How bad does driving kill me?
January 21, 2011 9:22 AM
How much time is removed from your statistical life expectancy for each hour you drive in a car?
Is there enough data out there about this? I bet there is.
People who drive all the time are more likely to die in car accidents. Therefore, as a group, their life expectancy will be less than people who don't drive all the time.
Now, let's assume that the population of the drivers and that of the non-drivers is sufficiently diverse that the driving is the only meaningful difference between the groups*.
Now, plot the hours spent driving vs life expectancy and draw a best-fit curve. What's that look like?
I want to be able to say "Look, I know it's only a four-hour-long drive, but it also removes a half-hour from my life expectancy, so I'm going to consider it a four-and-a-half-hour-long drive." Cuz that'd be fun.
Anyway, has this study been done? If not, where could I get the data?
* The drivers/non-drivers might have other differences besides driving. Maybe the drivers have more money than the non-drivers and will live longer. Or maybe the non-drivers get more exercise and will live longer. That sort of thing...
Is there enough data out there about this? I bet there is.
People who drive all the time are more likely to die in car accidents. Therefore, as a group, their life expectancy will be less than people who don't drive all the time.
Now, let's assume that the population of the drivers and that of the non-drivers is sufficiently diverse that the driving is the only meaningful difference between the groups*.
Now, plot the hours spent driving vs life expectancy and draw a best-fit curve. What's that look like?
I want to be able to say "Look, I know it's only a four-hour-long drive, but it also removes a half-hour from my life expectancy, so I'm going to consider it a four-and-a-half-hour-long drive." Cuz that'd be fun.
Anyway, has this study been done? If not, where could I get the data?
* The drivers/non-drivers might have other differences besides driving. Maybe the drivers have more money than the non-drivers and will live longer. Or maybe the non-drivers get more exercise and will live longer. That sort of thing...
Data on the number of accidents per mile driven are primarily available in the Federal Highway Administration’s annual publication Highway Statistics.1 This source contains data on the number of fatalities and injuries in accidents per mile driven. As for fatalities, there are 0.8 fatalities for each 100 million miles driven on interstate highways and 1.8 fatalities per 100 million miles driven on other principal arterial highways.
posted by jon1270 at 9:29 AM on January 21, 2011
posted by jon1270 at 9:29 AM on January 21, 2011
that said, there's lots of data out there....
Death rates by make/model.
FARS data.
posted by pjaust at 9:34 AM on January 21, 2011
Death rates by make/model.
FARS data.
posted by pjaust at 9:34 AM on January 21, 2011
On top of everything, this depends on your age and life expectancy when you get in the car. For example, if you only have half an hour to live, a car drive can't take an hour off your life. (The most you can lose in a car crash is equal to your counterfactual life expectancy at time of crash.)
So I'm actually not sure a straight-line estimate would be very illuminating here.
posted by grobstein at 9:40 AM on January 21, 2011
So I'm actually not sure a straight-line estimate would be very illuminating here.
posted by grobstein at 9:40 AM on January 21, 2011
As far as I know, all the risk information is calculated in miles, not minutes. You may have to do some weird data-set wrangling to get even near what you want (simple fatality rates doesn't tell you reduction in life expectancy).
If you have access to car or health insurance actuarial tables, someone may have already done this math.
posted by klangklangston at 9:42 AM on January 21, 2011
If you have access to car or health insurance actuarial tables, someone may have already done this math.
posted by klangklangston at 9:42 AM on January 21, 2011
If there are 0.8 fatalities per 100,000,000 miles as pjaust says, and you drive at an average of 50mph (I just made that up) then on an hourly basis, your chances of dying are:
100,000,000/50 = 0.8 fatalities per 2,000,000 hours.
So, per hour, your chances of dying are 0.8/2,000,000 = 4x10^-7, or 0.0000004 fatalities per hour.
The average life expectancy in the US is 78, and the average age of an American is presumably roughly have that, so we'll say 39.
That means, *if* you die in a car accident, you're sacrificing about 39 years of life, on average.
Since your chances of dying are 0.0000004, then the number of years you sacrifice per hour is 0.0000004 * 39 = 0.0000156.
If you convert that to minutes, then you're sacrificing 8.2 minutes of life per hour driven.
However, note that this only applies to hours *not yet driven* because the chances of you dying in a car accident that happened while driving in the past, given that you're still alive, are 0. The implications of this are that the 1,000 hours you spent driving last year didn't shorten your life at all, as you're still alive, however, the 1,000 hours you'll spend driving next year will statistically shorten your life by 8,200 hours, which is 5.7 days.
