Introduction
Tinnitus is the subjective impression of sound without any external source. This condition can lead to unpleasant sensations and has a substantial impact on the quality of life for persons who are experiencing it [1]. Tinnitus is considered as a separate condition in International Classification of Functioning, Disability and Health (ICF 2001) [2p.69]. The classification of tinnitus is beneficial in the diagnosis procedure and helps find the best treatment regimen for this condition. Tinnitus is categorized into two main types: subjective and objective tinnitus. Objective tinnitus can be heard by others, in addition to the person experiencing it. It is associated with numerous causes ranging from those related to vascular problems, muscle spasms, and dysfunction of the Eustachian tube. Subjective tinnitus, on the other hand, has no connection with outward forces and can only be heard by a sufferer. Sometimes it is referred to as “head noise” [3]. Studies show that tinnitus is quite common condition and affects 10% to 15% of the global population [4]. Tinnitus is an annoying condition that significantly burdens the quality of life. It causes irritability, sadness, anxiousness, frustration, insomnia, difficulty focusing, and social retract [5].
The main goal in the treatment of tinnitus is to reduce its impact. A number of therapies can be employed to manage tinnitus, including counseling, sound therapy, and the utilization of hearing aids. Hence, conducting a thorough assessment of the influence of tinnitus on patients is a crucial component of tinnitus therapy [6]. Tinnitus questionnaires are valuable instruments for assessing the response of individuals with tinnitus to their condition. They have been utilized in various ways, both in therapeutic settings and as components of research investigations [7]. A variety of questionnaires have been used internationally, including the Tinnitus Handicap Questionnaire (THQ) [8], the Tinnitus Handicap Inventory (THI) [9], and the Tinnitus Primary Function Questionnaire (TPFQ) [10]. The TPFQ is a survey designed to assess the impact of tinnitus on various behavioral aspects, consisting of 20 items categorized into concentration, emotions, hearing, and sleep, with values ranging from 0 to 100 [10]. The translation of tinnitus questionnaires into multiple languages necessitates psychometric validation and comparative analysis of the questionnaire outcomes across these languages. The crosscultural translations assess the influence of tinnitus on the quality of life of patients from diverse national, regional, and cultural backgrounds [7]. However, the TPFQ is unavailable in Central Kurdish, a language spoken by millions worldwide. This lack of a culturally appropriate assessment tool hinders the accurate assessment of tinnitus burden and the provision of effective treatment for individuals in Central Kurdish communities. Developing a Central Kurdish version of the TPFQ (TPFQ-CK) will bridge this gap, enabling healthcare professionals to effectively assess tinnitus and provide culturally competent care to Kurdish individuals. The TPFQ was translated into Central Kurdish by the cross-cultural adaptation and translation method suggested by Beaton, et al. [11].
This study aimed to assess the reliability and validity of the TPFQ-CK. Initially, the TPFQ was translated from English to the Central Kurdish language, yielding the TPFQ-CK. Then, an additional evaluation was performed to determine the reliability and validity of the TPFQ-CK.
Subjects and Methods
Study design
A total of 205 participants were included in this cross-sectional study between April 2022 and April 2023. The Sulaymaniyah University Research Ethics Board granted ethics approval prior to subject recruitment (Ethical code 7/29-4758- 2 was established on April 18, 2022). Prior to commencing data collection, informed consent was obtained from all participants. After obtaining written consent from the authors of the TPFQ, the study was carried out following the principles of the Helsinki Declaration.
