Science
17 January 2020 Vol 367, Issue 6475
http://www.sciencemag.org/current.dtl

Feature
Transparency on trial
By Charles Piller
Science17 Jan 2020 : 240-243 Full Access
Many clinical trial results aren’t posted publicly, as U.S. law requires—and a promised crackdown has fizzled.
Summary
Companies, universities, and other institutions that conduct clinical trials are required to record the results of most of them in a federal database, so that doctors and patients can see whether new treatments are safe and effective. But a Science investigation has found that many persist in not reporting those results, and the U.S. Food and Drug Administration (FDA) and National Institutes of Health (NIH) are doing little to nothing to enforce the reporting requirement—despite recent promises to the contrary. While many pharma companies have improved their reporting of trial results over the past few years, a large number of universities and academic medical centers continue to have bad records. Ironically, NIH itself is part of the problem—its top institutes for clinical trials have a poor record of reporting results of those trials for which they are responsible. Science checked all 4768 trials whose results have come due under legal requirements finalized by NIH and FDA 3 years ago. Overall, sponsors violated the reporting law more than 55% of the time. Yet FDA has never fined a trial sponsor for noncompliance and NIH has never publicly named or withdrawn a grant from violators, although such actions were vowed in 2016 when ClinicalTrials.gov’s requirements were clarified.