رفتاردرمانی شناختی برای حساسیت شنوایی: کارآزمایی تصادفی کنترل شده

Hyperacusis, defined as unusual intolerance to ordinary environmental sounds, is a common problem for which there are no controlled trials on psychological treatment. Given the avoidance strategies present in hyperacusis, and similarities with problems such as tinnitus and chronic pain, cognitive behaviour therapy (CBT) is hypothesized to be helpful for patients with hyperacusis. In this randomized controlled study of 60 patients with hyperacusis, CBT was compared with a waiting list control group using the Loudness Discomfort Level test (LDL), the Hyperacusis Questionnaire, the Hospital Anxiety and Depression Scales, the Quality of Life Inventory and an adapted version of the Tampa Scale of Kinesiophobia. There were significant between-group effects in favour of the CBT group on all measures except for the HADS anxiety scale. Between-group effect sizes were moderate to high, with Cohen's d = 0.67 and 0.69 per ear, respectively, for the primary measure LDL, and ranging from d = 0.32 to 1.36 for the secondary measures. The differences between groups ceased to exist when the waiting list group was treated later with CBT, and the treatment results were largely maintained after 12 months. In conclusion, CBT is a promising treatment for hyperacusis, although more research is necessary.

Hyperacusis as a primary problem has been defined as “unusual intolerance to ordinary environmental sounds” ( Vernon, 1987). Persons with hyperacusis are sensitive to sounds such as music, clatter, mechanical sounds and/or paper noises ( Andersson, Lindvall, Hursti, & Carlbring, 2002), and often protect themselves from sounds in different ways such as by wearing hearing protection (when taking the bus, for example) – even when there is no confirmed risk of hearing damage ( Baguley, 2003). The personal suffering related to hyperacusis has been described in the literature, with patients reporting feelings of fear, extensive use of ear protection devices, and avoidance of environments such as their places of work and settings for social activities ( Baguley and Andersson, 2007 and McKenna et al., 2010). A high percentage of sick leave from work has also been found in this group of patients ( Jüris, Andersson, Larsen, & Ekselius, 2013). Prevalence studies of hyperacusis are rare. In a Swedish study where data were collected via either a postal survey or the internet, the prevalence rates were 7.7% (n = 39) and 5.9% (n = 28), respectively, when excluding individuals with hearing impairment ( Andersson et al., 2002). In a Finnish study, where a broader definition of hyperacusis was used, the prevalence of self-reported hyperacusis was 17.2% ( Hannula, Bloigu, Majamaa, Sorri, & Mäki-Torkko, 2011). Little has been published concerning the aetiology and natural course of hyperacusis. Disturbed metabolism of 5-hydroxytryptamine (5-HT; serotonin) has been proposed as a mechanism in hyperacusis (Marriage & Barnes, 1995) and might also account for development of hyperacusis in depression and anxiety disorders (Attri & Nagarkar, 2010). The central gain hypothesis views hyperacusis as a result of a problematic compensatory gain process in the auditory pathways (Jastreboff & Hazell, 1993). Hyperacusis is reported to co-exist with many other conditions, including migraine and William's syndrome (Katzenell & Segal, 2001). There is also a large comorbidity between tinnitus and hyperacusis (Baguley, 2003), as 40 per cent of patients with tinnitus suffer from hyperacusis (Jastreboff & Jastreboff, 2000), and up to 86 per cent of patients with hyperacusis in clinical samples report tinnitus (Anari, Axelsson, Eliasson, & Magnusson, 1999). In a recent study almost half of the patients diagnosed with hyperacusis suffered from anxiety disorders (Jüris et al., 2013). The most common anxiety disorders were social phobia and agoraphobia, as measured with the