Behavioural techniques and rating scales

In short

• CBT has two halves, and the behavioural one is older. The cognitive half asks "is this thought true?"; the behavioural half asks "what happens if I act differently, and what do I learn?". Wolpe's method was published in 1958, before Beck's first papers.
• A stripped-down therapy did as well as the full one. In a 1996 trial, 150 people with depression received either only the behavioural part of cognitive therapy, that part plus work on thoughts, or the full therapy. The full version did no better, and negative thinking changed even in people who never worked on their thoughts directly.
• Behavioural activation is a structured treatment, not "just keep busy". It looks at what a person avoids and, step by step, brings them back to what matters to them. The UK's NICE guideline (2022) lists it among first-line options for depression.
• "Better than pills" is wrong. In one 2006 trial it worked about as well as an antidepressant in severe depression, but a reanalysis of other data did not confirm this. Decisions about medication belong with a doctor.
• Exposure doesn't work the way people used to think. How much anxiety drops during a session predicts the outcome poorly. Old fear isn't erased: a new "safe" memory forms alongside it, and the two compete. The point of an exercise is to test an expectation, not to wait until it feels easier.
• Safety behaviours keep fear going. Water in the bag, a seat by the door, a friend nearby bring relief now but stop you learning that the danger was overestimated.
• Measurement is part of CBT. Regular questionnaires slightly improve outcomes and help spot early when therapy isn't working. The PHQ-9 and GAD-7 are free to use; the BDI-II and BAI are copyrighted. A questionnaire is not a diagnosis.
• Care comes in steps. England's state service starts with the least intensive option that fits: guided self-help, a group, an online format. These are recommended options, not "CBT lite". With more severe depression, PTSD or social anxiety, treatment starts with high-intensity one-to-one therapy.

Timeline: from Wolpe to NHS Talking Therapies

Key milestones in the behavioural half of CBT and in outcome measurement. Details and sources are in the chapters below.

1. 1958

   Wolpe's book on reciprocal inhibition: systematic desensitisation

   Behavioural line

2. 1961

   The Beck Depression Inventory (BDI): CBT measures from the start

   Cognitive line

3. 1966

   Meyer describes exposure with prevention of rituals in OCD

   Behavioural line

4. 1970s

   Lewinsohn: depression as a shortage of reinforcement

   Behavioural line

5. 1996

   Jacobson and colleagues take cognitive therapy apart; the BDI-II appears

   Evidence

6. 2001

   Martell, Addis and Jacobson's book: behavioural activation as a treatment in its own right; the PHQ-9 appears

   Behavioural line

7. 2006

   Dimidjian's trial: activation, cognitive therapy and an antidepressant; the GAD-7 appears

   Evidence

8. 2008

   Craske and colleagues: exposure as new learning, not getting used to fear

   Behavioural line

9. 2010

   Pfizer opens free access to the PHQ-9 and GAD-7

   Evidence

10. 2014

    Craske and colleagues: strategies for exposure; a meta-analysis of behavioural activation (26 trials)

    Evidence

11. 2016

    COBRA: activation delivered by junior workers is no worse than CBT

    Evidence

12. 2019

    A reanalysis of the 1996 data finds no advantage for activation in severe depression

    Evidence

13. 2022

    NICE NG222: behavioural activation among first-line options for depression

    Evidence

14. 2023

    IAPT is renamed NHS Talking Therapies

    Evidence

The lines are a way of telling the story, not a strict classification (more on the "waves" in Episode 1).

How to read these notes: evidence labels

Every claim carries a label that shows how firmly you can stand on it.

These are standalone notes on CBT's behavioural techniques and rating scales. CBT stands for cognitive behavioural therapy. The label comes first in brackets, followed by the source: for example, proven by research; Cuijpers et al., 2007.

Where sources disagree, both versions are given. Questionnaire items are not reproduced here: the scales are described in our own words.

Two halves of CBT: thinking and doing

Cognitive techniques test a thought with words; behavioural techniques test it with experience. Both target the same mechanism: avoidance.

Episode 5 covered cognitive techniques: Socratic questioning and the thought record. There, a thought is examined with questions such as "what's the evidence?" or "how else could I see this?". Behavioural techniques ask something different: "what happens if I act differently, and what do I learn from it?".

The behavioural half of CBT is older than the cognitive one. Joseph Wolpe developed systematic desensitisation between 1948 and 1956, and his book came out in 1958 documented fact; Carona et al., 2023. Victor Meyer described exposure with prevention of rituals for obsessive-compulsive disorder (OCD) in 1966 documented fact; Ruggiero et al., 2018. Episode 1 covers this early history in detail.

Depression and anxiety share a mechanism that keeps the problem going: avoidance. Avoidance is anything that brings relief right now but makes life harder in the long run: not opening your email, not taking the underground, staying in bed instead of meeting a friend. The relief is real, so the avoidance sticks: behaviour is maintained by its consequences clinical consensus; Veale, 2008.

Behavioural techniques tackle avoidance in two ways:

• in depression, by bringing a person back to activities that give meaning, pleasure or a sense of "I managed it" (behavioural activation);
• in anxiety, by helping a person face what they fear and find out what actually happens (exposure).

In between sits the behavioural experiment: an action used to test a specific belief. It is a bridge between the behavioural and the cognitive halves.

Why depression shrinks a life: reinforcement and avoidance

The behavioural model of depression is simple: life holds less that is rewarding, the person withdraws, and life holds even less.

Lewinsohn's model

In the 1970s the psychologist Peter Lewinsohn proposed looking at depression through reinforcement. Reinforcement is a good consequence that follows an action: you call a friend and have a warm conversation; you finish a piece of work and feel you've managed it. In Lewinsohn's model, the key factor in depression is a low level of this kind of reinforcement, the kind that depends on the person's own actions (interpretation; Carvalho et al., 2011; Vanderkruik & Whisman, 2021). The key text is Lewinsohn (1974), "A behavioral approach to depression". A little earlier, in 1973, Charles Ferster described depression as a fall in reinforced behaviour and a rise in avoidance documented fact via secondary sources; the 1973 text itself was not opened.

