- Email: [email protected]
Cross-Cultural Adaptation and Validation of the Voice Handicap Index Into Greek
Cross-Cultural Adaptation and Validation of the Voice Handicap Index Into Greek *Meropi E. Helidoni, †,‡Thomas Murry, §Joanna Moschandreas, kChristos Lionis, {,#Athanasia Printza, and **George A. Velegrakis, *xk**Crete, {#Thessaloniki, Greece and yzNew York, New York Summary: The objective was to culturally adapt and validate the Voice Handicap Index (VHI) to the Greek language. The study design used was a psychometric analysis. The VHI was translated into Greek with cultural adaptations to accommodate certain words. The translated version was then completed by 67 subjects with various voice disorders and by a control group of 79 subjects. All the participants also completed a self-rating scale regarding the severity of their voice disorder. Statistical analyses demonstrated high internal consistency and high test-retest reliability both for the overall VHI score and for the functional, physical, and emotional domains of the VHI. A moderate correlation was found between the VHI and the self-rating severity scale. The subjects in the control group had lower scores compared to the subjects with voice disorders for the overall VHI score and for the three domains. Based on the internal consistency values and the test-retest reliability, the Greek version of VHI is a valid and reliable measure for use by Greek subjects with voice disorders. Key Words: Voice Handicap Index–Voice disorders–Greek Voice Handicap Index–Normal Voice. INTRODUCTION The Voice Handicap Index (VHI) is a self-administered voice handicap inventory that can be used with patients presenting a wide range of voice disorders. It has been shown to be a valid and reliable instrument for assessing patients’ self-perceived voice handicap.1 It has been translated into and validated for many languages2–11 and has been shown to be a consistent tool for assessing the validity and reliability of self-perceived voice handicap. In addition, it meets the criteria placed by the Agency for Healthcare Research and Quality for determining disability in speech-language disorders.12 The VHI was developed and validated by Jacobson et al.1 The VHI has three content domains representing functional, physical, and emotional aspects of voice disorders. There are 10 items corresponding to each domain, each item being measured on a five-point Likert-type scale (from 0 ¼ never to 4 ¼ always). A high score is expected to correspond to a more severe voice disorder. The VHI is usually completed within 5 minutes. The VHI is one of several tools that have gained popularity to aid in the assessment of voice disorders. Instrumental acoustic and aerodynamic measures have long been used to assess severity of the voice and changes with treatment. However, in recent years there has been a shift toward the development and use of patient-based assessment instruments for voice disorders to determine the patient’s perception of his/her voice Accepted for publication June 11, 2008. From the *ENT Department, University of Crete, Crete, Greece; yVoice and Swallowing Center, Columbia Presbyterian Medical Center, New York, New York; zClinical SpeechPathology in Otolaryngology, Columbia University, College of Physicians and Surgeons, New York, New York; xBiostatistics Lab, Faculty of Medicine, University of Crete, Crete, Greece; kClinic of Social and Family Medicine, Head of the Clinic of Social and Family Medicine, Faculty of Medicine, University of Crete, Crete, Greece; {2nd Otorhinolaryngology Department, Aristotle University of Thessaloniki, Voice Clinic, Thessaloniki, Greece; #2nd Otorhinolaryngology Department, Aristotle University of Thessaloniki, Papageorgiou Hospital, Thessaloniki, Greece; and the **Department of Otolaryngology of the University of Crete, Crete, Greece. Address correspondence and reprint requests to George A. Velegrakis, MD, University of Crete, Myrtias 3 Str., Heraklion, Crete 71409, Greece. E-mail: [email protected] Journal of Voice, Vol. 24, No. 2, pp. 221-227 0892-1997/$36.00 Ó 2010 The Voice Foundation doi:10.1016/j.jvoice.2008.06.005
handicap,1,13–18 which have all been shown to contribute to the assessment of voice handicap and offer insight to the patient’s perception of his/her voice problem. Of those, the VHI is currently the most widely used tool for patient selfassessment. Acoustic and aerodynamic measures provide useful information but they do not provide information on why dysphonic patients with similar diagnosis experience different levels of disability and handicap.1 For example, although some patients can tolerate the physical disability associated with reduced loudness in vocal fold paralysis, others cannot. Moreover, recovery from a voice disorder is also not uniformly perceived. Murry and Rosen19 have shown that improvement in VHI scores varies among patients depending on the type of treatment and the diagnosis of the disorder. Several studies have been carried out regarding the correlation of acoustic and aerodynamic measures to the VHI. In general, there is no strong relationship between acoustic or aerodynamic measures and the VHI. Hsiung et al20 investigated the relationship between voice laboratory measures and the VHI and found a poor correlation between them. Similar results were obtained by Wheeler et al,21 who correlated acoustic measures to the VHI and found that those measures could predict neither the overall VHI score nor the individual VHI items. Behrman et al22 also found a poor correlation between the harmonic-to-noise ratio and the VHI. The VHI has been used in clinical settings for self-evaluation of voice disorders and for pre- and posttreatment to assess treatment outcomes. Rosen and Murry23 investigated the use of the VHI in professional and recreational singers and in nonsingers (primarily teachers) with various voice disorders. Results showed that singers had lower VHI scores in comparison to nonsingers and professional singers had lower VHI scores in comparison to recreational singers. The results were interpreted to suggest that singers place different emphasis on their singing voice than on their speaking voice, whereas teachers focus on the speaking voice. This might be due to the fact that questions investigate voice disorders relevant to talking and not to singing. Several of the questions were particularly sensitive for singers. A low VHI score in singers should be taken under
