Pharmaceutical industry's drug research 'biased'

Since 2001, over 90,000 cyclist's have joined Cycling Forums to discuss topics from general cycling to equipment, training, racing and travel or vacation destinations (especially in europe during the tour de france). We also feature an great deals in our online store, 100's of articles, classifieds and product reviews.

http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda

* * *

Pharmaceutical industry's drug research 'biased'

By Geoff Dyer, May 30 2003 Financial Times, UK.

Research on new drugs sponsored by the pharmaceuticals industry is guilty of "systematic bias",
according to an article in today's British Medical Journal.

The report, which was based on an analysis of 30 clinical studies, says that drug company research
exaggerates the benefits of the products under investigation. Such studies can play an important
part in a doctor's decision to prescribe a certain treatment.

The article is part of a special issue of the BMJ, the leading medical journal in the UK, on the
relationship between drug companies and doctors.

Richard Smith, editor of the BMJ, said the issue of the magazine was not "anti-drug company".
However, he added: "Our central argument is that doctors, drug companies and, most importantly,
patients would all benefit from greater distance between doctors and drug companies."

The analysis of clinical research was conducted by Joel Lexchin at York University in Toronto. It
concluded that companies sometimes chose an inappropriate comparison for the trial, for instance by
using a higher dose of its product than that of a rival, which could bias the results. Research paid
for by a company was less likely to be published than research funded from other sources, it added.

Another article in the BMJ looking at research on five anti-depressant drugs submitted to Swedish
regulators said there was a tendency to report only the more favourable results.

Trevor Jones, director-general of the Association of the British Pharmaceutical Industry, said: "It
is important both to the industry, as producers of modern, innovative medicines, and to those who
prescribe them, that regular dialogue is maintained to ensure that patients benefit from the most
appropriate treatment for their condition."

Separately, health minister Hazel Blears said yesterday that, between 2000 and 2002, the government
had received eight reports of suicides and more than 1,000 reports of adverse reactions to the
anti-depressant Seroxat, which doctors believed could be linked to the drug.

The government launched a new inquiry last week into the class of drugs known as selective serotonin
re-uptake inhibitors, which includes Seroxat and Prozac, following claims by patients that they had
become addicted to the drugs or felt suicidal after taking them. Glaxo- SmithKline, which makes
Seroxat, says the drug has already been widely investigated.

A previous expert panel into the safety of the drugs was disbanded in March after it was disclosed
that two members owned shares in GlaxoSmithKline and a witness did not disclose all his links with
the industry.

* * *

http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda

"Theta" <theta@btinternet.com> wrote

> Richard Smith, editor of the BMJ, said the issue of the magazine was not "anti-drug company".
> However, he added: "Our central argument is that doctors, drug companies and, most importantly,
> patients would all benefit from greater distance between doctors and drug companies."

So the article itself is biased.

> It concluded that companies sometimes chose an inappropriate comparison for the trial, for
> instance by using a higher dose of its product than that of a rival, which could bias the results.

This is not bias. Comparison to a rival would simply be a different study. The reason drug companies
don't do comparisons to rivals is that often their own drugs are no better than their rivals' drugs.
Proving that your drug is just as good as a competitor's drug does not usually enhance sales.

Also, in the U.S., the FDA does not require that a new drug be better than existing drugs, only that
it be more efficacious than placebo. Clinical research is expensive, so why fund a study to prove
something that you don't need to prove?

> Research paid for by a company was less likely to be published than research funded from other
> sources, it added.

This objection ignores a fact of statistics. When a study shows that, within statistically
significant limits, a drug is effective, then that drug is effective. As long as the study is valid
(and this depends on many factors), then the statistics don't lie about the result. As a physician,
you have to read the report of the study and decide for yourself whether its design is a good one.
If you decide that it is, then you can take the conclusion of the study to the bank.

The reason that some studies are not published is that there's no financial motivation to publish a
study that proves that your drug is not effective. This stands to reason. But the studies that are
published, which demonstrate efficacy of the drugs, are no less valid for it.

On Sun, 29 Jun 2003 03:54:14 -0400, "Matt Beckwith" <matt@doctorbeckwith.com> wrote:

>"Theta" <theta@btinternet.com> wrote
>

>
>> Research paid for by a company was less likely to be published than research funded from other
>> sources, it added.
>
>This objection ignores a fact of statistics. When a study shows that, within statistically
>significant limits, a drug is effective, then that drug is effective. As long as the study is valid
>(and this depends on many factors), then the statistics don't lie about the result. As a physician,
>you have to read the report of the study and decide for yourself whether its design is a good one.
>If you decide that it is, then you can take the conclusion of the study to the bank.
>
>The reason that some studies are not published is that there's no financial motivation to publish a
>study that proves that your drug is not effective. This stands to reason.

