Nearly 60% of the clinical trials that the FDA reviewed during its approval processes for new drugs between 1998 and 2000 were not published in peer reviewed journals within 5 years after the drugs went to market, according to research published in PLoS Medicine.
24% of the non-published studies were “pivotal” trials focusing on safety and effectiveness in humans. Within this cohort, the PLoS study revealed publication bias: trials showing significant benefits for a new drug were more likely to be published than those showing no benefit.
Pharmaceutical companies sponsor the studies in question and they get to decide whether to pursue publication of trial results or squelch them.
PLoS study investigators concluded that Big Pharma’s failure to publish the results of all trials amounted to “scientific misconduct” that “harms the public good” because it prevents physicians from making informed decisions on behalf of their patients.
Things may be about to change at least a little however. The 2007 FDA Amendments Act mandated that substantial detail from all trials used in support an FDA approval process must be posted on the NIH clinical trials site. This is a good step, but maybe not enough. Clinicians may or may not recognize the unique value of the newly informed NIH site, much less use it. And it doesn’t address the problem with publication bias at all.