According to the model, reinforcement shrinks because there are fewer potentially rewarding events, they are harder to reach, the person lacks the skills to obtain them (social skills, for instance), or there are more unpleasant events (interpretation; Carvalho et al., 2011). The practical conclusion was to increase pleasant activities and social contact. Lewinsohn created the Pleasant Events Schedule for this purpose documented fact; Vanderkruik & Whisman, 2021.

Worth stressing: this is a model, a hypothesis about a mechanism. It doesn't explain everything about depression, but a very practical treatment grew out of it.

The avoidance cycle

In outline: a loss or a hard event → less that is rewarding → low mood → the person pulls back from tasks and people → even less that is rewarding.

In a 2008 review the psychiatrist David Veale lists five forms of avoidance in depression the author's own account; Veale, 2008:

TRAP and TRAC

Training materials on behavioural activation use two acronyms clinical consensus; behavioural activation training materials, Emory University Department of Psychiatry. They are attributed to the work of Martell, Addis and Jacobson, 2001 not verified against the primary source:

• TRAP: Trigger → Response → Avoidance Pattern;
• TRAC: Trigger → Response → Alternative Coping.

Rumination is avoidance too

Rumination means going over the same questions again and again: "Why am I like this?", "Why did this happen to me?", "If only I had…". In behavioural activation, rumination counts as a form of avoidance: the person seems busy with the problem but nothing changes the author's own account; Veale, 2008. Veale's technique is to turn "why" and "if only" questions into "how" questions. Not "why can't I get up?", but "how do I get up at 9:30 tomorrow?" the author's own account; Veale, 2008.

The 1996 trial: which part of therapy works?

A study that might have been expected to show the strength of full cognitive therapy showed that its behavioural part did just as well.

By the mid-1990s Beck's cognitive therapy for depression was well studied. It has several layers: activity scheduling, work on automatic thoughts, then work on deeper beliefs (schemas). The psychologist Neil Jacobson (1949–1999) and his colleagues asked which of these parts actually does the work.

This kind of study is called a component analysis: the therapy is split into parts, and the parts are compared with the whole.

150 outpatients with major depressive disorder were randomly assigned to three groups proven by research; Jacobson et al., 1996:

1. behavioural activation only;
2. behavioural activation plus work on automatic thoughts;
3. full cognitive therapy, including work on schemas.

All groups were treated by four experienced cognitive therapists.

The result. Full cognitive therapy did no better, either at the end of treatment or six months later. Behavioural activation and work on thoughts changed negative thinking and attributional style (how a person explains events) as much as the full therapy did proven by research; Jacobson et al., 1996. Attributional style predicted outcome in the activation group but not in the full-therapy group, which hints that the mechanisms may differ (interpretation).

The surprising part: the group that never worked on thoughts directly changed its thinking just as much. Sometimes the easier route to a thought runs through action. Exactly why remains unsettled (interpretation).

Caveats (interpretation). The trial was run by people who went on to develop behavioural activation, and such studies tend to favour the researchers' own method (researcher allegiance). Also, activation was delivered by cognitive therapists who knew the cognitive model well.

A treatment in its own right

After this trial, Jacobson's colleagues Christopher Martell and Michael Addis published Depression in Context (2001) documented fact; Buyck, 2002. The book presents behavioural activation as a standalone therapy: depression is understood in the context of a person's life, and the main target of treatment is avoidance (the authors' own account, per Buyck's review, 2002). Neil Jacobson died in 1999, before the book came out documented fact; obituary in Behavioural and Cognitive Psychotherapy. Today the main clinician's manual is Behavioral Activation for Depression by Martell, Dimidjian and Herman-Dunn documented fact; Guilford.

How behavioural activation works

A structured method with several steps, not advice to "pull yourself together". Below is how it is done in therapy.

A course usually takes 12 to 24 sessions clinical consensus; Veale, 2008. In the UK COBRA trial there were up to 20 sessions over 16 weeks plus up to four booster sessions documented fact; Mental Elf, 2016.

The steps

1. Explaining the model. Therapist and client go through the cycle "less reward → avoidance → worse still". The person sees that their state makes sense rather than "something being wrong with them".
2. Activity monitoring. Typically for a week, the person keeps a diary of what they did and how their mood was. Nothing changes yet; it's observation only. In Beck's cognitive therapy, each activity also gets two ratings: mastery ("how well did I manage?") and pleasure ("how much did I enjoy it?"). The diary shows how narrow life has become and how actions link to mood clinical consensus; Veale, 2008.
3. Values. What matters to the person in relationships, work or study, health, home, leisure. Not what they "should" do, but what they value. Activities are tied to these directions the author's own account; Veale, 2008.
4. Functional analysis. Working out what a behaviour does for the person. Veale describes an ABCDE format: antecedents; the behaviour and thinking processes; its immediate and unintended consequences; alternative actions linked to values; and their effect the author's own account; Veale, 2008.
5. An activity ladder. A list of tasks from easy to hard. Items are concrete and are not themselves a form of avoidance clinical consensus; Leonard & McCullars, 2020.
6. Scheduling. An activity is booked like an appointment with yourself: when, where, for how long. Not "I should go for a walk", but "Tuesday at 11:00, 15 minutes, to the park and back".
7. Working with rumination. "Why" questions become "how" questions the author's own account; Veale, 2008.
8. Troubleshooting at every session. Homework is more likely to get done when the client helps set it and the time and place are agreed in advance clinical consensus; Veale, 2008.

The core principle: act according to the plan, not according to your mood the author's own account; Veale, 2008. It is often summed up as "outside-in": behaviour changes first, and feelings catch up (interpretation; we could not confirm this wording in the primary source).

Common obstacles

Veale describes typical difficulties the author's own account; Veale, 2008:

• trying to change everything at once instead of taking small steps;
• waiting until you "feel like it" instead of following the plan;
• using problem-solving on hypothetical "what ifs" rather than real problems.

Another trap is filling the day with activity for its own sake. The aim is not more activity but getting back meaning, pleasure and a sense of mastery, and stopping avoidance.