Thus it follows that there should be a requirement that all data, pos or neg, be released.
see below.

>But the studies that are published, which demonstrate efficacy of the drugs, are no less
>valid for it.

That is quite incorrect. An individual study only gives some p value (probability) of a difference.
The best estimate of the true p is based on using all data. Withholding any data distorts the view.

A common way to assess that a result is significant is to use the 95% probability cutoff. But if you
do 20 tests that give a 95% probability of "success", statistics says that one of those is a fluke.
Again, this shows why hiding negative data biases the overall view.

bob

Let me get this straight. You are saying that choosing, voluntarily "an inappropriate comparison for
the trial" because you want to show those particular data is NOT bias?

Were you on the O.J. trial also?

"Matt Beckwith" <matt@doctorbeckwith.com> wrote in message news:bdm61l01jcl@enews4.newsguy.com...
> "Theta" <theta@btinternet.com> wrote
>
> > Richard Smith, editor of the BMJ, said the issue of the magazine was not "anti-drug company".
> > However, he added: "Our central argument is that doctors, drug companies and, most importantly,
> > patients would all benefit from greater distance between doctors and drug companies."
>
> So the article itself is biased.
>
> > It concluded that companies sometimes chose an inappropriate comparison for the trial, for
> > instance by using a higher dose of its product than that of a rival, which could bias the
> > results.
>
> This is not bias. Comparison to a rival would simply be a different
study.
> The reason drug companies don't do comparisons to rivals is that often
their
> own drugs are no better than their rivals' drugs. Proving that your drug
is
> just as good as a competitor's drug does not usually enhance sales.
>
> Also, in the U.S., the FDA does not require that a new drug be better than existing drugs, only
> that it be more efficacious than placebo. Clinical research is expensive, so why fund a study to
> prove something that you
don't
> need to prove?
>
> > Research paid for by a company was less likely to be published than research funded from other
> > sources, it added.
>
> This objection ignores a fact of statistics. When a study shows that, within statistically
> significant limits, a drug is effective, then that
drug
> is effective. As long as the study is valid (and this depends on many factors), then the
> statistics don't lie about the result. As a physician, you have to read the report of the study
> and decide for yourself whether
its
> design is a good one. If you decide that it is, then you can take the conclusion of the study to
> the bank.
>
> The reason that some studies are not published is that there's no
financial
> motivation to publish a study that proves that your drug is not effective. This stands to
> reason. But the studies that are published, which demonstrate efficacy of the drugs, are no less
> valid for it.

"Bob" <xyzbbruner@uclink4.berkeley.edu> wrote

> That is quite incorrect. An individual study only gives some p value (probability) of a
> difference. The best estimate of the true p is based on using all data. Withholding any data
> distorts the view.
>
> A common way to assess that a result is significant is to use the 95% probability cutoff. But if
> you do 20 tests that give a 95% probability of "success", statistics says that one of those is a
> fluke. Again, this shows why hiding negative data biases the overall view.

If a single study has a p value of less than 0.05, that means that there's a 95% chance that its
conclusion is correct. 95 is pretty high percentage. You wouldn't expect to be able to do another
study that has a p value less than 0.05 that proves the opposite. If that happened, it would
certainly be remarkable, and the result should not be hidden. But I bet it rarely occurs.

More likely, another study comes up with a p value slightly greater than .05. There is general
agreement that you haven't proven the hypothesis in that case. But that doesn't mean that the
hypothesis is false.

A better explanation for studies contradicting each other than invalidity of the conclusions is that
some studies are poorly designed. In fact, many studies are poorly designed.

A case in point: Studies regarding the efficacy of antidepressants often involve counseling. They
give counseling to the patients taking the drug, and they counsel the patients taking the placebo.
This is part of the reason why the placebo group improves. It's not really a placebo group, since
they got counseling.

Another example: A study to compare two antidepressants drops patients from the study who get
improvement in their depression within the first 2 weeks. The researchers are of the belief that
antidepressants can't possibly work that quickly. But antidepressants often work within the first
two days. The drug which works more quickly is going to come out looking worse in such a study.

