If you believe what you read in the peer-reviewed medical journals then almost all clinical trials of antidepressants have positive outcomes. To be precise, 94% of antidepressant trials appear to show that they are effective, according to an analysis published in the NEJM today.
But psychiatrist Dr Erick Turner suspected otherwise. The former FDA analyst took a more in-depth look at all recent clinical trial of antidepressants and found a very different picture. Out of 74 FDA registered studies of antidepressants, almost one in three (31%) went unpublished. And - surprise, surprise - the unpublished studies were predominantly negative ones that showed the drugs don’t work.
Publicaction bias: it’s a depressing scenario, if you pardon the pun. All but one of the positive trials found their way into medical journals, whereas all but three of the studies with negative or questionable outcomes either never saw the light of day or were published with a positive spin.
In reality, only 51% of antidepressant studies showed a positive effect. Or as Dr Turner told the Wall Street Journal: “There is a view that these drugs are effective all the time … I would say they only work 40% to 50% of the time, and [colleagues] would say, ‘What are you talking about? I have never seen a negative study.’ ”
The audit didn’t examine why negative trials don’t get published. “We cannot determine whether the bias observed resulted from a failure to submit manuscripts on the part of authors and sponsors, from decisions by journal editors and reviewers not to publish, or both,” the authors conclude.
The good news is that we now have clinical trial registers so that clinicians such as Dr Turner can monitor and report on publication bias. But the fact remains that patients are taking part in clinical trials for which the findings are not being disseminated in a timely and open manner to inform clinicians as to best practice.
A similar allegation has been made this week about the long awaited results from the trial of ezetimibe and simvastatin.
According to Dr Ben Goldacre writing on his Bad Science website, these delays and barriers to publication: “break the most fundamental and sacred contract in clinical research, the moral contract between patient and researcher. Patients are in the hands of researchers on the understanding they are taking modest risks to benefit mankind as a whole, to improve prescribing. “
Make sure the trials that are relied on have be registered before completion. If not registered then the findings should be closely scrutinised. In addition always ensure external validity when applying the findings of trials.
Posted by Dr Chris Topovsek on Thursday, 17 January 2008
If you believe what you read in the peer-reviewed medical journals then almost all clinical trials of antidepressants have positive outcomes. To be precise, 94% of antidepressant trials appear to show that they are effective, according to an analysis published in the 
