To describe the characteristics of brain-injured patients who were referred to a simulator-based driving evaluation, evaluate the different components of driving assessment and to search for factor of success on the driving simulator assessment.

We retrospectively analyzed data of brain-injured patients (205 after stroke and 66 after traumatic brain injury (TBI)) who performed a driving simulator assessment of fitness to drive in the neurorehabilitation unit of the Lille University Hospital. Data from disease characteristics, physical, visual and cognitive deficiencies and evaluation of the patient's behavior during the driving evaluation were collected.

Before the driving assessment, about 30% of either stroke or TBI patients resumed driving without any legal agreement and 20% declared they could present driving difficulties. Half of the stroke patients and 38% of TBI ones showed a motor deficit (most of the time a hemiparesia) and 19% a visual field defect. In the driving assessment, hand-foot coordination, visual exploration, divided attention, speed of the decision process and adaptability to unexpected events were more frequently impaired. In both groups, half of the patients were considered as able to drive on the basis of the synthesis of clinical and simulator data, whereas about 15% were not; a conclusion was not possible for the rest of the population who performed an on-road evaluation, a second simulator-based assessment or training driving sessions. Ability to drive was significantly associated with the patient's self-assessment (P=0.026) and the proxy's assessment (P=0.001), the presence of a visual field defect (P=0.001) and most of qualitative physical and cognitive evaluations during simulator driving assessment (P<0.01). Correlations between neuropsychological assessment and visual field hand and attention in one and behavioral evaluation in driving assessment are being processed. About 2/3 of patients have effectively resume driving (max 5 years follow-up). A significant proportion of those who showed objective limitations in the simulator-based assessment were subsequently allowed to drive by authorities, which raises the problem of the separation between assessment and decision processes.

Simulator-based driving assessment is of interest to assess fitness to drive in brain-injured patients.