Just after midnight on August 29, 2001, a white Pontiac missed a turn and plunged 400 feet from New York's Palisades Parkway into a swamp below. One of the four young men in the car died instantly. The other three would end up waiting six hours for help to arrive, despite having reached a 911 operator on a cellular phone.
Automatic crash notification (ACN) systems such as General Motor's OnStar might have brought help sooner in this case. An automatic call for help with location information would have gone out as soon as the crash was detected.
ACNs, which consist of a global positioning unit and a cell phone, do help to report and locate isolated crashes. But how great a benefit would we reap if all cars had ACNs? After all, it's rare for any crash site to go unnoticed for hours.
Fatality reduction would be small
Researchers at the Maine Medical Center and the Harvard Injury Control Research Center developed a mathematical model linking mortality risk to crash notification and response times. Based on data from more than 30,000 fatal crashes, the model predicts that shortening the notification times to a minute or less would reduce crash deaths by about 1 to 6 percent. A total of 400 to 1,700 lives would be saved each year, assuming perfect functioning of the ACN systems.
"It's a tangible and important benefit. At the same time, these results show the limitations of ACN," says Adrian Lund, the Institute's chief operating officer. "Emergency workers already find out about most crashes within a few minutes, and some of the crashes are so severe that the people are killed instantly. Besides, the predicted benefits depend on perfect functioning of the ACNs, which isn't likely."
Current ACN systems detect serious crashes by using information about frontal airbag deployment. Soon sensors for rear and side impacts will be added. When a crash is detected, a call is automatically placed to a service center, which then tries to contact the driver of the crash-involved vehicle. If there's no response or the driver verifies the crash, the call center notifies emergency personnel, providing the details and location of the crash.
A national ACN system is being advocated by the ComCARE Alliance, a nonprofit organization supported by medical nonprofits and the automotive and telecommunications industries. These and other advocates view ACN as a way to get patients to hospitals sooner. But in fatal crashes for which notification times are known, 75 percent of the urban events were reported within 4 minutes. Rural crashes were reported within 9 minutes. And even when ACN does reduce the time it takes for rescuers to be notified of crashes, the response times won't be any shorter. In remote areas, response times can be lengthy because rescuers can be stationed long distances from crash sites.
"The modesty of the mortality reductions predicted by the researchers' model can be attributed to the fact that, even without ACN, most notification times are short, especially relative to response times," Lund points out. "In the case of the Pontiac that crashed on the Palisades Parkway, ACN might have helped bring rescuers sooner, but even in this case no lives would have been saved."
Help also might have arrived sooner if another system — so-called enhanced 911 for cell phones, or E911—had been in operation. Under this system, the location of a cell phone is displayed to 911 operators during an emergency call. Such technology is scheduled for phase-in starting in 2003.
'Golden hour' might not be relevant
ComCARE Alliance, which supports both E911 and ACN, bases its recommendation of speedier notification times on a medical concept known as the "golden hour," according to which many trauma deaths can be prevented if appropriate medical treatment is applied within an hour of the injury. Based in part on common sense and in part on the experiences of trauma surgeons, this concept justifies much of the way injury is treated in this country. However, the "golden hour" might be less valid for car crash injuries than for other trauma cases.
Much of the basis of the concept comes from experience with the penetrating injuries sustained in wars. Quick medical attention to such injuries is important. In contrast, many crash injuries result from blunt trauma, for which quick medical treatment might not make much difference in the outcome.
Flawless performance isn't likely
The fatality reductions predicted for ACNs depend on the systems functioning perfectly. But when the National Highway Traffic Safety Administration tested a system over a five-year period in New York, a main finding was failure to notify emergency personnel of crashes in 5 out of 21 cases. There also were 31 false alarms.
"This means there would be 22,000 false alarms per year in a city with a million ACNequipped vehicles," Lund points out. OnStar's reliability should be better, if only because the service center should eliminate many of the false alarms. All of the successfully completed ACN calls in the test resulted in notification times of 2 minutes or less, but most of the times for the control group also were short. Eighty percent were less than 5 minutes, but 2 of the 25 notification times were longer than 30 minutes.
Another obstacle could be the cost
Automakers are installing ACNs as part of several telematics packages on high-end cars and SUVs. Besides some General Motors models, OnStar with its ACN component is in Acura, Audi, Lexus, and Subaru models. Some packages also include navigation systems, hands-free phones, and concierge services, each of which requires a monthly fee.
Ford recently announced that Sprint PCS VoiceCommand will provide similar services in Lincolns.
OnStar reports that 56 percent of its customers renew their subscriptions after the free trial period. The majority opt for the basic package, which includes ACN. If half of the owners of cars with systems in place aren't willing to pay the monthly premiums, the benefits will be lessened.