Most scientific papers are probably wrong

[Darth Wong]

I'm not saying Merck was being a paragon of virtue, I'm saying they had the opt of going public and covering their asses or hunkering down and waiting for the problem to eventually bite them in the ass. They chose the former, not because they were altruistic, but because the suits thought it would have a smaller negative impact on stock price.

That's because the problem was now short-term, and the risk imminent. How does that contradict anything I've been saying about their mentality? How does it contradict any of the studies showing that highly questionable practises are common in clinical trials? How does it contradict my statement that self-conducted research with no independent verification is not good science no matter how many times you say it is?

<snip lots of claims about Merck meeting FDA requirements and doing everything in their power to protect the public>

Funny that Dr. David Graham, the Associate Director for Science and Medicine in the Office of Drug Safety at the FDA, has a completely different attitude about this. He testified before the Senate Finance Committee that the FDA was not structurally designed to prevent major problems such as Vioxx, and that people in the FDA have often approved drugs that they know to have serious safety problems because they primarily serve the drug industry rather than public safety. If the safety risks were so difficult to ascertain, how did the drug end up causing tens of thousands of deaths according to the FDA?

He also seems to disagree about the impossibility of knowing any earlier about the risks. He testified about studies showing a 500% increase in heart attack risk that were conducted way back in 2000. And about your bullshit claim that Merck did a third safety study that it didn't have to do:

There have been a few reports since 2000 suggesting that there may be heart problems associated with Vioxx. Merck and the Food and Drug Administration have been monitoring the situation. Today's surprise announcement comes from a study Merck was doing for a totally different reason. It was looking at 2,600 patients to see if Vioxx prevented such problems as colon polyps. The findings of an increased risk for heart problems were discovered by scientists monitoring safety in the study.

Yet again, your argument is based on massive distortions of the facts. Merck did not do a third safety trial of their own volition: they were attempting to find new markets for the drug and accidentally made an uncomfortable discovery. And this "third study" was not the first news of these risks; these concerns had been raised four years before Merck pulled the drug from the market.

You can pretend that Merck is getting its ass kicked in court just because of stupid juries if you want, but the fact is that they did drop the ball, they did downplay risks, they did not try to conduct further safety studies even after independent studies suggested very serious public safety risks, and you are, once again, full of shit.

You originally claimed the companies stood to lose millions if their science didn't get the drug to market, I replied that they stood to lose billions when bad science bit them in the ass.

And I've since provided a fair bit of evidence that this does, in fact, happen, despite your claims to the contrary. Because your argument still hinges upon an assumption of business ethics and long-term thinking which does not necessarily exist in the boardroom. Not to mention the fact that you totally ignored my point about the lack of independent verification of results, which is textbook bad science.

generic drug manufacturers have to do their own independent clinical safety trials in order to do an ANDS in Canada (they only get to skip the efficacy trials for the active ingredient), so the cost of safety trials can't be that high or their drugs wouldn't be so damned cheap.

They're not. The average Canadian generic drug is 70% more expensive than the average American generic drug.

Does this stop the American pharmaceutical industry from howling about cheap Canadian generics? No.

[Darth Wong]

By the way, to address tharkun's argument that no pharmaceutical corporation would be stupid enough to risk major lawsuits for negligence, I would like to point out that prior to Enron, he would have undoubtedly argued that no corporation would be stupid enough to risk prosecution for incredibly blatant fraud either. Corporate suits have a history of short-term thinking and arrogant, foolish decisions, and then trying to use their influence to cover up for those decisions. Why do you think there's such a powerful political panic to shut down liability lawsuits even though the health care industry's liability payouts are not rising?

For the record the greatest arguments have NOT been about drug company lawsuits but rather about small and medium-practice doctors being sued in malpractice lawsuits. The argument (and I have no statistics to know if it is true or not) is that the rapidly rising cost of malpractice insurance is driving these doctors (who remain the first line of defense in preventing illness for most folks) out of buisness because they simply cannot afford the cost of malpractice insurance without charging exorbidant rates from their patients. Again that's the reasoning for the rush to limit medical liability lawsuits not liability from the drug companies (though they are certainly trying to get onboard).

And those arguments would be false too. Medical malpractice insurance industry payouts have been flat for the last five years, even though medical malpractice insurance premiums have skyrocketed over that same period. The insurance companies are fucking doctors up the ass and blaming liability lawyers. And people are falling for it because for some goddamned incredible reason, people keep giving big corporations the benefit of the doubt.

[tharkûn]

And those arguments would be false too. Medical malpractice insurance industry payouts have been flat for the last five years, even though medical malpractice insurance premiums have skyrocketed over that same period.

On average the premium price was playing catch-up over those years. Likewise the insurance companies used stock investments to hold down premium hikes, with lower average returns in the last five years they have been unable to do that.

Another fun thing is that the big cost isn't the payouts themselves, but the practice of defensive medicine. It costs the doctor, himself, virtually nothing to opt for the more expensive and unneeded procedure (like say a ceasarian delivery) in cases where it isn't warranted; however if he takes the more cost effective road (and in the case of ceasarians, the safer one) it can bite him in the ass (the headache is terrible if nothing else). Far more money is burnt avoiding the lawsuits in the first place because medical practitioners have few incentives not to practice defensive medicine than will ever be paid out in BS malpractice.

That's because the problem was now short-term, and the risk imminent. How does that contradict anything I've been saying about their mentality? How does it contradict any of the studies showing that highly questionable practises are common in clinical trials? How does it contradict my statement that self-conducted research with no independent verification is not good science no matter how many times you say it is?

