And again

David Martin <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

On 1/6/04 10:22 pm, in article
[email hidden], "James Annan"

Quoted message said:

No, the very basic error I was thinking of is the one they make in the
"60%" calculation. The ratio of the two numbers they calculate is
meaningless.

The drop in head injuries rate according to their stats is 3.6
percentage points. This is not a 3.6% drop in head injuries but a 13%
change (3.6/27.9)! However, the number of unhelmeted cyclists only
dropped by 5.8/84 = 7%. So according to their stats, the extra helmet
wearers did not only prevent their own head injuries but also stopped
injuries in a large number of those who were not wearing helmets!

I suppose you missed the point that the confounding factors are assumed to
apply to pedestrians as well and that pedestrian injuries typically act as a
good model for cyclist injuries. So comparing the change in cyclist HI with
the change in pedestrian HI, one presumes that the difference in change is
due to cyclists increased wearing of helmets.

No, I didn't miss that, which is why I used their figure for the
_extra_ gains made by cyclists, over and above that made by
pedestrians. Incidentally, none of the figures in the paper seem to
square with the data points or the regression line in the figure, but
I wasn't particularly addressing that initial part of their
calculation at all, even though it is also wrong in detail[1].

Quoted message said:

Quoted message said:

If that is not clear enough, consider a simple hypothetical: head
injury rates drop from 20% to 0% while helmet use increases from 0% to
100%. It is clear that helmet effficiency is 100% (all injuries are
prevented), and not 20/100=20% as Cook and Sheikh's method would
indicate.

No. You cannot take the extreme.

Sheesh. I just dropped in a simple numerical example to show the
bogosity of their calculation.

Quoted message said:

Assume cyclist HI drop from 50 to 25 percentage points. A 50% drop

You can stop right there, since according to the method in the paper
it is clear that they would refer to this as a 25% drop (50 - 25, not
the correct 25/50). This is the error which I am referring to. Their
3.6% drop in %HI (after accounting for pedestrian gains) is a number
of percentage points, not a % fall in injury rate. Got it yet? When
calculated as a % change in injury rate, it is double the % drop in
the proportion of unhelmeted cyclists. As soon as one realises this
point, it is clear that there are other factors present which exceed
any possible benefit of helmets, and that attributing any part of this
large overall injury reduction to helmets is basically guesswork.

I see you need to work through all the numbers rather than
understanding the essence of the mistake directly. Biologists!

[snip lots of tedious calculation]

Quoted message said:

the answer is so obviously bogus that it is
hard to credit the study.

So finally you agree with me. I'm happy for you. Do you understand
_why_ it is bogus, and how their 60% figure is wrong, or do I need to
go through it one more time?

Quoted message said:

Quoted message said:

I guess I get the prize...of writing to the journal and the authors to
explain their mistake.

Don't do it, you'll be laughed out of court unless you can come up with a
much better interpretation of the data than you already have. Mine would
need to be thoroughly worked over before letting anyone near it
professionally.

Don't be such a pompous ass, their formula for the 60% calculation is
obviously bogus, and the correct conclusion from that data, which I
have already explained, is that there _must_ be substantial
confounding factors that they have not taken into account since the
overall reduction in head injury rates is twice what would be provided
by 100% effective helmets. There is no better interpretation of this
data than to point out that it is wholly inadequate for the purpose.

James

[1] Note that there is also essentially the same error in the way they
calculate an "excess" 3.6 percentage points gain for cyclists, since
the drop in ped and cyclist %HI rates should also both be normalised
by their initial values before subtracting one from the other.
However, the effect of this error is probably small, since the ped and
bike values are quite similar, and it could perhaps be defended as an
insignificant approximation. I'm only addressing the error where they
calculate the efficiency of helmets as 3.6/5.8 = 60% which is a rather
elementary mistake to make for those who presume to write statistics
textbooks.

--
If I have seen further than others, it is
by treading on the toes of giants.
http://www.ne.jp/asahi/julesandjames/home/

Quoted message said:

Quoted message said:

AC wrote:
What I don't feel is likely is that Prof Sheikh simply fell into the
thrall of the BHIT and is uncritically regurgitating it's rhetoric -
aided and abbetted by a dodgy understanding of statistics.

Quoted message said:

Not quite sure if I'm reading you right, are you suggesting that Prof
Sheikh has a dodgy understanding of the stats? If so, how come he has
just managed to publish a textbook about the analysis of statistics for
the health industry?

No - I'm saying the opposite.

