![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
snousleBad news in my field - I only just heard about it today, after a 60 minutes episode on it aired last night. An earlier article in The Economist describes it pretty well. CBS called it "one of the biggest medical research frauds ever". Oh, joy.
This is exactly the kind of analysis I do so it hits rather close to home. What is disturbing is that the problems in the data were obvious and were found rather easily, but it took a very long time for the fraud to be fully exposed at the administrative level.
I must confess, when I started with my current client, my first reaction to their research was "this is way too good to be true", but after conducting a completely independent analysis reaching the same result as our academic founders had published, I was, and still am, confident in the correctness of the data. We have, to be sure, received data from collaborators with exactly the same kinds of errors mentioned in the Economist article, but there are simple consistency checks that flush them out immediately.
Makes me wonder what would happen if I had to issue a thumbs-down in the middle of a multimillion dollar project. I like to think it would halt everything in its tracks, but the pressures that surround high-impact research like this can cause people to behave in peculiar ways. There is no way this kind of fraud can continue forever - it will always become obvious in time - and Potti must have known early on that the gig would be up eventually. But, as they say, denial isn't just a river in Egypt. And the terrible thing is that the fraud propagated right through to actual clinical trials, affecting the lives of real people. This is certain to make future research much more difficult. It's already hard enough to satisfy IRBs, who are absurdly cautious, and this is going to make them even more paranoid. Nobody gets fired for research that never gets approved, and there is every chance that truly effective treatments will now get left on the cutting room floor.
In the meanwhile, I'm cheered by a jobs review in Nature, just arrived this morning, which says that I'm going to be infinitely employable forever. I have no idea why, ten years after the completion of the genome, there aren't hordes of university graduates clamoring for these positions, rendering me obsolete. But there aren't.
This is exactly the kind of analysis I do so it hits rather close to home. What is disturbing is that the problems in the data were obvious and were found rather easily, but it took a very long time for the fraud to be fully exposed at the administrative level.
I must confess, when I started with my current client, my first reaction to their research was "this is way too good to be true", but after conducting a completely independent analysis reaching the same result as our academic founders had published, I was, and still am, confident in the correctness of the data. We have, to be sure, received data from collaborators with exactly the same kinds of errors mentioned in the Economist article, but there are simple consistency checks that flush them out immediately.
Makes me wonder what would happen if I had to issue a thumbs-down in the middle of a multimillion dollar project. I like to think it would halt everything in its tracks, but the pressures that surround high-impact research like this can cause people to behave in peculiar ways. There is no way this kind of fraud can continue forever - it will always become obvious in time - and Potti must have known early on that the gig would be up eventually. But, as they say, denial isn't just a river in Egypt. And the terrible thing is that the fraud propagated right through to actual clinical trials, affecting the lives of real people. This is certain to make future research much more difficult. It's already hard enough to satisfy IRBs, who are absurdly cautious, and this is going to make them even more paranoid. Nobody gets fired for research that never gets approved, and there is every chance that truly effective treatments will now get left on the cutting room floor.
In the meanwhile, I'm cheered by a jobs review in Nature, just arrived this morning, which says that I'm going to be infinitely employable forever. I have no idea why, ten years after the completion of the genome, there aren't hordes of university graduates clamoring for these positions, rendering me obsolete. But there aren't.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2012-02-13 05:43 pm (UTC)no subject
Date: 2012-02-13 06:21 pm (UTC)Bioinformatics: softens hands while you do statistics!
no subject
Date: 2012-02-13 06:22 pm (UTC)no subject
Date: 2012-02-14 01:29 am (UTC)no subject
Date: 2012-02-14 07:42 am (UTC)Yes, I've been deceived, many times.
no subject
Date: 2012-02-14 02:29 pm (UTC)no subject
Date: 2012-02-14 03:29 pm (UTC)Although it is hard to fake data convincingly, this created enough confusion and reasonable doubt that it wasn't declared "fraud" for quite some time. Deliberate fraud of this kind is actually very rare, so the presumption was that everything was being done in good faith.
no subject
Date: 2012-02-15 04:46 pm (UTC)I haven't seen the climate deniers latch onto this yet, but I think they will in time.