Skip to content
MyFreud

Exposure Therapy: How Phobias Are Treated

Specific phobia is among the most treatable conditions in mental health, and the evidence says one properly run session is often enough. Here is what it takes.

5 min read

Pop-art illustration of two people sitting and talking in a room, one facing the viewer wearing glasses.

Key takeaways

  • Specific phobia responds to exposure therapy better than almost any other condition responds to almost any other psychological treatment. A meta-analysis of 33 randomised trials found exposure produced large effects and outperformed both placebo conditions and other active therapies.
  • A single session is frequently enough. In the original case series the average treatment time was a little over two hours, and 90 percent of patients were much improved or recovered when followed up an average of four years later.
  • A 2022 meta-analysis comparing single-session and multi-session exposure found no significant difference in outcome, while multi-session treatment took substantially longer. More sessions is not more treatment here.
  • What makes exposure work is doing it without the safety behaviours. Going in while gripping somebody's arm, keeping an exit in view or distracting yourself tends to blunt the learning the session exists to produce.
  • None of this transfers automatically to agoraphobia, social anxiety or panic disorder, which are separate conditions with separate evidence and worse single-session results.

Specific phobia is one of the most treatable conditions in mental health, and the treatment is exposure therapy. The part most people have not heard is how short it is: for a straightforward phobia, the evidence supports a single session of two to three hours rather than a course stretching over months.

That is an unusual claim in this field, so the rest of this is the evidence behind it and the conditions under which it holds.

What exposure therapy actually is

Exposure therapy is planned, repeated contact with the thing you are afraid of, arranged in an order you agree in advance, and carried out without the things you normally do to feel safer. It is not being thrown in at the deep end, and it is not talking about the fear until it goes away.

The active ingredient is testing a prediction. You go in expecting something specific to happen, usually that the fear will keep climbing until something gives, and you stay long enough to find out that it does not. Our overview of phobias and how they work covers why avoidance keeps that prediction permanently untested.

How well it works

The best summary is a meta-analysis by Wolitzky-Taylor and colleagues, which pooled 33 randomised treatment studies. [wolitzky-taylor-2008-meta] Exposure-based treatment produced large improvements against no treatment. It also outperformed placebo conditions and other active psychotherapies.

The second comparison is the one that carries weight. Beating a waiting list tells you people improve; beating a credible placebo and a rival therapy tells you something about the treatment itself.

What exposure is being compared against, and how it fares Illustrative
0 25 50 75 100 Relative improvement 88 Exposure therapy 58 Other active therapy 44 Placebo condition 18 No treatment

A schematic of the comparisons described in Wolitzky-Taylor et al. (2008), drawn to show ordering rather than measured effect sizes.

One session is usually enough

The single-session protocol comes from work published by Lars-Göran Öst in 1989. [ost-1989-one-session] Twenty consecutive patients were treated in one appointment combining graded contact with modelling, meaning the therapist demonstrated first. Mean treatment time was 2.1 hours. At follow-up averaging four years, 90 percent were much improved or completely recovered.

That was a small uncontrolled series, so the finding needed testing against the obvious alternative, which is that more sessions would do more. A 2022 meta-analysis by Odgers and colleagues compared single-session against multi-session exposure directly. There was no significant difference in outcome at the end of treatment or at follow-up, and effects were large for both. Multi-session treatment took significantly longer in total therapist time. [odgers-2022-single-vs-multi]

So the honest position is that a long course is not better, and for many people it is the thing that stops them starting.

What blunts it

The most common reason exposure does not work is that it was not quite exposure.

Safety behaviours. Holding somebody’s hand, keeping the exit in view, carrying medication you do not take, checking repeatedly that the dog is on a lead. Each of these gives the mind an alternative explanation for why nothing bad happened, so the prediction never actually gets tested.

Distraction. Looking away, counting, talking through it. Attention has to be on the thing for the learning to attach to it.

Leaving at the peak. Approaching, becoming frightened and escaping rehearses escape. This is the specific way unsupervised attempts backfire.

Doing it once, in one place. Fear that is unlearned in one room has a habit of returning in another. Varying the setting makes the new learning more portable.

Is this a specific phobia?

Tick what is true. This is a sorting prompt rather than a test, and it produces no diagnosis.

0 of 5 ticked

Where this does not apply

The evidence above is about specific phobia: a single identifiable object or situation. It does not transfer wholesale to the conditions it is most often confused with.

The first is agoraphobia, which is a fear of situations that would be hard to escape rather than of one thing, and it usually needs a longer, broader course.

