This chapter reviews some of the physical approaches that supplement the cognitive and behavioural repertoire: specifically, relaxation, controlled breathing, exercise and applied tension. The management of sleep problems is also considered.
Physiological responsiveness is one of the four interacting systems central to the CBT model, along with cognition, emotion, and behaviour. CBT may therefore include interventions focused on physiological symptoms, insofar as they are part of a maintenance cycle in the client’s formulation. Of course, physiological problems can equally well be tackled using cognitive, behavioural or physical methods (for example, tinnitus can be eased by changing catastrophic thinking), and by the same token, cognitive, behavioural or emotional problems can be tackled using physical methods. The important point is that the intervention, in whatever modality, should be derived from the formulation. We will look at a number of possible physical interventions, beginning with relaxation.
Physical tension can be part of the maintenance cycle for many problems – anxiety disorders, depression and sleep problems, among others. It may be part of a more general increase in arousal, which includes other physical symptoms, such as increased heart rate, light-headedness, heavy legs and trembling. Such elevated arousal can be reduced by using relaxation, either via specific relaxation exercises or by building pleasurable, relaxing activities into the day – for example, by having a soothing bath or a massage.
The benefits of relaxation should not be discounted, but it should be clear to both your client and yourself how the relaxing activity fits into the formulation. We emphasise this point because in some treatment approaches relaxation training was a line of first attack as part of an anxiety-management programme, with little attention paid to the interactions of tension with thoughts and behaviour. Nevertheless, relaxation has relevance within CBT and can be a powerful tool in testing out beliefs, as well as reducing symptoms directly.
A young man was experiencing anxiety symptoms across a range of situations and was finding some imminent exams highly stressful. Figure 10.1 shows the initial formulation of his problem, relating his beliefs about his work to his emotions, physical symptoms, his worrying thoughts and his behaviour. It is clear that if he could break into the cycle by reducing his physical symptoms, then he may be able to improve his concentration and work more productively, which in turn would allow him to feel less anxious.
Figure 10.1 A maintenance cycle for anxiety about exams
There are a number of approaches to teaching relaxation, most of which rely either on the progressive muscular relaxation described by Jacobson (1970) or the use of relaxing imagery or meditation (or all three). There is some evidence that matching relaxation method to the individual’s pattern of symptoms (for example, applied relaxation with clients with predominantly physiological symptoms) may improve outcome, but the evidence is by no means unequivocal (Michelson, 1986), and it is probably better to find out which approach suits your client by trying it out. There are many forms of recorded relaxation instruction available, and we would suggest that you find one you like for your own clients. Rather than describe a specific method, here are some general guidelines for acquiring relaxation skills:
Although a recording has been mentioned as a useful adjunct to learning to relax, the evidence suggests that it is better to go through the relaxation exercises with your client, rather than simply ask them to learn from a recording (Borkovec & Sides, 1979). This is partly because you can then observe the client as he begins to practise and pick up on errors from the start – for example, if he is sitting with tightly crossed legs, or is tightening up his arms and shoulders once another area is focused on. It is also helpful if he can ask questions or express reservations. For example, many clients are concerned that they cannot focus continually on their muscle groups but find their minds wandering; you can encourage them to simply note whatever thoughts come into their mind and to gently direct themselves to return to thinking about relaxation.
It is probably most useful to spend 10 or 15 minutes per session focusing on relaxation training, over perhaps five or six sessions if an approach like applied relaxation (Öst, 1987) is being used as a major component of treatment. The remainder of these sessions can then be used for other agenda items. In other cases, relaxation may play a more minor role, and you may only need to go over the relaxation procedure a couple of times. The effectiveness of the relaxation can be checked out by asking the client to monitor how relaxed he feels following his regular practice sessions and how he feels in whatever situations he applies relaxation. Figure 10.2 shows a possible diary for recording daily relaxation practice, and its effect.
Figure 10.2 Diary of relaxation practice, and its effect
Applications of relaxation in CBT
Where the client is too scared to carry out a behavioural experiment
Behavioural experiments often demand a great deal of courage, particularly if the prediction being tested includes elements such as, ‘If I do such-and-such, I will probably feel anxious, but I will not collapse/suffocate/jump over the edge …’ (or whatever other catastrophes the client fears). Relaxation can then be used as an aid to facing the feared situation (Rachman, Radomsky & Shafran, 2008) although only as a temporary solution because of the risk of relaxation becoming established as a safety behaviour.
