Statins and 'Publication Bias'

Interesting synopsis of article on 'publication bias', the 2005 regulation change, and statin/cholesterol studies:

 

In 2005, new regulations regarding the registration and publishing of clinical trials came into being.

 

A recent study discusses the fact that while earlier (prior to 2005) statin trials were overwhelmingly positive - trials since 2005 have have NOT been. Specifically, since 2005, all but one statin trial has yielded neutral or negative results. The one positive statin study, the JUPITER trial, has been the subject of considerable controversy; scrutiny of its findings reveal numerous implausible results which have caused some to question the role of the study's sponsors in the analysis and presentation of the data.

When vested interests are at play, studies can be designed (and interpreted) in a way that helps to guarantee a particular 'result' and conclusion. This has long been a known problem. For example, a recent report found that in studies of statins (cholesterol-reducing drugs), industry funded studies were 20 times more likely to report positive results for drug made by the study's sponsor. Additionally, study conclusions were 34 times more likely to be positive, too.

Another study of 546 trials found that 2/3rds of them were partially or fully funded by the pharmaceutical industry. Of the published trials of these studies, approximately 85% reported positive results. Rather slanted, wouldn't you say?

One potential explanation for the discrepancy in these publication figures is what is known as 'publication bias' - the practices of delaying publication or publishing some results but not others if they don't help the company's bottom line. For example: the 'ENHANCE' study, which found that two cholesterol-reducing drugs caused more blockage in arteries than one, required a US Congressional enquiry to force the publication of the results - two years after the trial ended.

The problem with publication bias, obviously, is that it can give us a very skewed view of reality. Some people may want that, of course. It's gratifying that these practices are being exposed and that it will be increasingly more difficult for drug companies and others with vested interests to engage in them.

(Derived from Dr. Briffa's article 'Increasing evidence shows that drug companies like to be selective about what they publish' , www.drbriffa.com)

Carl

http://nancycarl.blogspot.com/

http://picasaweb.google.com/carlrcraw