How well it works: the evidence on behavioural activation

The evidence for the method is strong, with important caveats. The biggest one: "better than antidepressants" is the wrong conclusion.

Meta-analyses

• Cuijpers and colleagues (2007): 16 studies, 780 participants. Compared with control groups the effect was large: 0.87 (95% confidence interval 0.60–1.15). Compared with cognitive therapy (10 studies), there was almost no difference: 0.02. The gains held at follow-up proven by research; Cuijpers et al., 2007.
• Ekers and colleagues (2014): 26 randomised trials, 1,524 participants. Against controls, SMD −0.74 (−0.91 to −0.56); against medication in 4 studies, −0.42 (−0.83 to −0.00). The authors themselves note that most studies were of low quality and follow-up periods were short proven by research; Ekers et al., 2014.

How to read these numbers. An effect size shows how far apart groups are in standard units: around 0.2 is small, 0.5 medium and 0.8 large (Cohen's conventional benchmarks). A large effect against a control usually shrinks when a method is compared with another active therapy.

The 2006 trial and what happened next

Sona Dimidjian and colleagues compared behavioural activation, cognitive therapy, an antidepressant and a placebo in 241 adults with major depression. The headline finding: among more severely depressed patients, behavioural activation was comparable to antidepressant medication, and both were significantly better than cognitive therapy proven by research; Dimidjian et al., 2006.

According to a critical summary of the trial, remission in severe depression was 56% after activation, 36% after cognitive therapy and 23% on the antidepressant; in moderate depression there were no significant differences. 44% of the antidepressant group dropped out, possibly because the paroxetine dose was raised to 40 mg within six weeks (interpretation; Baas & Ruhé, 2007; figures from the summary, as the full article could not be opened).

Then came a follow-up of people who had responded. In the first year, relapse was most common among those switched from medication to placebo (59%), and less common after cognitive therapy (39%) and activation (50%). Cognitive therapy protected significantly better than stopping medication; activation showed only a non-significant trend. Over two years, both therapies did at least as well as staying on medication proven by research; Dobson et al., 2008.

The key caveat. In 2019 Lorenzo-Luaces and Dobson reanalysed the 1996 trial data the same way and found no advantage for activation over cognitive therapy in severe depression. They state plainly that the 2006 results were not replicated proven by research; Lorenzo-Luaces & Dobson, 2019.

COBRA: it doesn't take a psychotherapist

The UK COBRA trial (2016) included 440 adults with depression. Behavioural activation was delivered by junior mental health workers after five days of training; CBT by professional therapists. After 12 months PHQ-9 scores had fallen by the same amount: a difference of 0.1 points (95% CI −1.3 to 1.5), so activation was no worse than CBT proven by research; Richards et al., 2016. The intervention itself cost about 21% less for activation documented fact; NIHR Evidence, 2016.

Limitations: a large share of participants were taking antidepressants, and this was not fully accounted for; about a third in each group attended fewer than eight sessions (interpretation; NIHR Evidence, 2016; Mental Elf, 2016).

What the guidelines say

In NICE's guideline on depression (NG222, 2022), behavioural activation appears among first-line options: group and individual formats for less severe depression, and the individual format for more severe depression, alongside CBT and antidepressants clinical consensus; NICE, 2022. CBT for depression as a whole is covered in Episode 7.

Exposure: from Wolpe to today

Exposure means facing, on purpose, what a person fears and avoids. What survives from the original idea is the fear ladder and the distress scale, not compulsory relaxation.

Systematic desensitisation

The South African psychiatrist Joseph Wolpe (1915–1997) worked from the idea of reciprocal inhibition: you can't be deeply relaxed and very anxious at the same time. The method went like this: the person learns deep muscle relaxation, a list of frightening situations is drawn up from mildest to worst, and while relaxed, the person imagines them one by one documented fact; Carona et al., 2023. Wolpe reported that nearly 90% of 210 patients were "apparently cured or much improved" the author's own account. That is the author's own report without a control group, not a clinical trial.

What has survived from Wolpe:

• the fear ladder (hierarchy): a list of situations in order of rising anxiety;
• the SUDS scale (Subjective Units of Distress) from 0 to 100. It is attributed to Wolpe and usually dated to his book The Practice of Behavior Therapy, 1969 not verified against the book.

What did not hold up. According to Craske and colleagues (2012), relaxation does not boost graded exposure, and exposure that isn't strictly graded works just as well as graded exposure proven by research, as summarised in Carona et al., 2023. The active ingredient seems to be facing the fear itself.

Kinds of exposure

documented fact; Torrents-Rodas et al., 2015

Exposure and response prevention (ERP) in OCD

OCD has its own version: the person faces what triggers the obsession and does not perform the ritual. A ritual can be hand-washing, a hidden mental act, or reassurance seeking: asking again and again for confirmation that everything is fine ("Are you sure you locked the door?" for the fifth time). In one of Craske's clinical examples, exercises were planned so that the patient's wife was not present and did not reassure him documented fact; Torrents-Rodas et al., 2015. OCD is covered in detail in Episode 9.

What the evidence shows

The common thread: large effects against a waiting list shrink noticeably against active treatments and in high-quality studies (interpretation). Worth remembering whenever you hear "evidence-based method". The evidence for CBT as a whole is covered in Episode 14.

Do you have to wait for anxiety to drop?

For a long time an exercise was thought to work only if anxiety came down. Research showed this isn't so, and explained why.

The old explanation: habituation

Habituation is a reduced response to something met repeatedly. The logic was that a person stays in the frightening situation until anxiety falls, and next time there is less of it. Emotional processing theory (Foa & Kozak, 1986) is linked to this: a drop in fear within and between sessions was taken as a sign of success (interpretation; Craske et al., 2008; Krause et al., 2022).

The new explanation: inhibitory learning

In 2008 Michelle Craske, a professor of psychology and psychiatry at UCLA, and her colleagues pulled the research together and reached an inconvenient conclusion: how much fear drops by the end of an exposure trial, or a series of them, does not predict the outcome proven by research; Craske et al., 2008, 2014.

