Tinnitus and Metacognitive Beliefs—Results of a Cross-Sectional Observational Study

Recent research has highlighted the role of metacognitions as a moderator for psychological distress in patients with chronic diseases. The present study investigates the role of metacognitions and worry in the association between tinnitus distress, anxiety, and depression. A cross-sectional study was carried out with a sample of tinnitus-outpatients who completed the Tinnitus-Handicap Inventory, Beck Anxiety Inventory, Beck Depression Inventory, Metacognition Questionnaire-30, Penn-State-Worry-Questionnaire. Associations of metacognitions, worries, tinnitus distress, anxiety and depression were investigated using structural equation models (SEMs). A sample of n = 107 patients was included in the study. In the first SEM, tinnitus distress significantly predicted depression (β = 0.68, p < 0.001) and anxiety (β = 0.47, p < 0.001). In the second model, worries and meta-cognitions were added as moderators. The explained variance substantially increased for depression (46 to 53%) and anxiety (22 to 35%) and the association of tinnitus distress with depression (β = 0.57, p < 0.001) and anxiety was weakened (β = 0.32, p < 0.001). Negative beliefs significantly predicted worries (β = 0.51, p < 0.001) and explained 41% of its variance. A good model fit for the final model was found (comparative fit index (CFI) = 0.98; (Tucker Lewis index) TLI = 0.96; root mean square error of approximation (RMSEA) = 0.067). Anxiety and depression in tinnitus patients might be influenced by worries, which is mainly predicted by negative beliefs about uncontrollability and danger of worries. Thus, psychotherapeutic approaches focused on alterations of metacognitions in patients with tinnitus should be investigated in future studies.3 (List three to ten pertinent keywords specific to the article yet reasonably common within the subject discipline.)


Introduction
Tinnitus is a multifactorial disorder and it involves the percept of a sound or sounds, such as ringing, roaring, whistling or buzzing, in the ear or head when no external source is present [1]. Tinnitus can be linked to many different etiologies, but it in 40% of patients it is considered idiopathic, as "no known events" are associated with the tinnitus onset [2] and often the approach to tinnitus requires a multidisciplinary team to manage the complexity of the symptoms. Subjective chronic tinnitus occurs in 10 to 15% [3] of the adult population and for some patients the sound is very debilitating. Tinnitus can also coexists with hidden hearing loss [4] and may be modulated by temporo-mandibular joint and neck disorders [5,6]. In case of hearing loss and tinnitus, cochlear abnormalities are considered the initial causal source and subsequently neural changes in the central auditory system maintain the tinnitus [3,7]. The perception of tinnitus may be constant or intermittent [8]. Anxiety and depression are frequently reported by patients with tinnitus [9][10][11][12] but the psychological mechanisms that drive this association are still unclear. In patients with physical health conditions, the symptoms of anxiety and depression are common and Brain Sci. 2021, 11, 3 2 of 12 recent reviews have highlighted the role of metacognition as a moderator for psychological distress in patients with chronic diseases [13,14].
The term metacognition indicates "the aspect of information processing that monitors, interprets, evaluates, and regulates the contents and processes of its organization" [15] and the Metacognitive Theory (MCT) [16] is based on the principle that the vulnerability and the extension of emotional disorders are associated with a dysfunctional style of thinking, named Cognitive Attentional Syndrome (CAS). CAS refers to repetitive negative thinking in the process of worrying and ruminating, a pattern of focusing attention on threat and dysfunctional coping strategies. Metacognitive beliefs support the CAS and they might be a relevant element associated with the process of adaptation to illness. The two main kinds of metacognitive beliefs are positive beliefs and negative beliefs. Positive beliefs about worry concern the benefits to engaging in it (e.g., "Worrying helps me cope"), while negative beliefs focus on the uncontrollability and danger of thoughts (e.g., "I could make myself sick with worrying"). Other metacognitions are cognitive confidence (e.g., "I have a poor memory"), cognitive self-consciousness (e.g., "I monitor my thoughts") and negative beliefs about the need to control thoughts (e.g., "If I did not control a worrying thought, and then it happened, it would be my fault").
Negative beliefs about worry seem to represent a common metacognitive pattern across pathologies such as Parkinson's disease [17,18], cardiac illness [19] or epilepsy [20,21]. Cognitive confidence, a measure of an individual's confidence in his own attention and memory might negatively affect coping strategies when the patient has chronic fatigue or neurological disorders, for example, feels mentally fatigued [14].
Previous studies have not investigated the role of metacognitive beliefs in tinnitus distress, but in a single group study [22] MCT was used to treat tinnitus suffers. Results indicated a reduction in the perception of the tinnitus, anxiety, and in the significance of annoying thinking. In addition, Caldirola and colleagues [23] examined the function of worry in patients with chronic tinnitus. In this article tinnitus handicap, anxiety and depressive symptoms were significantly associated with proneness to worry (linear regression models, p < 0.01).
This evidence suggests the need to analyze metacognitive beliefs and the tendency to worry in patients with tinnitus to improve treatments. The aim of the study was to investigate the role of metacognitions and worry in the association between tinnitus distress and the onset of anxiety and depression.