222 consideration because as suggested by the authors this may relate to the specific voice needs of singers. The VHI also provides additional information that cannot be obtained through visual imaging and objective measures. In the study by Rosen and Murry,24 the VHI was administered before and after treatment to three groups of voice disordered patients with unilateral vocal fold paralysis, vocal cysts/polyps, and muscle tension dysphonia. Results showed a significant reduce of voice handicap for all three groups after treatment. Previous investigations of the VHI have shown that the VHI is a valid and reliable tool for self-assessment of a voice disorder. As such, it may be considered an objective test much like the Medical Outcomes Study Short Form 36 (SF-36).25,26 The SF-36 is a self-administered health measure that assesses eight health concepts: physical functioning, role functioning physical, social functioning, bodily pain, mental health, role functioning emotional, vitality, and general health perceptions. The SF-36 is a general health-related quality of life measure that has been used to measure the impact of voice disorders on the patients’ health-related quality of life.27,28 It has been translated and validated in the Greek Language.29,30 Wilson et al27 and Krischke et al28 used the SF-36 to investigate the impact of dysphonia on the patients’ health status and quality of life. Results indicated that voice disorders significantly affect the health status and the health-related quality of life in patients with voice disorders. Benninger et al31 examined the relationship between the SF-36 and the VHI and found that they correlated in the domains of social functioning, mental health, and role functioning emotional. However, it is not a specific instrument for self-assessment of voice disorders, and consequently it does not cover a wide range of conditions regarding the voice. A search in the Greek Literature showed that no standardized measures exist in the Greek language regarding self-assessment of voice disorders. Self-assessment of voice disorders is important because it provides significant information on how a patient experiences a voice disorder, and investigates the impact of a voice disorder in daily life. In Greece, the evaluation of voice disorders is performed through subjective perceptual measures of voice quality by the clinician, through visual imaging, acoustic, and aerodynamic measures. Patient self-assessment is usually limited to simply asking the patient about the severity of their voice disorder. The translation and validation of a widely used measure for self-assessment of voice disorders in Greek is important to better assess the patient’s awareness of the problem, the desire to recover, and the importance the patient places upon the voice. Knowing these factors allows the clinicians to better plan a treatment program consistent with the overall degree of handicap. Purposes The purposes of the present study were to culturally adapt the VHI to the Greek Language, and to obtain measures of reliability of the Greek VHI from a group of individuals with voice disorders and a control group of subjects without voice disorders. Further, the purpose was to evaluate the psychometric properties of the Greek VHI overall score and each of the three domains.
Journal of Voice, Vol. 24, No. 2, 2010
MATERIALS AND METHODS Translation of the VHI The translation and cultural adaptation of the VHI were carried out according to the ‘‘Minimal Translation Criteria.’’32 The VHI was forward translated into Greek by two Greek speech and language therapists, who are proficient in the English and Greek languages. A reconciliation version of the two translations was developed with the help of a moderator. A Greek/English bilingual professional translator then back translated the reconciliation version of the VHI into English. The back translation was subsequently sent to one of the authors (TM) for review and comments. A Cognitive Debriefing Method32 was used to assess the subjects’ understanding of each item of the questionnaire, identify any problems with language they had difficulty with, and offer alternative suggestions for the rewording of specific items of the questionnaire. As part of this method, the Greek version of the VHI was pilot tested with seven subjects with voice disorders and subsequently, the VHI was amended according to their suggestions (Appendix A). For example, in item P4 ‘‘My voice sounds creaky and dry,’’ the word ‘‘shaky’’ was used instead of ‘‘creaky’’ because the word ‘‘creaky’’ is not used in Greek for voice and ‘‘shaky’’ is an equivalent word in meaning. Following the Cognitive Debriefing Method,32 the final version of the VHI, along with the subjects’ suggestions regarding the wording of the questionnaire was resolved by consensus review among the authors. The final Greek version of the VHI translated into English is shown in Appendix B.