The explanation is inhibitory learning. During extinction, the old link "underground → danger" is not erased; it stays intact. A new link "underground → safe" forms alongside it, and the two compete. Which one wins at a given moment depends on place, time and circumstances proven by research; Craske et al., 2014.

That is why fear can come back. Extinction research describes four forms of return of fear documented fact; Craske et al., 2014:

This is not therapy failing; it is how memory works.

What this means in practice

• The aim is to test an expectation. Before an exercise, the person spells out exactly what they fear ("I'll have an attack and pass out") and how strongly they believe it. The end of the exercise is set by the expectation being disconfirmed, not by fear coming down proven by research; Craske et al., 2014.
• Performance in the moment is not learning. The end of an exposure session "is just an index of what they performed in a moment", Craske says in an interview; learning is tested later the author's own account; SSCP interview with M. Craske.
• Variety. Exercises are done in different places, at different times, in a different order: this makes the new memory stronger.

In their 2014 review, Craske and colleagues list strategies to strengthen exposure documented fact; Craske et al., 2014, abstract:

1. expectancy violation: design the exercise to disconfirm the prediction as strongly as possible;
2. deepened extinction: combine several feared cues;
3. occasional reinforced extinction;
4. removal of safety signals;
5. variability: different order, length, cues;
6. retrieval cues that help recall the new learning;
7. multiple contexts;
8. affect labelling: putting feelings into words.

Limits of the model

Many of the strategies were tested in laboratory and analogue studies (people without a clinical diagnosis) rather than in large clinical trials (interpretation; Krause et al., 2022). A substantial number of people don't achieve clinically significant relief from exposure proven by research; Craske et al., 2014. Some protocols, for example for post-traumatic stress disorder (PTSD), still use SUDS and falling anxiety as a guide. It is a strong scientific model, not a final verdict on the old one.

The fear ladder: how it's built in therapy

A composite example to show the logic. The hierarchy is built together with a professional, and everyone's numbers differ.

First, three things are mapped out:

• what he avoids: the underground, especially at rush hour and on long stretches between stations;
• what exactly he fears: not "the underground" in general, but a specific outcome: "I'll have an attack, pass out, and nobody will help";
• which safety behaviours he uses: a friend nearby, water, a seat by the doors, calming music, checking his pulse.

SUDS anchor points

To rate consistently, anchor points are agreed: 0 — lying at home, completely calm; 25 — slight tension; 50 — noticeable anxiety, hard to concentrate; 75 — strong anxiety, want to leave; 100 — the worst I can imagine.

Ten rungs

Rung 10 combines two fears: crowds and body sensations. Craske calls this deepened extinction documented fact; Craske et al., 2014. Deliberately bringing on body sensations (interoceptive exposure) is part of panic protocols; with heart conditions, asthma or pregnancy such exercises are agreed with a doctor first clinical consensus.

Then and now

(interpretation based on Craske et al., 2014; Torrents-Rodas et al., 2015)

One exercise sheet

For rung 6, "one stop alone":

1. What do I expect? "An attack will start, I'll collapse, nobody will help."
2. How strongly do I believe it? 35%.
3. How will I know it happened? I'll end up on the carriage floor.
4. What am I dropping? The seat by the doors, calling a friend.
5. Do it: get on, stand in the middle, ride one stop, get off.
6. What happened? Heart pounding, SUDS 65, I didn't collapse and I got there.
7. What did I learn? "Strong anxiety is unpleasant but not dangerous; I can travel with it."
8. How strongly do I believe the original expectation now? 15%.
9. What's next? Repeat on a different line, then rung 7.

Common mistakes

• Getting off the train as soon as anxiety rises: this reinforces avoidance.
• Secretly keeping a safety behaviour ("I'm just keeping a pill in my pocket"). If the person links the outcome to the pill, it acts as a safety signal and gets in the way of learning; Craske discusses exactly this example the author's own account; SSCP interview with M. Craske.
• Practising rarely, once a month.
• Treating a day when anxiety didn't fall as a failure.
• Starting at the scariest rung without preparation.

Safety behaviours and behavioural experiments

Safety behaviours quietly keep fear alive. A behavioural experiment is a way to test a belief by doing something.

Safety behaviours

A safety behaviour is anything a person does to stop the feared thing from happening: carrying water, sitting near the exit, travelling only with a friend, keeping a pill on hand. A safety signal is an object or person that makes things feel safer: another person, the therapist, a phone, medication, water documented fact; Torrents-Rodas et al., 2015.

Why this is a problem: the person learns not "the underground is safe" but "the underground is safe if I have water". Without the water, the fear comes back. As early as 1991 Paul Salkovskis showed that instructions to drop safety behaviours improved outcomes proven by research; Salkovskis, 1991, via Craske et al., 2014.

The evidence is mixed, though: some studies found safety behaviours harmful, others did not sources disagree; Torrents-Rodas et al., 2015. The general recommendation is to identify them and phase them out — gradually, so the person doesn't drop out of therapy, or straight away if they are ready clinical consensus; Craske et al., 2014; Torrents-Rodas et al., 2015.

The behavioural experiment

A behavioural experiment is a planned action, in or between sessions, whose purpose is to gather new information to test a belief or build a new one. The main handbook is the Oxford Guide to Behavioural Experiments in Cognitive Therapy (Oxford University Press, 2004); the chapter on designing effective experiments is by James Bennett-Levy documented fact; Bennett-Levy, 2004. We couldn't open the handbook's text, so the steps below are the general outline as it is usually described not verified against the handbook:

1. The belief and how strongly you hold it, 0–100%.
2. A specific prediction: exactly what will happen and how I'll know.
3. An alternative explanation.
4. A plan: what, where, when; which safety behaviours I'll drop.
5. Carrying it out.
6. What actually happened.
7. What I learned; belief in the old and new ideas, 0–100%.

How is an experiment different from exposure? Classic exposure aimed at habituation; an experiment aims at testing a belief. In the inhibitory learning model the gap almost disappears: Craske writes that her approach overlaps with models in which exposure is used for belief disconfirmation, although it is broader documented fact; Craske et al., 2014. Both answer the question "what do I need to learn?".