Mini-International Neuropsychiatric Interview (MINI, Swedish version 5.0.0.) (Sheehan et al., 1998). Furthermore, the patients also displayed high scores on anxiety-related personality traits, measured with the Swedish universities Scales of Personality. As no underlying medical condition can be found in the large majority of affected patients (Baguley, 2003), researchers have suggested that hyperacusis is maintained and exacerbated by avoidance of sounds and increased anxiety (Schaaf, Klofat, & Hesse, 2003). It has been shown that persons who do not suffer from hyperacusis become more sensitive to sounds when they overprotect their ears (Formby, Sherlock, & Gold, 2003), while exposure to low-level noise treatment later desensitizes the same subjects. In another study, hyperacusis was found to be associated with noise-related avoidance behaviour and anxiety (Blaesing & Kroener-Herwig, 2012). This is in accord with the fear-avoidance model, which deals with fear of pain and focuses on the individual response of either confronting or avoiding the pain itself (Lethem et al., 1983 and Vlaeyen and Linton, 2000). The avoidance of pain is assumed to predict further avoidance and increased fear of pain, leading to inactivity, which in itself leads to further disability. This model has found support in the literature (Vlaeyen & Linton, 2012). In a study of healthy participants, fear-avoidance beliefs were quite rare. When present, however, they increased the risk of future pain episodes (Linton, Buer, Vlaeyen, & Hellsing, 2000). Fear-avoidance is also a risk factor for poor health in patients suffering from burns (Willebrand, Andersson, Kildal, Gerdin, & Ekselius, 2006). To our knowledge, there are no published randomized controlled trials (RCT) of any psychological treatment for hyperacusis. We assume that most patients who suffer from hyperacusis receive an audiological examination and some form of counselling, or are prescribed ear attenuation devices at their audiology clinic. An existing method for treating tinnitus, and also hyperacusis, is Tinnitus Retraining Therapy (TRT) (Jastreboff & Hazell, 1993), but there are few published controlled studies focusing on hyperacusis (Formby et al., 2013). Treatments involving measures to help patients avoid ear protection and promote exposure to increasing levels of pink noise have been reported to have good effect (Vernon, 1987). Cognitive behaviour therapy (CBT) is effective for a range of psychiatric disorders and could be considered the primary psychosocial treatment of choice for many, if not most, patients with mild to moderate psychiatric problems (Tolin, 2010). CBT is also effective as an adjunct for many somatic problems such as chronic pain (Eccleston et al., 2012). For the psychological problems associated with tinnitus, CBT is the treatment of choice (Hesser et al., 2011 and Martinez-Devesa et al., 2010). CBT is also effective for anxiety disorders (Hofmann & Smits, 2008), and has been proposed as a reasonable strategy for treating anxiety and stress associated with hyperacusis (Baguley, 2003), as there are similarities between anxiety disorders and hyperacusis. For persons with hyperacusis, avoidance behaviour protects the individual from the instant unpleasantness of certain sounds, but in the long run overprotecting the ears exacerbates hyperacusis (Vernon, 1987), and may lead to isolation and a depressed mood. Similar behavioural avoidance is also an important factor in the development of anxiety disorders, and motivates the use of exposure treatment (Murphy et al., 1997 and Salkovskis et al., 1999). The aim of this study was to investigate whether CBT could be helpful for patients with hyperacusis. The hypothesis was that patients with hyperacusis would benefit from CBT, as measured by loudness discomfort levels, hyperacusis symptoms, anxiety and depressive symptoms, quality of life, and fear of injury/reinjury due to exposure to sounds.