The entire pharmaceutical industry is long term. When you fudge the data you are years away from a profit. When you decide to no matter how badly you wish for the suits to be comic book characters they are not both able to weigh long term benifits from fudging the data today, but unable to to weigh the costs of data tommorrow. Rational players DO have a finicial incentive to get good science.

How does that contradict the studies showing questionable practices? It doesn't. A good part of the problem is that both sides (private and regulatory) don't have good solid scientific standards and that your studies are dipping back in history.

Funny that Dr. David Graham, the Associate Director for Science and Medicine in the Office of Drug Safety at the FDA, has a completely different attitude about this. He testified before the Senate Finance Committee that the FDA was not structurally designed to prevent major problems such as Vioxx, and that people in the FDA have often approved drugs that they know to have serious safety problems because they primarily serve the drug industry rather than public safety. If the safety risks were so difficult to ascertain, how did the drug end up causing tens of thousands of deaths according to the FDA?

Let's go over this again. The damage Vioxx does is cumulative, the longer you take the drug the worse off you are. It takes longer than the six months of typical use studied in the clinicals for the cumulative damage to show. Monitoring for chronic side effects requires lots of patients and long periods of study. Vioxx was put on fast track because it prevented many cases of GI toxicity, which also kills. The FDA though, based off animal data, that chronic effects would be minimal and that more lives would be saved by getting to market faster than the small chance of chronic problems. They were wrong.

Another fun point is that Vioxx was prescribed 105 million times to 20 million patients in the US. With 10,000 deaths that works out to 1 patient in 2,000 will have this type of reaction. That means that in a clinical trial of 8,000 patients you could expect 4 patients to eventually die who otherwise wouldn't have. Needless to say truncation artifacts will be present in the statistics.

So between this effecting "only" 1 in 2,000 patients and the need for chronic use to build up cumulative damage even good science would most likely miss the risk with a six month study. Even duplicating the trial (resulting in 41% better data) with an independent agency would still most likely have missed it.

He also seems to disagree about the impossibility of knowing any earlier about the risks. He testified about studies showing a 500% increase in heart attack risk that were conducted way back in 2000.

The initial six month trial happened before 2000. Vioxx was declared safe by the FDA in 1999. Merck brought forward the results of VIGOR, the previously mentioned trial in 2000 and the FDA looked over the data in early 2001. His opinion turned out to be correct, however his opinion was in the minority when the independent experts at the FDA reviewed the data.

Yet again, your argument is based on massive distortions of the facts. Merck did not do a third safety trial of their own volition: they were attempting to find new markets for the drug and accidentally made an uncomfortable discovery.

So? Clinical trials are expensive things and if you are going to do another one you may as well look for new markets to offset the cost. Likewise if you are going to do a clinical to look for new markets it behooves you to do a safety test as well. There is no reason not to get the data for both purposes.

And this "third study" was not the first news of these risks; these concerns had been raised four years before Merck pulled the drug from the market.

No, like I told you the first study was done and the FDA found no reason not to approve the drug (it did however see several reason to do so). VIGOR started after FDA approval and was completed in 2000, subsequent publication in NEJM followed. In early 2001 the FDA officially recommends CV warnings be issued to phsycians. Only after that did we get a three year study started that Merck flushed around eighteen months and withdrew Vioxx from the market. Oh here is a little tidbit you did leave out - this Febuary an independent advisory panel of doctors with the FDA reccommends returning Vioxx to the market with more stringent guidelines about dosage and CV warnings.

On a side note this same federal inquiry reccomends long overdue CV warnings for the older over the counter meds who kill in excess of 10,000 patients per annum.

You can pretend that Merck is getting its ass kicked in court just because of stupid juries if you want, but the fact is that they did drop the ball, they did downplay risks, they did not try to conduct further safety studies even after independent studies suggested very serious public safety risks

Merck released its data to the FDA, who has NEVER concluded that Vioxx is too dangerous for the market. It was a stupid jury, they convicted a rifleman for stabbing a man to death.

And I've since provided a fair bit of evidence that this does, in fact, happen, despite your claims to the contrary. Because your argument still hinges upon an assumption of business ethics and long-term thinking which does not necessarily exist in the boardroom. Not to mention the fact that you totally ignored my point about the lack of independent verification of results, which is textbook bad science.

Business ethics, nope. They can be as unethical as all hell in the boardroom, a 10% loss of equity still points to shoddy science being a bad business decision. I do require them to think longterm, of course in a business which has ten and fifteen year finicial plans that seems reasonable. All the drugs we are talking about today are results of business decisions in the late 80s and early 90s. I am curious if you could provide an example of what you consider to be evidence of long term thinking.

Does this stop the American pharmaceutical industry from howling about cheap Canadian generics? No.

Canadian generics are NOT cheaper.

Pharma bitches about your cheaper brand name drugs, not your generics. Pharma bitches about the higher cost of selling generics in Canada. The only reason the major pharma players (who predominately are not in the generic business) give a damn about your generics is because they want to prevent your brand names from being reimported and like to paint all Canadian medicine as potentially dangerous; of course only if crosses the border.

The point is drug trials cost serious money. With something like death you need thousands of people willing to risk their lives. When looking for chronic effects you need them pay for six to ten times longer. If you wanted to be sure of picking up that 1/2000 chance of having another Vioxx you would need to increase the trial sizes by an order of magnitude; you had have to increase their length by another order of magnitude. Depending on the thresholds you set, this could mean as much as two orders of magnitude higher clinical costs which will mean millions of dollars and thousands of lives lost while you wait the extra years for side effects to be found (assuming you don't let potentially life saving drugs have an expedited process like Vioxx went through).

