On­Road Driving Assessment of Older Adults:
A Review of the Literature

Prepared for the Justice Institute of BC
By
Kristina Kowalski, MSc
and
Holly Tuokko, PhD, RPsych

December 2007

On-Road Driving Assessment

EXECUTIVE SUMMARY
The components included in an on-road assessment will depend on the specific target
group of older adults to be assessed and the intended purpose of the evaluation. Many
different approaches are reflected in the literature and some well-developed standardized
measures are available, though none are generally accepted as the “gold standard”. The
key features to be considered in designing or evaluating an on-road assessment procedure
relate to course design (e.g., location of assessment, duration of testing, and route
characteristics), and driving behaviors and scoring of driving errors (e.g., raters,
subjective and objective measures). The characteristics of the route (e.g., four-way stop
intersections, yield situations, merge situations) will necessarily influence the driving
behaviors elicited and scored to determine the outcome of the assessment (e.g., pass/fail).
The scoring mechanisms can be subjective ratings involving observation and some
degree of judgment on the part of the examiner, or objective measurements, derived from
instruments that record various aspects of driving behavior such as Global Positioning
Systems, video-cameras, sensors, accelerometers, computers, and radar and video lane
tracking systems. Other important considerations, when designing or evaluating older
driver assessment procedures, include retraining and embedding on-road assessment
within a broader comprehensive evaluation of driver safety.

2

On-Road Driving Assessment

INTRODUCTION
Exactly what constitutes driving ability and how to best assess it are issues that are as yet
unresolved with respect to older drivers. It is apparent from the literature reviewed that
driving is a complex task and on-road driving is one component of a more detailed
comprehensive driving assessment process that is needed to fully evaluate the older
driver’s abilities and safety. This report will focus on issues related to on-road driving
assessment. As yet, no “gold standard” for assessing on-road driving has been identified
and used consistently in research related to older driver safety. Instead, a number of
different approaches have been taken to on-road driving assessment that differ with
respect to their procedures and intent. The purpose of this report is to describe the onroad driving assessment strategies articulated in the scientific literature relevant to older
adults. In so doing, the key features related to course design (location of assessment,
duration of testing, and route characteristics), driving behaviors and scoring of driving
errors (raters, subjective and objective measures) and other important considerations
relevant to older driver assessment will be addressed. A set of recommendations
concerning the design and implementation of an on-road assessment program for older
adults is provided.
METHOD
A systematic literature review of studies related to on-road driving assessment of older
adults was conducted. The literature review focused on existing research relevant to onroad driving assessment, including both closed course and open road (i.e., in-traffic)
methods, and issues related to scoring driving behaviors. Relevant research was
identified by searching online databases including Psych Info, Ageline, Medline, the Web
of Science, and Transportation Research Information Services (TRIS). An internet search
was used to locate additional scientific literature via Google and Google Scholar and
specific transportation research center websites. The following categories of search terms
were used: 1) aging terms (e.g., older adults, older drivers, elderly, aged, seniors, elders);
and 2) driving assessment terms (e.g., on-road driving assessment/test/evaluation/
examination/track; driving errors/mistakes, fitness to drive, driving performance,
methods/types of driving assessment). The literature search was restricted to research
published between 1990 and 2007.
In most of the reviewed literature, on-road driving assessments were used to study the
relations between driving and other off-road assessment methods (e.g.,
neuropsychological tests, driving simulators) or to compare differences in driving
performance across diagnostic groups (e.g., dementia, stroke, healthy elderly; Fox,
Bowden, & Smith, 1998). Fewer studies specifically addressed the development of onroad assessment methods to predict the actual driving performance (real world driving) of
older adults. For the purposes of this report, we have extracted the key elements of
course design and the evaluation of driving behaviors (e.g., scoring of driver errors)

3

On-Road Driving Assessment

from these various types of studies. In addition, we have included a brief section
highlighting other important considerations relevant to older driver assessment.
RESULTS
Target Population
The population of older adults aged 65 years and over is a heterogeneous group. Healthy
aging and many age-related diseases are associated with declines in sensory/perceptual,
physical/psychomotor and cognitive abilities that may compromise one’s ability to drive
safely (Anstey, Wood, Lord, & Walker, 2005; Eby, Trombley, Molnar, & Shope, 1998).
It is apparent from the literature that most discussion around driving assessment for older
adults has focused on detecting those drivers with cognitive impairment who are unsafe
to drive. Various groups (e.g., physicians, policy makers) have developed practice
guidelines acknowledging the importance of addressing the driving practices of people
with disorders that affect cognitive functioning (Dubinsky, Stein, & Lyons, 2000;
Johansson & Lundberg, 1997; Australian Society for Geriatric Medicine, 2003; Patterson
et al., 2001). However, it is also clear that medical conditions affecting other aspects of
functioning (e.g., vision, physical/motor functions) may affect driving as well, but
perhaps in different ways. That is, driving ability may be differentially impacted in
various groups of older adults. Given the breadth of disorders that may affect driving, the
Canadian Consensus Conference on Driving Evaluation of Older Drivers recommended
that referral for driving assessment is appropriate when there is uncertainty about ability
to drive safely due to health related factors (Korner-Bitensky, Gelinas, Man-Song-Hing,
& Marshall, 2005). The consensus group strongly agreed that these referrals for
assessment should include, but not be limited:
1) to those who have medical conditions that impact functioning (e.g., vision,
hearing, musculoskeletal, mental or neurological health); and
2) to specific diagnostic groups in which driving is frequently compromised
(e.g., stroke, Parkinson’s disease, mild-to-moderate dementia, cognitive
decline, or diabetes with peripheral neuropathy (Korner-Bitensky et al., 2005).
Another group appropriate for on-road driving evaluation is older adults who are either
seeking or being referred for driving re-training. Research on older drivers has
emphasized the need for interventions to increase older driver safety. A recent systematic
review of existing older driver training programs found evidence indicating educational
programs improve driver awareness and driving behavior, but do not necessarily reduce
crash rates (Kua, Korner-Bitensky, Desrosiers, Man-Son-Hing, & Marshall, 2007). It has
been suggested that older drivers need educational programs to teach compensatory
driving strategies (De Raedt & Ponjaert-Kristoffersen, 2000) and to enhance proactive
planning (Marottoli et al., 1998; Rudman, Friedland, Chipman, & Sciortino, 2006). Onroad driving assessment may be used to identify individuals who may benefit from driver
re-training and to identify particular driving behaviors as the focus for driver re-training
(Fox et al., 1998). Several studies have examined post-intervention changes in on-road
driving performance of older drivers (Ashman, Bishu, Foster, & McCoy, 1994; McCoy,
Tarawneh, Bishu, Ashman, & Foster, 1993; Ostrow, Shaffron, & McPherson, 1992)
4

