Demographic, medical, functional and physical performance data were collected on 174 Alabama drivers age 55 or older in order to study the factors associated with vehicle crashes among older drivers. Initial data were collected in 1991 and subjects were then followed through 1996 for incident vehicle crashes. 61 subjects experienced between one and four police-reported vehicle crashes during the study period. Variables related to function, medication use, affect, neurological disease and visuocognitive skills were associated with vehicle crash involvement in this cohort. After adjusting for age, race, days driven per week and gender, Cox proportional-hazards models showed that the following variables were associated with crash involvement: at least one crash before 1991; reported difficulty with yard work or light housework, or opening a jar; use of hypnotic medication; failing the useful field-of-view test; self-reported stroke or transient ischemic attack; and scoring within the depressed range on the Geriatric Depression Scale. These findings suggest that multifactorial assessments should be used to identify at-risk older drivers.
Abstract