Cyclist dies after crash during Tour of Colombia

Tony Raven said:

sports.yahoo.comnewsOpen ↗
cyclistdies&prov=ap&type=lgns Cyclist dies after crash
during Tour of Colombia

June 11, 2004 BOGOTA, Colombia (AP) -- A cyclist was killed
Friday when he crashed during the fifth stage of the Tour
of Colombia. Juan Barrero, a 31-year-old Colombian,
suffered serious head injuries after he got tangled up with
other cyclists and fell while negotiating a fast downhill
curve. He had a cardiac arrest and died while being
transported from a small hospital to a larger one, said
Orlando Cardona, director of the San Vicente de Paul
Hospital, located about 100 miles west of Bogota. Two other
cyclists were also injured in the crash, Cardona said.
``The fall was massive and Juan took the brunt of it,''
said his brother, Manuel Barrero, also riding in the tour.

And yes, he was wearing a helmet:

<http://www.dailypeloton.com/displayarticle.asp?pk=6391>

What should have been a day of epic cycling, with the main
contenders battling it out on the first real mountain stage
of the 54th Vuelta a Colombia, from Santa Rosa de Cabral to
Jericó over 171 km. turned a day of sadness and tragedy as
the 31-year-old cyclist Juan Barrero of the Alcaldía de Fusagasugá-
Juegos Nacionales 2004 team lost his life after falling
heavily on a descent, less than nine minutes into the stage.

Riding at a speed of about 80 km., the man got tangled up
with other cyclists and fell while negotiating a fast
downhill curve. He hit a rock and, despite wearing a helmet,
reportedly suffered serious head and thorax injuries, and
later had a cardiac arrest; Barrero died while being
transported from the Hospital de Santa Rosa de Cabal to the
larger "San Jorge de Pereira", located about 100 miles west
of Bogotá, said Hospital director Orlando Cardona.

Two other riders, Barrero's teammate Victor Hugo González,
who fractured his wrist and collarbone, and Óscar Santo
Álvarez, that had some bruises on his left arm and thigh,
were also involved and injured in the pile-up, Cardona said.
González was the only one that briefly commented "We were on
the descent between Santa Rosa and Chinchiná, everything was
going fine, but the road was wet, and all of sudden we found
ourselves on the ground". "The fall was massive and Juan
took the brunt of it," Juan's brother Manuel Barrero, also
riding in the tour, added.

Born in Facatativá on May 9, 1973, Juan Antonio Barrero had
been a professional rider for eight years, most of which in
the ranks of his current team. Barrero took part in several
editions of the Vuelta a Colombia as well as other races
like Clásico RCN, Vuelta a Chiriquí and the Tour of
Costarica, and more local events in his home country. He was
Cundinamarca Provincial Champion both in 1998 and the past
season, and this year notched up a stage victory at Clásica
de Fusagasugá.

From the cyclingnew.com site

'Death in Tour of Colombia

...The fall occurred some five minutes after the start, but
due to the descent the peloton was already descending at
high speed. Other riders said the turn was slick with some
type of liquid on the surface. Barrero suffered injuries to
the head, cervical area, and the thorax area. Ambulance
attention was immediate and all resuscitation measures were
given at the hospital. All three riders were wearing
appropriate helmets.'

Look out for all the 'cycle helmet again fails to save
cyclist in crash' headlines...

On 12 Jun 2004 14:50:54 -0700, [email hidden] (Howard)
wrote (more or less):

Quoted message said:

From the cyclingnew.com site

'Death in Tour of Colombia

...The fall occurred some five minutes after the start, but
due to the descent the peloton was already descending at
high speed. Other riders said the turn was slick with some
type of liquid on the surface. Barrero suffered injuries to
the head, cervical area, and the thorax area. Ambulance
attention was immediate and all resuscitation measures were
given at the hospital. All three riders were wearing
appropriate helmets.'

I doubt /any/ of them were wearing helmets appropriate
for cycling at 80Km/h down steep roads covered in a
slippy fluid.

They were probably wearing bicycle helmets, which are
appropriate for up to about 12mph (about 19Km/h)

An appropriate helmet would have been a heavy-duty motorbike
helmet. then again, that'd just have resulted in a spate of
'cyclists die from heat prostration' stories.

