Lancet and MMR, standards in medical publishing?

RDM_83 again

I know its a bit after the fact for this post but I'm interested in the situation of medical journals.

In my understanding of the case Dr Andrew Wakefield carried out the research, and this research was published in the Lancet back 1998

How was his work published in the when it was so blatantly unscientific, the sample size was tiny, no double blind and various other flaws.

Now my question is Lancet is "the premier medical journal" and this paper was published, where was the process of peer review?and if it did occur what does it say about the standards of medical publishing, with a sample size this small (even if you can't see the direct figures for the 'population' )its going to be clear from the stats if you look at them closely that there is an issue.
Now I understand the article was eventually pulled but still at a much later date.
To me this is a much more serious issue than one MD carrying out a flawed study it points to a journal with a complete lack of scientific rigor, I could not see a paper with similar flaws making it into any of the top flight journals in another field

oh ps Don't know much about this hoping to have my faith restored

mikhail

There are lots of kinds of medical studies.

If I have a non-blind, non-controlled trial with twelve patients where I give them a new medicine, and they all die the next week, that's pretty much a solid medical result.

A study doesn't have to be double-blind and have a control group to get published. Sometimes those controls aren't entirely possible, e.g. in accupuncture, where "fake accupuncture" is inevitably administered by someone who knows it's all bollox. It just has to outline its methodology clearly, and present results which appear significant.

The Wakefield paper specified what was done, and recommended that further (better) studies be carried out. They have been (e.g. a Danish study of something like a quarter of a million kids found no link between . But Wakefield distorted his results, and the media sensationalised the result. No peer review can detect or compensate for that.

(I am not a doctor. The above is based on recollections of things I've read by Ben Goldacre. I recommend his book and his blog for much more on the subject.)

RobFowl

The lancet have since withdrawn that paper and stated they should never have published it in the first place.
Their thinking at the time was that it was so controversial, not to publish could be viewed as an attempt to cover up..
The editor at the time has since moved on.

RDM_83 again

"Their thinking at the time was that it was so controversial, not to publish could be viewed as an attempt to cover up.."

Thanks that actually explains a lot about why they published it

Mikhail,
Have decent understanding of statistics so understand that if you have results you describe that you can say you've a 100% confidence in your hypothesis , but my problem was the publishing of a study that lacked the expected standard of experimental working
(otherwise every PhD would be coming out with a paper every fortnight)

Goldacre'S Guardian collumn is very good though it annoys me the way he quotes so much meta-analysis which have their own added flaws and problems especially whats called the File Drawer problem
http://en.wikipedia.org/wiki/Meta-analysis#Weaknesses

SomeDose

mikhail wrote: »

There are lots of kinds of medical studies.

If I have a non-blind, non-controlled trial with twelve patients where I give them a new medicine, and they all die the next week, that's pretty much a solid medical result.

A study doesn't have to be double-blind and have a control group to get published. Sometimes those controls aren't entirely possible, e.g. in accupuncture, where "fake accupuncture" is inevitably administered by someone who knows it's all bollox. It just has to outline its methodology clearly, and present results which appear significant.

The Wakefield paper specified what was done, and recommended that further (better) studies be carried out. They have been (e.g. a Danish study of something like a quarter of a million kids found no link between . But Wakefield distorted his results, and the media sensationalised the result. No peer review can detect or compensate for that.

(I am not a doctor. The above is based on recollections of things I've read by Ben Goldacre. I recommend his book and his blog for much more on the subject.)

Incidentally, and this isn't in any way a criticism of the profession, being a doctor does not automatically qualify someone as being competent in statistics or scientific research (a fact which drug companies love to exploit). Wakefield is an obvious example of this, although being ethically bankrupt was perhaps the greater concern in his case.

From time to time I see applications from consultants to use new drug therapies and, in some cases, the trials and studies they submit as "evidence" are complete tosh.

2Scoops

SomeDose wrote: »

Incidentally, and this isn't in any way a criticism of the profession, being a doctor does not automatically qualify someone as being competent in statistics or scientific research (a fact which drug companies love to exploit).

And that will include The Lancet's reviewers as well as the people who submit manuscripts! Wakefield's paper toned done the more controversial aspects to slip it under the radar, but a competent reviewer should still have pulled him up on it.