We tend to blame the meds.

But what if the real reason a patient isn’t improving… is because we’re treating the wrong thing?

In Difficult-to-Treat Depression, Chris Aiken opens Chapter 3 with a clinical punch:

“Sometimes the failure isn’t the antidepressant—it’s our own diagnostic prowess.”

That line stopped me.

Because misdiagnosis is one of the most under-discussed culprits in so-called “treatment resistance.”

And I don’t mean rare or exotic errors.

I mean: 

▸ Bipolar disorder hiding under unipolar labels 

▸ ADHD that’s been buried under low self-esteem 

▸ Substance use that we didn’t think to ask about 

▸ Trauma histories that masquerade as melancholia 

▸ “Failed” meds that were actually underdosed or cut too soon

Aiken reminds us that nearly half of patients labeled “treatment-resistant” may actually have undiagnosed bipolar disorder.

And that’s just one example.

What I love about this book is how high yield it is.

It’s not your usual textbook, with pages of throat-clearing on epidemiology and reminders of basic features you already know and will probably skim.

Every page counts.

Chapter 3 on Misdiagnosis is only four pages long—and yet it’s packed with clinical pearls you’ll use in practice the same day you read it.

This is the first in a short series where I’ll be sharing reflections on key chapters from the book.

Coming up: 

▸ How measurement tools can double remission rates—and why we still don’t use them 

▸ What to do when antidepressants don’t work, and how to rethink your next step 

▸ Underused augmentation tools, from lithium to celecoxib 

▸ And why staying well might matter more than full remission

Each post will highlight one chapter—and one mirror for our work.

There’s a phrase I keep coming back to in this one:

“Depression rarely travels alone.”

And when we forget that—when we tunnel in on “fixing the meds”—we can lose sight of the real work: figuring out who this person actually is, and what their suffering is trying to tell us.

Not every difficult case is misdiagnosed, of course.

But I wonder how many times we escalate treatment… when what’s really needed is a reframe.

A slower interview.

A wider lens.

A willingness to say: Maybe I got this wrong.

What’s your process when a patient isn’t getting better?

Do you double down—or zoom out?

▸ If this post stirred something, feel free to pass it along.

▸ Follow me (Daniel Carlat, MD) for grounded reflections on psychiatry in real-world practice.