On-Road Driving Assessment

The components included in an on-road assessment will depend on the specific target
group of older adults to be assessed and the intended purpose of the evaluation. For
example, many hospitals or rehabilitation centres employ driver rehabilitation specialists
(DRS), often occupational therapists, who are skilled in physical rehabilitation and
vehicle modification to accommodate physical disability. The driving assessments they
perform have been developed through clinical judgment and are based on subjective
criteria. Often the goal of driving assessments conducted in these settings is rehabilitative
in nature, focusing on the individual’s driving goals, and assessing driving in situations
relevant to those goals. This rehabilitative approach is likely to be of assistance in
identifying the driving strengths and weaknesses of the individual and possible driving
re-training strategies for that individual.
If the intent of the assessment is to reach a decision about the general driving safety of
the person, other approaches may be necessary. Such a decision requires that a
comparison be made to ‘reasonable road safety’, or to the performance of experienced
healthy drivers in similar driving situations. This requirement arises from a Supreme
Court of Canada ruling that standards for testing a driver with a disability cannot require
a level of driving safety that exceeds ‘reasonable road safety’. Reasonable road safety
denotes the level of road safety of healthy drivers who are permitted to drive. This ruling
does not imply that driver evaluation necessarily be the same for healthy and disabled
people; rather it only specifies that a higher standard of driving safety cannot be required
for disabled people than is required for healthy drivers.
Course Design
From the literature reviewed, it is clear that there are large differences in on-road
assessment practices in terms of on-road course design. This includes the location where
testing tasks place (e.g., closed course versus open road), duration of the assessment (30
minutes to over 90 minutes), and route characteristics. The characteristics of the route
(e.g., four-way stop intersections, yield situations, merge situations) will necessarily
influence the driving behaviors elicited and scored to determine the outcome of the
assessment (e.g., pass/fail). Driving behaviors, and methods by which they have been
scored, will be discussed after the key features of course design are examined.

Location
Within the literature reviewed, distinctions have been made between closed course and
open road in-car on-road driving evaluations. A closed course, or circuit, is a route
defined in a non-traffic or low traffic environment. Closed courses have been used for
two purposes: 1) to assess whether a client meets minimum standards of competence for
an open-road assessment (i.e., basic maneuvering or operational level skills; Fox et al.,
1998) and 2) to assess more complex operational aspects of driving such as non-routine
(emergency) driving events (Dobbs, 2005). All open road evaluations take place in

5

On-Road Driving Assessment

normal traffic situations, but may vary with respect to length, duration, and route
characteristics.
When examined in relation to open road driving performance, closed courses have
yielded little additional information to that obtained in traffic (Dobbs, 2005; Fox et al.,
1998). Moreover, closed courses are necessarily limited in that they do not provide
information about driver’s ability to interact with complex traffic patterns (Odenheimer et
al., 1994). In this way, they lack ecological validity. Closed course performance may not
generalize to “real world” driving for a number of reasons, perhaps the most salient of
which is that there is greater time for compensation of slowed reaction time to take place.
However, closed course evaluation may provide valuable information about car handling
skills and readiness to interact with traffic.
Duration
Professionals attending workshops at the Association of Driver Educators for the
Disabled (ADED) annual conference responded to survey questions on the process and
structure of their driving evaluation practices (Korner-Bitensky, Bitensky, Sofer, ManSon-Hing, & Gelinas, 2006) and 99 percent reported using on-road assessments greater
than 30 minutes in length, with 50 percent reporting length as 30-60 minutes and 49
percent indicating greater than 60 minutes. Participants in the Canadian Consensus
Conference on Driving Evaluation of Older Drivers agreed that a 45-60 minute on-road
evaluation generally provides sufficient time to assess various on-road driving maneuvers
and behaviors. However, they noted (i.e., strong consensus) that the driver evaluation
might need to be cut short if the evaluator or instructor perceives a threat to safety of the
individuals in the vehicle.
Route Characteristics
Another distinction that has been made within the literature reviewed is whether a driving
assessment route is standard (e.g., same components included in every assessment) or
not. From the ADED surveys, it was evident that, of the 94 percent of respondents who
reported performing on-road evaluations when assessing older drivers, the majority used
standard or fixed routes. Of note, a greater percentage of respondents indicated use of a
standard route in 2003 (78% said yes) than in 1997 (65% said yes).
The key route characteristics, or critical situations, included in an open road test have
varied from study to study. Typically, they are chosen to reflect situations in which older
adults report difficulty (Mallon & Wood, 2004) or that represent situations where normal
older adults, or older adults with medical conditions, have demonstrated difficulty
(Dobbs, 2005). Participants in the Canadian Consensus Conference on Driving
Evaluation of Older Drivers made a number of recommendations about open road course
design (see Table 1). In addition, strong consensus was reached concerning the
observation that, while existing on-road assessments had many good features, none
addressed the entire driving task.

