Naturalistic Studies (Driving In Traffic) continued...
The last on-road study including older drivers considered in this review was conducted by Mallon and Wood (2004). In this study, 137 participants in two vision groups and three age groups underwent an on-road driving assessment in a dual-brake vehicle. Ninety participants had normal vision and were divided into three age groups: young (mean age = 27); middle-aged (mean = 52), and older (mean = 68.9). A second group of older participants (mean age = 71) were diagnosed with visual impairment resulting from ocular disease. Prior to the on-road test, subjects underwent a 20- to 30-minute cognitive assessment using the Barry Rehabilitation Inpatient Screening of Cognition.
Driver performance was assessed for each subject by an occupational therapist and a driving instructor, both with a specialization in driver assessment. The route and maneuvers are described by the authors as “having sufficient duration and complexity to allow assessment of a variety of driving situations and maneuvers, and sufficiently challenging to allow manifestation of visual or cognitive deficits or both.” It was designed based on clinical open-road assessments in use by occupational therapists of a major rehabilitation center in Australia, closely matching an assessment by Odenheimer et al. (1994) where test scores correlated highly with cognitive ability, and high internal reliability was shown. The 15-km route involved particular driving situations and associated maneuvers at predetermined locations, and required approximately 50 minutes to complete. It consisted of city and suburban streets, simple and complex intersections, and exposed drivers to a range of traffic densities.
The occupational therapist recorded driving performance across a sequence of 106 locations. The associated driving situations and maneuvers were divided into the following nine categories, with number of locations noted in parentheses: roundabouts (3); merging (2); car parking (5); traffic-light-controlled intersections (20); non-traffic-light-controlled intersections, stop and yield (13); reversing (1); emergency brake (1); straight driving on single and dual carriageway (50); and lane changing (11). At each location, seven components of driving performance were assessed that included: general and blind spot observation, indication, braking-acceleration, lane positioning, gap selection, and approach. Failure of any aspect of performance resulted in failure of the whole task for a given location/maneuver.
For 84 percent of the locations (89 out of 106), the participant drove as instructed by the driving instructor (“directed navigation”). For 16 percent of the locations, participants were asked to find their way to a specific destination (“self-directed navigation”), using cues such as road signs and road markings to determine what route to take. Road test performance was defined as the overall score for the number of correct maneuvers out of a total of 106. In addition, the driving instructor provided an overall global rating of safety ranging from 1 to 10.
In Mallon and Wood’s study, all subjects performed better under instructor-directed navigation tasks than under the self-directed navigation tasks. There was a significant effect of age, with the older subjects performing worse on both the instructor-directed and self-directed navigation tasks than the two younger age groups. The older drivers with visual impairments made significantly more errors under both directed and self-directed navigation. Considering the number of errors made by each group for each category of location/maneuver, there were significant group differences at all locations except emergency braking. The young and middle-aged drivers performed better than the older drivers in both vision groups, and the older drivers with ocular disease performed worse than the normal-sighted older drivers for merging and straight driving locations. Age group differences in road test performance were greatest for traffic-signal controlled and non-signal controlled intersections, roundabouts, merging, lane changing, and straight driving.
The driving scores measured by the occupational therapist and the driving instructor were significantly positively related. The cognitive test scores and the occupational therapist’s assessment of driving performance were significantly positively related, as were the cognitive test scores and the self-directed component driving score. The cognitive test scores were also significantly positively related to the driving instructor’s safety rating. The cognitive tests, however, explained only 4.8 percent of the variance in the occupational therapist’s scores for directed navigation and 14.4 percent of the variance in the occupational therapist’s scores for self-directed navigation. This highlights the fact that cognitive tests cannot be substituted for on-road tests for determinations of fitness to drive.
The road test used in this study was sensitive to the number and types of driving errors made, and the locations/maneuvers where these errors occurred, making it a strong candidate for assessment of fitness to drive for future studies of older drivers and polypharmacy. The assessment reflected only the numbers and types of errors, however, not their severity. The study authors are currently examining a modified on-road scoring assessment that uses weighted criteria to denote degree of severity of driver errors.