Introduction
Tinnitus can be defined as the perception of sound in the absence of external auditory stimulation. This condition affects 10–15 per cent of the UK population, its incidence is increasing because of increased headphone usage and increased awareness;Reference Hoare, Kowalkowski, Kang and Hall1 there is no current cure. Characteristics of the perceived sound(s) can vary. For example, one person may experience whistling or humming sounds, whereas another may hear more complex sounds such as music. The sound can be intermittent or constant, unilateral or bilateral, and associated or not with hearing loss.Reference Phillips and McFerran2 Tinnitus is typically subjective (i.e. heard only by the affected individual) and rarely objective (i.e. heard by an examiner either with or without an assistive listening device; 1 per cent of presentations).Reference Hoare, Kowalkowski, Kang and Hall1 Objective tinnitus is typically caused by an internal stimulus, such as a vascular abnormality. The pathophysiology of tinnitus is poorly understood, but the neurophysiological model of tinnitus, as postulated by Jastreboff and Jastreboff in 2003, states that multiple systems in the brain are involved, with the auditory system playing a secondary role. This model, based on neurophysiology and behavioural neuroscience, is fundamental to tinnitus retraining therapy.Reference Jastreboff and Jastreboff3
Tinnitus retraining therapy
Tinnitus retraining therapy is a habituation technique for reducing the impact of tinnitus on an individual's life. It involves manipulating the limbic, autonomic and auditory systems so as to reduce the response to the abnormal stimulus. The goal is to decrease the sensations (via the central auditory system), emotions (via the limbic system) and behaviour (via the autonomic system) associated with tinnitus.Reference Jastreboff and Jastreboff3–Reference Henry, Jastreboff, Jastreboff, Schechter and Fausti5 Three steps are involved: taking a full patient history including daily activities; using assistive devices with wide-band noise to divert attention away from tinnitus; and psychological therapy via counselling. The benefits of assistive devices depend on factors including the severity of tinnitus (0–4 scale), the use of hearing aids, and the presence or absence of hyperacusis.Reference Henry, Schechter, Zaugg, Griest, Jastreboff and Vernon4 These factors are described in more detail in Table I.
Table I Protocol for tinnitus treatment*
* Taken from Jastreboff and Jastreboff.Reference Jastreboff and Jastreboff3 TRT = tinnitus retraining therapy
The counselling component of tinnitus retraining therapy is highly directive and structured, and is accompanied by protocols such as a specific follow up and initial review.Reference Henry, Schecter, Nagler and Fausti6 It is usually given 4–6 times in an 18-month period. This differs from other counselling approaches. For example, cognitive behavioural therapy (CBT) uses a short-term (8–10 weekly sessions) collaborative approach in which the patient and clinician identify and challenge negative thoughts.Reference Henry, Schechter, Zaugg, Griest, Jastreboff and Vernon4, Reference Kröner-Herwig, Biesinger, Gerhards, Goebel, Verena Greimel and Hiller7, Reference Wilson, Henry, Andersson, Hallam and Lindberg8
Cognitive behavioural therapy
Cognitive behavioural therapy is one of the most extensively researched areas of psychology.Reference Andersson and Lyttkens9 It was originally used to treat depression; however, it is now used for other disorders including anxiety, panic attacks and tinnitus.Reference Andersson, Porsaeus, Wiklund, Kaldo and Larsen10–Reference Kröner-Herwig, Frenzel, Fritsche, Schilkowsky and Esser14 The concept of CBT involves changing the patient's attitude to the disease that affects them so as to reduce symptom severity. The fundamental aim of CBT is to reduce the behaviours associated with tinnitus. This involves the therapist considering the patient's fears of tinnitus as a hypothesis which needs to be tested. This is done by first understanding how the thoughts associated with tinnitus are linked, and then modifying these thoughts and behaviours.Reference Rief, Weise, Kley and Martin15–Reference Andersson, Stromgren, Strom and Lyttkens19 Techniques used to achieve this objective include discussing evidence for and against the patient's beliefs, intentional manipulation to change the patient's thoughts, and providing relaxation techniques to use when the tinnitus is at its worst. In counselling sessions, these techniques are used by a trained psychotherapist or clinical audiological scientist.