[Darth Wong]

Oh for fuck's sake, your entire argument is a half-dozen different iterations of the same basic theme, which is:

1) The FDA is even more irresponsible than Merck
2) It would be astronomically expensive to discover these risks

The first argument is totally irrelevant to the argument that Merck is irresponsible; Dr. David Graham admitted before a Senate hearing that the FDA routinely approves drugs that it has serious safety concerns about. And the second argument is simply false; you mentioned the VIGOR study yourself; the fact that both Merck and the FDA continued to have no problem with this drug after the release of this study shows precisely how flawed the science behind drug approval and marketing is. And don't tell me that a 500% increase in the chance of death was not something that merited immediate response, or that Merck's study was due to the VIGOR study (according to the BBC, it was commissioned to find new markets for the drug, not to revisit the already-approved usage and safety).

[tharkûn]

1) The FDA is even more irresponsible than Merck

What are your standards for responsibility. What level of confidence do you think is appropriate (i.e. side effects hitting 1 in 10,000 people are demonstrated not to be in effect with 95% certainty)?

2) It would be astronomically expensive to discover these risks

Astronomical? Depends on your definition. It will be EXPENSIVE and TIME CONSUMING and PEOPLE WILL DIE while they wait for some drugs to come to market. There is a tradeoff between protecting the public through better screening and giving them better medicine through more timely introduction and more new pharmaceuticals.

And the second argument is simply false; you mentioned the VIGOR study yourself; the fact that both Merck and the FDA continued to have no problem with this drug after the release of this study shows precisely how flawed the science behind drug approval and marketing is.

They didn't have no problem. They put out specific CV contradictions which had they been followed by the prescribing doctors would have averted most of the deaths.

And don't tell me that a 500% increase in the chance of death was not something that merited immediate response

What do you propose to do about aspirin? It kills more people BY FAR. So does Tylenol. So does Ibuprofen.

or that Merck's study was due to the VIGOR study (according to the BBC, it was commissioned to find new markets for the drug, not to revisit the already-approved usage and safety).

Virtually all clinical trials have a safety component. The fact is while finding a new market they were revisiting the safety of common usage.

[Darth Wong]

What are your standards for responsibility.

Why don't you ask Dr. David Graham, the FDA director who says that the FDA's standards are far too low and testified to that effect in front of a Senate hearing? I take his word to be more authoritative on the matter than yours.

Astronomical? Depends on your definition.

Since the VIGOR study found these safety flaws years before the drug was pulled from the market, I will take your claims of the impossibility of knowing these hazards for what it's worth: absolutely nothing.

[tharkûn]

Why don't you ask Dr. David Graham, the FDA director who says that the FDA's standards are far too low and testified to that effect in front of a Senate hearing? I take his word to be more authoritative on the matter than yours.

I'm asking you:
What are your standards for responsibility. What level of confidence do you think is appropriate (i.e. side effects hitting 1 in 10,000 people are demonstrated not to be in effect with 95% certainty)?

I'm well aware of your opinion that the FDA is too irresponsible. I'm also aware of Dr. Graham's positions, some of which I agree with. However you have consistently claimed people aren't responsible and do nothing more than dismiss the costs of increased safety out of hand. Fine, quantify your position.

If we want to play the appeal to authority game, then should I just pull Dr. Kweder's testimony in response to Dr. Graham?

Since the VIGOR study found these safety flaws years before the drug was pulled from the market, I will take your claims of the impossibility of knowing these hazards for what it's worth: absolutely nothing.

VIGOR found reason for concern. VIGOR pitted Vioxx against Naproxen. Naproxen has known CV benifits. The question from VIGOR was it Vioxx that was killing more patients or was it Naproxen saving more lives? If the former then you restrict Vioxx, if the later then you treat the patients with something to receive the benifits of Naproxen as well as those of Vioxx. No definitive answer could be found, they issued CV warnings to doctors and still have not declared Vioxx unsafe.

When Merck ran the three year study it was a double blind against placebo (you know the most scientificly sound trial) for months the death disparity wasn't significant in the stastical sense. Around eighteen months the data showed a significant increase in risk of death against placebo. There is a reason why standard safety trials run against placebo, and why Merck withdrew in light of this more conclusive data and not VIGOR.

Frankly Mike your running away here. You call for an "immediate response" with increased risk of death. I then ask what you think should be done about aspirin, tylenol, and ibuprofen which have all killed more people than Vioxx, you fail to reply. I ask you to quantify your position on what responsible behaviour entails, you appeal to the authority of Dr. Graham, whom I do not recall quantifying such a position either. Buck up and delineate a hard, quantified level of safety you think is acceptable.

[Justforfun000]

Hehehe. One thing I can give you Tharkun, you stand your ground.

*jumps out of the path of flames*

[Darth Wong]

I'm asking you:
What are your standards for responsibility. What level of confidence do you think is appropriate (i.e. side effects hitting 1 in 10,000 people are demonstrated not to be in effect with 95% certainty)?

Since I'm not as qualified as Dr. Graham, why should I have to produce my own independent standards? Certainly, for automotive purposes, if one in two thousand cars killed the customer, we would consider it totally unacceptable. So you are obviously not interested in my criteria.

I'm well aware of your opinion that the FDA is too irresponsible. I'm also aware of Dr. Graham's positions, some of which I agree with. However you have consistently claimed people aren't responsible and do nothing more than dismiss the costs of increased safety out of hand. Fine, quantify your position.