One more practical point: Craske suggests moving work on thoughts to after exposure rather than before, so that the experience can surprise the author's own account; SSCP interview with M. Craske. That said, in one small study of people with claustrophobia, cognitive work before exposure did not worsen the outcome proven by research, a single study; Krause et al., 2022.

Relaxation and problem-solving

Two long-standing skills that are still part of CBT, though their role is not what it once was.

Relaxation

Progressive muscle relaxation means tensing and releasing muscle groups in turn. It was described by the physician Edmund Jacobson in Progressive Relaxation, first published by the University of Chicago Press in 1929; the 1938 edition is the one often cited documented fact; Jacobson, 1929, Center for Research Libraries catalogue record. He is a different person from Neil Jacobson of the 1996 trial; the two are often confused.

Applied relaxation (Lars-Göran Öst, 1987; not verified against the primary source) is the skill of relaxing quickly in real situations, trained in stages.

The role of relaxation today:

• as a required step before exposure: no, because it doesn't boost exposure proven by research; Craske et al., 2012, via Carona et al., 2023;
• as a treatment in its own right for generalised anxiety disorder (GAD): yes. NICE offers a choice of CBT or applied relaxation, usually 12–15 weekly one-hour sessions with a trained practitioner clinical consensus; NICE CG113. Psychological therapy for GAD overall gives g = 0.84 against control; there are few comparisons of applied relaxation with other methods, with some signs that CBT does better in the long term proven by research; Cuijpers et al., 2014. Anxiety disorders are covered in Episode 8.

Problem-solving

Problem-solving therapy is a step-by-step method for real problems: seeing the problem as a solvable part of life; defining it clearly; brainstorming options without judging them; weighing pros and cons; choosing; making a plan; carrying it out; reviewing the result. The method goes back to D'Zurilla and Goldfried's article "Problem solving and behavior modification" (documented fact; D'Zurilla & Goldfried, 1971, Journal of Abnormal Psychology, 78(1), 107–126; full text not opened).

Evidence in depression proven by research; Cuijpers et al., 2018: 30 trials, 3,530 participants. Against control, g = 0.79, but with very high variation between studies. In the 9 studies at low risk of bias, g = 0.34, comparable to other psychotherapies. In direct comparisons it came out slightly ahead of other therapies, but the authors suggest this may reflect researcher allegiance.

An important limit: problem-solving is for real tasks, not for hypothetical "what if everything falls apart tomorrow" worries the author's own account; Veale, 2008.

Why CBT keeps measuring

Questionnaires in CBT are not paperwork: they show whether therapy is working and when to change course.

CBT has been built on measurement from the start: the Beck Depression Inventory (BDI) appeared in 1961 documented fact; Beck Institute. Today this is called routine outcome monitoring: the client regularly fills in a short questionnaire and the therapist tracks the trend. Some systems compare a client's trajectory with the expected one and flag when they are "not on track".

Does it help?

A large 2021 meta-analysis (58 studies, 110 effect sizes, 21,699 patients) found that feedback from questionnaires slightly improves outcomes: d = 0.15 (95% CI 0.10–0.20); for clients who are not on track, d = 0.17. Drop-out also falls a little proven by research; de Jong et al., 2021. It's a small but real effect: regular measurement doesn't work miracles, but it helps catch early that therapy isn't working.

For anxiety disorders, NICE explicitly recommends routine outcome measurement and involving the client in reviewing how well treatment is working clinical consensus; NICE CG113.

Steps of care: low and high intensity

England's state psychological therapy service is organised in steps. That explains why one person is offered a book or an online course with a practitioner's support, while another goes straight to a course of one-to-one therapy.

England has a state-funded psychological therapy programme, NHS Talking Therapies; until January 2023 it was called IAPT documented fact; Nuffield Trust.

The least intensive intervention that fits

The programme follows a stepped care model: people are offered the least intrusive intervention appropriate to their needs clinical consensus; NHS Talking Therapies Manual, 2024. NICE's guideline on generalised anxiety disorder puts it as offering the least intrusive, most effective intervention first clinical consensus; NICE CG113, 1.2.1. The model works in both directions: people who don't recover at the lower step are stepped up to a course of high-intensity treatment, and stepped down when a less intensive option becomes more appropriate. The number of sessions should never be restricted arbitrarily clinical consensus; NHS Talking Therapies Manual, 2024.

Who delivers low intensity

Low-intensity treatment is delivered by psychological wellbeing practitioners (PWPs), and high-intensity treatment by high-intensity therapists (HITs) documented fact; NHS Talking Therapies Manual, 2024. PWPs are not psychotherapists; they have their own training route. At UCL, for example, it takes a year: one day a week at university and four days working in a service under supervision, learning low-intensity cognitive behavioural interventions delivered as guided self-help for people with mild to moderate depression and anxiety documented fact; UCL, PG Cert programme description. The programme counts guided self-help, digitally enabled therapy, psychoeducation groups and group physical activity programmes as low intensity documented fact; NHS Talking Therapies Manual, 2024. In practice this averages 3.7 sessions, against 8.3 for high intensity proven by research; Robinson, Kellett & Delgadillo, 2020; more in Episode 4.

Who needs high intensity from the start

Under NICE guidance, people with more severe depression, social anxiety disorder or PTSD should receive high-intensity treatment from the outset; for the latter two, low-intensity treatment lacks a strong evidence base clinical consensus; NHS Talking Therapies Manual, 2024, citing NICE. In OCD, low intensity fits only mild impairment clinical consensus; NICE CG31; more in Episode 9.

Formats: recommended options, not "CBT lite"

For less severe depression, NICE lists among the options guided self-help — printed or digital materials with support from a trained practitioner in person, by phone or online, usually 6–8 sessions — and group CBT: usually 8 sessions, two practitioners, around eight participants. The options were chosen on evidence of clinical and cost-effectiveness, and the choice is discussed with the person clinical consensus; NICE NG222, summarised in Kendrick et al., 2022. In the English programme, group CBT counts as high intensity and must be led by a trained high-intensity therapist. Digital options don't suit everyone, so they are always offered alongside others, including non-digital ones, and services should support an informed choice of format clinical consensus; NHS Talking Therapies Manual, 2024.

