Hang up and drive

July 12, 2005 9:56 AM

The use of a mobile phone up to 10 minutes before a crash is associated with a fourfold increased likelihood of crashing according to a forthcoming article in the British Medical Journal. There is no safety advantage in 'hands-free' devices. The authors interviewed 456 drivers in hospital ERs in Perth, and then accessed, with consent, their phone records. The authors had wanted to carry out the study in the United States, but the phone companies would not release customer billing records, even with a customer's consent. [Article pdf free for the next few days, but BMJ will charge for it after next week I think.]
posted by carter (47 comments total)

[I know this has been discussed before, but this seemed a particularly good source]
posted by carter at 9:58 AM on July 12, 2005

This study, and especially the reiteration of the lack of advantage to hands-free devices, makes me wonder about all of the other things that we do in cars that might similarly contribute to crashes but which are much more difficult to track.
posted by OmieWise at 10:03 AM on July 12, 2005

OmieWise: Like eating seven layer burritos, attempting to dicipline children in the back seat, and receiving oral sex (not even all at once)!
posted by aubilenon at 10:12 AM on July 12, 2005

oh, cool...now how do i get a text message alert eleven minutes before i'm going to have an accident so i can get off the phone? minority report?
posted by spicynuts at 10:15 AM on July 12, 2005

Let the lawsuits begin!
posted by Outlawyr at 10:21 AM on July 12, 2005

This bit boggles me:

The CTIA contends that following these guidelines can minimize risk:

• Assess whether a given call can wait.
• Do not take notes while driving.
• Do not talk while in heavy traffic.
• If possible, pull off the road and park in a safe location to use a mobile phone.

The Insurance Institute for Highway Safety disagrees, maintaining that its study shows that the simple act of talking on a cell phone while at the wheel significantly raises the risk of a serious crash, and following those guidelines will not lower that risk.

It seems to me that the first tip is to consider not making/taking so many calls and the fourth is to not drive while calling. Surely those, at least, would lower risk.

The other bit that boggles me, of course, is that people would try to take notes while driving often enough that it's on a list of 'don't's'.
posted by Karmakaze at 10:22 AM on July 12, 2005

ah, so it's not another post related to THAT BMJ?

Then I'm disappointed we are not going to discuss the virtues of phone temple implants.
posted by Laotic at 10:24 AM on July 12, 2005

In realted news, walking down the street gabbing into your cellphone you're about fifteen times as likely to walk straight into a parking meter as someone who's paying attention to where they're going.

Combine cell phone usage with being a teenager and I'll bet the odds of having an accident approach even money.
posted by fenriq at 10:34 AM on July 12, 2005

The authors had wanted to carry out the study in the United States, but the phone companies would not release customer billing records, even with a customer's consent.

Stop protecting my privacy!
posted by thedevildancedlightly at 10:36 AM on July 12, 2005

They include people who are off the phone up to 10 minutes before the crash? So I wonder if the risk is the same if you make a phone call just before you get in the car.
posted by sfenders at 10:46 AM on July 12, 2005

The authors had wanted to carry out the study in the United States, but the phone companies would not release customer billing records, even with a customer's consent.

Stop protecting my privacy!

Are you being deliberately obtuse? If someone doing a study asks you if you'd be willing to release your phone records to them for the study, you give them your consent, and your phone company still refuses, how is your privacy being protected?
posted by odinsdream at 10:50 AM on July 12, 2005

From the BMJ article: Using a hands-free phone is not any safer.

Not unless 3.8 is a smaller number than 4.9. Sure, even using a hands-free phone, your risk is almost 4 times greater than if not using a phone at all, but it IS safer than a hands-required phone.
posted by neuron at 10:55 AM on July 12, 2005

I suspect also that the phone industry are not all that interested in assisting research that might start cutting into a segment of their market.
posted by KirkJobSluder at 11:04 AM on July 12, 2005

odinsdream writes "Are you being deliberately obtuse? If someone doing a study asks you if you'd be willing to release your phone records to them for the study, you give them your consent, and your phone company still refuses, how is your privacy being protected?"

I think your irony detector is broken - please contact the nearest irony maintenance representative.
posted by nkyad at 11:05 AM on July 12, 2005

Good catch neuron. The statement that hand-free device was not a safety advantage as compared to not using one shocked me. Surely they are safer, I would guess significantly so. But, these guys are the one who did the study and have the stats---I am just surprised by them.