There are many other types of treatment for subjective tinnitus, with varying reliability. However, there is a lack of evidence to support the efficacy of one type over another.Reference Caffier, Haupt, Scherer and Mazurek20–Reference Schmitt, Kröner-Herwig and Patuzzi22 No treatment is used in clinical practice unless the patient's tinnitus severely impacts their quality of life, for example, by causing secondary depression and anxiety. Treatments not based on psychotherapy include medication, nutrient supplements, surgery, electrical stimulation and providing external sounds. Their efficacy and side-effect profiles vary, and in some instances are unknown. Thus, it is of great clinical importance to review the two main treatments for tinnitus: tinnitus retraining therapy and CBT.
Methods
A systematic literature review was undertaken to evaluate the efficacy of tinnitus retraining therapy and CBT in the treatment of tinnitus. A thorough methodology was employed to identify high-quality clinical evidence regarding both types of therapy. The methodology used was consistent with the systematic review recommendations of the Cochrane Back Review Group, and similar to those of the Agency for Health Care Policy and Research (‘AHCPR’) 1994 Back Guidelines and the National Institute of Occupational Safety and Health.Reference Higgins and Green23
Literature search
A search of two relevant databases, Medline (using Ovid) and the Web of Science, up to and including February 2013 was performed using the search terms ‘tinnitus’ AND ‘tinnitus retraining therapy’ and ‘tinnitus’ AND ‘cognitive behaviour therapy’. These databases were selected because they include journals that publish on subjects ranging from biomedical science to general science. The search was restricted to papers published in the English language. The first search term retrieved 2481 studies and the second retrieved 88 studies.
Studies were included if they involved human participants with tinnitus presenting as their main symptom; tinnitus was diagnosed; where tinnitus retraining therapy was used, protocols were implemented as outlined by Jastreboff and Jastreboff (Table I); had a randomised, controlled design; and participants were aged 16 years or older. Studies were excluded if they had fewer than 10 participants; patients presented with pulsatile tinnitus and delusional auditory hallucination; they were animal studies; or there was a high rate of participant drop out (> 25 per cent).
After discarding studies with non-human participants, a total of 14 studies remained (4 for tinnitus retraining therapy and 10 for CBT). These studies were critically appraised using evidence-based guidelines for the categorisation of medical studies. No additional articles were identified in the bibliographies of reports selected in the initial review. Other studies unrelated to the treatment of tinnitus were removed before appraisal by a format similar to the Agency for Health Care Policy and Research Back Guidelines. This left nine potential papers for critical appraisal.
Critical appraisal
The validity of systematic reviews can be defined as the degree to which their design and conduct are likely to prevent systematic errors of bias.Reference Higgins and Green23, Reference Moher and Olkin24 The four sources of systematic errors that might arise when evaluating the efficacy of tinnitus retraining therapy and CBT are described in detail in Table II. A simplistic approach was used to assess the validity of studies included in the review, based on criteria suggested in the Cochrane Handbook.Reference Higgins and Green23 Each paper was assessed and rated according to the level of bias risk, as summarised in Table III; any study classified as having a high bias risk was not reported in the literature review because its conclusions would be significantly weakened by the methodology used. All studies were scored according to the recommendations of the Cochrane Back Review Group. Scoring criteria are shown in Table IV. Inclusion required a score of 6 to ensure consistency with the Cochrane Back Review Group. All studies scored 6 or above. The final decision about inclusion for all study types was based on the overall quality of the literature retrieved; consequently, a flexible approach was used.
Table II Hierarchy of evidence on the effectiveness of healthcare interventions*
* Based on criteria developed by the Centre for Reviews and Dissemination.
Table III Criteria for assessing the validity of studies included in a systematic review
Table IV Criteria for evidence-scoring papers
TRT = tinnitus retraining therapy; CBT = cognitive behavioural therapy
Data analysis
Assessment of the data retrieved from all included studies showed that quantitative analysis would be inappropriate owing to the significant degree of heterogeneity. Therefore, data was subjected to a qualitative analysis.
