Skip to Main Content

This weekly column offers opinions on the latest pharmaceutical industry news.

The next time you take a Merck medicine and it doesn’t work, don’t expect a refund. The drug maker doesn’t offer a money-back guarantee.

But there are times, a high-ranking Merck official believes, when the company deserves to get its own money back. That’s when it licenses potential treatments from universities based on promising results from early clinical trials, and then can’t reproduce those results in subsequent experiments.

Collaborations between industry and academia fail to replicate initial data way too often. This undermines public confidence in science and wastes money and time as companies pursue dead ends based on flawed research.  As Merck’s chief medical officer sees it, a full or even partial money-back guarantee from universities could fix this.


“With such assurance, companies could proceed with a project more rapidly and more frequently,” Dr. Michael Rosenblatt wrote in a recent essay for Science Translational Medicine, a journal that’s widely read by researchers in academia and industry.

In Rosenblatt’s view, this kind of monetary incentive to deliver replicable data could act as a safeguard against sloppy investigators and institutions — “precisely the leverage points for change,” as he put it.


It’s a provocative idea, but it’s not workable and it risks alienating the scientific community at large.

Certainly there is much at stake for the pharmaceutical industry. As big drug makers continue to prune their own research and development budgets, they are increasingly partnering with academia on research. Academic labs are now, in effect, in the business of generating drug leads.

Rosenblatt pointed to recent studies showing that key findings in half or more of published reports in some scientific fields could not be reproduced. He also cited research showing that, on average, it takes two to six scientific personnel anywhere from one to two years of work in an industry lab to try to reproduce original studies, all at an estimated cost of around $500,000 to $2 million.

“There is a crisis occurring with irreproducible experiments,” said John LaMattina, a former head of R&D at Pfizer and now a senior partner at PureTech Health, a venture capital firm. “All of the issues that [Rosenblatt] raises are correct.”

That may be so, but Rosenblatt’s money-back guarantee solution remains contentious.

“I don’t know how else to say it, but I think the idea is crazy,” said Ken Kaitin, director of the Tufts Center for the Study of Drug Development, a nonprofit think tank in Boston that’s partly funded by the pharmaceutical industry.

“Academics don’t always do science in a way that is designed for commercial development,” Kaitin said. “Unfortunately, this implies you’ll punish the researchers for providing data that can’t be reproduced.”

Rosenblatt tries to sidestep this problem by giving universities options, such as replicating work before collaborating with a drug maker. This may encourage them to become more “careful” and convey only results in which they have “a high degree of confidence in order to avoid retraction and financial loss,” he wrote in his journal commentary.

But accomplishing this goal would require “some very clear definitions about what is reproducible, which may be difficult because there are different ways to validate results,” said Dr. Frank David, a former Wall Street consultant who now runs the Pharmagellan consulting firm in Milton, Mass.

And even if the research community could agree on a set of reproducibility standards, David added, “building what amounts to an insurance market doesn’t seem like the right solution.”

For now, Rosenblatt is simply floating an idea. A Merck spokeswoman said the company is not actively pursuing money-back guarantees. But clearly, the top doctor at the company is sending signals that the drug maker wants things done differently. And since Merck carries a lot of weight, universities are likely to pay close attention.

Just the same, clawbacks are more confrontational than collaborative. More important, the notion may undermine academic independence if universities worry about pleasing would-be industry partners.

If Merck wants better outcomes for its medicines, the company shouldn’t be singling out academic scientists. Why not look in the mirror and offer its own money-back guarantees for those of us unsatisfied with its drugs?

  • As usual crocodile tears keeps pouring down, as the cost of replicating a research, is nothing more than a fraction of bonuses that executives and CEOs cash every years for unaccountable performances.
    Indeed, the main problem lies in the plethora of not so qualified consultants, and executives, who do not know anything abount biology and medicine (and probably even how to make money). All people, handsomely paid to evaluate a priori whether a technology/discovery is actually worth pursuing. And why not, many “qualified” scientists with “solid industry experience” that might be just unable to replicate the results.

  • Better yet, why not offer money-back guarantees for when Big Pharma products turn out to be dangerous killers? Merck put so many people at risk – a risk they knew all about beforehand, from Ed Scolnick and Alise Reicin up to the CEO – with that Vioxx junk medicine. People who got injured had a chance at compensation, but regular purchasers got nothing for taking a huge risk.

  • merck fought vioxx rebates and continues to do so.

    of course vioxx just caused heart attacks and alzheimers and strokes and killed people.

    oh ya it was based on science so bad even the right wing nejm issued an official reprimand.

  • Dr Rosenblatt was referring to pre-clinical experiments, not clinical trials, in his essay (ie, basic and translational science). Historically, academia and industry has counted on the peer review process to validate published results. Further, young investigators used to routinely publish confirmatory, or rebuttal works. However the complexity of today’s biology, and perhaps greater focus on novelty, has resulted in many works being unchallenged. An alternative plan could be to ask the publishing journals to require confirmation in the hands of a second lab prior to publication.

Comments are closed.