publication bias – a post-script…

Posted on Sunday 22 May 2016

Wading around in the meta-analyses comparing antidepressant efficacy and safety among them can be like Kafka’s The Trial or McCarthy’s The Road. You’re never sure where you are or if you’ve arrived anywhere. One of the most quoted versions is Cipriani et al’s Comparative efficacy and acceptability of 12 new-generation antidepressants: a multiple-treatments meta-analysis that produced this chart.

…comparing efficacy and acceptability. The abcsissa [horizontal axis] is the rank from 1st down to 12th and the ordinate [vertical axis] is the probability of each rank for the drugs. Now recall this forest plot of Vortioxetine versus comparators [Duloxetine] from Cosgrove et al:
They went head to head with one of their least effective rivals – and lost. New isn’t necessarily better…

Sorry, the comment form is closed at this time.