Evidence-based trauma therapy
Which trauma therapies the guidelines actually recommend, what each one involves, how long it takes, and where coherence therapy honestly sits among them.
The short answer
The trauma therapies with the strongest evidence are Cognitive Processing Therapy (CPT), Prolonged Exposure (PE), EMDR and trauma-focused CBT. US Department of Veterans Affairs and Department of Defense guidance and UK NICE guidance all name these first, and everything else is measured against them.
That is a statement about evidence, not about fit. Someone whose trauma sits in a body-level startle pattern, or who has already completed a full course of CPT without the emotional charge lifting, is not helped by being told the best-evidenced option exists. This page sets out what each recommendation actually rests on, so you can tell the difference between "proven" and "right for me".
What counts as evidence
"Evidence-based" gets used loosely. When you compare trauma therapies, the claims usually fall into one of five tiers, and the tiers are not interchangeable.
- Named in clinical guidelines. A panel has reviewed the trials and put the treatment in a recommendation. This is the strongest practical claim available, because it already accounts for replication, sample sizes and dropout.
- Multiple randomized controlled trials. Independent teams, repeated findings, usually against a waitlist, treatment as usual or another active therapy.
- A single randomized trial. Real evidence, but one study cannot separate a treatment effect from the enthusiasm of the team that developed it.
- Open trials and case series. Everyone got the treatment, nobody got a comparison. Useful for showing something can work, unable to show that it does better than time, support or expectation.
- Theory and mechanism. A plausible account of how change happens, supported by laboratory science, without clinical outcome data in the population you care about.
Most disagreements about trauma therapy come from people comparing claims from different tiers as if they were the same kind of claim.
The first-line therapies
Cognitive Processing Therapy (CPT)
CPT treats post-traumatic stress as a problem of conclusions rather than of memory access. The memory is confronted, usually through a written account, and then the beliefs that formed around it get examined as "stuck points": it was my fault, I am permanently damaged, the world is not safe. Worksheets and Socratic questioning are used to test those beliefs against evidence.
- Typically twelve weekly sessions, with practice between them.
- Strongest fit when guilt, shame, self-blame or a shattered sense of safety dominate.
- Heaviest on homework of the first-line options, which some people value and others find impossible during a difficult period.
- Limitation: it works at the level of belief, so it can leave pre-verbal or body-level responses untouched.
Prolonged Exposure (PE)
PE works by habituation. You retell the traumatic memory in detail in session (imaginal exposure, recorded and listened back to), then gradually approach the situations you have been avoiding (in vivo exposure). The fear response is meant to decline through repeated, safe contact with what it is warning you about.
- Usually eight to fifteen sessions of about ninety minutes.
- Best supported for avoidance-dominant PTSD, where the symptoms are organised around not going near the memory or the places.
- The most demanding of the first-line options, and the one with the highest dropout in routine practice.
- Limitation: it produces new learning that inhibits the old response rather than rewriting it, which is why the response can return under later stress.
EMDR
EMDR pairs recall of a distressing memory with bilateral stimulation, usually guided eye movements, in a structured eight-phase protocol. Processing, desensitization and installation of a positive belief are worked through in sequence, with less verbal retelling and less homework than CPT or PE.
- Commonly eight to twelve sessions for a single trauma.
- Recommended by the World Health Organization, and first-line in the same guidance as CPT and PE.
- Suits people who cannot tolerate detailed narration or written accounts.
- Limitation: weaker fit for complex or developmental trauma with no clear index memory, and the mechanism is still argued about even where the outcomes are not.
Trauma-focused CBT (TF-CBT)
A structured, components-based treatment built for children and adolescents, involving the caregiver throughout. It combines psychoeducation, relaxation and affect regulation, gradual exposure to the memory, and cognitive processing, in a fixed sequence.
- Usually twelve to sixteen sessions.
- The best-evidenced option specifically for children and teens.
- Distinct from generic CBT: the trauma narrative and caregiver work are the active ingredients, not the thought records.
Side by side
| Approach | Guideline status | Typical length | Best fit |
|---|---|---|---|
| CPT | First-line | 12 sessions | Guilt, shame, stuck beliefs |
| PE | First-line | 8 to 15 sessions | Avoidance-dominant PTSD |
| EMDR | First-line | 8 to 12 sessions | Clear memories, low tolerance for retelling |
| TF-CBT | First-line (children, teens) | 12 to 16 sessions | Children and adolescents, with a caregiver |
| Present-centered therapy | Recommended alternative | About 10 sessions | Not ready to revisit the trauma |
| IFS | Single RCT for PTSD | Variable, often longer | Complex trauma, parts-based model |
| Somatic Experiencing | Small trials, mixed results | Variable | Body-based symptoms |
| Deep brain reorienting | Emerging | Variable | Startle and freeze patterns |
| Coherence therapy | Case evidence and mechanism | Variable | Patterns that survive the memory work |
Second-line and emerging options
These are not rejections. Each one exists because the first-line protocols leave a real group of people untreated: people who will not do exposure, people whose trauma has no single memory, people who finished a full course and still feel the old response firing.
Present-centered therapy works on current problems and coping rather than the trauma itself. It performs respectably in trials against trauma-focused treatments and is a reasonable route when revisiting the event is not tolerable yet. DBT and skills-based work come first when the problem is emotion regulation, self-harm or dissociation, because processing anything deeper is unsafe without stability. IFS has a small randomized trial for PTSD plus a growing clinical literature, and maps well onto complex trauma. Somatic Experiencing has several small trials with mixed findings. Deep brain reorienting is new enough that the published outcome data is thin.