Results Demography Demographic data for the final 60 patients are presented in Table 2. Their ages ranged from 18 to 61 years with a mean of 40.2 years (SD = 12.2). Forty-five (75%) patients were women. The average duration of hyperacusis was 12.2 years (SD = 15.2; range 1–60). The median duration was 5.0 years (calculated, as 10 patients claimed they had always suffered from hyperacusis). Table 2. Demography. CBT (n = 30) WL (n = 30) Marital Status (married/single) 14/16 18/12 Education >2 years university 16 20 <2 years university 5 0 Upper secondary school 9 8 Elementary school 0 2 Co-morbid tinnitus (y/n) 23/7 26/4 M (SD) M (SD) Age 38.3 (11.1) 42.1 (13.1) Hearing Threshold Right 11.2 (10.8) 10.3 (9.0) Hearing Threshold Left 13.3 (11.3) 12.0 (8.2) Table options The patients had mean hearing thresholds of 10.8 dB (SD = 9.9) in the right ear and 12.6 dB (SD = 9.8) in the left ear for the frequencies 500, 1000, 2000 and 3000 Hz (clearly within the range of normal hearing). There was an expected significant difference in the average hearing thresholds between the right and left ears, with the right ear outperforming the left (t = 2.76, p < 0.01). The mean LDLs ranged from 69.3 (SD = 12.1) to 76.0 (SD = 11.7) on average for the frequencies 250, 500, 1000, 2000, 3000 and 4000 Hz. There were no significant differences in LDL values between the right and left ears. Baseline differences Table 2 describes patient characteristics. T-tests showed no statistically significant differences between the groups either for the variables of sex, age and tinnitus, or for any of the outcome measures at baseline. In Table 3, mean values and standard deviations at baseline, pre-treatment for the waiting-list group, post-treatment and follow-up are presented for the outcome measures. Table 3. Means and standard deviations for baseline, pre-treatment (for the WL group only), post-treatment and follow-up measures. CBT WL Total M (SD) N M (SD) N N LDL Right ear Baseline 71.88 (10.75) 30 72.22 (10.79) 30 60 Pre-treatment (WL) – – 70.17 (12.19) 30 30 Post-treatment 78.55 (13.00) 30 79.06 (10.17) 30 60 12-month follow-up 80.09 (11.13) 30 79.89 (9.08) 30 60 LDL Left ear Baseline 73.47 (10.73) 30 72.94 (10.85) 30 60 Pre-treatment (WL) – – 70.44 (13.64) 30 30 Post-treatment 79.54 (12.80) 30 78.09 (9.95) 30 60 12-month follow-up 81.26 (10.17) 30 79.81 (9.87) 30 60 HQ Baseline 29.77 (5.49) 30 29.83 (6.33) 30 60 Pre-treatment (WL) – – 29.90 (6.24) 30 30 Post-treatment 21.50 (8.44) 28 25.71 (6.09) 28 56 12-month follow-up 20.21 (8.39) 28 24.52 (6.46) 27 55 HADS Anxiety Baseline 7.67 (4.66) 30 7.63 (3.64) 30 60 Pre-treatment (WL) – – 7.07 (4.03) 30 30 Post-treatment 5.75 (4.29) 28 5.86 (3.53) 28 56 12-month follow-up 5.78 (4.65) 27 6.07 (3.96) 27 54 HADS Depression Baseline 6.63 (3.84) 30 5.80 (3.75) 30 60 Pre-treatment (WL) – – 5.63 (4.00) 30 30 Post-treatment 4.21 (3.55) 28 4.68 (2.97) 28 56 12-month follow-up 3.81 (3.09) 27 4.63 (3.83) 27 54 TSK Baseline 44.30 (8.29) 30 42.87 (8.64) 30 60 Pre-treatment (WL) – – 41.17 (9.92) 30 30 Post-treatment 29.18 (7.49) 28 32.89 (9.51) 28 56 12-month follow-up 30.07 (6.58) 28 31.67 (10.08) 27 55 QOLI Baseline 1.49 (1.60) 30 0.97 (2.04) 30 60 Pre-treatment (WL) – – 1.30 (2.00) 30 30 Post-treatment 2.17 (1.38) 28 1.64 (1.95) 28 56 12-month follow-up 2.36 (1.35) 28 1.61 (1.98) 27 55 Table options Effect on the primary outcome measure The ANCOVAs showed a significant between-group effect on the LDL test for both ears (right ear F(1, 57) = 14.2, p < 0.001, left ear F(1, 57) = 11.6, p < 0.001), as illustrated in Fig. 2. Without