A client with a height phobia was planning to test out his prediction that he would be likely to jump off the cliff if he were to go onto a cliff path. He thought that he would be too anxious to do it ‘cold turkey’, so planned to use relaxation in the first instance as a way of getting himself onto the cliff path.
To test out a belief about whether symptoms have an organic basis or are anxiety-related
If a client’s unhelpful beliefs are focused on the aetiology of his symptoms, it may be possible to test out competing hypotheses by using relaxation.
A woman was fearful that her severe headaches were a symptom of a brain tumour. She practised relaxation on a daily basis and found that as she became more skilled at progressive relaxation, and gradually applied it in more stressful situations, the intensity and frequency of her headaches decreased. She recognised that this was more consistent with an anxiety explanation rather than a tumour explanation.
To interrupt vicious cycles where increased arousal interferes with performance
There are a number of problems where physical anxiety symptoms have a direct effect on performance of a task or function (see Chapter 4, Figure 4.8), and applying relaxation techniques may therefore be useful. For example, anxiety whilst speaking in public, and erectile problems, can both be affected in this way. Similarly, people can have difficulty eating if anxiety is interfering with their swallowing response. In each of these areas, relaxation may make a contribution insofar as it reduces arousal and allows the task to be performed without interference.
To give a break from tension/arousal
An obvious use of relaxation is where increased arousal is experienced as an unpleasant symptom in its own right. This can be the case with those who are chronically anxious and find the physical symptoms themselves unpleasant. In such cases, it is important to check whether the symptoms have an idiosyncratic meaning for your client. For example, does your client believe that the presence of chronic tension means that he is harming his immune system, or is it a sign that he is constitutionally deficient and ought not to have children, or that he can never hope to change? If the meaning of the symptoms seems to be distorted, then this should be tackled as described in Chapter 9. However, it may be that for your client the unpleasantness of the symptoms itself is stoking the problem, and that having a way to reduce this may increase a sense of mastery, which in turn boosts self-esteem, and onwards in a positive cycle. As ever, make sure that you and your client can make sense of what part the physical symptoms are playing in the problem: get out your diagram of the formulation, or draw it on your white board, clarify how the physical symptoms fit in, and hence, what role relaxation would have.
To end a stressful treatment session
If your client has had a difficult and stressful session, as for example, in restructuring traumatic images, then going through relaxation may help him to re-enter the everyday world, before he leaves the session. This can also enhance his confidence that he can experience, tolerate and manage strong emotions.
To provide an opportunity for pleasure
Your client may have insufficient opportunities for pleasant and rewarding activities. Many people thoroughly enjoy muscular or other relaxation, and fitting it into a busy schedule, doing something for himself, may result in improved mood for your client and increased energy for other activities.
To improve sleep
Relaxation can be a useful part of a programme to improve sleep hygiene, particularly if the client is in the habit of being active right up until bedtime (see the section below on sleep).
Common problems associated with using relaxation in CBT
The most common problem is that relaxation, as with many other strategies described as ‘coping techniques’, becomes established as a safety behaviour. This is only an issue where the client becomes trapped by the belief that, for example, ‘If I had not relaxed, I would have panicked and then I would have lost control/passed out/gone mad, etc.’. In essence, the client is left with the fear that he could have been overwhelmed by the problem if the relaxation had not allowed him to scrape through. The implication of this is that ultimately he has to face the problem without using relaxation, in order to demonstrate that even in those circumstances, while he might feel bad, there is no catastrophic outcome. It is worth bearing in mind that there is some evidence that the judicious use of safety behaviours may facilitate exposure to phobic objects or situations during treatment, and that this may not necessarily result in poorer outcome (Rachman et al., 2008).
Inability to relax if highly aroused
Past a certain level of arousal, it is very difficult to counter high arousal with relaxation. This can be an issue if the client is panicky, or is highly aroused for some other reason – as in PTSD, for example. In that case, it is more helpful to use a different strategy, for example a mindfulness approach (see Chapter 17) in which the symptoms are viewed with disinterest.
