The summary headline for the July press release from the Insurance Institute for Highway Safety read: “Crash test results: Two midsize utility vehicles earn `best pick’ ratings; two 1999 models don’t improve compared with their earlier versions.”
The institute goes on to say that the Mercedes M-Class and Lexus RX300 “are the best midsize sport-utilities the Insurance Institute for Highway Safety has tested in 40-m.p.h. frontal offset impacts.” On the other hand, it rates the Mitsubishi Montero Sport “poor.”
This was good news for Mercedes-Benz and Lexus, but a bad day for Mitsubishi because the institute’s ratings get attention.
The Arlington, Va.-based institute, comprising more than 70 of the largest insurance companies in the U.S., has an agenda that includes crash testing and evaluating vehicles; studying highways and recommending ways to make them safer; and scrutinizing drivers with an eye toward improving awareness, skills and attitudes.
One of its primary interests, however, is to keep the cost of vehicle repairs down and to avoid paying unnecessary claims, said Julie Rochman, institute vice president of communications.
“Absolutely. We exist to lower damages from crashes,” Rochman said. “It’s a win-win situation. Both the insurance companies and vehicle owners benefit.”
The institute grew out of the move toward a more scientific approach to safety spawned in the mid-1960s. Before this, the burden for reducing crashes and their consequences had been placed on motorists. It was felt if they improved their driving, there would be fewer accidents.
The scientific approach, including well-planned and executed laboratory tests plus the examination of the aftermath of real-life crashes, showed a need for safer vehicles and roadways. It suggested that vehicles with improved handling, crash-avoidance features and greater protection for passengers would add to the overall safety record.
To that end, the institute completed and opened a Vehicle Research Center in Ruckersville, Va., (near Charlottesville) in 1992. There it performs tests on vehicles the institute purchases from dealers.
“We try to test vehicles in groups or classes (like mid-size sport-utilities),” said Rochman. “We pick from a range of manufacturers and attempt to include best-sellers.”
Rochman said that because the institute’s budget is fixed, the types of tests it does vary, depending on current issues, such as air bags. For instance, if the institute is working with luxury vehicles, which they purchase from dealers, they will have less to spend on tests.
Rochman said the manufacturers are there when their vehicles are tested. The institute’s staff discusses the results with them afterward. The institute makes its results public but will add that manufacturers are addressing whatever problems the tests reveal. For example, when a door of a 1997 Mercedes-Benz E-Class opened late in the crash test, the manufacturer worked to repair it.
Rochman said typically a couple a months elapse between a test and the posting of the results. “All of our finalized research–i.e., we’ve gone over crash test data, figured out what it means for that vehicle and relative to other vehicles in the same class–is available to the public via our Web site as soon as we release the information. We update the Web site as we send things out to the media for public release.”
The institute says its 25-member staff is the only independent crash-testing operation in the U.S.
Others in research and safety agree that the institute is careful, accurate and not at all bashful.
“The institute has been very vocal on safety issues,” said David Cole, head of the office for the Study of Automotive Transportation at the University of Michigan in Ann Arbor. “They have communicated freely and effectively.”
Cole said the organization works hard to maintain objectivity, but it has focused on the auto industry, he added, not on the insurance industry.
Clarence Ditlow heads the Center for Auto Safety in Washington, D.C. Obviously, he said, the institute looks after the interests of its constituents in the insurance industry. But he applauded the institute’s offset crash tests and 5-m.p.h. bumper tests.
“We don’t quarrel about their practices, but we wish they had a broader social viewpoint,” Ditlow said. For example, he said, the institute doesn’t address fuel-economy issues and other environmental concerns.
Ditlow indicated he doesn’t care for the institute’s ratings approach to vehicle safety.
Dr. Lawrence Schneider, senior research scientist at the University of Michigan, also questioned the use of crash-test research.
“They look at things anecdotally,” Schneider said. The insurance institute will do a test, show the results and say, for example, that it resulted in a “bad” injury, when in other situations one might walk away from this crash.
“It’s not always the complete story,” said Schneider. He and other scientists are concerned about the completeness of crash tests. If test dummies were used, which models were they? How were results measured? Research, he said, might be better served if it were long-term in its nature and outlook. Going for short-term results–much of the institute’s research is conducted over six months or less– doesn’t tell the whole story.
“I’d love to see them do more long-term research, especially in biomechanical issues,” he said, adding that the institute may be a little guilty of churning out safety sound bites instead of sticking to long-term, more serious research.”
In addition to its crash testing, the institute is looking at ways to make urban arterial roads safer. It is concerned about air bags, headrests, elderly and teenage drivers, licensing practices and theft issues.
But its crash testing and vehicle ratings are what have drawn the greatest public attention.
The good thing about these crash test issues is they are based on science, said Tim Hurd, a spokesman for the National Highway Traffic Safety Administration. The institute’s tests can be duplicated. They also have an objective record in the claims filed by policyholders. Results can, however, be interpreted subjectively.
NHTSA and the insurance institute are Washington-area neighbors. Hurd said institute staff members frequently participate in the public comment period required for NHTSA’s proposed standards.
Automakers have to meet those standards once they are in effect. And they cannot ignore institute test results, even if they do not result in standards.
“(The insurance institute) provides information we have to address,” said Bill Eagleson, manager of North American regulation in the Automotive Safety Office at Ford Motor Co. “Our customers are paying attention.”
But Eagleson said consumers need to ask themselves questions about test results. Is what was being tested on a vehicle important to the buyer? How do vehicles in the same category compare with one another? It’s more than a matter of a vehicle having a rating of a certain number of stars.
The Ford spokesman said there are issues of laboratory tests versus real-world situations. If dummies are used, what is their level of biofidelity? There is greater precision when tests are repeated. How is data being interpreted?
Rochman said a test is not repeated unless the automaker in question can show there was a flaw in the test process. Such a retesting took place when a acceleration slated to be 26 m.p.h. was actually 40 m.p.h.
She said the automakers have run the same tests on their vehicles and know what the results should be before the institute does them. “We run our findings past them to see if there is any anomaly. If so, and they can prove that our findings are inconsistent with other existing knowledge–that our test was somehow flawed–then we will retest. We don’t release our findings until we are certain of our findings.”
She said the tests are based on European standards subjected to repeatability criteria.
“Ratings can have a place if they are properly characterized as to their meanings,” Eagleson said. “These test results are a relative measure of performance within a limited context.”
