Austin to Evaluate Local Emergency Room Data to Determine Whetheror Not to Implement an All-Ages Hel

[Helmut Springer]

SMS <[email protected]> wrote:
>> I fail to see any correlation between "seeing a lot of head
>> injuries" and being able to "judge of how an accident's mechanic
>> worked and which difference a certain type of helmet would have
>> made".
>
> Of course you do!

You snipped my request to elaborate...


> You've already made your decision, despite all the evidence, and
> you're desperately looking for a way to dispute not only all the
> previous studies, but the future ones as well, since you are well
> aware that it's unlikely that the Austin study will reach a
> different conclusion than the plethora of previous studies. No
> amount of expert evidence will convince you to admit the facts,
> even though you actually do know them.

....and you do not elaborate but only repeat your claim.

I understand you can't back your point with reason, feel invited to
prove me wrong.

--
MfG/Best regards
helmut springer

 

[Ozark Bicycle]

SMS wrote:
> Helmut Springer wrote:
>
> > I fail to see any correlation between "seeing a lot of head
> > injuries" and being able to "judge of how an accident's mechanic
> > worked and which difference a certain type of helmet would have
> > made".
>
> Of course you do! You've already made your decision, despite all the
> evidence, and you're desperately looking for a way to dispute not only
> all the previous studies, but the future ones as well, since you are
> well aware that it's unlikely that the Austin study will reach a
> different conclusion than the plethora of previous studies. No amount of
> expert evidence will convince you to admit the facts, even though you
> actually do know them.


That sounds like a good description of the early stages of Anti-helmet
Psycho Syndrome.

 

[Helmut Springer]

gds <[email protected]> wrote:
> 3) You are absolutely wrong in stating that an ER doc needs to
> know and understand all the issues that you list to make a
> judgement concerning the level of injury and the use of a helmet.
> He may need to know all that if he wants to say "how" the helmet
> prevented a more serious injury. But to say "from what I've
> observe injuries in those cyclist reporting the use of a helmet
> are less (or more or the same) requires none of that. It is an
> observation of end point and not an explantion of how one got
> there.

The "from what I've observe" due to the biased sample that is
observed makes the result void when talking about anything else but
exactly the observed sample of cyclists and incidents. Very basic
statistics. If I sit besides a 15% mountain road and measure power
output of riding by cyclists the resulting statistic will be
meaningless with regards to the overall German cycling population,
regardless of how well I measure.

--
MfG/Best regards
helmut springer

 

[Helmut Springer]

gds <[email protected]> wrote:
> Rather he is saying I've seen lots fo patients and here is what
> I've observed.
> That is not a biased sample. It is neither biased nor a sample. It
> is the observation (not a conclusion) accross the entire
> population of his experience.

If you do not know the correlation between his "population of
experience" (commonly called "sample") and the overall population
(commonly called "real world") you can't deduce anything from the
first for the second.

Now I don't see any reason why anyone in this discussion should care
about any specific small sample with unknown characteristics of some
ER doc. Maybe you can explain?


--
MfG/Best regards
helmut springer

 

[[email protected]]

gds wrote:
>
>
> 1)Frank, you are confusing the "discovery" process with the validation
> process. Most drugs are first discovered and only after that discovery
> are they put through clinical trials. It is only very recently that
> biomedical scince has reached the point where drugs are designed to do
> specific things. Until now various compunds were simply discovered to
> have an effect. And they were sold and marketed without understanding
> how the effect was accomplished.
> Do you think that the natives understood how quinine worked? Nope they
> just knew it worked.

I don't see how you can characterize this attempt at a study as being a
"discovery" process. First, helmets have obviously been discovered by
now. But more importantly, the physicians quoted make it clear that
they think it's a validation process. They are claiming it will
demonstrate the value of helmets. "... there's really no doubt what
the outcome is going to be," said Dr. Steve Berkowitz, CEO of St.
David's Healthcare."

(I note Berkowitz seems blissfully unaware of the evidence of his
bias!)


