Thursday I got an odd mail from Cyagen.
Cyagen offered me $100 or more in reward for citing their animal model services in an article. Even more surprising the reward was dependent on the impact factor of the journal I was publishing in.
Cyagen gives the following example:
If you published a paper in Science (IF = 30) and cite Cyagen Biosciences, you will be entitled to a voucher with a face value of $3,000 upon notification of the publication (PMID).
Thus the higher the impact factor the higher the reward.
Click to enlarge
First I thought this was link bait, because the mail was send from vresp.com. (spam reputation) and not cyagen.com.
But tweets from @doctorsdilemma and @PhytophthoraLab pointed me to this advertorial at Cyagen’s website.
It was real.
This doesn’t feel right, because it “smells like a conflict of interest”.
In clinical research we all know where conflicts of interest can lead to: biased results. Or according to Corollary 5 in the famous Iaonnides paper “Why Most Published Research Findings Are False”:
“The greater the financial and other interests and prejudices in a scientific field, the less likely the research findings are to be true.”
We expressed our concerns on Twitter but it took a while for Cyagen to respond to our concern (see Facebook-Q&A *).
Meanwhile Ben Goldacre from BadScience and Retraction Watch wrote a post about Cyagen’s “pay-for-citation program”.
Ben Goldacre emphasizes the COI-problem:
I would imagine that this is something journal editors will be interested in, and concerned by. We worry about “conflict of interest” a lot in science, and especially medicine: if someone has shares in a company, or receives money from it, and their publication discusses that company’s products, then this needs to be declared in the paper. If you get funding for a study then, again, you must declare it. If you have received payment for making an academic citation, then in my view this is clearly a source of funds that should be declared.
The most problematic point, according to Goldacre, is that (part of) the researchers “have received funds from Cyagen, in exchange for an academic citation.” He keeps an archive of the 164 papers that cite Cyagen, so one can always check whether “any one of them has declared receiving money from the company”.
(because in his opinion they should)
Some of the commenters to Goldacres’ post go even further and state:
“It’s one thing to fund a study. It’s another thing entirely to pay someone to write what you want — that’s not science, it’s PR. You’re not being puritanical at all, this is corruption, plain and simple.”
and
“This is an old and very well researched problem. TV makers get paid to have the popular detective drive a certain make of car, scoff a certain brand of drink and so on. That this kind of advertising is not innocent and that it works is beyond doubt. The media have established firm rules about this, science journals need just copy them.”
Although Retraction Watch lets both sides be heard, the title of its post “Researchers, need $100? Just mention Cyagen in your paper!” suggests researchers are being bribed by Cyagen to mention its product anywhere in the paper.
I agree that the Cyagen incentive is a (very curious) marketing trick, but it has nothing to do with ghostwriting or corruption and IMHO the authors of the 164 papers are not to blame and need not declare anything.
Let me explain why.
- First, as everybody familiar with benchwork knows, authors are required to mention the brand and source of the used reagents and tools (including strains of mice). This is for transparency and reproducibility. There can be huge differences between products and it is important to know which enzyme, primer or mouse strain someone has used.
- Authors only MENTION the products, reagents etc. that they have used. Mentioning a brand doesn’t mean an endorsement, although you probably wouldn’t choose a product if you thought it worked suboptimal.
Indeed the first two papers on the Cyagen citations list both mentionCyagen only ONCE in the materials and methods where it belongs
- The product itself is not being tested, it is just being used, something else is being tested. (it would be different if the discount was for a diet supplement that was being tested)
- The authors don’t receive money, they receive a discount on their next purchase. This is common practice in the lab when you are a regular customer.
- You don’t write COI’s for getting (normal) discounts.
Still, I do think that this incentive is rather inappropriate and not a very wise thing to do (even from a marketing point of view).
It is not the discount that is problematic, neither is the mention of the product’s name and brand in the articles (as this is required).
It is the coupling of the discount (erroneously called “reward”) to “a mention” (erroneously called “citation”) that is not a proper thing to do and even more so, the coupling of the height of the discounts to the impact factor of a journal.
The idea behind it is that Cyagen not only wants to thank its customers, it wants to “increase the number of publications featuring Cyagen as a product or service provider to add strength to our company’s reputation and visible experience”.
Of course, Cyagen’s reputation is not linearly dependent on the number of the publications and certainly not the citation score. Reputation depends on other things, like the quality of your product, client satisfaction and … your reputatIon.
It now seems that the entire marketing campaign will just have the opposite effect: it will harm Cyagen’s reputation (and perhaps also that of the researchers).
My advice to Cyagen is to directly remove the incentive from their website and stop the unsolicited mailing.
And please, Cyagen, stop to defend your P(C)R-tactics, just admit it was a wrong approach and learn from it.
* Later I found out that the a Q&A on Facebook had been extracted from an interview a blogger was still in the process of doing with a product manager of Cyagen. Apparently social media etiquette is another thing Cyagen doesn’t master very well.
Update post:2015-08-16
Laika's MedLibLog 
Leave a comment