Relaxation experienced as losing control
Some clients experience relaxation as anxiety-provoking rather than anxiety-reducing, often because it feels as though they are losing control. If this is the case, it may be worth exploring the meaning of losing control and then use relaxation to test your client’s predictions about what might happen if he did let go.
Debbie had strong beliefs about the value of being in control, and this was associated with many intrusive thoughts about everyday financial issues that were not in her immediate control. She always tried to make sure that she stuck within the rules, and did not like to give herself time to ‘play’ in case it went too far. She feared that if she did not keep herself under control, she might ‘lose it’ and become irresponsible. She agreed to experiment with giving herself time to practice relaxation, initially to find out whether she could choose how much control she retained.
Hypersensitivity to small bodily changes
When they first begin to practice relaxation, many clients pick up on small bodily changes of which they were previously unaware. This may occasionally create or increase an attentional bias towards bodily changes, which can be interpreted as indicating a risk to health of some kind (see Chapter 4). If this happens, it should be explored and tested in the same way as other distorted thinking, and hence may provide a useful opportunity to practise evaluating a negative thought.
Charles became aware of tingling in his fingers when he was practicing his relaxation, and was fearful that this may be an indication of the onset of a stroke. He investigated this with his therapist by initially doing some highly distracting arithmetic, counting backwards in 7s, and looking at the effect on the tingling; then focusing on the pressure in his elbow leaning against his chair, and on what sensations he experienced (not tingling, but certainly sensations he had not been aware of until he paid attention to them). This was followed by a discussion about the effect of attention on the perception of benign physiological ‘symptoms’, and how that could apply to his tingling fingers.
Despite these possible drawbacks, there are many imaginative ways that relaxation can be used within CBT to disrupt maintenance cycles causing distress to clients.
The catastrophic misinterpretation of benign physical symptoms is the central process of one well-established model of panic (Clark, 1986). One of the benign symptoms often implicated is hyperventilation, i.e. breathing at a high rate and volume. This can result in symptoms that mimic the person’s panic attacks (for example, shortness of breath, light-headedness, feeling hot, unsteadiness) and which are liable to be interpreted catastrophically as indicating imminent death, collapse, madness and so on. Salkovskis, Jones and Clark (1986) developed a strategy of controlled breathing to allow the client to reattribute their symptoms to a more benign cause (i.e. that it is a symptom of anxiety) and, hence break into the vicious circle of misinterpretation that was maintaining the panic attacks. This strategy can be used in order to develop a shared formulation with the client and also as one technique in a graded approach to dealing with panic attacks.
Applications of controlled breathing in CBT
Developing a shared formulation using controlled breathing
The following steps can be used in developing a formulation with the client about the role of over-breathing in panic attacks:
Managing panic symptoms through controlled breathing
Having derived a formulation involving the role of breathing, the client can then be taught to manage the symptoms. Initially, the therapist teaches controlled breathing, first by following a recording at a moderate breathing rate, and subsequently at a slower rate. The therapist then:
This approach is described in more detail in Clark (1989).
Using controlled breathing in behavioural experiments when the client is too scared to continue
As with relaxation, controlled breathing can be used as a short-term coping strategy to allow a client to carry out a behavioural experiment that he would otherwise be too scared to do. He could then move on in a graded way to do the experiment without controlled breathing.
Problems associated with using controlled breathing in CBT
As with relaxation, it is very important that the client does not use the controlled breathing as a safety behaviour but instead employs it only as a short-term strategy. If he always uses controlled breathing when he feels panicky, he may continue to believe that ‘If I had not done my breathing, I would have collapsed/gone mad … etc.’. Ultimately he has to face the panic symptoms without using controlled breathing, so that he can learn that while he might feel uncomfortable, the feelings are tolerable, and there is no catastrophic outcome.
Using controlled breathing when feeling panicky
Some clients are unable to use controlled breathing when feeling very panicky but may learn to do so if they have more practice when they are not feeling panicky. It is important that there is a shared understanding with the therapist of why it may be helpful.