"Bob" <xyzbbruner@uclink4.berkeley.edu> wrote in message news:3eff1768.6386828@agate.berkeley.edu...
> On Sun, 29 Jun 2003 03:54:14 -0400, "Matt Beckwith" <matt@doctorbeckwith.com> wrote:
>
> >"Theta" <theta@btinternet.com> wrote
> >
>
> >
> >> Research paid for by a company was less likely to be published than research funded from other
> >> sources, it added.
> >
> >This objection ignores a fact of statistics. When a study shows that, within statistically
> >significant limits, a drug is effective, then that
drug
> >is effective. As long as the study is valid (and this depends on many factors), then the
> >statistics don't lie about the result. As a
physician,
> >you have to read the report of the study and decide for yourself whether
its
> >design is a good one. If you decide that it is, then you can take the conclusion of the study to
> >the bank.
> >
> >The reason that some studies are not published is that there's no
financial
> >motivation to publish a study that proves that your drug is not
effective.
> >This stands to reason.
>
> Thus it follows that there should be a requirement that all data, pos or neg, be released.
> see below.
>
>
> >But the studies that are published, which demonstrate efficacy of the drugs, are no less valid
> >for it.
>
> That is quite incorrect. An individual study only gives some p value (probability) of a
> difference. The best estimate of the true p is based on using all data. Withholding any data
> distorts the view.
>
> A common way to assess that a result is significant is to use the 95% probability cutoff. But if
> you do 20 tests that give a 95% probability of "success", statistics says that one of those is a
> fluke. Again, this shows why hiding negative data biases the overall view.

Witholding data from a pulished study, or witholding a whole study?

On Sun, 29 Jun 2003 13:59:40 -0400, "Matt Beckwith" <matt@doctorbeckwith.com> wrote:

>"Bob" <xyzbbruner@uclink4.berkeley.edu> wrote
>

At the outset... i agree with much of your discussion of trials below. I am focusing here on certain
areas where discussion seems warranted.

I think we also would agree...

We are talking about trials that are "done well", and reported with sufficient detail that we can
interpret them.

The cutoff of p = 0.05 is somewhat arbitrary. It is not that 0.04 is "good" and 0.06 is "bad".

>> That is quite incorrect. An individual study only gives some p value (probability) of a
>> difference. The best estimate of the true p is based on using all data. Withholding any data
>> distorts the view.
>>
>> A common way to assess that a result is significant is to use the 95% probability cutoff. But if
>> you do 20 tests that give a 95% probability of "success", statistics says that one of those is a
>> fluke. Again, this shows why hiding negative data biases the overall view.
>
>If a single study has a p value of less than 0.05, that means that there's a 95% chance that its
>conclusion is correct. 95 is pretty high percentage. You wouldn't expect to be able to do another
>study that has a p value less than 0.05 that proves the opposite. If that happened, it would
>certainly be remarkable, and the result should not be hidden. But I bet it rarely occurs.
>
>More likely, another study comes up with a p value slightly greater than .05. There is general
>agreement that you haven't proven the hypothesis in that case. But that doesn't mean that the
>hypothesis is false.
>

It is fundamental statistics that the p=0.05 from the one trial means there is a 1 in 20 chance that
the result is "by chance". And the only way to know is to get more data.

>A better explanation for studies contradicting each other than invalidity of the conclusions is
>that some studies are poorly designed. In fact, many studies are poorly designed.
>

But you seem to imply that the study that proves the drug works is the valid study. Surely you did
not mean that.

If two studies are done...

Either

they are comparable, and then the data can be combined.

Or

they are not comparable, in which case knowing the full details of both is more information.
One of the great hidden variables in throwing around all the statistics is the assumption that
the population being treated is uniform. Of course, we all know it is not -- but that makes it
much harder.

As to studies being poorly designed, that is another issue. And the only way we know is if it is
published in full. But simply being done differently does not mean poorly designed.

The heart of the original post was the question of whether all data should be reported, in
particular whether negative data should be reported. Not reporting it creates a false record. It
seems to me appropriate that the FDA should require that all data obtained for a drug should be
properly reported. (I am not into company bashing; I spent many years as a research scientist for a
biotech company, and dealt with the frustrations of working a drug to market.)

>A case in point: Studies regarding the efficacy of antidepressants often involve counseling. They
>give counseling to the patients taking the drug, and they counsel the patients taking the placebo.
>This is part of the reason why the placebo group improves. It's not really a placebo group, since
>they got counseling.
>
>Another example: A study to compare two antidepressants drops patients from the study who get
>improvement in their depression within the first 2 weeks. The researchers are of the belief that
>antidepressants can't possibly work that quickly. But antidepressants often work within the first
>two days. The drug which works more quickly is going to come out looking worse in such a study.