Earlier you wrote:

"If such a highly qualified person can fall for the bent stats and
outright lies told by BHIT, it isn't surprising that so many other people
do. Mind you, I'm not too sure how much faith I'd put in a book by an
author who has so completely failed to analyse the evidence."

Quoted message said:

And having now read the article, see thread "and yet again", Prof Sheikh
seems firmly in the "don't lets look at evidence that doesn't support our
preconceived ideas" camp.

I don't come to that conclusion from reading that paper. What I take away
from it is a clear framework for thinking about the ethical issues
involved (useful whatever your stance), an impression that many of the
arguments are contentious, and some starting points to investigate them
further (there are links to papers that do not support his views - they
might not be the ones that you would have chosen). He openly takes a
stance, but I don't feel bullied into accepting it.

Quoted message said:

He doesn't even mention the whole population, long term evidence, why
not?

I'm not sure what evidence you think he should have included. Remember
that this is an article primarily about ethics - the guy will have strict
word limits and his article would have been hard to follow if he hadn't
kept reasonably close to the point.

If you want to read something that should be comprehensive look for a
review paper.

AC

Peter Amey said:

Peter Clinch said:

anonymous coward said:

Professor Sheikh has been researching bicycle helmets for years - it's
more likely that BHIT listen to what he has to say, than the other way
round.

Then he really ought to manage better than "The consensus is that they
reduce anything up to 88% of serious head injuries if they are used".

I fear (wearing my pedant's hat rather than my cycling helmet) that the
statement is absolutely correct as it stands. Because "anything up to"
includes 0 (no effect) and also negative numbers (icnreases risk). All
the statement does is place an upper bound on the potential change in
hazard.

I withdraw my comment above and trade will send in my pedant's hat for a
full service. I was misled by the "anything up to" when the really
innacurate bit is "The consensus is...". Starting the sentence with
those words is an attempt to make axiomatic that there is a consenus.
The rest is just sleight of hand intended to imply that the (non
existent) consensus tends toward the 88% of the scale.

One of the most important aspects of the whole H debate is that /there
is no consensus/.

Peter

On Tue, 01 Jun 2004 22:41:35 +0100,

David Martin said:

First let me admit my professional background. I am a research scientist

Thanks for your feedback.

Quoted message said:

Quoted message said:

When the
decrease in head injury rate for cyclists and peds were compared were
they normalised to account for risk exposure (size, make up of population,
motor traffic behaviour, road conditions and total miles travelled for
both years).

Does it need to be? It was as a proportion of total injuries resulting in
admission. This automatically corrects for distance travelled/exposure, size
of population etc.

I think it does. Measuring change in hospitalisation numbers over time and
then attributing them to one putative injury-prevention device without looking
at other risk factors or the make-up and actions of the groups (and other
parties) doesn't seem very rigorous.

Also the helmet wearing number was stated as 22% (which comes from Bryan-Brown
and Christie's survey in 1999). This is only correct for busy sites, the
same survey showed helmet-wearing at 8.2% on quieter roads.

Quoted message said:

Quoted message said:

Of course his paper may answer some of these points but I would be
surprised if one could design an unbiased study that could
supply figures that would strongly support helmet compulsion.

It is a very short paper, and a very simplistic analysis that with a bit of
care could probably have been done better.

From what I have seen the number of papers pro and anti helmet compulsion
seem pretty equally split. Some of them arrive at diametrically opposite
conclusions from the same base data.

--
Andy Leighton => [email hidden]
"The Lord is my shepherd, but we still lost the sheep dog trials"
- Robert Rankin, _They Came And Ate Us_

On 2/6/04 3:22 am, in article
[email hidden], "James Annan"

Quoted message said:

No, I didn't miss that, which is why I used their figure for the
_extra_ gains made by cyclists, over and above that made by
pedestrians. Incidentally, none of the figures in the paper seem to
square with the data points or the regression line in the figure, but
I wasn't particularly addressing that initial part of their
calculation at all, even though it is also wrong in detail[1].

I misread your original post. My apologies. It was late. I was tired. As
they say, don't drink and derive.

I see we do actually agree. The numbers quoted are completely bogus.

What would have helped their analysis is using the data split into regions.
They obviously have some data of this sort (looking at the regression plot).

Quoted message said:

Quoted message said:

Quoted message said:

If that is not clear enough, consider a simple hypothetical: head
injury rates drop from 20% to 0% while helmet use increases from 0% to
100%. It is clear that helmet effficiency is 100% (all injuries are
prevented), and not 20/100=20% as Cook and Sheikh's method would
indicate.

No. You cannot take the extreme.