Quoted message said:

Look out for all the 'cycle helmet again fails to save
cyclist in crash' headlines...

--
Cheers, Euan Gawnsoft: gawnsoft.co.srgawnsoft.co.srOpen ↗
Symbian/Epoc wiki: html.dnsalias.nethtml.dnsalias.netOpen ↗ Smalltalk
links (harvested from comp.lang.smalltalk)
html.dnsalias.netsmalltalkOpen ↗

Quoted message said:

They were probably wearing bicycle helmets, which are
appropriate for up to about 12mph (about 19Km/h)

I'm not quite sure where this 12mph figure comes from? In
the Snell standards (smf.orgstds.htmlOpen ↗) helmets
are tested by simulating a 2.2 metre fall onto a flat anvil
or 1.3 metres onto other surfaces. It's a long time since I
did any physics so this may be incorrect:

S=0.5at^2 = 2.2=0.5*10*t^2 t = sqrt(2.2/(0.5*10)) V =
at = 6.6 m/s

which corresponds to about 23kmh ~= 15mph which is in the
right ball park?

also a 1.3m drop ~= 11.4mph impact

I'm left wondering how well this simulates being thrown over
the handlebars, where you have your whole body weight behind
your head - the tests just use a 5kg dummy head.

If you go ice skating and aren't better than me, you'll find
that it doesn't hurt much more if you fall over at speed
compared to if you fall over whilst standing still - unless
you slide into something.

On the same principle, presumably a 12mph helmet might give
some useful protection if you fall off at high speed but
scrape to a halt on the road without hitting anything else.
Or else fall off, scrape almost-but-not-quite to a halt,
then hit something?

Quoted message said:

An appropriate helmet would have been a heavy-duty
motorbike helmet.

Interestingly, motorcycle helmet drop distances are 3m and
2.2m respectively - not as large a difference as I expected.
Though I guess there's more to their differences than impact
absorbtion

Quoted message said:

then again, that'd just have resulted in a spate of
'cyclists die from heat prostration' stories.

On a related topic, I wonder how big a drop in cycling we
would see if helmets were made compulsory in the uk? I can
see how wearing a cycle helmet might be onerous in
Australia because it's hot. I imagine that's less of a
problem here (so far).

Incidentally (to state my position), I currently wear a
helmet about 50% of the time. I'm opposed to helmet
compulsion but undecided whether to wear one myself. I am
interested in the theory that helmets increase torsional
injuries, and I accept they're far from the panacea they're
made out to be.

AC

There are a number of differing standards for helmets, the
highest one give a 15 MPH rating, the others 12 MPH or so.
In reality many helmets when independenty tested fail to
even meet these standards and when incorrectly worn quite
possibly no protection at all.

The main point about the impact absorbtion abilities of
helmets is that the sort of impacts they can absorb, be they
sustained in a low or a high speed fall, are unlikely to be
life threatening in any case. In an impact high enough to be
life threatening a helmet has such a low ability to absorb
impacts it will make little or no difference and may well
just fail catastrophically.

As the energy that has to be dissipated in a crash rises
with the square of the speed a rough estimate of the forces
involved can be calculated as follows:

At 12 MPH the force would be 12 squared x mass. Lets give
the mass, a standard value of 1 so we can compare the
effects of speed. This gives a value of 144 at 12 MPH.

At 42 mph the force would be 42 squared x 1 = 1764.
Substracting the tested ability of a helmet to absorb energy
from this this leaves 1620, which still equates to 40.2 MPH,
still very likely to result in a fatality.

When it comes to 'road safety' helmets are a serious
distraction away from the real issues: reducing the
number of collisions that occur and reducing vehicle
speeds to a low enough level that colisions are
survivable when they do occur.

For one I would accept the compulsory wearing of helmets to
give protection from minor injuries, and the odd rare
serious injury, at speeds below 15 MPH if vehicle speeds
were restricted to 15 MPH to prevent all those serious
injuries that occur above 15 MPH. (Given current police
guidelines 15 MPH would be enforced at almost 20 MPH but
with ISA systems that guarantee motor vehicles could not
exceed 20 MPh I might support a universal 20 MPH limit in
towns and on country lanes).

anonymous coward said:

I'm not quite sure where this 12mph figure comes from? In
the Snell standards (smf.orgstds.htmlOpen ↗) helmets
are tested by simulating a 2.2 metre fall onto a flat
anvil or 1.3 metres onto other surfaces.

