Efficacy of new generation antidepressants: differences seem illusory

Area of Science:

• Psychiatry
• Clinical Psychology
• Evidence-Based Medicine

Background:

• Cipriani et al. used network meta-analysis to assess 12 new-generation antidepressants for major depression.
• Their findings suggested differential efficacy among antidepressants, but methodological criticisms were raised.

Purpose of the Study:

• To evaluate the reliability of network meta-analysis findings regarding antidepressant efficacy.
• To determine the rate of false positive results in network meta-analyses under the null hypothesis.

Main Methods:

• A Monte Carlo simulation replicated Cipriani's network meta-analysis.
• 1000 simulations were run assuming no true differences between the 12 antidepressants.
• The frequency of false positive findings indicating antidepressant superiority was calculated.

Main Results:

• Over 70% of simulations incorrectly identified one or more antidepressants as superior when no true differences existed.
• This indicates a high rate of false positive results in network meta-analysis.

Conclusions:

• Network meta-analysis may generate unwarranted claims of antidepressant superiority.
• There is limited evidence supporting significant efficacy differences between antidepressants based on current trial data.
• Caution is advised when interpreting multiple comparison analyses from network meta-analyses.