Where coherence therapy fits
Coherence therapy is not first-line, and this site is not going to pretend otherwise. It has case evidence and a strong mechanistic basis in memory reconsolidation, and it does not have the randomized trials that would put it alongside CPT, PE or EMDR. Our own assessment is on the evidence page, including what is missing.
What it addresses is a specific gap. The first-line treatments are good at reducing post-traumatic symptoms. They are less reliable at dissolving the emotional learning that formed around the event: the implicit conclusions about safety, worth and what happens when you need something from another person. When someone completes EMDR and the memory no longer distresses them, yet the same pattern shows up in every relationship, that learning is the thing still running. Coherence therapy targets it directly by reactivating the schema and setting up the disconfirming experience that allows it to be rewritten rather than suppressed.
That is a claim about mechanism and about the cases practitioners report, not a claim of equivalent evidence. If you are choosing a first treatment for PTSD, choose from the first-line list. If you have done that and something is still running, coherence therapy is a reasonable conversation to have, along with the IFS and body-based options.
Matching therapy to situation
| Your situation | Where to start |
|---|---|
| One event, clear intrusive memories, nightmares | CPT, PE or EMDR. All three are defensible first choices. |
| You cannot face retelling it in detail | EMDR, or present-centered therapy first if even that is too much. |
| Guilt, shame or self-blame dominate | CPT. The stuck-point work is built for exactly this. |
| Avoidance has taken over your life | PE, if you are willing to do the exposure work with support. |
| Childhood abuse, neglect, no single memory | Phased treatment: stabilization first, then a longer trauma-focused phase. IFS or coherence therapy alongside. |
| Emotion regulation is the emergency | DBT or skills work first. Deeper processing after stability. |
| Symptoms improved but the pattern stayed | Coherence therapy, IFS, or somatic work, depending on where it lives. |
| Freeze, startle, shutdown that talk therapy misses | Somatic Experiencing or deep brain reorienting. |
For a fuller account of each modality, see types of therapy for trauma. For the choice framed the other way round, by what your symptoms look like rather than by what is proven, see best therapy for trauma.
Questions to ask a therapist before you start
The answer you want is a named protocol, not "trauma-informed". These are the questions that separate a trained provider from an interested one.
- Which specific trauma treatment are you trained in: CPT, Prolonged Exposure, EMDR or trauma-focused CBT?
- Where did you do that training, and did it include supervised cases?
- How will you measure whether this is working? Trauma protocols use symptom measures for a reason.
- What happens if it is not working after eight or ten sessions?
- What do you do first if I become destabilized, dissociate or self-harm?
- How much work happens between sessions, and what does that look like?
You are also allowed to say that a particular approach sounds intolerable. A therapist trained in one protocol and honest about its limits is more useful than one who claims to do everything.
What the evidence doesn't settle
Guideline recommendations are averages over trials, and trials are mostly of adults with single-incident trauma, run by research teams on people who agreed to be randomized. Three gaps matter when you are the one choosing.
Dropout is underweighted. A treatment that works for everyone who completes it and is abandoned by a third of starters is not obviously the best first choice for you. Prolonged Exposure in particular loses a meaningful share of people to the exposure work itself.
Complex and developmental trauma is under-studied. Most trials enrol adults with a defined index event. If your trauma happened repeatedly, in childhood, at the hands of the people who were supposed to protect you, the trials were not designed around you, and phased treatment is the norm for good reason.
Equivalent averages hide individual differences. When two treatments come out level in a trial, that is a statement about the group. Some people do far better in one. If eight or ten sessions of a well-delivered first-line therapy have not moved anything, the sensible reading is not that trauma is untreatable, it is that this level of the problem has not been reached yet.
Common questions
What is the most effective therapy for trauma?
CPT, Prolonged Exposure, EMDR and trauma-focused CBT, in that order of trial volume rather than of superiority. They are all first-line. Which one is most effective for you depends on your symptoms, what you have already tried and how much direct work with the memory you can tolerate.
Does CBT work for trauma?
Trauma-focused CBT does, and CPT and PE are both forms of it. General CBT that treats trauma as a set of unhelpful thoughts, without working on the memory, has weaker evidence for post-traumatic stress. This is the most common reason people say CBT did nothing for their trauma: they had the generic version.
Is EMDR better than CPT?
Most head-to-head trials find them roughly equivalent. EMDR asks less narration and less homework. CPT gives more structure for guilt, shame and a shattered sense of safety. Neither is reliably ahead of the other.
Is coherence therapy evidence-based?
Not in the strict sense. The science it rests on, memory reconsolidation, is well established, and there is published case evidence, but no large randomized trials for PTSD. It is not first-line. It is best used for the emotional learning that survives a first-line course.
How long does trauma therapy take?
Twelve sessions for CPT, eight to fifteen for Prolonged Exposure, eight to twelve commonly for EMDR, twelve to sixteen for trauma-focused CBT with a child. Complex trauma usually takes longer, and often starts with a stabilization phase before any trauma processing.
How do I find a trauma-informed therapist?
Filter for training in a named protocol rather than for the phrase "trauma-informed", then use the questions above on a first call. Our guide to finding a coherence therapy therapist covers what to look for if you want a provider who works this way specifically.
Stay in the loop
Get notified when we publish new articles on coherence therapy, memory reconsolidation, and lasting emotional change.
No spam. Unsubscribe anytime.