An article in today’s Wall Street Journal reports that retractions of medical studies are on the rise. For a variety of reasons, including outright fraud, more and more publications are retracting published studies. One instance cited in the article dates to 2003 and that study prompted a change in the way many doctors prescribe anti hypertensive medications. Turns out it wasn’t true and instead of helping patients, the combination may in fact harm them.
Since 2001, while the number of papers published in research journals has risen 44%, the number retracted has leapt more than 15-fold, data compiled for The Wall Street Journal by Thomson Reuters reveal.
That’s quite an increase. The article mentions Andrew Wakefield and his now discredited research on a link between autism and vaccines, but that’s not the only case of bad research passing through the process and The Lancet is not the only publication to be duped.
There is a lot of competition for money for research grants and even more pressure for researchers to produce a break through study. Human nature being what it is, some people are going to yield to the temptation to “fudge” things to get a study published. It’s fraud and should be treated as such both civilly and criminally. In addition it undermines the credibility of the researchers and confidence in the research itself.
The article contains more details, but requires a paid subscription so a link won’t be of much benefit. There is however a blog that tracks retractions. Appropriately enough it’s called Retraction Watch. That there needs to be such a blog should be a cautionary tale about blind reliance on published research, especially a single “breakthrough” study.
Gregor Mendel would be proud.
Scientists are supposed to follow the science where ever it leads, but some seem to follow the science only when it leads to money.
As with paramedics, when we start increasing the numbers to satisfy political goals, we start scraping the bottom of the barrel.
As more and more people publish research, more of it will come from less qualified, less competent, less conscientious people.
This also reflects on the quality of the education that fails to produce an understanding of science that would lead to more competent research.
The peer review system seems to rely on volunteer peer review and maybe we need to pay for a higher quality of peer review. The peer review system certainly has a lot of flaws. While many of these should have been caught by peer reviewers, how many were published in journals that have to rely on what cannot pass peer review at any major journal?
Some of these retractions are due to plagiarism, rather than any problem with the data. While this is an ethical problem, it says absolutely nothing about the results of the study, which do not depend on the wording of parts of the the paper. If a paper that demonstrates that a drug is an excellent treatment has portions that are plagiarized, that does not mean that the drug does not work. It does mean that we should look very closely at the data and be suspicious, but 2 + 2 does not equal 5 because a paper contained some portions that were cut and pasted by an assistant, while the primary investigators were not aware of the copying.
If research were better done, I would not have such an easy time finding fault with so many published papers.
.
This article has been discussed on Respectful Insolence, a blog written by a doctor who is both a cancer surgeon and researcher. Scientific fraud and journal article retractions.
Some important points that he makes:
The highest quality journals appear to have been maintaining their high level of quality.
–
0.035% of all science papers?
That could be restated as 99.965% of all science papers are not affected by this.
Even Ivory Snow isn’t this pure.
This is a reason to be skeptical of research?
No.
We should always read everything with skepticism, but is there anything else with such a low rate of fraud?
.