• How dangerous are cell phones to drivers?

    So yesterday the NTSB recommended a national ban on cell phone use for drivers:

    The National Transportation Safety Board recommended Tuesday that all states and the District ban cellphone use behind the wheel, becoming the first federal agency to call for an outright prohibition on telephone conversations while driving.

    Distracted driving, some of it due to cellphone use, contributed to an estimated 3,092 deathsin highway crashes last year, according to the National Highway Traffic Safety Administration.

    “No call, no text, no update, is worth a human life,” said NTSB Chairman Deborah A.P. Hersman. “It is time for all of us to stand up for safety by turning off electronic devices when driving.”

    Regular readers of the blog will know that such language doesn’t sit well with me. As you know, the number one killer of children in the US is accidents, and yet no on ever says such things as “driving just isn’t worth killing so many children”.  Ignoring the comment, though, just how dangerous are cell phones compared to other distractions?

    To get some actual data, I headed over to the NHSTA Fatality Analysis Reporting System’s Query Tool. I asked for the number of vehicles/drivers involved in fatal crashes 2010, broken down by distractions just prior to the crash. Here’s a graph of the output (click to enlarge):


    The first thing to note is that the vast majority of accidents (77%) don’t involve distractions. An additional 8% are categorized as “unknown if distracted”. Further large categories include “not reported” at 7% and “Distraction/Inattention, Details Unknown” at 3%. So that’s a total of 95% where there was no distraction, where it’s unknown if there was a distraction, or details of a distraction are unknown.

    How do things look if we eliminate those categories? Like this:

    I colored in all cell phone related categories as red. Are they there? Yes. But even combined together, they aren’t close to as risky as being “lost in thought”. Altogether, in 2010, 373 vehicles/drivers were linked to cell phone related distractions. That’s less than 1% of all of them.

    So are cell phones distracting to drivers? Yes. But are they dangerous to a level necessitating the rhetoric seen above? I don’t know. That’s partly because I don’t understand what the threshold is for such language. Eating and drinking were linked to 51 vehicles/drivers in fatal crashes. Will the NTSB declare that eating in a car isn’t worth a human life and call for a ban on food and drink? Radio and climate controls were linked to 49 vehicles/drivers in fatal crashes. Will the NTSB declare that listening to music or driving at a comfortable temperature aren’t worth a human life and call for a ban on radios and air conditioning?

    No? Why not? What’s the threshold for a nationwide ban? How many lives are too many?

    I don’t know what the answer is. But I think that people who make declarative statements and set policy should be prepared to offer an answer. Alternatively, they could present the data in absolute as well as relative terms, and allow people to make their own decisions.


    • By this logic, we wouldn’t be spending so many billions on airport security theater and would go back to the pre Sept 2001 system.

    • I don’t disagree with you in your point about policy makers and declarative statements, but we certainly should consider the source of the FARS data and the human reporting factor involved to come to the conclusion that fatal accidents (hereafer referred to as FAs) involving cell phones is probably under-reported.

      A quick Google search provides the FARS explanatory brochure, found here: . The brochure indicates that FARS is an aggregate of state reported data, including: police accident reports, state highway department data, death certificates, coroner/ME reports, hospital medical reports, and emergency medical services reports. It would take a bit of looking into these reporting mechanisms, but one could guess that a person directly involved in the FA could be 1) ashamed to admit that a cell phone was involved 2) too traumatized to remember and report accurately or 3) outright lying to protect their driving record or change at a favorable insurance claim, or that of a related party. Those not directly involved (i.e. onlookers) probably can’t tell if a person is using a cell phone prior to the FA given that their point of attention likely begins following the catalystic event (in this case, cell phone use).

      Since supporting data is always nice, the query tool turned up that, in 2009, 72.4% of fatalities were vehicle occupants, 13.2% were motorcyclists, and 14.4% were non-motorists. Of the majority category, 72.1% of fatalities were the driver and 27.7% were the passenger (with 0.3% thrown in as Unknown Occupant). Therefore, it follows that most of the people doing the first-hand reporting to the police or emergency responders are passengers in the vehicle or vehicles involved in the FA. These folks likely fall into categories 1 and 2, and possibly 3 – leading to under-reporting the actual event catalyst.