Sample and Setting
A sample of n = 107 consecutive outpatients was included in this cross-sectional study. Patients were evaluated between April 2018 and April 2019 at the Tinnitus Center of European Hospital (Rome). Inclusion criteria were (a) presence of tinnitus, (b) age over 18 years, (c) speaking Italian fluently and (d) absence of cognitive impairment or major psychiatric/neurological disorders (schizophrenia, Alzheimer's disease, Parkinson's disease). All patients were evaluated with otorhinolaryngological and psychological examination, audiometry and questionnaires as part of the routine clinical practice. Based on the duration of tinnitus, patients were classified as with acute tinnitus (<3 months), subacute tinnitus (3-6 months) and chronic tinnitus (>6 months). The study was conducted in accordance with the Declaration of Helsinki. The protocol of the study was approved by the Ethics Committee of University of L'Aquila (ID number 42:2020, protocol number 128446). Informed consent was obtained by every patient before filling out the questionnaires.

Tinnitus Sample Case History (TSCH)
The TSCH is a standardized questionnaire of 35 items to collect sociodemographic and clinical data in tinnitus research [24]. The Italian version of the TSCH was used to collect significant information, such as patient background (i.e., age, gender), tinnitus char-Brain Sci. 2021, 11, 3 3 of 12 acteristics (i.e., loudness, pitch, perception at onset, tinnitus location, percentage of awake time aware of tinnitus) and related medical conditions (i.e., hearing impairment, noise annoyance, vertigo/dizziness, headache, Temporomandibular joint disorders, cervical disorders).

Tinnitus Handicap Inventory (THI)
The THI [25] is a self-report questionnaire widely used in tinnitus research to assess the impact of tinnitus in daily life. It consists of 25 items and 3 subscales (functional, emotional and catastrophic). The Italian THI version has a good reliability (α = 0.94) and validity [26] and was used to assess tinnitus severity graded from slight (grade 1) to catastrophic (grade 5). A THI score > 36 was considered to indicate a decompensated tinnitus [27,28].

Penn State Worry Questionnaire (PSWQ)
The PSWQ [35,36] is a 16-item self-administered scale to measure worry. The items are scored on a 5-point Likert-type scale (1-not at all typical of me, to 5-very typical of me). Based on the final total score, the patients' worries can be classified as low , moderate (40-59) and high (60-80).

Tonal Audiometry
The presence of hearing loss was investigated with tonal audiometry. Pure tone audiometry was carried out using a clinical audiometer (Madsen Itera II, GN Otometrics) and patients were evaluated into an audiological cabin. Normal hearing was defined by threshold < 25 dB HL in all frequencies tested between 250 Hz and 8.000 Hz.

