Pre-preliminary 2009 Bicyclist Fatality Report

[update: this posting has been superseded -- please see:

Manner and Fault in Bicyclist Traffic Fatalities: Arizona 2009

which has detailed, finalized stats and discussion]

First, some perspective: In the state of Arizona, approximately 1,000 people are killed per year in traffic collisions of all types. The number of cyclist (usually called a “pedalcyclist” in the jargon) fatalities fluctuated between15 and 36 per year over the past twenty years, with an average of about 25/year.

[as a sidebar, Arizona total traffic fatalities which have been as high as 1,293 just a couple of years ago, were down to 937 for 2008.  The exact cause of this happy trend is a matter of great debate, e.g. the effect of economic recession, and photo-enforcement. Even after this dramatic reduction Arizona roads remain significantly more dangerous than US averages ]

I have become increasingly frustrated by what seems to me to be short-shrift paid to analysis of crashes resulting in a cyclist’s serious injury/fatality Read the rest of this entry »

It’s official; 2009 state-level NHTSA traffic fatality figures published

Get them here.

Here’s a typical national news story, from the LaTimes Traffic Deaths and Injuries Plummet in 2009: “Fatalities drop 9.7% from 2008 as the number of deaths dips to its lowest point since 1950, the Transportation Department says.” The official toll for 2009 is 33,808.

Closer to home, the total for 2009 in Arizona of 807 traffic fatalities represents a significant year-over-year drop that began in 2006, when there were 1,293 fatalities.

Per mile (VMT) figures won’t be available for awhile; it appears Arizona, which is consistently more dangerous than US averages will continue to close the gap.

Per capita figures show Arizona, again, consistently more dangerous than US averages, but continuing to improve.

Bicyclist Statistics

Bicyclists, statistically did not fare well in 2009. At 25 deaths, that is 6 higher than 2008, bucking the overall trend. Though the usual caution applies, the number of cyclist deaths is (thankfully) quite small, so a variation of just a few makes large percentage differences, and trends are harder to discern year-over-year.

Prior to the official release, i was aware of 16 fatalities; I am now aware of another 9 but they are  identified only by date time and (usually) location — but not name.  I was shocked to find out that Phoenix had 9 fatalities, of which I only knew about 4 previously. How can that be?

You can view a summary spreadsheet of each of the 25 fatalities here.

There is a detailed report available: Manner and Fault in Bicyclist Traffic Fatalities: Arizona 2009.

Arizona “Benefits”?

The Arizona Republic’s take included the odd conclusion that “Arizona benefits from being a younger state” thus the roads here are newer thus safer.

Arizona benefits from being a younger state. Because most development here is relatively recent, the roads are newer and designed to safer, more modern standards.

That means wider lanes and shoulders, better signs, smoother curves and banks, more guard rails and more innovations such as rumble strips, which are ruts in the sides of highways that alert drivers when they veer off the road.

“These are things people drive by every day which they may or may not notice. But they all contribute to make our roads safer,” said Laura Douglas, an Arizona Department of Transportation spokeswoman.

Rumble strips, for example, reduce the accident rate by a third, she said. ADOT also paints extra-thick road stripes, installs new guard rails that cushion crashes and uses larger, easier-to-see traffic signals, Douglas said.

Arizona traffic death toll drops to a 16-year low 9/11/2010, The Arizona Republic

This might be true in isolated examples, such as the rumble strips. Overall, though, this ignores the human-behavior dimension of driving. So, e.g. newer roads are much wider and straighter and the likely result is that drivers will drive faster. Maybe you’ll get relatively fewer wrecks but the ones that occur will be more violent as a result. If you look at the state-level NHTSA figures you will find the safest state is….. drumroll please… Massachusetts! A very old state. Arizona’s VMT rate is over twice as deadly as Massachusett’s. The disparity in per capita rate, since Arizonans drive more miles, is even worse.

Massachusetts happens to be the safest state in the US, but it’s not an outlier, the relationship holds up generally — states where most development pre-dated automobiles have far lower death rates, and vice versa.

“Better” roads also have a vicious circle effect of raising the number of miles driven, thus exposing one to more risk, albeit a decreasing risk per mile. In other words, dwelling on rate per VMT is misleading. Likewise, “better” vehicles, which are from an engineering perspective are much safer, have not yielded the expected improvement. Since as usual, human-behavior kicks in and drivers, knowing their vehicles are “safer”, (unconsciously or not) drive just a little more risk.

