Key Points
Question
Can specific patient and clinical characteristics predict response to cognitive behavioral therapy (CBT) in patients with tinnitus?
Finding
This secondary cohort analysis of a single-arm clinical study including 88 adult patients with tinnitus who underwent CBT found that participants with extreme tinnitus bother and moderate to high anxiety levels experienced the greatest associated reduction of tinnitus severity.
Meaning
These findings suggest that CBT as a treatment for tinnitus may be more effective in select subgroups of patients based on severity of symptoms and clinical psychosocial impact of tinnitus.
Abstract
Importance
Clinical guidelines recommend cognitive behavioral therapy (CBT) as a treatment for tinnitus. However, patient response to CBT is variable, and currently, there are no known predictors of response to CBT treatment for tinnitus.
Objective
To identify the clinical predictors of patient response to CBT for treatment of tinnitus.
Design, Setting, and Participants
This was a secondary cohort analysis of a single-arm clinical study including adults with chronic bothersome tinnitus recruited from Washington University School of Medicine in St Louis (Missouri) from September 2019 to February 2023. Participants completed an 8-week group CBT program with a licensed clinical psychologist. Each week consisted of 2.5 hours of CBT, amounting to 20 hours of total CBT participation, primarily delivered through a virtual platform. Conjunctive consolidation was used to create a predictive classification system for response to CBT based on tinnitus bother and anxiety levels.
Main Outcome and Measure
Response to CBT was predefined as a 13-point or greater decrease in the Tinnitus Functional Index (TFI) survey score.
Results
The study sample included 88 adult patients (median [IQR] age, 59 [49-66] years; 47 [53%] females and 41 [47%] males) with chronic bothersome tinnitus, of whom 53 (60%) had at least 13-point decrease in TFI and were considered to be responders. In univariable and multivariable logistic regression analyses, high to moderate anxiety level and severe tinnitus bother were associated with treatment response (adjusted odds ratio: anxiety, 3.33; 95% CI, 0.90-12.30; tinnitus bother, 12.08; 95% CI, 1.48-98.35). The clinical stratification system showed good predictive and discriminative ability (χ2 for linear trend = 20.0; C statistic = 0.75; 95% CI, 0.65-0.85).
Conclusions and Relevance
The findings of this study show that assessment of bother and anxiety levels in patients with tinnitus may be useful for identifying those who are more likely to respond to CBT. Before incorporation into clinical practice, future research should externally validate this finding in a separate population.
This secondary cohort analysis identifies clinical predictors of patient response to cognitive behavioral therapy for the treatment of tinnitus.
Introduction
Tinnitus is the perception of a sound in the absence of an external auditory stimulus.1,2,3 Tinnitus may impact quality of sleep, impair concentration, and interfere with the ability to communicate.4,5 The severity of distress from tinnitus can vary from mild to severe. According to the US National Health Interview Survey data,6 11 in 100 individuals experience tinnitus; that is, an estimated 27 million people in the US. Tinnitus has been the most prevalent and most claimed military service−related disability filed for compensation since the Vietnam War.7 Patients with tinnitus report willingness to utilize additional health care, undergo invasive treatments, and spend a considerable amount of money to alleviate the severity of tinnitus, resulting in increased health care expenditures per patient.8,9
There are a variety of treatments for tinnitus. Hearing aids and sound maskers reduce tinnitus distress by redirecting auditory focus away from tinnitus.10 Several behavioral modification therapies, including cognitive behavioral therapy (CBT),11,12 mindfulness-based cognitive therapy,13,14 and acceptance and commitment therapy15 have demonstrated effectiveness in reducing tinnitus severity and symptom burden. A meta-analysis of 10 randomized clinical trials demonstrated the efficacy of CBT in reducing tinnitus distress.16 Therefore, CBT has emerged as a recommended treatment for tinnitus.17 However, a percentage of patients—from 35% to 43%—do not respond to CBT,11,18 and no clinical factors have been identified that are associated with response to treatment.
This study aimed to identify patient characteristics associated with treatment response to CBT and to create a system that can be used to classify patients based on their likelihood to respond to CBT. Identifying patients who are more likely to respond to CBT may aid in shared patient-clinician decision-making, increase patient satisfaction, and improve efficacy in clinical trials by enabling more effective assessment of treatment efficacy within unique prognostic subgroups.
Methods
This study presents the secondary analysis of data from a single-arm longitudinal clinical study assessing individualized factors associated with tinnitus burden. The primary analyses were preregistered (https://osf.io/uq9c5/registrations) and will be reported separately. This study was reviewed and approved by the Washington University Human Research Protection Office, and signed informed consent was obtained from all study participants. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
Study Population
Study participants were recruited from outpatient clinics at Washington University Department of Otolaryngology. A total of 88 participants were enrolled from September 2019 to February 2023. Eligibility criteria included individuals from 20 to 70 years old, subjective tinnitus duration of 6 months or greater, and substantial tinnitus-related distress. Participants were excluded if the tinnitus was related to cochlear implantation, retrocochlear lesion, or Meniere disease; if they reported or had a history of acute or chronic unstable psychiatric or medical condition or preexisting depression; or if they had a prior trial of CBT for tinnitus treatment.