Pop into your LBS some time and try to find a Snell
certified helmet. The helmet makers pushed through a much
lower set of standards to allow them to produce something
which was (a) cheaper to make for the same sale price and
(b) more acceptable to users who don't like having
their heads boiled when the ambient temperature is
above freezing.

--
Guy
===
May contain traces of irony. Contents liable to settle after
posting. chapmancentral.co.ukchapmancentral.co.ukOpen ↗

Victory is ours! Down with Eric the Half A Brain!

Howard said:

There are a number of differing standards for helmets, the
highest one give a 15 MPH rating, the others 12 MPH or so.
In reality many helmets when independenty tested fail to
even meet these standards and when incorrectly worn quite
possibly no protection at all.

The main point about the impact absorbtion abilities of
helmets is that the sort of impacts they can absorb, be
they sustained in a low or a high speed fall, are unlikely
to be life threatening in any case.

I've read a few newspaper articles to the effect that
someone has been punched, fallen and hit their head against
a kerbstone and died. I wonder whether low-speed impacts
generally _are_ innocuous, or whether we're so good at
protecting our noggins that they _seem_ innocuous because we
manage to avoid or mitigate most of them?

Quoted message said:

In an impact high enough to be life threatening a helmet
has such a low ability to absorb impacts it will make
little or no difference and may well just fail
catastrophically.

As the energy that has to be dissipated in a crash rises
with the square of the speed a rough estimate of the
forces involved can be calculated as follows:

At 12 MPH the force would be 12 squared x mass. Lets give
the mass, a standard value of 1 so we can compare the
effects of speed. This gives a value of 144 at 12 MPH.

At 42 mph the force would be 42 squared x 1 = 1764.
Substracting the tested ability of a helmet to absorb
energy from this this leaves 1620, which still equates to
40.2 MPH, still very likely to result in a fatality.

You haven't addressed my main point at all, which is that in
real-life collisions the energy of any individual impact
does not have to be equal to the total energy theoretically
available in a collision occuring at that speed.

For example, imagine a van pulls out of a side junction, and
I hit it sideways on (90 degrees) at 12mph. Let's assume for
the sake of argument that it does what the label says and
can prevent me from suffering brain damage under these
circumstances.

Now imagine that a car pulls over to park on my side of the
road, and I have a head-on collision with it with a closing
speed of 24mph. If the windscreen is at an angle of 30
degrees my head will impact it at only 12mph (24 x sin(30))
so I may still be OK. As you've pointed out, a 24mph
collision has to burn off 4x the energy of a 12mph
collision so I'll still be moving, but if I'm lucky I may
get away with it.

This is all horribly artificial, but I have hit a van at
maybe 8mph without hitting my head at all. So presumably a
helmet might help in scenario 1 even if the impact speed is
a bit higher than 12 mph.

Quoted message said:

When it comes to 'road safety' helmets are a serious
distraction away from the real issues: reducing the number
of collisions that occur and reducing vehicle speeds to a
low enough level that collisions are survivable when they
do occur.

I too would put helmets third. But I genuinely haven't yet
made up my own mind whether I think helmets are a 'might as
well' option or 'useless and possibly dangerous'.

Quoted message said:

For one I would accept the compulsory wearing of helmets
to give protection from minor injuries, and the odd rare
serious injury, at speeds below 15 MPH if vehicle speeds
were restricted to 15 MPH to prevent all those serious
injuries that occur above 15 MPH. (Given current police
guidelines 15 MPH would be enforced at almost 20 MPH but
with ISA systems that guarantee motor vehicles could not
exceed 20 MPh I might support a universal 20 MPH limit in
towns and on country lanes).

Shouldn't your limit be 7.5mph - so that no collision can
have a closing speed of more than 15mph?