> 2)You are also confusing real world ER visit behaviour with whatever
> you think it might be. Most studies I've seen report just the opposite
> of what you are suggesting. That is the poor and uninsured use ER's and
> the wealthier and insured use other modes of getting care.
> Of course, if the injury is really severe that doesn't apply and the
> ambulance goes to the ER.But since you posit that there are virtually
> no serious injuries from cycling that shouldn't mean much.

I'd like a citation on the "most studies" parrt of that.

The study that generated the absurd "85%" claim, Thompson, R., Rivara,
F., & Thompson, D., A Case-Control Study of Effectiveness of Bicycle
Safety Helmets, New England Journal of Medicine - May 25, 1989, was
immediately criticized about the obvious differences between cases and
controls. Non-helmeted cyclists were much more likely to be minority,
riding on streets, falling on hard (not grassy) surfaces, struck by
cars, etc. They had about three times the number of broken legs as the
helmeted cyclists. In response, T, R & T claimed they had
computationally adjusted for such confounding variables, but one
statistician to whom they released their data found errors in their
computations.

Furthermore, the percentage of helmeted cyclists in that (and other)
case-control studies greatly exceeded the observed percentage of
helmeted cyclists in contemporary street surveys. That fact alone
indicates the helmeted folks (or helmeted kids) were more likely to
show up at the ER.

Now, we can speculate why. Some have said that, especially in that
early study, it was because helmets were on especially nervous folks
(or children of nervous parents). Others have claimed that the helmets
induce head impacts, or at least crashes. Those are possible, but it's
always seemed to me that the insurance angle is a likely cause of the
discrepancy.

> 3) You are absolutely wrong in stating that an ER doc needs to know and
> understand all the issues that you list to make a judgement concerning
> the level of injury and the use of a helmet. He may need to know all
> that if he wants to say "how" the helmet prevented a more serious
> injury. But to say "from what I've observe injuries in those cyclist
> reporting the use of a helmet are less (or more or the same) requires
> none of that. It is an observation of end point and not an explantion
> of how one got there.

If he is to judge whether a helmet would have made a difference, ISTM
he needs to know quite a lot about helmets, and about the specific
crash. Would an auto body mechanic be able to say whether anti-lock
brakes would have prevented a particular accident? Very, very
unlikely!

And, BTW, the ER doc would certainly have to recognize the possibility
of sampling bias due to self selection! Again, the method they're
using would give credit to the red cotton cycling caps!

I see no evidence these docs are even dimly aware of such matters.
>From what I know from my ER doc friend, they are trained in body
repair. But unless they are into real research (and have specialized
help) they don't know much about the fine points of experimental
design.

- Frank Krygowski

 

[SMS]

gds wrote:

> So, let us say that he keps these rudimentary records and that over the
> past 25 years he has observed a total 100 head injuries in cyclists.
>>From his notes he can determines that for whatever reason the average
> severity of of injury in non helmeted cyclist is greater than in
> helmeted cyclists. He can report that as a valid observation. What he
> can not do is say that helmets "caused" the injuries to be less severe
> nor can he say that from these observations he believes that this will
> hold true for all other ER docs. But his observation based on recorded
> information is well within his area of expertise is a valid summary
> observation.

The fact is that the ER data is about the best data you're going to get.
You can't do a double-blind study on helmet effectiveness, and the
population studies are meaningless because there are so many
uncontrolled variables. ER doctors, that see a lot of head injury
patients, are the best source of unbiased data, which is why so many
studies use them.

I think that most of us agree that legislation should not be the result
of the evaluation of the ER studies. Are there any adult helmet laws at
all in the United States? The approach to stopping helmet laws should
not be to attack ER data, but to continue to stress the relatively low
number of accidents where a helmet would have had any effect.

 

[Tom Keats]

In article <[email protected]>,
SMS <[email protected]> writes:
> "http://keyetv.com/topstories/local_story_284175358.html"
>
> I think that the second line of the story was supposed to say "They will
> now get some from local hospitals," rather than "They will not get some
> from local hospitals."
>
> This is likely bad news for the AHZ's, as the current ER data already
> proves the benefits of helmet use when a head impact accident occurs.