I always assumed the problem with cellphones were that one hand might not be on the wheel and the head and eyes were not in the position to see properly through all the windows and mirrors. I never thought that attention is a problem; talking on the phone doesn't distract me from being aware of traffic, at least. The hands free devices free both hands to be on the wheel and keeps people from doing that bended neck thing that gets the head out of position. So I would assume they would be much safer. I guess a lot of people have problem with attentiveness more than optimal driving positioning.
posted by dios at 11:10 AM on July 12, 2005

dios: I never thought that attention is a problem; talking on the phone doesn't distract me from being aware of traffic, at least.

A big problem with perception, as noted by both Buddhist monks and cognitive psychologists, is that most people are never really aware on how much attention or time is being spent on a particular task. Our subjective experience of attention is extremely flawed. You only really get a good idea about how it works from experimental studies.

I suspect this is true of a lot of tasks in HCI as well. I'm fond of noting that we've replaced the 3 Martini lunch with Microsoft Office.
posted by KirkJobSluder at 11:16 AM on July 12, 2005

Someone should do a study to see how much the rate of traffic accidents increases when they play Autobahn on the radio.
posted by sfenders at 11:18 AM on July 12, 2005

KirkJobSluder, the difference being people can understand me now when I swear at MS Office as opposed to the post 3 martini lunch when I'd be just about incoherent.

dios, you assume a lot.
posted by fenriq at 11:20 AM on July 12, 2005

Dios, speaking for myself, when I talk on the phone while driving and an unexpected traffic situation occurs, I always lose the thread of my conversation. It might work the other way round, but perhaps my survival instinct overrides.

I can imagine there are people who become so involved with their call that they pay less attention to the road, and hence the risk.

By the way, with phone on hands-free, I crane forward towards the central microphone, because the callers complain about the sound quality. I bet that that is more dangerous than just holding the phone to my year. With my right hand.
(supposing the car is an automatic.)
posted by Laotic at 11:25 AM on July 12, 2005

dios writes "I always assumed the problem with cellphones were that one hand might not be on the wheel and the head and eyes were not in the position to see properly through all the windows and mirrors. I never thought that attention is a problem; talking on the phone doesn't distract me from being aware of traffic, at least."

The results that hands-free is still pretty dangerous are not new, I know they've been documented before. Which does indicate that attention is compromised somewhere along the line. It does seem strange to me, as well, though.
posted by OmieWise at 11:27 AM on July 12, 2005

Is this paper peer reviewed?

Participants 456 drivers aged =17 years who owned or used mobile phones and had been involved in road crashes necessitating hospital attendance between April 2002 and July 2004.
there is no control here

How do they account for people who spoke on their cellphones and didn't crash their cars? Are they literaly saying that 80% of the people who got into car accidents had been speaking on their cellphones when they had an accident, or what?

It isn't resonable to extrapolate the results of this study to people who don't crash their car into things.

The only way to do this test is to count the total number of people who drive at a particular time and didn't crash and find the ratio of crash/no crach for cellphone users, and crash/nocrash for not cellphone users.

The statistics here just don't add up to me.
posted by delmoi at 11:35 AM on July 12, 2005

A lot of what we think about traffic issues is also really counter-intuitive. People think that bike lanes improve safety, when the vast majority of bike accidents involve confusion about right of way at intersections. People think that vehicle size increases safety, in spite of the fact that vehicle structure is more important. (I'd rather drive a small vehicle with crush zones than a over-styled pickup.)

demoi: How do they account for people who spoke on their cellphones and didn't crash their cars? Are they literaly saying that 80% of the people who got into car accidents had been speaking on their cellphones when they had an accident, or what?

By including participants who were not using the phone prior to the crash. And certainly, as a pseudo-experiment there are some issues on this. However, the statistics and methods are sound.

The only way to do this test is to count the total number of people who drive at a particular time and didn't crash and find the ratio of crash/no crach for cellphone users, and crash/nocrash for not cellphone users.

For a variety of reasons. This is a bit difficult to obtain. On the other hand, this does support experimental studies involving driving simulations.
posted by KirkJobSluder at 11:45 AM on July 12, 2005

Yeah, I'm sending you that information now on my Blackberry and I should arrive in five or six ...screeeeechk....

.
posted by buzzman at 11:48 AM on July 12, 2005

Reminds me of Episode 33 of MythBusters. They did better driving drunk than they did talking on a phone.
posted by pzarquon at 11:50 AM on July 12, 2005

You're not driving a phone booth.

You're not driving a restaurant.

You're not driving a beauty parlor.

You're not driving a day care center.

