"Conclusions: Reboxetine is, overall, an ineffective and potentially harmful antidepressant. Published evidence is affected by publication bias, underlining the urgent need for mandatory publication of trial data."
Here's the BMJ paper
Insider wonders if Ben will take this further.
