I agree completely with this assessment that the non-publication of trial results that may damage the prospects for a drug or device is a problem. I agree also that the solution might be to have penalties for non-publication.
Drug and device companies don’t want unfavorable data published because that will complicate their marketing strategies. Scientists don’t want poor results published because it might jeopardize their grants and promotion gravy train. It’s a sad situation all around, because patients, families, and doctors (most of whom are pretty clueless about trials results anyway, unless they were written up by USA Today) don’t get a full picture.
This leaves decision makers vulnerable to marketing campaigns and to the deals concocted between health plans or pharmacy benefit managers (PBMs) and the industry. Even those deals are, by definition, skewed, because the the coverage and reimbursement decisions (including things such as placement on a particular tier of covered drugs) is predicated on incomplete data.
This problem is not limited to just clinical trial results. I remember a paper I worked on many years ago, involving a completely forgettable dermatologic product. The drug’s sales manager had been tasked with getting a paper published showing the cost-effectiveness of the product, as a prelude to marketing it to both dermatologists and health plans and PBMs.
Well, the consulting team I was on assembled the model and the data, and the results were, shall we say, very unimpressive. We told him that were we to write a paper based on the model as developed, the entire enterprise would cause people to shrug their shoulders.
On a teleconference that I can remember like it was yesterday, this man-child with the maturity of a high school student, huffed and puffed, audibly shuffled papers on his desk, and then ripped the consulting team leader for building a lousy model that would not produce a publishable paper.
The consulting team leader, unwilling to endure a loss of the contract, committed to reconfiguring the model, and, eventually, a paper was produced in a published in a journal than I am sure no one reads. But, hey, it’s a peer-reviewed, published paper.
I’ll be writing more about the army of consultants who advise the drug industry on marketing and messaging how they have raised deception to an art form.
The non-publication of a significant proportion of trials skews the evidence base, particularly if failed trials are less likely to be published.
Failure to publish is also a significant ethical problem.