Is it ethical/legal to withhold information in clinical publications?

[Lost In Transcription]

Ah so that actually what is routinely going on then. Thank you. That's all I need to know. I kinda figured that was the case because some people seem to be evasive about it.

Routine is a bad word choice. Not unheard of is more accurate.

And doesn't change the justification.

---

Members don't see this ad.

 

[Lawpy]

This is the second time you've posted about questionable decisions in research. Are you a creative troll or are you surrounded by questionable people?

It's likely the former, as they like posting contrarian threads/replies. So far I've seen these two questionable research ones, an anti-vaccination argument, and a flat-earther argument.

Don't forget about the following gems

Proposal for a better MCAT
MMI - consent
https://forums.studentdoctor.net/threads/medical-school-ranking-and-sanctuary-cities.1235141/

 

[Lost In Transcription]

Don't forget about the following gems

Proposal for a better MCAT
MMI - consent
https://forums.studentdoctor.net/threads/medical-school-ranking-and-sanctuary-cities.1235141/

Oh there's more. Especially in the SPF..

 

[TelemarketingEnigma]

I would like to bump this thread to remind people that statistics and confident intervals mean jack**** if the underlying assumptions are questionable, no matter how much resources are spent, how authoritative is the source.

This seems relevant to this thread. these posts are entertaining.

 

[deleted781806]

Don't forget about the following gems

Those are quite a doozy too, geez.

But yes, as above, just because that the current scientific climate doesn't encourage studies to confirm reproducible results in practice (somewhat due to cost prohibitions) does not mean that we can just throw the principle of "your results should be reproducible, also don't lie in your papers" out the window.

 

[deleted781806]

And you are quite a miser in your confident intervals, +/-20 is more on point. With that in mind, do you think then that the claim of lack skill is legit? Case in point. This is an semi-unrelated paper. I crossed some labels out because reasons.

Also, I am a little upset about the mild deception in posting this data table, which is sinking to a lower level than I would have expected. If you look at the original publication (removing those fields was pointless fyi), those "primary outcome score" columns are presenting the data as means +/- standard deviation, NOT the 95% confidence interval. The scores were a composite of different analyses applied to the treatment groups, and the p-value was calculated using a non-parametric Wilcoxon signed-rank test. The confidence intervals that were actually shown on the table obviously overlap and were not reported to be statistically significant.

Your portrayal of large standard deviations as large 95% confidence intervals to the scale of "+/- 20" in the given example table is a blatant misrepresentation of what is going on, and it's ridiculous that you would try to pass it off as such.

*EDIT*: Furthermore, your portrayal of "On that table, each of the 250+ data point costs hundred of thousands of dollars and tremendous coordination." is a blatant falsehood, there were only 134 cases studies on that particular table. Trying to say "[to] deviate from their non-human models..." is another falsehood, the subjects being studied were human babies (like geez dude), as well as saying the paper is 10+ years old even though it was published 6 years ago is just another white lie in an overall broken argument and misrepresentation of good data and good science produced and published recently.

tl;dr: Your given example is flat-out misconstrued through multiple lies and I would post the paper if it wasn't for some obscure TOS violation against posting links.

*Further edit*: Also, to use a study that was carried out across FIVE different clinical institutions with 102 contributors (listed in the supplement, 14 co-authors listed on the paper, representing the group) as an example of some single hospital trying to convince patients to come to them is equally ridiculous just on a fundamental level.

 

[Lucca]

Moving to topics in healthcare

 

[Stagg737]

The paper is like what, 10+ years old and is still being quoted/studied. The authors of that table are so influential that if you want to apply for a grant to further research the methodology and deviate even a tiny bit from their non-human models, you can kiss goodbye to that grant. Everything goes through them. So no, it is not a crappy paper. And,again, I have already stated that it wasn't the paper in the OP.

Did you mean all valid results must be reproducible or that all results must be reproduced by other groups? There is a significant difference here. If I publish something, and, to the best of my ability, detail every single steps, my paper would still be crap because I had 30 data points and each cost about 5k in, material only, to collect, that is not taken in accounts the practices to perfect the technique. In other words, if the paper is cost prohibitive (in labor or capital) to replicate in other institutions, it is automatically crap? I mean you are a med student I should have experience this but if the procedure calls for a closure of an anastomosis leak to be done within say, 2 minutes (lol), and there are like 3 people in the fields that can do it (with video evidence), 2 of whom are the authors, while the 3rd is not interested in the research so they won't do it, does it mean that such procedure is crap because it cannot be done by the lesser surgeons?

