March's PharmedOut Fodder
Problems with Clinical Trials
We rely on clinical trials of drugs to be held to the highest scientific standards. FDA uses trial results to approve treatments for their safety and efficacy, and doctors trust peer-reviewed journals to share accurate information about new treatments.
But in February, and in many months before that, we have repeatedly heard about trials that aren't up to par, whether in their design or reporting of results.
Here are three top ways clinical trials can be manipulated:
1. Industry influences study reports. This was a primary concern raised in Dr. Robert Califf's nomination for FDA commissioner. His research institute at Duke University took ample pharmaceutical industry funding, and during confirmation hearings, he was questioned about industry's role in studies.
Califf explained that while drug companies consult in study designs, academic researchers have the "final say." That may be, but the link between industry funding and more favorable results has been proven repeatedly.
2. Not reporting at all. Since the U.S. does not require the registration or reporting of all clinical trials, researchers often is not published. Last month, it became apparent that this is systemic at academic institutions: One-third of clinical trials conducted at 51 top U.S. universities and academic hospitals were never published in a peer-reviewed journal or government registry, according to a new BMJ study. Two-thirds of all the trials were not published within two years.
As one of the study authors wrote in NPR, "Not reporting results violates the basic principle of the scientific method. It hurts patients, society and science. It also dishonors the people who gave their consent and bore the risk of participating in the studies. Who would agree to be in a study with the knowledge that the results wouldn't be shared?"
All clinical trial results should be reported.
3. Outcome switching. Before starting a clinical trial, researchers are supposed to pre-specify which outcomes they are studying. Doing this makes it easier to see whether outcomes have been switched. Researchers shouldn't pick and choose which outcomes to reveal.
Journals are supposed to check for outcome switching, but too often, they do not, and cherry-picked outcomes get published, presenting false impressions of drugs. As the COMPare project reports, a recent PLOS One review of 137 randomized controlled trials found that 18% had some form of discrepancy relating to the primary outcome, 15% changed the definition of the primary outcome, and 64% of the trials had discrepancies between pre-specified and reported outcomes for non-primary endpoints.
The FDA, academic institutions, and medical journals must commit to raising expectations for clinical trials. If they don't, who will?