You're driving a car.

Pay attention and DRIVE for chrissake.
posted by Relay at 11:51 AM on July 12, 2005

Well duh, I just have to end my call at least 10 minutes before I have an accident. Next?
posted by jalexei at 11:58 AM on July 12, 2005

I'm wondering whether talking on the phone while driving is a sign of being a bad driver in general as much as it is a specific cause of bad driving. The sort of person who chats on the phone to his auntie about bunions while racing a dangerous machine at 100 kph between carloads of children may just be the sort of person who does other stupid things while racing a dangerous machine at 100 kph between carloads of children. Maybe this helps to explain why drivers are still more likely to crash 10 minutes after using the phone -- they aren't still addled because of the call, their minds are always somewhere else.

So would it be fair for insurance companies (or police) to check your phone records after an accident, see whether you were on the phone at or just before the time of the accident, and raise your rates (or file charges) accordingly?
posted by pracowity at 12:08 PM on July 12, 2005

Relay wins thread.
posted by wakko at 12:08 PM on July 12, 2005

Cognitive Daily Article: Why are cell phones so dangerous for driving?


Strayer and Johnson found no difference between people who used a handheld or hands-free cell phone, and no difference between radio/audiobook listeners and the driving-only condition. However, the cell-phone talkers missed more than twice as many red lights as the other participants:


And this is particularly interesting:

But does any conversation lead to driving errors? In a second experiment, Strayer and Johnson used a similar apparatus, but instead of using red and green lights, they had participants drive over “easy” and “difficult” courses. The volunteers were first asked to simply repeat words to the experimenter over the telephone. Next they were asked to generate a new word starting with the last letter of the word the experimenter gave them (for example if the experimenter said “salmon,” the volunteer could respond “nicotine”). The results were as follows:

[and a chart follows which shows that creating words is worse than repeating words which is worse than doing neither.]
posted by callmejay at 12:32 PM on July 12, 2005

Don't drive exhausted either - it's as bad as driving drunk.
posted by agregoli at 12:39 PM on July 12, 2005

By including participants who were not using the phone prior to the crash. And certainly, as a pseudo-experiment there are some issues on this. However, the statistics and methods are sound.

What makes them so "sound"? Are they saying that the vast majority of people involved in car accidents were talking on their cell phone when it happened? They don't say that, in fact (from what I can tell) they don't say anything about their method, and how they did their calculations at all. Only the end result.

But all we saw was the abstract, which contains a simple assertion. I'd rather see more about how they did their calculations. Papers with bad math get published all the time, sadly.
posted by delmoi at 12:58 PM on July 12, 2005

delmoi, you can get the full paper by clicking on "Online First PDF" in one of the right hand columns, under "This Article."
posted by carter at 1:11 PM on July 12, 2005

Whether or not this particular study is scientifically above reproach, the phenomenon responsible is well-documented. It's called inattention blindness, and we chatted about it a while ago here.
posted by soyjoy at 1:21 PM on July 12, 2005

delmoi, it's not the likelihood of getting in a crash relative to if you're on cell phones, but the likelihood of having been on cell phones relative to if you're in a crash. Those are different sets of numbers.
posted by mystyk at 1:25 PM on July 12, 2005

Hrmm... Again, talking only to people who are in crashes seems to me as if it would skew the results some. Basically, what they're saying is there's a correlation between being a bad driver (getting in accidents) and talking on a cell phone. I would say it's more distracting to me having a conversation with 2 other people in the car, than it is having a phone conversation with one person. But that's just me. So no talking while you're in my car, dammit!
posted by antifuse at 3:10 PM on July 12, 2005

delmoi: Are they saying that the vast majority of people involved in car accidents were talking on their cell phone when it happened?

No RTFA.

They don't say that, in fact (from what I can tell) they don't say anything about their method, and how they did their calculations at all. Only the end result.

The paper deals sufficiently with their recruitment methods and their means of making comparisons. RTFA.
posted by KirkJobSluder at 3:35 PM on July 12, 2005

I'm in 100% agreement with Relay. It's a 3000-pound object with 3000 pounds worth of inertia (yes, i know pounds are not mass, sheesh), limited maneuverability and limited stopping distance. You have voice mail, don't answer the phone while you're driving.

People just don't understand that driving is not a "minor attention" activity. You really need to pay attention.

Of course, there are just a lot of people out there who just can't @#$%ing SHUT UP FOR 10 MINUTES IN A ROW... there are a few people who I see on my commute every day who are always on the phone every time I see them. Once I drove my whole commute (45 minutes in LA traffic, when I'm in a car as opposed to my motorcycle) within a couple of cars of a person who was on the phone the entire time.

