The Politics of "Doing Science"

[Pangloss]

I've been reading Steven Milloy's "Junk Science Judo", and I have to say it's pretty intriguing. I do understand the criticisms, and I think a true skeptic has to recognize that someone who is allegedly on the payroll of a company like ExxonMobil or the tobacco companies cannot objectively speak out against global warming or the dangers of second-hand smoke inhalation.

 

But it's really the larger issue that I'm interested in anyway. Specifically, this allegation by Milloy that the practice of science in this country (the US) has become a matter of scare tactics and what he calls "junk science".

 

I'm not sure I agree with him about the extent of the problem, but it's not hard to see the logic in what he's saying. Scientists get ahead these days by doing single epidemiologic studies (which right off is a concern -- one study is not science!) which they take straight to the nearest media outlet.

 

The problem is that the government, media, and scientists are ostensibly trying to err on the side of caution. But in the process of erring on the side of caution, they end up so radically shifting the terrain that the entire issue is settled before any further science is done to try and verify the claim!

 

So for example a study comes out claiming that there "may" be a correlation between Vitamin E and heart disease. Next thing you know the FDA or EPA is passing regulations, and then voila -- millions of dollars are being spent. Then another study comes out that completely contradicts the evidence, and we're back to square one -- only the new regulations and budgets stay intact.

 

It's enough to make one wonder if any real science is even done anymore! Is it all just faulty epidemiology?

 

What do you guys think? Am I the only one who cringes these days when the talking heads open a new story with "a new study released today by...." This can't be good for science!

[swansont]

The irony of Milloy having a column (and book) about Junk Science is that Milloy is a shill, and presents science within an ideological framework. So "junk" science becomes science where he doesn't like the answer.

 

The larger problem of not trusting studies is related. If ideology has already dictated what the results have to be, you can't trust it. Sadly, this has infiltrated government-sponsored work, which is supposed to be free from such interference. It has become increasingly perverted by politics in what conclusions are allowed to be drawn, and especially so in the recent past. You get decisions like not approving (or delaying) medical treatments or not dispensing medical advice (or dispensing wrong information) because of political and ideological stances; IOW you'd rather kill people than help them, because you disapprove of their ideals or behavior, or because helping business is worth a few lives. Recent legislation has turned the system backwards with regard to burden-of-proof, or requiring kinds of proof/evidence that science cannot provide.

 

I'm in the middle of reading "The Republican War on Science" by Chris Mooney. It's good, but scary.

[scicop]

Another problem is that of funding. When a new relationship or finding comes out, those findings tend to dictate which projects are financed. So using the "Vitamin E and Heart Disease" relationship you suggested as an example, if something like that is put out there, then it becomes a hot topic. The NIH for example will put a call out for grants that focus on the dissecting out the relationship. Scientist who are always in need of funding, design projects so that they can have access to that money to study the Vit E and Heart Disease Relationship.

 

A few years go by, and papers are published by the grants that were given out for the study of that topic. Scientist have built their careers studying that topic and perhaps some the fruits of their labor have seen some light in the early stages of clinical development. Now, BAM a new study is put out that contradicts the finding. Do you really think NIH is going to stop funding for ongoing studies? No. Do you think NIH will provide money to further investigate the contradictions? NO! Why? because yo uhave to look at WHO the NIH (or any health policy agency for that matter) is!

 

Here's is who they are: Scientist and Clinicians who biased to the orignial findings.

 

Why?:

 

NIH grant (also NSF) funding is dictated by a peer-review process by scientist/clinicians that have already established themselves in the field. So, if grants are comming that challenges their original findings, they will not approve the grant! It could jeopardize their future funding/careers. They don't want to look as though they were barking up the wrong tree.

 

Its a way of self promoting/survivial among scientists.

 

Is it bad? yes. Ethical? NO. But its peer-review, and in this case unlike the legal system, they are not blind.

 

So, wanna say that global warming is a problem? don't count on getting money from a gov't agency.

[Pangloss]

The irony of Milloy having a column (and book) about Junk Science is that Milloy is a shill, and presents science within an ideological framework. So "junk" science becomes science where he doesn't like the answer.

 

Yeah as I mentioned before I'm definitely keeping that in mind -- I've seen this criticism peppered around in various places, common enough that it's worth noting. But....

 

 

I'm in the middle of reading "The Republican War on Science" by Chris Mooney. It's good, but scary.

 

In terms of finding an objective baseline viewpoint, what's the difference? Isn't that just exchanging one ideology for another? Cato/Milloy/libertarians aren't Republicans, either, so that title (which I've seen associated with criticism of Milloy elsewhere as well) makes me suspect on that basis as well -- why is he lumping them in together? Is it because of the corporate associations with Cato, and just ignoring the ideological differences?

