Psychometric properties of the Italian Tinnitus Functional Index (TFI)

Abstract

Various questionnaires are used to assess the impact of tinnitus on the quality of life. The Tinnitus Functional Index (TFI) has excellent properties for scaling the severity of tinnitus and treatment-related changes in both clinical and research settings. The aim of this study was to evaluate the psychometric properties of the Italian version of the TFI with particular emphasis on factor analysis, internal consistency, reliability and validity. The original English version of the TFI was translated into Italian using the translation/back-translation process; 137 participants who were recruited at the Tinnitus Clinic in Milan and had suffered from tinnitus for at least three months (39.4% females, age: 18-80 years, mean age: 48.26, SD: 14.08) completed the Italian version of the TFI, the Tinnitus Handicap Inventory (THI), the Beck Depression Inventory-Primary Care Version (BDI-PC) and the Numeric Rating Scale of annoyance (NRS-A). Of these patients, 57 completed the TFI again at a second visit 7-14 days later, before undergoing any intervention, in order to provide data for reproducibility assessment. The psychometric properties were investigated using exploratory factor analysis and internal consistency and test-retest reliability instruments. The convergent validity of the TFI was evaluated using correlation coefficients obtained from the remaining measurements. The Italian TFI has a four-factor structure that was somewhat different from the original. The internal consistency proved to be good (0.92 ≤ α ≤ 0.96) as did the test-retest reliability (0.79 ≤ α ≤ 0.85). In terms of convergent validity, the TFI showed high correlations with the THI (r = 0.77) and the NRS-A (r = 0.70) scores, and moderate correlations with the BDI-PC scores (r = 0.46). The difficulties encountered when attempting to reproduce the original eight-factor structure were consistent with other studies in which the TFI was translated into European languages. In spite of this, the factorial structure of the Italian version of the TFI was characterised by high levels of reliability and validity. Overall, the Italian adaptation of the TFI was shown to be suitable to measure the impact of tinnitus on the daily lives of individuals.

Correlation Between Chronic Tinnitus Distress and Symptoms of Depression: A Systematic Review

“…The tinnitus distress was assessed by using different multi-item questionnaires as listed as follows: The THI was used in 20 out of 33 studies ( 42 , 52 – 54 , 56 , 58 , 60 – 66 , 68 – 70 , 74 ), TFI by seven ( 45 , 46 , 50 , 52 , 55 , 57 , 68 ), TQ by five ( 44 , 48 , 51 , 56 , 73 ), TRQ by two ( 49 , 71 ), mini-TQ by two ( 59 , 72 ), and both THQ ( 66 ) and TPFQ by one study ( 66 ). The depressive symptoms were also assessed by using different multi-item questionnaires.…”

“…The depressive symptoms were also assessed by using different multi-item questionnaires. The BDI was used in 15 studies ( 42 , 52 – 54 , 60 – 63 , 65 , 66 , 68 , 69 , 74 ), HADS-D was used by seven studies ( 43 – 46 , 59 , 71 , 72 ), CES-D by four studies ( 16 , 51 , 55 , 73 ), SDS by three studies ( 47 , 50 , 58 ), SCL-90R ( 48 , 56 ). The MMPI and the PHQ-9 were both used once in two different studies ( 49 , 57 ) ( Tables 4–6 ).…”

“…The 15 out of 33 studies (45%) used the BDI to report depression severity ( Table 3 ) ( 42 , 52 – 54 , 60 – 66 , 68 – 70 ). Correlations between measures of tinnitus distress and depression severity were calculated using Pearson's test in 9 out of 15 studies ( 42 , 52 – 54 , 63 , 65 – 68 ), Spearman's tests in four out of 15 studies ( 60 , 62 , 69 , 70 ) and univariate regressions analysis in two of 15 studies ( 61 , 64 ).…”

“…The 15 out of 33 studies (45%) used the BDI to report depression severity ( Table 3 ) ( 42 , 52 – 54 , 60 – 66 , 68 – 70 ). Correlations between measures of tinnitus distress and depression severity were calculated using Pearson's test in 9 out of 15 studies ( 42 , 52 – 54 , 63 , 65 – 68 ), Spearman's tests in four out of 15 studies ( 60 , 62 , 69 , 70 ) and univariate regressions analysis in two of 15 studies ( 61 , 64 ). The 13 out of 15 studies reported a statistical significant positive correlation, which varied between 0.51 and 0.73 in the studies using Spearman correlation and between 0.46 and 0.7 in the studies using Pearson's correlation ( Table 3 ) ( 42 , 52 – 54 , 60 – 62 , 64 , 65 , 67 , 69 , 70 ).…”

“…For example, in the 15 studies who used the BDI, only 4 studies referred to the original BDI manual for interpretation of the score ( 53 , 54 , 60 , 61 ). Some studies provided different interpretations of scores using the same measure ( 63 , 65 , 67 , 69 ) and most studies did not provide a definition at all of a cut-off for a clinical relevant depressive symptom scores ( 42 , 52 , 62 , 66 , 68 , 69 ). The wide variety of questionnaires, definitions and cut-off values used to measure and define (symptoms of) depression could all contribute to the wide range of reported prevalence numbers of clinically relevant depressive symptom scores (4.6–41.7%).…”