Medical Evidence

[TheDBird90]

Advertisement - Members don't see this ad

This thread relates to my earlier post, "Medical Journals." I found some very interesting news that I forgot to share in that thread.

What is your response to the claim that, "much of the scientific literature, perhaps half, may simply be untrue?"

(source - http://www.collective-evolution.com...-journal-half-of-all-the-literature-is-false/)

 

[efle]

All you can really do is learn to be very critical of study design, and carefully note who is presenting the information to you. Especially with open access online journals, piles of garbage will get published after "peer review". But, that does not indicate that half of the content in journals like NEJM and Lancet are garbage!

 

[Goro]

As an editor of the journal that gave us Andrew Wakefield's bogus claims that the MMR vaccine was unsafe, I guess he should know.

A lot of crap science gets published, but the beauty of Science is that it's self-correcting, and things that aren't true or don't work get put into their place eventually. This comes from someone who spent a good deal of time trying to replicate some findings in a Nature paper (that I suspect were fabricated). No, I'm not sharing.

 

[LizzyM]

We make recommendations and treatment decisions based on the best available information while acknowledging that our knowledge is imperfect and some things that are thought to be true today may turn out to be untrue upon further investigation.

For example, in the early 1990s when many of you were in utero, the advice to pregnant women was to avoid eating peanuts or peanut butter and not to give peanut products to little ones in order to prevent sensitivity (allergy) to peanuts. Now, the latest advice is that kids should be exposed to peanuts at a very young age to prevent allergies. http://www.annallergy.org/article/S1081-1206(16)31164-4/fulltext

Part of life-long learning on the part of physicians is because we realize that knowledge evolves over time as we attempt to understand what is true.

 

[efle]

spent a good deal of time trying to replicate some findings in a Nature paper (that I suspect were fabricated)

Looking into how frequent failure to replicate is was a big shock for me.

To anyone interested, I first found some crazy info about Psych and then cancer drugs too , it's pretty mind blowing to think a majority of what you see even in "legit" sources would not hold up.

 

[JustAPhD]

As an editor of the journal that gave us Andrew Wakefield's bogus claims that the MMR vaccine was unsafe, I guess he should know.

A lot of crap science gets published, but the beauty of Science is that it's self-correcting, and things that aren't true or don't work get put into their place eventually. This comes from someone who spent a good deal of time trying to replicate some findings in a Nature paper (that I suspect were fabricated). No, I'm not sharing.

Looking into how frequent failure to replicate is was a big shock for me.

To anyone interested, I first found some crazy info about Psych and then cancer drugs too , it's pretty mind blowing to think a majority of what you see even in "legit" sources would not hold up.

That cancer one was the article I was going to mention. 47/53 landmark cancer papers couldn't be replicated by Amgen. I attended a lecture given by a ex-scientist from Amgen who was a part of that program to replicate these findings. He gave a brilliant talk highlighting specific (but anonymous to us) cases of data that was represented poorly. He stopped short of saying it was outright fraud, but to the room, it was quite clear.

The sad part is that some of the very high-impact journals (looking at you, Science and Nature) prioritize innovation so much that sometimes the actual data gets overlooked.

 

[Goro]

Indeed. It took almost 80 years for cancer surgeons to learn than lumpectomies were just as effective as radical mastectomies.

We make recommendations and treatment decisions based on the best available information while acknowledging that our knowledge is imperfect and some things that are thought to be true today may turn out to be untrue upon further investigation.

For example, in the early 1990s when many of you were in utero, the advice to pregnant women was to avoid eating peanuts or peanut butter and not to give peanut products to little ones in order to prevent sensitivity (allergy) to peanuts. Now, the latest advice is that kids should be exposed to peanuts at a very young age to prevent allergies. http://www.annallergy.org/article/S1081-1206(16)31164-4/fulltext

Part of life-long learning on the part of physicians is because we realize that knowledge evolves over time as we attempt to understand what is true.

 

[Goro]

NIH has the same mania in grant apps.

The sad part is that some of the very high-impact journals (looking at you, Science and Nature) prioritize innovation so much that sometimes the actual data gets overlooked.

 

[AnatomyGrey12]

As an editor of the journal that gave us Andrew Wakefield's bogus claims that the MMR vaccine was unsafe, I guess he should know.

A lot of crap science gets published, but the beauty of Science is that it's self-correcting, and things that aren't true or don't work get put into their place eventually. This comes from someone who spent a good deal of time trying to replicate some findings in a Nature paper (that I suspect were fabricated). No, I'm not sharing.

That's funny because in my Cell Bio class our prof had us go through a Nature paper and come up with our own opinions on the validity. The purpose of this was that ultimately the data presented in the paper did not support the conclusion presented by the authors, it actually showed the opposite