Analyses were conducted to determine the effectiveness of air bags on driver fatalities using Fatal Accident Reporting System (FARS) data fro 1985-1993. Two types of analyses were used to estimate air bag effectiveness. Driver fatalities in frontal crashes were compared with driver fatalities in nonfrontal crashes for cars with manual or automatic door-mounted lap-shoulder belts and cars with those belts plus a driver air bag. Driver fatal crash rates per 10,000 registered vehicles were compared for cars with manual or automatic door-mounted lap/shoulder belts and cars with those belts plus a driver air bag. In the analysis based on the location of damage, the likelihood of a driver fatality was reduced by 14 percent in cars with air bags and belts compared with the same cars with belts only, based on 23 percent fewer than expected fatalities in front and front-angle crashes. In the analysis based on vehicle registration data, fatality rates for front and front-angle crashes were reduced by 19 percent in cars with air bags and belts compared with the same cars equipped with belts only, and by 15 percent in all crashes. Restricting the comparison vehicles and years to account for differences in vehicles age and increasing belt use produced slightly lower estimated: 14 percent reduction in front and front-angle crashes and 13 percent reduction in all crashes. (A)
Abstract