The research, reported in today's issue of The New England Journal of Medicine, examined crash reports and telephone billing records of several hundred Canadian drivers with cell phones who had collisions during a 14-month period. The authors calculated that within a few minutes after beginning a call in their cars, drivers were 4.3 times more likely to have an accident than they were when their phones were not in use.
That increased risk, they conclude, is "similar to the hazard associated with driving with a blood alcohol level at the legal limit" and drivers should consider additional "road safety precautions" while talking on the devices.
In a surprising coincidental finding, University of Toronto physicians Donald Redelmeier and Robert Tibshirani also determined that cell-phone "units that allowed the hands to be free offered no safety advantage over hand-held units."
The authors emphasized that their study does not prove that cell phones cause accidents; it only indicates that use is associated with increased risk. There are numerous possible contributory factors, Redelmeier said yesterday, and "we don't know anything about what (the drivers) were doing with their radios or coffee cups" at the time of their collisions.
"It's very difficult with this kind of study to establish causality," noted Michael Goodman of the National Highway Traffic Safety Administration, which is planning to issue a two-year comprehensive review of international research on the subject in late March. "It could be," he said, "that the very circumstance that gets the person on the phone - being late for a meeting, involved in an argument, being distraught - itself could precipitate the crash."
Goodman, the technical manager for NHTSA's forthcoming study, said there is no definitive measure of "the magnitude of the problem, if there is a problem."
Nor, he said, is it clear whether any risks posed by car-phone use outweigh the incontestable benefits they provide in notifying police, fire and ambulance services of highway emergencies. Nonetheless, he said, "the process of alerting the public to some of these concerns is very important."
In a statement in response to the Toronto study, the Cellular Telecommunications Industry Association (CTIA) concurred with the "conclusion that drivers should be vigilant." But the group noted that while the cell-phone market has been expanding drastically (in 1995, the Toronto researchers write, "the number of new subscribers in the United States exceeded the birth rate") there has been no apparent corresponding rise in accidents.
"From 1988 to 1995," CTIA observed, "wireless phone users grew 1,685 percent to nearly 34 million subscribers. During the same period, injuries resulting from auto accidents decreased by 16.6 percent and fatalities fell by 26 percent."
The Toronto study is only the second prominent paper to have been published on the subject in North America.
ast year, Rochester (N.Y.) Institute of Technology researcher John M. Violanti and colleagues reported a smaller-scale investigation that found a sample of persons with cell phones in their cars had an approximately 34 percent greater chance of having an accident than did a similarly sized sample of drivers without phones. Other studies evaluating driver performance while using cell phones have shown the devices to be somewhat distracting, but not more so than a spirited conversation or tuning the radio.
At present, NHTSA's Goodman said, issues such as car-phone operation are typically addressed at the state level. Although no state has laws on the book regulating their use, some have legislation pending. Several foreign countries - including Israel, Brazil, Switzerland and parts of Australia - prohibit cell-phone use while driving.
In an accompanying editorial in the journal, Malcolm Maclure of the Harvard School of Public Health and Murray A. Mittleman of Beth Israel Deaconess Medical Center in Boston, caution that "laws banning all telephone use while driving would be premature without better data and a thorough assessment of the contribution of car telephones to roadway safety."
However, they write, the cell phone industry "now has an ethical obligation to include warnings and advice with their products and mailed bills; to support and assist in further engineering, ergonomic and epidemiologic research; and to provide easy-to-dial toll-free numbers for reporting road hazards and unsafe driving."
The Toronto study derived its sample from drivers involved in collisions who came to an accident reporting center, as required by law, between July 1994 and August 1995. Each had been involved in a crash with "substantial property damage" averaging around $1,000 Canadian, but no injury. The researchers identified 699 drivers with cell phones who were willing to have their telephone records released for use in the project, and then compared the time of their accidents (as determined by police reports and calls to emergency services) with cell-phone use as shown in the phone logs. The subjects, who averaged four calls a day of 2.5 minutes mean duration, were much more likely to have had their accidents within a few minutes of initiating a call.