By Ryan S. Sultan, MD
Assistant Professor of Clinical Psychiatry, Columbia University Irving Medical Center
Double Board-Certified in Adult Psychiatry & Child/Adolescent Psychiatry
Published:
Quick answer. EMDR and trauma-focused CBT show comparable efficacy for PTSD, and the 2023 VA/DoD Clinical Practice Guideline places EMDR, prolonged exposure, and cognitive processing therapy at its highest tier. The practical differences are that EMDR requires substantially less detailed verbal narration of the trauma and assigns no between-session homework, while prolonged exposure assigns exposure practice and cognitive processing therapy assigns written worksheets. Dropout in randomized trials was 18% for EMDR and 22% for prolonged exposure, with overlapping confidence intervals. General CBT without a trauma focus performs less well for PTSD than either. |
Both Sit in the Highest Recommendation Tier
The 2023 VA/DoD Clinical Practice Guideline reviewed each trauma therapy individually rather than as a class, and it recommends three at its highest level: EMDR, prolonged exposure, and cognitive processing therapy. The World Health Organization named trauma-focused cognitive behavioural therapy and EMDR as the only psychotherapies indicated for PTSD across children, adolescents, and adults (WHO, 2013).
Head-to-head trials show comparable efficacy. Choosing between them turns on what each treatment asks of the patient, which is a question about fit rather than about ranking.
CBT Means Two Different Things Here
Patients comparing EMDR against CBT are usually comparing it against one of two things, and the distinction changes the answer.
Trauma-focused CBT covers prolonged exposure and cognitive processing therapy. These are structured protocols that work directly on the traumatic memory, and they carry the guideline recommendations described above. When trials compare EMDR against CBT for PTSD, this is what they mean.
General CBT targets present-day thoughts and behaviors without a trauma focus. It is effective for depression, anxiety disorders, insomnia, and a wide range of other presentations. For PTSD specifically it performs less well than the trauma-focused protocols, and the guidelines reflect that. Notably, therapies without a trauma focus show lower dropout than trauma-focused ones, which is a real advantage that has to be weighed against lower efficacy on the trauma itself.
A patient told they are "doing CBT" for PTSD is worth asking which one. Someone receiving general CBT for a trauma presentation is receiving a treatment the guidelines rank below the alternatives.
What the Head-to-Head Evidence Shows
A meta-analysis of 26 randomized controlled trials found a moderate pooled effect for EMDR on PTSD symptoms, with Hedges's g of −0.662 (95% CI −0.887 to −0.436), alongside effects on depression of g = −0.643 and on anxiety of g = −0.640 (Chen and colleagues, PLoS One, 2014).
In a community-based comparison of EMDR and prolonged exposure, 7 of 10 EMDR participants achieved a 70% or greater reduction in PTSD symptoms, against 2 of 12 in the prolonged exposure arm, with no dropouts in the EMDR group and 3 of 10 in the comparison group (Ironson and colleagues, Journal of Clinical Psychology, 2002). In a trial in female assault survivors, EMDR and prolonged exposure were equally effective (Rothbaum and colleagues, Journal of Traumatic Stress, 2005).
Samples at this scale support the conclusion that the treatments perform comparably. They do not support a superiority claim in either direction, and clinicians or websites making one are going beyond the data.
The Practical Comparison
| EMDR | Prolonged Exposure | Cognitive Processing Therapy | |
|---|---|---|---|
| Detailed verbal narration of the trauma | Substantially less | Central to the protocol | Written trauma account |
| Between-session homework | None | Exposure assignments | Worksheets |
| Typical protocol length | Varies widely by trauma type | 8 to 15 sessions | 12 sessions |
| Session length | 60 to 90 minutes | 60 to 90 minutes | 50 to 60 minutes |
| Dropout in randomized trials | 18% (95% CI 12–24) | 22% (95% CI 16–28) | Pooled trauma-focused rates |
| Temporary symptom worsening | Not separately reported | 20.0% | 28.6% |
| Guideline tier (VA/DoD 2023) | Highest | Highest | Highest |
Across 116 randomized trials, pooled dropout from psychological therapies for PTSD was 16% (95% CI 14 to 18). For EMDR the rate was 18% (95% CI 12 to 24) across 21 trials, and for prolonged exposure 22% (95% CI 16 to 28) across 22 trials (Lewis and colleagues, European Journal of Psychotraumatology, 2020). Those confidence intervals overlap substantially, and the gap should be read as small.
What Actually Decides It
Three factors carry most of the weight in practice.
How much detailed retelling you can tolerate. Prolonged exposure requires narrating the traumatic event in detail, repeatedly. EMDR requires substantially less verbal narration. Patients who have avoided treatment for years specifically because they cannot imagine describing what happened often find EMDR the more accessible starting point, and that accessibility is a clinical consideration rather than a preference.
Whether homework is realistic. EMDR completes its work within sessions. Prolonged exposure assigns exposure practice between sessions and cognitive processing therapy assigns written worksheets, both of which require time, privacy, and a degree of stability that not every patient has in the middle of a trauma presentation. For a patient working two jobs or managing small children, the homework requirement is frequently what determines whether a protocol is finished.
What else is going on. Untreated ADHD impairs the sustained attention and working memory that all three protocols require, and the resulting stall gets misread as resistance. Active substance use, unstable sleep, and an unsafe living situation each change the sequencing. These are the judgments that benefit from a diagnostic evaluation before a modality is selected.
