Insurers, regulators at odds over auto crash tests, safety

By Jane O'Donnel
USA Today

A vehicle's score in a highly publicized insurance industry crash test may not be a good indicator of the chance of survival in a crash, the results of a new study suggest.

Occupants in vehicles rated good are three times less likely to die if they hit a vehicle that is rated poor in the Insurance Institute for Highway Safety's test. But occupants of vehicles rated anything other than poor have about the same survival chances.

The timing and results of the institute's new study, released yesterday, are important because federal regulators said this week they are concerned the test results may increase demands for large vehicles that are more threatening to small vehicles in crashes.

The National Highway Traffic Safety Administration says it has found that bigger vehicles that do better in the insurance institute's crash test are sometimes built stiffer and therefore potentially more dangerous to smaller vehicles.

Adrian Lund, the institute's chief operating officer, dismisses the NHTSA concerns, saying there is "absolutely no credible evidence" to suggest vehicles with better scores are doing more harm to other autos.

The insurance institute says its offset test — so named because a portion of the front end of a vehicle hits part of a barrier — is a better simulation of a common real-world crash than the head-on crash test done by the government.

NHTSA said Tuesday that its research found only that occupants in a vehicle with a good score on an offset test might have a reduced risk of lower-limb injuries and possibly neck injuries in small adults.

The insurance institute has been struggling to prove the life-saving value of its test for more than two years. Its study is the first to compare frontal crash test results with real crashes that are reported in a federal fatality database.

Vehicles can be rated good, acceptable, marginal or poor. Very few are rated poor these days.

The insurance institute acknowledged that the link between its test results and a reduction in fatalities was most evident in frontal crashes between similar-sized vehicles.

David Harless, a Virginia Commonwealth University economist who studies car crashes, says the insurance institute's study shows its ratings are of only limited use.

Harless said the obvious question is: "What meaning do the crash test scores have if I don't happen to meet these circumstances (and) am in a car-to-pickup crash?"

"To the extent that consumers look to these ratings, they think of them as being a general indicator of safety in a serious crash," he said.

Lund acknowledges there are "an awful lot of kinds of crashes." But he says that while the findings don't "translate directly into other kinds of crashes ... , the additional protection would translate."

Automakers, who redesigned many vehicles to do well in tests, have mixed feelings about them.

Vann Wilber, safety director for the Alliance of Automobile Manufacturers, says the ratings could mislead consumers because they require so many caveats. But manufacturers are working to improve their ratings, he said, and it "should be a good thing for consumers."