A 2009 study by the Insurance Institute for Highway Safety found that midsize SUVs with stronger roofs, as measured in quasi-static tests, had lower risk of ejection and lower risk of injury for nonejected drivers. The objective of the present study was to determine whether a similar association exists for other vehicle groups.

Twelve small passenger cars were evaluated according to Federal Motor Vehicle Safety Standard 216 test conditions extended to 10 inches of plate displacement. Crash databases in 14 states provided more than 20,000 single-vehicle rollover crashes involving these vehicles. Logistic regression analyses were used to evaluate the effect of roof strength on the rate of driver injury while assessing and controlling for the effects of driver age, vehicle stability, state, and other factors where necessary.

Small cars with stronger roofs had lower overall rates of serious injury, lower rates of ejection, and lower rates of injury for nonejected drivers. Although the effect on ejection was somewhat smaller for cars than for SUVs, the overall pattern of injury results was consistent. For roof strength-to-weight ratio measured at 5 inches (SWR5), a one-unit increase (e.g., from 2.0 to 3.0) was associated with a 22% reduction in risk of incapacitating or fatal driver injury in single- vehicle rollovers. This compares with a 24% reduction estimated for a similar change in roof strength among midsize SUVs.

The association between vehicle roof strength and occupant injury risk in rollover crashes appears robust across different vehicle groups and across roof SWR5 values, varying from just more than 1.5 to just less than 4.0. If roofs were to increase in strength by one SWR5, a 20-25% percent reduction in risk of serious injury in rollovers would be expected. Still, even if all vehicle roofs were as strong as the strongest roof measured, many rollover injuries still would occur, indicating the need for additional research and countermeasures.