Statistical Analyses
Demographics for the sample are presented with means and standard deviations (SD). The relationship of tinnitus distress, meta-cognitions, and depression, anxiety and worries were investigated with two structural equation models (see Figure 1). In model A the direct influence of tinnitus distress on depression and anxiety was tested. In model B, worries were added to the model as moderators for this relationship and meta-cognitions were added as predictors for worries. The association of the five assessed meta-cognitions with tinnitus distress, anxiety, depression and worries were tested with Pearson correlation coefficients prior to inclusion to the model. Meta-cognitions were added to the model if a significant association was found with at least one of the dependent variables. Pearson's Brain Sci. 2021, 11, 3 4 of 12 chi-squared test (χ 2 ), the comparative fit index (CFI) and root mean square error of approximation (RMSEA) with lower and higher bounds of the 95% confidence interval (CI) were calculated to determine the model's goodness of fit. To evaluate whether the empirical data fit closely with the theoretical model, the p-value of Close Fit (PCLOSE) was calculated based on the RMSEA values, with values of p > 0.05 indicating close fit and p < 0.05 indicating worse than close model fit. Acceptable goodness of fit was defined as RMSEA values of < 0.08 and CFI values > 0.90. p-values < 0.05 (two-sided) were considered statistically significant. Statistical analyses were performed with IBM SPSS (v22.0) and SPSS AMOS (v24.0).
Brain Sci. 2021, 11, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/brainsci worries were investigated with two structural equation models (see Figure 1). In model A the direct influence of tinnitus distress on depression and anxiety was tested. In model B, worries were added to the model as moderators for this relationship and meta-cognitions were added as predictors for worries. The association of the five assessed meta-cognitions with tinnitus distress, anxiety, depression and worries were tested with Pearson correlation coefficients prior to inclusion to the model. Meta-cognitions were added to the model if a significant association was found with at least one of the dependent variables. Pearson's chi-squared test (χ 2 ), the comparative fit index (CFI) and root mean square error of approximation (RMSEA) with lower and higher bounds of the 95% confidence interval (CI) were calculated to determine the model's goodness of fit. To evaluate whether the empirical data fit closely with the theoretical model, the p-value of Close Fit (PCLOSE) was calculated based on the RMSEA values, with values of p > 0.05 indicating close fit and p < 0.05 indicating worse than close model fit. Acceptable goodness of fit was defined as RMSEA values of < 0.08 and CFI values > 0.90. p-values < 0.05 (two-sided) were considered statistically significant. Statistical analyses were performed with IBM SPSS (v22.0) and SPSS AMOS (v24.0).

Figure 1.
Structural equation models to test the moderating effect of worries and meta-cognitions on the relationship of tinnitus distress with anxiety and depression. Model A depicts the direct association of tinnitus distress with the dependent variables (anxiety, depression). Model B depicts the model with worries as a moderator of the association of tinnitus distress with the dependent variables (anxiety, depression) and meta-cognitions as predictors for worries.

Results
A total of n = 107 patients were included in the analyses (60.7% male). A total of 77.1% of patients reported chronic tinnitus and 63.6% had a hearing loss. Details on patient characteristics are reported in Table 1. Based on the total THI score, 19.6% of the patients were classified as very mild (grade I), 28.0% mild (grade II), 29.0% moderate (grade III), 15.9% severe (grade IV), and 7.5% very severe (grade V). Based on the cut-off of 36 on the THI, 52.3% of the patients reported a decompensated tinnitus.

Meta-Cognitions in Patients with Tinnitus
The highest correlations between MCQ-subscales were found for negative beliefs about the controllability of thoughts and danger of worry (MCQ neg) with beliefs about the need to control thoughts (MCQ NC) and with cognitive self-consciousness (MCQ CSC). No association was found between cognitive confidence (MCQ CC) and negative beliefs about the controllability of thoughts and danger of worry (MCQ neg), beliefs about the need to control thoughts (MCQ NC), and cognitive self-consciousness (MCQ CSC) ( Table 2). No gender differences were found for the MCQ-30 subscales, except for the need to control thoughts, with males scoring higher on this subscale (11.3 vs. 9.9 points; t = 2.03; p = 0.045). None of the MCQ-30 subscales were associated with patients age, subjective tinnitus loudness, or awareness of tinnitus (all p > 0.05). However, both the negative beliefs about uncontrollability and danger of worries (r = 0.26; p = 0.016) as well as the need for control (r = 0.26; p = 0.016) were significantly associated with a higher level of annoyance be the ear noise.

The Association of Tinnitus Distress with Meta-Cognitions, Anxiety, Depression and Worry
The results of our study evidenced that higher levels of tinnitus distress were significantly associated with negative beliefs about the controllability of thoughts and danger of worry and beliefs about the need to control thoughts. Anxiety, depression, and worries were significantly associated with negative beliefs about the controllability of thoughts and danger of worry, cognitive confidence and beliefs about the need to control thoughts. For worries, an additional positive correlation was observed with higher cognitive selfconsciousness. Overall, strongest correlations with tinnitus distress and anxiety, depression and worries were found for negative beliefs about the controllability of thoughts and danger of worry. For details see Table 3.

Direct Association of Tinnitus Distress with Depression and Anxiety
In the first step, the direct associations of tinnitus distress with depression and anxiety were investigated by calculation of a SEM. Tinnitus distress significantly predicted depression (p < 0.001, β = 0.68) and anxiety (p < 0.001, β = 0.47) and explained 46% and 22% of the variance, respectively. Strong intercorrelations were found between depression and anxiety (r = 0.56) (Figure 2).

