Effects of Cognitive Behavioural Treatments
A Systematic Review and Meta-Analysis
Description of Treatments Evaluated in Randomized Controlled Trials
A large number of more or less different psychological treatments have been subjected to stringent evaluation in randomized controlled trials. The diversity of methods reflects to a large extent the general picture of treatments used for specific phobias in general. Thus, even if dental phobia is one of the few specific phobias where confrontation with the phobic situation may entail pain, researchers in the field have not found it necessary to develop specific treatments for this phobia. However, for dental phobia there may be an important distinction whether or not the treatment involves exposure in vivo. Below follows a description of the most important of these treatment methods.
The oldest form of cognitive behaviour therapy (CBT) for phobias is systematic desensitization (SD), developed by Joseph Wolpe (1958) and it has three components. It starts with a shortened form of progressive relaxation in which the patient learns to relax the gross body muscles. In parallel with the relaxation training the second component means that the therapist and patient construct a hierarchy of anxiety-arousing situations covering all the important aspects of the patient’s phobia. Each situation is given an anticipated Subjective Unit of Disturbance (SUD) value from 0–100 and rank ordered based on this. Finally, in the third component the therapist present the lowest situation from the hierarchy to the relaxed patient who imagines being in that situation. This is done in gradually longer time periods, usually 10, 20 and 30 seconds. If the patient can imagine the situation for 30 seconds without feeling anxious that particular situation is considered to be desensitized. The same procedure is then repeated for all the remaining situations in the patient’s hierarchy.
In dental phobia SD was first used by Shaw and Thoresen (1974). The difference from the standard SD was that the scenes in the hierarchy were presented via audio tapes instead of verbally by the therapist. Gatchel (1980) made further changes in that he used a group format and employed a version of SD called self-control desensitization. A standard hierarchy of 16 situations was used but these situations were presented in general terms and the patients were instructed to fill them out with individual details. Finally, each imagination of a situation lasted for 60 seconds and the patients were trained to use relaxation as a coping skill. Moore and Brødsgaard (1994) presented relaxation instructions via an audio cassette and anxiety scenes employing eight 30-second videotaped dental scenes as developed by Carlsson, Linde and Öhman (1980). All examples have as general features that patients are trained in relaxation and anxiety-arousing situations are presented in a gradual fashion, but not in vivo as is done in exposure.
When a treatment has the phobic patient confronting his or her anxiety-arousing situations in reality, the method is called exposure. It can be done in imagination (in vitro) or in real life (in vivo). The exposure can also be performed gradually, as is usually the case nowadays, or directly with the highest situation in the patient’s hierarchy, which is called flooding. The purpose of exposure is for the patient to learn that the anxiety reactions to the phobic stimuli, e.g. palpitations and subjective fear, are reduced (habituation takes place) as long as the patient remains in the situation instead of escaping from it.
The flooding type of exposure was mainly used in the 1970s–80s and is illustrated in the dental phobia area by Mathews and Rezin (1977), with imaginal flooding and Gauthier et al. (1985) who used auditory stimuli recorded during actual dental procedures. However, flooding has almost completely been replaced by gradual exposure since it is a very hard treatment emotionally, both for patient and therapist, but without leading to better effects than gradual exposure.
Gradual exposure is illustrated by the Moore (1991) and Moore and Brødsgaard (1994) studies, but called clinical rehearsal. However, the same method is described somewhat differently in these two studies. In the first study the author talks about ‘direct exposure to threatening dental situations or dental instruments in gradual, approximating steps’ (p. 231), whereas in the second study it is described as ‘direct, but simulated exposure to dental situations or dental instruments in gradual, approximating steps’ (p. 259). This means that either the authors used two versions of the treatment but called it the same or it was just one version and they realized that they needed to qualify the description with the words ‘but simulated’ in the second study.
The most clear-cut example of exposure in vivo comes from the Haukebø et al. (2008) study. In this study the exposure in vivo treatment starts with a fairly low situation in the individual patient’s anxiety hierarchy and then gradually continues up the ‘ladder’ as soon as the lower situation has been achieved by the patient. The exposure treatment is usually carried out across a number of sessions (e.g. five), but it can also be done in one prolonged so-called one-session treatment (OST), which is described in detail in Chapters 9–10 of this volume. This treatment is a combination of exposure in vivo, modelling and brief cognitive therapy.