imputation, values were similar (right ear F(1,55) = 12.5, p < 0.001, left ear F(1,55) = 10.1, p < 0.01). Between-group effect sizes (Cohen's d) were moderate, d = 0.67 for the right ear and d = 0.69 for the left ear. The within-group effect size in the CBT group was also moderate, d = 0.56 for the right ear and d = 0.51 for the left ear. Effect sizes are presented in Table 4. Full-size image (26 K) Fig. 2. Results for the LDL-test at the four assessments: 1 = baseline; 2 = post treatment for the CBT-group and pre-treatment for the WL-group; 3 = post-treatment for the WL-group, and 4 = 12-month follow-up for both groups. Significant differences exist between CBT- and WL-groups at assessment 2 for both ears. Figure options Table 4. Effect sizes, Cohen's d. WL-group has received CBT at the 12-month follow-up. Within-group effect sizes Between-group effect sizes CBT WL CBT vs WL LDL Right ear Post-treatment 0.56 0.18 0.67 12-month follow-up 0.75 0.90 0.02 LDL Left ear Post-treatment 0.51 0.20 0.69 12-month follow up 0.74 0.79 0.14 HQ Post-treatment 1.16 0.01 1.13 12-month follow-up 1.35 0.85 0.58 HADS Anxiety Post-treatment 0.49 0.15 0.32 12-month follow-up 0.46 0.25 0.07 HADS Depression Post-treatment 0.73 0.04 0.38 12-month follow-up 0.90 0.26 0.24 TSK Post-treatment 1.91 0.18 1.36 12-month follow-up 1.90 0.95 0.19 QOLI Post-treatment 0.46 0.16 0.51 12-month follow-up 0.59 0.16 0.44 Table options Effect on the secondary outcome measures After controlling for differences in pre-test scores, there were significant group effects in favour of the treatment group on all secondary outcome measures except for the HADS anxiety scale (see Table 3). The Hyperacusis Questionnaire The ANCOVA showed significant group effects for hyperacusis severity (F(1,55) = 27.5, p < 0.001). The between-group effect size was large, d = 1.13, and the within-group effect size for the CBT group was also large, d = 1.16. The Hospital Anxiety and Depression Scale The ANCOVA showed significant group effects for the depression scale (F(1, 55) = 8.1, p < 0.01), but not for the anxiety scale (F(1, 55) = 3.9, p = 0.054), although there was a trend in favour of the CBT group. Between-group effect sizes for depression were small, d = 0.38, and the within-group effect size for the CBT group was moderate, d = 0.73. The Quality of Life Inventory A significant treatment effect was shown on the QOLI (F(1,55) = 4.3, p < 0.05). The between-group effect size was moderate, d = 0.51, and the within-group effect size for the CBT group was small, d = 0.46. The Tampa Scale of Kinesiophobia (adapted to hyperacusis) The ANCOVAs showed significant group effects (F(1, 55) = 47.1, p < 0.001). The between-group effect size was large, d = 1.36, and the within-group effect size for the CBT group was also large, d = 1.91. Comparison between the treatment condition and the waiting-list group after receiving treatment When the waiting-list group later received CBT, the outcomes were similar. No significant differences were observed (see Table 4 for effect sizes). Follow-up The follow-up assessment was conducted 12 months after treatment, and was completed by 55 patients. Improvements were maintained at the group level for all measures in the CBT group. At follow-up the differences due to group on the HQ were significantly different in favour of the treatment group (F(1,52) = 5.3, p < 0.05). The HADS anxiety scale showed a significant improvement in the treatment group from baseline to follow-up, and also in the waiting-list group from pre-treatment to follow-up after they had also received CBT.