Subjects Consecutive sampling was used in the recruitment of participants. Forty-five outpatients, who attended the ENT clinic of the University Hospital of Crete for voice disorders between February and October 2007, were invited to participate in the study. Over the same time period, 22 subjects attending a private medical ENT office in Athens were also invited to participate. The subjects with voice disorders did not have previous laryngeal surgery nor did they have other recent surgeries at the time they completed the VHI. There were no refusals. The sample consisted of 18 males (27%) and 49 females (73%), a sex ratio of almost 1:3. The mean age of the voice disorder group was 47 years (SD 15 years), the age range being from 22 to 74 years. The principal investigator and an otolaryngologist classified the subjects in the voice disorder group into the following four broad subgroups according to their diagnosis: 1. ‘‘Mass lesions’’ subgroup (34 subjects, 51%): 18 with vocal nodules, 13 with vocal polyps, two with vocal fold cysts, one with leukoplakia. 2. ‘‘Inflammatory disorders’’ subgroup (15 subjects, 22%): Seven with Reinke’s edema, two with vocal fold hemorrhage, six with chronic laryngitis. 3. ‘‘Neurogenic disorders’’ subgroup (14 subjects, 21%): 10 with vocal fold paralysis, two with vocal fold paresis, one with superior laryngeal nerve paresis, one with spasmodic dysphonia.
Meropi E. Helidoni, et al
Validation of the Voice Handicap Index Into Greek
4. ‘‘Hyperfunctional disorders’’ subgroup (four subjects, 6%): four with muscle tension dysphonia. A control group consisting of 79 subjects participated in the study. These subjects were assembled from persons accompanying the voice disorder patients, those attending the ENT department for reasons other than a voice disorder, and clinical staff members. The mean age of the 79 subjects constituting the control group was 38 (SD 12.5) years (age range 19–70 years). Thirty-two of the control group subjects (40%) were men and 47 (60%) were women. The subjects in the control group did not report any voice complaints on the day of the completion of the VHI, did not present any voice disorders, had not undergone voice therapy, and had not lost any work days due to voice disorders in the past. The Ethical Committee of the Medical School of the University of Crete approved the study. All subjects were informed of the aims of the study and of the anonymity and confidentiality of the data. All subjects provided written consent before participating. Procedure The Greek VHI was completed twice with a period of approximately 2 weeks between each administration. An interval period of 2 weeks was selected because no substantial change was expected to take place in the health condition of the subjects within that period. All subjects reported no medical changes between the first and second administration. The actual interval period ranged from 5 to 28 days. The mean time interval between administrations was 13 days (SD ¼ 3 days) in the group of subjects with voice disorders and 14 days (SD ¼ 3 days) in the control group. When the subjects completed the second VHI, they had no access to their responses from the first VHI. As in the original study by Jacobson et al,1 all the subjects completed a four-point self-rating scale regarding the perception of their voice, in which 0 corresponded to a normal voice, 1 to a mild voice disorder, 2 to a moderate voice disorder, and 3 to a severe voice disorder. Statistical analysis The internal consistency of the Greek version of the VHI was assessed using Cronbach’s alpha coefficient. A value greater than 0.8 is considered ‘‘good’’ and greater than 0.9 ‘‘excellent.’’33,34 A value greater than 0.7 is often considered satisfactory. The test-retest reliability, that is, the repeatability of the questionnaire, was assessed by estimating the intraclass correlation coefficient (ICC). The ICC is a measure of the reproducibility of a variable and is defined as the proportion of variability that is associated with differences between subjects, as a proportion of the total variability (between- and withinsubject variation).35 The ICC was also calculated for each domain separately. Pearson’s rank correlation coefficient was calculated to assess the degree of association between the VHI score and the self-rating dysphonia severity scale. The distributions of both overall and individual domain VHI scores in the group of subjects with voice disorders were positively skewed. Possible correlations between VHI scores and the self-rating scale and age were assessed using the
223
nonparametric Spearman’s rho test. Comparisons of scores in the four voice disorder diagnosis subgroups were made using the nonparametric Kruskal-Wallis test and the distributions of VHI scores in male and female subjects with voice disorders were compared using the nonparametric Mann-Whitney test. Possible differences in scores between the three 10-item domains were assessed using the paired-sample t test. Average individual domain and overall VHI scores in voice disordered and control groups were compared using the nonparametric MannWhitney test. The average age of subjects in voice disordered and control groups was compared using the independent-sample t test and possible sex differences were assessed using the chisquared test of independence. The significance level was set to 0.05 throughout. The Statistical Package for the Social Sciences, Version 15 (SPSS Inc, Chicago, IL) was used for all statistical analyses.