AC

anonymous coward said:

I've read a few newspaper articles to the effect that
someone has been punched, fallen and hit their head
against a kerbstone and died. I wonder whether low-speed
impacts generally _are_ innocuous, or whether we're so
good at protecting our noggins that they _seem_ innocuous
because we manage to avoid or mitigate most of them?

Quite probably a bit of both. I'd guess /most/ are innocuous
(but can still be painful) but you can get really unlucky
and catch a sensitive spot at the wrong angle. I've heard of
a case of a death slipping getting out of the bath, when the
victim would presumably not have been in a Big Hurry and
probably fairly with it. Plus it is a natural reaction to
protect one's head.

Quoted message said:

Now imagine that a car pulls over to park on my side of
the road, and I have a head-on collision with it with a
closing speed of 24mph. If the windscreen is at an angle
of 30 degrees my head will impact it at only 12mph (24 x
sin(30))

No, it will impact it at 24 mph. There may well be a
deflection effect so you don't soak up as much energy, but
it'll still be travelling at 24 mph. And the deflection
effect may or may not be a good thing, if you remember all
that stuff about rotational effects etc.

Quoted message said:

This is all horribly artificial

Isn't it just. So that's why whole population studies tend
to be treated as more useful than individual what-ifs. They
don't show any recognisable aid from lids in KSIs.

Quoted message said:

I too would put helmets third. But I genuinely haven't yet
made up my own mind whether I think helmets are a 'might
as well' option or 'useless and possibly dangerous'.

I personally think neither of the above. "Might as well"
assumes there are no downsides, but if you find that a
snugly fitting shell of non-breathable material (vents can
only do so much) with a snug chinstrap has zero comfort
difference then your head is very different to mine. I think
they are potentially useful in reducing the effects of
painful but minor accidents, but they give a level of
discomfort 100% of the time, so it's a case of "you choose,
you lose". I'm not in the habit of falling off when I'm on
the roads, so these days I tend not to bother. I am in the
habit of falling off the MTB if I'm Going For It, and
comfort is a minor issue in those cases anyway, so I will
wear it there.

So I regard them for most of the cycling I do as "not useful
enough to overcome the downside of being relatively
uncomfortable all the time", which isn't the same as either
of your options above.

Pete.
--
Peter Clinch Medical Physics IT Officer Tel 44 1382 660111
ext. 33637 Univ. of Dundee, Ninewells Hospital Fax 44 1382
640177 Dundee DD1 9SY Scotland UK net
[email hidden] dundee.ac.uk~pjclinchOpen ↗

In "]news:[email hidden],
Peter Clinch <[email hidden]> typed:

Quoted message said:

anonymous coward said:

Now imagine that a car pulls over to park on my side of
the road, and I have a head-on collision with it with a
closing speed of 24mph. If the windscreen is at an angle
of 30 degrees my head will impact it at only 12mph (24 x
sin(30))

No, it will impact it at 24 mph. There may well be a
deflection effect so you don't soak up as much energy, but
it'll still be travelling at 24 mph. And the deflection
effect may or may not be a good thing, if you remember all
that stuff about rotational effects etc.

And you'll probably want something to lubricate the
deflection to avoid the rotational effect. If you're lucky,
then you've got a nice head of hair to do it with.

Hopefully, also, your head will be retracting away from the
impact site to mitigate the collision as well.

Like Pete says, these are reasons why it's pretty
complicated to try and do it by imagining what happens in
the case of a collision and instead look at the data.

A

Peter Clinch said:

anonymous coward said:

I've read a few newspaper articles to the effect that
someone has been punched, fallen and hit their head
against a kerbstone and died. I wonder whether low-speed
impacts generally _are_ innocuous, or whether we're so
good at protecting our noggins that they _seem_ innocuous
because we manage to avoid or mitigate most of them?

Quite probably a bit of both. I'd guess /most/ are
innocuous (but can still be painful) but you can get
really unlucky and catch a sensitive spot at the wrong
angle. I've heard of a case of a death slipping getting
out of the bath, when the victim would presumably not have
been in a Big Hurry and probably fairly with it. Plus it
is a natural reaction to protect one's head.

