Apparently, these drugs can cause some people to kill themselves. But they also cause some users to recover, meaning they're less likely to commit suicide. It wasn't clear how exactly this balance was measured, but the net effect can be positive.
Still raises some hard questions along the lines of the trolley experiment. Who gets to live, and who gets to die? And what about the fact that we don't even know who it is that will get better or worse?
Regarding coding, the coder is only really able to quit because the market is good. Plenty of people along the same economic chain that led to this are in more precarious job circumstances.