When CBT Is the Better Choice
Cognitive processing therapy suits patients whose trauma is organized around beliefs they can articulate and want to argue with. Guilt following a decision made under threat, self-blame after an assault, and shattered assumptions about safety and trust all respond well to a protocol built to examine them directly. Patients who find written work clarifying rather than burdensome tend to do well here.
Prolonged exposure suits patients whose avoidance is the dominant feature and who are prepared to confront it methodically. It has the largest evidence base of the three and the longest track record in veteran populations.
General CBT is the better choice when the presenting problem is depression, an anxiety disorder, or insomnia rather than trauma, and when trauma is secondary or historical. It is also a reasonable holding position for a patient who is not yet stable enough for trauma processing.
When EMDR Is the Better Choice
EMDR suits patients who cannot tolerate detailed narration, who cannot realistically complete homework, and who have tried a talk-based therapy that stalled. It also suits patients whose trauma is stored more somatically than verbally, where the presenting complaint is physical activation rather than an articulable belief.
The dropout figures are relevant here. A treatment that is finished is worth more than a treatment with a marginally larger effect size that a patient leaves in week four.
Accessing Either Treatment in New York
Both treatments are widely available in New York City, and most providers of either are therapists who do not prescribe. That arrangement works. It introduces a coordination gap when medication and trauma processing need to be sequenced against each other, which is common.
Trauma presentations are frequently misdiagnosed, and the misdiagnoses change treatment. Hypervigilance and concentration difficulty resemble ADHD. Dissociation is misread as inattention. Complex PTSD is often misdiagnosed as borderline personality disorder, which alters how a patient is treated by every clinician who reads the chart afterward. A diagnostic evaluation by a psychiatrist before a modality is chosen prevents a course of the wrong protocol.
Dr. Sultan provides EMDR, CBT, and CPT-informed work as the treating psychiatrist at the Chelsea office in Manhattan, and by telehealth in New York, Virginia, Montana, and Florida. Consultations can be arranged through Integrative Psych.
Related Reading
- EMDR therapy: a complete clinical guide
- EMDR with a psychiatrist in NYC
- CBT in NYC
- PTSD and complex PTSD treatment
- What happens after an EMDR session
- Medication, therapy, or both
Frequently Asked Questions
Is EMDR better than CBT for PTSD?
Head-to-head trials show comparable efficacy, and the 2023 VA/DoD Clinical Practice Guideline places EMDR, prolonged exposure, and cognitive processing therapy at the same highest tier. The choice turns on fit. EMDR requires less detailed verbal narration and assigns no between-session homework.
What is the difference between EMDR and CBT?
EMDR works on the traumatic memory using bilateral stimulation across an eight-phase protocol and assigns no homework. Trauma-focused CBT works on the memory through detailed narration in prolonged exposure or through written accounts and worksheets in cognitive processing therapy. General CBT targets present-day thoughts and behaviors without a trauma focus and performs less well for PTSD specifically.
Can you do EMDR and CBT at the same time?
Running two trauma protocols concurrently is generally avoided, since it makes it impossible to tell what is producing a change. Sequencing them is common, and combining EMDR with non-trauma-focused work such as CBT for insomnia or DBT skills is routine.
Which has a lower dropout rate, EMDR or CBT?
Dropout was 18% for EMDR and 22% for prolonged exposure across randomized trials, with confidence intervals that overlap substantially. Trauma-focused therapies as a group show higher dropout than therapies without a trauma focus.
Does EMDR work if CBT did not?
Frequently. A stalled course of talk-based therapy is one of the more common reasons patients arrive for EMDR, and the differences in what each protocol demands mean that failure of one does not predict failure of the other. A stalled course is also worth examining for an untreated comorbidity, since conditions such as ADHD impair the attention every trauma protocol requires.
How many sessions does each take?
Prolonged exposure typically runs 8 to 15 sessions and cognitive processing therapy 12. Published EMDR protocols vary widely by trauma type, and for single-incident adult trauma a course of weekly 60 to 90 minute sessions over roughly three months is a reasonable planning assumption.
References
- US Department of Veterans Affairs and Department of Defense. VA/DoD Clinical Practice Guideline for the Management of PTSD and Acute Stress Disorder. 2023. ptsd.va.gov
- World Health Organization. Guidelines for the Management of Conditions Specifically Related to Stress. Geneva, 2013.
- Chen Y-R, Hung K-W, Tsai J-C, et al. Efficacy of eye-movement desensitization and reprocessing for patients with posttraumatic stress disorder: a meta-analysis of randomized controlled trials. PLoS One. 2014;9(8):e103676. PMID 25101684
- Ironson G, Freund B, Strauss JL, Williams J. Comparison of two treatments for traumatic stress: a community-based study of EMDR and prolonged exposure. J Clin Psychol. 2002;58(1):113-128.
- Rothbaum BO, Astin MC, Marsteller F. Prolonged exposure versus eye movement desensitization and reprocessing (EMDR) for PTSD rape victims. J Trauma Stress. 2005;18(6):607-616.
- Lewis C, Roberts NP, Gibson S, Bisson JI. Dropout from psychological therapies for post-traumatic stress disorder (PTSD) in adults: systematic review and meta-analysis. Eur J Psychotraumatol. 2020;11(1):1709709. PMID 32284816
- Larsen SE, Wiltsey Stirman S, Smith BN, Resick PA. Symptom exacerbations in trauma-focused treatments: associations with treatment outcome and non-completion. Behav Res Ther. 2016;77:68-77.
- Lee CW, Cuijpers P. A meta-analysis of the contribution of eye movements in processing emotional memories. J Behav Ther Exp Psychiatry. 2013;44(2):231-239. PMID 23266601