That you consistently insist on referring to helmet skeptics as
"zealots" implies that you are a PHZ. Of course, you're not a
zealot. Neither are the skeptics. But this sort of escalatory
dialogue accomplishes nothing but contention (which also
accomplishes nothing more than a mere ego massage for warmongers.)

I note the more intelligent helmet skeptics (which excludes myself)
typically don't resort to contentious language, but rather prefer
to rebut with rationality and consideration.

Skeptics: 9 Pro Helmet People: 1 (2 on, 2 out, top of the
999th inning)

Out here in the real world I know/admire/love many people who
insist on wearing their helmets, or insist on not wearing any
helmet, or just plain don't care one way or another. But
they don't make a big steenkin' agendum about it. They just
love to ride, same as myself.

> IMVAIO, the city council needs to not look at just ER data,

IMO city councils need to keep the fire hydrants working,
the cops doing their jobs properly, deal with real estate
variances, and sell dog licenses. What do city councillors
have to do with ERs anyways?

And "data" is not Information:
http://en.wikipedia.org/wiki/Case-control_study

City Council needs ER data like I need Endless Shrimp
at Red Lobster.

> but at the
> relatively small number of serious accidents. By looking only at the
> subset of accidents, they are not looking at the big picture, but then
> politicians rarely do.

Do you sell helmets as well as coffee stuff and
overblown bike lights?

I don't really expect an answer, as I know you've killfiled me.


have a good weekend,
Tom

--
-- Nothing is safe from me.
Above address is just a spam midden.
I'm really at: tkeats [curlicue] vcn [point] bc [point] ca

 

[[email protected]]

SMS wrote:
>
> The fact is that the ER data is about the best data you're going to get.
> You can't do a double-blind study on helmet effectiveness, and the
> population studies are meaningless because there are so many
> uncontrolled variables.

<sigh> False, and explained many times.

The population studies are time-series studies. They keep track of the
head injury data over a period of years, perhaps decades, and note what
sorts of changes happen when helmet use jumps by a large amount. (For
example, one recent study looked at every population where helmet use
suddenly jumped by a minimum of 40% - in all cases, due to the
imposition of MHLs. It found no benefit.)

The "controlled variable" in such a study is helmet use. What are the
"uncontrolled variables"? The usual list is: increasing motor vehicle
traffic; popularity of video games and other distractions from cycling;
larger distances between home and work; climate change; and so on.

But these effects are very easily distinguished from that of helmet
use! When a MHL causes helmet use to rise 40% or more, it happens
within a few weeks. You do NOT get an entire population to move ten
miles further away in a few weeks. You do NOT get a massive increase
in video game activity in a few weeks. You do NOT get tremendous
increases in traffic in a few weeks.

The helmet use changes are effectively step changes. The other changes
are gradual changes. If helmets have beneficial effects, the benefits
should appear as step changes.

The fact that the benefits have NOT appeared indicates that the
promises of benefits are false. It's really that simple - just as it
was for Hormone Replacement Therapy.

Steven, you said "No amount of expert evidence will convince you to
admit the facts, even though you actually do know them." You _must_
have been projecting your own faults on others, because several people
have explained the above to you several times.

You MUST be remaining willfully ignorant.

- Frank Krygowski

 

[SMS]

peter wrote:

> Self-selected group studies are notorious for reaching false
> conclusions since it's impossible to know all the ways in which the
> self-selection has biased the results of the study.

The ER studies remove the self-selection completely. They aren't
determining the overall number of head injuries, they are evaluating,
within the head injuries that come into the ER, whether or not a helmet
would have made a difference in the severity of the injury, something
that they are uniquely qualified to do.

Their only conclusion is whether or not a helmet is effective in
head-impact crashes serious enough to warrant an emergency room visit.
The real problem comes when they try to extrapolate this conclusion into
public policy.

In reality, it's the population studies, so often incorrectly cited by
AHZs, that have a high degree of self-selection. The most infrequent
cyclists have removed themselves from the pool of cyclists, the total
miles that the population cycles barely changes, then the AHZs calculate
the number of injuries per cyclists and proclaim that there was no
effect from helmets, even though the number of cyclists allegedly fell
by a significant percentage.