Quoted message said:

Now imagine that a car pulls over to park on my side of
the road, and I have a head-on collision with it with a
closing speed of 24mph. If the windscreen is at an angle
of 30 degrees my head will impact it at only 12mph (24 x
sin(30))

No, it will impact it at 24 mph. There may well be a
deflection effect so you don't soak up as much energy, but
it'll still be travelling at 24 mph.

You make it sound as if it's almost the same thing, but of
course it isn't.

A few people on this list have fallen off on the flat at
speeds of 24mph+ and are here to tell the tale; I'd wager
that none of us have ridden directly into a brick wall at
24mph (unless they flew over the top of it or were riding a
recumbent). Hitting a sloped surface has got to be somewhere
in between.

Quoted message said:

And the deflection effect may or may not be a good
thing, if you remember all that stuff about rotational
effects etc.

Point taken.

Quoted message said:

Quoted message said:

This is all horribly artificial

Isn't it just. So that's why whole population studies tend
to be treated as more useful than individual what-ifs.
They don't show any recognisable aid from lids in KSIs.

I tend to think the population studies are so fraught with
uncertainties and confounding factors that it's difficult to
draw any firm conclusions from them. Do bike helmets help?
Is autism linked to MMR? Do whisky or olive oil drinkers
make you live longer? What about drinking whisky and olive
oil together?

Take the seatbelt paradox - seatbelts may not make the roads
safer, but if I get into a car as a passenger I'm probably
well advised to wear one. It's a while since I read John
Adam's book, but IIRC the population statistics would have
suggested I shouldn't bother.

I seem to be very argumentative today. Maybe some fresh air
will help (lunchtime bike ride),

AC

anonymous coward said:

I tend to think the population studies are so fraught with
uncertainties and confounding factors that it's difficult
to draw any firm conclusions from them.

A criticism which applies to a very much greater degree to
the small-scale prospective studies which suggest benefit.
If you're going to discount the evidence, then you need to
discount /all/ of it :-)

Quoted message said:

Do bike helmets help? Is autism linked to MMR? Do whisky
or olive oil drinkers make you live longer? What about
drinking whisky and olive oil together?

These things show correlations between things. The
population level studies on bike helmets show /no/
correlation - that's the point. You can argue for ever about
whether autosm is caused MMR or just happens to manifest at
around the time the MMR is administered; in the case of
helmets what you have is no discernible effect (see Tony's
"spot the year" charts posted earlier).

Quoted message said:

Take the seatbelt paradox - seatbelts may not make the
roads safer, but if I get into a car as a passenger I'm
probably well advised to wear one. It's a while since I
read John Adam's book, but IIRC the population statistics
would have suggested I shouldn't bother.

Invalid comparison: Adams' population statistics show that
for optimum safety everybody except the driver should wear
one :-) As a passenger you will probably not materially
affect the outcome of the journey.

--
Guy
===
May contain traces of irony. Contents liable to settle after
posting. chapmancentral.co.ukchapmancentral.co.ukOpen ↗

Victory is ours! Down with Eric the Half A Brain!

anonymous coward said:

You make it sound as if it's almost the same thing, but of
course it isn't.

All I'm saying is that the impact velocity is not reduced
according to a simple bit of trigonometry, and I said that
because that is the case.

Quoted message said:

I tend to think the population studies are so fraught with
uncertainties and confounding factors that it's difficult
to draw any firm conclusions from them.

The point about whole population studies is you have a very
big sample base so that fact alone tends to round off a lot
of the awkward corners. In a single case /that/ is where you
are so fraught with uncertainties you can't usefully deduce
much except for very specific cases.

Quoted message said:

Do bike helmets help? Is autism linked to MMR? Do whisky
or olive oil drinkers make you live longer? What about
drinking whisky and olive oil together?

In the first case, very probably not as far as KSIs are
concerned. In the other cases you're not asking such simple
questions and there are plenty of other variables that may
influence your results. For example, olive oil is just one
thing in a diet so you can't really isolate it. For MMR,
there were autistic folk before there was MMR and the causes
weren't well understood to start with so it could be all
sorts of other things. But with helmets we knew what the
rates were before everyone was wearing them and after
everyone was wearing them where compulsion was introduced.
If they have a significant effect you'd expect to see it
when the figures come in, and we don't. There will be other
variables, but nothing like as many, or of such
significance, as the rest of a diet aside from olive oil.