RESULTS The overall median VHI score in the group of 67 Greek subjects with voice disorders was 32 (mean 37, SD 21.3), the range being from 4 to 97 out of a maximum possible score of 120. Median scores for the functional, physical, and emotional domains were 9, 18, and 6, respectively (Table 1), out of a maximum possible score of 40. Eight patients (12%) had a score of zero for the functional and/or the emotional domains, whereas the minimum score for the physical domain was 2. Of the 67 subjects with voice disorders, 13 (19%) perceived their voice as being severely impaired, 29 (43%) moderately impaired, 22 (33%) mildly impaired, and 3 (5%) normal, according to the self-rating scale. The overall estimated internal consistency of the VHI was excellent (a ¼ 0.95) and the three separate domains had internal consistency ranging from 0.88 to 0.91 (see Table 2). The estimated ICC for the overall VHI score was 0.96 with 95% confidence interval (CI) from 0.93 to 0.97, whereas the estimates for the three domains were as follows: 0.93 (95% CI: 0.89–0.96), 0.88 (95% CI: 0.82–0.93), and 0.93 (95% CI: 0.88–0.96) for the functional, physical, and emotional domains, respectively. Also, the overall VHI score positively correlated to the self-rating scale with Spearman’s rho equal to 0.55 (P < 0.0001, n ¼ 67). Each of the three separate domains (with rho ¼ 0.49, 0.52, and 0.51 for the functional, physical, and emotional domains, respectively, and P < 0.0001 in each case) was also positively correlated to the self-rating scale. The median VHI score was 66.5 in the neurogenic subgroup (min 7, max 97), 39 in the hyperfunctional subgroup (min 7, max 70), 29.5 in the mass lesions subgroup (min 10, max 72), and 29 in the inflammatory disorders subgroup (min 4, max 57). VHI score distribution differed to a statistically significant extent between the four diagnosis subgroups (P ¼ 0.042), these distributions being illustrated in Figure 1. Subjects with neurogenic disorders appeared to have higher functional domain scores with median score 19 (min 0, max 30) compared to 5.5, 10, and 10.5 in the mass lesion (min 0, max 20), inflammatory (min 0, max 19), and hyperfunctional (min 0, max 20) subgroups, respectively. The distribution of functional domain scores differed significantly between the four diagnosis
224
Journal of Voice, Vol. 24, No. 2, 2010
TABLE 1. A Summary of Scores for the Functional, Physical, and Emotional Domains and Overall, in the 67 Subjects With Voice Disorders Who Completed the Greek Version of the VHI VHI Domain (Maximum Possible Score)
Mean (SD)
Median (Q1, Q3)
Minimum, Maximum Score
Functional (40) Physical (40) Emotional (40) VHI overall (120)
10 (7.7) 18 (7.4) 9 (8.9) 37 (21.3)
9 (3, 14) 18 (13, 23) 6 (2, 16) 32 (21, 51)
0, 30 2, 33 0, 36 4, 97
Q1 ¼ lower quartile; Q3 ¼ upper quartile
subgroups (P ¼ 0.003), and are illustrated in Figure 2. The physical and emotional domain scores did not differ to a statistically significant extent among the four diagnosis subgroups (P ¼ 0.14 in each case). There was a difference in age distribution among the four subgroups (P < 0.0001), with median ages being 36, 57, 59, and 55 years in the mass lesion, inflammatory, neurogenic, and hyperfunctional subgroups, respectively. Age, however, was not correlated to either the individual domain scores or the overall VHI score (rho ¼ 0.10, P ¼ 0.45, rho ¼ 0.11, P ¼ 0.40, rho ¼ 0.12, P ¼ 0.35, and rho ¼ 0.33, P ¼ 0.79 for domains F, P, and E and the total score, respectively). The median VHI score in male subjects was 29 (min 4, max 27) and in female subjects 33 (min 10, max 97). The distribution of F, P, E, and total scores did not differ according to gender (with Mann-Whitney P values of 0.44, 0.15, 0.38, and 0.19, respectively). The average physical domain score was significantly higher than average scores in the functional and emotional domains, with mean difference between physical and functional domain scores being 8 (95% CI: 6–10) units and between physical and emotional domain scores again 8 (95% CI: 7–10) units. In contrast, the mean difference between functional and emotional domain scores was 0 (95% CI: 1 to 2) units. Forty (51%) of the control group subjects had an overall VHI score of zero, the range of scores being from 0 to 14. When the three separate domains were considered, 61 (77%), 49 (62%), and 68 (86%) of subjects had a score of zero in the F, P, and E domains, respectively. The maximum scores were 7, 7, and 3 for the F, P, and E domains, respectively. On the self-rating
FIGURE 1. Dot plots of individual VHI scores according to diagnosis group: mass lesions (n ¼ 34), inflammatory disorders (n ¼ 15), neurogenic disorders (n ¼ 14), and hyperfunctional disorders (n ¼ 4). The horizontal lines represent the medians. scale, 71 (90%) subjects perceived themselves as having a normal voice, whereas the remaining eight (10%) subjects stated having mild voice disorder. In the control group, both, overall scores and scores in each of the separate domains were negatively correlated to age, with rho ¼ 0.40 (P < 0.001) for the correlation of age with the overall VHI score, and rho ¼ 0.34 (P ¼ 0.02), 0.30 (P ¼ 0.008), and 0.29 (P ¼ 0.01) for the F, P, and E domains, respectively. Score distributions, however, did not differ between male and female control group subjects (P ¼ 0.81, 0.95, 0.17, and 0.62, respectively). The controls were 8 years younger, on average, than those in the voice disordered group, the 95% CI for the difference being from 4 to 13 years. VHI score distributions differed to a statistically significant extent between the voice disorder and the control group, both overall and for each of the F, P, and E domains separately (Mann-Whitney P < 0.0001 for each domain and overall). DISCUSSION The results of the present study indicate that the Greek version of the VHI has strong internal consistency for subjects with voice disorders. These findings are in agreement with the original study by Jacobson et al1 and with the Chinese version of the