 

Let me know if that book is any good, I'll add it to the list. I'd like to hear more about his basic premise as well -- if you want to start a thread on it here that'd be great.

[Pangloss]

Ok I wanted to reply to this separately because the above was more in the nature of an aside.

 

The larger problem of not trusting studies is related. If ideology has already dictated what the results have to be' date=' you can't trust it. Sadly, this has infiltrated government-sponsored work, which is supposed to be free from such interference. It has become increasingly perverted by politics in what conclusions are allowed to be drawn, and especially so in the recent past. You get decisions like not approving (or delaying) medical treatments or not dispensing medical advice (or dispensing wrong information) because of political and ideological stances; IOW you'd rather kill people than help them, because you disapprove of their ideals or behavior, or because helping business is worth a few lives. Recent legislation has turned the system backwards with regard to burden-of-proof, or requiring kinds of proof/evidence that science cannot provide.

[/quote']

 

Okay so you think Milloy has a point, then, if I'm reading you right, you just don't necessarily agree with his specific examples? That's kinda how I'm reading it.

 

Sci those were some interesting points as well.

[ecoli]

I want to know what labs are publishing results like this. In my lab, my PI would NEVER allow anyone to even think about publishing their results unless we got the results many many times. And then, only if the experimental design was flawless.

[Pangloss]

I'm glad to hear that, thanks for posting it. Exactly the sort of reality check I was hoping to get here to balance this book out.

 

Would you guys like me to post some specific examples from the book? That might be an interesting discussion.

[scicop]

I want to know what labs are publishing results like this. In my lab, my PI would NEVER allow anyone to even think about publishing their results unless we got the results many many times. And then, only if the experimental design was flawless.

 

Thats mighty noble of your PI. Not all labs are like that, I've personally seen instances where n of 1 was published. Not kidding. When grants are on the line and publications are needed, you'll be amazed at what scientist can get away with.

[ecoli]

Thats mighty noble of your PI. Not all labs are like that, I've personally seen instances where n of 1 was published. Not kidding. When grants are on the line and publications are needed, you'll be amazed at what scientist can get away with.

 

I'm surprized something like that can get past a peer-review board. I hope rubish like that doesn't get into the higher-quality journals at least.

[Pangloss]

Perhaps the problem lies in the fact that CNN doesn't have a peer review board.

 

Even more to the point, the major journals are now essentially compromised as well, not only be subscriber and advertising revenue, but by the undercutting of standards due to overwhelming pressure to produce revelatory new articles.

[swansont]

Ok I wanted to reply to this separately because the above was more in the nature of an aside.

 

 

 

Okay so you think Milloy has a point' date=' then, if I'm reading you right, you just don't necessarily agree with his specific examples? That's kinda how I'm reading it.

[/quote']

 

 

I haven't read Milloy's book, only a few of his columns. Sometimes his objections have merit, often times they don't. It all depends on his preconceived position. If the science conflicts with his ideology, it'll invariably be deemed junk; the problem in those cases is that the objections aren't always valid from a scientific standpoint, and his rebuttals/objections suffer from misquotations, misrepresentations and logical fallacies.

[swansont]

In terms of finding an objective baseline viewpoint' date=' what's the difference? Isn't that just exchanging one ideology for another? Cato/Milloy/libertarians aren't Republicans, either, so that title (which I've seen associated with criticism of Milloy elsewhere as well) makes me suspect on that basis as well -- why is he lumping them in together? Is it because of the corporate associations with Cato, and just ignoring the ideological differences?

 

Let me know if that book is any good, I'll add it to the list. I'd like to hear more about his basic premise as well -- if you want to start a thread on it here that'd be great.[/quote']

 

The basic gist is that the Republicans are demonstrably worse abusers than the Democrats, not that Dems are angels who never politicize science.

 

Attempted legislation of the Gingrich "Contract with America" Republicans doing away with the Office for Technology Assessment and other maneuvers. Then there's the scientific creationism and later the intelligent design movements (the wedge document and "teach the controversy!" even though none really exists). Many, many example of helping to enable industry (including tobacco to challenge policy by discrediting science (or scientists, when the science couldn't be), or make science jump unreasonable hurdles, or "democratizing" science (my terminology, not his). (And I haven't gotten to the last section of the book, on "Dubya")

 

 

Treating science as a democracy, such that all viewpoints are equally valid and should be heard, is inconsistent with how science actually works. You can always find a credentialed person with a crackpot viewpoint, but that one "expert" doesn't merit equal time in considering what the valid conclusions are.

[zyncod]

From Science's submission policies

 

Prior Publication

Science will not consider any paper or component of a paper that has been published or is under consideration for publication elsewhere. Distribution on the Internet may be considered prior publication and may compromise the originality of the paper as a submission to Science. Please contact the editors with questions regarding allowable postings this policy.

