According to the New England Journal of Medicine, you are four times more likely to be involved in a car crash just by talking on a phone while driving.1 If something is four times more likely to get you fired or trash your car, better judgement should tell you to avoid it.
There is other news though: Virginia Tech Transportation Institute (VTTI) reports that talking on your phone is only 1.3 times more dangerous than driving alone.2
Wait: how can it be that the New England Journal of Medicine and the Virginia Tech Transportation Institute can differ by 300% on their findings? Something seems wrong with that kind of difference, right?
Take a second to run a Google search on dangers of talking on phone while driving, and what do you find? Injury centers at major universities, government safety councils, and law offices: all of them lamenting the fourfold risk increase associated with talking on a phone. No mentions of the Virginia Tech numbers though. Is it because the New England Journal of Medicine is a more reliable source with a more reliable study? Let’s find out.
The New England Journal of Medicine first published their oft-cited article on February 13th, 1997: twenty years ago at the time of my writing. The study was performed in Toronto, Canada, by consenting persons completing a questionnaire and the researchers subsequently examining their phone records. The 699 participating subjects had automobile collisions resulting in “substantial property damage but no personal damage.”
By contrast, the Virginia Tech Transportation Institute published its study in September of 2009. The study continuously monitored 203 commercial vehicle drivers. Monitoring was performed by five different cameras and a host of sensors. The study gathered ≈ 3 million miles of “continuously collected kinematic and video data, and represents the most comprehensive naturalistic CMV driving set in the world.”2 [emphasis added] (pg. xviii)
The VTTI’s study, commonly known as Olson, et al, is one of the most famous and most referenced distracted driving studies in existence. The famous statistic “texting and driving is 23x more dangerous than just driving” . . . that’s from this study. “Glancing at a phone for 4.6 seconds at 55 mph is the equivalent of traveling the length of a football field blind” . . . Same study again; in fact, 4.6 seconds was chosen because the study determined that the average driver spent 4.6 seconds out of a 6 second interval not looking forward when texting. In fact, the VTTI numbers are cited again and again for texting and driving, but the numbers for talking and driving emerge from a study that is 12 years older and much less reliable.
It would seem that if the odds of a crash were genuinely four times greater when a driver is talking on a phone, then number of traffic injuries and fatalities would have soared during the massive rise of cellular phone access for the majority of Americans. Instead, in the U.S., we went from 3.5 million cellular phone subscribers in 1989 to 300 million subscribers in 20103 (almost 10,000% increase), yet traffic fatalities per 100 million Vehicle Miles Traveled (VMT) dropped from 2 to 1 (a 50% decrease) during the same time period.4 Injury crashes also diminished from about 150 per 100 million VMT to very near 75 per 100 million VMT. 4 (figure 1 and figure 2 respectively). Simply put, phone conversations are not as dangerous as advertised in mainstream media.
Nor is this problem limited to talking on the phone while driving . . .
Voice-to-text technology was studied at Texas A&M University in 2013. The results showed driver response time and eyes-on-road time taking equally significant hits while typing a text or using the voice-to-text technologies. Predictably, this was announced and echoed by news outlets across the country: Huffington Post and Scientific American among them.
The Scientific American article began by describing the intuitive idea that it must be less distracting to use voice-to-text because it allows a driver to keep their eyes on the road. The author then mentions the Texas A&M study and says “The results surprised me—and troubled me. Turns out it makes absolutely no difference whether you text hands-free or by voice.”5 Really? No difference?
The article begins by describing a recent phone review in which the author examined a phone that voice activates without touching a button (“OK, Google”). He then fails to note that the phones used during the study were not equipped with this technology. He also fails to mention that many of the participants had never used voice-to-text prior to the study, and were given only a little time to familiarize themselves with it. Also, the LED that drivers “responded” to was completely missed just as often when people were driving without distraction at all as when they were texting manually, and less than that using Siri.6
More significant than these observations, however, are the findings of the VTTI study which stated that “ . . . talking or listening on a hands-free phone provided a significant protective effect (OR = 0.4). A similar significant protective effect was found for using a CB radio (OR = 0.6)” (p. 148). For those who are not statistics geeks, the “OR” is an “odds ratio”, and if it is less than one here, it means that it is safer to do the experimental task (talking on the phone hands free) than the control task (just driving). By the time of the Scientific American’s article publication, VTTI had already addressed the differences between findings by saying:
“These results show conclusively that a real key to significantly improving safety is keeping your eyes on the road. In contrast, “cognitively intense” tasks (e.g., emotional conversations, “books‐on‐tape”, etc.) can have a measurable effect in the laboratory, but the actual driving risks are much lower in comparison.” 7
Sometimes, the right answer really is the simple one.
Now, I don’t want to give everyone the idea that you can’t create distractions and danger by talking on the phone or using voice-to-text technology . . . you can become distracted by anything. I personally don’t like the idea of talking on a phone handset while driving, and I love hands free so much I installed after-market media/CD players in two vehicles just to have it. The point is that news outlets and government agencies should not have to “cherry pick” their numbers in attempts to sensationalize a point or “scare” people into certain behavior standards. If a behavior really is dangerous, shouldn’t it be enough to use the best and most accurate information, and to appeal to people honestly?
Then again: people don’t have a great reputation for responsible choices, no matter what facts are provided. Thanks for reading, and please be sure to post your comments below.
1. New England Journal of Medicine article:
2. VTTI Driver Distraction in Commercial Vehicle Study (Olson et al.)
3. Cell phone subscription source:
4. 2010 crash stats: for
5. Scientific American Article
David Pogue November 2013
6. Texas A&M Study Voice-to-Text:
7. VTTI reactionary article: