Should I Get Medication First? CBT vs. Antidepressants for Depression — What the Evidence Shows

"Should I get medication first?" — The clinical evidence behind your answer

You've probably been in this moment: a client referred for depression sits down for the first session and asks, "Should I see a psychiatrist and get on medication before we start therapy?" You know psychotherapy works — but presenting the data that backs that up, with confidence and precision, is a different skill.

Rush, Beck, Kovacs, and Hollon (1977) provided that data for the first time. In the first randomized controlled trial (RCT) to compare cognitive therapy (CT) directly against an antidepressant (imipramine), 78.9% of the cognitive therapy group improved at 12 weeks, compared with 22.7% of the medication group. This article walks through that study, its clinical implications, and the five decades of evidence that have accumulated since.

What is cognitive therapy? Beck's model

Cognitive therapy (CT), developed by Aaron Beck, rests on the model that depression is maintained by the negative cognitive triad — a pessimistic view of the self, the world, and the future.

Cognitive therapy is a structured, short-term treatment. It typically aims to conclude within 12–20 sessions, with a strong emphasis on skill acquisition built around between-session homework — thought records and behavioral experiments.

Rush et al. (1977): the first RCT comparing cognitive therapy and medication

Sample characteristics: 41 outpatients with unipolar depression, with a mean chronic course of 8.8 years; 75% presented with suicidal ideation. This was not a mild sample — it was a chronic, clinically severe population.

Design: Participants were randomly assigned to 12 weeks of either cognitive therapy or imipramine (a tricyclic antidepressant).

Key result: 78.9% of the CT group reached marked improvement or full remission, versus 22.7% on medication. Dropout was higher in the medication group, and at six-month follow-up CT gains were maintained with a lower relapse rate than the medication group.

How to interpret these results

The study was published in 1977 and carries real methodological limits — a small sample (41) and single-therapist effects among them. In the decades since, far larger trials have compared CT with pharmacotherapy, and the conclusion has largely converged on equivalence.

That carries two important clinical implications.

First, the current scientific consensus is that psychotherapy is as effective as medication for depression. The dramatic 78.9% vs. 22.7% gap in Rush et al. (1977) was not reproduced in later trials, but the study was decisive in establishing the premise that psychotherapy is genuinely effective in its own right.

Second, cognitive therapy tends to keep working after treatment ends. When medication is discontinued, relapse risk rises; the skills learned in cognitive therapy, once internalized, continue to operate without ongoing treatment.

Explaining psychotherapy to your client: practical steps

1. Inform the treatment choice

When a client asks, "Which should I do first — medication or therapy?", it helps to convey that both are evidence-based options.

"Both psychotherapy and medication are effective for depression. Which one fits you depends on several factors we can look at together."

2. Describe how psychotherapy differs

Help the client understand the difference between the two.

3. Judge by severity and level of functioning

In severe major depression — for example, with psychotic features or marked functional impairment — medication or inpatient care may take priority. Psychotherapy presupposes a level of functioning that allows the client to participate safely in sessions.

"Is it feeling almost impossible to keep up with daily life right now?"

That question is a first clue to whether psychotherapy alone is viable.

4. Keep combined treatment on the table

Medication and psychotherapy are not mutually exclusive. There is evidence that for moderate-to-severe depression, combined treatment outperforms either one alone. If a client is already on medication, adding psychotherapy can strengthen the effect.

5. "The skills stay with you after the medication stops"

When a client fears becoming dependent on medication, explain the durable nature of psychotherapy.

"The skills you build here stay with you after we finish — they become part of how you handle things. That's what makes psychotherapy different."

After Rush et al. (1977): what changed over 50 years

Despite its small sample, Rush et al. (1977) was a turning point that redirected psychotherapy research. In the years since, cognitive behavioral therapy (CBT) has been tested against medication for depression in dozens of randomized controlled trials, repeatedly confirming broad equivalence.

Notably, the large RCT by DeRubeis et al. (2005) found cognitive therapy and paroxetine (an antidepressant) equally effective in moderate-to-severe depression, and Hollon et al. (2005) showed that relapse rates after treatment ended were significantly lower in the CT group than in the medication group.

The current clinical consensus runs like this: for mild-to-moderate depression, CBT is appropriate as a stand-alone first-line treatment. For severe depression, medication or combined treatment may deliver a faster initial response.

With this context in hand, you can answer the client's question more precisely — explaining, with the data behind you, that the recommendation may shift depending on "how much you're struggling right now."

Lay out the evidence-based options with your client

Rush et al. (1977) is nearly 50 years old, but the message it established still holds. For depression, psychotherapy is effective independently of — and on par with — medication.

When a client asks, "Should I get medication first?", you can now answer with confidence: "Psychotherapy is genuinely effective too. And what you learn here stays with you after treatment ends." That single sentence hands the client a real, additional choice. Recording the rationale behind each treatment decision and tracking the client's response over time in your case notes or EHR lets you ground your clinical judgment in data rather than memory.