A 500% increase of heart-attack risk is more than enough. That doesn't satisfy you? Too fucking bad. You're just stalling for time and playing games; neither you or I as individuals has the credentials to be determining such things; standards are set by committees and it is not something that somebody sits down in his basement and rattles off, regardless of his education. But if there is widespread doubt within the regulatory agencies that the standards are strict enough, then to put it simply, the standards are not strict enough. And when you have an FDA director testifying that a lot of people in the organizations have misgivings about some of their drug approvals, that is more than enough evidence to indicate that something is wrong.

If we want to play the appeal to authority game, then should I just pull Dr. Kweder's testimony in response to Dr. Graham?

What difference would it make? When in doubt, responsible people err on the side of public safety, not on the side of industry convenience. Merck didn't do that, and you are arguing that we shouldn't either.

When Merck ran the three year study it was a double blind against placebo (you know the most scientificly sound trial)

Wrong. The most scientifically sound trial is one which has undergone independent verification of results. NO medical trials are required to contain an independent verification component, yet their results are considered so authoritative that they are deemed final and binding for the purpose of regulatory approval. THAT IS BAD SCIENCE. In any other field of science, a study, no matter how carefully conducted, is not considered authoritative alone; it must be verified. I have made this point several times now, to deafening silence from you. It speaks directly to the thread subject, whereas you keep trying to distract from that same subject. So please, answer the fucking point, you deceptive little ass-wipe.

Frankly Mike your running away here.

That's rich, coming from a lying little asstard who has been trying to change the subject from "is this bad science" to "is this legal".

You call for an "immediate response" with increased risk of death. I then ask what you think should be done about aspirin, tylenol, and ibuprofen which have all killed more people than Vioxx, you fail to reply.

Because I ran out of time for the post and had to leave. Or didn't you notice that I went off-line for many, many hours after that? But as long as you're going to be a sophistic asshole about it, why don't you tell me whether these casualties were caused by proper or improper use? Single-use or combinations? And if so, why don't you tell me why this should mitigate the point about questionable science and attention to consumer safety at all?

Here's a quote about ibuprofen:

The researchers found that those taking ibuprofen were almost twice as likely as those taking aspirin alone to die by 1997. That meant that for every 1,000 patients treated, there were 12 extra deaths a year when ibuprofen was taken with aspirin.

For heart-related deaths, ibuprofen was linked to three extra deaths per 1,000 patients treated per year.

Wyeth Consumer Healthcare, the maker of Advil, said "this study does not provide a definitive answer" to whether ibuprofen interferes with the effects of aspirin.

Wyeth spokesman Fran Sullivan argued that recent clinical research papers "suggest that the use of ibuprofen at the same time as low-dose aspirin does not alter the risk from heart disease."

You were hoping to use Ibuprofen as an example against my argument that the industry is not sufficiently concerned with public safety?

Any time you're talking about something that doubles the risk of a problem, it's certainly not something that should be immediately brushed off, at least not without further study. And yet ... guess what they're doing!

[Trytostaydead]

It's cited from PubMed. It's obviously based on medical studies, and we've all known for a long time that medical studies are heavily influenced by nonscientific factors and poor controls. Hell, the fact that they're generally paid for by the company which stands to lose millions of dollars if they don't produce the desired results is rather damning evidence against their integrity already.

Pubmed is the NIH archive, meaning just about ANYTHING somehow relating to humans will be archived there in an accessible databse to researchers all over the world. You may also want to check up on what "medical studies" actually are and how broad of a definition that term is, not to mention your statement regarding funding.

Odds are if you published in anything that could remotely be possibly related to humans or of human interest, you'll find yourself in pubmed. Pubmed is one of the preimere archives for researchers to go to. It is up to the researcher to determine how valid the study is. Oftentimes that means checking what journal that article was published in. Who the author is. What institute they came from (usually meaning which university are they affiliated with). How many people have cited their paper already? How many citations did they use. Has anyone reproduced their data in some way?

[Darth Wong]

Pubmed is the NIH archive, meaning just about ANYTHING somehow relating to humans will be archived there in an accessible databse to researchers all over the world.

How does that change the point that if it's in PubMed, it's more likely to relate to medicine?

[Nova Andromeda]

Pubmed is the NIH archive, meaning just about ANYTHING somehow relating to humans will be archived there in an accessible databse to researchers all over the world.

How does that change the point that if it's in PubMed, it's more likely to relate to medicine?

--Pubmed indexs nearly all the research journals that I've ever searched for (genetics, biochemistry, and EM are just some of the fields I play in), and I wouldn't be surprised if half of all research articles turned out to be wrong or flawed in some way. There is a lot of crappy science out there, but in the end it gets overturned (if it's worth consideration in the first place). In addition, those people publishing bad science (as opposed to overlooking something subtle, etc.) hurt their career and aren't taken seriously over the long term.

[Trytostaydead]

Pubmed is the NIH archive, meaning just about ANYTHING somehow relating to humans will be archived there in an accessible databse to researchers all over the world.

How does that change the point that if it's in PubMed, it's more likely to relate to medicine?

The point is, Pubmed is an amazing tool for researchers to use, and IS used all over the country and internationally as well. The works archived through pubmed is not a reflection upon its integrity.