Translation and adaptation of TPFQ to Kurdish
The TPFQ was cross-culturally translated into a Kurdish version using the method described by Beaton, et al. [11]. In order to achieve this objective, we followed the subsequent steps: The initial stage involved the independent forward translation of the text into the Kurdish language by two bilingual native Kurdish-speaking individuals. One of the translators has professional translation expertise, while the other had a medical background. The second stage involved arranging a meeting between the two translators and the leader of the investigation to identify any discrepancies between the two translated versions. Subsequently, they reached a consensus on a single Kurdish version. The third phase involved the reverse translation of the translated TPFQ-CK by two proficient linguists who were fluent in both Kurdish and English languages and living in Sulaymaniyah, Iraq. They were unaware of the original English rendition of TPFQ-20. The assessment of discrepancies between the back-translated version of TPFQ-CK and the original TPFQ was conducted at this stage. The fourth phase was the formation of an expert committee, comprising our study group, 9 otolaryngologists, a methodologist, a Kurdish language specialist, and translators. Their task was to conduct a comparative analysis between the original TPFQ-20 and all translations and back translations of this questionnaire which were obtained from the previous stages and steps of translation and adaptation of TPFQ. This analysis aimed to develop the pre-final version of TPFQ-CK. The questionnaire items were individually assessed on a scale ranging from 0 (indicating improper translation) to 100 (indicating perfect translation). If each evaluator assigned a score below 80, the translation was deemed unsatisfactory and necessitated a new translation. Consensus was achieved for the translation of each item when the maximum score for each item surpassed 80. The fifth stage involved conducting a pre-testing or pilot study of the pre-final version of the Central Kurdish Questionnaire. It was performed on 54 participants, consisting of 35 individuals without any history of hearing loss or tinnitus and 19 individuals with tinnitus. The objective of this pilot study was to assess the transparency and flexibility of the questionnaire items. These people were excluded from the study. They indicated that they easily understood the questionnaire topics without any trouble and expressed their belief that the questionnaire covered significant aspects relevant to tinnitus. The pilot investigation revealed that the administration of the established transadapted TPFQ-CK took about 15–20 minutes. The sixth phase involved transcribing the previous procedures and analyzing the interviews conducted with the patients in order to compose the final version of TPFQ-CK. The Central-Kurdish version of TPFQ can be found in Supplementary Material (in the online-only Data Supplement). The final version of TPFQ-CK was subsequently given to 205 patients suffering from chronic tinnitus who met the specified criteria for inclusion and provided written informed consent to participate in this research.
On the day of participation, all participants of the study underwent a comprehensive assessment that included a full medical history, general physical examination, neurological examination, ear examination, and a basic audiological evaluation. Additionally, they completed the Central Kurdish version of the TPFQ-CK, which consists of 20 items that assess various aspects of tinnitus, specifically focusing on concentration, emotion, hearing, and sleep. The THQ was also administered. The scores ranged from 0 to 100. The participants were instructed to assign a score to a single number, with 0 indicating complete disagreement and 100 indicating perfect agreement. The examiner offered further clarification and explication of the questionnaire items for certain participants who has have a limited educational background or illiterate. Subsequently, we assessed the accuracy and validity of the Central Kurdish version by administering it to a group of Kurdish people who suffer from tinnitus, and verified its reliability by detailed statistical analysis.
Assessment of psychometric properties of TPFQ-CK
Participants
To determine the sample size, we used the minimum number of subjects per variable, which was 10 [12,13]. We estimated that 205 participants would be sufficient. The study comprised 205 individuals who experienced subjective tinnitus for 3 months or more. They were recruited from a publicly funded medical center and a private audiology clinic in Sulaymaniyah, Iraq. The TPFQ-CK and the THQ were administered to the participants, and a pure-tone audiometry (PTA) was performed. A test-retest was conducted 3 to 7 days after the initial assessment to ensure the reliability of measurements across different cultural contexts, contributing to the validity and trustworthiness of the instrument. The exclusion criteria included any cognitive or linguistic difficulties, objective tinnitus, and any underlying medical cause.
There is no published questionnaire specifically designed for tinnitus assessment in our language. For this reason, the THQ-Central Kurdish version was employed for comparison with the TPFQ-CK. Our THQ was undergoing the same process of cross-cultural adaptation as the TPFQ-CK, by using the methods described by Beaton, et al. [11]. But the results have not yet been published. Therefore, the name of THQ was used instead of THQ-CK for the time being.