Variables
Baseline patient demographic characteristics (age, gender, race and ethnicity), tinnitus description, and hearing characteristics were collected; demographic information was self-reported. Audiogram results were not obtained, and objective measures (eg, pure-tone average and word recognition score) were not available. The following patient surveys were administered: Tinnitus Functional Index (TFI),19 Global Bother Score (GBS),20,21 and Patient-Reported Outcomes Measurement Information System (PROMIS).22
Tinnitus Functional Index
The TFI is a validated 25-item questionnaire with 8 intended domains: intrusiveness, sense of control, sleep, auditory, relaxation, quality of life, and emotional distress resulting from tinnitus severity.19 Each item is rated on a 10-point scale. The total score is scored between 0 and 100; the raw score is divided by 25 and multiplied by 10. The minimal clinically important difference has been determined to be a change of 13 points.23
Global Bother Score
The degree of tinnitus bother was measured using the GBS, a derivation of the Clinical Global Impression Scale.20 A 5-point scale ranging from not bothered to extremely bothered allows for the classification of severity degree and degree of bother. The GBS has been determined to be a valid and reliable treatment outcome measure.21
Patient-Reported Outcomes Measurement Information System
To explore the psychosocial impact of tinnitus, participants completed US National Institute of Health PROMIS survey including the following domains: sleep, anxiety, emotional support, anger, social isolation, social participation, fatigue, sleep-related impairment, and general life.22 The PROMIS is a collection of PROMIS questionnaire items that captures information about different aspects of quality of life. Standardized T scores for each domain listed are computed with a mean (SD) score of 50 (10).
Cognitive Behavioral Therapy Sessions
Patients were enrolled in a primarily online group CBT program provided by the Department of Psychiatry at Washington University School of Medicine in St Louis. Overall, the treatment is based on exploring the relationship between an individual’s thoughts, feelings, and behaviors in relation to the tinnitus symptoms they are experiencing. The program was adapted from a CBT manual with the assistance of the manual’s authors.24 Relaxation techniques are the primary skill taught.
The CBT curriculum was based on 8 sessions completed during 10 weeks that totaled 20 hours of CBT participation. eTable 1 in Supplement 1 outlines the session-by-session content of the CBT curriculum. To ensure compliance, study members monitored class attendance.
Classification of Data
Responders to CBT
The primary outcome measure was response to CBT as measured by the TFI. For the purposes of this study, response to tinnitus was defined as a decrease of 13 or more points from the preintervention TFI survey score to the postintervention TFI survey score.
Degree of Tinnitus Bother
The degree of tinnitus bother was measured using the GBS. Participants were categorized as having mild (1-2 points), moderate (3 points), or severe (4-5 points) tinnitus based on their preintervention GBS survey score.
Psychosocial Impacts of Tinnitus
To explore the psychosocial impact of tinnitus, participants completed the PROMIS domains. Participants were categorized based on each measure’s median T-score cut point with moderate to severe classified as 1SD from the median; mild classified as within 1 SD from the median; and none classified as less than the median.25
Statistical Analysis
Descriptive statistics were used to explore the characteristics of the study population and summarize the distribution of characteristics for responders and nonresponders. Univariable logistic regression was used to identify demographic and tinnitus characteristics associated with response to CBT. Clinically and statistically significant prognostic variables were included in a multivariable logistic regression model, and then independently significant variables were combined in a cluster-type technique known as conjunctive consolidation.26
Conjunctive Consolidation
Conjunctive consolidation is a type of cluster analysis used to group categories of variables based on their clinical and statistical similarities.26,27 The overall goal is to create meaningful clusters that can inform clinical decision-making, improve the understanding of complex relationships between variables, and enhance the development of predictive models or classification systems. The process begins with creating a correlation matrix table with the discrete categories of the 2 variables under analysis. Each variable’s effect on the outcome of interest is reported separately, allowing for the identification and visualization of how each variable contributes to the overall outcome. One or more categories of the 2 separate variables are then conjoined to form combined categories or cells. The rate of the outcome of interest is calculated within each defined conjoined cell or group, where a conjoined cell represents a specific combination of categories of the 2 variables included in the conjunctive consolidation. Cells are examined for clinical and statistical similarities; clinical similarity might involve shared characteristics such as symptom patterns, whereas statistical similarity refers to similar outcome rates or statistical measures. Cells that exhibit both clinical and statistical similarities are consolidated into larger, more comprehensive groups, helping to simplify the data structure and highlight meaningful patterns. The grouped effects and consolidated cells are used to develop staging or classification systems based on patient or symptom characteristics, allowing for more tailored and precise categorization.