Furthermore, the statement that medical studies are heavily influenced, poor controls, etc and "the fact that they're generally paid for by the company.." is quite an erroneous statement. Yes, absolutely, pharmaceutical companies have been known to put out data which only shows their product in a good light (which is why the FDA is now making them release everything), but that is the company and their research. But how does that "generally" apply to a researcher working say at UCLA mapping out the human genome? Or someone stimulating a nerve in a basement lab at some university research facility? Or a doctor compiling a list of followups after a particular surgery? Or someone mapping out the genetic factors for say cystic fibrosis? Are these medical studies heavily influenced by nonscientific factors and poor controls? Or controlled by large companies afraid to lose millions?

[Darth Wong]

The point is, Pubmed is an amazing tool for researchers to use, and IS used all over the country and internationally as well. The works archived through pubmed is not a reflection upon its integrity.

Since the integrity of PubMed is not the subject of this thread, why is this relevant?

Furthermore, the statement that medical studies are heavily influenced, poor controls, etc and "the fact that they're generally paid for by the company.." is quite an erroneous statement. Yes, absolutely, pharmaceutical companies have been known to put out data which only shows their product in a good light (which is why the FDA is now making them release everything), but that is the company and their research.

Conceded, I should have said "pharmaceutical studies" instead of "medical studies".

[Trytostaydead]

The point is, Pubmed is an amazing tool for researchers to use, and IS used all over the country and internationally as well. The works archived through pubmed is not a reflection upon its integrity.

Since the integrity of PubMed is not the subject of this thread, why is this relevant?

Because the original statement I was quoting suggested that since the article is cited from PubMed is "obviously based on medical studies" and therefore has some "rather damning evidence against their integrity already." The fact that it or any article from PubMed is not a reflection on its validity, bias, etc, but the fact it was published in a readily available journal that the NIH will archive it.

[Darth Wong]

The point is, Pubmed is an amazing tool for researchers to use, and IS used all over the country and internationally as well. The works archived through pubmed is not a reflection upon its integrity.

Since the integrity of PubMed is not the subject of this thread, why is this relevant?

Because the original statement I was quoting suggested that since the article is cited from PubMed is "obviously based on medical studies" and therefore has some "rather damning evidence against their integrity already." The fact that it or any article from PubMed is not a reflection on its validity, bias, etc, but the fact it was published in a readily available journal that the NIH will archive it.

Oh for fuck's sake, can't you read? I wasn't saying that everything in PubMed is wrong; I was saying that here's a potential explanation for how this guy could say that so many scientific papers are wrong: based on the venue in which his paper is being published, he's working in a field where that may actually be the case.

[Trytostaydead]

Oh for fuck's sake, can't you read? I wasn't saying that everything in PubMed is wrong; I was saying that here's a potential explanation for how this guy could say that so many scientific papers are wrong: based on the venue in which his paper is being published, he's working in a field where that may actually be the case.

Ah, my bad. I misconstrued what you wrote.

[tharkûn]

Since I'm not as qualified as Dr. Graham, why should I have to produce my own independent standards?

Dr. Graham never produced these standards either, so far as I know. I'm asking for an objective, quantifiable standard you think is good.

Certainly, for automotive purposes, if one in two thousand cars killed the customer, we would consider it totally unacceptable. So you are obviously not interested in my criteria.

If you think drugs which kill 1 in 2000 patients is a good objective and quantifiable standard, then that is fine. All I ask now is what type of confidence level you want with that, assuming you think that is a fair objective standard for all drugs. Do I need to be 95% confident that my drug doesn't kill 1 in 2,000 patients, or 99% or higher?

A 500% increase of heart-attack risk is more than enough. That doesn't satisfy you?

In a drug comparison study? No. If I compare Aspirin against Tylenol I'm not just looking at which pain releiver is safer. I'm also including Aspirin's blood thinning effects, etc. A 500% risk against placebo would certainly be indicative, but against a drug which is already proven to save lives, no it is not enough. That tells me that my drug is either killing people or the other drug is saving people my drug doesn't. It would definately be cause for another look at safety - either a new clinical against placebo or a review of patient outcomes if the drug has a long enough history to go that route.

But if there is widespread doubt within the regulatory agencies that the standards are strict enough, then to put it simply, the standards are not strict enough. And when you have an FDA director testifying that a lot of people in the organizations have misgivings about some of their drug approvals, that is more than enough evidence to indicate that something is wrong.

Not argueing there. However your point was that the drug company's science was bad. Dr. Graham doesn't dispute the numbers coming out of Merck, indeed he uses them extensively. He argues that the FDA sets the bar too low and that the requisite data Merck provided was misinterpreted.

Because I ran out of time for the post and had to leave. Or didn't you notice that I went off-line for many, many hours after that? But as long as you're going to be a sophistic asshole about it, why don't you tell me whether these casualties were caused by proper or improper use? Single-use or combinations? And if so, why don't you tell me why this should mitigate the point about questionable science and attention to consumer safety at all?

You made a reply, if you haven't the time to reply to everything, then simply take the few seconds to post to that effect when you reply.

Let's start with aspirin for specifics:
1. About 500 aspirin related deaths are intentional or accidental overdoses, the rest are attributable to renal failure, GI distress, allergy, etc.
2. Aspirin is always in combination. It shows interaction with vitamin C and several other dietary components. Only the exceptionally rare cases where the victim was fasting prior to death will show a classical single-use response.
3. I will tell you about what mitigates these deaths, after you tell me what you think the appropriate response of a drug that kills as many people as aspirin (currently it is sold like candy).

You were hoping to use Ibuprofen as an example against my argument that the industry is not sufficiently concerned with public safety?

What do you think should be done about Ibuprofen. I know the risks, I even know the backstory and some of the politics, my question is GIVEN these risks what action would be responsible on the part of Ibuprofen manufacturers and the FDA?