TABLE 2. Cronbach’s Alpha Coefficient and the ICC Coefficients (With 95% CL) and Spearman’s Rank Correlation Coefficient (to Assess Correlation With the Self-Rating Scale) for Each of the Three Separate Domains and for the Overall VHI Score (n ¼ 67)
Domain
No. of Items
Cronbach’s Alpha Coefficient
Functional Physical Emotional Overall VHI
10 10 10 30
0.882 0.845 0.911 0.945
ICC (95% CI)
Correlation With Self-Rating Scale, r (P Value)
0.93 (0.89–0.96) 0.88 (0.82–0.93) 0.93 (0.88–0.96) 0.96 (0.93–0.97)
0.49 (<0.0001) 0.52 (<0.0001) 0.51 (<0.0001) 0.55 (<0.0001)
Meropi E. Helidoni, et al
Validation of the Voice Handicap Index Into Greek
FIGURE 2. Dot plots of individual functional domain scores according to diagnosis group: mass lesions (n ¼ 34), inflammatory disorders (n ¼ 15), neurogenic disorders (n ¼ 14), and hyperfunctional disorders (n ¼ 4). The horizontal lines represent median scores. VHI.2 The three domains had strong internal consistency as found in the Hebrew-translated version of the VHI.8 The Greek VHI also had excellent overall test-retest reliability. These findings are in agreement with the original study by Jacobson et al1 and with the Chinese version of the VHI.2 The test-retest reliability was found to be slightly lower for the physical domain, compared to the other two domains. This was also found in the Hakkesteegt et al’s study.10 In the voice disorder group, higher physical domain scores were found on average, in comparison to the functional and emotional domains, also in agreement with previous reports.3,5,10 It may be that subjects with voice disorders are more familiar with their physical symptoms and when confronted with statements about functional or emotional issues, they do not associate them specifically with a voice disorder. In other words, the physical symptoms are the prominent perceptual parameters patients directly associate with dysphonia. As in the original study by Jacobson et al,1 there was a moderate correlation between the overall VHI scores and the subjects’ self-rating of the severity of their voice disorder, supporting the construct validity of the Greek VHI. This finding supported the prediction of Jacobson et al1 and Lam et al2 that higher self-perception of voice handicap would result in higher scores in the functional, physical, and emotional domains of the VHI. This finding is also in agreement with the Chinese version of the VHI that also found a significant correlation with the subjects’ self-rating of dysphonia severity.2 When the subjects were grouped according to the type of voice disorder, the neurogenic subgroup had the highest overall scores, followed by subjects in the hyperfunctional subgroup, the mass lesion subgroup, and the inflammatory subgroup, in that order. Even though these results can only be preliminary due to the small numbers of subjects in each group, they are in agreement with the findings of previous studies. Hsiung et al3 reported that the glottic insufficiency group received the highest VHI scores in that study, followed by the vocal mass
225
group and the functional voice disorder group. Rosen et al24 evaluated three dysphonic groups of patients before and after treatment, and found that the highest pretreatment handicap scores were obtained by patients with unilateral vocal fold paralysis followed by patients with benign vocal fold lesions. The lowest scores were returned by patients with muscular tension dysphonia. Results obtained by Lam et al2 and Amir et al8 also indicated that the neurogenic group had the highest VHI scores. The Lam et al’s2 and the Amir et al’s8 study results were similar to the current study that found the lowest scores occurred in the inflammatory group. Thus, as can be seen from the present study and previous studies, there is overwhelming evidence that indicates the VHI can be affected by the type of the voice disorder.8 Patients sense the greatest handicap when there is a neurogenic-based voice disorder. The data from the present study indicate that the VHI may be a sensitive tool to identify voice disorders. With regard to the control group, VHI scores were significantly lower than the voice disorder group. These findings agree with those of Guimara˜es and Abberton,5 Amir et al,8 and Lam et al2 according to which subjects with voice disorders had significantly higher scores in relation to a control comparison group of subjects without voice disorders. The current subjects in the control group were not age matched to the voice disorder subjects; they were 8 years younger as compared to the subjects in the voice disorder group. Older control group subjects tended to have lower scores. However, in the voice disorder group age and gender were not correlated to