Direct Association of Tinnitus Distress with Depression and Anxiety
In the first step, the direct associations of tinnitus distress with depression and anxiety were investigated by calculation of a SEM. Tinnitus distress significantly predicted depression (p < 0.001, β = 0.68) and anxiety (p < 0.001, β = 0.47) and explained 46% and 22% of the variance, respectively. Strong intercorrelations were found between depression and anxiety (r = 0.56) (Figure 2).

Meta-Cognitions and Worries as Moderators for the Association of Tinnitus Distress with Anxiety and Depression
In the second step, the PSWQ total score and MCQ-30 subscales were added to the model as moderators. As shown in Table 2, significant correlations for four of the five MCQ-30 subscales with anxiety depression and worries were found, namely positive beliefs about worry, negative beliefs about the controllability of thoughts and danger of worry, cognitive confidence, beliefs about the need to control thoughts and cognitive selfconsciousness with the factors of model one were found. Thus, four of the five MCQ-30 subscales were added to the model.

Meta-Cognitions and Worries as Moderators for the Association of Tinnitus Distress with Anxiety and Depression
In the second step, the PSWQ total score and MCQ-30 subscales were added to the model as moderators. As shown in Table 2, significant correlations for four of the five MCQ-30 subscales with anxiety depression and worries were found, namely positive beliefs about worry, negative beliefs about the controllability of thoughts and danger of worry, cognitive confidence, beliefs about the need to control thoughts and cognitive self-consciousness with the factors of model one were found. Thus, four of the five MCQ-30 subscales were added to the model.

Discussion
Anxiety and depression are frequently reported by patients with tinnitus. The psychological mechanisms that drive the association between tinnitus distress and anxiety and depression are still unclear. Many tinnitus pathophysiology models have been developed to investigate the mechanisms of tinnitus awareness and distress, such as the neurophysiological model, the habituation theory, the fear-avoidance model and the cognitive model [1]. The Cognitive Behavioral Model of Tinnitus proposes the interaction between tinnitus and negative thoughts, negative emotions, attention and monitoring, safety behavior, and beliefs [37]. The aim of the present study was to investigate the role of metacognition and worry in the association between tinnitus distress, anxiety and depression. As hypothesized, we found that metacognitive beliefs and worry predicted the development of anxiety and depression in patients with tinnitus. Studies using connectivity neuroimaging techniques like diffusion MRI and resting-state fMRI have been devoted to exploring the neuroimaging relationship between tinnitus and auditory-limbic interactions [38]. A strong connectivity between the ventral and dorsal prefrontal cortical network was recently demonstrated by fMRI in the regulation of action selection based on metacognition but this has not yet been shown in tinnitus research [39].
Almost a third of the patients in our study reported depressive symptoms while almost half of them reported relevant levels of anxiety. In accordance with previous studies [9][10][11][12], increased levels of tinnitus distress were associated with higher levels of