Quantitative Evaluation of New Predictive Classification System
Quantitative evaluation was performed on the variables included within conjunctive consolidation to evaluate the predictive and discriminative ability of each stage within the classification system.25,27 Monotonicity of the survival gradient refers to the unidirectional decrease in response rates across cells. The range of overall response gradient refers to the difference between the strata of the classification with the highest and lowest response rate. Predictive reduction in error is a statistical criterion measured using the λ coefficient that quantifies the ability of 1 variable to predict another. The coefficient of determination (R2) is a statistical measure that determines the proportion of the variance in the dependent variable that can be explained by the independent variables within a regression model.28,29 The χ2 for linear trend quantitatively assesses the linear trend within cells.25 The C statistic measures the discriminative ability of classification system, and it is equivalent to the area under the ROC (receiver operating characteristics) curve.30
Statistical analysis was performed using SPSS, version 29.0 (IBM Corp). Variables with an α ≤ .1 were selected to be included in the multivariate logistic regression model. All other variables were assessed at an α = .05 and calculated with a confidence interval.
Results
This study included 88 participants (median [IQR] age, 59 [49-66] years; 47 [53%] were female and 41 [47%] were male), of whom 53 (60%) responded to CBT as a treatment for tinnitus. Demographic and baseline tinnitus information of the total study population and treatment responders are shown in Table 1; this information was self-reported. The psychosocial effects of tinnitus, measured by the PROMIS domains, of the total study population are shown in Table 2.
Table 1. Response to Cognitive Behavioral Therapy (CBT) as a Function of Baseline Participant Characteristics.
| Characteristic | No. (%) | Proportion difference, % (95% CI)b | |
|---|---|---|---|
| Participants | CBT respondersa | ||
| Participants | 88 (100) | 53 (60) | NA |
| Age, y | |||
| ≤50 | 23 (26) | 17 (74) | 19 (−5 to 37) |
| >50 | 65 (74) | 36 (55) | |
| Gender | |||
| Female | 47 (53) | 30 (64) | 9 (−11 to 28) |
| Male | 40 (46) | 22 (55) | |
| Nonbinary | 1 (1) | 1 (100) | NA |
| Racec | |||
| White | 77 (91) | 45 (58) | 17 (−19 to 38) |
| Otherd | 8 (9) | 6 (75) | |
| Duration of tinnitus, ye | |||
| ≤5 | 33 (39) | 18 (55) | 16 (9-38) |
| 6-15 | 27 (32) | 17 (63) | |
| ≥16 | 24 (29) | 17 (71) | |
| Loud noise exposure | |||
| No | 32 (36) | 21 (66) | 8 (−13 to 28) |
| Yes | 56 (64) | 32 (57) | |
| Previous hearing aid use | |||
| No | 80 (91) | 48 (60) | 3 (−31 to 29) |
| Yes | 8 (9) | 5 (63) | |
| Current hearing aid use | |||
| No | 84 (95) | 52 (62) | 37 (−9 to 60) |
| Yes | 4 (5) | 1 (25) | |
| Tinnitus onset | |||
| Sudden | 26 (30) | 14 (54) | 9 (−12 to 30) |
| Gradual | 62 (70) | 39 (63) | |
| Frequency of tinnitus | |||
| Constant | 65 (74) | 38 (58) | 7 (−16 to 27) |
| Fluctuating | 23 (26) | 15 (65) | |
| Tinnitus loudness | |||
| Mild | 27 (31) | 13 (48) | 25 (−10 to 49) |
| Moderate | 50 (57) | 32 (64) | |
| Severe | 11 (13) | 8 (73) | |
| Tinnitus interference with sleep | |||
| Yes | 55 (63) | 38 (69) | −24 (2.5 to 42) |
| No | 33 (37) | 15 (45) | |
| Effort to ignore tinnitus | |||
| Slight | 64 (73) | 36 (56) | 26 (−6 to 43) |
| Moderate | 13 (15) | 8 (62) | |
| Substantial | 11 (13) | 9 (82) | |
| Global bother score | |||
| Mild | 10 (11) | 2 (20) | 55 (19 to 74) |
| Moderate | 50 (57) | 30 (60) | |
| Severe | 28 (32) | 21 (75) | |
Response to CBT (ie, responder) was predefined as a 13-point or greater decrease in the Tinnitus Functional Index survey score.
Proportion difference (95% CI) column displays the difference in the proportion of responders between the category with the highest response rate and the category with the lowest response rate, along with the corresponding 95% CI around the difference.
Missing values, n = 3.
Other includes Asian, Black, more than 1 race, and unknown.
Missing values, n = 4.