the overall VHI score and to the three domains (F, P, and E). Future studies of the Greek population may lead to increased understanding of how this population responds to voice changes and to treatments. The proportion of male subjects in the voice disorder group of the present study was low. This reflects a general clinical impression in Greece. Although one can speculate on the reasons for the higher prevalence of voice disorders in females, additional research into lifestyle, employment, and environment is needed to reach definitive answers. It has been postulated that the physical aspects of higher fundamental frequency36 and reduced amounts of hyaluronic acid37 may ultimately play a role in the increased incidence of voice disorders in females compared to males. In recent years, the importance for the use of patient-centered measures along with the objective and perceptual measures for the evaluation and treatment effectiveness of voice disorders has been cited.38,39 By using the VHI in Greece, clinicians will have a better understanding of the reasons why patients seek help for their voice. With a valid and reliable patient questionnaire, the clinician may better address the problems confronting the patient. The opportunity to provide a multidimensional voice assessment for the voice-disordered patients invites a more effective treatment program in accordance with the patients’ needs.
CONCLUSIONS The Greek translation of the VHI appears to be a valid and reliable assessment of patients’ self-perception of voice handicap. In this study, a control group and a voice-disordered group demonstrated that the Greek VHI significantly discriminated
226 between the two groups of subjects. Application of the Greek VHI provides an additional valid tool for the clinician to understand the manner in which patients perceive their voice problem. Further work with the Greek VHI may also show that this assessment tool can also provide valid and reliable outcome data when used before and after treatments. REFERENCES 1. Jacobson BH, Johnson A, Grywalski C, et al. The Voice Handicap Index (VHI): development and validation. Am J Speech Lang Pathol. 1997;6: 66–70. 2. Lam PKY, Chan KM, Ho WK, Kwong E, Yiu EM, Wei WI. Cross-cultural adaptation and validation of the Chinese Voice Handicap Index-10. Laryngoscope. 2006;116:1192–1198. 3. Hsiung MW, Lu P, Kang BH, Wang HW. Measurement and validation of the Voice Handicap Index in voice-disordered patients in Taiwan. J Laryngol Otol. 2003;117:478–481. 4. Nawka T, Wiesmann U, Gonnermann U. Validierung des Voice Handicap Index (VHI) in der deutschen Fassung. HNO. 2003;51:921–929. 5. Guimara˜es I, Abberton E. An investigation of the Voice Handicap Index with speakers of Portuguese: preliminary data. J Voice. 2004;18:71–82. 6. Woisard V, Bodin S, Puech M. The Voice Handicap Index: impact of the translation in French on the validation. Rev Laryngol Otol Rhinol. 2004;125:307–312. 7. Pruszewicz A, Obrebowski A, Wiskirska-Woznica B, Wojnowski W. Complex voice assessment—Polish version of the Voice Handicap Index (VHI). Otolaryngol Pol. 2004;58:547–549. 8. Amir O, Ashkenazi O, Leibovitzh T, Michael O, Tavor Y, Wolf M. Applying the Voice Handicap Index (VHI) to dysphonic and nondysphonic Hebrew speakers. J Voice. 2006;20:318–324. 9. Amir O, Tavor Y, Leibovitzh T, et al. Evaluating the validity of the Voice Handicap Index-10 (VHI-10) among Hebrew speakers. Otolaryngol Head Neck Surg. 2006;135:603–607. 10. Hakkesteegt MM, Wieringa MH, Gerritsma EJ, Feenstra L. Reproducibility of the Dutch version of the Voice Handicap Index. Folia Phoniatr Logop. 2006;58:132–138. 11. Nunez-Batalla F, Corte-Santos P, Senaris-Gonzalez B, Liorente-Pendas JL, Gorriz-Gil C, Suarez-Nieto C. Adaptation and validation to the Spanish of the Voice Handicap Index (VHI-30) and its shortened version (VHI-10). Acta Otorrinolaringol Esp. 2007;58:386–392. 12. Agency for Healthcare Research and Quality. Criteria for determining disability in speech-language disorders. Evidence Report/Technology Assessment. January 2002, Number 52. 13. Rosen CA, Lee AS, Osborne J, Zullo T, Murry T. Development and validation of the Voice Handicap Index-10. Laryngoscope. 2004;114:1549–1556. 14. Hogikyan ND, Sethuraman G. Validation of an instrument to measure Voice-Related Quality of Life (V-RQOL). J Voice. 1999;13:557–569. 15. Gliklich RE, Glovsky RM, Montgomery WW. Validation of a voice outcome survey for unilateral vocal cord paralysis. Otolaryngol Head Neck Surg. 1999;120:153–158. 16. Ma EP, Yiu EM. Voice activity and participation profile: assessing the impact of voice disorders on daily activities. J Speech Lang Hear Res. 2001;44:511–524. 17. Deary IJ, Wilson JA, Carding PN, Mackenzie K. VoiSS A patient-derived Voice Symptom Scale. J Psychosom Res. 2003;54:483–489. 18. Deary IJ, Webb A, Mackenzie K, Wilson JA, Carding PN. Short, self-report voice symptom scales: psychometric characteristics of the Voice Handicap Index-10 and the Vocal Performance Questionnaire. Otolaryngol Head Neck Surg. 2004;131:232–235. 19. Murry T, Rosen CA. Outcome measurements and quality of life in voice disorders. In: Rosen CA, Murry T, eds. Philadelphia, PA: WB Saunders; 2000:905–916. 20. Hsiung MW, Pai L, Wang HW. Correlation between voice handicap index and voice laboratory measurements in dysphonic patients. Eur Arch Otorhinolaryngol. 2002;259:97–99.