I am not sure what point you are trying to make with those. Both raise interesting issues. Whatever
is done, the trials should be reported -- with full explanation of what was done.

bob

"Bob" <xyzbbruner@uclink4.berkeley.edu> wrote

> But you seem to imply that the study that proves the drug works is the valid study. Surely you did
> not mean that.

Yes I did. If a study has a p value less than 0.05, and proves that the drug works, then there is a
19 in 20 chance that the drug works.

If another study, with similar design, has a p value of 0.06, that doesn't mean that the drug
doesn't work. It just means that that study failed to prove that the drug works.

So you can't say that the study that does not prove that the drug works is something which casts
doubt on the study that proves that it does work.

> The heart of the original post was the question of whether all data should be reported, in
> particular whether negative data should be reported. Not reporting it creates a false record. It
> seems to me appropriate that the FDA should require that all data obtained for a drug should be
> properly reported. (I am not into company bashing; I spent many years as a research scientist for
> a biotech company, and dealt with the frustrations of working a drug to market.)

Yes, I would tend to agree that this would improve things. The more information we have, the better
conclusions we can draw.

Matt Beckwith <matt@doctorbeckwith.com> wrote in message = news:bdoqda02chg@enews4.newsguy.com...
> "Bob" <xyzbbruner@uclink4.berkeley.edu> wrote
>=20
> > But you seem to imply that the study that proves the drug works is =
the
> > valid study. Surely you did not mean that.
>=20
> Yes I did. If a study has a p value less than 0.05, and proves that =
the
> drug works, then there is a 19 in 20 chance that the drug works.
>=20
> If another study, with similar design, has a p value of 0.06, that =
doesn't
> mean that the drug doesn't work. It just means that that study failed =
to
> prove that the drug works.
>=20
> So you can't say that the study that does not prove that the drug =
works is
> something which casts doubt on the study that proves that it does =
work.
>=20
> > The heart of the original post was the question of whether all data should be reported, in
> > particular whether negative data should be reported. Not reporting it creates a false record. It
> > seems to me appropriate that the FDA should require that all data obtained for a drug should be
> > properly reported. (I am not into company bashing; I spent many years as a research scientist
> > for a biotech company, and dealt with the frustrations of working a drug to market.)
>=20
> Yes, I would tend to agree that this would improve things. The more information we have, the
> better conclusions we can draw.
>=20 20
Not sure what you guys are trying to prove in this thread, but you can't get to discussing the
statistical validity until after = establishing that the study's base is statistically random. =20

AFIN none of the drug co reports qualify. In ALL, the participants are pre-selected from
'applicants' and not all 'applicants' are 'eligible'. This alone will bias the end report in
whatever direction they wish it to go.

Al

I think we have discovered a new politician in the making.

"Al Hephy" <ahephy@freewweb.invalid> wrote in message news:1056979093.992320@savina...

Matt Beckwith <matt@doctorbeckwith.com> wrote in message news:bdoqda02chg@enews4.newsguy.com...
> "Bob" <xyzbbruner@uclink4.berkeley.edu> wrote
>
> > But you seem to imply that the study that proves the drug works is the valid study. Surely you
> > did not mean that.
>
> Yes I did. If a study has a p value less than 0.05, and proves that the drug works, then there is
> a 19 in 20 chance that the drug works.
>
> If another study, with similar design, has a p value of 0.06, that doesn't mean that the drug
> doesn't work. It just means that that study failed to prove that the drug works.
>
> So you can't say that the study that does not prove that the drug works is something which casts
> doubt on the study that proves that it does work.
>
> > The heart of the original post was the question of whether all data should be reported, in
> > particular whether negative data should be reported. Not reporting it creates a false record. It
> > seems to me appropriate that the FDA should require that all data obtained for a drug should be
> > properly reported. (I am not into company bashing; I spent many years as a research scientist
> > for a biotech company, and dealt with the frustrations of working a drug to market.)
>
> Yes, I would tend to agree that this would improve things. The more information we have, the
> better conclusions we can draw.
>
>
Not sure what you guys are trying to prove in this thread, but you can't get to discussing the
statistical validity until after establishing that the study's base is statistically random.

AFIN none of the drug co reports qualify. In ALL, the participants are pre-selected from
'applicants' and not all 'applicants' are 'eligible'. This alone will bias the end report in
whatever direction they wish it to go.

Al

theta@btinternet.com (Theta) wrote in message
news:<18e81488.0306281036.2949dd98@posting.google.com>...
> http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
>
> * * *
>
> Pharmaceutical industry's drug research 'biased'
>

The medical journals should actively and carefully screen out *all* the studies that are not done by
independent researchers.

If you work for a drug company your studies may be useful in the company newsletter.