Table 2. Response to Cognitive Behavior Therapy (CBT) as a Function of Baseline PROMIS Domain.
| Baseline PROMIS domaina | No. (%) | Proportion difference, % (95% CI)c | |
|---|---|---|---|
| Participants | CBT respondersb | ||
| Participants | 88 (100) | 53 (60) | NA |
| Anxiety | |||
| None to mild | 72 (82) | 38 (53) | 41 (16 to 53) |
| Moderate to severe | 16 (18) | 15 (94) | |
| Depression | |||
| None to mild | 78 (89) | 46 (59) | 11 (−21 to 33) |
| Moderate to severe | 10 (11) | 7 (70) | |
| General life satisfaction | |||
| Low to average | 10 (11) | 7 (70) | 11 (−21 to 33) |
| High to very high | 78 (89) | 46 (59) | |
| Sleep disturbance | |||
| None to mild | 72 (82) | 41 (57) | 18 (−9 to 37) |
| Moderate to severe | 16 (18) | 12 (75) | |
| Sleep-related impairment | |||
| None to mild | 66 (75) | 35 (53) | 29 (5 to 45) |
| Moderate to severe | 22 (25) | 18 (82) | |
| Fatigue | |||
| None to mild | 77 (88) | 43 (56) | 35 (4 to 48) |
| Moderate to severe | 11 (12) | 10 (91) | |
| Social isolation | |||
| None to mild | 84 (95) | 51 (61) | 11 (26 to 47) |
| Moderate to severe | 4 (5) | 2 (50) | |
| Anger | |||
| None to mild | 82 (93) | 49 (60) | 7 (−31 to 33) |
| Moderate to severe | 6 (7) | 4 (67) | |
| Emotional support | |||
| Low to average | 7 (8) | 6 (86) | 28 (−11 to 44) |
| High to very high | 81 (92) | 47 (58) | |
Abbreviation: PROMIS, Patient-Reported Outcomes Measurement Information System.
The cut point was defined as 1 SD from population median.
Response to CBT (ie, responder) was predefined as a 13-point or greater decrease in the Tinnitus Functional Index survey score.
Proportion Difference (95% CI) column displays the difference in the proportion of responders between the category with the highest response rate and the category with the lowest response rate, along with the corresponding 95% CI around the difference.
As the data in Table 1 and Table 2 show, response to CBT treatments was associated with several demographic and clinical characteristics in the univariate analysis. Tinnitus loudness, effort to ignore tinnitus, GBS, anxiety, depression, fatigue, and sleep impairment were sequentially entered into a multivariable logistic regression model. After controlling for other characteristics in the model (Table 3), individuals with severe level of bother from tinnitus had a higher odds of responding to CBT (aOR = 12.1; 95% CI, 1.5-98.4) and individuals with moderate to severe levels of anxiety had higher odds of responding to CBT (aOR = 3.3; 95% CI, 0.9-12.3). While these estimates are imprecise, the large effect sizes warranted further investigation.
Table 3. Multivariate Model of Predictors of Response.
| Variables | aOR (95% CI) |
|---|---|
| Loudness of tinnitus | |
| Mild | 1 [Reference] |
| Moderate | 2.47 (0.76-8.02) |
| Severe | 1.73 (0.19-15.38) |
| Effort to ignore tinnitus | |
| Slight | 1 [Reference] |
| Moderate | 0.32 (0.06-1.60) |
| Significant | 2.04 (0.25-16.58) |
| Global Bother Score | |
| Mild | 1 [Reference] |
| Moderate | 7.82 (1.25-49.18) |
| Severe | 12.08 (1.48-98.35) |
| Baseline PROMIS Domain a | |
| Anxiety | |
| None to mild | 1 [Reference] |
| Moderate to severe | 3.33 (.90-12.30) |
| Depression | |
| None to mild | 1 [Reference] |
| Moderate to severe | 0.39 (0.10-1.52) |
| Fatigue | |
| None to mild | 1 [Reference] |
| Moderate to severe | 2.08 (0.59-7.37) |
| Sleep-related impairment | |
| None to mild | 1 [Reference] |
| Moderate to severe | 3.39 (0.84-13.62) |
Abbreviations: aOR, adjusted odds ratio; PROMIS, Patient-Reported Outcomes Measurement Information System.
Cut point was defined as 1 SD from population median.
Functional and Clinical Severity Classification
The conjunction of GBS (severe, moderate, and mild) and anxiety level (none, mild, and moderate to severe) is shown in the Figure. The 28 participants (31.8%) with severe bother from tinnitus had a higher rate of treatment response (75.0%) than the 49 participants (55.9%) with moderate bother (59.0%) and the 10 participants (11.4%) with mild bother (20.0%). The 16 participants (18.2%) with moderate to severe anxiety had a higher rate of treatment response (94.0%) than the 54 participants (61.4%) with mild anxiety (54.0%) and the 17 with no anxiety (47.0%).
