Study: More Than 50 Percent Of Drug Approval Trials Go Unpublished

A new study is out in PLoS Medicine wherein researchers present evidence that 57 percent of drug studies submitted to the FDA for approval of a drug go unpublished. Conversely, 43 percent of drug studies examined by the researchers were published. The study examined 909 studies related to 90 drugs approved by the FDA between 1998 and the end of 2000.

This study was apparently spurred in part by recent studies establishing that half of anti-depressant approval trials had gone unpublished. It looks like the same kind of publication bias that exists there exists with all drugs. So you have to wonder if the effects sizes of all drugs are wildly overstated, as were anti-depressants.

To me, this is simply unacceptable. I hope this study gets a lot of attention.