The only way to recreate accidents and their effects
properly for study
is... well, actually you can't do it. There are far more
things you have to account for in a single incident than
you have to account for at a population level, because
the population level accounts for all the accidents
people really had.

Quoted message said:

Take the seatbelt paradox - seatbelts may not make the
roads safer, but if I get into a car as a passenger I'm
probably well advised to wear one. It's a while since I
read John Adam's book, but IIRC the population statistics
would have suggested I shouldn't bother.

If you get into a car as a passenger then you have no
control over the vehicle, so effectively all else is equal.
And if all else is equal then you're better off wearing a
seatbelt. Where the seatbelt safety argument breaks down is
where all things aren't equal: a driver may drive faster and
brake later because they feel safer with their seatbelt on,
for example, but that's irrelevant to a passenger.

Whole population studies here show no effect for the
passenger but OTOH the nullifying effect is mainly produced
by the driver's actions. Stokers aside, on a cycle you're
in charge.

Pete.
--
Peter Clinch Medical Physics IT Officer Tel 44 1382 660111
ext. 33637 Univ. of Dundee, Ninewells Hospital Fax 44 1382
640177 Dundee DD1 9SY Scotland UK net
[email hidden] dundee.ac.uk~pjclinchOpen ↗

Peter Clinch said:

If you get into a car as a passenger then you have no
control over the vehicle, so effectively all else is
equal. And if all else is equal then you're better off
wearing a seatbelt. Where the seatbelt safety argument
breaks down is where all things aren't equal: a driver may
drive faster and brake later because they feel safer with
their seatbelt on, for example, but that's irrelevant to a
passenger.

There are undoubtably situations where both seat belts and
bicycle helmets reduce one's survival chances. Take for
example this accident:

topgear.com03Open ↗

From the pictures, it's astonishing the driver survived. I'd
suggest
that wearing a seatbelt would have given rise to a less
fortunate outcome.

There are hints that bike helmets *may* increase the risk of
rotational injury in certain circumstances, although I'm not
aware of any direct case studies on this.

As you say, the key difference is that we have good,
statistically sound data that wearing a seat belt increases
one's chances of surviving a car journey. We do not have
such data for the wearing of bike helmets, and if there were
a significant advantage, this evidence should be jumping off
the pages at us.

I wonder if the driver of the car linked to above wears a
seatbelt now?

--
Mark.

Mark Tranchant said:

There are undoubtably situations where both seat belts and
bicycle helmets reduce one's survival chances. Take for
example this accident:

topgear.com03Open ↗

From the pictures, it's astonishing the driver survived.
I'd suggest
that wearing a seatbelt would have given rise to a less
fortunate outcome.

This is indeed the case, but since you can't predict the
sort of accident you'll have before it's too late, knowing
that such cases exist isn't of much use. You've got to go
with the overall probabilities every time for the best
chance, though that doesn't guarantee you won't be caught in
one of the uncommon accidents where your precautions work
against you.

Quoted message said:

There are hints that bike helmets *may* increase the risk
of rotational injury in certain circumstances, although
I'm not aware of any direct case studies on this.

I think any such thing would be affectively impossible. You
need reliably controlled data to do useful studies, and you
aren't going to get Ethics Committee approval for the sort
of experiment that provides reliably controlled data for
brain injuries! Just taking victims off the streets, you
won't really know enough about enough accidents to pin down
the hows and whys well enough unless there's an epidemic of
people with their necks screwed off, but AFAICT there isn't.

Quoted message said:

As you say, the key difference is that we have good,
statistically sound data that wearing a seat belt
increases one's chances of surviving a car journey.

If (and only if) all else is equal we have good evidence
that in a notional road accident a seat belt is more likely
to do us favours than not, and that's not /quite/ the same
thing as you've written above.

Quoted message said:

I wonder if the driver of the car linked to above wears a
seatbelt now?

Interesting question. If she's dopey enough to drive at
100mph on a wet road who can predict what she "thinks" about
safety questions? Hopefully she learned that doing the ton
in the wet is Not a Cunning Plan and isn't particularly
concerned about any seat belt aspect.