Figure. Response to Cognitive Behavioral Therapy as a Function of Tinnitus Bother and Anxiety Level.
aDenominators are the number of patients in each category, and numerators are the corresponding number of cognitive behavioral therapy responders.
GBS was conjoined with anxiety levels to create the Functional Severity classification system according to response rates. The 16 individuals with moderate to severe degree of bother and moderate to severe anxiety had the highest rate of treatment response (94%), defining the “I” classification. The 22 individuals with severe bother and mild to no anxiety levels had a 72% response rate, defining the “II” classification. The 39 individuals with moderate bother and mild or no anxiety levels had a 49% response rate, defining the “III” classification. The 10 individuals with mild bother, regardless of anxiety levels, had the lowest treatment response rate (20%), defining the “IV” classification. The resulting gradient between group I and IV is a 74% difference in response rates.
Several additional variables, including age, gender, sleep disturbance, depression, anger, and tinnitus loudness and duration, were included in exploratory conjunctive consolidation but did not yield an increased gradient.
Quantitative Performance of New Predictive Classification System
The predictive and discriminative ability of the final staging system was quantitatively evaluated (Table 4). The final Functional Severity classification system displayed monotonicity of response rate with a linear trend when moving from stage I to III, with a response rate gradient of 74%, and χ2 for linear trend of 20.0. The classification system demonstrated good predictive ability by reducing the error in prediction of response by 45% and explaining about 28% of the variability in treatment response (λ = 0.45; R2 = 0.28;). The classification system had good discriminative ability with C statistic of 0.75 (95% CI, 0.65-0.85). The testing characteristics of the Functional Severity classification system include a specificity of 80%, sensitivity of 60%, a positive predictive value of 0.82, and a negative predictive value of 0.57.
Table 4. Quantitative Evaluation of Predictive Classification Systems.
| Classification | Anxiety | Tinnitus bother | Functional severity |
|---|---|---|---|
| Monotonicity of response gradient | Yes | Yes | Yes |
| Range of overall response gradient, % | 47 | 55 | 74 |
| Predictive reduction in error, % | 29 | 31 | 45 |
| Proportionate reduction in variance | .11 | .13 | .28 |
| χ2 for Linear trend | 7.6 | 8.5 | 20.0 |
| C statistic (95% CI) | .65 (.53-.76) | .65 (.54-.77) | .75 (.65-.85) |
To internally validate our Functional Severity classification system, split-sample validation was performed.31 A random sample of 50% of the study participants was selected and conjunctively consolidated with the same classification system. The quantitative evaluation of the internal validation is shown in eTable 2 in Supplement 1.
Discussion
This study identified individuals with severe tinnitus bother and moderate to high anxiety levels as the most likely to respond to CBT. Tinnitus bother and anxiety levels were combined to create a prediction index, with the highest responders being individuals with extreme tinnitus bother and moderate to high anxiety levels. Patients with mild tinnitus bother, regardless of anxiety level, showed the lowest responsiveness to CBT. These findings underscore the importance of considering both tinnitus-specific and psychosocial factors in treatment planning. The prediction system enhances the ability for tailoring interventions based on individual patient characteristics, specifically for patients likely to respond to CBT. In contrast, the relatively low negative predictive value suggests the prediction system is less likely to correctly identify nonresponders. That is, the system successfully identifies people who should be strongly encouraged toward CBT, but does not provide a basis for discouraging patients from pursuing it.
Medical literature is replete with studies demonstrating the efficacy of CBT in decreasing tinnitus distress and bother.11,32 In a meta-analysis of 19 randomized trials, Landry et al32 found improvement in tinnitus-related quality of life, depression, and anxiety after CBT treatment compared to no intervention. As a result, the American Academy of Otolaryngology–Head and Neck Surgery released management guidelines for tinnitus and recommended CBT as treatment for bothersome tinnitus with a duration of more than 6 months.17 A unique internet-based delivery of CBT was investigated by Beukes et al11 in a noninferiority trial of 44 patients and demonstrated a 57% response rate. Although CBT has demonstrated effectiveness for a variety of psychological conditions and is well-reported in the medical literature, no study to date has identified predictors of CBT treatment response in patients with tinnitus.