6

On-Road Driving Assessment

Table 1. Critical Components of an on-road driving assessment route
Strong Consensus
Four-way stop intersections
Two-way stop intersection
Left turns
Right turns
Stop sign
A merge that requires an increase in speed
Roadway requiring lane positioning
Route requiring changing of lanes
Roadways requiring varying speeds to > 70 km/hr
Yield situation (sign where available)
Environment requiring backing up
Following another vehicle
Moderate Consensus
Traffic lights
Merging speed of 70 km/hr plus
Parallel parking not a critical component (required in some provinces)
The strong agreement reached at the consensus conference on many of the critical
characteristics of an on-road assessment route provides a basis for a core set of elements
for sites to consider when assessing older drivers.
Another distinction made in the literature concerning in-car, on-road driving evaluations
is whether or not they are standardized. A standardized driving evaluation involves a set,
or fixed, route designed to fulfill requirements for traffic density and specified driving
behaviors (e.g., maneuvers) that may be reproduced in other locations. The developers of
these evaluations argue that knowledge beyond personal observation and consensus
opinion is needed, and the scientific method provides a means to acquire that knowledge.
These test developers each address the need for validated performance-based driving
evaluations that may serve as a basis for driving recommendations that are fair and
consistent. For each of these evaluations, specific driving behaviors (see below) are
observed, broken into components, and rated, often several times. The properties of the
evaluation are examined to address issues of reliability (i.e., the same results will be
obtained on repeated assessments or with different raters) and validity (i.e., the outcome
evaluation distinguishes between groups of drivers or relates to other measures of driving
ability). Some evaluations have been developed more thoroughly than others. Examples
of standardized driving evaluations are summarized in Appendix A.
It is important to note that from the reports of professionals in the ADED study (KornerBitensky et al., 2006), only two respondents used a published (standardized) road test, in
both instances the Miller Road Test (see below; Carr, Jackson, Madden, & Cohen, 1992)
The authors speculated that the cost of these tools may be prohibitive or they may be
lacking in terms of practicality or face validity.

7

On-Road Driving Assessment

The Vehicle
Another issue related to course design is the vehicle in which the driver undertakes the
evaluation. Although it has been argued that drivers are disadvantaged if not allowed to
drive their own cars, there are a number of advantages to using the same vehicle (i.e.,
examiner’s car) for each evaluation. First, the use of the same car contributes to the
standardization of the evaluation process by ensuring that the technical specifications of
the vehicle are the same for each participant. Perhaps most importantly, cars with dualcontrol brakes or control adaptations (e.g., engine switch-off) enhance the safety of those
in the vehicle. In addition, if the car is fully modifiable, the driver can try various
adaptations before engaging in open road evaluation and before investing in adaptations
to his/her own car.
Critical Driving Behaviors and Scoring of Driving Errors
Critical Driving Behaviors
As noted earlier, the characteristics of the route will necessarily influence the driving
behaviors elicited, or maneuvers undertaken, and scored to determine the outcome of the
assessment (e.g., pass/fail). In the literature reviewed, a number of different critical
driving behaviors, or maneuvers, were identified as important. For example, Justiss,
Mann, Stav, & Velzo’s (2006) behind-the-wheel driving assessment involved broad,
general categories of critical behaviors. Their categories were as follows:
1) Vehicle position (e.g., the position of the vehicle forward and backward in
relation to other vehicles, objects and pavement markings);
2) Lane maintenance (e.g., lateral position of vehicle during driving maneuvers
and while stopped);
3) Speed regulation (e.g., ability to follow speed limits and control over
acceleration and braking of the vehicle);
4) Yielding (e.g., giving the right of way);
5) Visual scanning of the driving environment;
6) Gap acceptance (e.g., choosing the safe distance to cross in front of oncoming
traffic); and
7) Adjustments to stimuli/traffic signs (e.g., the ability to respond appropriately
to driving situations).
Other studies have used more specific maneuvers (e.g., left hand turns, yields, merges,
lane changes) selected because of their associations with older driver crashes. Although
there is wide variation in the types of critical behaviors used across studies, the members
of the Canadian Consensus Conference on Driving Evaluation in Older Drivers reached
strong consensus that certain critical driving behaviors should be included in a
comprehensive driving evaluation (Table 2).

8

On-Road Driving Assessment

Table 2. Critical Driving Behaviors of an on-road driving assessment
Strong Consensus
− Speed maintenance at various speeds according to road conditions and in respect of
signage
− Maintaining lane positioning
− Stopping at red traffic lights
− Stopping at stop signs
− Not stopping at green lights
− Merging at appropriate speed with visual scanning and awareness of critical stimuli
− Appropriate lane positioning during turns Backing up–awareness of surrounding
stimuli by checking behind both sides
− Slowing to potential hazards
− Yielding where appropriate
− Not spending excessive time at intersections
− Respecting a “space cushion” around the car, that is, the distance in front and back of
the vehicle and also on either side of the vehicle
− Maintaining driving performance with introduction of “cognitive burden” (for
example, responding to conversational questions from the evaluator)
Scoring of Driving Errors
Scoring procedures used to evaluate the on-road driving performance of an older adult
vary considerably. There are a number of scoring issues related to the rater (e.g., how
many raters are needed and who should do the rating) and the scoring mechanism/rating
system used (e.g., subjective evaluations and objective measures).
The Rater
Some on-road assessments have used a single rater to score driving performance, while
other studies have used multiple raters (Fox et al., 1998). Fox et al. (1998) recommend
that a driving instructor be seated in the front seat to provide instructions to the driver and
to intervene if necessary to maintain vehicle safety. Meanwhile, the back seat rater can
devote all their attention to observing and rating the driving maneuvers performed by the
examinee. In addition, as Fox et al. (1998), note this permits the rater to have a better
view of some critical aspects of driving that cannot be properly viewed from the front
seat (e.g., visual scanning). On the other hand, others (e.g., Justiss et al., 2006) have
developed scoring procedures that allow one rater to both adequately observe driving
behaviors and intervene to maintain vehicle safety. Justiss et al. (2006) developed a
driving performance form (i.e., checklist of errors under specific maneuvers), based on
feedback from driving rehabilitation specialists, so a single evaluator could observe and
rate driving errors and attend to the driver and driving environment simultaneously.
Often the raters are driving instructors, researchers or occupational therapists. It is
widely agreed that a comprehensive driving assessment, of which on-road assessment is a
component, should be conducted by a health professional that has specific expertise in
9