Unbelievable.

I think that some towns have made it illegal to talk on a hands-required phone while driving. Bravo.
posted by zoogleplex at 3:36 PM on July 12, 2005

In case anyone else is tempted to trot out the "you're just as distracted talking to someone in the car" line, you might first want to read the link to a previous thread I posted above.

Because the answer is no, you're not.
posted by soyjoy at 7:02 PM on July 12, 2005

as i said in a previous post, the cell phone is a good idea gone horribly wrong.
posted by brandz at 7:25 PM on July 12, 2005

Now while I don't like people driving with their phones, I have to take issue with the idea that talking on a hands-free phone is very different from talking with someone in the car. Soyjoy, your link didn't work so I'm still believing this... having studied attention and cognitive resource management a little bit in school it seems that the two are totally equivalent. In fact, it is reasonable to suggest that if a person is in the car they are a physical and visual presence as well and would be more distracting.

Someone tell me how a hands-free set is worse than normal conversation, if it is an intuitive system anyway.
posted by BlackLeotardFront at 8:04 PM on July 12, 2005

well, I would think one factor would be that the other party to the conversations is also risking their life, so maybe their comments would be better timed?
posted by Iax at 8:57 PM on July 12, 2005

You know, seeing that our roads are getting SO much safer these days, let's figure out more ways to distract drivers attention!

The combination of cell phones and driving is a plague. There should be a national law against it and auto makers should install devices that scramble a cell phone signal when the engine is running.

But oh no you say! This is a violation of our civil liberties!
Well tell that to the family of the safe, responsible driver that is killed by a cell phone distracted driver.
posted by cpchester at 9:31 PM on July 12, 2005

I think that some towns have made it illegal to talk on a hands-required phone while driving. Bravo.
It is illegal here in Australia to drive while using a non hands-free phone, including using the phone to send SMS messages.

The thinking behind speaking on the phone being more dangerous is that, when speaking to a person in the car, they can see what is going on and instinctively adjust, while a person on the other end of the phone just babbles on as you careen over a cliff.

zoogleplez, a lot of people use mobile phones to turn otherwise "wasted" commuting time into productive time - arranging appointments, networking, blah blah blah.

There should be a national law against it and auto makers should install devices that scramble a cell phone signal when the engine is running.
Good idea - ban using the phone while driving. While we are at it, let's ban smoking, changing CDs, listening to music, talking, etc etc. It should be illegal to do anything while driving except sit with straight back, hands at 10 to 2, perfectly alert and concentrating totally on the job at hand.
posted by dg at 9:51 PM on July 12, 2005

Just because it's hard or impossible to ban doing some things while driving doesn't mean we should give up and allow everything while driving.

Eh. This argument will go away when people are no longer allowed to drive while "driving" -- when cars become automated and you just have to punch in a destination and sit there and wait for the car to get you there. Which I hope happens soon.
posted by pracowity at 11:14 PM on July 12, 2005

That reminds me of a Zen teaching. The Master said: "when I'm eating, I'm eating, when I'm walking, I'm walking".

I know there are people who think they can do a thousand wonders at the same time, and they actually can in some occasions, but the fact is and this article shows right the opposite.

I bet only a soft music would be healthy while driving a car. Because it's anti-stress, it helps you keep calm while in a heavy traffic. Any kind of talking, even with people inside the car, means distracting from driving. That means crashing. The 10 minutes statistic shows that your mind doesn't move on from the conversation and keep you distracted for a long time after the call. And that means crashing again.
posted by nandop at 6:18 AM on July 13, 2005

I have to take issue with the idea that talking on a hands-free phone is very different from talking with someone in the car. Soyjoy, your link didn't work so I'm still believing this... having studied attention and cognitive resource management a little bit in school it seems that the two are totally equivalent.

The link did work. It was to a MetaFilter thread where this whole topic was picked apart and put back together. The link within the FPP had died, but I guess I assumed anyone who still needed precise official confirmation would check for the key terms and read the study, rather than continuing to reassert their own gut feeling as the final word.
posted by soyjoy at 7:47 AM on July 13, 2005

The thinking behind speaking on the phone being more dangerous is that, when speaking to a person in the car, they can see what is going on and instinctively adjust, while a person on the other end of the phone just babbles on as you careen over a cliff.

What about talk radio?
posted by SAC at 8:59 PM on July 13, 2005