Validation and Development of Voice Handicap Index-Throat for Koreans Open Access

Throat symptoms are one of the most common complaints of patients referred to general physicians and can significantly influence health-related quality of life, thus requiring appropriate treatment [1, 2]. Throat symptoms can be caused by laryngopharyngeal reflux, nonspecific mucosal hyperactivity, and lesions in the throat area [3‒5]. Additionally, throat symptoms such as throat clearing, globus sensation, and irritated throat are frequently reported by patients with dysphonia [6]. Throat symptoms may be considered the cause of functional voice disorders or the outcome of voice misuse or abuse [7]. Since throat symptoms are a relatively common cause of voice disorder, specific investigations on throat symptoms may be necessary for patients with dysphonia [8].

Three self-assessments (the Glasgow-Edinburgh Throat Scale, Pharyngeal Reflux Symptom Questionnaire, and Reflux Symptom Index) are commonly used to evaluate throat symptoms [9‒11]. However, these have been developed to measure problems with more diagnosis-specific characteristics (including globus and reflux) [6, 8]. Although the Voice Symptom Scale (VoiSS) questionnaire addressed pharyngeal symptoms, it was not specifically designed for this but to assess vocal symptoms [6].

To evaluate and develop an instrument that could simplify patients’ estimation of throat symptoms and consider their relation to voice problems simultaneously, Lyberg et al. [6] developed and validated the Voice Handicap Index-Throat (VHI-T) scale based on the Voice Handicap Index (VHI) because the VHI is widely used in clinical work and an instrument that is easy to distribute and analyze [6, 12‒14]. The VHI-T scale has been translated and validated into Persian and Brazilian Portuguese to date [8, 15]. This throat scale consists of 10 items in addition to the VHI.

Using the VHI-T scale helps to determine the category of discomfort in patients with throat symptoms and ascertain the degree of throat symptoms experienced by the patients [8, 15]. Also, VHI-T could simplify the patients’ estimation of symptoms from the throat and to consider their relation to voice problems simultaneously [6]. Despite these advantages, no published Korean version of the VHI-T (VHI-T_k) scale exists.

Thus, this study aimed to develop, validate, and analyze the reliability of the VHI-T_k. To do this, we translated the VHI-T scale into Korean, developed a Korean version of the VHI-T, and verified its validity and reliability.

Lyberg et al. [6] developed the original version of VHI-T in Swedish, but each item of the VHI-T was presented in English as well as Swedish in their study. Therefore, in our study, each item of the VHI-T written in English was adapted and translated into Korean. To do so, we referred to the guidelines in Kristjansson’s research [16]. First, two bilingual English translators and one speech-language pathologist (SLP) fluent in English were asked to translate the English version of the VHI-T into Korean language. Second, one English translator and three SLPs reviewed and agreed on the VHI-T translated into Korean. The translated version of the VHI-T was then back-translated into English. Finally, considering the suitability of the forward-backward translation, the VHI-T_k was produced (online suppl. material 1; for all online suppl. material, see https://doi.org/10.1159/000535108). Similar to the original VHI-T, the VHI-T_k includes 10 items. Each item was estimated on a five-point numerical scale (0 = never; 1 = almost never; 2 = sometimes; 3 = almost always; 4 = always). Higher scores represent more severe throat symptoms.

This study was approved by the Institutional Review Board of Kosin University (KU IRB 2022-0057). In total, 103 patients who visited the Pusan National University Hospital in South Korea with dysphonia and throat problems between May 2023 and July 2023 were selected. All diagnoses were made by an otolaryngologist and two SLPs. The patients were diagnosed with one of the following: functional dysphonia (N = 18; defined by vocal fatigue as a cardinal symptom, without any pathological laryngeal findings), mass in the larynx (N = 44; defined as the presence of leukoplakia, cyst, polyp, cancer, Reinke’s edema, papilloma, or nodule of the vocal folds), neurological voice disorder (N = 18; palsy of the vocal folds and spasmodic dysphonia), and throat problems, not themselves complaining of voice problems (N = 23; globus sense, chronic cough, laryngopharyngeal reflux, nonspecific mucosal hyperreactivity) This diagnostic classification was based on the study by Alizadeh et al. [8].

The control group comprised 27 healthy volunteers without dysphonia, chronic cough, or throat symptoms. There was no significant difference in age and sex ratios between the two groups (p = 0.951; 0.944, respectively). Patient information is shown in Table 1.

All participants completed the following three questionnaires at their initial visit (the VHI, VoiSS, and VHI-T). Since the VHI and VoiSS were adapted and validated in Korean (Korean version of the VHI [K-VHI] and the Korean version of the VoiSS [K-VoiSS]), the K-VHI and K-VoiSS were used in this study, and these two questionnaires were used to analyze concurrent validity based on previous studies [6, 8, 17, 18]. Additionally, for analyzing the test-retest reliability, the case group patients revisited after 2 weeks.