Results
Study selection
Of the 14 studies identified in the literature search, 9 met the evidence-scoring criteria for inclusion into the review (Table V). Studies that involved animals, did not follow the strict protocols of tinnitus retraining therapy or lacked acceptable outcome measures were excluded. The nine selected studies were scored for validity and strength of evidence. All scores were consistent with those of the Cochrane Back Review Group.
Table V Validity and strength of evidence scores*
* A higher score indicates stronger evidence. CBT = cognitive behavioural therapy; TRT = tinnitus retraining therapy
Excluded studies
Both excluded CBT studies had a high drop-out rate (Table VI). All three excluded tinnitus retraining therapy study exclusions failed to adhere to the strict protocols listed by Jastreboff and Jastreboff.Reference Jastreboff and Jastreboff3 In fact, they used a modified version of the original tinnitus retraining therapy methodology described in Table I.
Table VI Studies excluded upon further analysis
CBT = Cognitive behavioural therapy; TRT = tinnitus retraining therapy
Assessment of bias
All selected studies were randomised controlled trials, and no significant risk of bias was identified. However, the risk of bias was unclear in three studies.Reference Andersson, Porsaeus, Wiklund, Kaldo and Larsen10, Reference Henry and Wilson11, Reference Kröner-Herwig, Frenzel, Fritsche, Schilkowsky and Esser14
Adverse effects
No adverse effects related to treatments were reported in any of the selected studies.
Sensitivity of analysis
One study included a biofeedback technique combined with CBT.Reference Weise, Heinecke and Rief16 Therefore, a careful analysis of this study's results was made.
Other factors
Factors that may affect the validity of a study include the gender, age, socioeconomic status and cultural diversity of participants. In one study, 95 per cent of participants were male and all had a military background.Reference Henry, Schechter, Zaugg, Griest, Jastreboff and Vernon4 Although it is possible that these factors have no effect on treatment success, different cultural groups and genders are likely to have different ideologies which can skew results, especially for CBT.
Characteristics of selected papers
The selected studies used heterogeneous units; therefore, quantitative analysis using common statistical programs such as the Statistical Package for the Social Sciences (‘SPSS’) or Statistical Analysis System (‘SAS’) would be inappropriate. Outcome measures for each of the studies are shown in Table VII.
Table VII Outcome measures*
* All studies used at least one form of standardised, validated questionnaire.
Efficacy of tinnitus therapy
Studies using CBT and tinnitus retraining therapy show that these therapies are beneficial for treating tinnitus. The results varied because of the different validated questionnaires used. Each study into CBT compared this therapy with a control group and used at least one validated questionnaire. In contrast, the tinnitus retraining therapy study used three validated questionnaires and compared this therapy with tinnitus masking therapy. Both tinnitus masking therapy and tinnitus retraining therapy were beneficial. Assessment of tinnitus masking therapy using the validated Tinnitus Handicap Questionnaire showed a significant improvement in tinnitus severity within three months, but no significant improvement after six months. Tinnitus retraining therapy was slower to improve symptoms than tinnitus masking therapy, but had a significantly better effect after six months. The Tinnitus Handicap Questionnaire showed that tinnitus was improved by 1118 points after tinnitus retraining therapy and by 300 points after tinnitus masking therapy. Only one study into CBT used the Tinnitus Handicap Questionnaire.Reference Henry and Wilson11 It showed a significant reduction in the level of tinnitus-associated stress but not in subjective tinnitus loudness or depression; however, this was not shown quantitatively. Quality of life scores showed a significant improvement in tinnitus in all eight CBT studies. However, subjective tinnitus loudness was not improved by CBT and was not addressed in the tinnitus retraining therapy study. Improvements in depressive symptoms and general mood were evaluated in six of the CBT studies but not in the tinnitus retraining therapy study. There was significant reduction in depression scores for CBT intervention groups relative to controls; however, there was no significant effect in two studies with low baseline depression scores.Reference Kröner-Herwig, Frenzel, Fritsche, Schilkowsky and Esser14, Reference Weise, Heinecke and Rief16 One study showed an improved quality of life score after 6 months of CBT, but this decreased at 12 months.Reference Kröner-Herwig, Frenzel, Fritsche, Schilkowsky and Esser14
Discussion
This systematic literature review aimed to determine whether tinnitus retraining therapy and CBT are effective in the management of tinnitus, and which treatment technique can produce the best outcomes. Since this review specifically assessed subjective tinnitus, the outcome measures were subjective; tinnitus loudness and quality of life were the primary outcome measures. Questionnaires were used to assess primary outcomes in all studies. These assessed the loudness of tinnitus perception and how tinnitus affects the patients' quality of life. Questionnaires on depression and anxiety were used only in CBT studies.
