The Driver's Role in Fatal Two-Car Crashes: A Paired “Case-Control” Study
Level 4 - case-series / case-control
Paired case-control study design using registry data
OpenAlex W1898908442 · doi:10.1093/oxfordjournals.aje.a116017
What was done
This population-based paired case-control study analyzed driver characteristics associated with driving errors resulting in fatal motor vehicle collisions. Data from the Fatal Accident Reporting System were evaluated for 6,506 two-car crashes (81% of 7,993 eligible events), comparing the driver who initiated the crash (case) against the passively involved driver (control) within the same event to control for environmental conditions and traffic exposure.
What was found
Alcohol was the strongest predictor of crash initiation (OR = 11.5; 95% CI 9.57-13.9), with risk elevated at the lowest blood alcohol concentrations and rising as levels increased. Age showed a U-shaped association, with lowest initiation risk among drivers aged 40-49 years, rising to an OR of 3.35 (95% CI 2.72-4.13) in teenagers and 22.1 (95% CI 14.2-34.5) in drivers over 80 years. Additional risk factors were driving without a valid license (OR = 2.16; 95% CI 1.72-2.73), not wearing a seat belt (OR = 1.54; 95% CI 1.35-1.75), and having a crash within the prior year (OR = 1.21; 95% CI 1.07-1.38).
Why it matters
Using a paired within-crash design eliminates confounding by environmental conditions and exposure, demonstrating that fatal driving errors are strongly concentrated in specific high-risk driver demographics and behaviors.
Limits
The study is limited to fatal two-car collisions and may not generalize to single-vehicle or non-fatal crashes. Reliance on routine registry data introduces potential misclassification of crash initiation or underreporting of alcohol use, and 19% of eligible fatal crashes were excluded.
Cited by
- supports The distribution of motor vehicle accidents by driver age follows a U-shaped function rather than a linear distribution.