TC

tcomeau <tunderbar@hotmail.com> wrote in message =
news:b550f406.0306300954.2f4cccaa@posting.google.com...
> theta@btinternet.com (Theta) wrote in message =
news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> >=20
> > * * *
> >=20
> > Pharmaceutical industry's drug research 'biased'
> >=20
>=20
> The medical journals should actively and carefully screen out *all* the studies that are not done
> by independent researchers.
>=20
> If you work for a drug company your studies may be useful in the company newsletter.
>=20
> TC

You are on the right track. But not far enough. An independent evaluation a year or so back showed a
conclusively a bias in favor of the pharmaceutical even if they only FUND the=20 study as compared
to a completely independent evaluation.

I may still have a copy of it around here somewhere and might try to find it if someone insisted. It
was totally dismaying.

Whether sub-conscious or deliberate, it is clear that the only reliable data is going to come from
fully independent reports. Getting the FDA or any other politically motivated agency to realize this
is going to be a mighty task.

Al

"tcomeau" <tunderbar@hotmail.com> wrote in message
news:b550f406.0306300954.2f4cccaa@posting.google.com...
> theta@btinternet.com (Theta) wrote in message
news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> >
> > * * *
> >
> > Pharmaceutical industry's drug research 'biased'
> >
>
> The medical journals should actively and carefully screen out *all* the studies that are not done
> by independent researchers.

Of course. We wouldn't want them to judge the studies by their merit, replicability, or peer
review, would we.

> If you work for a drug company your studies may be useful in the company newsletter.

And their merit is irrelevant? Such bias you show. Why not ignore all the conspiracy theories and
allow the normal peer review process to show what is rubbish and what might be significant? It's the
best judge we have to do this. Amateurs posing all sorts of emotional theories are pretty
misleading.

"Al Hephy" <ahephy@freewweb.invalid> wrote in message news:1057009021.956017@savina...

tcomeau <tunderbar@hotmail.com> wrote in message
news:b550f406.0306300954.2f4cccaa@posting.google.com...
> theta@btinternet.com (Theta) wrote in message
news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> >
> > * * *
> >
> > Pharmaceutical industry's drug research 'biased'
> >
>
> The medical journals should actively and carefully screen out *all* the studies that are not done
> by independent researchers.
>
> If you work for a drug company your studies may be useful in the company newsletter.
>
> TC

You are on the right track. But not far enough. An independent evaluation a year or so back showed a
conclusively a bias in favor of the pharmaceutical even if they only FUND the study as compared to a
completely independent evaluation.

I may still have a copy of it around here somewhere and might try to find it if someone insisted. It
was totally dismaying.

Whether sub-conscious or deliberate, it is clear that the only reliable data is going to come from
fully independent reports. Getting the FDA or any other politically motivated agency to realize this
is going to be a mighty task.

I wonder if we are confusing "scientific research" with drug efficacy testing. The regulator that
you elected should make the latter a valid process, I believe your regulator in the States is
perhaps not up to the job (according to the stories I hear from your countrymen). Scientific
research is self regulating.

"Muush" <waaa@waaa.waaa> wrote in message news:<yP4Ma.144$w14.3032@news-server.bigpond.net.au>...
> "tcomeau" <tunderbar@hotmail.com> wrote in message
> news:b550f406.0306300954.2f4cccaa@posting.google.com...
> > theta@btinternet.com (Theta) wrote in message
> news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> > >
> > > * * *
> > >
> > > Pharmaceutical industry's drug research 'biased'
> > >
> >
> > The medical journals should actively and carefully screen out *all* the studies that are not
> > done by independent researchers.
>
> Of course. We wouldn't want them to judge the studies by their merit, replicability, or peer
> review, would we.
>

If their peers were aware of the researchers biases and interests, I would think that they would
review them in the correct light instead of assuming that they are unbiased and simply interested in
true science, which they are obviously not.

> > If you work for a drug company your studies may be useful in the company newsletter.
>
> And their merit is irrelevant? Such bias you show.

Damned right!

> Why not ignore all the conspiracy theories and allow the normal peer review process to show what
> is rubbish and what might be significant? It's the best judge we have to do this. Amateurs posing
> all sorts of emotional theories are pretty misleading.

The normal peer review process is hamstrung by the large number of apparently unbiased studies that
are, in fact, biased. Obviously the pharmas are spending millions (if not billions, combined) to
ensure that studies show what they want them to show.

The first step to correct this is to assume that researchers are biased and expose the bias up
front. The journal must make it policy that all researchers declare their interests explicitly
and fully.

