Medication, Therapy, or Both?

Home > Medication vs. Therapy vs. Both

I provide both, so I have no financial reason to steer you toward either. Here is what the evidence supports, by condition.

The short version

Where the answer is genuinely "either"

Mild to moderate depression. Head-to-head trials show comparable response rates for CBT and antidepressants. The meaningful difference is what happens after treatment stops: psychotherapy shows lower relapse rates, plausibly because skills persist while drug levels don't.

Reasonable grounds for choosing: prior response, time availability (therapy demands more of it), tolerance for side effects, pregnancy or planned pregnancy, and preference. Preference is a legitimate clinical input, not a concession.

Where medication is not optional

Bipolar disorder. No psychotherapy prevents mania. Therapy is genuinely valuable for adherence, early warning recognition, and the psychological work of living with the illness — but as an addition, never a substitute.

Related and frequently gotten wrong: treating bipolar depression with an antidepressant alone risks precipitating mania. This is one of the more common errors I see in second opinions.

Schizophrenia and psychotic disorders. Antipsychotic treatment is foundational. CBT for psychosis has real evidence as an adjunct. Where treatment resistance is established, clozapine is the most effective option and is substantially underused — I have published on why.

Where the modality matters more than the category

OCD. "Therapy" is not the answer; ERP is. General supportive therapy can worsen OCD by delivering reassurance, which functions as a compulsion. This is the single most consequential distinction on this page.

PTSD. EMDR or trauma-focused CBT. General talk therapy about trauma is not the same treatment.

Emotion dysregulation. DBT specifically.

Asking "should I do therapy" without specifying which therapy is like asking whether to take medication without specifying which.

ADHD, specifically

Medication has the larger effect size, and the international registry evidence I reviewed in a 2025 JAMA Psychiatry editorial shows medicated periods associated with reductions in criminal convictions (32–41%), substance-related emergency events (31–35%), and motor vehicle crashes (38–42%).

But medication does not address the executive skill deficits, the emotional dysregulation, or twenty years of accumulated belief that you are unreliable and lazy. Those respond to CBT and DBT skills.

The common pattern is someone on a well-titrated stimulant who describes treatment as helping a lot but not being enough. That gap is usually therapy-shaped.

The practical argument for both in one place

When both are indicated, split care means two clinicians who rarely speak. Your therapist doesn't know why the dose changed; your prescriber doesn't know what happened in session.

For straightforward cases that's fine and often cheaper. For complex ones it's where treatment stalls.

Related

Medication management · CBT · DBT · EMDR · ERP · Psychiatrist vs. psychologist vs. therapist

Frequently Asked Questions

Is therapy or medication better for depression?

For mild to moderate depression they perform comparably in head-to-head trials, with psychotherapy showing lower relapse rates after treatment ends. For severe depression, combination treatment outperforms either alone.

Do I have to take medication if I see a psychiatrist?

No. A psychiatrist evaluates and recommends; you decide. For a number of conditions psychotherapy is first-line and medication isn't indicated at all.

Can therapy alone treat bipolar disorder?

No. Medication is foundational for bipolar disorder — no psychotherapy prevents mania. Therapy is a valuable addition for adherence, early warning recognition, and adjustment, but not a substitute.

Why do I need a specific type of therapy?

Because modality matters as much as the decision to do therapy at all. OCD requires exposure and response prevention specifically; general talk therapy can worsen it. PTSD requires trauma-focused treatment. Emotion dysregulation requires DBT.

Is medication enough for ADHD?

Often not entirely. Medication has the larger effect size but doesn't address executive skill deficits, emotional dysregulation, or the accumulated self-narrative. Many well-treated patients describe a residual gap that is therapy-shaped.

Can one clinician provide both?

Some psychiatrists still do. Most no longer provide psychotherapy. When both are indicated, one clinician removes the coordination gap that fragments split care.