Slightly off-topic, is it also your opinion that any experiment done using the Hadron Collider is crap because there is only 1 in the world and only a few can access it?

I am not asking that to arguing with you. I really want to know. I find this insinuation that Western Blot and surgical cases are similar because they are both "data point" ridiculous.

First part: I wasn't talking about this specific paper. I was talking about the fact that you were trying to justify the paper's validity by indirectly stating "It's in a high-influence journal, so it's obviously a legit study". Which is an incorrect statement and a fallacy. I'm not saying the paper is bad, I'm saying the argument for defending the paper was bad.

As to the bolded, wtf are you talking about? I never inferred anything you wrote about here. I was saying that there is a difference between leaving out a variable that may be seen as irrelevant and reporting that they used some other procedure than what they actually used. Cutting edge research that is only being done at one institution or which isn't reproducible everywhere because of lack of resources or technology isn't the issue I was addressing. It has nothing to do with the study being reproducible. It has to do with straight up lying in the paper (you said that the study reported using saline when they used another fluid instead), which is unethical and illegal. The rest of the questions/arguments are irrelevant.

 

[wizzed101]

Also, I am a little upset about the mild deception in posting this data table, which is sinking to a lower level than I would have expected. If you look at the original publication (removing those fields was pointless fyi), those "primary outcome score" columns are presenting the data as means +/- standard deviation, NOT the 95% confidence interval. The scores were a composite of different analyses applied to the treatment groups, and the p-value was calculated using a non-parametric Wilcoxon signed-rank test. The confidence intervals that were actually shown on the table obviously overlap and were not reported to be statistically significant.

Your portrayal of large standard deviations as large 95% confidence intervals to the scale of "+/- 20" in the given example table is a blatant misrepresentation of what is going on, and it's ridiculous that you would try to pass it off as such.

*EDIT*: Furthermore, your portrayal of "On that table, each of the 250+ data point costs hundred of thousands of dollars and tremendous coordination." is a blatant falsehood, there were only 134 cases studies on that particular table. Trying to say "[to] deviate from their non-human models..." is another falsehood, the subjects being studied were human babies (like geez dude), as well as saying the paper is 10+ years old even though it was published 6 years ago is just another white lie in an overall broken argument and misrepresentation of good data and good science produced and published recently.

tl;dr: Your given example is flat-out misconstrued through multiple lies and I would post the paper if it wasn't for some obscure TOS violation against posting links.

*Further edit*: Also, to use a study that was carried out across FIVE different clinical institutions with 102 contributors (listed in the supplement, 14 co-authors listed on the paper, representing the group) as an example of some single hospital trying to convince patients to come to them is equally ridiculous just on a fundamental level.

You are so butthurt it's adorable! I don't know what to say because even though I know that you were talking about me, you were quoting someone else conjured by your delirium. I never said any of those. I wish I could see your face typing that^.^

First part: I wasn't talking about this specific paper. I was talking about the fact that you were trying to justify the paper's validity by indirectly stating "It's in a high-influence journal, so it's obviously a legit study". Which is an incorrect statement and a fallacy. I'm not saying the paper is bad, I'm saying the argument for defending the paper was bad.

Ah, but I my original quote you directed to some clowns who would probably jump in and bash that paper simply because it was I who posted it. And yes, I can slap them down by revealing where it came from, easily. It wasn't mean to be a general comment. But let me say this again for the x times: I wanted to show aldol some clinical data with a large deviation, among the papers I have on hand, that one seemed best: the amount of data was huge and the deviation was large. The way you wrote that reply indicated some... uhm... misunderstanding.

As to the bolded, wtf are you talking about? I never inferred anything you wrote about here. I was saying that there is a difference between leaving out a variable that may be seen as irrelevant and reporting that they used some other procedure than what they actually used. Cutting edge research that is only being done at one institution or which isn't reproducible everywhere because of lack of resources or technology isn't the issue I was addressing. It has nothing to do with the study being reproducible. It has to do with straight up lying in the paper (you said that the study reported using saline when they used another fluid instead), which is unethical and illegal. The rest of the questions/arguments are irrelevant.

I reread your comment, my bad lol. I thought you were saying that the only good sciences are those reproducible. Those trolls made me confused.

 

[freak7]

You are so butthurt it's adorable! I don't know what to say because even though I know that you were talking about me, you were quoting someone else conjured by your delirium. I never said any of those. I wish I could see your face typing that^.^

Would you mind not saying stuff like that? It's extremely condescending and rude.

