Nature News pointed out a study in the September 2, 2009 issue of the Journal of the American Medical Association (abstract free, article will cost you) about the failure of clinical trials to be registered at ClinicalTrials.gov. There’s no legal obligation to register (it is a requirement for publication in certain medical journals), but hiding the presence of these trials makes it easier to not disclose negative results. As the study and the Nature piece indicate, there’s also a disconnect between the stated goals of registered trials and how they are written up in journals. The switch in stated goals is rarely caught, and much like the lack of registration, clouds the knowledge gained from research.
There’s a two-fold problem with this. First is the ability to know what drugs aren’t effective so patients don’t take them. Second is the impact on the state of scientific knowledge in the field. We can learn from both, as I’ve tried to suggest in earlier posts. But without knowing what didn’t work, it’s hard to learn from it. That’s why the registry makes sense, and why the lack of registrations is unfortunate. It’s not fraud, it’s not exactly misrepresentation, but it is the hiding of knowledge that can help us learn.