Journal of Voice, Vol. 24, No. 2, 2010 21. Wheeler KM, Collins SP, Sapienza CM. The relationship between VHI scores and specific acoustic measures of mildly disordered voice production. J Voice. 2006;20:308–317. 22. Behrman A, Sulica L, He T. Factors predicting patient perception of dysphonia caused by benign vocal fold lesions. Laryngoscope. 2004;114: 1693–1700. 23. Rosen CA, Murry T. Voice Handicap Index in singers. J Voice. 2000;14: 370–377. 24. Rosen CA, Murry T, Zinn A, Zullo T, Sonbolian M. Voice Handicap Index change following treatment of voice disorders. J Voice. 2000;14:619–623. 25. Ware JE, Sherbourne CD. The MOS 36-Item Short-Form Health Survey (SF-36): I. Conceptual Framework and Item Selection. Med Care. 1992;30:473–483. 26. McHorney CA, Ware JE, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31:247–263. 27. Wilson JA, Deary IJ, Millar A, Mackenzie K. The quality of life impact of dysphonia. Clin Otolaryngol. 2002;27:179–182. 28. Krischke S, Weigelt S, Hoppe U, et al. Quality of life in dysphonic patients. J Voice. 2005;19:132–137. 29. Pappa E, Kontodimopoulos N, Niakas D. Validating and norming of the Greek SF-36 Health Survey. Qual Life Res. 2005;14:1433–1438. 30. Anagnostopoulos F, Niakas D, Pappa E. Construct validation of the Greek SF-36 Health Survey. Qual Life Res. 2005;14:1959–1965. 31. Benninger MS, Ahuja AS, Gardner G, Grywalski C. Assessing outcomes for dysphonic patients. J Voice. 1998;12:540–550. 32. Medical Outcomes Trust. Trust introduces new translation criteria. Trust Bull. 1997;5:1–4. 33. Feinstein AR. Clinimetrics. New Haven: Yale University Press; 1987. 180. 34. Kianifard F. Evaluation of Clinimetric Scales: basic principles and methods. Statistician. 1994;4:475–482. 35. Streiner DR, Norman GR. Health Measurement Scales: A practical Guide to Their Development and Use. 3rd ed.). New York: OUP; 2003. 133–137. 36. Woo P. Quantification of videostrobolaryngoscopic findings—measurements of the normal glottal cycle. Laryngoscope. 1996;106(Suppl 79):1–27. 37. Butler JE, Hammond TH, Gray SD. Gender-related differences of hyaluronic acid distribution in the human vocal fold. Laryngoscope. 2001;111: 907–911. 38. Dejonckere PH, Bradley P, Clemente P, et al. A basic protocol for functional assessment of voice pathology, especially for investigating the efficacy of (phonosurgical) treatments and evaluating new assessment techniques. Eur Arch Otorhinolaryngol. 2001;258:77–82. 39. Hogikyan ND, Rosen CA. A review of outcome measurements for voice disorders. Otolaryngol Head Neck Surg. 2002;126:562–572.
Appendix A Voice Handicap Index Physical part
Q4. My voice sounds creaky and dry. Emotional Q3. I find part other people don’t understand my voice problem. Emotional Q5. I am less part outgoing because of my voice problem. Emotional Q9. My voice part makes me feel incompetent.
Q4. My voice sounds shaky and dry. Q3. I find that other people do not comprehend the problem of my voice. Q5. I am less social because of the problem of my voice. Q9. My voice makes me feel insufficient.