How outcomes are measured there

NHS Talking Therapies calculates outcomes from two short questionnaires:

• "caseness": a PHQ-9 score of 10 or more, or a GAD-7 score of 8 or more, at the start of treatment;
• recovery: moving from caseness at the start of treatment out of caseness at the end documented fact; Moller et al., 2019.

In 2017/18, 50.8% recovered documented fact; Moller et al., 2019. In May 2026, across England: recovery 50.7%, reliable improvement 68.6%, reliable recovery 48.1%; 50,884 people finished a course of treatment that month documented fact; NHS England, 2026. Reliable improvement is a change larger than measurement error; in the programme's metrics it means a fall of at least 6 points on the PHQ-9 or at least 4 on the GAD-7 (documented fact; IAPT Manual, Appendices and helpful resources, NHS England, 2018, updated 2019, table 1; MindStep, 2018).

Criticism (interpretation): "recovery" based on two questionnaires is a simplification; there is pressure to hit targets; and outcomes differ markedly between deprived and affluent areas — in 2024/25 about 40.5% in the most deprived areas versus 52.4% in the least deprived documented fact; Nuffield Trust.

Rating scales: what they measure and which you can use

A questionnaire shows how strong symptoms are; it doesn't diagnose. Each scale also has its own legal status: some are free to use, others are protected.

Three rules

1. A scale is not a diagnosis. A questionnaire shows how marked symptoms have been over a recent period (for the PHQ-9, GAD-7 and WHO-5, the last two weeks). A diagnosis is made by a professional through a clinical interview. The copyright holders say so themselves: a low WHO-5 score is a reason for further assessment documented fact; WHO, WHO-5; the PSS is not a diagnostic tool and has no cut-offs documented fact; PhenX; the PCL-5 does not replace a structured interview documented fact; ISTSS.
2. Look at the trend, not one data point. A high score is a reason to see a professional; a low score doesn't mean you shouldn't, if you feel bad.
3. Items from protected scales are not published. They can be named and described in your own words.

Scales that are free to use

How accurate the cut-offs are proven by research:

• PHQ-9. In a study of 6,000 patients (580 of whom were also interviewed by a mental health professional), a score of 10 or more identified major depression with 88% sensitivity and 88% specificity (Kroenke et al., 2001). Sensitivity is the share of people with the condition whom the test picks up; specificity is the share of people without it whom the test correctly leaves out.
• GAD-7. A score of 10 or more identified GAD with 89% sensitivity and 82% specificity (Spitzer et al., 2006). But the same cut-off also flags panic disorder (74% / 81%), social anxiety (72% / 80%) and PTSD (66% / 81%) (Kroenke et al., 2010, via CamCOPS). So the GAD-7 is a general anxiety indicator, not a "test for GAD" (interpretation).

Why the PHQ-9 and GAD-7 are free. They were developed by Robert Spitzer, Janet Williams and Kurt Kroenke with support from Pfizer; in 2010 the company made them available "without copyright restriction and at no charge", with translations in nearly 80 languages documented fact; Pfizer, 2010; News-Medical, 2010. How well each specific translation has been validated was not checked for these notes.

Scales you can't simply download and post

Using a scale sensibly

If you fill in a free scale, for example together with a professional:

1. complete it at the same interval, such as every two weeks;
2. note the date and score, and look at the trend;
3. answer for the last two weeks, not just today;
4. if the score keeps rising or is above the "moderate" cut-off, see a professional;
5. bring your results to your first appointment: they help.

Try this: one postponed task

A one-minute exercise to feel the principle "act on the plan, not the mood". It is not treatment and not a substitute for a professional.

This is how behavioural activation tests the link between action and mood: instead of waiting until you "feel like it", take a small step and see what changes.

Check yourself and key takeaways

Short questions with answers, then the essentials on one page.

Check yourself

1. What did the 1996 component analysis of cognitive therapy show? — The full therapy did no better than activation alone or activation plus work on thoughts; negative thinking changed equally.
2. Why is "behavioural activation beats antidepressants" wrong? — It rests on a single 2006 trial; a reanalysis of the 1996 data found no advantage in severe depression.
3. What are TRAP and TRAC? — Trigger → response → avoidance pattern, and trigger → response → alternative coping.
4. Why is a drop in anxiety during exposure a poor indicator? — It doesn't predict the outcome: the old memory isn't erased, and what matters is new learning, which shows up later.
5. When does an exercise end in the inhibitory learning approach? — When the expectation has been tested, even if anxiety is still high.
6. How does a behavioural experiment differ from exposure? — An experiment aims to test a specific belief; in the modern model of exposure the two converge.
7. Which scales are free to use and which are protected? — PHQ-9, PHQ-2, GAD-7: free; WHO-5 and PSS-10: non-commercially; BDI-II, BAI, Y-BOCS, ISI, WSAS: protected.
8. Why does item 9 of the PHQ-9 need special attention? — It asks about thoughts of death or self-harm; any answer other than "not at all" means talking to a professional straight away.

Key takeaways

• Behavioural techniques change how people feel through action and experience; the shared enemy in depression and anxiety is avoidance.
• Behavioural activation is a treatment in its own right with good evidence and a place in NICE guidance, but it is not "better than pills".
• Exposure works as new learning: test the expectation and drop safety behaviours, rather than waiting for relief.
• Fear can return, and that is not failure but a property of memory; that's why exercises are repeated in different places and at different times.
• Relaxation doesn't boost exposure, but applied relaxation is a treatment in its own right for GAD.
• Regular measurement slightly improves therapy; a scale is not a diagnosis; each scale has its own legal status.
• Stepped care: start with the least intensive option that fits and step up if it doesn't help; supported self-help and groups are recommended formats, not a substitute for "real" therapy.
• Severe depression, trauma, OCD, frequent panic attacks and eating disorders call for a professional.

