D2 said:Of course, I'm sure he's seen the data and I'm equally sure he'll find
a way to refute the data. Which goes to my point: statistics are not
static, they are dynamic and subjective as hell and thus worthless to
'prove' anything.
No, you are missing the point. What you cite is a secondary source,
Frank and I have looked at the original paper from which those figures
are taken. We have also looked at the original papers from which the
claims of no benefit are taken.
On the one hand you have small-scale studies which say that there is a
difference in head injury rate between helmeted and unhelmeted
cyclists. On the other hand you have figures from whole populations
which say that there is no discernible change in recorded head injury
rates as helmet use rises.
Both are right.
What is wrong is to claim that the small-scale studies predict that
getting non-wearers to wear helmets will reduce their incidence of
head injury to the same as the voluntary wearers. And the studies
themselves steer clear of making such a claim.
It is helmet promoters who claim that if non-wearers wear helmets,
they will exhibit the injury profile of wearers. They say that the
small-scale studies are predictive - but they are wrong, and it's been
proven that they are wrong because several places have achieved
increases of 40 percentage points or more in helmet use over a very
short period through legislation and *not one* of those places shows
any reduction in the proportion of cyclists' injuries which are head
injuries as a result.
So the science is fine (up to a point), it's the representation of the
science by people with an agenda which is a problem.
Now you may say that Frank and I have an agenda, and you'd be right,
but you have to remember that both of us used to be enthusiastic
advocates of helmets and have changed our view radically in response
to the data we've seen. Not second-hand reports of data, but source
data from the studies themselves.
But I absolutely encourage you to make up your own mind. Here's a
reading list for you:
http://www.chapmancentral.co.uk/Web/public.nsf/Documents/Research_Evidence
Quoted message said:Frank cites an impressive amount of information to support his cause,
but why is his data any more valid than the stuff I found? It took me
very little time to dig out the stats I posted and there was a lot
more. Frank invited me to do some research, I did and found plenty of
information that contradicts his anti-helmet position. I haven't read
Frank's reply yet but I'm sure he'll counter with an intelligent
message heavy with even more numbers.
Frank's data is more valid than what you found because what you found
is not data, it's someone's use of data to promote an agenda. From
your perspective there's no difference, because to you both are third
parties, but that just means you need to do what Frank did and go back
to the source data.
There are a few reasons the 1989 Seattle study, in particular, gets my
goat (and Frank's too).
First, it was written by a group who were already strongly pro-helmet
when they started, and the preamble makes it clear that they set out
to prove their case rather than to investigate.
Second, it arrives at its conclusions by comparing entirely different
groups of cyclists. If helmets "prevent" 85% of head injuries then
they also "prevent" black skin, low income, riding on roads instead of
bike trails, riding unaccompanied, and being hit by cars. They also
"prevent" 72% of broken legs. All this is taken direct from the
authors' source data.
Third, it does not in any case claim that helmets "prevent" x% of
injuries, as noted above, but it is generally represented as if it
does.
Fourth, it assumes a helmet wearing rate in the general population
which one of the co-authors knew was four or five times above the
actual level, because he had conducted counts on the streets of
Seattle at the same time as the study data was being collected.
Fifth, the figures it comes up with have never, to my knowledge, been
matched or exceeded by any other study of this or any other type.
Given the criticisms above, which have been published in the medical
press, why are these figures still quoted? Answer: because according
to Randy Swart of BHSI the figures are "so ingrained in the injury
prevention [sic] community that a change would not be helpful". Got
that? It's more important to be helpful to those trying to sell
helmets, than to be honest.
Sixth, the terms "head injury" and especially "brain injury" are used
in a dishonest way. The public perception of brain injury is
something which leaves you with permanent intellectual impairment.
These injuries are incredibly rare; what the study means by brain
injury is *concussion* - which, in almost every case, has no lasting
consequences. The use of the term brain injury is extremely unlikely
to be understood by the majority of laymen as meaning concussion, but
that is what it means. There is precious little evidence that any
helmet on sale then or now will mitigate the kind of brain injury
which leads to permanent intellectual impairment, whatever the
handwringers might say.
So there's a few reasons why, in my view, any web page or publication
which uses the figure 85% or 88% should immediately be torn up and
thrown away. Either it is written by someone ignorant of the facts,
or it is written by someone who knows the facts and chooses to ignore
them. And in this case I am using the transitive definition of facts,
in the sense that it is incontrovertibly true that the 1989 Seattle
study is poor science and anybody who takes a genuine interest in the
subject must know it.
Quoted message said:Why should I or anyone else accept his data and conclusions over the
stuff I found? I can print a pile of research and so can Frank and,
taken at face value, both piles will provide plenty of compelling
reading for anyone who wishes to make a decision regarding the use of a
bicycle helmet.
No you can't print a pile of research, you can print a pile of
citations and maybe abstracts but the majority of those papers are
copyright and not available on the web. Frank and I have accumulated
large libraries of helmet papers, we buy them, we email authors for
copies, we get them from libraries. I have a library of about 700
documents, including a couple of hundred published studies.
Do read some of the sources, and look for a couple of things:
See if they give a realistic, objective appraisal of the risk of
cycling compared to other activities. X thousand injuries is big, but
how big compared to, say, driving or walking on city streets?
Objectively, cycling is not very dangerous.
See if they genuinely claim that helmets can prevent the most serious
injuries. In most cases the mean ISS for the "prevented" injuries is
very low and in any case the helmeted community tend to have a lower
mean ISS than the unhelmeted, due to their being the risk-0averse ones
to start with.
See how many injuries involving motor traffic are included. The
Thompsons claim that helmets prevent motor traffic injuries, based in
part on data which includes _no_ helmeted cyclists in collision with
cars! Yet almost all serious and fatal injuries are sustained in
collision with motor vehicles (and it's usually the motor driver's
fault).
See if they name the sources, and if the claims they make are verified
by those sources. I recently found out that here in the UK the helmet
promoters have been using a figure for child fatalities which is an
"estimate based on under reporting" (of child fatalities? in a western
democracy?) and is actually five times the real figure as published by
the Government. They claimed it came from a Parliamentary report, but
they were not being terribly honest about that...
OK, one last point. Here is a paragraph from a helmet standard of
1989 (standards are a lot less stringent now).
"It (the standard) specifies requirements for helmets intended for
use by pedal cyclists on ordinary roads, particularly by young
riders in the 5 years to 14 years age group, but which may also
be suitable for off the road. It is not intended for high-speed
or long distance cycling, or for riders taking part in
competitive events. The level of protection offered is less than
that given by helmets for motor-cycle riders and is intended to
give protection in the kind of accident in which the rider falls
onto the road without other vehicles being involved."
I will send you a $10 certified money order if you can find me a
single manufacturer which claims, on any cycle helmet sold in the USA
or Europe, that it provides meaningful protection in collisions
involving motor vehicles.
Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk
88% of helmet statistics are made up, 65% of them at CHS, Puget Sound