On-Road Driving Assessment

driving evaluation (Korner-Bitensky et al., 2005). In the United States, the term driver
rehabilitation specialist is used to identify an individual who plans, develops, coordinates
and implements driving services for individuals with disabilities (American Medical
Association, 2003).
The Scoring Mechanisms
The scoring mechanisms used to assess driving performance can be divided into
subjective ratings and objective measurements of driving performance. Subjective
ratings include scoring methods that involve observation and some degree of judgment on
the part of the examiner. Objective measurements are derived from instruments that
record various aspects of driving behavior such as video recorders or Global Positioning
System (GPS) technology, thereby eliminating, or greatly reducing, the need for
subjective judgment on the part of the examiner. The type of route used in the assessment
influences validity of the scoring procedures. In their discussion of scoring procedures
used in driving evaluation of brain injured populations, Fox et al. (1998) point out that
scoring driving behaviors on non-standard routes produces results of questionable
validity, because each driver that is assessed is not exposed to the same opportunities for
making driving errors. Another critical issue related to scoring driving performance was
raised by Brown and Ott (2004) who point out that it is not always possible to assess and
to ascertain how an older driver would handle some driving errors because the evaluator
must intervene to maintain their own safety under certain conditions.
Subjective Ratings
In the literature reviewed, a wide variety of subjective methods of scoring driving
procedures have been employed. The range of procedures used has varied from simple
tallies of errors (or correct maneuvers) made and time taken to execute particular
maneuvers, to more elaborate rating procedures of global performance or individual
driving behaviors (e.g., recording type, severity and location or error, qualitative
descriptions or rating scales; (e.g., recording type, severity and location or error,
qualitative descriptions or rating scales; Fox et al., 1998). Rating scales may vary from 2
point (e.g., safe/unsafe or correct/incorrect) to 5 point scales or larger (Dobbs, Heller, &
Schopflocher, 1998; Fox et al., 1998; Justiss et al., 2006; Mallon & Wood, 2004). Often
assessment of driving performance is based solely on clinical judgment, therefore,
leaving room for discrepant evaluations across different raters (Dobbs, 2005). As
discussed above, standardized methods of testing and scoring improve the validity of onroad evaluation (Dobbs, 2005; Fox et al., 1998).
Objective Measurements
Objective assessment of driving performance is made possible by the use of instrumented
vehicles (Rizzo, Jermeland, & Severson, 2002). An instrumented vehicle is equipped
with instrumentation to take recordings of a variety of aspects of driving (Rizzo et al.,
2002). These technologies include GPS, video-cameras, sensors, accelerometers,
computers, and radar and video lane tracking systems. An instrumented vehicle can be
differentiated from a dual-brake vehicle that is equipped with controls to allow
10

On-Road Driving Assessment

intervention by the examiner. The technologies used in instrumented vehicles can be used
in conjunction with ratings by a driving evaluator or without an observer in the car. The
latter allows for driving to be assessed in a more naturalistic way by eliminating the
concern that drivers might alter their driving behavior due to awareness of their
evaluation (Baldwin, Duncan, & West, 2004; Porter & Whitton, 2002). Many of the
devices used (e.g., GPS receivers, the Driving Monitor System) are relatively inexpensive
and can be installed fairly easily (Baldwin et al., 2004; Porter & Whitton, 2002). Using a
combination of GPS and video cameras allows for measurements of vehicle position,
heading, velocity, acceleration, and deceleration to be obtained, and lane changes,
distance to oncoming cars, and head motion (depending on where the cameras are
mounted) to be monitored. All modern vehicles record speed, emission controls, and
vehicle performance, while some “smarter” cars allow reporting of additional items
including seatbelt use, headlight use, climate and traction control, wheel speed, and ABS
activation (Rizzo et al., 2002).
Classification Systems
Following an on-road driving assessment, older driver are often provided with an overall
or global classification of driving typically, though not always, determined by subjective
evaluation by the examiner, rather than based on a quantified overall driving score
(Justiss et al., 2006). Often this global classification is a simple pass/fail or safe/unsafe
(Fox et al., 1998; Justiss et al., 2006). Others suggest broader classification systems that
identify those with potential for remediation. For example, Justiss et al. (2006) rated
overall performance on their behind-the-wheel assessment on a 4 point scale: safe (3)
safe with restrictions or recommendations (2), unsafe remediable (1) and unsafe not
remediable (0). Participants in the Canadian Consensus Group on Driving Evaluation of
Older Drivers strongly agreed that evaluation outcome be classified into three categories:
Pass (performed safely); Poor/Indeterminate (with possible potential for remediation);
and Do Not Pass (unsafe/non-remediable; Korner-Bitensky et al., 2005).
Other important considerations in the assessment of older drivers
Retraining
Incorporating retraining into open road evaluation is another approach to driving
assessment that was proposed by Brouwer and Von Zomeren in 1992. They suggested
training impaired drivers in strategic (e.g., route planning) and tactical level (i.e.,
maneuvers) compensatory behaviors until training goals are reached or a preset
maximum number of lessons is exceeded. In this context, unsafe driving would be
defined as the inability to reach the criterion as a safe driver within the limits of available
training and technical adaptations. This approach has the advantage of “building in”
retaining to the driving assessment procedure so that the driver is given every opportunity
to improve. The disadvantages of this approach are the potential cost of providing the
lessons, especially since there is little published research to guide in the implementation
of such a program.