And like pjaust says, there are lots of confounding factors. Are you driving a safe or unsafe car? Are you older or younger than average age (the older you get, the fewer years you're risking)? Are you a relatively safe or dangerous driver?
Someone else can take this farther than I have, I'm sure. I am also not a statistician, I may have left things out.
posted by tylerkaraszewski at 9:49 AM on January 21, 2011
100,000,000/50 = 0.8 fatalities per 2,000,000 hours.
So, per hour, your chances of dying are 0.8/2,000,000 = 4x10^-7, or 0.0000004 fatalities per hour.
The average life expectancy in the US is 78, and the average age of an American is presumably roughly have that, so we'll say 39.
That means, *if* you die in a car accident, you're sacrificing about 39 years of life, on average.
Since your chances of dying are 0.0000004, then the number of years you sacrifice per hour is 0.0000004 * 39 = 0.0000156.
If you convert that to minutes, then you're sacrificing 8.2 minutes of life per hour driven.
However, note that this only applies to hours *not yet driven* because the chances of you dying in a car accident that happened while driving in the past, given that you're still alive, are 0. The implications of this are that the 1,000 hours you spent driving last year didn't shorten your life at all, as you're still alive, however, the 1,000 hours you'll spend driving next year will statistically shorten your life by 8,200 hours, which is 5.7 days.
And like pjaust says, there are lots of confounding factors. Are you driving a safe or unsafe car? Are you older or younger than average age (the older you get, the fewer years you're risking)? Are you a relatively safe or dangerous driver?
Someone else can take this farther than I have, I'm sure. I am also not a statistician, I may have left things out.
posted by tylerkaraszewski at 9:49 AM on January 21, 2011
People who drive all the time are more likely to die in car accidents. Therefore, as a group, their life expectancy will be less than people who don't drive all the time.
How do you know this is true? What if the people who don't/can't drive are less healthy?
Population averages have little to do with your own individual risk.
posted by mandymanwasregistered at 10:33 AM on January 21, 2011
How do you know this is true? What if the people who don't/can't drive are less healthy?
Population averages have little to do with your own individual risk.
posted by mandymanwasregistered at 10:33 AM on January 21, 2011
I mean wouldn't it be more fun to find out that specific road's traffic volume and # of road deaths and go from there? C'mon!
posted by mandymanwasregistered at 10:47 AM on January 21, 2011
posted by mandymanwasregistered at 10:47 AM on January 21, 2011
... however, the 1,000 hours you'll spend driving next year will statistically shorten your life by 8,200 hours, which is 5.7 days.
You mean 8,200 minutes, right?
posted by Johnny Assay at 11:04 AM on January 21, 2011
You mean 8,200 minutes, right?
posted by Johnny Assay at 11:04 AM on January 21, 2011
Here's a study from the journal Medical Decision Making. From the abstract: "The baseline analysis suggested that 1 hour spent driving was associated with approximately 20 minutes of additional lost time in life expectancy due to the potential of a crash."
I don't know of any numbers on reduced life expectancy because of the decreased physical activity associated with driving, but here's something related.
posted by neal at 11:12 AM on January 21, 2011
I don't know of any numbers on reduced life expectancy because of the decreased physical activity associated with driving, but here's something related.
posted by neal at 11:12 AM on January 21, 2011
What's the underlying reason for wanting this info?
Life is risk. You are guaranteed to die from it. The question is when and how you'll live your life in ways that will let you thrive. Everything has risk and sometimes obsessing about the potential risks is worse than the risks themselves.
posted by wkearney99 at 7:20 AM on January 22, 2011
Life is risk. You are guaranteed to die from it. The question is when and how you'll live your life in ways that will let you thrive. Everything has risk and sometimes obsessing about the potential risks is worse than the risks themselves.
posted by wkearney99 at 7:20 AM on January 22, 2011
What's the underlying reason for wanting this info?
I hate driving and I want a new silly thing to say in order to talk trash about driving.
posted by Galaxor Nebulon at 7:01 AM on January 24, 2011
I hate driving and I want a new silly thing to say in order to talk trash about driving.
posted by Galaxor Nebulon at 7:01 AM on January 24, 2011
« Older How to obtain a Spanish police report from abroad? | Name of a defunct 80's restaurant chain? Newer »
This thread is closed to new comments.
Also, bear in mind that walking is probably more dangerous on a per-mile basis than driving, as is cycling. If you gotta get somewhere, a car is a relatively safe way to do it.
posted by pjaust at 9:28 AM on January 21, 2011