Outcome measures
The TPFQ-CK questionnaire was completed by each participant in 15–20 minutes. The patients indicated that they easily understood the questionnaire topics without any trouble and expressed their belief that the questionnaire covered significant aspects relevant to tinnitus. The researchers were present while the questionnaires were being completed and assisted with any questions the participants had about its components.
Statistical analysis
The collected results were organized into a table and subjected to statistical analysis using a personal computer equipped with SPSS software (version 26; IBM Corp., Armonk, NY, USA). The reliability and validity of the TPFQ-CK questionnaire were examined to evaluate its cross-cultural adaption. The data were displayed in the form of mean, standard deviation (SD), expressed as numerical values and/or percentages. We computed Cronbach’s α coefficient, intraclass correlation coefficient (ICC), and test-retest reliability to evaluate the internal consistency of the measurements. The ICC is a numerical measure ranging from 0 to 1, any ICC result above 0.9 signifies outstanding reliability and below 0.5 signify weak reliability [14]. Construct validity of the TPFQ-CK was evaluated by conducting a Spearman’s correlation test between the TPFQ-CK score and THQ score. Spearman’s correlation can be used for assessing both test-retest reliability and construct validity [15,16]. Criterion validity is evaluated through the assessment of the concordance correlation coefficient (CCC), as well as the examination of ceiling and floor effects. Ceiling and floor effects were taken into account for the total TPFQ-CK score and subscales, where a score above 97 and below 3 was considered. The Bland-Altman plot is utilized to evaluate the agreement or concordance between the first and second assessments of TPFQ-CK and can be a valuable tool for detecting potential outliers in data comparing two quantitative measurement methods [17]. The values used to describe the correlations were categorized as weak if they were less than 0.3, adequate if they were between 0.5 and 0.7, and high if they were higher than 0.7 [18].
Results
Patients
A sample of 205 clients, consisting of 102 females and 103 males, were included. The average age of the patients was 47.43±14.32 years. The majority of the patients resided in urban areas (56.6%), were employed (34.1%), and were housewives (33.7%). Additionally, 22.4% were illiterate, and 51.7% had only non-academic education. Regarding previous occupations, 45.4% of participants worked in noisy environments. Tinnitus was reported as affecting the left ear in 35.6% of cases, the right ear in 31.2%, and both ears in 33.2%. The tinnitus was described as a ringing sound in 41.5% of cases and a whooshing sound in 22.9%. It was found that tinnitus increased in silence for 62.9% of clients, while only 12.2% experienced increased tinnitus in noisy environments. Additionally, tinnitus was continuous in 64.4% of cases. The duration of tinnitus was less than 1 year for 38.5% of clients, 1 to 2 years for 25.4%, and 3 to 5 years for 17.1%. Table 1 shows the frequency distribution of tinnitus characteristics.
TPFQ-CK scores’ summary
The mean total score of TPFQ-CK was 39.25±24.89 in the first and 38.71±24.72 in the second evaluation (mean difference: 0.54±0.37). The mean concentration factor score was 43.82±24.16 in the first evaluation and 43.14±24.02 in the second. The mean emotion factor score was 41.23±25.81 in the first evaluation and 40.71±25.69 in the second. The mean hearing factor score was 38.50±30.61 in the first evaluation and 38.02±30.25 in the second. The mean sleep factor score was 33.44±28.05 in the first evaluation and 32.96±27.70 in the second. The second evaluation resulted in very small changes in all subscale values (Table 2).
Reliability
Internal consistency
The internal consistency of the TPFQ-CK was assessed by computing Cronbach’s α coefficient. The Cronbach’s α coefficient for the TPFQ-CK total score was 0.933, indicating a high level of consistency. The Cronbach’s α scores for all factors were 0.951, 0.949, 0.959, and 0.960, respectively (Table 3).
Table 4 displays the score of each TPFQ-CK’s questions and the correlation between each question and the total score for the 20-item version. The correlation between the TPFQ-CK’s questions and the total score varied from 0.519 to 0.754, with a median value of 0.627.