Any time you're talking about something that doubles the risk of a problem, it's certainly not something that should be immediately brushed off, at least not without further study. And yet ... guess what they're doing!

In the US system once a drug is considered safe it is the government's responsibility to prove it is no longer safer.

I whole heartedly concur with your call for further study, as did the scientists running the third trial for Vioxx, who tried to do just that from the beginning (having some small bits of human decency afterall).

There are medical reasons the FDA continues to advocate for the return of Vioxx to the market and economic ones why Merck doesn't wish to do so.

[kaikatsu]

I know jack all about medical research, but I know a small bit about law and economics, and I will say this: An unethical, immortal person who is trying to make the most money out of his company through unethical practices could possibly do the following.

1.) Try to get plausable denyability of CRIMINAL behaviour through research. It's not about if the drug works, it's demonstrating that "we didn't know." This is remarkably easy to do. As I'm sure Mike can attest, a mechanical engineering failure can take years to sort out just what wasn't factored in -- the human body is INFINITELY more complex.

2.) Launch the drug, watch stock price soar if the drug takes off.

3.) At first sign of bad news, bail and sell while stock price is still soaring, while leaving enough of a window that no one can claim insider trading (something which is not easy to proove, the Martha Stewart fiasco excepted)

4.) Shrug as the crash hurts INVESTORS. The board rides the wave. They saw it coming. Buy back the now reduced stock price. Repeat.

I'm not saying EVERY drug company will do this, but investment scams are easy. If the drug works? Hey, they still made good money. If it crashes? No problem, THEY knew what was coming before anyone else did.

tharkûn, your notion that "they are too greedy to worry about a suit" is flawed because a suit can only hit a -company-. An insider with a good accountant and solid CYA principles can easily get money out of this kind of cycle.

As an alternate example, do you really think the people on the top know what's going on? A middleman researcher who has blown sixty million dollars on a new drug and needs to explain where all that money went is going to want to pull strings to get positive research done, and he might not own a whit of stock in the company.

Taking that into account with Mike's data, and I think you've got a pretty clear example that a.) the threat of lawsuits don't always have to deter the people in charge and b.) lawsuits AREN'T detering the people in charge.

[tharkûn]

I know jack all about medical research, but I know a small bit about law and economics, and I will say this: An unethical, immortal person who is trying to make the most money out of his company through unethical practices could possibly do the following.

Okay cite an example of when that was ever done in the drug industry. All trades by executive officers are tracked and it should be trivial to show them selling out before impending trouble.

4.) Shrug as the crash hurts INVESTORS. The board rides the wave. They saw it coming. Buy back the now reduced stock price. Repeat.

Who hires the board? INVESTORS. Repetitively soaking your investors is not practical. They will fire your ass and replace you. Even the most criminal Enron only soaked their investors once. Again I'd like to repeat my challenge for evidence.

I'm not saying EVERY drug company will do this, but investment scams are easy. If the drug works? Hey, they still made good money. If it crashes? No problem, THEY knew what was coming before anyone else did.

Okay show me the proof. Executives bailing before impending doom is trivial to prove. Frankly though such a course of action risks both civil and criminal liability.

tharkûn, your notion that "they are too greedy to worry about a suit" is flawed because a suit can only hit a -company-. An insider with a good accountant and solid CYA principles can easily get money out of this kind of cycle.

Show me three drug insiders who did.

As an alternate example, do you really think the people on the top know what's going on? A middleman researcher who has blown sixty million dollars on a new drug and needs to explain where all that money went is going to want to pull strings to get positive research done, and he might not own a whit of stock in the company

Yes I do. Normally a drug that reaches clinicals has a VP with direct access. Further by the time you reach human trials it isn't sixty million, it is in the hundred millions and the guys running the show switch off (your drug discovery people are different than your clinical guys). Frankly the people at the front end have already done their job and expect between 9/10 and 19/20 of their efforts to be in vain. The people at the back end look at the preceding costs as sunk, which they are. Immediate concerns of the people doing the research are not to BS your clinicals; the only people who could possibly want that are at the top, and as I've already stated a rational, objective view from the top now sees multi-billion dollar lawsuits looming if they do.

Frankly this is more retconning. You've already decided that they are guilty of bad science and are coming up with convoluted reasons why they'd do these things. First you have to assume they are immoral, then greedy, then incompotent - now fraudulent with idiot investors.

[kaikatsu]

Okay cite an example of when that was ever done in the drug industry. All trades by executive officers are tracked and it should be trivial to show them selling out before impending trouble.

See, unfortunately, this isn't the kind of thing I can provide documented evidence for. I do know quite a bit of antidotical evidence. The reason why I can't provide documented evidence is as follows: if I COULD demonstrate it, they would already be charged, thus nullifying my point. Likewise, if they got away with it, it was because the insider trading was not tracked.

I realize this is a weak argument, but I am attempting to show a plausable mechanism under which condoning questionable data would still be profitable.

Insider trading by proxy is almost impossible to proove. I say this on hearsay evidence so I'm sure you'll dispute it -- just those economic professors I've had have commented on it a lot. To show the person guilty you have to FIND a link, then demonstrate it, and a savvy blue collar criminal will think ahead in terms of years.

I WISH I could provide figures, but we don't know how much money is lost every year to white collar crime. Unlike a blue collar robbery, there's no one there to file a statement. No matter how anal accountants are there's always money that gets misappropriated at best, and outright stolen at worst if people are savvy. Likewise, I can't show the number of times board members have WILLFULLY taken money from investors and gotten away with it.

