Researchers are coming to some troubling conclusions about air bags: The technology that saves lives also causes injuries and, in some cases, fatalities.
The problem is the speed with which air bags must inflate. The misconception is that air bags are like soft pillows. In fact, they're more like rocket-propelled balloons that shoot out at speeds of up to 200 miles an hour. The people most at risk: children, the elderly, and short drivers.
The Big Three automakers last week proposed a change in government requirements that would allow them to install air bags that use less force. While manufacturers are working to reduce the risk of injury from air bags, these improvements are probably years away from implementation.
Because some 50 million vehicles are already equipped with the devices, a coalition of air-bag suppliers, insurance companies, the US government, and safety groups has launched a $10 million campaign to educate drivers about the hazards. By using better safety habits, the coalition argues, drivers can minimize air-bag dangers.
"What we're struggling with is the education problem," says Michael Jablonski of Cleveland-based TRW Inc., one of the world's largest manufacturers of air-bag systems.
The force of air-bag deployment causes injuries in more than 4 out of 10 cases, reports the Insurance Institute for Highway Safety, a research group in Arlington, Va., funded by auto insurers. In 96 percent of the cases, these injuries are minor. The bags pose the biggest threat to people who sit close to where the bag deploys. This includes short drivers who sit up close to the steering wheel and children who perch on the front passenger seat.
According to the National Highway Traffic Safety Administration (NHTSA), air bags have killed 42 people since 1990 - 19 adults and 23 children. Especially disturbing is that several of these incidents occurred while vehicles were moving at low speeds.
"There is evidence that bags are saving lives," says Susan Ferguson, vice president of research for the Insurance Institute. "But they also have the propensity to injure."
On balance, they do far more good. The NHTSA credits them with saving at least 1,500 lives since auto manufacturers began installing them in the mid-'80s. But air bags are secondary restraining devices that complement rather than replace seat belts, safety advocates say. Between 1982 and 1994, seat belts saved an estimated 65,000 lives.
Thus, the most important driving tip is to wear a seat belt, safety advocates say. Most air-bag fatalities have occurred to drivers not wearing seat belts. In the case of children, they should always ride in the back seat. If they must ride in front - in a pickup truck, for example - they should ride in front-facing child seats.
At least six air bag-related fatalities involved children riding up front but in rear-facing child seats. According to the NHTSA, the other children killed were out of position at the time of the crashes because they were not wearing seat belts, were perching near the dashboard and the air bag, or were wearing only the lap portion of the seat belt.
Although consumer and safety groups agree what steps the public should take, they don't agree on what the government and automakers should do to reduce air-bag injuries.
Under current federal regulations, air bags are designed to save the life of an average-size man not wearing his seat belt who is traveling at 30 miles per hour. That requires the aggressive propulsion of today's air bags. But a smaller person, traveling at a slower speed, may be at risk of air-bag injury during an accident, concludes a study for Transport Canada.
The auto manufacturers would continue to conduct crash tests at 30 m.p.h., but they would modify test conditions to adjust for variables such as force of impact. They argue that children will be less likely to get hurt if air bags don't inflate so fast. But the NHTSA and the consumer-funded Center for Auto Safety aren't so sure.
Air-bag manufacturers are working on next-generation bags to solve some of these problems. "Smart" air bags will be able to sense a person's weight, whether he or she is belted or unbelted, the position of the seat, and the severity of the accident. Then they will react accordingly, sometimes deploying at a slower rate than today's go-or-no-go devices.
"Everybody is working very seriously on that," says Mr. Jablonski. But because of the lead times auto manufacturers require, consumers won't see these sophisticated devices until after the turn of the century, he adds.