11

On-Road Driving Assessment

Other components of a comprehensive older driver evaluation
There are many related assessment procedures that are employed to evaluate or predict
real world driving performance. These include office-based assessments of drivingrelevant abilities (e.g., cognitive, physical, motor, health related factors and medical
conditions), tests of driving specific knowledge (e.g., rules of the road, traffic sign
recognition), driving simulator testing, and self assessment or report from others (e.g.,
questionnaire about driving ability, computerized self-assessment programs). Although
many of these procedures (e.g., performance on cognitive measures or driving simulators)
have not been found to reliably predict real-world driving performance (e.g., Dubinsky,
Stein, & Lyons, 2000; Duchek et al., 2003; Whelihan, DiCarlo, & Paul, 2005), the
participants of the Canadian Consensus on Driving Evaluation in Older Drivers
recommend that a comprehensive driver evaluation is needed and that it include both offand on-road methods of evaluation (Korner-Bitensky et al., 2005). The group achieved
moderate to strong consensus regarding the components to be included in a
comprehensive driving evaluation (Table 3). For specific choice of tools to be included in
the off-road assessment portion of a comprehensive driver evaluation, see KornerBitensky et al. (2005). A discussion of the advantages and disadvantages of these related
assessment methods is outside the scope of this report but the strengths and weaknesses
of these methods can be found in the papers by Fox et al. (1998), Rizzo, Reinach,
McGehee and Dawson (1997), Rizzo, McGehee, Dawson and Anderson (2001), and The
British Psychological Society (2001).
Table 3. Comprehensive Driving Evaluation
Strong Consensus
Pre-road driving assessment should be conducted by same individual who is conducting
the on-road assessment.
The pre-road driving assessment should include measures of cognition, visual perception,
and physical and motor status.
Driving specific information should be gathered including driving history, medical
history, self-perception of driving performance, and physical and motor status.
A comprehensive test of driver knowledge should be included in a pre-road assessment.
Moderate Consensus
The pre-road driving assessment should include measures of vision, reaction time, and
behavior.
An on-road driving assessment should not be conducted if poor results are obtained on
pre-road assessment tools.
A closed route course should be conducted with individuals whose pre-road assessment
suggests a high risk.

Limitations of on-road assessment
On-road driving assessment has several important weaknesses including increased safety
risk for the driver, examiner, and other road users and possibly biased evaluations (e.g.,

12

On-Road Driving Assessment

Fox et al., 1998; McKenna, Jefferies, Dobson, & Frude, 2004; Schultheis, Hillary, &
Chute, 2003). In addition, standard on-road examination procedures may involve less
demanding environmental conditions (e.g., predetermined routes, less busy times) and
decision making processes (e.g., instructions/cues provided by examiner) compared to the
real world (Shultheis et al., 2003; The British Psychological Society, 2001). Moreover,
driving observed during an on-road testing situation may not accurately represent an
individual’s true behavior in everyday driving situations (e.g., more caution, motivation
to drive safely, awareness of evaluation; Fox et al., 1998; Pietrapiana et al., 2005; The
British Psychological Society, 2001). Despite these disadvantages, open road driving
assessment is generally agreed upon as the most ecologically valid method to evaluate
driving ability of older drivers, particularly if undertaken within the context of a
comprehensive older driver evaluation.
CONCLUSIONS
The components included in an on-road assessment will depend on the specific target
group of older adults to be assessed and the intended purpose of the evaluation. Many
different approaches are reflected in the literature and some well-developed standardized
measures are available, though none are generally accepted as the “gold standard”.
RECOMMENDATIONS
Most of the following recommendations are made based on the assumption that the intent
of the on-road driving assessment is to make recommendations regarding the safety of
older drivers. We have made these recommendations under the headings contained in the
body of the report. It is recommended that:
Target Population
• the target population that will take part in the older driver assessment and the
purpose of the assessment be clearly articulated and defined. The requisite
elements for inclusion in the assessment will be dependent on the target group(s)
and the purpose;
Course Design
• a brief closed course examination of vehicle handling skill precede any on-road
driving assessment to identify those with inadequate skills to enter traffic and to
allow for vehicle modifications to be fitted for those that need them before
entering traffic;
• a 45-60 minute on-road evaluation be employed;
• the Critical Components of an On-Road Driving Assessment from the Canadian
Consensus Conference on Driving Evaluation in Older Drivers be used in the
development of on on-road assessment for use with older adults. This will allow
for consistency with other sites across Canada and facilitate international
interactions;
• the critical components (i.e., route characteristics) selected for the assessment
should not be so difficult as to contrast unfavorably with ordinary licensing test