FARS

Cindie Holub’s death on March 1,  (Cindy’s death was written up on bicyclelaw.com, also see 2010 fatalities), from injuries sustained in a Feb 24 collision with a garbage truck caused me to look up the rule for categorization purposes. “To be included in this census of crashes, a crash had to involve a motor vehicle traveling on a trafficway customarily open to the public, and must result in the death of a person (occupant of a vehicle or a nonmotorist) within 30 days of the crash.” from DOT HS 811 137.

The US DOT runs a very elaborate, publicly available, query-able database for every traffic fatality in the US called FARS — Fatality Analysis and Reporting System. Read the rest of this entry »

New Crash Forms / ALISS database

As of Jan 1, 2009(?); Arizona has new forms. The old “Arizona Traffic Accident Report” will now be an “Arizona Crash Report” (as an aside, if you don’t know why that is significant, please see here). You can see what’s on the new form here, in a presentation by Rick Turner; includes the tantalizing bullet point “Customers Will Be Able to Query, Analyze and Retrieve Their Own Crash Data”. Read the rest of this entry »

Incidence of Pedestrian and Bicyclist Crashes by Hybrid Electric Passenger Vehicles

This ponderously-name technical report from the NHTSA, Incidence of Pedestrian and Bicyclist Crashes by Hybrid Electric Passenger Vehicles [DOT HS 811 204 pdf here]

Has a bunch of interesting tidbits. It’s obviously leading or suggesting that quieter motorvehicles, particularly at low speeds where tire noise would be less significant, have a tendency to not be heard by cyclists or peds thus leading to more crashes. Sounds plausible. The difference may explain some human behavioral factors of operators of bicycles; such as why cyclists rarely make a complete stop, yet rarely get seriously injured in those situations. (Motorist, too, rarely stop but that’s another story)

What mainly caught my eye was a juicy dataset as describe in the METHODS section “State crash files from NHTSA’s State Data System (SDS)… The SDS includes all police-reported crashes, regardless of the injury or crash outcomes”. Though the SDS actually only contains records for 32 states, arizona not being one of them.

Who’s Responsible?

There is a claim floating around that some study has concluded that motorists are responsible for some 90% of car-bike collisions.

This would be a lot higher than is generally appreciated. I’ve grappled with this a little bit before in Understanding Collision Summaries, where I pointed out an inexplicably high proportion of  “other” violations assigned to bicyclists.

So far, I’ve found a page at projectfreeride.org with a table that is said to be source from Tomlinson, David. Conflicts Between Cyclists and Motorists in Toronto, Canada. Link to a .pdf on the Velomondial.net.

The same claim can be found in a newslettery article dated Aug 19, 2009 on a University of Toronto website entitled Smart Cycling. the information was supplied by a physcian, Dr. Chris Cavacuti, who is also involved with projectfreeride. And a correction with that article that was posted Aug 26:

In the interview, Dr. Cavacuiti is quoted as saying “The [Toronto Collision] study concluded that cyclists are the cause of less than 10 per cent of bike-car accidents”. Dr. Cavacuiti has asked us to make readers aware that the Toronto Collision study was actually designed to look at the cause of bicycle/motorist collisions but not culpability.

It is actually several studies conducted by the Charles Komanoff and member of the Right of Way organization in New York that concluded that concluded that cyclists were strictly culpable for less than 10 per cent of bike-car accidents.

Dr. Cavacuiti would like to apologize for any confusion this error may have caused.

On the projectfreeride page,  in a statement summarizing Tomlinson’s findings, the page at projectfreeride says “In fact, cyclists are the cause of less than 10% of bike-car accidents in this study”. Is that really what Tomlinson found? Or should the correction mentioned above be also applied to the projectfreeride page too?

This claim got picked up by the Freakonomics blog, garnering wide exposure.

Skepticism at the commuteorlando blog. Links the 90% claim back to Komanoff’s group Killed by Automobile paper. More links here on cycledog.

(more to come…)

MAG Stats

There was a nice little piece in the local West Valley paper, Surprise a safe zone for bicycle riders.

They make some rather outlandish claims, though.

According to figures recently released by Maricopa County Association of Governments, Surprise has one of the lowest rates of injuries and fatalities in the county.

Surprise has an average rating of 5.08 per 100,000 of its population for bicyclist injuries and fatalities and 6.31 for pedestrian injuries and fatalities from 2003 through 2007, the latest statistics…

This contrasts with other neighboring cities such as Peoria — with ratings of 16.43 for bicyclists and 11.71 pedestrians, and Glendale — with 26.68 for bicyclists and 31.27 for pedestrians.

Tempe ranks as the most dangerous city for bicyclists with a rating of 93.57 injuries and fatalities per 100,000 and 54.5 for pedestrian injuries and fatalities.

