lederhosen: (Default)
lederhosen ([personal profile] lederhosen) wrote2006-04-06 02:09 pm
Entry tags:

A slight misunderstanding

Via Mediawatch, transcript of '9 AM with David and Kim', discussing the recent UK drug trial disaster:

David Reyne: Some of these guys were given a placebo.

Dr. David Ritchie: Correct

David Reyne: I don’t really understand what a placebo is, but it seems to have, to have saved them! And wouldn’t it make sense that every time a trial like this takes place, that there’s a placebo on hand.


*sigh*

Exercise yesterday: 10km. Total 191km/115mi. Captured by Barrow-wights.

[identity profile] tyggerjai.livejournal.com 2006-04-06 05:53 am (UTC)(link)
The look on Dr. Ritchie's face in that still is priceless.

Though "drug trial that went terribly wrong" is a somewhat dubious phrase. If the trial showed that the drug is dangerous and not suitable for public consumption, then surely it went terribly *right*, albeit with unfortunate consequences? Isn't this akin to the Feynman gripe about "the experiment was a failure because it didn't support our hypothesis", as opposed to "because it didn't adequately test our hypothesis"?

sol.
.

[identity profile] lederhosen.livejournal.com 2006-04-06 06:00 am (UTC)(link)
Well, certainly 'wrong' from the perspective of those who got the drug...

I agree that the whole point of testing is to identify risks. OTOH, there've been claims that this one was exacerbated by poor procedures - had the patients been dosed one by one over a longer period, rather than all at once, there'd presumably only have been one person harmed instead of six. But I don't know enough about how such things are run to say whether that's a practical safeguard to apply.

[identity profile] tyggerjai.livejournal.com 2006-04-06 06:06 am (UTC)(link)
Ah. If it's poorly implemented, then yes, that's wrong.

I used to work for a clinical trials company. There's an illusion of independence, but so long as your data meets FDA (yes, FDA - the Australian legislation basically says "See FDA section 11", or whatever) requirements you're effectively working for Glaxo Smith Kline.

It was a very depressing experience.

You do simultaneous tests so you can say you tested it on 6 people and get the drug to market first. To be fair, you have to go through a hell of a lot of trials before you even start thinking about testing on humans, but no, heaven forbid anything increase your time to market ....

sol.
.

[identity profile] lederhosen.livejournal.com 2006-04-06 06:28 am (UTC)(link)
And, yeah, it's easy for the rest of us to be wise after the fact. In the real world, scheduling eight patients for eight different days isn't necessarily trivial. I can speculate about whether it was conducted with adequate care, but in the end I'm not in a position to make that judgement with certainty.