Pete.
--
Peter Clinch Medical Physics IT Officer Tel 44 1382 660111
ext. 33637 Univ. of Dundee, Ninewells Hospital Fax 44 1382
640177 Dundee DD1 9SY Scotland UK net
[email hidden] dundee.ac.uk~pjclinchOpen ↗

in message <[email hidden]>, Mark Tranchant

(') said:

There are undoubtably situations where both seat belts and
bicycle helmets reduce one's survival chances. Take for
example this accident:

topgear.com03Open ↗

Browsing that site as displacement activity because I have a
particularly heavy workload just now, I came across this:

<URL:http://www.topgear.com/content/jsp/bigpic.html?car=/co-
ntent/my_topgear/prangs/01/04/bigImages/img01.gif>

Enjoy.

Oh yes, before anyone asks, the frequency of my posting to
Usenet _is_ in direct proportion to the size and urgency of
my workload. You all better hope I get some let-up, soon.

--
[email hidden] (Simon Brooke)
jasmine.org.uk~simonOpen ↗

' ' <------- this blank intentionally spaced left

On Mon, 14 Jun 2004 14:36:48 +0100, Mark Tranchant
<[email hidden]> wrote in message
<[email hidden]>:

Quoted message said:

topgear.com03Open ↗

The pictures in that site are a munument to stupidity,
aren't they? So many of them are clearly the result of
physics: 1 testosterone: nil. And the explanations? Not -
ahem - universally plausible.

Guy
--
May contain traces of irony. Contents liable to settle after posting.
chapmancentral.co.ukchapmancentral.co.ukOpen ↗

88% of helmet statistics are made up, 65% of them at Washington University

Peter Clinch said:

anonymous coward said:

You make it sound as if it's almost the same thing, but
of course it isn't.

All I'm saying is that the impact velocity is not reduced
according to a simple bit of trigonometry, and I said that
because that is the case.

Velocity is a vector and includes a directional component;
speed is a scalar.

Quoted message said:

Quoted message said:

I tend to think the population studies are so fraught
with uncertainties and confounding factors that it's
difficult to draw any firm conclusions from them.

The point about whole population studies is you have a
very big sample base so that fact alone tends to round off
a lot of the awkward corners. In a single case /that/ is
where you are so fraught with uncertainties you can't
usefully deduce much except for very specific cases.

Quoted message said:

Do bike helmets help? Is autism linked to MMR? Do whisky
or olive oil drinkers make you live longer? What about
drinking whisky and olive oil together?

In the first case, very probably not as far as KSIs are
concerned. In the other cases you're not asking such
simple questions and there are plenty of other variables
that may influence your results. For example, olive oil
is just one thing in a diet so you can't really isolate
it. For MMR, there were autistic folk before there was
MMR and the causes weren't well understood to start with
so it could be all sorts of other things. But with
helmets we knew what the rates were before everyone was
wearing them and after everyone was wearing them where
compulsion was introduced. If they have a significant
effect you'd expect to see it when the figures come in,
and we don't. There will be other variables, but nothing
like as many, or of such significance, as the rest of a
diet aside from olive oil.

The only way to recreate accidents and their effects
properly for study
is... well, actually you can't do it.

You can try - that's what crash-test dummies and the like
are for. For example, I think I'm right in saying that most
of the evidence against bull bars is based on kinematics and
the epidemiological data against them is relatively weak.

Quoted message said:

There are far more things you have to account for in a
single incident than you have to account for at a
population level, because the population level accounts
for all the accidents people really had.

Quoted message said:

Take the seatbelt paradox - seatbelts may not make the
roads safer, but if I get into a car as a passenger I'm
probably well advised to wear one. It's a while since I
read John Adam's book, but IIRC the population statistics
would have suggested I shouldn't bother.

If you get into a car as a passenger then you have no
control over the vehicle, so effectively all else is
equal. And if all else is equal then you're better off
wearing a seatbelt.

This is my reasoning too - taken from JA.

Quoted message said:

Where the seatbelt safety argument breaks down is where
all things aren't equal: a driver may drive faster and
brake later because they feel safer with their seatbelt
on, for example, but that's irrelevant to a passenger.

