Objective: To estimate the association between past medical contacts and the risk of vehicle collision in a population of older drivers from the province of Quebec, Canada.
Design: Case-crossover study.
Setting: Quebec.
Participants: 111 699 older drivers involved in at least one vehicle collision between January 1988 and December 2000.
Main outcome measures: For each driver, the risk of having a vehicle collision while exposed and not exposed to a medical contact was compared. Separate conditional logistic regression analyses were conducted for all drivers and in four diagnostic-specific subgroups.
Results: The study found a weak but statistically significant increased risk of all collisions being associated with a medical contact within 1 month before the collision, for all drivers (OR=1.10, 95% CI 1.08 to 1.11) and for drivers with diabetes (OR=1.07, 95% CI 1.03 to 1.11).
Conclusion: Older drivers who have a collision are more likely to have been in contact with a physician shortly before the collision. These findings suggest that there might be an opportunity to detect medical conditions that put older drivers at higher risk of collision; however, further research is needed to assess the potential effectiveness and practical modalities of screening.