      Now as to how this impacts things – you would have to estimate the degree of underreporting and apply the adjustment to determine just how dangerous the cell phone categories actually are relative to other, non-cell phone categories. Of course, that’s all contingent on the assumption that it’s just cell phone use that’s under-reported, which is likely not the case. Most people will likely put the FA in the “Not Distracted” category, worried that admitting distracted driving will put them on the path of criminal or civil liability. All in all, the source data may make this database of good intent, but not such practical use unless your a policy maker looking to make declarative statements based on potentially biased statistics.

      • I also suspect under-reporting is common. However, if cellphone usage is a significant contributor in fatal accidents, I would expect there to be at least some correlation with cellphone subscriber growth. The number of cellphone subscribers has grown substantially in the past 15 years (http://www.infoplease.com/ipa/A0933563.html). However, the number of fatal accidents seems to be constant or perhaps even declining over the same time-period (http://www-fars.nhtsa.dot.gov/Main/index.aspx).

        I am not stating this as evidence to support an argument, but rather asking the question to see if I am overlooking something.

    • This logic is faulty. Let’s say that drinking boiling acid has a 100% chance of accident, but is something only a single person has ever attempted to do while driving. The fact that it would be at the bottom of the chart with a single death does not mean that it “isn’t close to as risky as being lost in thought”. You have to divide these chart values by their total population incidence when not involved in an accident to determine their “risk”. So, while your argument has some interesting properties, without examining details such as “drinking coffee in the car may be 10x as common as using a cell phone while driving”, you cannot rely on these items being comparable.

    • Dr. Carroll~
      I listen to you all the time on Stand Up. While I’m inclined to agree with you, I have to wonder how accurate these statistics are. How are we to know that people actually report the reasons accurately? Many people might be inclined to tell their insurance company or the police that they were “lost in thought” rather than admit they were using their cell phones at the moment of an accident. All of these causes would have to be self-reported, yes?

      Also, even if we take it that only 1% of accidents are caused by cellphones, why does that diminish the affect on people? If only 1% of accident fatalities are caused by drunk driving, we would still try to prevent drunk driving, yes?

      If you’re arguing that its dishonest to fight cellphone use when eating and other causes do way more damage, I understand that.

      I think the most important question here is whether cellphone/texting laws have any impact on accident rates. Thoughts?

      • It’s the inconsistency. I agree that the statistics are imperfect. But that’s the point. Why make policy on bad data? And why pick cell phones and not radios?

        Make the case.

        • Here in the UK we have a ban on using a hands-on cellphone for calling or texting. It’s based on solid research – the headline being you’re four times more likely to crash. Here’s a report from our main accident charity:

          We call them mobile phones – better to search on that for references rather than cellphone.

    • I am a sample of one, but in my experience talking on a cell phone is roughly as distracting as talking to a passenger, but nowhere near as distracting as trying to text or dial on a handset. I have almost crashed texting/typing/reading email, but never talking on the phone.

    • You might give more weight to these numbers if your family were to be permanently shattered because another driver had an “important” call to make.

    • What about my number one distraction, pretty girls jogging?

    • Will we ever see information on how many lives have been saved by cell use. For instance; the call to avoid a road from a friend, the talk that keeps a tired driver awake the few minutes needed to get to a hotel or home.
      And also, like your stuff on Pete’s show, thanks

    • I’m a retired police officer from the UK and visit Florida two or three times a year and find your current debate fascinating. Over here cell phone use has been banned for a few years now because it was scientifically proved beyond doubt that talking or texting on a phone whilst driving is as bad, if not worse than driving whilst under the influence of drink or drugs. Part of the statute gave police the powers to seize cell phones from all parties involved in fatal or potentially fatal crashes to have them forensically analysized to prove or disprove that a driver was on his/her phone at the time of the crash. We don’t rely on people telling the truth in such circumstances because they invariably lie!! The statistics went through the roof and the courts started jailing offenders for 5 years or more. The police are also empowered to stop a driver they see using a cell phone whilst driving and issue a ticket for £60 (about $90) plus three points on their licence. No other offence needs to be seen first in order to stop and prosecute. I’m not saying its stopped the practice completely but over in the UK using a cell phone whilst driving has become as socially unacceptable as drink driving.

    • You might give more weight to these numbers if your family were to be permanently shattered because another driver had to change the station.

      You might give more weight to these numbers if your family were to be permanently shattered because another driver had to get something out of the glove box.

      You might give more weight to these numbers if your family were to be permanently shattered because another driver had to eat his hamburger.

      Need I continue?