Who hires the board? INVESTORS. Repetitively soaking your investors is not practical. They will fire your ass and replace you. Even the most criminal Enron only soaked their investors once. Again I'd like to repeat my challenge for evidence.

The problem is assuming that the investors aren't busy trying to dump the falling stock. By the time anyone gets around to voting you've got a large number of people that were not involved in the drop, and most people who are part of a large mutual fund/diversified portfolio won't be tracking that one particular stock with a great deal of attention.

However, I will agree that repeatedly soaking investors is a bad, short term idea. That being said, it happens. That being said, bank robbery is a terrible idea too, with a very high rate of capture. People still do it.

Okay show me the proof. Executives bailing before impending doom is trivial to prove. Frankly though such a course of action risks both civil and criminal liability.

I'm not sure that "trivial" to proove is the correct term. Rather, I would say we have only CAUGHT those people for whom it WAS trivial to proove. To use an analogy, you might demonstrate that rape trials carry a high conviction rate, but rapes themselves have a low rate of report.

I'm not the only person that thinks this way. These ideas are coming from an economics professor I had, but to me they make a lot of sense. Here's a link of some similar figures.

http://www.corporatepolicy.org/issues/crimedata.htm

I realize that this talks about white collar crime in GENERAL, as opposed to merely insider trading. I'll do some looking around for more data to back the point.

Show me three drug insiders who did.

Just to re-iterate, you are asking me to show you three drug insiders who scammed investors for LOTS of money, and then who NEVER GOT CAUGHT. Believe me, I'd love nothing more than to demonstrate this... but the closest I could get would be to show someone that evaded capture for ten to fifteen years -- and I'm not about to do research on that because there have been reforms since then, and we are talking about the current situation.

To reverse the situation, how can you proove that EVERY drug insider DOES get caught? That's not something you can proove.

Yes I do. Normally a drug that reaches clinicals has a VP with direct access. <snip> Immediate concerns of the people doing the research are not to BS your clinicals...

I think this point of view is idealistic. Testers which tend to give positive results do tend to get hired back more often. However I am unable to demonstrate this at the time, so I will retract this point. My experience with falsifying and/or hush-hushing problems and bugs is with a smaller business that had a more "networked" environment, so I may have overstepped my bounds with that statement.

Frankly this is more retconning. You've already decided that they are guilty of bad science and are coming up with convoluted reasons why they'd do these things. First you have to assume they are immoral, then greedy, then incompotent - now fraudulent with idiot investors.

To be more accurate, I'm demonstrating a plausable reason for WHY they might be involved in such crime. Wong has, IMHO, pretty conclusively evidenced that it DOES happen. You are saying there are great incentives for people NOT to perform these actions deliberately, and every time it has happened it has been a special case, because in GENERAL no one would risk the suits.

I merely wish to show that, perhaps, there are plausable reasons to, as you said, risk the suits -- these reasons being that they MAY be crafty and skilled at manipulating stock, OR they are merely incompetent, OR they have idiotic investors. At any given time any one of these may be true -- I am not saying that they are all true every time, nor guaranteed to be.

You call it retconning for me to have "already decided that they are guilty of bad science" but I think that Mike's data shows that they ARE. You are, if I understand you correctly, saying that it's not bad science because to do so would be to go against powerful economic incentives.

[tharkûn]

See, unfortunately, this isn't the kind of thing I can provide documented evidence for.

Sure you can. By law every time a board member or executive officer sells or buys their own stock it is recorded by the FCC. Every single instance in the last two years was reported to the FEC and is availible for inspection here (note you will want to click on the insider trading link which may take a year or two to load). If an executive makes a major trade not that long before news breaks about the dangers of a drug their company makes it is recorded and reported.

The reason why I can't provide documented evidence is as follows: if I COULD demonstrate it, they would already be charged, thus nullifying my point. Likewise, if they got away with it, it was because the insider trading was not tracked.

All insider trader is tracked by law. The above link gives you EVERY single instance in the US for the last two years.

Insider trading by proxy is almost impossible to proove. I say this on hearsay evidence so I'm sure you'll dispute it -- just those economic professors I've had have commented on it a lot. To show the person guilty you have to FIND a link, then demonstrate it, and a savvy blue collar criminal will think ahead in terms of years.

Okay so now you are saying that we don't just have amoral, greedy bastards commiting felonies, but we have other amoral, greedy bastards entering into conspiracies to commit felonies. Further ALL these people are such criminal masterminds that they have never dicked up and been caught.

Somehow of course the corrupt executive is going to have to get his payment from the proxy. I don't see all that many fool proof methods for doing so.

The problem is assuming that the investors aren't busy trying to dump the falling stock. By the time anyone gets around to voting you've got a large number of people that were not involved in the drop, and most people who are part of a large mutual fund/diversified portfolio won't be tracking that one particular stock with a great deal of attention.

BS. Many stocks are held by long term investors who simply weather such storms. Many such investors are doing so through the services of mutual funds, investment managers (who have thousands of accounts or more), and other managed accounts. Those people will not be selling and some of the professionals they hire to manage their money will look into these things. Likewise you have the billionaire investors (like Warren Buffet) who scoop up stock from mismanaged companies and then fire the management.

However, I will agree that repeatedly soaking investors is a bad, short term idea. That being said, it happens. That being said, bank robbery is a terrible idea too, with a very high rate of capture. People still do it.

Show me an example of repetitive soaking.

I'm not sure that "trivial" to proove is the correct term. Rather, I would say we have only CAUGHT those people for whom it WAS trivial to proove.