Sources

1. Baas, K. D., & Ruhé, H. G. (2007). Gedragsactivatie is effectief bij de behandeling van depressie [summary of Dimidjian et al., 2006]. Tijdschrift voor Psychiatrie, 49(6), 411. https://www.tijdschriftvoorpsychiatrie.nl/media/2/articles_1642pdf.pdf
2. Bennett-Levy, J. (2004). Devising effective behavioural experiments. In Oxford Guide to Behavioural Experiments in Cognitive Therapy. Oxford University Press. doi:10.1093/med:psych/9780198529163.003.0002
3. Buyck, D. (2002). Review: Depression in Context (Martell, Addis, Jacobson, 2001). Primary Care Companion J Clin Psychiatry, 4(5), 201. https://psychiatrist.com/read-pdf/24563
4. Carona, C., Ramos, K., & Salvador, C. (2023). Psychotherapy by reciprocal inhibition: Wolpe's unique legacy. BJPsych Advances, 29, 218–222. https://www.cambridge.org/core/product/A34DE39AC4C20EA598209C8150F9117F/core-reader
5. Carpenter, J. K., Andrews, L. A., Witcraft, S. M., Powers, M. B., Smits, J. A. J., & Hofmann, S. G. (2018). Cognitive behavioral therapy for anxiety and related disorders: A meta-analysis of randomized placebo-controlled trials. Depression and Anxiety, 35(6), 502–514.
6. Carvalho, J. P., et al. (2011). The Reward Probability Index: Design and validation of a scale measuring access to environmental reward. Behavior Therapy, 42(2), 249–262.
7. CamCOPS documentation (University of Cambridge): PHQ-9, GAD-7, Y-BOCS, WSAS, PCL-5, EDE-Q. https://camcops.readthedocs.io/
8. Craske, M. G., Kircanski, K., Zelikowsky, M., Mystkowski, J., Chowdhury, N., & Baker, A. (2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46, 5–27.
9. Craske, M. G., Treanor, M., Conway, C. C., Zbozinek, T., & Vervliet, B. (2014). Maximizing exposure therapy: An inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23. doi:10.1016/j.brat.2014.04.006. https://escholarship.org/content/qt5x26t4nd/qt5x26t4nd.pdf
10. Cuijpers, P., van Straten, A., & Warmerdam, L. (2007). Behavioral activation treatments of depression: A meta-analysis. Clinical Psychology Review, 27(3), 318–326. doi:10.1016/j.cpr.2006.11.001
11. Cuijpers, P., Sijbrandij, M., Koole, S., Huibers, M., Berking, M., & Andersson, G. (2014). Psychological treatment of generalized anxiety disorder: A meta-analysis. Clinical Psychology Review, 34, 130–140.
12. Cuijpers, P., de Wit, L., Kleiboer, A., Karyotaki, E., & Ebert, D. D. (2018). Problem-solving therapy for adult depression: An updated meta-analysis. European Psychiatry, 48(1), 27–37.
13. de Jong, K., Conijn, J. M., Gallagher, R. A. V., Reshetnikova, A. S., Heij, M., & Lutz, M. C. (2021). Using progress feedback to improve outcomes and reduce drop-out, treatment duration, and deterioration: A multilevel meta-analysis. Clinical Psychology Review, 85, 102002. https://research.vu.nl/en/publications/using-progress-feedback-to-improve-outcomes-and-reduce-drop-out-t/
14. Dimidjian, S., Hollon, S. D., Dobson, K. S., et al. (2006). Randomized trial of behavioral activation, cognitive therapy, and antidepressant medication in the acute treatment of adults with major depression. Journal of Consulting and Clinical Psychology, 74(4), 658–670. doi:10.1037/0022-006X.74.4.658
15. Dobson, K. S., Hollon, S. D., Dimidjian, S., et al. (2008). Randomized trial of behavioral activation, cognitive therapy, and antidepressant medication in the prevention of relapse and recurrence in major depression. Journal of Consulting and Clinical Psychology, 76(3), 468–477.
16. Ekers, D., Webster, L., Van Straten, A., Cuijpers, P., Richards, D., & Gilbody, S. (2014). Behavioural activation for depression: An update of meta-analysis of effectiveness and sub group analysis. PLoS ONE, 9(6), e100100.
17. Jacobson, N. S., Dobson, K. S., Truax, P. A., Addis, M. E., Koerner, K., Gollan, J. K., Gortner, E., & Prince, S. E. (1996). A component analysis of cognitive-behavioral treatment for depression. Journal of Consulting and Clinical Psychology, 64(2), 295–304. doi:10.1037/0022-006X.64.2.295
18. Krause, K. L., Koerner, N., & Antony, M. M. (2022). Cognitive restructuring before versus after exposure. Behavior Modification, 46(6), 1432–1459.
19. Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606–613.
20. Leonard, R., & McCullars, A. (2020). Treating depression during COVID-19: Building connections and increasing engagement. Rogers Behavioral Health (webinar).
21. Lewinsohn, P. M. (1974). A behavioral approach to depression. In R. J. Friedman & M. M. Katz (Eds.), The Psychology of Depression. Wiley (not opened).
22. Lorenzo-Luaces, L., & Dobson, K. S. (2019). Is behavioral activation (BA) more effective than cognitive therapy (CT) in severe depression? A reanalysis of a landmark trial. International Journal of Cognitive Therapy, 12, 73–82. https://link.springer.com/article/10.1007/s41811-019-00044-8
23. Martell, C. R., Addis, M. E., & Jacobson, N. S. (2001). Depression in Context: Strategies for Guided Action. W. W. Norton.
24. The Mental Elf (2016). Behavioural activation not inferior to CBT for depression: the COBRA RCT. https://www.nationalelfservice.net/treatment/cbt/behavioural-activation-not-inferior-to-cbt-for-depression-the-cobra-rct/
25. MindStep (2018). Table 1: outcome definitions. BMC Psychiatry. https://bmcpsychiatry.biomedcentral.com/articles/10.1186/s12888-018-1987-1/tables/1
26. Moller, N. P., Ryans, G., Rollings, J., & Barkham, M. (2019). The 2018 UK NHS Digital annual report on the IAPT programme: A brief commentary. BMC Psychiatry, 19, 252.
