Systematic review of outcome domains and instruments used in clinical trials of tinnitus treatments in adults

To develop a self-report tinnitus handicap measure that is brief, easy to administer and interpret, broad in scope, and psychometrically robust. A standardization study of a self-report tinnitus handicap measure was conducted to determine its internal consistency reliability and convergent and construct validity. Audiology clinics in tertiary care centers in two sites. In the first investigation, 84 patients reporting tinnitus as their primary complaint or secondary to hearing loss completed the 45-item alpha version of the Tinnitus Handicap Inventory (THI). In the second investigation, 66 subjects also reporting tinnitus completed the 25-item beta version. Convergent validity was assessed using another measure of perceived tinnitus handicap (Tinnitus Handicap Questionnaire). Construct validity was assessed using the Beck Depression Inventory, Modified Somatic Perception Questionnaire, symptom rating scales (annoyance, sleep disruption, depression, and concentration), and perceived tinnitus pitch and loudness judgments. From the alpha version of the THI, we derived a 25-item beta version with the items grouped into functional, emotional, and catastrophic subscales. The total scale yielded excellent internal consistency reliability (Cronbach's alpha = .93). No significant age or gender effects were seen. Weak correlations were observed between the THI and the Beck Depression Inventory, Modified Somatic Perception Questionnaire, and pitch and loudness judgments. Significant correlations were found between the THI and the symptom rating scales. The THI is a self-report measure that can be used in a busy clinical practice to quantify the impact of tinnitus on daily living.

There is widespread recognition that consistency between research centres in the ways that patients with tinnitus are assessed and outcomes following interventions are measured would facilitate more effective co-operation and more meaningful evaluations and comparisons of outcomes. At the first Tinnitus Research Initiative meeting held in Regensburg in July 2006 an attempt was made through workshops to gain a consensus both for patient assessments and for outcome measurements. It is hoped that this will contribute towards better cooperation between research centres in finding and evaluating treatments for tinnitus by allowing better comparability between studies.

The psychometric properties of a tinnitus handicap questionnaire are reported. There were two phases in this study. In Phase I, 87 questions were administered to 100 tinnitus patients. From their responses, 59 items that were either redundant, insensitive, or had low item-total correlations were eliminated. In Phase II, the resulting 27-item questionnaire was administered to 319 patients. Fifty-three of these patients also completed psychological and psychophysical measures that were used to validate the questionnaire. A factor analysis of patients' responses revealed a three-factor structure. These three factors appeared to reflect the physical, emotional, and social consequences of tinnitus (Factor 1), hearing ability of the patient (Factor 2), and the patients' view of tinnitus (Factor 3). Although the 27-item questionnaire had high internal consistency reliability and validity as reflected by correlations with life satisfaction and depression scales, it is recommended that only the items on the Factor 1 and the Factor 2 subscales be scored because of the low internal consistency reliability of the Factor 3 subscale. This questionnaire can be used to compare a patient's tinnitus handicap with the norm, identify specific areas of handicaps, and to monitor a patient's progress with particular treatment programs.