Presence of physical disorder
Hyperventilation is not recommended in a number of physical conditions, unless medical supervision is available. These include atrial fibrillation, asthma and chronic obstructive pulmonary disease, epilepsy and pregnancy.
Hypersensitivity to small changes in breathing
The client may develop a heightened awareness of minor changes in his breathing, and so care must be taken to interpret these benignly and not as signs of dys-function or precursors of panic (see, for example, the catastrophic misinterpretation shown in Figure 4.5, and the management of panic disorder described in Chapter 14).
Too tense to breathe evenly
It is easier to start the cycle if you suggest that the client focuses initially on the out-breath, as the lungs are relatively empty and the body pushes for an in-breath.
Extensive research over the past 20 years has established the efficacy of exercise in the treatment of depression (Craft & Landers, 1998), and the NICE guideline on the treatment of depression (NICE, 2004a) recommends that all patients with mild depression should be advised of the benefits of a structured exercise programme. The effect of exercise on depression may be mediated by an increase in endorphins, but it may be associated with other effects of increased exercise, many of which may also be important for clients with anxiety problems (Taylor, 2000). For example, exercise can provide distraction, or can increase self-esteem, as when engaging in a competitive sport. The question is whether intervening with exercise would disrupt a maintenance cycle for your client, and it is worth looking out for symptoms that could be addressed in this way, particularly as exercise can become self-maintaining once the basic skills are learned.
Applications of exercise in CBT
The best-established application of exercise is with depression, where apart from the direct effect of increased endorphins, exercise may also provide opportunities for pleasurable and satisfying activity, promoting improved mood. It is often helpful to begin in a graded way, as depressed clients may feel tired much of the time, and may be doubtful that they have the energy for exercise – and may have given it up for that reason. This can be tested out in an experiment, and as it may be early in treatment, this can serve as an example of the empirical nature of CBT.
The sense of competence derived from exercise may be relevant for someone with low self-esteem (Fox, 2000).
Chronic fatigue syndrome (CFS)
Chronic fatigue syndrome is a condition defined by persistent fatigue unrelated to exercise, not relieved by rest, and accompanied by other symptoms such as headache, and muscular and joint pains. Exercise can be central to a graded programme where the client can test out predictions about fatigue (Silver, Surawy & Sanders, 2004).
For clients with chronic anxiety, or in chronically stressful situations, it can be helpful to test out the benefits of exercise on tension levels. This can be especially helpful with younger clients who may not take to relaxation.
There is good evidence of the effects of exercise on sleep, as long as it is regular and is not used close to bedtime, when it tends to be arousing (see below).
Health anxiety or panic disorder
Many clients with health anxiety or panic disorder have beliefs about the risks exercise poses to their health. For example, one man believed that if he exercised, his heart rate would increase, and this would increase his risk of a heart attack. It can be very important to invalidate these beliefs via experiments focused on exercise.
It can be helpful for clients with anger problems to test out the effects of exercise on tension levels, particularly if this is followed by a soothing activity such as a relaxing bath.
Problems associated with using exercise in CBT
In some disorders, such as eating disorders and body dysmorphic disorder, exercise is overvalued because of its perceived effect on body shape and weight control. In these cases, you should be circumspect about using it to tackle associated problems such as tension or low self-esteem.
Presence of physical disorder
Exercise is not recommended in a number of physical conditions, such as cardiovascular problems. You should ask your client whether he is aware of any condition that would be a contra-indication for exercise, and seek medical advice if necessary.
Although many anxious clients feel as though they are about to pass out, some people, particularly those with phobic anxiety about blood or injury, frequently actually faint in response to their anxiety (Öst, Sterner & Fellenius, 1989): an initial increase in blood pressure (typical of anxiety) is followed by a sudden decrease, leading to fainting (Öst, Sterner & Lindhal, 1984). In applied tension, the client is taught to increase his blood pressure by tightening the muscles in his arms, legs and torso for a few seconds and then to return the muscles to normal. He is then taught to identify the signs of a drop in blood pressure (provoked, for example, by exposing him to photographs of blood or other injuries) and to reverse the decrease by using applied tension (Öst & Sterner, 1987).
