Review of Route Infrastructure and Risk of Injury
20 views
Skip to first unread message
John Forester
Nov 1, 2012, 5:39:46 PM11/1/12
to BicycleDriving
There has been considerable commentary about the paper by Treschke et al
titled Route Infrastructure and the Risk of Injury to Bicyclists,
American Journal of Public Health, current issue or thereabouts. The
paper looks stupendously important, with its long list of public health
professionals authoring it and using very complicated mathematical
programs, and the use of data about two points in each trip: where the
cyclist crashed, and another point where he didn't crash. Much is made
of the claim that cycle tracks are far and away the safest kind of
bicycling facility. However, what we actually have before us is the
result when total ignorance of traffic engineering combines with bicycle
advocacy ideology. I have written a review of this paper and posted it
on my website. For convenience to current email readers, I post only the
Conclusions section here.
Conclusions
I find that this paper (Treschke et al, Route Infrastructure and the
Risk of Injury to Bicyclists) does not deserve serious consideration for
two kinds of defects: the combination of incompetent traffic-engineering
with ideological argument. I offer two examples of this combination.
When considering the small reduction in crash rate produced by bike
lanes versus the larger reduction produced by the absence of parked
cars, the authors chose to advocate that produced by the bike lanes that
they favor.
When considering the astonishing reduction in crash rate produced by
cycle tracks, as shown by their data, the authors chose to proclaim that
reduction as genuine for cycle tracks in general.
In the bike-lane issue, the authors chose to proclaim the effect of bike
lanes, small though it was, rather than the larger effect of the absence
of parking. That combines traffic-engineering incompetence with
ideological propaganda.
In the much more impressive cycle-track issue, the authors proclaimed
enormous crash reduction without informing the readers of the two
relevant facts. First, that their data came from only one installation.
Second, that that installation was not along a typical city street but
in the only situation in which a plain cycle track could possibly be
safe, a place without crossing or turning movements by motorists,
cyclists, or pedestrians. The authors refer to the forty-year-old
cycle-track controversy as if they had studied it, but clearly they
don�t understand it. It is clear that the authors have such faith in the
cycle-track concept that its astonishing data failed to alert them to
investigate why such data was reported.
The authors� failure to understand the difficulties of cycle tracks and
the only conditions in which cycle tracks may be safe constitute
traffic-engineering incompetence.
The authors� proclamation of the great safety of cycle tracks and their
failure to be alerted to problems with the source of their cycle-track
data demonstrate the improper influence of ideological considerations.
The URL for my full review of the Treschke paper is at:
http://johnforester.com/Articles/Facilities/Infrastructure%20&%20Injuries.pdf
--
John Forester, MS, PE
Bicycle Transportation Engineer
7585 Church St. Lemon Grove CA 91945-2306
619-644-5481 fore...@johnforester.com
www.johnforester.com
titled Route Infrastructure and the Risk of Injury to Bicyclists,
American Journal of Public Health, current issue or thereabouts. The
paper looks stupendously important, with its long list of public health
professionals authoring it and using very complicated mathematical
programs, and the use of data about two points in each trip: where the
cyclist crashed, and another point where he didn't crash. Much is made
of the claim that cycle tracks are far and away the safest kind of
bicycling facility. However, what we actually have before us is the
result when total ignorance of traffic engineering combines with bicycle
advocacy ideology. I have written a review of this paper and posted it
on my website. For convenience to current email readers, I post only the
Conclusions section here.
Conclusions
I find that this paper (Treschke et al, Route Infrastructure and the
Risk of Injury to Bicyclists) does not deserve serious consideration for
two kinds of defects: the combination of incompetent traffic-engineering
with ideological argument. I offer two examples of this combination.
When considering the small reduction in crash rate produced by bike
lanes versus the larger reduction produced by the absence of parked
cars, the authors chose to advocate that produced by the bike lanes that
they favor.
When considering the astonishing reduction in crash rate produced by
cycle tracks, as shown by their data, the authors chose to proclaim that
reduction as genuine for cycle tracks in general.
In the bike-lane issue, the authors chose to proclaim the effect of bike
lanes, small though it was, rather than the larger effect of the absence
of parking. That combines traffic-engineering incompetence with
ideological propaganda.
In the much more impressive cycle-track issue, the authors proclaimed
enormous crash reduction without informing the readers of the two
relevant facts. First, that their data came from only one installation.
Second, that that installation was not along a typical city street but
in the only situation in which a plain cycle track could possibly be
safe, a place without crossing or turning movements by motorists,
cyclists, or pedestrians. The authors refer to the forty-year-old
cycle-track controversy as if they had studied it, but clearly they
don�t understand it. It is clear that the authors have such faith in the
cycle-track concept that its astonishing data failed to alert them to
investigate why such data was reported.