TC

"Muush" <waaa@waaa.waaa> wrote in message news:<xg5Ma.1$o74.95@news-server.bigpond.net.au>...
> "Al Hephy" <ahephy@freewweb.invalid> wrote in message news:1057009021.956017@savina...
>
> tcomeau <tunderbar@hotmail.com> wrote in message
> news:b550f406.0306300954.2f4cccaa@posting.google.com...
> > theta@btinternet.com (Theta) wrote in message
> news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> > >
> > > * * *
> > >
> > > Pharmaceutical industry's drug research 'biased'
> > >
> >
> > The medical journals should actively and carefully screen out *all* the studies that are not
> > done by independent researchers.
> >
> > If you work for a drug company your studies may be useful in the company newsletter.
> >
> > TC
>
> You are on the right track. But not far enough. An independent evaluation a year or so back showed
> a conclusively a bias in favor of the pharmaceutical even if they only FUND the study as compared
> to a completely independent evaluation.
>
> I may still have a copy of it around here somewhere and might try to find it if someone insisted.
> It was totally dismaying.
>
> Whether sub-conscious or deliberate, it is clear that the only reliable data is going to come from
> fully independent reports. Getting the FDA or any other politically motivated agency to realize
> this is going to be a mighty task.
>
> I wonder if we are confusing "scientific research" with drug efficacy testing. The regulator that
> you elected should make the latter a valid process, I believe your regulator in the States is
> perhaps not up to the job (according to the stories I hear from your countrymen). Scientific
> research is self regulating.

I think that the responsibility lies primarily with the journals themselves. It is their credibility
in the scientific community that will be (has been?) destroyed by the "scientists" that are on the
various industries payrolls.

But this is not to diminish the responsibility of the corrupt researchers themselves with no sense
of scientific propriety or personal integrity, nor does it diminish the responsibilty of the pharmas
and food industry that has a far reaching responsibility to the consumers health.

DB Allison and other "researchers" on pharma and food industry payrolls have published many, many
papers on scientific research and also many involving drug efficacy tests. Neither type of research
from this individual and his ilk should have ever been published without a complete listing of their
interests in the field. If their interests had been listed in complete detail, I am quite certain
that their peers, and the press, would all have summarily dismissed the majority of the papers, and
rightly so.

TC

"tcomeau" <tunderbar@hotmail.com> wrote in message
news:b550f406.0307012040.27423985@posting.google.com...
> "Muush" <waaa@waaa.waaa> wrote in message
news:<yP4Ma.144$w14.3032@news-server.bigpond.net.au>...
> > "tcomeau" <tunderbar@hotmail.com> wrote in message
> > news:b550f406.0306300954.2f4cccaa@posting.google.com...
> > > theta@btinternet.com (Theta) wrote in message
> > news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > > > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> > > >
> > > > * * *
> > > >
> > > > Pharmaceutical industry's drug research 'biased'
> > > >
> > >
> > > The medical journals should actively and carefully screen out *all* the studies that are not
> > > done by independent researchers.
> >
> > Of course. We wouldn't want them to judge the studies by their merit, replicability, or peer
> > review, would we.
> >
>
> If their peers were aware of the researchers biases and interests, I would think that they would
> review them in the correct light instead of assuming that they are unbiased and simply interested
> in true science, which they are obviously not.

Huh? How can this affect the replication attempts of the peers? They go on results, merit, not
personalities or wild-eyed conspiracy theories.

> > > If you work for a drug company your studies may be useful in the company newsletter.
> >
> > And their merit is irrelevant? Such bias you show.
>
> Damned right!

So what are you babbling on about? No study results are taken at face value. Every man and his
dog in the particular field is trying to show the original findings are wrong. This is how
science works.

> > Why not ignore all the conspiracy theories and allow the normal peer
review
> > process to show what is rubbish and what might be significant? It's the
best
> > judge we have to do this. Amateurs posing all sorts of emotional
theories
> > are pretty misleading.
>
> The normal peer review process is hamstrung by the large number of apparently unbiased studies
> that are, in fact, biased.

Huh? There are dozens, if not thousands of labs worldwide in each field trying to disprove ALL
findings, even their own. Far better to find the fault in your own study before your peers do. I can
remember way back a wise old head of a university department bringing young researchers to tears by
strongly pointing out their study design or result analysis errors.

>Obviously the pharmas are spending millions (if not billions, combined) to ensure that studies show
>what they want them to show.

Are you talking of scientific research, or drug registration trials?

> The first step to correct this is to assume that researchers are biased and expose the bias
> up front.

Which is what the scientific method does and has been doing for centuries.

