 tests to compare the percentages of licensed drivers (Table 2). There was no statistically significant difference among the three groups in the percentage of drivers who held a driver’s license at any time (see Table 2). In the year after the visual field test, there was a significant difference in the percentage of drivers with a valid license. Drivers in Group C who did not meet the driver licensing guidelines for visual fields were less likely to have a valid driver’s license after their visual field test (see Table 2). Moreover, drivers in Group C were less likely to retain their license in the year after the visual field test (see Table 2).
Analysis of Drivers With a Valid License
Crashes and traffic violations were the outcome measures used to determine whether there was an increased safety risk for the drivers with VFI. Table 3 summarizes the number of crashes and traffic violations for each group. None of the drivers with a canceled driver’s license had a crash or traffic violation. All incidents were by individual drivers, except 1 driver in Group C had two violations. This finding indicates that the data were not skewed by a single driver with multiple incidents.
Number of Crashes and Traffic Violations
There was no significant difference among the groups in the rate of crashes (p = .402) or traffic violations (p = .108). The Monte Carlo method was used to expand the sample size to 5,000, and there remained no difference among the groups. The p value was .645 for accidents and .101 for violations.
Discussion
Although previous studies (Bowers et al., 2005; Elgin et al., 2010; Papageorgiou et al., 2012; Parker et al., 2011; Racette & Casson, 2005; Tant et al., 2002; Wood et al., 2009, 2011) showed a range of driving performance with drivers with VFI on assessment, there was little information about the probability that these drivers would return to driving. Previous research has focused on the assessment, or the driver’s performance at a single point in time, whereas in our study we reviewed crash rates and traffic violations for 1 yr. For 338 drivers, there was only one crash and eight traffic violations within the year after the visual field test. The single crash occurred with a driver who did not have VFI. During the year after a visual field test, drivers with VFI did not have an increased probability of a crash or traffic violation compared with drivers without VFI. Moreover, 76 drivers with VFI who did not meet the licensing standards but retained a valid license had no crashes.
Our findings suggest that drivers with VFI who did not meet the driver licensing standards were less likely to have or to retain a driver’s license in the year after their visual field test compared with drivers without VFI. In this study, all groups had a lower percentage of licensed drivers in the year after the visual field test. Without qualitative data, it is unknown why some drivers without VFI did not retain their license and whether this outcome is reflective of natural fluctuations in the general population. Although 76% of the drivers with VFI retained their license, this percentage may change on the basis of other functional impairments or interventions provided.
The main limitations of the study relate to the data that were available. If the driver licensing process was followed, then all drivers who did not meet the visual standards (Group C) would have completed a Functional Driver Evaluation with an occupational therapy practitioner or driver rehabilitation specialist. Unfortunately, it cannot be determined whether all drivers in Group C completed a Functional Driver Evaluation or whether drivers in Groups A and B completed a Functional Driver Evaluation for reasons unrelated to the VFI. It is unknown whether in-clinic or on-road rehabilitation was provided to any of the drivers and whether this training influenced the probability of retaining a driver’s license. Finally, given that accident data were collected for only 1 yr after testing, definitive conclusions about long-term driving behavior in drivers with VFI cannot be determined on the basis of available data.
This research suggests the need for a variety of future studies. Further experimental research is needed to determine whether rehabilitation enables more people with VFI to return to driving. It is recommended that future studies expand the data collection to include both quantitative and qualitative data. Details on the types of traffic violations would permit analysis on whether the violations are consistent among the groups or whether violations occur that are consistent with unsafe driving behaviors associated with drivers with VFI (Bowers et al., 2005; Elgin et al., 2010; Tant et al., 2002; Wood et al., 2009, 2011).
Although the number of drivers who retained a driver’s license was known, the number of drivers who continued to drive and their driving frequency were not known. Analyzing the number of incidents compared with the number of kilometers driven would strengthen future research, because there is a known low mileage bias that indicates an increased rate of crashes in drivers with the lowest rate of driving exposure (Langford et al., 2008, 2013; Staplin et al., 2008; Winfred et al., 2005).
For occupational therapy practitioners and driver rehabilitation specialists, this study suggests that drivers with VFI who completed Functional Driver Evaluations did not have an increased probability of crashes in the next year. In keeping with best practice, it is recommended that generalist occupational therapy practitioners and physicians screen and refer appropriate drivers with VFI for a Functional Driving Evaluation. This screening could include assessments of contrast sensitivity, processing speed, and visual attention and task switching, which have correlated with on-road driving performance (Bowers et al., 2005; Elgin et al., 2010; Wood et al., 2009).
Implications for Occupational Therapy Practice
The results of this study have the following implications for occupational therapy practice:
Clients with visual field impairment have the potential to drive. Generalist occupational therapy practitioners may screen and refer appropriate drivers for a Functional Driving Evaluation.
Occupational therapy practitioners who practice as driver rehabilitation specialists may provide Functional Driving Evaluations for drivers with visual field impairment.
Further research on driving patterns among drivers with visual field impairment may strengthen rehabilitation strategies and recommendations.
Conclusion
Drivers with VFI did not have a higher probability of having a crash or traffic violation over a 1-yr period. The majority of drivers with VFI retained their driver’s license, indicating fitness to drive is not determined solely on diagnosis. It is recommended that drivers with VFI be referred for an individualized, functional assessment through a driver rehabilitation center.
Footnotes
Acknowledgments
This research was not funded and was completed in fulfillment of Lisa Kristalovich’s Master of Science in Rehabilitation Science degree. W. Ben Mortenson acknowledges a New Investigator Award from the Canadian Institutes of Health Research.