Another treatment for dental phobia performed in just one session of 90 minutes was described by Thom, Sartory and Jöhren (2000). This consists of anxiety management training, imaginal exposure to phobic situations and stress-inoculation training.
A multimodal behavioural treatment first described by Berggren and Carlsson (1984) and used in a randomized controlled trial (RCT) by Berggren and Linde (1984) consists of progressive relaxation and desensitization to 30-second videotaped dental treatment scenes in combination with biofeedback training. This treatment is carried out by a psychologist during about eight sessions and followed by two sessions of adjusted conventional dental treatment by a dentist. During this phase the patients are encouraged to apply the techniques they learned during the first phase with the psychologist. In the next RCT from the same research group (Harrison, Berggren and Carlsson 1989) the same treatment was called systematic desensitization. In the third study (Hammarstrand, Berggren and Hakeberg 1995) the method was called psychophysiological therapy and in the fourth (Berggren, Hakeberg and Carlsson 2000) and fifth studies (Lundgren, Carlsson and Berggren 2006) the name used was relaxation therapy. A detailed description of this method is provided in Chapter 8 of this volume.
Getka and Glass (1992) used a six-session behavioural treatment which consisted of a combination of videotaped modelling, relaxation training, videotaped desensitization and self-paced in vivo practice in the dental clinic.
Cognitive behaviour therapy
Cognitive therapy was developed by Beck, Emery and Greenberg (1985) for anxiety disorders in general and by Clark (1986) for panic disorder in particular. With the emergence of cognitive therapy the focus shifted from features of the dental situation to the patients’ cognitive misinterpretations (catastrophic beliefs), helping patients to challenge the evidence for their beliefs and substituting them with realistic interpretations that are based on real evidence. Behavioural experiments are used to test specific beliefs during therapy sessions and often as homework assignments between sessions. This type of treatment was used in an RCT for dental phobia by Willumsen, Vassend and Hoffart (2001) across 10 sessions, but they also used exposure in accordance with standard dental procedures.
A much briefer treatment, just one session of one hour, was used by De Jongh et al. (1995) as a preparation for dental treatment. The treatment was focused on negative cognitions that were evident after the patient had answered the Dental Cognitions Questionnaire. The intervention was focused on cognitions for which patients rated at least 70 per cent believability. This was achieved by asking the patient for the evidence he or she had for the specific thought, challenging this evidence and helping the patient to view the situation in another, realistic way. In those cases where the patient clearly lacked information about the dental procedure in question such information that could help correct the negative cognition was provided.
Another example of this approach is the Haukebø et al. (2008) study, where the patients’ catastrophic beliefs were identified and challenged in a dental treatment situation in which the dental treatment was an integrated part of the fear treatment, delivered by a specially trained dentist. This approach was tested in a five- as well as one-session approach.
There are a number of treatment methods called coping techniques which are characterized by teaching the patient some kind of skill that can be used in the actual anxiety-arousing situation in order to stop the anxiety reactions from escalating and eventually abort them altogether.
Applied relaxation (AR) was developed by Öst in the late 1970s and used in a first RCT on social phobia in 1981 (Öst, Jerremalm and Johansson 1981) and later in an RCT for claustrophobia (Öst Johansson, and Jerremalm 1982) and agoraphobia (Öst, Jerremalm and Jansson 1984). A group version of AR was used in an RCT of dental phobia (Jerremalm, Jansson and Öst 1986) and the individual version in the study by Willumsen et al. (2001). AR can be seen as a course across 8–10 sessions, starting with progressive relaxation, which takes about 15 minutes to instruct. This component instructs patients to briefly tense and then relax the specific muscle groups included. In the second component, release-only relaxation which takes 7–8 minutes, the tension instruction is deleted and patients are just instructed to relax the muscle groups. The third component, cue-controlled or conditioned relaxation, teaches the patient to connect the self-instruction ‘Relax’ with the state of relaxation and takes 2–3 minutes. In the fourth component, differential relaxation, the patient is taught how to be relaxed in the muscle groups that are not used for the activity at hand. In this phase it usually takes the patient 60–90 seconds to become relaxed. The fifth component, rapid relaxation, has been reduced to four simple instructions and at the end of practising during this phase it takes the patient 20–30 seconds to become relaxed in natural, but non-anxiety situations. The last component, application training, allows the patient to apply the skill of relaxing rapidly in anxiety-arousing phobic situations. In these situations the patients will experience the initial anxiety reaction, apply the AR for as long as they need and notice that the reaction is reduced and aborted altogether. In the Jerremalm et al. (1986) study the patients were instructed to apply the skills during dental treatments with an ordinary dentist outside the research project.