Sheesh. I just dropped in a simple numerical example to show the
bogosity of their calculation.

Misread the point. Sorry. Now I see what point you are making I am entirely
in agreement.

Quoted message said:

Quoted message said:

Assume cyclist HI drop from 50 to 25 percentage points. A 50% drop

You can stop right there, since according to the method in the paper
it is clear that they would refer to this as a 25% drop (50 - 25, not
the correct 25/50). This is the error which I am referring to. Their
3.6% drop in %HI (after accounting for pedestrian gains) is a number
of percentage points, not a % fall in injury rate. Got it yet? When
calculated as a % change in injury rate, it is double the % drop in
the proportion of unhelmeted cyclists. As soon as one realises this
point, it is clear that there are other factors present which exceed
any possible benefit of helmets, and that attributing any part of this
large overall injury reduction to helmets is basically guesswork.

I see you need to work through all the numbers rather than
understanding the essence of the mistake directly. Biologists!

I missed your original point on the underlying mistake (it was a poorly
presented paper, hard enough to work out what they had done, let alone why
it was wrong.) BTW I put numerical examples in for the benefit of others to
explain why the original data is unreliable.

Quoted message said:

[snip lots of tedious calculation]

Quoted message said:

Quoted message said:

the answer is so obviously bogus that it is
hard to credit the study.

Quoted message said:

So finally you agree with me. I'm happy for you. Do you understand
_why_ it is bogus, and how their 60% figure is wrong, or do I need to
go through it one more time?

I didn't actually see where they had got their 60% figure from. (reading
things late at night after a few beers is not a good idea.)

Quoted message said:

Quoted message said:

Quoted message said:

I guess I get the prize...of writing to the journal and the authors to
explain their mistake.

Quoted message said:

Quoted message said:

Don't do it, you'll be laughed out of court unless you can come up with a
much better interpretation of the data than you already have. Mine would
need to be thoroughly worked over before letting anyone near it
professionally.

Don't be such a pompous ass,

</pompous ass>

Quoted message said:

their formula for the 60% calculation is
obviously bogus, and the correct conclusion from that data, which I
have already explained, is that there _must_ be substantial
confounding factors that they have not taken into account since the
overall reduction in head injury rates is twice what would be provided
by 100% effective helmets. There is no better interpretation of this
data than to point out that it is wholly inadequate for the purpose.

Agreed 100%. Is it worth writing now or is it too late? Would it raise the
issue in the wrong way or actually be beneficial?

Maybe it would be worth contacting the authors directly to find out what
their original data was and their justification for arriving at the figure.

Apologies for the misunderstanding. Notwithstanding the disagreement, I
think we do have plenty of ammunition for a critical letter if it would be
appropriate.

...d

Quoted message said:

[1] Note that there is also essentially the same error in the way they
calculate an "excess" 3.6 percentage points gain for cyclists, since
the drop in ped and cyclist %HI rates should also both be normalised
by their initial values before subtracting one from the other.
However, the effect of this error is probably small, since the ped and
bike values are quite similar, and it could perhaps be defended as an
insignificant approximation. I'm only addressing the error where they
calculate the efficiency of helmets as 3.6/5.8 = 60% which is a rather
elementary mistake to make for those who presume to write statistics
textbooks.

in message <[email hidden]>, James

Annan (') said:

If that is not clear enough, consider a simple hypothetical: head
injury rates drop from 20% to 0% while helmet use increases from 0% to
100%. It is clear that helmet effficiency is 100% (all injuries are
prevented), and not 20/100=20% as Cook and Sheikh's method would
indicate.

Please...

I know you don't mean this, but even so best not to say it. If helmet
use rose to 100% and injuries dropped to 0% then it would be clear that
something was preventing injuries. But even then, absent a very careful
review of other possibilities, it would not be evidence of the efficacy
of helmets.

--
[email hidden] (Simon Brooke) http://www.jasmine.org.uk/~simon/
/-\ You have discovered a security flaw in a Microsoft product. You
|-| can report this issue to our security tesm. Would you like to
| | * Be completely ignored (default)?
| | * Receive a form email full of platitudes about how much we care?
\_/ * Spend hours helping us fix this problem for free?

in message <[email hidden]>, Andy Leighton

(') said:

From what I have seen the number of papers pro and anti helmet
compulsion
seem pretty equally split. Some of them arrive at diametrically
opposite conclusions from the same base data.

Isn't this in itself evidence that there really isn't any good data, and
that people in the field are just arguing from positions of prejudice?

