These days, most vehicles are designed so that, if you get into a crash at about 35 mph -- even if you run straight into a solid, immovable wall -- you're likely to walk away. But recent crash tests and an analysis by the Insurance Institute for Highway Safety concluded that if you crash into the "underride guards" on the backs of tractor-trailers at that speed, the consequences can be deadly.

These rear guards are the primary countermeasure for reducing deaths and injuries when a passenger vehicle crashes into the back of a tractor-trailer, says Adrian Lund, IIHS president.

The Institute is petitioning the federal government to require stronger underride guards that will remain in place during a crash -- and to mandate guards for those large trucks and trailers that are not currently required to have them.

In 2009, 70 percent of the 3,163 people who died in all large truck crashes were occupants of cars or other passenger vehicles, Lund said. In many of these crashes, the upper part of a passenger vehicle's cabin was crushed when the body of the truck or trailer smashed through the vehicle's safety cage.

"The way passenger cars are designed, if you crash into another passenger car, the front-end structures can withstand and distribute a tremendous amount of crash energy, in a way that minimizes injuries for vehicle occupants," says Lund. "But hitting the back of a large truck is a totally different situation. Your vehicle could be one that earns top marks in frontal crash tests, but if that truck's underride guard fails -- or if the truck doesn't have one at all -- your chances of walking away from even a relatively low-speed crash are not good at all."

In the Institute's recent study, it took case files from the Large Truck Crash Causation Study -- a federal database of roughly 1,000 real-world crashes in 2001-03 -- and analyzed them to identify patterns of crashes involving heavy trucks and semi-trailers with and without underride guards. Only 22 percent of those crashes either didn't involve underride, or had only negligible effect. In 23 of the 28 cases in which someone in the passenger vehicle died, there was severe or catastrophic underride damage -- meaning the entire front end (or more) of the vehicle slid beneath the truck.

The National Highway Traffic Safety Administration (NHTSA) estimates that about 423 people in passenger vehicles die every year when their vehicles strike the backs of large trucks, and that more than 5,000 passenger vehicle occupants are injured in such crashes.

Based on that data, the Institute conducted crash tests at various speeds to gauge how serious the injuries to passengers could be at those speeds. The Institute's crash tests evaluated three semi-trailer rear guards that comply with current U.S. government regulations. Two of the trailers were also certified to meet Canadian requirements, which are more stringent than U.S. regulations when it comes to strength and energy absorption. In the tests, a 2010 Chevrolet Malibu was crashed into the rear of the parked trailers.

Lund stresses that the purpose was not to assess the Malibu's crash-worthiness. The Malibu has already earned a Top Safety Pick designation from the Institute and got a 5-star safety rating in the National Highway Traffic Safety Administration's New Car Assessment Program.

"The purpose was to ascertain if some underride guards perform better than others, and to find out what crash speeds and underride designs produced different types of failure," says Lund. "The damage to the cars in some of these tests was so devastating that it's hard to watch the footage of the tests without wincing. If these had been real-world crashes, there would be no survivors."

Decapitation is a very real possibility in underride crashes. In three of the crash tests the heads of the dummies in the car either smashed into the trailer or into the car's hood after the hood was ripped off and was pushed into the occupant compartment. One such test involved a Hyundai Translead trailer whose underride guard bent forward, sheared its attachment bolts, and broke after the Malibu hit it in the center rear at 35 mph. This was the weakest of the guards that were tested.

Meanwhile, a trailer made by Wabash National Corp., which was equipped with a guard that met Canadian requirements, successfully prevented underride of the Malibu's passenger compartment in a center-rear test at 35 mph. The Wabash's guard was the strongest of the three that were tested. A trailer made by Vanguard National Trailer Corp. ranked in the middle.

"The Wabash trailer had strong attachments, which is what kept the guard in place," says Lund. "As a result, the guard engaged the Malibu's front end, which allowed the car's structure to absorb the crash and distribute the energy of the crash. If that was a real-world incident, that would have been a survivable crash."

The Institute also conducted tests with overlaps of 50 percent and 30 percent to discern what would happen if a car crashed into the trailer with only part of its front end instead of a full head-on crash.

In a 35-mph test with a 50 percent overlap, the guard on the Vanguard trailer allowed severe underride. Its guard meets both U.S. and Canadian standards. In contrast, the Wabash trailer's guard successfully prevented underride in the same test. The outcome for the Wabash was different when the overlap was reduced to 30 percent. The end of the guard that was struck in that 30 percent test bent forward, allowing for significant underride.

This test showed that even the strongest guard leaves as much as half of the rear of the trailer vulnerable to severe underride, says Lund, adding that the guard only worked as intended when the striking car engaged the center.

"Under current certification standards, the trailer, underride guard, bolts, and welding don't have to be tested as a whole system," Lund says. "That's a big part of the problem. Some manufacturers do test the guards on their trailers. We think all guards should be evaluated this way. At the very least, all rear guards should be as strong as the best one we tested.

"Absent regulation, there's little incentive for manufacturers to improve underride countermeasures," concludes Lund. "So we hope NHTSA will move quickly on our petition."