The authors� failure to understand the difficulties of cycle tracks and
the only conditions in which cycle tracks may be safe constitute
traffic-engineering incompetence.
The authors� proclamation of the great safety of cycle tracks and their
failure to be alerted to problems with the source of their cycle-track
data demonstrate the improper influence of ideological considerations.
The URL for my full review of the Treschke paper is at:
http://johnforester.com/Articles/Facilities/Infrastructure%20&%20Injuries.pdf
--
John Forester, MS, PE
Bicycle Transportation Engineer
7585 Church St. Lemon Grove CA 91945-2306
619-644-5481 fore...@johnforester.com
www.johnforester.com
Tricia Kovacs
Nov 1, 2012, 6:15:45 PM11/1/12
to bicycle...@googlegroups.com
I'm was just registering for the Journal of Health so I could download
and read the Teschke paper.
I told my husband (who is not a member of BicycleDriving) about John's
conclusions on the study.
I just had to share his stupendously astute comment: "living on earth is
much safer than not living on earth".
Tricia Kovacs
and read the Teschke paper.
I told my husband (who is not a member of BicycleDriving) about John's
conclusions on the study.
I just had to share his stupendously astute comment: "living on earth is
much safer than not living on earth".
Tricia Kovacs
Tricia Kovacs
Nov 3, 2012, 1:11:53 PM11/3/12
to bicycle...@googlegroups.com
Dear John,
Thank you for writing a review of the Teschke study. I read both the
study and your review and I agree with your analysis of the study. I do
have one question on a statement in your review:
Significant Characteristics
However, the presence of bike lanes showed only a small reduction in
crash probability as shown by odds ratio.
Teschke Study section titled Discussion
Bike lanes on major streets with no parked cars and off-street bike
paths had nearly half the risk of the reference.
This does seem significant to me.
But with you, I question the method of the study. They randomly selected
another portion of the cyclist's route where he/she didn't crash and
state that type of roadway is safe? I also question the sample size.
There were only 2 crashes in the cycletrack and only 10 incidents where
a cycletrack is a control site (no crash). Is that a large enough sample
size to be statistically significant when the total number of crashes
was 690? If nobody is riding on the cycletracks, then sure, they're safe.
I do wish we could get before/after studies of roads where cycletracks
exist. I would like to see before/after counts of cyclists and
motorists, and before/after crashes (both car/bike and bike/other). How
difficult would that be on cycletracks in NYC or Washington DC? I've
commented on this forum that I don't understand why FHWA experimental
studies are not required for cycletracks. If they were required, we
should have plenty of quality data to evaluate the safety of
cycletracks. Does anyone know of such studies?
Tricia
On 11/1/2012 5:39 PM, John Forester wrote:
Thank you for writing a review of the Teschke study. I read both the
study and your review and I agree with your analysis of the study. I do
have one question on a statement in your review:
Significant Characteristics
However, the presence of bike lanes showed only a small reduction in
crash probability as shown by odds ratio.
Teschke Study section titled Discussion
Bike lanes on major streets with no parked cars and off-street bike
paths had nearly half the risk of the reference.
This does seem significant to me.
But with you, I question the method of the study. They randomly selected
another portion of the cyclist's route where he/she didn't crash and
state that type of roadway is safe? I also question the sample size.
There were only 2 crashes in the cycletrack and only 10 incidents where
a cycletrack is a control site (no crash). Is that a large enough sample
size to be statistically significant when the total number of crashes
was 690? If nobody is riding on the cycletracks, then sure, they're safe.
I do wish we could get before/after studies of roads where cycletracks
exist. I would like to see before/after counts of cyclists and
motorists, and before/after crashes (both car/bike and bike/other). How
difficult would that be on cycletracks in NYC or Washington DC? I've
commented on this forum that I don't understand why FHWA experimental
studies are not required for cycletracks. If they were required, we
should have plenty of quality data to evaluate the safety of
cycletracks. Does anyone know of such studies?
Tricia
On 11/1/2012 5:39 PM, John Forester wrote:
John Forester
Nov 3, 2012, 2:00:04 PM11/3/12
to bicycle...@googlegroups.com
Tricia has been misled by the wording of the paper. The reference type
is major street with parked cars and no special bicycle facility. Tricia
quotes the paper: "Bike lanes on major streets with no parked cars and
the infrastructure differences, bike lanes or bike paths. However, far
more of the difference between the two major street categories (major
street with parked cars and no special bicycle facility and the major
street with no parked cars and bike lanes) was produced by the absence
of parked cars than was produced by the bike lane stripe. As shown by
the plotted diagram. When you are dealing with ideologues this is the
kind of misrepresentation that you have to watch out for.
