Cognitive behavioural therapy

Cognitive behavioural therapy is the most widely available talking therapy in the UK and the one with the largest evidence base behind it. It is also the most widely misdescribed, usually as a matter of replacing negative thoughts with positive ones.

It is not that, and the difference is the whole of the method. CBT does not ask whether a thought is cheerful. It asks whether a thought is accurate, and whether believing it is useful — and quite often the answer is that the gloomy thought is partly true, which positive thinking has no way of handling. Being told to look on the bright side when something is genuinely wrong is not a therapy; it is an instruction to lie to yourself, and it is why people who have been offered the "think positive" version come away convinced the whole approach is glib.

The model

CBT works from a simple observation: in any given situation, what a person thinks, how they feel, what their body does and how they behave all feed into one another. A thought lowers the mood, the lowered mood makes the thought more believable, the behaviour that follows removes any chance of finding out whether the thought was true, and round it goes.

Take the example of expecting to fail at something before starting. The expectation arrives, the energy goes out of the attempt, attention moves to everyone else in the room rather than to the task, and the performance suffers — which then confirms the original expectation. If the attempt succeeds anyway, it gets written off as a fluke, so the belief survives either way.

That loop is what CBT treats. Note what it does not claim: that the thought caused the problem in the first place, or that the person is thinking wrongly on purpose. The question is not where a pattern came from but what is keeping it going now — because that is the part available to be changed.

The cognitive half, done properly

A thought record is the standard tool, and it has more steps than most descriptions give it.

• The situation. Where, when, who, what was happening.
• The feeling, and how strong. Named and rated, because "anxious, 80%" is workable where "bad" is not.
• The thought that carried the charge. Out of the several thoughts present, one usually does most of the emotional work. Finding it is the skill.
• The evidence for it. Taken seriously rather than brushed aside.
• The evidence against it — including the things that get routinely discounted.
• An alternative that fits all of that evidence. Not a positive thought: a more accurate one, which is frequently less dramatic rather than more optimistic. "I found that difficult and I got through it" rather than "I was brilliant".
• The feeling, re-rated. Often down rather than gone, which is the realistic result.

It is written down, at the time or close to it, because the whole exercise depends on catching thoughts that otherwise pass unexamined. Doing it in one's head generally means doing the comfortable half.

Alongside this, therapy usually involves learning to recognise a handful of recurring patterns — catastrophising, all-or-nothing thinking, assuming what others are thinking, predicting the future, taking the blame for things outside one's control, dismissing evidence that does not fit, treating a feeling as proof, and the tyranny of what one "should" be doing. Naming a pattern as it happens takes a surprising amount of its force away, which is most of why the naming is taught.

The behavioural half, which is the engine

This is the part left out of almost every short account of CBT, and for many problems it does more work than the thought records.

Behavioural activation, for depression. Depression removes the motivation to do the things that would lift the mood, so waiting to feel like it guarantees doing less, which lowers the mood further. The treatment inverts the usual order: activity is scheduled and done before the motivation arrives, starting small enough to be possible on a bad day, with attention paid to what actually shifts anything. It sounds mechanical and it is one of the better-evidenced treatments for depression there is.

Graded exposure, for anxiety and phobias. Anxiety is maintained by avoidance: every avoided situation prevents the discovery that the feared thing either does not happen or is survivable. So feared situations are approached deliberately, in an order the person chooses, starting manageable and staying with each step long enough for the anxiety to come down on its own — which it does. The point is not to endure it heroically but to learn something that cannot be learned by argument.

The subtlety is safety behaviours: the small precautions that make a situation bearable — sitting near the exit, rehearsing every sentence, keeping a bottle of water to hand, going only with a particular person. They feel like coping and they preserve the fear, because whatever goes well gets credited to the precaution rather than to the situation being survivable. Dropping them is part of the treatment.

Exposure with response prevention, for obsessive-compulsive disorder, is the same logic applied to compulsions: the intrusive thought is allowed to be present without performing the ritual that usually follows it. It is demanding and it is the treatment that works.

Behavioural experiments cut across all of this. Rather than debating whether a prediction is true, it gets tested: the prediction is written down, something is done, and the result is compared against it. One experiment often shifts a belief that weeks of discussion left intact.