Problems associated with using applied tension in CBT
Once again, the major risk of using applied tension is that it can function as a safety behaviour. It is important to help the client to view applied tension as a helpful thing to do when his blood pressure drops, just as it is helpful to look both ways before crossing the road – i.e. based on a reasonable caution about the consequences of not doing so.
Presence of physical disorders
Before using applied tension, therapists should seek medical advice about any client who is pregnant or has a known physical disorder, particularly hypertension or a cardiovascular condition.
CBT and sleep
We will now turn to problems with sleep, implicated in many mental-health problems as well as being common in the general population.
Insomnia is a problem experienced by 10% of adults, and 20% of those over 65 years of age at any one time (Espie, 2010) and can include delayed onset of sleep; difficulty staying asleep, with multiple awakenings; and waking too early. It can be secondary to a range of physical and psychiatric conditions, although most evaluations of psychological interventions have focused on insomnia as the primary condition.
Many of the early CBT treatments focused on relaxation as a way of reducing physical arousal levels, even though client reports about insomnia often emphasise mental arousal – for example, ‘I lie calmly in bed, but my thoughts are racing’, or ‘All the worries of the day come into my mind’. Accordingly, there has been an increasing emphasis on cognitive approaches to sleep problems (Harvey, 2002), as well as attention to other physiological and behavioural aspects. We shall therefore look at the processes which are currently thought to be involved in poor sleeping, taking as an example Cara, who was highly successful in business as well as being committed to her teenage children. She was chronically unable to fall asleep, and also woke repeatedly in the night and so she had relatively few hours sleep.
Processes implicated in poor sleeping
Unhelpful automatic thoughts and beliefs in bed or by day
Cara believed that if she did not have six hours sleep then she would be unable to think productively or relate effectively to her family or colleagues; that she ought to be able to control her sleep, as she controlled other aspects of her life; and that any tiredness she experienced in the daytime was attributable to her insomnia (rather than a result of a busy day with no planned breaks and no time for relaxation).
Safety behaviours, including monitoring internally and externally
When in bed, Cara repeatedly checked the clock; she wore ear-plugs and used a special cushion which she believed increased her control over falling asleep; she monitored her body for signs of wakefulness.
In the daytime, she tried to avoid complex work following a bad night; she monitored her body for signs of tiredness and for signs of poor concentration.
Poor stimulus and temporal control of sleep behaviour
When she went to bed, Cara took a book and an iPod in case she could not sleep; she spent long periods of time wakeful in bed, reading and listening to music (poor stimulus control).
If she had a disturbed night, she would lie in, if possible, and go to bed early the next night, often when not sleepy (poor temporal control).
Increased mental arousal and possibly physical arousal
As she lay in bed, Cara had many worrying thoughts and tried to work out solutions to them (mental arousal).
Poor sleep hygiene
In order to tire herself out, Cara went to the gym after her evening meal had settled; she had a glass of whiskey in order to settle herself but then busied herself with household tasks before going to bed.
Distorted perception of sleep
Cara significantly underestimated how long she slept, and judged how well she had slept in terms of how she felt on waking, at a time of ‘sleep inertia’ when very many people feel heavy and tired.
Interventions for sleep problems
As with other problems, interventions are planned on the basis of a detailed assessment and formulation, taking account of whatever maintaining cycles seem most relevant for the particular client. The most common interventions include:
Re-evaluation of unhelpful or distorted thoughts and beliefs
Unhelpful cognitions can be tackled using verbal challenging and behavioural experiments.
Cara identified her thoughts about needing to be in control as particularly distressing, and challenged these verbally by looking at alternative perspective on controlling sleep (was it common for other people to be able to control when they slept? What was the evidence that they could/could not? What would she say to a friend who said that she should be able to control her sleeping? What were the pros and cons of thinking that way?)
Cara also tested the accuracy of the thought about needing six hours sleep by doing a behavioural experiment. For a number of days she recorded how many hours sleep she had, and then did ratings of how tired she felt at work, and rated how productive she had been in the morning and afternoon of each day. Although she was not ‘blind’ to the amount of sleep she had had, she still found that generally she was reasonably productive on most days at work. Although she felt tired on some days, the factor that made most difference to productivity was whether she was doing ‘desk’ work after lunch, when she concentrated poorly, whether she had had a good night or not. By contrast, she was highly effective if she was doing ‘people’ tasks, whether or not she had slept well.