Is bicycling really almost twenty times more dangerous in Tempe relative to Surprise? One imagines it has more to do with the amount of cycling relative to population. And other distortions, e.g. maybe ASU students aren’t counted as Tempe residents.

Here is MAG’s Transpo Committee page, to see the data referred to in the story, open page Crash Trends in the MAG Region 2001-2007 and then click on “Injuries and fatalities per 100K population”, and a table with per-city data pops into the same page below the existing stuff (thanks to Sarath for pointing that out).

Crashes are the Leading Cause of Death…

I found this image lurking on the NHTSA website. MVCs (Motor Vehicle Collisions) are always the leading cause, though the exact ages vary from year to year, e.g. from Motor Vehicle Traffic Crashes as a Leading Cause of Death in the United States, 2000 “motor vehicle traffic crashes are the leading cause of death for every age 2 through 33″. Note that this ranking is all inclusive; thus it includes things like suicide, homicide, and so forth.

Just like everything else in life, there are some nuances that are worth understanding. The simplest distinction is between internal (think disease) and external (think any sort of accidental death; car crash, drowning, falling…). These distinctions are detailed in the technical report, e.g. Motor Vehicle Traffic Crashes as a Leading Cause of Death in the U. S., 1997 (emphasis added):

“As a major external cause, traffic crashes are the prime cause of accidental death in the United States, and this has been true for many years. Thus, for persons of all ages, traffic crashes alone in 1997 caused almost one-half of all accidental deaths that occurred….  “

For example, from Exhibit 5 here are the top 5 causes of “accidental death” for both sexes combined. MVCs DOMINATE the rankings.

1. Motor Vehicle Traffic Crashes 42,340
2. Falls  15,477
3. Poisoning  10,163
4. Other and Unspecified Causes (including suffocation which was #4) 5,207

…

MVC’s (Motor Vehicle Collisions) are so horrifically high, that they have even snuck into the debate over Universal Health Care in the US. It seems that “unnatural” causes of death (MVC being the prime category) are so high in the US that they have significantly depressed our life expectancy. By adjusting the life expectancy data for all the OECD countries (except Luxembourg), the US catapults from last to first place! This is all according to University of Iowa researchers  Robert L. Ohsfeldt and John E. Schneider.

More Photo-enforcement in the WSJ

On the heels of last weeks “front pager” — Jenkins thows in his two cents in today’s column The War on Short Yellows. His punditry is undoubtedly astute: “One Arizona sheriff recently proved you could get elected by opposing speed cameras”. He should have stopped there, since his analysis of safety is lacking. Firstly, he either doesn’t know, or doesn’t let on, the scope of the problem. To put it simply, traffic collisions are the leading cause of unnatural death for all Americans (link to reference here)… this is a huge problem.

And the problem is even worse in Arizona; something he either doesn’t know or doesn’t care about. Arizona rates (even after some fairly large improvements in recent years) far above US averages in both per capita fatalities, and fatalities per 100M VMT. So it should probably come as no surprise that the authorities in Arizona are trying out things like photo-enforcement. Which he, reflexively, believes is basically a jack-booted government gone wild.

He goes with the typical cannard — that supposedly the collisions prevented represent only a small fraction of all collisions. His exact stat was “Consider: Red-light running and speeding, the two main uses of traffic cameras, are implicated in fewer than 8% of accidents”.  He doesn’t reveal a source (possibly a talking point from the NMA?), I’m guessing it is 3% + 5%, and also guessing it’s the national causastion survey. In any event, the weakness is that these collisions are far more freqently fatal. Arizona has a particularly high fatal red-light running rate.

He even brings up Britian, yet he either doesn’t know, or doesn’t let on that Britian experienced a precitious decline in fatality rates through the 1990′s — coincident with the rise in photo-enforcement. Are the two related? One wonders, but Jenkins apparently doesn’t care or wonder. By the way, fatality rates are far below US rates (both per capita, and per VMT).

His solution? lengthen yellow lights. This would undoubtedly reduce violations. But unless the yellow is “short” (shorter than engineering standards) there’s no indication this would reduce collisions, though. And as to the other ten’s of thousands of deaths annually? Well he doesn’t even have a suggestion for that.

2007 Arizona claims enormous improvement in VMT fatality rate

I guess it takes a long time for the VMT state-by-state rates to trickle out — but they are all here. The numbers are close to, but not the same as, ADOT’s 2007 Motor Vehicle Crash Facts.

The VMT fatality rate for 2007 is 1.69 fatalities per per 100 million Vehicle Miles Traveled. Read the rest of this entry »