Test-retest reliability
We assessed test-retest reliability using Spearman’s correlation (since our data was abnormally distributed, we used Spearman’s correlation) and the ICC. A Spearman’s correlation coefficient close to 1 with a significant p-value suggests that, on average, scores are stable and there is a strong positive relationship between individual scores across assessments. Table 5 shows that the TPFQ-CK has excellent test-retest reliability. The ICC for the total score (ICC=1.000) demonstrated excellent reliability. For the concentration factor, the ICC was 0.998, indicating excellent consistency and reliability. For the emotion factor, the ICC was 0.999. For the hearing factor, the ICC was 1.000. The statistics for the sleep subscale showed excellent consistency and reliability with an ICC of 0.999.
Bland-Altman plot
To evaluate the agreement or concordance between the first and second evaluation of TPFQ-CK, we used Bland-Altman plot. Fig. 1 displays the excellent concordance between test and retest evaluations.
Validity
Construct validity
A Spearman’s correlation coefficient was computed by comparison the total score of TPFQ-CK with THQ. Fig. 2 demonstrates a strong correlation of 0.895 between total scores of TPFQ-CK and THQ. Also, strong correlation was observed between the TPFQ-CK’s total score, its four factors and the THQ (r=0.895, 0.843, 0.858, 0.784, and 0.798; p<0.001). No meaningful association was found between the bilateral average of PTA and the TPFQ-CK total score and its subscales. The information is provided in Table 6.
Criterion validity
To assess the validity, Lin’s CCC was calculated, yielding a value of 0.999 (95% CI: 0.999 to 1.000) for the TPFQ-CK total score. This indicates a robust correlation and agreement between the two evaluations. No ceiling or floor effects were detected (0%).
Strong concordance between two measures was demonstrated for the concentration factor (CCC=0.999, 95% CI: 0.999 to 0.999). No ceiling or floor effects were detected (0%). Similarly, a strong concordance was observed for the emotion factor (CCC=0.999, 95% CI: 0.999 to 0.999), and no ceiling or floor effects were detected. For the hearing factor, the CCC was 0.999 (95% CI: 0.999 to 1.000), and for the sleep factor, the CCC was 0.999 (95% CI: 0.999 to 0.999), indicating strong concordance between measures. The percentage of responded questionnaires with scores <3 (floor) and >97 (ceiling) for the TPFQCK were recorded at two different times during one week. Scores on the questionnaire range from 0 (no problem) to 100 (worst problem imaginable).
Discussion
The successful cross-cultural adaptation of the TPFQ is a crucial step in enhancing the questionnaire’s utility and applicability in diverse linguistic and cultural settings. This study aimed to modify the TPFQ to suit different cultural and linguistic environments while maintaining its psychometric properties [11]. The adaptation process involved meticulous translation, back-translation, and expert evaluation. This meticulous technique aims to ensure that the questionnaire maintains semantic, idiomatic, and conceptual similarity across different languages. The findings of our study demonstrate that the adapted questionnaire maintains its language and cultural suitability [19].
The study modified the TPFQ to suit Kurdistan cultural context and assessed TPFQ-CK’s reliability and validity, demonstrating its effectiveness in evaluating subjective tinnitus outcomes in Kurdish-speaking people in Iraq.
In assessing reliability, we observed that the TPFQ-CK had outstanding reliability and showed good item-total correlations. The internal consistency, as demonstrated by Cronbach’s α, indicated that the questionnaire items are coherent and measure a common construct consistently across the sample. The Cronbach α coefficient for the total score of the TPFQ-CK was 0.933, suggesting that the TPFQ-CK consistently yields comparable results within the same context. The Cronbach’s α of the original TPFQ (0.89) was little lower than it. The Cronbach’s α subscale values for concentration, emotion, hearing, and sleep were all greater than 0.9, which is higher than the original TPFQ values of 0.86, 0.9, 0.9, and 0.93 [10].