I'm not asking for a legal proof of fraud. I'm asking for examples of executives who sold their stock in a short timeframe before the bad news and price correction. All their trades are recorded and reported. If you think Merck executives dumped stock before informing the FDA of Vioxx troubles, then show me the SEC filings they dumped stock just before.

I will leave the conspiracy to defraud shareholders with as yet unnamed method of payout alone for now.

I'm not the only person that thinks this way. These ideas are coming from an economics professor I had, but to me they make a lot of sense. Here's a link of some similar figures.

Advocacy group. A quick glance spotted several BS claims. However you are ranging pretty far afield here. The accusation isn't that white collar crime occurs, or even that it occurs in drug companies. It is that drug companies, or their executives, have incentives to persue bad science.

Just to re-iterate, you are asking me to show you three drug insiders who scammed investors for LOTS of money, and then who NEVER GOT CAUGHT.

Caught, didn't get caught, insufficient evidence to convict ... doesn't matter to me. Just point me to the SEC filings just prior to their bad science going public.

Believe me, I'd love nothing more than to demonstrate this... but the closest I could get would be to show someone that evaded capture for ten to fifteen years -- and I'm not about to do research on that because there have been reforms since then, and we are talking about the current situation.

You could wade through the SEC filings in the days/weeks/months prior to a drug having problems and show that yes executives were making suspicious transactions in that time period. More realisticly you could look for investor reports, particularly in cases where leadership struggles were occuring, and find someone elses work showing that an executive sold off prior. I'm not asking you to prove guilt beyond a reasonable shadow of doubt, I'm asking that you prove that transaction which might possibly have happened in the manner you suggested occured.

To reverse the situation, how can you proove that EVERY drug insider DOES get caught?

I can prove that any given insider didn't sell off a block of stock just prior the drug going tits up. Show us some who have sold large blocks of stock under suspiscious circumstances; as you know your claim, your burden of proof.

I think this point of view is idealistic. Testers which tend to give positive results do tend to get hired back more often.

We've gotten bonuses for giving negative results.

To be more accurate, I'm demonstrating a plausable reason for WHY they might be involved in such crime.

Please. At the end of the day your little theory fails to remember that pushing shoddy science isn't the only way they can soak the investors. From creative bookkeeping to misstating earnings/losses to using acquisition information (which is ridiciously common in the drug industry) are all viable ways of defrauding investors - and these HAVE been demonstrated. If you were the great white collar criminal why in hell would you dick with the clinical trials? There are far more people involved, far greater reason for the rare scientist with a conscious to blow the whistle, and gratuitiously longer papertrails. Even if you manage not to blow the game, you still face hell trying to predict when the drug will go tits up and assuming this were really likely that someone else wouldn't be doing the same damn thing with another drug and destroy your 'profits' before you can cash out.

Of all the ways to defraud the investors, this has got to be one of the most likely to fail. If you have THIS much influence why are you dicking around with something THIS convoluted?

Wong has, IMHO, pretty conclusively evidenced that it DOES happen.

Wong has provided ample evidence that new side effects are found after drugs are declared safe. Wong has also provided evidence that some clinical trials were insufficiently rigorious. NOWHERE has he given even minimal evidence of fraudulent conduct. Indeed when I orginally chided him about the fraud and falsification angle he corrected me that his concerns were about negligence.

You are saying there are great incentives for people NOT to perform these actions deliberately, and every time it has happened it has been a special case, because in GENERAL no one would risk the suits.

No, not all that long ago shoddy science paid out. When a drug was found to have new side effects the stock price didn't take as serious a hammering. The market response has changed so the direction of profit motive has changed. The cost/benifit analysis between "get it to market" and "don't get the company sued" has shifted.

I merely wish to show that, perhaps, there are plausable reasons to, as you said, risk the suits -- these reasons being that they MAY be crafty and skilled at manipulating stock,

In which case it boggles the mind why they are dicking around with clinical trials rather than just traditional defrauding of the investors, particularly given the longer papertrails.

OR they are merely incompetent

Right but not so incompotent to be tossed by the shareholders, but compotent enough to see a benifit in dicking with the science

OR they have idiotic investors

The stastical likelihood that each of the millions of investors in Merck (for example) is an idiot is astronomical. Even if we restrict ourselves to the 1000 largest investors the idea that all of them fall under the idiotic category defies the logic of the bell curve (evening assuming 90% of all investors are idiots, you still end up with 1.7 EE -44% chance of them all being idiots).

At any given time any one of these may be true -- I am not saying that they are all true every time, nor guaranteed to be.

And I'm saying this is all BS retconning of the fact that you have already decided they are evil and are groping at straws to find something beyond comic book supervilliany.

You call it retconning for me to have "already decided that they are guilty of bad science" but I think that Mike's data shows that they ARE.

Mike's "data" shows that Mike thinks it is bad science for a drug which has to have certain chemical activity to fit into the opiate receptor looks like another drug which also fits into the opiate receptor

Mike has some good data, unfortunately much of it is old and conditions have changed since then.

You are, if I understand you correctly, saying that it's not bad science because to do so would be to go against powerful economic incentives.

No I'm saying I challenged Mike to provide three examples from the last five years, he hasn't done so yet.

When we go over Vioxx I'm explaining exactly what the damn trials meant and am still waiting to hear what quantified standard Mike thinks should be imposed on them.

If you have specific examples of bad science, then I will go over them from a scientific perspective (i.e. how likely was this side effect to be found in a 5,000 patient, 12 month clinical trial). However the incentive to fudge the science has already shifted the other way - you stand to lose more by fudging the science than by encouraging good science.