Discussion
Anxiety and depression are frequently reported by patients with tinnitus. The psychological mechanisms that drive the association between tinnitus distress and anxiety and depression are still unclear. Many tinnitus pathophysiology models have been developed to investigate the mechanisms of tinnitus awareness and distress, such as the neurophysiological model, the habituation theory, the fear-avoidance model and the cognitive model [1]. The Cognitive Behavioral Model of Tinnitus proposes the interaction between tinnitus and negative thoughts, negative emotions, attention and monitoring, safety behavior, and beliefs [37]. The aim of the present study was to investigate the role of metacognition and worry in the association between tinnitus distress, anxiety and depression. As hypothesized, we found that metacognitive beliefs and worry predicted the development of anxiety and depression in patients with tinnitus. Studies using connectivity neuroimaging techniques like diffusion MRI and resting-state fMRI have been devoted to exploring the neuroimaging relationship between tinnitus and auditory-limbic interactions [38]. A strong connectivity between the ventral and dorsal prefrontal cortical network was recently demonstrated by fMRI in the regulation of action selection based on metacognition but this has not yet been shown in tinnitus research [39].
Almost a third of the patients in our study reported depressive symptoms while almost half of them reported relevant levels of anxiety. In accordance with previous studies [9][10][11][12], increased levels of tinnitus distress were associated with higher levels of depression and anxiety. Additionally, women reported higher values in anxiety and depression which may confirm a specific gender effect for the assessed psychological comorbidities [9,40,41].
Higher levels of tinnitus distress were associated with negative beliefs about the controllability of thought, as well as with a higher need to control thoughts. Anxiety, depression, and worries were significantly associated with negative metacognitive beliefs, cognitive confidence and beliefs about the need to control thoughts. For worries, an additional positive correlation, was observed with higher cognitive self-consciousness. Our results confirmed the prominent role of metacognitive beliefs in the CAS maintaining psychological disorders [16].
In our study we also tested the hypothesis that meta-cognitions influence the tendency for worries, which in itself may be a moderator for the relationship of tinnitus distress with anxiety and depression. As hypothesized, the inclusion of worries led to a substantial increase in the explained variance in both depression and anxiety. Furthermore, interesting data were reported in model B of the two structural equation models, where the PSWQ total score and MCQ-30 subscales were added as moderators. The overall explained variance substantially increased for depression (59%) and anxiety (44%) and the direct association of tinnitus distress with both depression and anxiety was significantly weakened after inclusion of worries, indicating that worries might indeed moderate the relationship between tinnitus distress, depression and anxiety. Worries on the other hand were significantly predicted by meta-cognitions which explained 41% of the variance of the variable worries. However, this association was mainly caused by the factor negative beliefs about the uncontrollability and danger of worries, while the other included MCQ-30 subscales were not significant in the structural equation model. Our results thus confirmed that worry and metacognitive beliefs are related to the development and maintenance of anxiety and depression symptoms [42,43] and that negative beliefs are a significant predictor of these symptoms in different illness types [17,44,45].
Tinnitus is usually considered chronic after 6 months [1]. In many cases the natural history of tinnitus is the habituation to the auditory sensation, but in approximately 20% of patients the tinnitus is considered bothersome [46]. Bothersome tinnitus affects sleep [47], concentration and psychological stability [40,41,[48][49][50][51]. In a previous study, Budd and Pugh [52] reported that the adaptation or habituation to tinnitus may occur over time as a result of avoidance of maladaptive coping strategies, rather than due to the patients active use of effective coping strategies. Worry is a kind of maladaptive coping strategy that maintains attention on threats, and it has an important function in the CAS [16]. As highlighted in the present study, worries are significantly associated with increased anxiety and depression in tinnitus patients and negative beliefs may be a driving factor in this association. The results thus confirmed that the importance of dysfunctional metacognitive beliefs and the tendency to worry are relevant factors associated within the process of adapting to a chronic condition illness, which has also been highlighted in recent studies on psychological distress in patients with chronic diseases [13,14].
As for now, the standard technique of Cognitive Behavioral Therapy (CBT) [53] focusses on thought content instead of on the thinking process. However, psychological interventions centered on the MCT [16] could facilitate treatments for patients with tinnitus using specific techniques as Detached Mindfulness and Attentional Training Technique (ATT) [54] to manage worry. These techniques develop the activation of metacognitive knowledge, attentional flexibility and a decentered relationship with thoughts. Patients can learn to observe their thoughts and feelings and experience them, without automatically identifying with them. In particular, the ATT could be adapted to help patients shift attention from tinnitus to other stimuli. The finding that negative thoughts and dysfunctional cognitions appears to be linked to a worse experience of tinnitus was recently investigated by other authors using the Tinnitus Cognitions Questionnaire (TCQ) [55] and the Tinnitus Cognitions Scale (T-Cog) [56]. In the present study we focused our attention on the thought process using the MCQ to demonstrate that worry and metacognition play a fundamental role in maintaining the disorder even in patients with tinnitus.
This study has some limitations. First, we could not find a significant difference between worry and metacognition in acute and chronic because we evaluated a high number of patients with chronic tinnitus (77.1%). Second, our sample is small and we did not investigate the presence of other medical comorbidities (diabetes, insomnia, migraine, hypertension) and tinnitus pitch and loudness. Third, we did not analyze the results based on particular tinnitus subtypes. Additionally, the study was conducted as a cross-sectional observational study and thus no causal conclusions can be drawn from the study. However, the observed correlations give indications for future research and allow valuable theoretical conclusions.

Conclusions
Anxiety and depression in tinnitus patients might be influenced by worries, which are mainly predicted by negative beliefs about uncontrollability and danger of worries. Thus, psychotherapeutic approaches focused on alterations of metacognitions in patients with tinnitus should be investigated in future studies.