The comparison between the crash site and the control site did not imply
that the control site was safe. It is a matter of probabilities. On one
roll of the dice, that one trip, there was a crash at one site but no
crash at the other. There is no way to say that the crash site was
unsafe, or the control site was safe, because on a different trip,
another roll of the dice, the results could be different. Therefore, the
data from one trip is insufficient to produce probabilities or odds
ratios (which are slightly different). Only by using the data from many
rolls of the dice, many different trips, can such results be produced,
and that is what the investigators attempted to do. So far as I know,
the statistical processing by the referenced statistical package gets
done correctly.
The errors that I note in the study are not those of data collection and
statistical processing, which may be (I don't know) considered quite
proper in the field of public health. Hence the reason that the editors
and referees of the AJPH accepted the paper. However, the investigation
was not one of public health, but one of traffic engineering, in
particular the sub-specialty called bicycle traffic engineering, about
which the authors, the editors, and the referees had no expert knowledge
but quite a lot of ideology. Hence the errors that they made.
I don't know whether the two crashes recorded for cycle track usage can
produce a significant result. I rather doubt it, although I don't know
the statistical processing done. But I was not about to claim that that
result was not significant on a statistical level, not when other
criticism was quite valid.
is major street with parked cars and no special bicycle facility. Tricia
quotes the paper: "Bike lanes on major streets with no parked cars and
off-street bike paths had nearly half the risk of the reference." This
wording was calculated, consciously or not, to emphasize the effect of
the infrastructure differences, bike lanes or bike paths. However, far
more of the difference between the two major street categories (major
street with parked cars and no special bicycle facility and the major
street with no parked cars and bike lanes) was produced by the absence
of parked cars than was produced by the bike lane stripe. As shown by
the plotted diagram. When you are dealing with ideologues this is the
kind of misrepresentation that you have to watch out for.
The comparison between the crash site and the control site did not imply
that the control site was safe. It is a matter of probabilities. On one
roll of the dice, that one trip, there was a crash at one site but no
crash at the other. There is no way to say that the crash site was
unsafe, or the control site was safe, because on a different trip,
another roll of the dice, the results could be different. Therefore, the
data from one trip is insufficient to produce probabilities or odds
ratios (which are slightly different). Only by using the data from many
rolls of the dice, many different trips, can such results be produced,
and that is what the investigators attempted to do. So far as I know,
the statistical processing by the referenced statistical package gets
done correctly.
The errors that I note in the study are not those of data collection and
statistical processing, which may be (I don't know) considered quite
proper in the field of public health. Hence the reason that the editors
and referees of the AJPH accepted the paper. However, the investigation
was not one of public health, but one of traffic engineering, in
particular the sub-specialty called bicycle traffic engineering, about
which the authors, the editors, and the referees had no expert knowledge
but quite a lot of ideology. Hence the errors that they made.
I don't know whether the two crashes recorded for cycle track usage can
produce a significant result. I rather doubt it, although I don't know
the statistical processing done. But I was not about to claim that that
result was not significant on a statistical level, not when other
criticism was quite valid.
Tricia Kovacs
Nov 3, 2012, 5:40:56 PM11/3/12
to bicycle...@googlegroups.com
John,
I see what you mean. Their data showed little difference between
major street, no parking, bike lane
and
major street, no parking, no infrastructure
Thanks for the explanation.
After reading the paper, I started to think about all the "major" crashes I've had or friends have had. Here they are listed in chronological order (the earlier ones were as a kid):
Tricia
I see what you mean. Their data showed little difference between
major street, no parking, bike lane
and
major street, no parking, no infrastructure
Thanks for the explanation.
After reading the paper, I started to think about all the "major" crashes I've had or friends have had. Here they are listed in chronological order (the earlier ones were as a kid):
- my shoe laces untied getting wrapped around the pedal
- carrying a purse over my shoulder which fell down in my front
wheel (really stupid)
- friends riding a tandem downhill with faulty brakes
- friend getting doored by a parked car
- falling the first time I used clipless pedals
- overlapping wheels with a cyclist in front of me who slowed down when a dog ran in front of her
- overlapping wheels when a group slowed down (for unknown reason) and I was too busy socializing
- husband hitting a pothole and doing an endo
- wiping out on a tandem when we slipped on a wet edge line stripe
(this was while my husband was still recovering from surgery from the
pothole crash and I felt terrible)
- husband getting T-boned by a distracted driver (husband was visible (bright clothes, proper position) and thought he made eye contact)
- hitting a patch of mud on the downhill of a bike trail
Tricia
Reply all
Reply to author
Forward
0 new messages