 

[Helmut Springer]

gds <[email protected]> wrote:
> Helmut Springer wrote:
>> If you do not know the correlation between his "population of
>> experience" (commonly called "sample") and the overall population
>> (commonly called "real world") you can't deduce anything from the
>> first for the second.

Please read and understand this very basic principle of statistics.


> You continue to confuse the creation of individual data points and
> some measure of a number of some such data points.

I'm not confusing anything. This discussion is about significance
of statistics, and you are obviously even aware that this is not
given:


> Now let's say we construct a proper sample of all ER doc's in the
> US and get them to do such observations and to report them. If we
> construct our sample properly then we can surely extrapolate to
> say something like "it looks like in the US that in ER visits the
> severity of head injuries is less in those who report using a
> helmet vs. that report not using a helmet." Again, one can not say
> that helmet use caused this lower severity nor would it be OK to
> exptrapolate the findings beyond the US.

So in the course of this discussion your just adding noise, correct
me if I'm wrong.


--
MfG/Best regards
helmut springer

 

[peter]

SMS wrote:
> peter wrote:
>
> > Self-selected group studies are notorious for reaching false
> > conclusions since it's impossible to know all the ways in which the
> > self-selection has biased the results of the study.
>
> The ER studies remove the self-selection completely.

Clearly false. These studies suffer from self-selection in two ways.
1) The portion of the population that chooses to wear a helmet when
cycling is self-selected, and 2) except for the small percentage of
accident victims who are unconscious and therefore unable to either
accept or decline treatment, the decision of whether to appear at the
ER is a self-selected one - and people will differ considerably on how
severe the injury has to be before seeking professional medical
treatment.
....
> they are evaluating,
> within the head injuries that come into the ER, whether or not a helmet
> would have made a difference in the severity of the injury, something
> that they are uniquely qualified to do.

My daughter has undergone some ER training already in med school, but
it is all about how to evaluate the severity of an injury and the
treatment to apply. There has been no training at all on how to
determine what effect a hypothetical helmet would have had on the
accident - nor would I see this as something that should be part of an
ER physician's training.
....
> In reality, it's the population studies, so often incorrectly cited by
> AHZs, that have a high degree of self-selection. The most infrequent
> cyclists have removed themselves from the pool of cyclists, the total
> miles that the population cycles barely changes, then the AHZs calculate
> the number of injuries per cyclists and proclaim that there was no
> effect from helmets, even though the number of cyclists allegedly fell
> by a significant percentage.

False again. The evaluations of the impact on cycling particpation
from MHLs has come from traffic surveys in a given area taken before
and after imposition of the MHL. Someone who cycles only infrequently
is unlikely to heppen to pass by one of the survey locations and
therefore if he stops cycling it won't have much effect on the survey
results. OTOH, an active commuter and/or recreational rider may pass
by multiple survey sites during the study period and therefore has a
much greater impact on the results. In reality, such cyclist traffic
counts are a representative measurement of the mileage covered by
cyclists in the area. If each cyclist started riding twice as often or
twice as far then the counts would also double even though the actual
number of individual cyclists remained the same.

 

[SMS]

peter wrote:
> SMS wrote:
>> peter wrote:
>>
>>> Self-selected group studies are notorious for reaching false
>>> conclusions since it's impossible to know all the ways in which the
>>> self-selection has biased the results of the study.
>> The ER studies remove the self-selection completely.
>
> Clearly false. These studies suffer from self-selection in two ways.
> 1) The portion of the population that chooses to wear a helmet when
> cycling is self-selected, and 2) except for the small percentage of
> accident victims who are unconscious and therefore unable to either
> accept or decline treatment, the decision of whether to appear at the
> ER is a self-selected one - and people will differ considerably on how
> severe the injury has to be before seeking professional medical
> treatment.

Again, what they are evaluating is whether or not the people that
present themselves to the emergency room, with head injuries from a
bicycle accident, that weren't wearing a helmet, would have fared better
with a helmet. That's all they are trying to determine. The severity of
injury before seeking ER treatement would be material if they were
trying to determine the number of accidents with head injuries, but
that's not what they are studying.