The obtained excellent internal consistency is likely due to the cross-cultural translation, this guarantees the conceptual similarity of the original TPFQ. This is accomplished through a substantial decrease in random errors. Furthermore, the original TPFQ lacked information concerning the reliability of outcomes when the assessment was administered multiple times. The test-retest reliability confirmed the stability of the questionnaire over time. We evaluated the test-retest reliability of TPFQ-CK by employing the ICC, a statistical metric utilized to quantify the resemblance between measurements acquired from the same participants. Hence, it is more suitable to utilize the ICC for assessing test-retest reliability. Based on the Landis and Koch categorization [20], test-retest reliability can be categorized as outstanding (ICC more than 0.8), good (ICC located between 0.6 and 0.8), fair (ICC located between 0.4 to 0.6), poor (ICC located between 0.2 to 0.4), or bad (ICC less than 0.2). The test-retest reliability of TPFQ-CK in our study was excellent, as indicated by its ICC statistic, which was extremely close to 1. This result is in line with the findings of Arefi, et al. [21] in Persian version of TPFQ reported an ICC of 0.975.
We use the Bland-Altman plot to evaluate the agreement or concordance between the first and second assessments of TPFQ-CK. It indirectly provides information about the validity and reliability of the new method or assessor. Good agreement in the plot suggests good validity (the new method measures what it should) and reasonable reliability (consistent measurements). All of the data of difference and mean of first and second assessment of TPFQ-CK was scattered between upper 95% CI (1.2790808) and lower 95% CI (-0.198681) around the mean (0.5402) which is obtained from one-sample t-test, which indicate that the two sets of measurements or assessments are in good agreement, consistent with each other, reliable, and statistically significant.
Regarding the validity, content validity is a vital aspect of questionnaire adaptation. The expert committee, consisting of bilingual professionals in otolaryngology, audiology, psychology, and linguistics, played a pivotal role in ensuring that the adapted TPFQ preserved the content validity of the original version. The inclusion of expert perspectives helped identify and resolve any linguistic or cultural nuances that might have affected the questionnaire’s comprehensibility or relevance Arian Nahad, et al. [22].
In order to determine a constant linear correlation between the two measures, we calculated Spearman’s rho and CCC (results closer to 1 indicated higher validity). For assessing construct validity, a Spearman’s correlation coefficient was computed by comparing the TPFQ-CK total score with THQ. A strong association of 0.895 was observed between TPFQ-CK and THQ. The overall scores of the TPFQ-CK, concentration, emotion, hearing, and sleep subscales showed a strong correlation with the THQ (r=0.895, 0.843, 0.858, 0.784, and 0.798; p<0.001). Xin, et al. [6] showed significant correlations between the TPFQ total score, concentration, emotion, hearing, and sleep with the THQ (r=0.73, 0.70, 0.73, 0.56, and 0.56; p<0.001), our result is higher than that of the Chinese version of TPFQ. In addition, the ceiling and floor effects were incorporated to eliminate the potential for extreme outliers in the TPFQ-CK results.
The criterion validity was assessed by the evaluation of Lin’s concordance correlation coefficient between two measures, between the first and second evaluations of the TPFQ-CK total score, concentration, emotion, hearing, and sleep, all demonstrated a robust correlation and agreement between the two evaluations, and the CCC was 0.999 for all of them. According to Akoglu [23] and Lin [24], when assessing the level of agreement and consistency between two measures, an ICC and CCC value close to one indicates an extremely high level of agreement and consistency between the two measures. In our study, the ICC and CCC of the total score and factors were near one, indicating an extremely high level of agreement and consistency between the two measures.
The study sample was drawn from a specific region (Sulaymaniyah governorate, Iraq), which may limit the generalizability of the findings to other Kurdish-speaking populations. Possible cultural subtleties and differences in how tinnitus is understood may be present among individuals who speak Central Kurdish, which could impact their answers to the questionnaire.
In conclusion, the TPFQ-CK is a valid and reliable assessment tool for evaluating the influence of tinnitus on the quality of life of Central Kurdish speaking individuals with tinnitus. Our findings emphasise the necessity for additional assessments of TPFQ-CK to validate its efficacy in various clinical contexts.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
Print
