HUMAN OS WIKI · 01 · UNDERSTANDING YOURSELF

WHICH THERAPY FOR WHAT?

CBT, DBT, ACT, EMDR, ERP, IFS — the acronyms make therapy feel like a maze. The honest evidence is simpler and stranger than the marketing: for a few conditions a specific therapy is clearly first-line, and for most, who you work with matters more than which method they name.

By Jared Ohman 9 min read Last updated June 2026 Source: THF Research

When bona-fide therapies are compared head-to-head, outcome differences are often small to nonexistent — yet the therapeutic alliance is among the most consistent predictors of whether therapy works at all. — Synthesis of comparative psychotherapy research (Wampold and colleagues)

SHORT ANSWER

For a few conditions, a specific therapy is clearly first-line: dialectical behavior therapy (DBT) for borderline personality disorder, exposure and response prevention (ERP) for OCD, and EMDR or trauma-focused CBT for PTSD. CBT is the most-researched approach and a first-line option for depression, anxiety, and phobias. But for most conditions, when "bona fide" therapies are compared head-to-head, outcome differences are small — and the therapeutic alliance (the trust between you and your therapist) is one of the most consistent predictors of whether therapy works. So the practical rule is: match the therapy to the condition where the evidence is specific, and everywhere else, prioritize the fit with the therapist and whether your progress is actually being measured.

GET THE FREE PRINTABLE ↓ One page, wallet-card layout. Free. One email below, no spam, unsubscribe in a click.

The problem

You finally decide to get help, and you hit a wall of acronyms. CBT, DBT, ACT, EMDR, ERP, IFS — each with a confident website saying it's the answer. You don't know if the differences are real or marketing, whether you need a specific kind, or whether you've just wasted three sessions with the wrong one. So you either freeze, or you pick at random and hope.

The honest evidence is both clearer and stranger than the marketing implies. It comes down to two truths that sound contradictory and aren't: for a handful of conditions, the specific therapy you choose genuinely changes the outcome — and for most of the rest, who you work with matters more than which method they name.

Hold both and the maze collapses into a simple decision.

The mechanism

Decades of comparative research keep landing on the same uncomfortable finding: when you put two legitimate, well-delivered therapies head-to-head for the same problem, the outcome differences are often small. The thing that does predict whether therapy works, across modalities, is the therapeutic alliance — the trust and collaboration between you and your therapist. That's the "common factors" view, and the evidence for it is strong.

But that's only half the truth, and the half people over-extend. For specific conditions, the specific technique matters a great deal:

OCD → ERP. Obsessive-compulsive disorder responds to exposure and response prevention — an offshoot of CBT whose active ingredient is the exposure. Talk therapy without it tends to underperform.

Borderline personality disorder → DBT. Dialectical behavior therapy was built specifically for BPD's emotional dysregulation and self-destructive urges, and is the most empirically supported treatment for it.

PTSD → EMDR or trauma-focused CBT. These are the most reliably supported, with strong symptom reduction; for some single-event traumas the gains can be rapid.

Depression, anxiety, phobias → CBT is the most-researched first-line option (with others, including ACT and good generic counseling, also effective for many).

So the synthesis is not "therapy type doesn't matter" and not "you must find the one perfect method." It's: match the therapy to the condition where the evidence is specific, and prioritize fit and measured progress everywhere.

One honest caveat: this is an orientation, not medical advice or a diagnosis. The right plan is one you build with a qualified clinician.

The operating system

Five steps to choose without getting lost in the acronyms.

STEP 01

Start with the condition, not the brand

Before comparing methods, get clear on what you're actually treating. The acronyms only sort themselves once there's a target. "I want therapy" is a maze; "I have OCD" or "I'm processing a trauma" points almost directly at the right modality.

If you don't have a clear picture yet, an assessment with a clinician is the first step — diagnosis is what makes the modality question answerable.

STEP 02

For specific conditions, ask for the first-line therapy by name

Where the evidence is condition-specific, choosing the modality matters — so ask for it directly. OCD: "Do you do ERP?" BPD: "Are you DBT-trained?" PTSD: "Do you do EMDR or trauma-focused CBT?" A therapist who treats these without the matching approach is a reason to keep looking.

For OCD especially, a generalist doing open-ended talk therapy can leave the core untreated. Insist on exposure-based work.

STEP 03

Everywhere else, prioritize fit over brand

For depression, anxiety, general distress, and most of what brings people to therapy, the modality is less decisive than the relationship. Optimize for fit: do you feel understood, safe, and able to be honest with this person? That alliance is one of the strongest predictors that the work will help.

A great-on-paper modality with a therapist you don't click with will usually underperform a "lesser" method with someone you trust.

STEP 04

Vet the therapist, not just the acronym

Ask how they actually work: what's their training in the approach, how do they structure treatment, and — crucially — do they use measurement-based care (briefly tracking your symptoms over time)? A therapist who measures is one who'll notice when something isn't working and adjust.

"How will we know if this is helping?" is one of the best questions you can ask in a first session. A good answer involves something more than vibes.

STEP 05

Track progress, and be willing to switch

Give it a fair run, but watch the trend. If there's no meaningful movement after roughly 6–8 weeks of consistent work, raise it directly with your therapist — adjust the approach, or consider a different fit. Therapy not working with one person or method is information, not a verdict that therapy can't help you.

Switching after a fair trial isn't failure or disloyalty. The goal is your outcome, not finishing a particular method.