Objectives: To compare the frequency of motor vehicle crashes of drivers aged 50 and older with a diagnosis of dementia with that of a group without dementia in the 3 years before and 3 years after an index hospital admission using the Western Australian Data Linkage System (WADLS).
Design: Retrospective population-based study.
Setting: De-identified data were obtained from Western Australian Hospital Morbidity Data System and the Western Australian Death Registrations using the WADLS from 2004 to 2010. The Integrated Road Information System was used to identify individuals involved in a crash as the driver from 2001 to 2013.
Participants: Individuals with dementia with an index hospital admission (n = 1,666, 34%) and individuals without dementia (n = 3,636, 66%) who had been involved in at least one motor vehicle crash as the driver from 2001 to 2013.
Measurements: Involvement in a police-reported crash as the driver.
Results: The occurrence of one or more crashes as the driver in the dementia group (43% had a crash as the driver) was higher in the 3 years before the index hospitalization than in the comparison group (30% had a crash as the driver). The risk of a crash was 93% less for those with dementia in the 3 years after an index hospital admission with dementia than for those without dementia (incidence rate ratio = 0.07, 95% confidence interval = 0.06-0.09) compared to the previous 3 years, after adjusting for relevant confounders.
Conclusion: Although older drivers may give up driving after a diagnosis of dementia, they may be at greater risk of crashing before diagnosis or in the early stages of dementia. Better methods are needed to identify at-risk drivers with early dementia and prevent crashes.
Keywords: data linkage; dementia; motor vehicle crashes.
© 2016, Copyright the Authors Journal compilation © 2016, The American Geriatrics Society.