Stress inoculation training (SIT) was developed by Meichenbaum (1977) and consists of three components. In the first, which is educational, the therapist presents a conceptualization of anxiety and in the second the patient is taught various coping skills (cognitive, physiological or behavioural) that can be used in counteracting anxiety reactions. The final component allows the patient to practise the acquired coping skills while being exposed to anxiety-arousing stimuli. SIT has been used in RCTs for dental phobia by Moses and Hollandsworth (1985), Getka and Glass (1992), Berggren et al. (2000) and Lundgren et al. (2006) even if the latter two called the treatment cognitive therapy. A version of SIT, called self-instructional training and focusing completely on the cognitive component, was used in the study by Jerremalm et al. (1986). Another version used by Ning and Liddell (1991) was called anxiety management training and consisted of relaxation training and cognitive restructuring in combination with imaginal exposure. However, none of these approaches implies real experience with the feared dental treatment situation during the phobia treatment.
A treatment developed by Bandura and co-workers in the late 1960s early 1970s (e.g. Bandura et al. 1974) is called modelling and can be carried out in various ways, e.g. video presentation or in vivo. In this treatment models that the patients can identify with age- and gender-wise are successfully performing the behaviours that the phobic patients are avoiding. Often, this is combined with exposure in vivo where the patients in the second phase perform these behaviours, first with the help of the therapist and then on his or her own. In dental phobia modelling has been used in RCTs by Shaw and Thoresen (1974) and Wroblewski, Jacob and Rehm (1977).
Various treatment methods have only been used in one RCT each. Progressive relaxation and electromyographic biofeedback were employed by Miller, Murphy and Miller (1978) and hypnotherapy by Hammarstrand et al. (1995).
Was exposure in vivo used?
For specific phobias in general exposure in vivo is the treatment with the strongest evidence-base (Wolitzky et al. 2008). Thus, it is interesting to look at the use of this method in RCTs for dental phobia. The clearest example of exposure in vivo is the Haukebø et al. (2008) study in which both the one- and the five-session treatment were completely carried out in vivo, in a framework of dental treatment, by a dentist trained to do the phobia treatment. It was supplemented with cognitive procedures helping the patient to challenge cognitive distortions (catastrophic beliefs). In the Gothenburg studies (Berggren and Linde 1984; Berggren et al. 2000; Hammarstrand et al. 1995; Harrison et al. 1989; Lundgren et al. 2006) the in vivo exposure was restricted to two sessions with a fear clinic dentist, following the initial eight sessions with a psychologist. Berggren and Linde (1984) described the content of these sessions as ‘adjusted conventional dental treatment using perceived control and reattribution of negative stimuli’ (p. 1223). Getka and Glass (1992) used two (out of six) sessions for ‘self-paced in vivo practice’ and, as described above, concerning the Moore (1991; Moore and Brødsgaard 1994) studies it is difficult to ascertain whether the exposure was real or simulated. All the other studies that used exposure employed some kind of imaginal or in vitro. However, the majority of RCTs in dental phobia have used treatment without exposure in vivo or in vitro.
Criteria for inclusion of studies
In order for a study to be included in the systematic review and meta-analysis the participants had to be randomly assigned to one treatment group and one control group, or at least two active treatment conditions. Thus, one-group studies with pre- and post-treatment assessment or quasi-experimental designs were not included. Table 12.1 presents an overview of the included studies. There is a total of 22 RCTs published between 1974 and 2008, making 0.63 RCTs per year, which is not a particularly impressive research ou/>