 

[Seldon]

Can someone please kill this thread? It's all gotten really out of hand.

 

[deleted781806]

I don't know what to say because even though I know that you were talking about me, you were quoting someone else conjured by your delirium. I never said any of those.

Everything I placed in quotes was from previous posts by you, I did not exaggerate and I did not lie like yourself. Let anyone with eyes see for themselves.
The paper you referenced is free to view from the New England Journal of Medicine, and all of my points are easily verifiable.

 

[deleted781806]

I don't know what to say because even though I know that you were talking about me, you were quoting someone else conjured by your delirium. I never said any of those.

Please enlighten me as to which post of yours that I referenced was a fabrication of my "delirium".

I understand your point but consider these: there is only 2 hospitals in the entire US that can do this procedure. And this paper, I suspect, is meant to be a promotion to get patients in.

And you are quite a miser in your confident intervals, +/-20 is more on point. With that in mind, do you think then that the claim of lack skill is legit? Case in point. This is an semi-unrelated paper. I crossed some labels out because reasons.

On that table, each of the 250+ data point costs hundred of thousands of dollars and tremendous coordination.

The paper is like what, 10+ years old and is still being quoted/studied. The authors of that table are so influential that if you want to apply for a grant to further research the methodology and deviate even a tiny bit from their non-human models, you can kiss goodbye to that grant.

 

[wizzed101]

Please enlighten me as to which post of yours that I referenced was a fabrication of my "delirium".

The statistics you quoted in that first paragraph had nothing to do with what I intended to show. I merely wanted an example of clinical data with large CI. That's it. The deception is something your fit made up.

Plus, I stated very clear multiple times that the paper mentioned in the OP was not the NEJM paper. You have decided that you've been duped from the beginning for who knows what reason. If that is not delirium, I don't know what is ^^

Nitpicking: I said ~250 data point. So, I added up the 2 columns, it was a typo. You can see that later I said 150. I am sorry hurt your feeling because I said human = data point and made a typo . Your reaction to that is amusing though. And I mistook the time because I remembered the cases beginning in 2002/2003 or something. They only published everything way later. I am so sorry that I only read it once. But it seemed that you did your HW going through all of that. Good for you ^^

And lastly, the trial was based on an animal model, like how it should be. You can't seriously think that people ask for grant to repeat that? That's just a trivia that you need not know about, btw. It was just to show how influential the authors of the NEJM paper are.

Would you mind not saying stuff like that? It's extremely condescending and rude.

Why not? They come here to troll me. I can do whatever I want to retaliate.

 

[freak7]

Why not?

It's not nice.

They come here to troll me.

They don't.

I can do whatever I want to retaliate.

You can't.

 

[deleted781806]

I merely wanted an example of clinical data with large CI. That's it.

You didn't provide one. "Large" standard deviations =/= Large CI. That's stats 101. The difference in the two groups was also found to be statistically significant using sound methodology, not just "oh well the SDs overlapped, therefore p > 0.05" because that's not how statistics work in the slightest.

I stated very clear multiple times that the paper mentioned in the OP was not the NEJM paper

Yes, you stated very clearly that it was not the NEJM paper, but all argument that persisted after mentioning the graph was in direct relation to the NEJM paper and not the OP, as you *also* clearly stated later on.

And lastly, the trial was based on an animal model, like how it should be.

What trial? The NEJM paper, that you were referencing in the post concerning a "non-human model" was a study looking into prenatal vs postnatal surgical correction of spina bifida in infants. Really hard to see how you are going to stretch that as an "animal model". Quote from you would be the one that goes like this: "The authors of that table are so influential that if you want to apply for a grant to further research the methodology and deviate even a tiny bit from their non-human models". Unless your grammar was so *incredibly* shoddy that you used a pronoun wrong, you can't deflect that.

All argument post-table was centered around the NEJM table provided, you said it yourself. But now that you are caught in blatantly misrepresenting data to suit your narrative, you will, of course, deny that it was truly what you were talking about. You can't win the argument anymore so you will 1) use insults in order to make yourself seem superior to the situation and 2) will deny anything/everything you said was actually sincere or related to the actual discussion at hand. It's not hard to predict how you were going to proceed, and honestly it's a little disappointing that you continue to act in that way. So in any case, I will just let me posts stand in testament to my position and everyone will be able to see that you're just a simple troll, like we all knew anyways.

 

[wizzed101]

It's not nice.

que?