--
[email hidden] (Simon Brooke) http://www.jasmine.org.uk/~simon/

;; not so much a refugee from reality, more a bogus
;; asylum seeker

David Martin said:

Agreed 100%. Is it worth writing now or is it too late? Would it raise the
issue in the wrong way or actually be beneficial?

Since you seem to know what you are on about, it would be worth writing a
proper letter to the editor of the journal pointing out the problems with the
paper. Such letters often get published alongside the on-line paper. Would
also be worth giving a copy to the CTC and cyclehelmets.org so it can go on
their website.

Tony

Simon Brooke said:

in message <[email hidden]>, Andy Leighton

(') said:

From what I have seen the number of papers pro and anti helmet
compulsion
seem pretty equally split. Some of them arrive at diametrically
opposite conclusions from the same base data.

Isn't this in itself evidence that there really isn't any good data, and
that people in the field are just arguing from positions of prejudice?

Absolutely - I also said in a previous post that it is extremely difficult
to gain good data due to the difficulties of designing an adequate study.

--
Andy Leighton => [email hidden]
"The Lord is my shepherd, but we still lost the sheep dog trials"
- Robert Rankin, _They Came And Ate Us_

Looking at the PDF of the IP online paper, there is a wonderful
discontinuity in the data when the version 10 injury codes began to be used.
The regression is carried out across this regardless.

Interesting. I may use this paper as an example of how not to do it.

...d

On Wed, 02 Jun 2004 07:00:08 +0100, anonymous coward
<[email hidden]> wrote (more or less):
....

Quoted message said:

I don't come to that conclusion from reading that paper. What I take away
from it is a clear framework for thinking about the ethical issues
involved (useful whatever your stance),

It's even worse then. It's a paper on /ethics/, which generates no
new data.

This means that the 88% figure the Prof. claims is part of the
consensus can /only/ be the 88% figure which was repudiated even by
its original researchers!

This surely /cannot/ be ethical.

--
Cheers,
Euan
Gawnsoft: http://www.gawnsoft.co.sr
Symbian/Epoc wiki: http://html.dnsalias.net:1122
Smalltalk links (harvested from comp.lang.smalltalk) http://html.dnsalias.net/gawnsoft/smalltalk

in message <[email hidden]>, Tony Raven

(') said:

David Martin said:

Agreed 100%. Is it worth writing now or is it too late? Would it
raise the issue in the wrong way or actually be beneficial?

Since you seem to know what you are on about, it would be worth
writing a proper letter to the editor of the journal pointing out the
problems with the
paper. Such letters often get published alongside the on-line paper.
Would also be worth giving a copy to the CTC and cyclehelmets.org so
it can go on their website.

Wot 'e said.

Bad arguments ought to be countered. I don't mean that in a partisan way
- bad arguments against helmets use should be countered just as much as
bad arguments for. This actually is quite important because lives
really are at stake. We really do need to know whether helmets can be
beneficial and if so in what circumstances they may be beneficial - and
in what circumstances they may be harmful. We don't really know this
because we don't have good data but in the meantime this sort of
water-muddying nonsense does not help.

--
[email hidden] (Simon Brooke) http://www.jasmine.org.uk/~simon/
Just as defying the law of gravity through building aircraft requires
careful design and a lot of effort, so too does defying laws of
economics. It seems to be a deeply ingrained aspect of humanity to
forever strive to improve things, so unquestioning acceptance of a
free market system seems to me to be unnatural. ;; Charles Bryant

Simon Brooke said:

Wot 'e said.

Bad arguments ought to be countered. I don't mean that in a partisan way
- bad arguments against helmets use should be countered just as much as
bad arguments for. This actually is quite important because lives
really are at stake. We really do need to know whether helmets can be
beneficial and if so in what circumstances they may be beneficial - and
in what circumstances they may be harmful. We don't really know this
because we don't have good data but in the meantime this sort of
water-muddying nonsense does not help.

As an example here are the comments the BMJ published last time Cook and
Sheikh wrote an article:

http://bmj.bmjjournals.com/cgi/content/full/322/7293/1063

Tony

Tumbleweed said:

"Tony Raven" <[email hidden]> wrote in message

[quoting someone else]

Quoted message said:

Quoted message said:

Bicycle helmet laws in other countries, such as Australia and New Zealand,
have reduced cycling-related head injuries and fatalities with good
uptake, they say.

Is that true or not?

Yes, in much the same way that banning pedestrians from all roads
would reduce pedestrian injuries and fatalities. Perhaps we should do
that, too.

