Page 2 of 2

Posted: Wed Feb 06, 2008 7:40 pm
by BioDocFL
'Ethics' are relative.

Posted: Wed Feb 06, 2008 8:21 pm
by HarryZ
BioDocFL wrote:'Ethics' are relative.
Absolutely....but far too often they take a seat behind the power of the dollar :cry:

Posted: Wed Feb 06, 2008 8:32 pm
by BioDocFL
'If you agree the data should not be shared, are you not stating that it is more important that a specific researcher have the chance of a publication than to have the information made more quickly available to help people with cancer?'

It should be pointed out that the process of publication is one of the critical steps in the whole process of research. It involves peer review which is a means of quality control. Only a minority of papers get published, at least in reputable journals. This is a self-governing process that is necessary to avoid dubious data, faulty conclusions, and improper intrepretation of the data. Valid data does eventually get shared. Publication in a controlled, although slow cumbersome process, is necessary. Without the process, we could have the tobacco industry releasing their 'data' and having it accepted with as much credability as any thoroughly peer reviewed study in a reputable journal. It is the researchers who are contributing their time unpaid to conducting the peer review process in order to protect the public from faulty work, and yes, to protect their own areas of expertise from charlatans. Peer review of potential publications is just as important as the original peer review of grant applications that decides whether to fund the work in the first place. Should we also just put the grant money out there for anyone who comes along without any review? Peer review for publishing is not perfect, but what is a better approach and how do you implement and monitor an alternative? And how do you keep from demotivating researchers if it takes away from recognition of their work? Researchers are self-motivated and accomplishment oriented. They have put a lot of time and effort into education and training. And they are under extreme pressure in their work and are typically paid less than if they had gone into business or law, for example. Maybe there is some altruism in researchers but not to the point of their own detriment. They need a fair system of motivators just like in any other field.

Wesley

Re: researchers don't want to share!

Posted: Wed Feb 06, 2008 11:45 pm
by NHE
Here are some lighter thoughts on the process of peer review... :roll:

http://www.thisisms.com/ftopict-2757.html

NHE

Posted: Thu Feb 07, 2008 8:49 am
by art
HarryZ wrote:
The researchers fear is that someone else will find out something important about cancer sooner than they would. Explain to me how that is something we should be fearing.
Art,

In a perfect world this would be great but even in the grass roots research world, research data is guarded. Being able to publish results first and having a better opportunity to obtain the limited research money that results from that is what the "game" is all about. I don't like it any more than you do but that is unfortunately the reality. My cousin's son is a PhD micro-biology cardiac system researcher and it's all about being first and getting that almighty grant money.
We'll have to disagree that this is only realizable in a perfect world. I'm amassing empirical evidence to the contrary, so you might need to reconsider at some point. If we sit back and say "well, that's just how it is" we'll be stuck with this sort of situation. If we don't put up with it, it will change. But I'll admit, it sure is easier to just say "that's how it is."
HarryZ wrote:
I'm sorry that researchers have created a system for themselves (yes, they created it) that would "punish" them for being slower, but the solution is not to hide data! If they *had* to share the data, the reward system would change very quickly.
That would be nice but I doubt we will ever see that happen in our lifetime.
I urge you to not be so defeatist. It's happening. PLoS, human genome project, our project, and now the mandate to make publications available for free after publication (with time delay, but it's a start. The NIH is even getting in on things and requiring data to come back and be shared for certain programs they fund.

Let's all keep demanding this, refuse to give to organizations that don't require this (except the IRS of course, unless you want to go to jail :-/)!

The people with the money won't suffer if this were to change, therefore I contend that it is changable. The researchers aren't going to stop taking your money if they have to share. They'll grumble, but there isn't enough money around for them to just go somewhere else.