Many patients who receive CBT benefit from associated reductions in psychological distress. However, a substantial minority of patients experience no benefit; thus, further investigation into predictors of response is warranted. In a prospective cohort of 228 individuals being treated with CBT for tinnitus, patient characteristics including increased baseline symptom severity and education level were associated with response.18,33
Predictors of CBT response have been investigated for treatment in other disorders, including depression and anxiety. In a multicenter cohort study of pediatric anxiety disorders, increased symptom severity and presence of comorbid conditions decreased the likelihood of CBT response by half.34 The association of increased symptom severity and worse CBT outcomes has been replicated in treatment of depression and anxiety.35,36 There are conflicting results in the literature regarding the association between baseline symptom severity and response to CBT across conditions. However, baseline patient motivation consistently predicts treatment response to CBT.37,38,39 Higher rates of CBT response were identified in patients with higher treatment motivation and more positive expectations.40,41 In a randomized clinical trial of 157 participants, Alfonsson et al38 found that individuals with higher motivation experienced a statistically significant reduction in stress symptoms after CBT treatment. Despite the heterogeneity of treatment response as reported within the published medical literature, patients experiencing heightened distress are likely to be more motivated to engage actively in therapeutic strategies aimed at alleviating symptoms.
The Functional Severity classification system predicts patient response to CBT based on tinnitus bother and anxiety levels. The system may partly reflect the motivation for tinnitus relief among the study participants. The distress associated with tinnitus can act as a powerful motivator, driving individuals to actively participate in CBT exercises, cognitive restructuring, and relaxation techniques. Alternatively, individuals with mild tinnitus bother may not perceive the same level of urgency or necessity to engage fully in the therapeutic process. The lack of tinnitus distress may result in reduced motivation to invest time and effort in the recommended CBT interventions. Consequently, this diminished motivation may contribute to this study’s observed lower responsiveness to CBT in individuals with mild tinnitus bother, even when anxiety levels are elevated. The combination of tinnitus bother and anxiety levels translates to the degree of patient motivation, and therefore, response to CBT.
This Functional Severity classification system, when validated in a larger set of patients, can be used by clinicians and researchers. Physicians are often confronted with the need to make treatment recommendations for patients with tinnitus. This classification system suggests that physicians should highly recommend CBT treatment for patients with bothersome tinnitus and high anxiety. However, patients with lower levels of bother and anxiety should not be discouraged from pursuing CBT based on the testing characteristics of this system. Future researchers may use this classification system to improve the conduct of clinical research when designing clinical trials investigating tinnitus treatments. By incorporating a nuanced understanding of clinical tinnitus characteristics into trial protocols, researchers can enhance the precision and effectiveness of interventions, ultimately advancing the field of tinnitus management.
Limitations
This study has several limitations. This study population was mostly composed of White participants (91%), reflecting the patient population of a metropolitan city in the US Midwest region; this may limit its generalizability to rural or community-based settings. The relatively small sample size is associated with imprecise prognostic estimates and prevents definitive conclusions. Finally, although internal validation was performed, external validation was not performed and needs to be completed to fully assess this classification system’s performance.
Conclusions
This secondary cohort analysis of a single-arm clinical study identified extreme tinnitus bother and moderate to high anxiety levels as predictors of response to CBT for tinnitus. After it has been externally validated, the Functional Severity classification system may be used as a guiding framework for tinnitus treatment recommendations and tinnitus research. This classification system offers insight into providing CBT treatment counseling to patients with tinnitus and designing future randomized clinical trials.
eTable 1. CBT Curriculum Session-by-Session Content
eTable 2. Internal Validation Quantitative Evaluation
eFigure. Model Performance in Predicting Response to CBT
Data Sharing Statement
References
