The non-profit group Informed for Life (www.informedforlife.org) has just released its combined RISK scores for many 2006 model-year vehicles. The group's RISK scores use a calculated composite of NHTSA and IIHS crash ratings, weighted according to actual fatality risk in accidents, along with known factors that raise the fatality risk in vehicles, like weight, the presence of stability control, and the presence of a side-curtain airbag.

Although the group isn't yet well known to consumers or industry insiders, it's filling a gap in the availability of straightforward crash ratings. "So far, it's been a purely personal effort on my part," said the group's president, Michael Dulburger, a Conn. mechanical engineer who started the project about five years ago when he was shopping for a new car and found that none of the established crash-ratings systems corresponded to actual injury or fatality risk. When considering the purchase of a Ford F-150, he found that NHTSA gave the truck a five-star crash rating; he was later amazed to find out that the IIHS had ranked it 'poor.' Without any professional ties to the auto industry, Dulburger set out to devise his own system; since then, he's been funding the project himself - and devoting much of his spare time to entering data and honing a system that would better correlate with real-world statistics.

After poring over actual crash data and what existing ratings meant, he saw some glaring deficiencies that rendered some aspects of either IIHS or NHTSA ratings meaningless to a typical consumer. "Even within each source they don't really tell you how to apportion your concern," said Dulberger, who pointed out that rear impacts account for only three percent of all fatalities, while they're given what appears to be equal footing in their ratings. On the other hand, frontal impacts account for 43 percent of fatalities, rollovers 28 percent, and side impacts 26 percent, according to NHTSA. "The IIHS still doesn't even look at rollover. And when you look at the NHTSA side-impact ratings, they mention that they don't consider head injuries. What good is that?"

"The NHTSA also tells you that you shouldn't compare vehicle ratings in different weight classes. But that's not the way that people shop for vehicles… And the full ratings range isn't even used - for instance with NHTSA almost every car is four or five stars," commented Dulburger, who referred to a 30:1 fatality spread from the model with the highest fatality rate to that with the lowest, according to NHTSA.

"There are 117 traffic fatalities a day in the U.S. What you're in makes a huge difference. If we could stay away from the models with the worst fatality rates, we'd greatly affect that number."

"Automakers have figured out that safety is a selling point," said Dulburger, who referred to the frequent mention of crash ratings in ads. "Five stars is such a small part of the overall picture. You need to be looking at so much more."

Dulburger admits that the group's system doesn't take into account other factors like the varying demographics and driving attitudes of buyers for each model, but he said that it's the closest match to how the cars protect in real-world crashes.

The combined RISK scores are now indexed for 2006 so that 100 represents the fatality-weighted average - the lower the score, the better. See the Web site for complete details of how the scores are calculated. Below are the best and worst of '06, according to the system. -Bengt Halvorson

Lowest RISK scores:

Acura RL - 56
Lincoln Town Car - 58
Volvo S80 - 61
Acura TL - 62
Acura MDX - 66
Ford Crown Victoria - 66
Mercury Grand Marquis - 66
Mitsubishi Galant - 66
Lincoln LS - 68
Acura TSX - 68


Highest RISK scores:

Honda Insight - 167
Scion xA - 158
Mazda B-Series - 152
Ford Ranger - 152
Toyota Tundra - 150
Toyota Tacoma - 149
Ford Focus - 149
Nissan Sentra - 148
Nissan Frontier - 148
Kia Spectra - 146

http://www.informedforlife.org/