We use medications for reasons that don’t match their names. That’s why some clinicians are turning to Neuroscience-Based Nomenclature (NbN)—a system that classifies meds by mechanism instead of indication.

We reviewed Psymatik in The Carlat Psychiatry Report this month, as part of our occasional coverage on how AI and digital tools are changing the way we prescribe.

Her business was failing and her marriage was falling apart. She wanted to OD on her insulin.

We’re trained to treat mental illness—not to navigate the courtroom.

You start a medication. The patient feels better in two weeks. You think: “Finally. This one works.” But six weeks later, the magic’s gone.

“When I’m manic, I feel like I put on Superman’s vest.” 

That’s how one patient explained why he didn’t want to stay on lithium. He wasn’t denying the crashes. He just wasn’t ready to give up the highs.
Another described it differently:
“If you take too much of it you're slightly underwater, but if you take too little you're floating three feet above the water.”

You’re sitting with a patient.

They’ve had terrible luck with sexual side effects.
And you’re silently trying to recall which SSRI is "less bad" on GI issues.
It’s not that you don’t know the data—it’s that it’s never all in one place.

How many times have you prescribed Abilify for depression and then stumbled over what to call it?

Do you say “antidepressant”?
“Mood stabilizer”?
Or do you tell the truth—“It’s an antipsychotic”—and watch your patient flinch?
We’ve all had this moment. The treatment fits. The label doesn’t.