>The journal must make it policy that all researchers declare their interests explicitly and fully.

As far as I know, they do, but this makes no difference to the peer review crucification if there is
a flaw in any published study.

"tcomeau" <tunderbar@hotmail.com> wrote in message
news:b550f406.0307012031.540955c6@posting.google.com...
> "Muush" <waaa@waaa.waaa> wrote in message
news:<xg5Ma.1$o74.95@news-server.bigpond.net.au>...
> > "Al Hephy" <ahephy@freewweb.invalid> wrote in message news:1057009021.956017@savina...
> >
> > tcomeau <tunderbar@hotmail.com> wrote in message
> > news:b550f406.0306300954.2f4cccaa@posting.google.com...
> > > theta@btinternet.com (Theta) wrote in message
> > news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > > > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> > > >
> > > > * * *
> > > >
> > > > Pharmaceutical industry's drug research 'biased'
> > > >
> > >
> > > The medical journals should actively and carefully screen out *all* the studies that are not
> > > done by independent researchers.
> > >
> > > If you work for a drug company your studies may be useful in the company newsletter.
> > >
> > > TC
> >
> > You are on the right track. But not far enough. An independent evaluation a year or so back
> > showed a conclusively a bias in favor of the pharmaceutical even if they only FUND the study as
> > compared to a completely independent evaluation.
> >
> > I may still have a copy of it around here somewhere and might try to find it if someone
> > insisted. It was totally dismaying.
> >
> > Whether sub-conscious or deliberate, it is clear that the only reliable data is going to come
> > from fully independent reports. Getting the FDA or any other politically motivated agency to
> > realize this is going to be a mighty task.
> >
> > I wonder if we are confusing "scientific research" with drug efficacy testing. The regulator
> > that you elected should make the latter a valid process, I believe your regulator in the States
> > is perhaps not up to the job
(according
> > to the stories I hear from your countrymen). Scientific research is self regulating.
>
> I think that the responsibility lies primarily with the journals themselves. It is their
> credibility in the scientific community that will be (has been?) destroyed by the "scientists"
> that are on the various industries payrolls.

The peer review process involves peers from all areas, so it doesn't matter where the scientists who
do the bench work derive their remuneration or grant money. I suspect you are confusing "science"
with drug certification applications.

> But this is not to diminish the responsibility of the corrupt researchers themselves with no sense
> of scientific propriety or personal integrity, nor does it diminish the responsibilty of the
> pharmas and food industry that has a far reaching responsibility to the consumers health.

How can "corrupt" researchers publish and survive the peer review process?

> DB Allison and other "researchers" on pharma and food industry payrolls have published many, many
> papers on scientific research and also many involving drug efficacy tests. Neither type of
> research from this individual and his ilk should have ever been published without a complete
> listing of their interests in the field. If their interests had been listed in complete detail, I
> am quite certain that their peers, and the press, would all have summarily dismissed the majority
> of the papers, and rightly so.

You obviously don't understand the peer review process. Other researchers must be able to replicate
the studies published. No journal wants to publich something that will be shot down in flames the
next day. Your conspiracy theories could only be valid if there were no peer review.

tunderbar@hotmail.com (tcomeau) wrote in message
news:<b550f406.0307012031.540955c6@posting.google.com>...
> "Muush" <waaa@waaa.waaa> wrote in message news:<xg5Ma.1$o74.95@news-server.bigpond.net.au>...
> > "Al Hephy" <ahephy@freewweb.invalid> wrote in message news:1057009021.956017@savina...
> >
> > tcomeau <tunderbar@hotmail.com> wrote in message
> > news:b550f406.0306300954.2f4cccaa@posting.google.com...
> > > theta@btinternet.com (Theta) wrote in message
> news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > > > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> > > >
> > > > * * *
> > > >
> > > > Pharmaceutical industry's drug research 'biased'
> > > >
> > >
> > > The medical journals should actively and carefully screen out *all* the studies that are not
> > > done by independent researchers.
> > >
> > > If you work for a drug company your studies may be useful in the company newsletter.
> > >
> > > TC
> >
> > You are on the right track. But not far enough. An independent evaluation a year or so back
> > showed a conclusively a bias in favor of the pharmaceutical even if they only FUND the study as
> > compared to a completely independent evaluation.
> >
> > I may still have a copy of it around here somewhere and might try to find it if someone
> > insisted. It was totally dismaying.
> >
> > Whether sub-conscious or deliberate, it is clear that the only reliable data is going to come
> > from fully independent reports.