- 1.Piccirillo JF, Rodebaugh TL, Lenze EJ. Tinnitus. JAMA. 2020;323(15):1497-1498. doi: 10.1001/jama.2020.0697 [DOI] [PubMed] [Google Scholar]
- 2.Baguley D, McFerran D, Hall D. Tinnitus. Lancet. 2013;382(9904):1600-1607. doi: 10.1016/S0140-6736(13)60142-7 [DOI] [PubMed] [Google Scholar]
- 3.Bauer CA. Tinnitus. N Engl J Med. 2018;378(13):1224-1231. doi: 10.1056/NEJMcp1506631 [DOI] [PubMed] [Google Scholar]
- 4.Kennedy V, Wilson C, Stephens D. Quality of life and tinnitus. Audiol Med. 2004;2(1):29-40. doi: 10.1080/16513860410027349 [DOI] [Google Scholar]
- 5.Zeman F, Koller M, Langguth B, Landgrebe M; Tinnitus Research Initiative database study group . Which tinnitus-related aspects are relevant for quality of life and depression: results from a large international multicentre sample. Health Qual Life Outcomes. 2014;12(1):7. doi: 10.1186/1477-7525-12-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Batts S, Stankovic KM. Tinnitus prevalence, associated characteristics, and related healthcare use in the United States: a population-level analysis. Lancet Reg Health Am. 2024;29:100659. doi: 10.1016/j.lana.2023.100659 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.United States Department of Veterans Affairs . Annual Benefits Report Fiscal Year 2022. Accessed June 27, 2024. https://www.benefits.va.gov/REPORTS/abr/
- 8.Rademaker MM, Stegeman I, Hooiveld M, Stokroos RJ, Smit AL. Patients with tinnitus use more primary healthcare compared to people without tinnitus. Sci Rep. 2021;11(1):17299. doi: 10.1038/s41598-021-96607-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tyler RS. Patient preferences and willingness to pay for tinnitus treatments. J Am Acad Audiol. 2012;23(2):115-125. doi: 10.3766/jaaa.23.2.6 [DOI] [PubMed] [Google Scholar]
- 10.Simonetti P, Vasconcelos LG, Gândara MR, Lezirovitz K, Medeiros ÍRT, Oiticica J.. Hearing aid effectiveness on patients with chronic tinnitus and associated hearing loss. Braz J Otorhinolaryngol. 2022;88(Suppl 3):S164-S170. doi: 10.1016/j.bjorl.2022.03.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Beukes EW, Andersson G, Allen PM, Manchaiah V, Baguley DM. Effectiveness of guided internet-based cognitive behavioral therapy vs face-to-face clinical care for treatment of tinnitus: a randomized clinical trial. JAMA Otolaryngol Head Neck Surg. 2018;144(12):1126-1133. doi: 10.1001/jamaoto.2018.2238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Robinson SK, Viirre ES, Bailey KA, et al. A randomized controlled trial of cognitive-behavior therapy for tinnitus. Int Tinnitus J. 2008;14(2):119-126. [PubMed] [Google Scholar]
- 13.Roland LT, Lenze EJ, Hardin FM, et al. Effects of mindfulness based stress reduction therapy on subjective bother and neural connectivity in chronic tinnitus. Otolaryngol Head Neck Surg. 2015;152(5):919-926. doi: 10.1177/0194599815571556 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Philippot P, Nef F, Clauw L, de Romrée M, Segal Z. A randomized controlled trial of mindfulness-based cognitive therapy for treating tinnitus. Clin Psychol Psychother. 2012;19(5):411-419. doi: 10.1002/cpp.756 [DOI] [PubMed] [Google Scholar]
- 15.Westin VZ, Schulin M, Hesser H, et al. Acceptance and commitment therapy versus tinnitus retraining therapy in the treatment of tinnitus: a randomised controlled trial. Behav Res Ther. 2011;49(11):737-747. doi: 10.1016/j.brat.2011.08.001 [DOI] [PubMed] [Google Scholar]
- 16.Hesser H, Weise C, Westin VZ, Andersson G. A systematic review and meta-analysis of randomized controlled trials of cognitive-behavioral therapy for tinnitus distress. Clin Psychol Rev. 2011;31(4):545-553. doi: 10.1016/j.cpr.2010.12.006 [DOI] [PubMed] [Google Scholar]
- 17.Tunkel DE, Bauer CA, Sun GH, et al. Clinical practice guideline: tinnitus. Otolaryngol Head Neck Surg. 2014;151(2)(suppl):S1-S40. [DOI] [PubMed] [Google Scholar]
- 18.Rodrigo H, Beukes EW, Andersson G, Manchaiah V. Internet-based cognitive-behavioural therapy for tinnitus: secondary analysis to examine predictors of outcomes. BMJ Open. 2021;11(8):e049384. doi: 10.1136/bmjopen-2021-049384 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chandra N, Chang K, Lee A, Shekhawat GS, Searchfield GD. Psychometric validity, reliability, and responsiveness of the Tinnitus Functional Index. J Am Acad Audiol. 2018;29(7):609-625. doi: 10.3766/jaaa.16171 [DOI] [PubMed] [Google Scholar]
- 20.Busner J, Targum SD. The clinical global impressions scale: applying a research tool in clinical practice. Psychiatry (Edgmont). 2007;4(7):28-37. [PMC free article] [PubMed] [Google Scholar]
- 21.O’Leary MP. Validity of the “bother score” in the evaluation and treatment of symptomatic benign prostatic hyperplasia. Rev Urol. 2005;7(1):1-10. [PMC free article] [PubMed] [Google Scholar]