 

[[email protected]]

On 20 Oct 2006 19:02:37 -0700, [email protected] wrote:

>
>Steven, you said "No amount of expert evidence will convince you to
>admit the facts, even though you actually do know them." You _must_
>have been projecting your own faults on others, because several people
>have explained the above to you several times.
>
>You MUST be remaining willfully ignorant.

This should be no surprise.

Bill Sornson, another of the pro-helmet/pro-MHL crowd, is on record as
saying her prefers _not_ to examine the data about helmet
(in)effectiveness.

 

[Johnny Sunset aka Tom Sherman]

SMS wrote:
> peter wrote:
> > SMS wrote:
> >> peter wrote:
> >>
> >>> Self-selected group studies are notorious for reaching false
> >>> conclusions since it's impossible to know all the ways in which the
> >>> self-selection has biased the results of the study.
> >> The ER studies remove the self-selection completely.
> >
> > Clearly false. These studies suffer from self-selection in two ways.
> > 1) The portion of the population that chooses to wear a helmet when
> > cycling is self-selected, and 2) except for the small percentage of
> > accident victims who are unconscious and therefore unable to either
> > accept or decline treatment, the decision of whether to appear at the
> > ER is a self-selected one - and people will differ considerably on how
> > severe the injury has to be before seeking professional medical
> > treatment.
>
> Again, what they are evaluating is whether or not the people that
> present themselves to the emergency room, with head injuries from a
> bicycle accident, that weren't wearing a helmet, would have fared better
> with a helmet. That's all they are trying to determine. The severity of
> injury before seeking ER treatement would be material if they were
> trying to determine the number of accidents with head injuries, but
> that's not what they are studying.

What QUALIFICATIONS do EMERGENCY ROOM PHYSICIANS have to determine if a
thin expanded polystyrene shell SIGNIFICANTLY reduced head injuries in
an accident?

Will these EMERGENCY ROOM PHYSICIANS be working with ACCIDENT
RECONSTRUCTION PROFESSIONALS to determine the approximate velocity
vector of the cyclist's head at the time of impact? And if not, how
will they come to any valid conclusion?

--
Tom Sherman - Here, not there.

 

[peter]

SMS wrote:
> peter wrote:
> > SMS wrote:
> >> peter wrote:
> >>
> >>> Self-selected group studies are notorious for reaching false
> >>> conclusions since it's impossible to know all the ways in which the
> >>> self-selection has biased the results of the study.
> >> The ER studies remove the self-selection completely.
> >
> > Clearly false. These studies suffer from self-selection in two ways.
> > 1) The portion of the population that chooses to wear a helmet when
> > cycling is self-selected, and 2) except for the small percentage of
> > accident victims who are unconscious and therefore unable to either
> > accept or decline treatment, the decision of whether to appear at the
> > ER is a self-selected one - and people will differ considerably on how
> > severe the injury has to be before seeking professional medical
> > treatment.
>
> Again, what they are evaluating is whether or not the people that
> present themselves to the emergency room, with head injuries from a
> bicycle accident, that weren't wearing a helmet, would have fared better
> with a helmet. That's all they are trying to determine.

That part is well understood. What you're missing is that due to the
self-selection that occurs before the patients ever get to the ER the
data available to the ER researchers is biased in ways that make it
impossible for them to make that determination. The old "garbage in,
garbage out" rule applies here.

What the ER staff can observe is that they have two groups of patients:
group A that was wearing helmets at the time of the crash, and group B
that wasn't. They also can evaluate how frequent and how severe the
head injuries are of group A and group B. Based on that they then
conclude that group A had some percentage fewer head injuries than
group B and, at least in the studies that I've seen, they ascribe that
difference between the two groups to the wearing of helmets. That
would be fine if in fact helmet wearing were the only difference
between groups A and B. But in the real world there are likely to be
many other differences between the two groups and the varying injury
experience may well be due to those other differences and have nothing
to do with helmet wearing. In the case of the Seattle study it was
subsequently found that people in group A were from much higher
socio-economic strata than those in group B, they were far more likely
to have health insurance, were more likely to be riding recreationally
in quiet suburbs as opposed to transportationally in the city, and many
other differences between the groups besides their choice of head
covering.

