julia.endo said:

The version of this that has an answer is narrower than the version being asked.

All true, with one condition: that curve is for people who reached the dose on schedule. Anyone who slowed the ladder for tolerability is on a different, flatter curve, and comparing yourself with the published mean will make you feel like a non-responder when you are not.

MariaRD said:

The dose-response is real but shallow at the top.

Saving this. It is the first explanation that did not require me to already understand it. I will report back once I have actually tried it.

julia.endo said:

The version of this that has an answer is narrower than the version being asked.

I will push back on the "any working dose is fine" framing. The maintenance evidence sits overwhelmingly at the top studied dose, and the extension data shows regain tracking dose reduction rather than tracking stopping. Holding low is reasonable; pretending it is evidentially equivalent is not.

The figures, for anyone assembling their own picture. If you are going to change something, change one thing and give it long enough to express itself. Four weeks is the usual minimum for anything pharmacological on this board, and two weeks of data has told you almost nothing.

Moderator note: good thread. Keeping it here rather than moving it, because the question is general enough to be useful.