 

In addition, reporting the main findings of a paper in the mass media may compromise the novelty of the work and thus its appropriateness for Science. Authors are free to present their data at scientific meetings but should not overtly seek media attention or give copies of the figures or data from their manuscript to any reporter, unless the reporter agrees to abide by Science's press embargo. If a reporter attends an author's session at a meeting and writes a story based only on the presentation, such coverage will not affect Science's consideration of the author's paper. (For more information, please see the embargo entry in the Science Contributors FAQ.)

 

The fact is, most of the information that gets out into the media early is not going to be that good anyway, because the authors are compromising their chances to publish in a good journal. The last autism-vaccine "news" is an example of this, where the news story was apparently based on a poster presented at a conference. Not even a talk, a poster.

 

If it's important enough to warrant a news story, it'll be published in a respectable journal before it gets out to the media. Otherwise, the news story is not worth the paper it's printed on.

[swansont]

The problem is that the government' date=' media, and scientists are ostensibly trying to err on the side of caution. But in the process of erring on the side of caution, they end up so radically shifting the terrain that the entire issue is settled before any further science is done to try and [i']verify the claim[/i]!

 

So for example a study comes out claiming that there "may" be a correlation between Vitamin E and heart disease. Next thing you know the FDA or EPA is passing regulations, and then voila -- millions of dollars are being spent. Then another study comes out that completely contradicts the evidence, and we're back to square one -- only the new regulations and budgets stay intact.

 

From what I'm reading (Mooney's book and elsewhere), this really hasn't been the case.

 

Supplements, for example, aren't subject to the FDA in the same way that drugs are. You don't have to demonstrate effectiveness, you just have to be careful not to claim too much. (e.g. don't claim that Ginkgo-Biloba will improve cognitive functions, just have a testimonial that "I take it and now I'm more alert and my memory has improved!"). Look at what happened with ephedra.

 

In cases where some effect has been found to be harmful, it's a lot easier to strike down the rules regulating the substance than it used to be. Read up on the Data Quality Act. It makes it much easier to question the science on which policy is based; all you need is a study or two (who cared if they're funded by the affected industry?) that contradict the other studies upon which the policy was based.

[Pangloss]

Examples of the opposite seem to abound as well, though, don't they? Products like fen-phen, silicon breast implants, Alar, etc, removed or withdrawn based on initial scares that subsequently turned out to be false. But still we have lawsuits and regulations flying around based on the assumption that the scare was true.

 

How can we stop the initial scare from wiping out a product before we get a chance to follow-up with an in-depth study? While at the same time providing cautionary information to the public in such a way as to take advantage of the warning without panicing and destroying a useful product?

[Mokele]

You have no idea how happy I am right now that my field has zero applicability, zero media appeal, and can be fully funded for a decade for less than the cost of a used car.

 

Mokele

[padren]

Science is generally relatively expensive, and the studies that get done must inevitably be the ones viewed as the most relevant to someone with money to give.

Some more generous funders probably only need to know the science is being done well and will add to the scientific knowledge of mankind, and that is what they want so they are happy to objectively fund any science that adheres to that.

Other funding is likely to be more forthcoming if it deals with a "scary" issue that is hyped to make it even more scary, or if it is in the interests of a commercial company such as big tabacco or big oil.

 

It does make sense that the need for money creates an inherent need to pitch studies that get the attention of people with money as a general "pressure" but the question is how manageable is it.

We have a "pressure" to hurt each other the more densely we are packed into cities and we get road rage, but it is a pretty minor effect when most people deal with the pressure just fine.

I think the principle is sound, the question I am more curious about is the impact and pervasiveness of it.

[swansont]

Examples of the opposite seem to abound as well' date=' though, don't they? Products like fen-phen, silicon breast implants, Alar, etc, removed or withdrawn based on initial scares that subsequently turned out to be false. But still we have lawsuits and regulations flying around based on the assumption that the scare was true.

 

How can we stop the initial scare from wiping out a product before we get a chance to follow-up with an in-depth study? While at the same time providing cautionary information to the public in such a way as to take advantage of the warning without panicing and destroying a useful product?[/quote']

 

The problem in trials is the "democratization" I discussed earlier. Each side trots out an "expert" or two, and they make their case to the judge and/or jury, who are not experts and possibly not particularly competent in science. And so many of the subtleties of statistics and science are lost on them, and it sounds like a 50-50 proposition, ending up with who is more persuasive, not who has the better science. Not surprisingly, sometimes the professional crank-for-hire has the better sales pitch.

 

The situation has improved in recent years, with the ability of judges to consult independent experts and get an unbiased view of what the prevailing science is actually saying.