13

On-Road Driving Assessment

•

•
•
•

but challenging enough to allow manifestation of any cognitive or perceptual
deficits incompatible with safe traffic participation;
a standardized procedure that has a fixed route with predefined driving behaviors
(maneuvers) and scoring criteria be used. This procedure should be conducted in
the same manner for all older drivers to the extent possible given constraints due
to disability and should be conducted in as similar an environment as possible. It
is imperative that assessments used for making recommendations regarding the
safety of drivers be reliable and valid;
driver retraining be built into the assessment with a maximum number of lessons
specified before retraining is terminated;
a dual control car that is fully modifiable be used to conduct on-road driving
assessments. This is essential for safety reasons and safety overrides concerns
about validity of using vehicle unfamiliar to driver;
a self-directed navigation component be considered for inclusion in the
assessment. This task requires the participant to actively engage in multiple tasks
involving visual scanning, divided attention, planning and judgment. This requires
a much higher degree of cognitive and visual ability than simply following
directions;

Critical Driving Behaviors and Scoring of Driving Errors
•

•
•

•
•
•
•
•

the Critical Behaviors During On-Road Driving Assessment from the Canadian
Consensus Conference on Driving Evaluation in Older Drivers be used in the
development of an on-road assessment for use with older adults. This will allow
for consistency with other sites across Canada and facilitate international
interactions;
the driving behaviors sampled for use in the assessment must be sufficient in
terms of duration and complexity to allow observation of several driving
situations and maneuvers. This will enhance reliability and validity;
two raters, a driving instructor seated in the front seat to provide instructions to
the driver and to intervene if necessary to maintain vehicle safety and an
independent rater in the back seat, be employed to ensure accurate scoring of
driving errors;
other technologies be used only as determined by the purpose of the assessment;
there be a large number of independently scored items, to enhance reliability and
provide independence of one score from the next;
the scoring procedure be simple with specification of the correct responses to
enhance reliability and ensure identical standards of performance are used for
each assessment;
the outcome of the driver assessment be classified into three categories: Pass
(performed safely); Poor/indeterminate (with possible potential for remediation);
Do Not Pass (unsafe/non-remediable);
specific information about driving behaviors be obtained to determine needed
vehicle modifications, training requirements and compensatory potential.

Other important considerations in the assessment of older drivers
14

On-Road Driving Assessment

•
•
•
•

in keeping with the Canadian Consensus group, related off-road assessment
procedures, along with an on-road evaluation, be considered for inclusion in a
comprehensive driving evaluation of older adults;
off-road/pre-road driving assessment procedures not be used to make a decision
about an individual’s driving safety on their own;
off-road assessment methods be used to screen for individuals who might be
appropriate for on-road driving testing or to supplement information provided by
an on-road assessment;
research on the development and implementation of an older driver assessment
program be undertaken to:
o ascertain validity of the assessment procedure for ethical and legal purposes;
o monitor and improve program design (ongoing program evaluation);
o examine driving ability in different diagnostic groups (e.g., dementia, stroke,
TBI) with the aim of characterizing deficient driving skills ;
o examine the relations between off-road assessment procedures (e.g.,
psychological and medical tests) and on-road driving performance;
o examine the criterion validity of standardized on-road assessment including
sensitivity and specificity of judgment of safe driving as this is an area that
requires further development in healthy as well as impaired populations;
o examine on-road assessment as a criterion measure for intervention/training
effects (e.g., classroom education, simulators training, etc).

15

On-Road Driving Assessment

Appendix A: Selected standardized measures
Name
Developer
Where used
Driver
Jones (1978)
University of
Performance
Southern
Test
California

The Miller
Road Test

Carr et al.
(1992)

Washington
University, St.
Louis, Missouri

Performancebased Driving
Evaluation
(PDE)

Odenheimer et
al. (1994)

Geriatric
Research
Education and
Clinical Center,
Department of
Veterans
Affairs, West
Roxbury,
Massachusetts
Department of
Veterans
Affairs,
California

Fitten et al.
(1995)

The
Washington
University
Road Test
The
DriveABLE
Road Test

Hunt et al.
(1997)

Dobbs et al.
(1998)

Description
-reliability and standardization
was examined in normal drivers;
- standardized route, several
critical maneuvers rated several
times,;
-observation of about 110 items of
behavior
-administered to various groups of
drivers (e.g., different ages) and
examined for test-retest reliability
-used with older and younger
adults
-not conducted within the
complexity of an open-road
environment
-a fixed route, in traffic, and in
range of driving situations;
-68 scored tasks
-the correlation between the intraffic score on the PDE and a
driving instructor’s global rating
was .74.

-hospital internal roadway to
ensure low-level traffic
conditions;
-2.7 km in length;
-six stages, each presenting a
different degree of driving
complexity;
-wide-angle camera on mounted
on the roof of the car;
-scoring criteria for specific
performance and more general
aspects of driving.
Washington
-partially standardized evaluation
University, St.
where the driver follows a preLouis, Missouri determined route in low-volume
traffic.
Developed in
-most well-developed of the
Edmonton; now standardized road tests;
used in many
-Originally, consisted of a closed
provinces and
course with emergency-type

16

On-Road Driving Assessment

states

situations and an open road course
with a driving instructor and an
observer in the car as well as
video recorders mounted on the
car (one to record driver, one to
record the road);
-dual control car used;
-people of different age groups
and older adults with and without
cognitive impairment;
-participants also administered
neuropsychological and
rehabilitation evaluations to
ensure the control participants
were mentally and physically
normal;
-this information was also used to
examine the relations between
these measures and the open road
measures;
-150 driving errors were scored - -the errors that distinguished
between normal and impaired
groups were identified as
competence-defining errors;
-frequency, severity and locations
in the course where these errors
were made were examined and
used to set rules for laying out a
road course that would have the
attributes necessary for effective
evaluations of medically at-risk
drivers;
-to avoid discrimination, the fail
criterion was set so that the person
was making errors that place him
outside the range of competencedefining errors displayed by
healthy, normal drivers in that
situation.
-closed course and self-directed
navigation instructions (i.e., take
me to …) did not contribute to the
discrimination of the impaired
versus healthy groups and so were
abandoned;