IIRC John Adams argued that the population statistics showed
no reduction in road deaths after seatbelt wearing became
mandatory - if you take drink-driving and similar factors
into account.

This could be because:

1) All else stayed equal, but on average seat belts confer
no benefit.

2) Seat belts are effective, but drivers became more
reckless nullifying their advantages.

3) Car performance increased to match the rate of seatbelt-
wearing, rate so collision speeds became higher.

4..) Other explanations.

In:

adamsmith.org.ukpdfOpen ↗
files/risky-business.pdf

John Adams argues case (2) convincingly. He draws heavily on
whole-population studies and I'm not arguing that he's wrong
to do so. However, he also draws on small-scale interviews,
case-control studies and crash test data to make his case.

Quoted message said:

Whole population studies here show no effect for the
passenger but OTOH the nullifying effect is mainly
produced by the driver's actions.

Did JA have enough population data that he could make this
claim without recourse to other types of study? Not in the
link I've given above. But perhaps he has enough data to do
so (reading "Risk" is on my todo list).

AC

you know? said:

anonymous coward said:

I tend to think the population studies are so fraught
with uncertainties and confounding factors that it's
difficult to draw any firm conclusions from them.

A criticism which applies to a very much greater degree to
the small-scale prospective studies which suggest benefit.
If you're going to discount the evidence, then you need to
discount /all/ of it :-)

Quoted message said:

Do bike helmets help? Is autism linked to MMR? Do whisky
or olive oil drinkers make you live longer? What about
drinking whisky and olive oil together?

These things show correlations between things. The
population level studies on bike helmets show /no/
correlation - that's the point. You can argue for ever
about whether autosm is caused MMR or just happens to
manifest at around the time the MMR is administered; in
the case of helmets what you have is no discernible effect
(see Tony's "spot the year" charts posted earlier).

Would even a large protective effect necessarily show up?
Especially given that minor and major head injuries are
grouped together (i.e. all the population studies use a
blunt tool).

Suppose helmets were 60% effective at preventing head
injuries. If c. 30% of admissions include a head injury, if
40% more cyclists wore a helmet, we might only expect to see
a reduction of 100x(0.6*0.4*0.3) = 7.2% in the total number
of injuries that resulted in a head injury, or a 25%
decrease in the number of head injuries.

By eyeball, the amount of year-to-year variation in Tony's
graphs is pretty high compared to 25%. And we know this
data is affected by a decrease in cycling and other
confounding factors.

Quoted message said:

Quoted message said:

Take the seatbelt paradox - seatbelts may not make the
roads safer, but if I get into a car as a passenger I'm
probably well advised to wear one. It's a while since I
read John Adam's book, but IIRC the population statistics
would have suggested I shouldn't bother.

Invalid comparison: Adams' population statistics show that
for optimum safety everybody except the driver should wear
one :-) As a passenger you will probably not materially
affect the outcome of the journey.

There could be a similar argument for cycle-helmets.
Assume that due to helmet compulsion, there are fewer
cyclists and the balance of leisure v. transportational
cyclists changes. Perhaps drivers will be less cyclist-
aware and may believe that cyclists are now invulnerable -
causing them to drive more dangerously and increase the
proportion of accidents that result in a head injury.
Perhaps cyclists who were less likely to have the sort of
accident resulting in a head injury will stop cycling.
Even if helmets are quite effective they may not be good
enough to counter the increased risk, and the proportion
of head injuries will go up.

If this scenario were accurate I would draw two conclusions:

Cycle-helmet compulsion is wrong and dangerous

As a cyclist, I should wear a helmet

It wouldn't be inconsistent with a level / increasing
trend in the proportion of injuries that include head
injuries (though there may be data that I'm not aware of
to counter it).

AC

Simon Brooke [email hidden] opined the following...

Quoted message said:

in message <[email hidden]>,
Mark Tranchant Browsing that site as displacement activity
because I have a particularly heavy workload just now, I
came across this:

<URL:http://www.topgear.com/content/jsp/bigpic.html?car=/-
content/my_topgear/prangs/01/04/bigImages/img01.gif>

Cracking. ;-)

Jon