SPSS Statistics 25.0 (IBM Statistical Product and Service Solution, SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. The content validity of the VHI-T_k was evaluated by two laryngologists and two SLPs who did not participate in the translation and adaptation of the VHI-T_k. They were asked to estimate the appropriateness of the translations and applicability of each item of the VHI-T_k using a Likert scale (1 = not applicable, 5 = very applicable). To evaluate face validity, 20 participants (10 patients and 10 healthy volunteers) were asked to rate the importance of each item using a Likert scale (1 = not important, 5 = very important). In this study, the cutoff score of validity was defined as 4, and each item was considered valid if it was higher than the cutoff score [19]. To evaluate the construct validity of the VHI-T_k, we compared the VHI-T_k scores between the case and control groups using an independent samples t test. To evaluate the concurrent validity of the VHI-T_k, we implemented the correlation analysis among the VHI-T_k, K-VHI, and K-VoiSS using Spearman’s correlation coefficients.

The internal consistency and test-retest reliability of the VHI-T_k were also analyzed. For internal consistency analysis, Cronbach’s alpha correlation coefficients were calculated for each item. For Cronbach’s alpha correlation coefficients, values of 0.7, 0.8, and 0.9 were considered acceptable, good, and excellent, respectively [20]. For test-retest reliability analyses, we compared the results of the first and second questionnaires using a paired t test for the case group. Patients in the case group who experienced symptom changes were excluded from the test-retest reliability analysis. Thus, the test-retest reliability analyses were based on 20 patients in the case group who reported that they did not experience a symptom change for 2 weeks. Finally, we implemented a one-way ANOVA to assess differences in K-VHI, K-VoiSS, and VHI-T_k scores among the four diagnosis types in the case group.

To evaluate the content validity of the VHI-T_k, two laryngologists and two SLPs assessed the appropriateness of the translations and applicability of each item of the VHI-T_k using a Likert scale. The content validity ratio for the ten items was between 4.25 and 4.50. This indicates that the VHI-T_k has content validity. To evaluate the face validity of the VHI-T_k, 20 participants (10 patients and 10 healthy volunteers) rated the importance of each item using a Likert scale. The face validity ratio for the ten items was found to be between 4.05 and 4.40. This indicates that the VHI-T_k has face validity.

To analyze the construct validity of the VHI-T_k, we compared the VHI-T_k scores between the case and control groups using an independent samples t test (Table 2). The VHI-T_k scores of the case group (13.9 ± 7.8) were significantly higher than those of the control group (1.9 ± 1.0) (p < 0.001). To verify the concurrent validity of the VHI-T_k, a correlation analysis among the VHI-T_k, K-VoiSS, and K-VHI was conducted. The VHI-T_k scores were significantly correlated with scores from the K-VHI (functional, 0.393; physical, 0.414; emotional, 0.310; and total, 0.416) and the K-VoiSS (0.361) (p < 0.001).

The internal consistency of the VHI-T_k showed a high Cronbach’s alpha correlation coefficient for the VHI-T_k (0.908), with a total correlation ranging from 0.895 to 0.901. The test-retest reliability of the VHI-T_k was implemented in 20 patients in the case group. Total VHI-T_k scores did not change significantly between the first and second visits (first: 13.0 ± 5.9, second: 12.8 ± 5.4, p = 0.505).

We further analyzed the mean scores of K-VHI, K-VoiSS, and VHI-T_k according to the four diagnosis types in the case group. The scores of K-VHI and K-VoiSS were the highest in the neurological voice disorder group, and one-way ANOVA revealed significant difference in mean scores according to diagnosis types (range of p value: 0.000–0.009).

For the scores of VHI-T_k, one-way ANOVA revealed significant difference in mean scores according to diagnosis types (p = 0.025). Further analysis with the Tukey HSD test for the throat scale indicated statistically significant differences between the mean scores for the neurological voice disorder group and the throat problems group.

Throat symptoms are frequently reported in patients with dysphonia [21]. The VHI-T scale was developed to evaluate the patients’ self-estimation of throat symptoms and simultaneously consider their relationship to voice problems. This study aimed to verify the validity and reliability of the VHI-T_k. To our knowledge, this study is the first to validate the VHI-T in the Korean population.

We confirmed the content and face validity of the VHI-T_k. This indicated the translation suitability, and the VHI-T_k appeared to measure the degree of throat symptoms related to voice problems among Koreans.

To verify the construct validity of the VHI-T_k, we compared the K-VHI, K-VoiSS, and VHI-T_k scores of the case and control groups. Additionally, to verify the concurrent validity of the VHI-T_k, we implemented correlation analysis among the VHI-T_k, K-VHI, and K-VoiSS. We found that these scores of the case group were significantly higher than those of the control group. Also, the VHI-T_k scores were significantly correlated with the K-VHI and K-VoiSS. These results were similar to those of previous studies [6, 8]. Alizadeh et al. [8] reported that the VoiSS and VHI-T scores were significantly higher in the case group than in the control group, and the VHI-T was significantly correlated with the VoiSS. Regarding these results, they reported these two questionnaires are similar in that these examine the vocal tract discomfort. Lyberg et al. [6] reported that the three subscales of VHI and VHI-T scores were significantly higher in the case group than in the control group, and the VHI-T was correlated with the three subscales of VHI. Regarding these results, they reported that the VHI-T is originally designed as an instrument for patients to estimate perceived throat problems and is a good complementary tool to the VHI.

