University of Groningen: Pharmaceutical companies still paint too rosy a picture of antidepressant effectiveness

More than one million Dutch people use antidepressants. Before these drugs are put into the market, their effectiveness is extensively studied in clinical trials. Although these trials are considered the gold standard, in 2008 they were shown to paint far too rosy a picture of the effectiveness of antidepressants. Researchers concluded that there was a ‘publication bias’: especially research showing a positive effect was published, while research showing no effect was published much less often. And when the latter studies were published, the results were usually written down in such a way that it appeared that the drugs did work.

New research
Fourteen years later, epidemiologist Ymkje Anna de Vries and her colleagues investigated whether this publication bias has decreased. They recently published their findings in the scientific journal PLOS Medicine.

In the past fourteen years, four new antidepressants have entered the American market. De Vries examined thirty registered clinical trials for these antidepressants. Fifteen of these trials found a positive effect, meaning that the drug under investigation worked significantly better than a placebo. Although all the studies in which a positive effect was found had been published, only seven of the fifteen studies without a positive effect had been published correctly and transparently. Six studies had not been published at all and two studies had been published but falsely reported a positive effect.

The study by De Vries shows that the situation is better than in 2008. At that time, far fewer studies in which antidepressants did not have a sufficient effect were actually published. De Vries: “But there is still a problem. Because of this bias, antidepressants may work less well than you would think.” According to the researcher, it is important that pharmaceutical companies are closely monitored: “These companies mainly have a profit motive, so there should actually be some kind of watchdog behind them to ensure that they report their research results as neatly as possible.”

Comments are closed.