clinical trials- subsequent publications

[Jani]

---

Members do not see this ad.

This was posted previously with not much response so am writing again,

With research clinical trials does the investigators/ co-ordinators have the right to publish the results on their own?

If so what precautions do you take to ensure the pharmaceutical company sponsoring the trial is going to be happy with your own production.

Thanks for any comments or views.

 

[mpp]

I would think it would depend on the agreement with whomever is paying for the research. If they are paying, they obviously have a right to read whatever it is you are going to publish about their product and perhaps refuse to have it published if they don't want it. If you want the pharmaceutical company to be happy, show them what you are publishing.

 

[jalabert]

What do you mean by happy with your own production?

The issue of publication is going to vary between drug, stage of development and sponsoring organization. Some investigators will have the right to publish, some will not, and in some cases, it depends on the results. If results look good, the company will sign off on publication, if not, publication is stifled. This is typically negotiated at the time of contract development. For example, some academic institutions will not negotiate a contract that stipulates right of refusal to publish in the event of unflattering results, OTOH, other institutions will.

With regard to your own production, if you mean, how do you make sure the company is happy with your results, you can't. What you can do is make sure there are no holes in the research so that, if the results don't look favorable, it could not be ascribed to poor design or control. Conversely, if results look good, you want them to be reproducible or hold up to intense scrutiny. Companies are always happy with your work when the results look favorable for their product, but this doesn't always happen. So, you just have to be diligent, ethical, stick to your guns and let the chips fall where they may. Some might be tempted to creatively analyze data, or use some other technique to produce more favorable results, but this can ultimately come back to bite you.

 

[Jani]

thanks for your comments both, the point I was trying to make was here we are trying bring something to print which is part of a large randomised study. The pharmaceutical company introducing the drug has its own way of interpreting and releasing data say for eg : combined blood pressure tablets Vs monotherapy etc. How do you make sure the numbers ( more importantly clinical significance) dont contradict.

secondly do you go through the ethical committee etc.

 

[Kalel]

My understanding is that the FDA regulates many of the pharm company's clinical trials because they are ultimately used to determine whether or not the drug is approved by the FDA or not. Drug companies frequently do publish trials after they have been approved, these trials usually are *much* more shady; some border more along the lines of being an infomerical rather then an actual research study. It's unfortunate, because no practioner has the time to read every study that is published in his or her specialty. This can lead to a lot of misinformation being fed to them by pharm reps and physicians that they hire ($500-1,000 speaking fee) to talk about their "studies". Sometimes, editorial boards of respected journals are able to screen these studies out by not publishing them, but there are still a number that get through and are somehow published (pharm co's support journals through ad payments). Some classic examples of this are studies between the same classes of drugs like zyrtec vs claritin vs clarinex, or the studies illustrating the "difference" between the PPI's. They are often poorly run, heavily biased, and sometimes make conclusions that have little statistical significance. One of my psych professors told us that many of the antidepressants used to run several clinical trials before the FDA approved them because they had difficulty getting statistical significance between their drug and the placebo. They don't tell you about the prior trials that showed no statistical difference in their advertising or in their published study, so it's difficult to tease out what is authentic and what is not. Another problem is that many of the pharm co's run their trials through private doctor's office's, with the doctors being paid to recruit their own patients for clinical trials. This leads to the conflict of interest of the doctor wanting to continue receiving money from the drug company to supplement their salary (which can be a significant amount, I know of a private GI doc who has a nurse researcher and supplements his income with ~100K from drug company research) and producing valid results for the medical community. I'm not anti-pharm company research though, because they provide the funds for most of the legitimate clinical research as well.

 

[jalabert]

Kalel brings up a lot of points that confuse the issue to a certain extent. You will get no argument that there are some less than ethical physician researchers "producing" data that is ultimately beneficial to their pocket book. My original point was that a researcher should play it straight, then you don't have to worry about making people or companies happy. Sure you want to continue obtaining research money, but if you are good and have a good rep, money will come from some source.

The point about not publishing or promoting previous non statistically sig data, is a bit of a grey area IMHO. I see no problem in a series of studies that tend to show practical significance (say p of 0.1) leading, ultimately, to a study showing statistical significance of p < .05. Just because the p was too high to be statistically sig in previous studies does not mean there is no practical significance to a particular treatment. Learn from mistakes, or poor design and improve to get stat significance. That is how science is done. OTOH, if there were a series of studies done that did not even approach statistical sig, and out of the blue one shows up with a p<.05, I would be sceptical. With regard to the first scenario though, stats can be a double edged sword.

 

[jalabert]

Originally posted by Jani
thanks for your comments both, the point I was trying to make was here we are trying bring something to print which is part of a large randomised study. The pharmaceutical company introducing the drug has its own way of interpreting and releasing data say for eg : combined blood pressure tablets Vs monotherapy etc. How do you make sure the numbers ( more importantly clinical significance) dont contradict.

secondly do you go through the ethical committee etc.

It is difficult to comment since, due to the nature of your question, you cannot mention specifics. If you have a sponsored study, though, I would imagine your advisor, or whom ever is running the study, will have a better idea of how to handle the specific situation than anyone who will be posting here. Honestly, if you are worried about results contradicting what the pharmaceutical company is trying to market, you need to think about your motivation for research. (I will step up onto my soapbox for a second) Research should be done to find out the truth about whatever the research question is. If the truth contradicts marketing claims, so be it. Again, I go back to my original statement that, if the study is tight and well designed you cannot worry about how happy the pharmaceutical compay will be. (getting off the soap box). Now, in the real world the researcher doesn't want to bite the hand that is feeding him, but if results are contradictory to marketing claims you need to determine why that is. If the study had holes, and is in effect, crap, put the results in a box and chalk it up to experience. If it is tight though, it needs to see the light of day. Depending on the nature of the results, you may be saving some lives, which is more important than contract money.