What a course looks like

CBT is structured and time-limited, usually somewhere between six and twenty sessions depending on the problem, and it has a recognisable shape.

Early sessions build a formulation — a shared map of what is happening and what is keeping it going — which is drawn collaboratively rather than handed over, and revised as things become clearer. Each session then has an agenda agreed at the start, and a review of what was tried since the last one.

There is work between sessions, and it matters more than the hour itself. An hour a week cannot change a pattern that operates all day; the therapy happens in the week and the session plans and reviews it. This is also the commonest point of failure, and worth being honest about: depression removes exactly the capacity the homework requires. A good therapist starts small enough that it is possible rather than setting tasks that confirm a sense of failure — and if the tasks are consistently too much, saying so is useful information rather than an admission.

Progress is usually tracked with short questionnaires, so improvement is visible rather than a matter of impression. The ending is planned, and typically produces a written summary — what the problem was, what helped, what the early warning signs are and what to do about them. That document is the point of the whole course: the aim of CBT is to leave someone able to do it for themselves.

What it is used for

UK guidance recommends CBT, in one adapted form or another, for depression, generalised anxiety, panic, social anxiety, specific phobias, health anxiety, obsessive-compulsive disorder (with response prevention), and post-traumatic stress disorder (in a trauma-focused form). It is adapted for eating disorders, used as an addition to medication in psychosis and bipolar disorder, and applied to chronic pain, irritable bowel syndrome and the distress that accompanies long-term physical illness.

It is also the first-choice treatment for insomnia — CBT for insomnia is more effective in the long run than sleeping tablets, which work briefly and create their own problems, and NICE has recommended a digital programme as an option in primary care. Anyone who has been offered sleeping tablets and not told this has been short-changed.

The forms it comes in

Through NHS talking therapies, CBT is delivered at two levels. Guided self-help pairs a workbook or digital programme with a practitioner who checks in regularly — brief contact, and considerably more effective than working through something alone. Individual CBT with a therapist is the fuller version, offered where the problem warrants it. Group courses exist for several conditions and are not a lesser option; for some problems the group is part of the treatment.

Several relatives are worth knowing by name. Mindfulness-based cognitive therapy is recommended for preventing relapse in recurrent depression. Acceptance and commitment therapy works on the relationship to difficult thoughts rather than on their content. Dialectical behaviour therapy, compassion-focused therapy and metacognitive therapy each developed to address something standard CBT handled less well.

When it is not the right choice

Honesty here is more useful than enthusiasm.

CBT suits people who get on with structure, written tasks and a focus on the present. Some find it too mechanical, or feel that the difficulty is the shape of a whole life rather than a set of maintaining loops — and for them a counselling or psychodynamic approach is a legitimate and better choice rather than a failure to engage.

It also attracts a fair criticism: a therapy that works on how a person interprets their circumstances can slide into treating the interpretation when the circumstances are the problem. Someone whose low mood is the reasonable consequence of insecure work, an unaffordable home or a relationship that is harming them does not have distorted thinking. Good practice includes practical problem-solving and names the real constraint as real — and anyone whose therapy appears to be locating the fault in their attitude is entitled to say so in the session.

And it does not always work. Around half of those who complete a course of treatment in NHS talking therapies recover by the service's own measure, a further group improve without reaching that threshold, and some get nothing from it. "CBT did not help me" is an ordinary clinical outcome and a reason to try something different, not evidence of having done it wrong.

Getting it

In England, NHS talking therapies can be referred to directly, free, without a GP appointment or a diagnosis. Elsewhere in the UK the route is through a GP. Mental health sets out both, along with the crisis lines.

Privately, "CBT therapist" is no more a protected title than any other, so the marker to look for is accreditation with the BABCP, which maintains the UK's accredited CBT register and can be searched directly. Choosing a therapist covers how to check, and what to be wary of.

For anyone wanting to start before an appointment comes through, the self-help CBT literature is unusually good — the better workbooks are written by the clinicians who developed the treatments, and some library services stock them on recommendation from health professionals. Apps vary enormously in quality, and the ones with evidence behind them tend to be the ones a service offers rather than the ones that advertise.

See also the first therapy session and yoga and mental health.