17

On-Road Driving Assessment

-have tackled a number of the
challenges of standardization,
including the need for procedures
to ensure every person receives
the same evaluation outcome
regardless of testing site and the
constant monitoring of the
administration of driving tests.
-recently, this procedure has been
validated for a large group people
across a wide range of ages that
manifested a variety of medical
conditions;
-new course for road-test was
developed and validated against
an off-road (in office) test battery;
-important feature of this test is
that by identifying the attributes
of a road course that reveal
competence-defining driving
errors of medically impaired
drivers, criteria for “unsafe to
drive” meet the requirements of
the Supreme Court of Canada’s
ruling concerning ‘reasonable
road safety’.
Mallon &
Wood (2004)

Queensland
University of
Technology,
Brisbane,
Australia

18

-similar to PDE of Odenheimer et
al. (1994) with the addition of
self-directed navigational
component to highlight
difficulties of older drivers;
-15 km, predetermined route in
city and suburban streets, with
simple and complex intersections,
and a range of traffic densities;
-operational maneuvers are
performed at pre-designated
locations;
- a spectrum of situations that
drivers would typically encounter
when they interact with traffic on
a daily basis;
-for 16% of locations, the
participant was asked to find
her/his way to specific destination

On-Road Driving Assessment

Behind-thewheel Driving
Performacne
Assessment

Justiss et al.
Velozo, (2006)

National Older
Driver
Research and
Training
Centre,
University of
Florida

19

(self-directed navigation task),
requiring the participant to
actively engage in multiple tasks
involving visual scanning, divided
attention, planning and judgment;
-all their groups (differing by age
and visual impairment) performed
better when followed directions,
all showed more errors under selfdirected navigation and errors in
self-directed navigation increased
by age and visual impairment;
-as yet unknown whether the task
will discriminate between healthy
and impaired drivers
-driving behaviors (vehicle
position, lane maintenance, speed
regulation, yeilding, signaling,
visual scanning, adjustment to
stimuli and gap acceptance) on a
fixed route with low, moderate
and high grades of maneuver
complexity
- scored from 0 (physical
intervention required) – 3 (no
errors
-examined criterion validity,
interrater reliability, internal
consistency, and test-retest
reliability

On-Road Driving Assessment

REFERENCES

American Medical Association (2003). Physicians guide to assessing and counseling
older drivers. Retrieved 20 December 2007.
Anstey, K. J., Wood, J., Lord, S., & Walker, J. G. (2005). Cognitive, sensory and
physical factors enabling driving safety in older adults. Clinical Psychology
Review, 25(1), 45-65.
Ashman, R., Bishu, R., Foster, B., & McCoy, P. (1994). Countermeasures to improve the
driving performance of older drivers. Educational Gerontology, 20(6), 567-577.
Association, A. M. (2003). Physicians guide to assessing and counseling older drivers.
Retrieved 20 December 2007.
Baldwin, K. C., Duncan, D. D., & West, S. K. (2004). The driver monitor system: A
means of assessing driver performance. Johns Hopkins APL Technical Digest,
25(3), 269-277.
Brouwer, W., & Van Zomeren, E. (1992, Oct 5-6). Evaluation of driving in traumatically
brain-injured persons. Paper presented at the Second International Congress on
Objective Evaluation in Rehabilitation Medicine, Montreal, Canada.
Brown, L. B., & Ott, B. R. (2004). Driving and dementia: A review of the literature.
Journal of Geriatric Psychiatry and Neurology, 17(4), 232-240.
Carr, D., Jackson, T. W., Madden, D. J., & Cohen, H. J. (1992). The effect of age on
driving skills. Journal of the American Geriatrics Society, 40(6), 567-573.
De Raedt, R., & Ponjaert-Kristoffersen, I. (2000). Can strategic and tactical
compensation reduce crash risk in older drivers? Age and Ageing, 29(6), 517-521.
Dobbs, A. R. (2005, June 5-8). The development of a scientifically based driving
assessment and standardization procedures for evaluating medically at-risk
drivers Paper presented at the Canadian Multidisciplinary Road Safety
Conference XV, Fredericton, NB.
Dobbs, A. R., Heller, R. B., & Schopflocher, D. (1998). A comparative approach to
identify unsafe older drivers. Accident Analysis and Prevention, 30(3), 363-370.
Dubinsky, R. M., Stein, A. C., & Lyons, K. (2000). Practice parameter: risk of driving
and Alzheimer's disease (an evidence-based review): report of the quality
standards subcommittee of the American Academy of Neurology. American
Academy of Neurology, 54, 2205-2211.
Duchek, J. M., Carr, D. B., Hunt, L., Roe, C. M., Xiong, C., Shah, K., et al. (2003).
Longitudinal driving performance in early-stage dementia of the Alzheimer type.
Journal of the American Geriatrics Society, 51(10), 1342-1347.
Eby, D. W., Trombley, D. A., Molnar, L. J., & Shope, J. T. (1998). The assessment of
older drivers' capabilities: A review of the literature: University of Michigan:
Transportation Research Institute.
Fitten, L. J., Perryman, K. M., Wilkinson, C. J., Little, R. J., Burns, M. M., Pachana, N.,
et al. (1995). Alzheimer and vascular dementias and driving - a prospective road
and laboratory study. Journal of the American Medical Association, 273(17),
1360-1365.
Fox, G., Bowden, S., & Smith, D. (1998). On-road assessment of driving competence
after brain impairment: Review of current practice and recommendations for a