Social anxiety involves ongoing evaluation by other people, which is not something you can arrange a controlled two-hour encounter with in the same way.

Panic disorder centres on the bodily sensations rather than an external trigger, so the exposure is to the sensations themselves.

And emetophobia is a well-known awkward case, because the feared outcome is genuinely difficult to arrange and largely internal.

What the research does not establish

The 1989 series was twenty patients with no control group. Its value is in having been replicated and refined since, not in the original numbers.

Almost all of these trials were run in specialist centres by therapists trained in the protocol. The gap between that and an ordinary appointment is real.

And the trials measure people who agreed to have exposure therapy. Exposure is unpleasant by design, a proportion of people decline it or drop out, and results from those who completed it will always look better than what an unselected person can expect.

What helps

Ask for it by name. Say specific phobia and say exposure therapy. That phrasing gets you to the right treatment faster than describing the fear.

Ask how long it will take. A practitioner proposing twelve weekly sessions for a straightforward single-object phobia is not matching the evidence, and it is a reasonable question to put directly.

Write down the prediction before you start. What exactly do you think will happen. The session works by testing that sentence, and a vague version cannot be tested.

Drop one safety behaviour at a time. If dropping all of them at once is too much, dropping them in order is the compromise that keeps the learning intact.

Book a top-up rather than starting over if the fear returns. Return of fear is expected and does not undo what was learned.

When to seek help

Speak to a doctor or a mental health professional if a fear is limiting what you do, if you are turning down work, travel or medical care because of it, or if avoiding it has quietly become a significant piece of organisation in your life. Specific phobia has an unusually good prognosis, which makes the case for raising it stronger than for many things people bring to an appointment.

If you are having thoughts of harming yourself, treat that as urgent and contact your local emergency services or a crisis helpline.

How MyFreud can help

MyFreud is useful for the part exposure depends on: writing down what you expected to happen, and then what did. Most people remember the fear accurately and the outcome vaguely, which is exactly backwards for this kind of learning, and a log fixes it.

Download MyFreud and start today: App Store or Google Play.

Frequently asked questions

Does exposure therapy actually work for phobias?

Yes, and it is one of the clearest results in clinical psychology. A meta-analysis pooling 33 randomised trials found exposure-based treatment produced large improvements compared with no treatment, and also beat placebo conditions and other active psychotherapies rather than merely matching them. That last comparison is the informative one, because it rules out the explanation that people simply improve when somebody pays attention to them.

How long does it take to treat a phobia?

Often one session of two to three hours. The original single-session protocol averaged a little over two hours, and a 2022 meta-analysis found single-session exposure performed as well as multi-session courses while taking significantly less total time. That is unusual enough to be worth stating plainly: for a specific phobia, a course of weekly appointments stretching over months is not the standard the evidence supports.

Is exposure therapy traumatic?

It is uncomfortable and it is not meant to be traumatic, and the difference is control. Exposure is planned with you, graded so you choose the order, and stopped when you decide rather than when the fear ends. The version that does harm is the one where somebody is surprised or held in place, which is not what the protocols describe. It is fair to expect a genuinely unpleasant couple of hours and reasonable to weigh that against a fear you have organised your life around.

Can I do exposure therapy on my own?

Some people can and self-help versions have reasonable support, but there is a specific way it goes wrong unsupervised. If you approach the thing, become frightened and then leave, you have rehearsed escape rather than learned anything, and the fear usually comes back stronger. The rule that matters is staying until the expectation you went in with has been tested. If you cannot reliably do that alone, the session is worth booking.

Why did my fear come back after I got over it?

Return of fear is a well-recognised phenomenon and it does not mean the treatment failed. What exposure builds is a new association that competes with the old one rather than erasing it, so the original can resurface with time, stress, or a change of context. The practical response is a brief top-up rather than starting again, and doing the exposure in several different places and situations the first time makes it less likely.

References

  1. 1.Wolitzky-Taylor KB, Horowitz JD, Powers MB, Telch MJ ( 2008). Psychological approaches in the treatment of specific phobias: a meta-analysis. Clinical Psychology Review. doi:10.1016/j.cpr.2008.02.007
  2. 2.Öst LG ( 1989). One-session treatment for specific phobias. Behaviour Research and Therapy. doi:10.1016/0005-7967(89)90113-7
  3. 3.Odgers K, Kershaw KA, Li SH, Graham BM ( 2022). The relative efficacy and efficiency of single- and multi-session exposure therapies for specific phobia: a meta-analysis. Behaviour Research and Therapy. doi:10.1016/j.brat.2022.104203