Previous studies reported that the VHI-T demonstrated internal consistency and test-retest reliability [6, 8]. In this study, the VHI-T_k was also found to have the internal consistency and test-retest reliability. Cronbach’s alpha correlation coefficients were greater than 0.8 for internal consistency. Thus, the VHI-T_k seems to be an appropriate tool for clinical use in Korean speaking populations, also being patient-friendly and convenient to administer and evaluate [6].

Furthermore, our results presented that the case group mainly have complaints on the physical subscale in the K-VHI. This is in accordance with the findings of Lyberg et al. [6] and Belafsky et al. [10]. Belafsky et al. (2002) observed a decrease on the physical subscale after treatment of throat problems. To be able to discriminate the patient’s problems, it might be more rewarding to evaluate the subscale scores of the VHI, rather than the total score [22]. Considering this, the VHI-T_k may become a useful clinical instrument that may help to discriminate problems that might be either coexisting or occurring separately.

Finally, we implemented a one-way ANOVA to assess differences in the K-VHI, K-VoiSS, and VHI-T_k scores among the four diagnosis types in the case group. The K-VHI and K-VoiSS scores were the highest in the neurological voice disorder group. Previous study mentioned that glottis insufficiency including neurological voice disorders is difficult to cure and causes the most severe handicaps in patients’ life [12]. In their study, the scores of VHI were the highest in the glottis insufficiency group including neurological voice disorders. For the VoiSS, there has been no study comparing scores across diagnosis, but a previous study reported that there was a positive correlation between the VoiSS and VHI [18]. Future research needs to be conducted on the VoiSS scores according to the diagnosis. In contrast to the K-VHI and K-VoiSS, the VHI-T_k scores were the highest in the throat problems group, and there was a statistically significant difference for the VHI-T_k scores between the neurological voice disorder group and throat problem group. In a previous study comparing the VHI-T scores of the six groups (phonasthenia, benign lesions of the vocal folds, neurological laryngeal motility disorder, benign goiter, throat problems, and control group), it was found that the throat problem group had the highest VHI-T score. Moreover, for the VHI-T scores, a statistically significant difference was observed between the mean scores for the neurological laryngeal motility disorder group and the throat problem group in a previous study [6].

Through the results of our study, it was confirmed that K-VHI and K-VoiSS can be efficiently used in evaluating voice problems and that VHI-T_k is useful in measuring throat problems. This suggests it may be useful to utilize VHI and VoiSS as well as VHI-T to estimate patients’ symptoms from the throat and to consider their relation to voice problems simultaneously. The limitations of our study are as follows. First, because there were no studies comparing the scores of VHI, VHI-T, and VoiSS among various groups, it is difficult to generalize the results of our study. There is a need to conduct research related to this. Second, there were differences in various factors between the previous study and this study (including difference in the number of patients for each diagnosis type, age, and degree of problem). Future studies that consider these factors are needed. Third, research studies are needed to confirm whether the VHI-T_k can be used with various questionnaires (e.g., VHI-10, Voice-Related Quality of Life, Voice Outcome Survey, VoiSS, etc.) as well as VHI and VoiSS.

However, our results showed that the VHI-T_k, in combination with the K-VHI may make it possible to discriminate between voice and throat problems and help patients express both categories of concerns simultaneously [6]. Lyberg et al. [6] mentioned that a deeper insight in the problems may lead to an increased understanding of patients with throat complaints, with or without voice complaints. Additionally, they reported that this knowledge may be helpful in designing clinical interventions [6].

In this study, we validated the VHI-T_k questionnaire for measuring self-perceived voice and throat problems in a Korean population. Our findings demonstrated that the VHI-T_k is a valid and reliable questionnaire for measuring self-perceived voice and throat problems. The VHI-T_k is expected to be useful in revealing a category of symptoms that are common in patients with dysphonia in clinical practice. Future studies with a larger sample size and various diagnosis types are required to further verify the clinical usefulness of the VHI-T_k.

This study was approved by the Institutional Review Board of Kosin University (KU IRB 2022-0057). Written informed consent was obtained from each subject after the details of the study were explained. This research was conducted ethically in accordance to the principles of the Declaration of Helsinki.

There are no conflicts of interest to declare.

This study was financially supported by the research fund of Kosin University (2022).

YeonWoo Lee conceived of the presented idea, processed and analyzed the data, and drafted the manuscript. DongWon Lim and JaeWon Kim were involved in data collection. GeunHyo Kim was involved in the overall design and planning of the research and interpretation of these results.

Data are not publicly available due to ethical reasons. Further inquiries can be directed to the corresponding author.