20

On-Road Driving Assessment

standardized examination. Archives of Physical Medicine and Rehabilitation, 79,
1288-1296.
Hunt, L. A., Murphy, C. F., Carr, D., Duchek, J. M., Buckles, V., & Morris, J. C. (1997).
Environmental cueing may affect performance on a road test for drivers with
dementia of the Alzheimer type. Alzheimer Disease & Associated Disorders, 11,
13-16.
Johansson, K., & Lundberg, C. (1997). The 1994 International Consensus Conference on
Dementia and Driving: A brief report. Alzheimer Disease & Associated
Disorders, 11, 62-69.
Jones, M. (1978). Driving performance measures for safe performance curriculum (No.
78-3). Los Angeles: University of Southern California.
Justiss, M. D., Mann, W. C., Stav, W., & Velozo, C. (2006). Development of a behindthe-wheel driving performance assessment for older adults. Topics in Geriatric
Rehabilitation, 22(2), 121-128.
Korner-Bitensky, N., Bitensky, J., Sofer, S., Man-Son-Hing, M., & Gelinas, I. (2006).
Driving evaluation practices of clinicians working in the United States and
Canada. American Journal of Occupational Therapy, 60(4), 428-434.
Korner-Bitensky, N., Gelinas, I., Man-Song-Hing, M., & Marshall, S. (2005).
Recommendations of the Canadian Consensus Conference on driving evaluation
in older drivers. Physical & Occupational Therapy in Geriatrics, 23(2/3), 123144.
Kua, A., Korner-Bitensky, N., Desrosiers, J., Man-Son-Hing, M., & Marshall, S. (2007).
Older driver retraining: A systematic review of evidence of effectiveness. Journal
of Safety Research, 38(1), 81-90.
Mallon, K., & Wood, J. M. (2004). Occupational therapy assessment of open-road
driving performance: Validity of directed and self-directed navigational
instructional components. American Journal of Occupational Therapy, 58(3),
279-286.
Marottoli, R. A., Richardson, E. D., Stowe, M. H., Miller, E. G., Brass, L. M., Cooney, L.
M., et al. (1998). Development of a test battery to identify older drivers at risk for
self-reported adverse driving events. Journal of the American Geriatrics Society,
46(5), 562-568.
McCoy, P., Tarawneh, M., Bishu, R., Ashman, R., & Foster, B. (1993). Evaluation of
countermeasures for improving driving performance of older drivers.
Transportation Research Record, 1405, 72-80.
McKenna, P., Jefferies, L., Dobson, A., & Frude, N. (2004). The use of a cognitive
battery to predict who will fail an on-road driving test. British Journal of Clinical
Psychology, 43, 325-336.
Medicine, A. S. f. G. (2003). Australian Society for Geriatric Medicine Position
Statement No. 11. Driving and dementia. Australian Society for Geriatric
Medicine, 22, 46-50.
Odenheimer, G. L., Beaudet, M., Jette, A. M., Albert, M. S., Grande, L., & Minaker, K.
L. (1994). Performance-based driving evaluation of the elderly driver: Safety,
reliability, and validity. Journal of Gerontology. Medical Sciences, 49(4), M153159.

21

On-Road Driving Assessment

Ostrow, A., Shaffron, P., & McPherson, K. (1992). The effects of a joint range-of-motion
physical fitness training program on the automobile driving skills of older adults.
Journal of Safety Research, 23(4), 207-219.
Patterson, C., Gauthier, S., Bergman, H., Cohen, C., Feightner, J., Feldman, H., et al.
(2001). The recognition, assessment and management of dementing disorders conclusions from the Canadian Consensus Conference on Dementia. Canadian
Journal of Neurological Sciences, 28(Suppl 1), S3-S16.
Pietrapiana, P., Tamietto, M., Torrini, G., Mezzanato, T., Rago, R., & Perino, C. (2005).
Role of premorbid factors in predicting safe return to driving after severe TBI.
Brian Injury, 19(3), 197-211.
Porter, M. M., & Whitton, M. J. (2002). Assessment of driving with the global
positioning system and video technology in young, middle-aged, and older
drivers. Journals of Gerontology Series a-Biological Sciences and Medical
Sciences, 57(9), M578-M582.
Rizzo, M., Jermeland, J., & Severson, J. (2002). Instrumented vehicles and driving
simulators. Gerontechnology, 1(4), 291-296.
Rizzo, M., McGehee, D. V., Dawson, J. D., & Anderson, S. N. (2001). Simulated car
crashes at intersections in drivers with Alzheimer disease. Alzheimer Disease and
Associated Disorders, 15(1), 10-20.
Rizzo, M., Reinach, S., McGehee, D., & Dawson, J. (1997). Simulated car crashes and
crash predictors in drivers with Alzheimer disease. Archives of Neurology, 54(5),
545-551.
Rudman, D. L., Friedland, J., Chipman, M. L., & Sciortino, P. (2006). Holding On and
Letting Go: The Perspectives of Pre-seniors and Seniors on Driving SelfRegulation in Later Life. Canadian Journal on Aging, 25(1), 65-76.
Schultheis, M., Hillary, F., & Chute, D. (2003). The Neurocognitive Driving Test:
Applying technology to the assessment of driving ability following brain injury.
Journal of Head Trauma Rehabilitation, 17(1), 38-47.
The British Psychological Society (2001). Fitness to drive and cognitive: A document of
the multidisciplinary working party on acquired neuropsychological deficits in
fitness to drive 1999. Leicester.
Whelihan, W. M., DiCarlo, M. A., & Paul, R. H. (2005). The relationship of
neuropsychological functioning to driving competence in older persons with early
cognitive decline. Archives of Clinical Neuropsychology, 20(2), 217-228.

22