I’m continually wondering about the risks involved in the outdoor activities I love. I probably won’t ever get a clear answer for most of them, because most have too many variables and too little data available. Still, I keep coming back to the questions. Shouldn’t it be possible to at least come up with a pie-in-the-sky guess at the degree of risk inherent in an activity? I don’t know, but I do know one thing I would need first: a basis of comparison. My impulse is always to compare outdoor risks with the risk of an activity nearly everyone participates in, driving. So today made my way to the National Center for Statistics and Analysis, and downloaded the 2004 Traffic Safety Facts Annual Report.
My first question is, what is the risk of dying inherent in the activity? Clearly we need to know the number of participants and the number of fatalities associated with the activity to answer this. For driving, there’s another place to answer this question, the the Odds of Dying (2002) page at National Safety Council. For that year, the odds of a “car occupant” dying were 1 in 17,625. Doing the same calculation for 2004 from NCSA report, the odds that year were 1 in 8,862. Wow, is driving getting much more risky? It turns out 2002 was a record low year for auto fatalities in the US. Maybe we should look at a few years, using prior NCSA reports and US population estimates:
|Year||Odds of dying as a car occupant|
|1992||1 in 7755|
|1993||1 in 7677|
|1994||1 in 7584|
|1995||1 in 7445|
|1996||1 in 7429|
|1997||1 in 7494|
|1998||1 in 7639|
|1999||1 in 7604|
|2000||1 in 7573|
|2001||1 in 7623|
|2002||1 in 7528|
Wow, much different results for 2002 that way. An illustration, I guess, that risk estimates will vary wildly even with loads of data. There’s also an assumption here that the entire US population are “car occupants” at some point – seems likely enough for this guesswork. Even if the real risk is half, or twice this estimate, at least we have something.
It would also be nice to do some numbers for injuries, not just fatalities. Scanning the reports, I can see that numbers for injuries are even sloppier than for fatalities. I can see that the odds of injuries appear to be declining since 1996, so let’s take the low number and say there are about 75 injuries for every fatality. If we estimate the risk of dying in a car as 1 in 8,000, the risk of injury would be 1 in 106.
Now I would like to take these numbers and do some guessing about what the odds are for other activites, but that’s for another leisurely morning…