Reduction in safety behaviours
Insofar as safety behaviours often prevent the disconfirmation of unhelpful beliefs about sleep (e.g. ‘I sleep so badly I could never manage without my ear-plugs’), they need to be removed so that beliefs about the importance of sleep can be softened.
Cara began by moving her clock out of sight, so that she could not check the time. She also stopped wearing her ear-plugs. Although it was initially difficult not knowing what the time was, she recognised that it reduced the pressure to get to sleep. Similarly, she found that she was not disturbed if she did not wear her ear-plugs, and felt that this further decreased her sense of being an insomniac.
Improved stimulus and temporal control
This intervention has been extensively evaluated since Bootzin’s (1972) paper. The approach was based on the notion that consistent cues are necessary to allow the client to clearly differentiate environmental and behavioural sleep cues from non-sleep cues, and to allow the body to acquire a consistent sleep rhythm. There are debates about whether this hypothesis is correct, but the approach has been shown to be effective in rapidly reducing sleep-onset latency. It can be difficult for clients to follow, because in the early stages of the procedure, tiredness may increase. Clients may need a great deal of encouragement to continue. A good description of the procedure can be found in Espie (1991), and the main elements include the following:
Decrease mental arousal
Insomnia may be maintained by a failure to ‘tie up loose ends’ before going to bed so that unresolved issues from the day rush into the mind. It may be helpful for your client to set aside time in the evening (though not immediately before going to bed) to experiment with writing down/thinking through issues from the day, including their emotional impact. If this is not sufficient, and there are themes that recur or are more troubling, then it may be helpful to rehearse cognitive re-evaluations of these thoughts (as outlined in Chapter 9) so that the client is prepared in advance.
A woman frequently worried about being an unsatisfactory parent, and the ways in which this would affect her children later in life. She looked at the evidence for her being a bad parent, and at the determinants of her children’s future success. She then summarised this in a phrase that she thought encapsulated the alternative view, namely, ‘I only have to be good enough, no one can be perfect; giving them love is the most important thing.’ She kept the more detailed evidence recorded in her therapy notebook.
Decrease physical arousal
Although the evidence for increased physical arousal in insomnia is unclear, many studies have shown that progressive muscular relaxation has some impact on sleep-onset latency, on total time slept and, importantly, on the quality of sleep perceived by the client. Furthermore, many clients enjoy muscular relaxation. It seems to make no difference which relaxation approach is used.
Poor sleep hygiene
For most clients, general information and advice would include:
Although none of these elements would be sufficient to maintain chronic insomnia, they could each exacerbate sleeping problems.
Problems in sleep management
Strategies that are initially useful for sleep hygiene can develop into safety behaviours. Encourage your client to experiment with using the strategies flexibly, to challenge the belief that they are essential for his well-being.
Undisclosed drug use
Your client may not respond to psychological interventions if he is using drugs that promote wakefulness.
Sleep problem secondary to another problem, or not insomnia
A sleep problem may not respond to psychological interventions if it is secondary to another psychiatric or physical condition. In such cases, the primary disorder should be treated, using a cognitive behavioural approach if appropriate, for example with depression. Neither will it respond if it is a sleep disorder not classified as insomnia (e.g. sleep apnoea, nocturnal myoclonus or restless legs).
o to reduce general arousal
o to provide pleasure
o to facilitate exposure to anxiety-provoking situations.
o relaxation functioning as a safety behaviour
o relaxation being experienced as losing control
o hypersensitivity to small bodily changes.
Review and reflection:
Taking it forward:
Benson, H. (2000). The relaxation response. New York: Avon Books.
This is a revised edition of a 1970s classic guide to systematic relaxation and meditation. It offers ideas for developing a range of different relaxation exercises.
Espie, C. A. (2010). Overcoming insomnia and sleep problems: a self-help guide using cognitive behavioural techniques. London: Constable Robinson.
An easy to read self-help book which is based on practical applications of CBT.