Posted: Thu Feb 07, 2008 9:58 am
by art
BioDocFL wrote: [Peer review] is a self-governing process that is necessary to avoid dubious data, faulty conclusions, and improper intrepretation of the data.
I don't think you can say it is "necessary." You might argue that it is "sufficient," but there are many variations on the concept that could be used if things were different. Peer review is highly flawed and is by no means a guard against bad science. Papers are regularly retracted and results unrepeatable. An ideal review process would have a much lower rate of this. Other industries have different review processes to maintain quality in areas that are sometimes more important to human safety than a genetics study (e.g. airplane construction, venture capital, or even the FDA for drug studies)
BioDocFL wrote: Valid data does eventually get shared. Publication in a controlled, although slow cumbersome process, is necessary. Without the process, we could have the tobacco industry releasing their 'data' and having it accepted with as much credability as any thoroughly peer reviewed study in a reputable journal.
Well, the quoted article claims that the data is NOT always shared. As long as data is clearly labeled as reviewed or not reviewed then having the tobacco company's raw data out there would be a good thing. You don't think every statistician on the planet wouldn't be analyzing that data to find things? We'd get a much better result than peer review. And I've seen how papers get "thoroughly" reviewed - unpaid reviewers who have too much to do, and sometimes an agenda to put forth. We could do better. But anyway, this has nothing to do with sharing of data. Put it out there and let anyone have at it. You can keep your peer review if that makes you happy.
BioDocFL wrote: It is the researchers who are contributing their time unpaid to conducting the peer review process in order to protect the public from faulty work, and yes, to protect their own areas of expertise from charlatans.
They aren't protecting the public from faulty work. At most they are protecting the publication from an embarrassing retraction. There are no consequences from the public to the reviewer for letting a bad paper through. The publication suffers humiliation and probably never uses those reviewers again, but there isn't a legal responsibility there.
BioDocFL wrote: Peer review for publishing is not perfect, but what is a better approach and how do you implement and monitor an alternative?
Eee - that's a whole 'nuther quagmire from sharing. I have ideas, and as I mentioned above, there are models from other industries that could be tried. We don't need to have a one-size fits all model (I know how people like to). Have different levels and qualities. But get the data out there.
BioDocFL wrote:And how do you keep from demotivating researchers if it takes away from recognition of their work? Researchers are self-motivated and accomplishment oriented.
You answer your own question with the statement that followed it. Keep in mind that first publication does not have to be sole "accomplishment."

We might want to start a separate thread about peer review. This one was about sharing and I don't see any reasons given for not sharing in your response. You point out the need to consider the results of unreviewed analysis of shared data, which is valid. But I don't think that the intent of those who want data shared is to avoid review. Nor do I think that any of the harms you put forth are inherent in sharing if you allow for adopting new or altered methods of review.

Posted: Thu Feb 07, 2008 10:09 am
by HarryZ
Art,
We'll have to disagree that this is only realizable in a perfect world. I'm amassing empirical evidence to the contrary, so you might need to reconsider at some point. If we sit back and say "well, that's just how it is" we'll be stuck with this sort of situation. If we don't put up with it, it will change. But I'll admit, it sure is easier to just say "that's how it is."
I admire and respect the hard work you are doing to try and change the existing situation. I have followed MS research for over 40 years now and have seen how small the progress it has made during this time. Trying to introduce new concepts and treatments outside the established world of MS medicine is a very tough grind. One only need to look at what happened to Prokarin when it first made its appearance in 1999 to see how the established hierarchy circled the wagons to protect their domain! I wish you the best of luck because I know it will be a long, hard road.
I urge you to not be so defeatist. It's happening. PLoS, human genome project, our project, and now the mandate to make publications available for free after publication (with time delay, but it's a start. The NIH is even getting in on things and requiring data to come back and be shared for certain programs they fund.
I'm not a "defeatist"...perhaps a "realist" would be more appropriate. My wife had MS for 36 years. I can't explain how hard we tried to push for different paths to be taken in the MS world. She died last September from complications that MS brought to her health. I'll be glad to have someone like yourself carry on in attempting to change the current system.
The people with the money won't suffer if this were to change, therefore I contend that it is changable. The researchers aren't going to stop taking your money if they have to share. They'll grumble, but there isn't enough money around for them to just go somewhere else.
The people with money will continue to be happy as long as their money machine continues to operate and they don't feel threatened by what you are doing. If they see that their castle's foundation is under attack, just watch at what they will do to protect it. I saw first hand what the NMSS did when they felt Prokarin might get too many MS patients' attention.

Harry

Posted: Thu Feb 07, 2008 2:02 pm
by BioDocFL
Two things that come to mind that make accessing raw data by outsiders (ie, not of the original research team) complex are HIPAA and hacking. To protect the one and protect against the other require a lot of effort by the hosting institution.

I believe that the peer process does protect against some faulty data (of course not all) by reviewing grant applications, IRBs reviewing the original protocol set ups, and the editorial boards and manuscript reviewers checking over articles before they are published. It is not perfect but researchers are sincere in their efforts and probably the best trained for the critical eye needed.

Wesley