If the people in one group tend to be upper-middle class kids whose
parents take them to get medical treatment "just in case" for just
about any little cycling mishap, then the patients in this group will
tend to have a much lower percentage of head injuries when compared to
a group that will only visit an ER after accidents that result in
severe injuries.
If the first group is also more likely to wear helmets, then the result
of an ER case-control study will be that helmets are effective, even if
the helmets themselves make no difference at all.

 

[[email protected]]

SMS wrote:
>
>
> Again, what they are evaluating is whether or not the people that
> present themselves to the emergency room, with head injuries from a
> bicycle accident, that weren't wearing a helmet, would have fared better
> with a helmet. That's all they are trying to determine. The severity of
> injury before seeking ER treatement would be material if they were
> trying to determine the number of accidents with head injuries, but
> that's not what they are studying.

Even if the ER docs were not biased due to the observed (but
unappreciated) differences between the well-off and the poor, they are
unlikely to be able to make the judgement of what a helmet _would_ have
done. At least, if we're talking about head injuries of any
consequence.

If a physician sees, for example, evidence of internal bleeding or
swelling of the brain, how is he expected to judge whether the minimal
protection offered by thin styrofoam would have prevented that? How
can he know enough about helmet performance to tell?

My guess is the typical ER physician literally thinks bike helmets
prevent almost all head injury - that is, 85%. He doesn't know that
there is real scientific doubt about their effectiveness. He certainly
doesn't know any of the details of helmet certification. And he
doesn't even see many _serious_ bike head injuries upon which to base
his judgement. (Remember, serious cycling head injuries are quite
rare, far more rare than serious auto passenger head injuries.)

IOW, he is in no way equipped to make that judgement regarding any
particular serious head injury.

But my prediction is that any given ER doc won't be asked about many
serious head injuries. He'll see many more cyclists with road rash on
the knees and hands. Some of those will also have a scrape on the
head. The ER doc will recognize the minor head scrape as being
something even a flimsy styrofoam helmet would prevent, and he'll
sternly make his pronounement: "Nurse, this is yet another head injury
a helmet would have prevented!"

City council will not know that the head scrape was no more serious
than the knee scrape. They'll interpret "head injury" as "debilitating
brain injury," just as intended by promoters of this so-called "study."

The only serious question I have about this is: Was the entire scheme
concocted by Bell Sports?

- Frank Krygowski

 

[Ozark Bicycle]

[email protected] wrote:

<snipped>

> The only serious question I have about this is: Was the entire scheme
> concocted by Bell Sports?
>

Now we see that paranoia is a component of Anti-helmet Psycho Syndrome.


> - Frank Krygowski

^^^^^^^^^^^^^^^^^^^^^^^^
The Grand Poohbah of the Anti-helmet Psychos.

 

[Bill Sornson]

[email protected] wrote:

> My guess is the typical ER physician literally thinks bike helmets
> prevent almost all head injury - that is, 85%. He doesn't know that
> there is real scientific doubt about their effectiveness.

Boy, he sure sounds stupid. Wonder how he became a physician?!?

 

[Ozark Bicycle]

Bill Sornson wrote:
> [email protected] wrote:
>
> > My guess is the typical ER physician literally thinks bike helmets
> > prevent almost all head injury - that is, 85%. He doesn't know that
> > there is real scientific doubt about their effectiveness.
>
> Boy, he sure sounds stupid. Wonder how he became a physician?!?


It'a all part of the the Bell Sports sponsored World Wide Conspiracy to
Promote Helmets. ;-)

 

[[email protected]]

Ozark Bicycle wrote:
> Bill Sornson wrote:
> > [email protected] wrote:
> >
> > > My guess is the typical ER physician literally thinks bike helmets
> > > prevent almost all head injury - that is, 85%. He doesn't know that
> > > there is real scientific doubt about their effectiveness.
> >
> > Boy, he sure sounds stupid. Wonder how he became a physician?!?
>
>
> It'a all part of the the Bell Sports sponsored World Wide Conspiracy to
> Promote Helmets. ;-)

This is a little like attending a panel discussion on medical ethics,
and having it crashed by two little boys who escaped from Problem Child
Day Care.

- Frank Krygowski