Cognitive behavioural therapy
This review found that CBT is ineffective in reducing the loudness of subjective tinnitus. A common explanation is that the treatment is unlikely to directly or indirectly improve the pathophysiology of subjective tinnitus. The model for CBT is that it modifies the patient's response to tinnitus, for example, by reducing the behaviour associated with sensing tinnitus, and not the tinnitus itself. However, CBT was reported to be beneficial when used as part of a holistic approach. Although, CBT did not reduce the loudness of tinnitus, it improved the patient's behavioural response to the sounds. Thus, patient scores showed mood and quality of life improvements. The study by Weise et al. contained the most participants (n = 111) and showed the greatest improvement on quality of life and depression via a global tinnitus annoyance questionnaire.Reference Weise, Heinecke and Rief16
Tinnitus retraining therapy
Only one study used the original, standardised approach for tinnitus retraining therapy. It comprised a large number of participants (n = 123), with very few dropping out (n = 5). This study showed that tinnitus retraining therapy of varying duration is beneficial for treating tinnitus of varying severity, and is more beneficial for treating severe tinnitus than mild-to-moderate tinnitus. Furthermore, 12 months of tinnitus retraining therapy was more effective than 3 and 6 months of therapy. In this study, tinnitus masking therapy was more effective than tinnitus retraining therapy in the short term (3–6 months), but this was reversed at 12 months and significantly so at 18 months. Thus, tinnitus retraining therapy should be used as a long-term therapy.
Tinnitus retraining therapy vs cognitive behavioural therapy
It was difficult to compare these interventions because different standardised questionnaires had been used in different studies. However, it is clear that both types of intervention have important roles in managing subjective tinnitus. It is currently impossible to determine which is the more effective treatment because of the distinct lack of research into tinnitus retraining therapy and the absence of standard units for quantitative analysis.
Conclusion
This systematic literature review shows that both CBT and tinnitus retraining therapy can reduce the negative impact of subjective tinnitus by improving patients' quality of life. However, the loudness of subjective tinnitus is not improved by CBT and has not been assessed for tinnitus retraining therapy. Furthermore, studies using the original tinnitus retraining therapy model and homogeneous outcome measures are lacking. Therefore, there is a need for more randomised controlled trials of tinnitus retraining therapy to enable a quantitative analysis of CBT vs tinnitus retraining therapy to be made.
Implications for research
For tinnitus retraining therapy, there is a significant lack of research that adheres to the protocols used by its founder. It is not advisable to include modified tinnitus retraining therapy methods in literature reviews because the validity of results will be compromised relative to those of tinnitus retraining therapy. Furthermore, the number of validated questionnaires should be restricted to increase homogeneity in outcome measures. Longer follow up would also be beneficial because variations in the effects of therapy on tinnitus can continue for up to 18 months.
Implications for clinical practice
The loudness of subjective tinnitus is not improved by CBT and has not been assessed following tinnitus retraining therapy. However, both CBT and tinnitus retraining therapy significantly reduce the negative impact of subjective tinnitus on patients' quality of life. Tinnitus retraining therapy and CBT cannot currently be effectively compared, with good validity; however; tinnitus retraining therapy is significantly more effective in treating tinnitus than tinnitus masking therapy.