- 22.Bevans M, Ross A, Cella D. Patient-Reported Outcomes Measurement Information System (PROMIS): efficient, standardized tools to measure self-reported health and quality of life. Nurs Outlook. 2014;62(5):339-345. doi: 10.1016/j.outlook.2014.05.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Meikle MB, Henry JA, Griest SE, et al. The Tinnitus Functional Index: development of a new clinical measure for chronic, intrusive tinnitus. Ear Hear. 2012;33(2):153-176. doi: 10.1097/AUD.0b013e31822f67c0 [DOI] [PubMed] [Google Scholar]
- 24.Beukes EW, Andersson G, Manchaiah V, Kaldo V. Cognitive behavioral therapy for tinnitus. Plural Publishing; 2020. [Google Scholar]
- 25.HealthMeasures. PROMIS Score Cut Points. 2024. Accessed June 28, 2024. https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points
- 26.Feinstein AR. Multivariable Analysis: An Introduction. Yale University Press; 1996. [Google Scholar]
- 27.Neely JG, Lieu JE, Sequeira SM, et al. Practical guide to understanding multivariable analyses, part b: conjunctive consolidation. Otolaryngol Head Neck Surg. 2013;148(3):359-365. doi: 10.1177/0194599812471500 [DOI] [PubMed] [Google Scholar]
- 28.Piccirillo JF, Wells CK, Sasaki CT, Feinstein AR. New clinical severity staging system for cancer of the larynx: five-year survival rates. Ann Otol Rhinol Laryngol. 1994;103(2):83-92. doi: 10.1177/000348949410300201 [DOI] [PubMed] [Google Scholar]
- 29.Austin PC, Steyerberg EW. Interpreting the concordance statistic of a logistic regression model: relation to the variance and odds ratio of a continuous explanatory variable. BMC Med Res Methodol. 2012;12(1):82. doi: 10.1186/1471-2288-12-82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Chicco D, Warrens MJ, Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Comput Sci. 2021;7:e623. doi: 10.7717/peerj-cs.623 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Steyerberg E. Clinical Prediction Models: A Practical Approach to Development, Validation, and Updating. Springer; 2009. doi: 10.1007/978-0-387-77244-8 [DOI] [Google Scholar]
- 32.Landry EC, Sandoval XCR, Simeone CN, Tidball G, Lea J, Westerberg BD. Systematic review and network meta-analysis of cognitive and/or behavioral therapies (CBT) for tinnitus. Otol Neurotol. 2020;41(2):153-166. doi: 10.1097/MAO.0000000000002472 [DOI] [PubMed] [Google Scholar]
- 33.Rodrigo H, Beukes EW, Andersson G, Manchaiah V. Predicting the outcomes of internet-based cognitive behavioral therapy for tinnitus: applications of artificial neural network and support vector machine. Am J Audiol. 2022;31(4):1167-1177. doi: 10.1044/2022_AJA-21-00270 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hudson JL, Keers R, Roberts S, et al. Clinical predictors of response to cognitive-behavioral therapy in pediatric anxiety disorders: the genes for treatment (GxT) study. J Am Acad Child Adolesc Psychiatry. 2015;54(6):454-463. doi: 10.1016/j.jaac.2015.03.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Carter JD, Crowe MT, Jordan J, McIntosh VV, Frampton C, Joyce PR. Predictors of response to CBT and IPT for depression; the contribution of therapy process. Behav Res Ther. 2015;74:72-79. doi: 10.1016/j.brat.2015.09.003 [DOI] [PubMed] [Google Scholar]
- 36.Sunderland M, Wong N, Hilvert-Bruce Z, Andrews G. Investigating trajectories of change in psychological distress amongst patients with depression and generalised anxiety disorder treated with internet cognitive behavioural therapy. Behav Res Ther. 2012;50(6):374-380. doi: 10.1016/j.brat.2012.03.005 [DOI] [PubMed] [Google Scholar]
- 37.Treasure JL, Katzman M, Schmidt U, Troop N, Todd G, de Silva P. Engagement and outcome in the treatment of bulimia nervosa: first phase of a sequential design comparing motivation enhancement therapy and cognitive behavioural therapy. Behav Res Ther. 1999;37(5):405-418. doi: 10.1016/S0005-7967(98)00149-1 [DOI] [PubMed] [Google Scholar]
- 38.Alfonsson S, Olsson E, Hursti T. Motivation and treatment credibility predicts dropout, treatment adherence, and clinical outcomes in an internet-based cognitive behavioral relaxation program: a randomized controlled trial. J Med Internet Res. 2016;18(3):e52. doi: 10.2196/jmir.5352 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Lombardi DR, Button ML, Westra HA. Measuring motivation: change talk and counter-change talk in cognitive behavioral therapy for generalized anxiety. Cogn Behav Ther. 2014;43(1):12-21. doi: 10.1080/16506073.2013.846400 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Kodal A, Fjermestad KW, Bjelland I, et al. Predictors of long-term outcome of CBT for youth with anxiety disorders treated in community clinics. J Anxiety Disord. 2018;59:53-63. doi: 10.1016/j.janxdis.2018.08.008 [DOI] [PubMed] [Google Scholar]
- 41.Cowan MJ, Freedland KE, Burg MM, et al. ; ENRICHD Investigators . Predictors of treatment response for depression and inadequate social support–the ENRICHD randomized clinical trial. Psychother Psychosom. 2008;77(1):27-37. doi: 10.1159/000110057 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. CBT Curriculum Session-by-Session Content
eTable 2. Internal Validation Quantitative Evaluation
eFigure. Model Performance in Predicting Response to CBT
Data Sharing Statement