Cochrane

> > Getting the FDA or any other politically motivated agency to realize this is going to be a
> > mighty task.
> >
> > I wonder if we are confusing "scientific research" with drug efficacy testing. The regulator
> > that you elected should make the latter a valid process, I believe your regulator in the States
> > is perhaps not up to the job (according to the stories I hear from your countrymen). Scientific
> > research is self regulating.
>
> I think that the responsibility lies primarily with the journals themselves. It is their
> credibility in the scientific community that will be (has been?) destroyed by the "scientists"
> that are on the various industries payrolls.

Are you a philanthropic scientist - one who works exclusively for the intrinsic bvalue of the
pursuit or are you paid for your work?

> But this is not to diminish the responsibility of the corrupt researchers themselves with no sense
> of scientific propriety or personal integrity,

So any science done for pay is be definition only done by currupt and unethical psuedo-scientists?

> nor does it diminish the responsibilty of the pharmas and food industry that has a far reaching
> responsibility to the consumers health.
>
> DB Allison and other "researchers" on pharma and food industry payrolls have published many, many
> papers on scientific research and also many involving drug efficacy tests. Neither type of
> research from this individual and his ilk should have ever been published without a complete
> listing of their interests in the field. If their interests had been listed in complete detail, I
> am quite certain that their peers, and the press, would all have summarily dismissed the majority
> of the papers, and rightly so.

Perhaps you'd like to provide an example or two or three of Allison's work and then provide a
scientific assessment of the works deficiencies. I haven't found anything with DB Allison as the
lead author that reflects a drug efficacy trial.

js

"Muush" <waaa@waaa.waaa> wrote in message news:<Q6wMa.182$gw4.3809@news-server.bigpond.net.au>...
> "tcomeau" <tunderbar@hotmail.com> wrote in message
> news:b550f406.0307012040.27423985@posting.google.com...
> > "Muush" <waaa@waaa.waaa> wrote in message
> news:<yP4Ma.144$w14.3032@news-server.bigpond.net.au>...
> > > "tcomeau" <tunderbar@hotmail.com> wrote in message
> > > news:b550f406.0306300954.2f4cccaa@posting.google.com...
> > > > theta@btinternet.com (Theta) wrote in message
> news:<18e81488.0306281036.2949dd98@posting.google.com>...
> > > > > http://groups.msn.com/psychbusters Decoding Psychiatric Propaganda
> > > > >
> > > > > * * *
> > > > >
> > > > > Pharmaceutical industry's drug research 'biased'
> > > > >
> > > >
> > > > The medical journals should actively and carefully screen out *all* the studies that are not
> > > > done by independent researchers.
> > >
> > > Of course. We wouldn't want them to judge the studies by their merit, replicability, or peer
> > > review, would we.
> > >
> >
> > If their peers were aware of the researchers biases and interests, I would think that they would
> > review them in the correct light instead of assuming that they are unbiased and simply
> > interested in true science, which they are obviously not.
>
> Huh? How can this affect the replication attempts of the peers? They go on results, merit, not
> personalities or wild-eyed conspiracy theories.
>
> > > > If you work for a drug company your studies may be useful in the company newsletter.
> > >
> > > And their merit is irrelevant? Such bias you show.
> >
> > Damned right!
>
> So what are you babbling on about? No study results are taken at face value. Every man and his
> dog in the particular field is trying to show the original findings are wrong. This is how
> science works.
>
> > > Why not ignore all the conspiracy theories and allow the normal peer
> review
> > > process to show what is rubbish and what might be significant? It's the
> best
> > > judge we have to do this. Amateurs posing all sorts of emotional
> theories
> > > are pretty misleading.
> >
> > The normal peer review process is hamstrung by the large number of apparently unbiased studies
> > that are, in fact, biased.
>
> Huh? There are dozens, if not thousands of labs worldwide in each field trying to disprove ALL
> findings, even their own. Far better to find the fault in your own study before your peers do. I
> can remember way back a wise old head of a university department bringing young researchers to
> tears by strongly pointing out their study design or result analysis errors.
>
> >Obviously the pharmas are spending millions (if not billions, combined) to ensure that studies
> >show what they want them to show.
>
> Are you talking of scientific research, or drug registration trials?
>
> > The first step to correct this is to assume that researchers are biased and expose the bias up
> > front.
>
> Which is what the scientific method does and has been doing for centuries.
>
> >The journal must make it policy that all researchers declare their interests explicitly
> >and fully.
>
> As far as I know, they do, but this makes no difference to the peer review crucification if there
> is a flaw in any published study.

You guys are the sorriest bunch of naive bastards I've seen in years.

TC