[Update: this article was written way back in 2011; since then i continued to update links to the yearly Publication; the 2017 data is currently most recent as of 9/17/2019]
Each year, the USDOT, NHTSA (United States Dept of Transportation / National Highway Traffic Safety Administration) issues a report called Traffic Safety Facts: Bicyclists and Other Cyclists xxxx. The report comes out about 18 months after the close of the calendar year under review.
I wonder if this / this data would be more useful if it were on a simple spreadsheet — I extracted one of the tables last year and it was a P.I.T.A.
NHTSA publications currently lists an edition of this report for the year 2017, all the way back to 1993:
They tend to list the most recent 10-year period, so e.g. 2011 lists 2002-2011 data
One of the things in each Traffic Safety Facts publication is a 50 state, state-by-state breakdown; which otherwise is pretty hard to find (individual states are easily found here). As a by-product of that stat, the state’s fatality rate per population is listed, see arizona-has-the-highest-cycling-fatality-rate? for some comments on that particular statistic. They also give some standard breakdowns as to age, gender, urban/rural, month of the year, etc. In the 2014 edition, Tucson and Phoenix were singled out as… “The large cities with the highest pedalcyclist fatality rates were both in Arizona: Tucson (11.36 pedalcyclist fatalities per 1,000,000 people) and Phoenix (7.16 pedalcyclist fatalities per 1,000,000 people),” p. 8. Ouch. But without exposure data, this is hardly useful information.
A stat they have just started publishing (since 2009) each year is the “(pedalcyclist) percent of overall fatalities” — reports from before 2009 listed a 10 year line-graph of total pedalcyclist fatals. The percentage seems better and more relevant.
To (cherry?) pick one datapoint; the decrease in traffic fatalities (2010, I think i was referring to) compared to 2007 has been dramatic, a 20% decline in only 3 years! There is much speculation as to why, e.g. the economic recession in the US is thought to play a large role, vehicle improvements continue to advance; more cars have airbags; large SUV rollovers are killing fewer of their own occupants due to Electronic Stability Control (see suvs-becoming-less-deadly). These engineering improvements allow motorists to “crash safe”.
Unfortunately if you unpack the numbers into broad categories, the NHTSHA breaks them down into three (see Table 2 of Traffic Safety Facts 2010, reproduced in the comment below):
- Motor Vehicle Occupants: i.e. drivers and passengers of enclosed vehicles
- Non-vehicle Occupants: includes bicyclists, but is predominantly pedestrians
The differences are rather stark; while MV occupants have enjoyed a 24% decrease in fatalities , Non-vehicle occupants have seen only a 9% decline.
There could be a lot of reasons for this disparity… perhaps for the same reasons that bad economic conditions are having the effect of reducing the number of MV occupant fatalities , they are also perhaps causing more people to bike/walk.
I am not aware of anyone looking at this, though it seems like a big deal.
Stats for injuries can also be derived from the same table of information (table 2 of the general 2010 TSF); it shows a continuation of the disturbing trend but interestingly it also shows a disparity between injury rates vs. fatality rates. For injuries, MV occupants are down 10%, while Non-occupant injuries are UP 5%. Note this is still the same “swing” of 15 percentage points. But it does also beg the question of why fatalities are falling faster than injuries. Better trauma care?
Various standard measures of exposure listed in Table 1 of TSF are Resident Poulation, Number of Licensed Drivers, Number of Registerd MVs, Vehicle Miles Traveled have all, as expected, not changed much in the course of 3 years. The former former 3 have all increased slightly, while VMT decreased slightly (+3, +2, +1, -2 % respectively). Fatalities per VMT is perhaps the most closely-watched measure of traffic safety; and is currently at a historic low of 1.11 Fatals per 100Million VMT.
None of these exposures really tells us what might be happening differently with non-occupants; so we’re left to speculate/hypothesis — e.g. perhaps due to the bad economy, the decrease in mile traveled by MV was partially replaced by more bicycle or walking trips. And since there are proportiately more MV trips/miles traveled compared to bikeing/walking, a small number of displaced miles has the effect of increasing bike/walk exposure markedly.
Traffic Safety Facts Annual Reports
There’s also a much longer document put out each year called the Traffic Safety Facts FARS/GES Annual Report; so the library page has a link to search for just TSF FARS/GES annual reports. E.g. the most recent one available as of this writing is 2012, 812032.pdf (2011 is 811754AR.pdf) and runs about 200 pages. One guesses this contains *all* the data presented in the specialty TSFs, e.g. motorcyclists, bicyclists, speeding, alcohol, etc.