Also, banning marriage would reduce the number of people murdered by
their spouse, which must be a good thing.

regards, Ian SMith
--
|\ /| no .sig
|o o|
|/ \|

Gawnsoft said:

On Wed, 02 Jun 2004 07:00:08 +0100, anonymous coward
<[email hidden]> wrote (more or less):
...

Quoted message said:

I don't come to that conclusion from reading that paper. What I take away
from it is a clear framework for thinking about the ethical issues
involved (useful whatever your stance),

It's even worse then. It's a paper on /ethics/, which generates no
new data.

Is the study of ethics such a bad thing?

Quoted message said:

This means that the 88% figure the Prof. claims is part of the
consensus can /only/ be the 88% figure which was repudiated even by
its original researchers!

I was referring to the paper referenced in the thread 'and yet again'
which does reference RTR but never includes any numbers - I think you may
be referring to the BBC article?

The 'and yet again' article does reference RTR but it also references
an article debunking RTR. I concede it references more articles supporting
helmet effectiveness than it does articles against...

Patrick Herring said:

I think this is BHIT et al running out their big guns in the run-up to
the 18th of June re-reading.

They certainly are. There was the BBC article, and an article in Metro
this morning.

Something I note is the use of statistics. People see things like "33%
[give or take a bit] of cycle injuries serious enough for hospital are
head injuries" and seem to conclude that cycling is really dangerous,
and about a third of cyclists will hurt their heads. Its like the people I
once saw in a supermarket choosing an "80% fat free" product over another
because "Its fat free". People seem mystically awestruck by percent signs.

I had a look at a pro-helmet web site today, and its catchline was
something like "There are 5000 chances to fall off every mile". They
assumed that you will crash. Their whole ethos was that cycling was
dangerous. Their reccomendations for child carriers were quite extreme,
and went as far as not recomending taking a small child on a bike at
all. I've seen a couple of friend's small children (roughly 2 years old)
quite enjoying being taken for rides.

- Richard

--
_/_/_/ _/_/_/ _/_/_/ Richard dot Corfield at ntlworld dot com
_/ _/ _/ _/
_/_/ _/ _/ Time is a one way street,
_/ _/ _/_/ _/_/_/ Except in the Twilight Zone.

"Tony Raven" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Sigh!

Tony

http://news.bbc.co.uk/2/hi/health/3756385.stm

Call for mandatory cycle helmets

Whilst not normally being a conspiracy theorist, I am beginning to wonder if
there isn't something in the possibility that a lot of eminent people who
have promoted helmets are going to look very silly indeed if the whole
population, long term research about cycle helmets is ever accepted.

Just why would so many people ignore the most reliable evidence, and quote
evidence from sources known to be inaccurate?

Only a theory you understand, but why else?
--
cheers

Rich

"burt" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

Whilst not normally being a conspiracy theorist, I am beginning to wonder if
there isn't something in the possibility that a lot of eminent people who
have promoted helmets are going to look very silly indeed if the whole
population, long term research about cycle helmets is ever accepted.

Just why would so many people ignore the most reliable evidence, and quote
evidence from sources known to be inaccurate?

Only a theory you understand, but why else?

It's a combination of mental inertia and the difficulty of overcoming
"common sense".

Lots of people took some time coming round to my idea that the disk
brake/qr design is faulty, it was not that the evidence grew slowly
over time but that their minds eventually warmed up to the facts.

I recently saw an ogden nash quote to the effect that the doors to a
bigot's mind opens outwards, so the pressure of facts only keeps it
more firmly closed. IMO substitute "human" for "bigot" and it would be
about right. It is very rare to find someone who can form anything
approaching a dispassionate assessment of the evidence for any
nontrivial and emotive issue, especially when that entails abandoning
a long-held belief. And I'm only talking about people who are
basically intelligent in the first place, those who wibble on with
"but it could save a child's life..." (or equivalently in the disk
brake debate cannot understand a simple force diagram) aren't even
worth starting on.

James

James Annan said:

It's a combination of mental inertia and the difficulty of overcoming
"common sense".

"All truth goes through three steps: First, it is ridiculed. Second, it is
violently opposed. Finally, it is accepted as self-evident."
Arthur Schopenhauer

Tony

Tony Raven said:

James Annan said:

It's a combination of mental inertia and the difficulty of overcoming
"common sense".

"All truth goes through three steps: First, it is ridiculed. Second, it is
violently opposed. Finally, it is accepted as self-evident."
Arthur Schopenhauer

One shouldn't worry about being ridiculed, but it obviously doesn't follow
that everything that is ridiculed is truth