27. NHS England / NHS Digital (2026). NHS Talking Therapies monthly statistics, performance May 2026: outcomes. https://digital.nhs.uk/data-and-information/publications/statistical/nhs-talking-therapies-monthly-statistics-including-employment-advisors/performance-may-2026/outcomes
28. NICE (2011, updated). Generalised anxiety disorder and panic disorder in adults: management (CG113). https://www.nice.org.uk/guidance/cg113
29. NICE (2022). Depression in adults: treatment and management (NG222). https://www.nice.org.uk/guidance/ng222
30. NIHR Evidence (2016). Simpler, cheaper therapy — behavioural activation can be as good as CBT for treating depression. https://evidence.nihr.ac.uk/alert/simpler-cheaper-therapy-behavioural-activation-can-be-as-good-as-cbt-for-treating-depression/
31. Nuffield Trust. NHS Talking Therapies (IAPT) programme. https://www.nuffieldtrust.org.uk/resource/improving-access-to-psychological-therapies-iapt-programme
32. Öst, L.-G., Havnen, A., Hansen, B., & Kvale, G. (2015). Cognitive behavioral treatments of obsessive-compulsive disorder: A systematic review and meta-analysis of studies published 1993–2014. Clinical Psychology Review.
33. Pearson Assessments: Beck Anxiety Inventory; Beck Depression Inventory-II (product pages).
34. Pfizer (2010). Pfizer to offer free public access to mental health assessment tools to improve diagnosis and patient care. https://www.pfizer.com/news/press-release/press-release-detail/pfizer_to_offer_free_public_access_to_mental_health_assessment_tools_to_improve_diagnosis_and_patient_care
35. PhenX Toolkit. Perceived Stress (protocol 180801). https://www.phenxtoolkit.org/protocols/view/180801
36. Reid, J., Laws, K., Drummond, L., Vismara, M., Grancini, B., Mpavaenda, D., & Fineberg, N. A. (2021). Cognitive behavioural therapy with exposure and response prevention in the treatment of obsessive-compulsive disorder: A systematic review and meta-analysis. Comprehensive Psychiatry (as summarised in BJPsych Open).
37. Richards, D. A., Ekers, D., McMillan, D., et al. (2016). Cost and outcome of behavioural activation versus cognitive behavioural therapy for depression (COBRA): A randomised, controlled, non-inferiority trial. Lancet, 388(10047), 871–880.
38. Ruggiero, G. M., Spada, M. M., Caselli, G., & Sassaroli, S. (2018). A historical and theoretical review of cognitive behavioral therapies. Journal of Rational-Emotive & Cognitive-Behavior Therapy, 36, 378–403.
39. Society of Clinical Psychology (SSCP). Transcript of "Maximizing Exposure Therapy for Anxiety Disorders" — interview with Michelle Craske. https://sscpweb.org/page-18148
40. Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092–1097.
41. Torrents-Rodas, D., Fullana, M. A., Vervliet, B., Treanor, M., Conway, C., Zbozinek, T., & Craske, M. G. (2015). Maximizar la terapia de exposición: un enfoque basado en el aprendizaje inhibitorio. Revista de Psicopatología y Psicología Clínica, 20(1), 1–24.
42. Vanderkruik, R., & Whisman, M. A. (2021). A cross-cultural examination of pleasant events and depressive symptoms. Journal of Clinical Psychology, 77, 1384–1393.
43. Veale, D. (2008). Behavioural activation for depression. Advances in Psychiatric Treatment, 14, 29–36. doi:10.1192/apt.bp.107.004051
44. Wolitzky-Taylor, K. B., Horowitz, J. D., Powers, M. B., & Telch, M. J. (2008). Psychological approaches in the treatment of specific phobias: A meta-analysis. Clinical Psychology Review, 28, 1021–1037.
45. World Health Organization. WHO-5 Well-Being Index (licence CC BY-NC-SA 3.0 IGO).
46. Behavioural and Cognitive Psychotherapy. "To do so much": Neil S. Jacobson, 23 February 1949 – 2 June 1999 (obituary). https://www.cambridge.org/core/journals/behavioural-and-cognitive-psychotherapy/article/to-do-so-much-neil-s-jacobson-23-february-1949-2-june-1999/9B704BA2FD7B2ECB26C5A3ECC0E8DEF4
47. NHS England (2024). NHS Talking Therapies for anxiety and depression manual (v7; first published 2018). https://england.nhs.uk/wp-content/uploads/2018/06/NHS-talking-therapies-manual-v7-1.pdf
48. Kendrick, T., Pilling, S., Mavranezouli, I., Megnin-Viggars, O., Ruane, C., Eadon, H., & Kapur, N. (2022). Management of depression in adults: Summary of updated NICE guidance. BMJ, 378, o1557. https://discovery.ucl.ac.uk/10152602/3/Mavranezouli_BMJ%20NICE%20Depression%20summary%20accepted.pdf
49. NICE (2011, updated). Generalised anxiety disorder and panic disorder in adults: management (CG113), recommendations 1.2.1, 1.2.14. https://nice.org.uk/guidance/cg113/chapter/1-guidance
50. NICE (2005). Obsessive-compulsive disorder and body dysmorphic disorder: treatment (CG31), via OCD-UK: https://www.ocduk.org/nice-guidelines-for-the-treatment-of-ocd/
51. Robinson, L., Kellett, S., & Delgadillo, J. (2020). Dose-response patterns in low and high intensity cognitive behavioral therapy for common mental health problems. Depression and Anxiety, 37(3), 285–294. https://eprints.whiterose.ac.uk/154899/1/CBT_Dose_Response_pre-print_2019.pdf
52. UCL. Low Intensity Cognitive Behavioural Interventions for Common Mental Health Problems PG Cert (PWP training programme description). https://www.ucl.ac.uk/prospective-students/graduate/taught-degrees/low-intensity-cognitive-behavioural-interventions-common-mental-health-problems-pg

If you need help now

If your life is in danger, call your local emergency number: 112 in most of Europe, 911 in the US and Canada, 999 in the UK, 000 in Australia. For support lines by country, see Help now.

Image credits

This page uses no third-party images.