New generation antidepressants achieve almost no benefit compared with placebo in mild to moderate depression, with slightly more benefit in severe depression but only because of less response to placebo, a meta-analysis of clinical trial data has shown.
Researchers analysed all available data from clinical trials submitted to the US Food and Drug Administration for the licensing of four selective serotonin or serotonin-noradrenaline reuptake inhibitors—fluoxetine (Prozac), venlafaxine (Efexor), nefazodone (Serzone), and paroxetine (Seroxat, Paxil).
They analysed the degree to which people improved in relation to the initial severity of the depression in people randomised to drug or placebo.
Results showed almost no difference between the effects of drug treatment and placebo at moderate levels of initial depression, rising to a relatively small difference in patients with severe depression. On average, the antidepressants improved the score on the Hamilton scale of depression by 1.8 points more than placebo (PLoS Medicine 2008;5:e45).
The effect of drug treatment reached conventional criteria for clinical significance, defined by the National Institute for Health and Clinical Excellence as a difference in score of three points, only for patients at the upper end of the very severely depressed category, who had baseline scores of more than 28.
Additional analysis showed that these patients responded less well to placebo than patients with less severe depression, rather than responding better to antidepressants.
Irving Kirsch, professor of psychology at the University of Hull and lead author of the study, said, “Using complete datasets and a substantially larger dataset than previously reported, we found that the overall effect of new generation antidepressant medications is below recommended criteria for clinical significance.”
He added, “Efficacy reaches clinical significance only in trials involving the most extremely depressed patients, and this pattern is due to a decrease in the response to placebo rather than an increase in response to medication.”
Although the trials reached statistical significance for the effects of the drugs compared to placebo, as required by the FDA for drugs to be licensed, they did not reach clinical significance, he explained. “Maybe these drugs should be evaluated in terms of clinical criteria as well as statistical criteria,” he said.
He noted that the analysis failed to show any difference between the drugs included in terms of their efficacy or in the relation between severity of depression and efficacy.
The study was based on full datasets for 35 clinical trials, including unpublished data, involving a total of 5133 people with depression. The results were obtained from the FDA under the Freedom of Information Act. By including data from unpublished as well as published trials, the researchers set out to avoid any bias that might arise from some trials not being published. They were unable to include sertraline (Zoloft, Lustral) and citalopram (Cipramil) in their analysis because they had originally planned because they were unable to obtain all the relevant data for these drugs.
Professor Kirsch noted that an internal memo from Paul Leber, a member of staff with the FDA at the time, which was included in the information he analysed, said that doctors, patients, and payers ought to know that many trials failed to show a significant difference between antidepressants and placebo.
In a statement, Eli Lilly, the company that markets fluoxetine, said, “Extensive scientific and medical experience has demonstrated that fluoxetine is an effective antidepressant. Since its discovery in 1972, fluoxetine has become one of the world’s most studied medicines.
“More than 40 million people suffering from depression have been treated with fluoxetine in over 100 countries around the world. Lilly is proud of the difference fluoxetine has made to millions of people living with depression.”