They don't.

Not an argument.

You didn't provide one. "Large" standard deviations =/= Large CI. That's stats 101. The difference in the two groups was also found to be statistically significant using sound methodology, not just "oh well the SDs overlapped, therefore p > 0.05" because that's not how statistics work in the slightest.

Did I just tell you I only read the paper once and that I picked that table because it had a large +/- number relative to the mean? Besides, the person I was responding to understood the point. I realized that the edited the wrong part of the table where there was a big overlap but was too lazy to redo it. And again, the point was made.

Yes, you stated very clearly that it was not the NEJM paper, but all argument that persisted after mentioning the graph was in direct relation to the NEJM paper and not the OP, as you *also* clearly stated later on.

? The only reason I kept repeating that was because other post someone came in and mistook the NEJM paper as the one in the OP. There is exactly one purpose for the inclusion of that table: it had a large +/- number relative to the mean.

What trial?

Did you actually read it or you were just raging while trying to nitpick me?

The NEJM paper, that you were referencing in the post concerning a "non-human model" was a study looking into prenatal vs postnatal surgical correction of spina bifida in infants. Really hard to see how you are going to stretch that as an "animal model". Quote from you would be the one that goes like this: "The authors of that table are so influential that if you want to apply for a grant to further research the methodology and deviate even a tiny bit from their non-human models". Unless your grammar was so *incredibly* shoddy that you used a pronoun wrong, you can't deflect that.

You have this habit of assuming people always refer to exact detail when it fits your narrative of deception. First, you assumed that the improvement I mentioned in the OP paper was the same as the alleged omission. Now you assume that I was talking about grants for research using human subjects (!). But I have already given you a hint: all clinical trials must be based on animal models.

All argument post-table was centered around the NEJM table provided, you said it yourself. But now that you are caught in blatantly misrepresenting data to suit your narrative, you will, of course, deny that it was truly what you were talking about. You can't win the argument anymore so you will 1) use insults in order to make yourself seem superior to the situation and 2) will deny anything/everything you said was actually sincere or related to the actual discussion at hand. It's not hard to predict how you were going to proceed, and honestly it's a little disappointing that you continue to act in that way. So in any case, I will just let me posts stand in testament to my position and everyone will be able to see that you're just a simple troll, like we all knew anyways.

What data did I mispresent? Did I even present any data? There is a difference between "here are the data supporting my point" and "here is an example of what clinical data look like." I used that table out of convenience but if you want to argue with me that clinical data typically have very narrow CI to the tune of 70+/- 2 as the person I was replying to purported, you are going to hurt yourself even more as I have demonstrated that I have no mercy. I think I am getting better at being civil. I didn't call you names or made many snide remarks ^^

And besides, I really just wanted that person to see that the variations in clinical results were huge. I think I succeeded.

All in all.

You can't.

Adorable

 

[deleted781806]

I have demonstrated that I have no mercy. I think I am getting better at being civil. I didn't call you names or made many snide remarks ^^

 

[Lost In Transcription]

@wizzed101 the sad pwrt is that every time you post, people are giving you a chance to redeem yourself.

You do yourself in. Then just none of us who know you are surprised.

It's beyond laughable that you think you've won in any sense any of these arguments. I'm always shocked by your deludedness.

 

[freak7]

Thank you mod gods.

 

[deleted781806]

Thank you mod gods.

Praise be.

(Now I just have to deal with the eternal embarrassment of stringing myself along for such a long time)

 

[freak7]

Praise be.

(Now I just have to deal with the eternal embarrassment of stringing myself along for such a long time)

That's pretty much why I kept my posts so short. I know a bait when I see one.

 

[Lost In Transcription]

Praise be.

(Now I just have to deal with the eternal embarrassment of stringing myself along for such a long time)

Eh, you'd be surprised at the number of students that read these things. Good to set the record straight (sadly).

 

[deleted781806]

Eh, you'd be surprised at the number of students that read these things.

That was the only reason that I continued for so long, at least others reading it might be able to learn something from all of this.

 

[freak7]

Eh, you'd be surprised at the number of students that read these things. Good to set the record straight (sadly).

That was the only reason that I continued for so long, at least others reading it might be able to learn something from all of this.

So we're in agreement? It's totally ok to lie and withhold information in a pub.

 

[Lost In Transcription]

So we're in agreement? It's totally ok to lie and withhold information in a pub.

You too get an A in reading comprehension!

 

[freak7]

You too get an A in reading comprehension!

Me read good!