Meropi E. Helidoni, et al
227
Validation of the Voice Handicap Index Into Greek
Appendix B Voice Handicap Index Original version Instructions These are statements that many people have used to describe their voices and the effects of their voices on their lives. Circle the response that indicates how frequently you have the same experience. 0: never, 1: almost never, 2: sometimes, 3: almost always, 4: always Greek Translation (final version)
. . 0:
, 1:
, 2:
, 3:
, 4:
Back – Translation into English (final version) Instructions Many people have used the following sentences in order to describe their voice and its effects on their lives. Circle the answer that indicates how often you have the same experience. 0: never, 1: almost never, 2: sometimes, 3: almost always, 4: always
Original Version F1. My voice makes it difficult for people to hear me. F2. People have difficulty understanding me in a noisy room F3. My family has difficulty hearing me when I call them throughout the house F4. I use the phone less often than I would like to
Greek Translation (final version)
Back-Translation into English (final version) F1. It’s hard for other people to hear me because of my voice. F2. It’s hard for other people to understand what I am saying in a noisy room. F3. It’s hard for my family to hear me when I call them inside the house. F4. I use the phone less often than I would like to. F5. I have the tendency to avoid being with groups of people because of my voice. F6. I talk to friends, neighbors or relatives less often because of my voice. F7. People ask me to repeat what I am saying when we speak face-to-face. F8. The difficulties of my voice limit my personal and social life. F9. I feel excluded from discussions due to my voice. F10. The problem of my voice causes me loss of income. P1. When I speak I become short of breath. P2. The sound of my voice varies during the day. P3. People ask me: “What’s wrong with your voice?” P4. My voice sounds shaky and dry. P5. I feel that I must make an effort to speak.
F1. . F2. . F3. . F4. .
F5. I tend to avoid groups of people because of my voice F6. I speak with friends, neighbors, or relatives less often because of my voice F7. People ask me to repeat myself when speaking face-to-face F8. My voice difficulties restrict my personal and social life F9. I feel left out of conversations because of my voice F10.My voice problem causes me to lose income
F5. . F6.
, .
F7. . F8. . F9. . F10.
, .
P1. I run out of air when I talk P2. The sound of my voice varies throughout the day P3. People ask, “What’s wrong with your voice?” P4. My voice sounds creaky and dry P5. I feel as though I have to strain to produce voice P6. The clarity of my voice is unpredictable
P1. P2.
. .
P3.
“
;”
P4. P5.
. .
P6. I cannot foresee the clarity of my voice.
P6. .
P7. I try to change my voice to sound different
P7. I try to change my voice in order to be heard differently. P8. I make a great effort to manage to speak.
P7. .
P8. I use a great deal of effort to speak
P8.
P9. My voice is worse in the evening P10. My voice ‘gives out’ on me in the middle of speaking E1. I am tense when talking to others because of my voice E2. People seem irritated with my voice
P9. P10. . E1.
E3. I find other people don’t understand my voice problem E4. My voice problem upsets me E5. I am less outgoing because of my voice problem E6. My voice makes me feel handicapped
E3. E4. E5.
E7. I feel annoyed when people ask me to repeat
E7.
. .
P9. My voice sounds worse at night. P10. My voice weakens while I speak. E1. I feel stress when I talk to other people, because of my voice. E2. People seem to be annoyed by my voice.
. E2. .
E3. I find that people do not comprehend the problem of my voice. E4. The problem of my voice upsets me. E5. I am less social because of the problem of my voice. E6. My voice makes me feel inferior.
. . / . E6. . .
E8. I feel embarrassed when people ask me to repeat E9. My voice makes me feel incompetent E10. I am ashamed of my voice problem
E8. . E9. E10.
. .
E7. I am annoyed when people ask me to repeat what I said. E8. I feel embarrassed when people ask me to repeat what I said. E9. My voice makes me feel insufficient. E10. I am ashamed of the problem of my voice.
handicap,1,13–18 which have all been shown to contribute to the assessment of voice handicap and offer insight to the patient’s perception of his/her voice problem. Of those, the VHI is currently the most widely used tool for patient selfassessment. Acoustic and aerodynamic measures provide useful information but they do not provide information on why dysphonic patients with similar diagnosis experience different levels of disability and handicap.1 For example, although some patients can tolerate the physical disability associated with reduced loudness in vocal fold paralysis, others cannot. Moreover, recovery from a voice disorder is also not uniformly perceived. Murry and Rosen19 have shown that improvement in VHI scores varies among patients depending on the type of treatment and the diagnosis of the disorder. Several studies have been carried out regarding the correlation of acoustic and aerodynamic measures to the VHI. In general, there is no strong relationship between acoustic or aerodynamic measures and the VHI. Hsiung et al20 investigated the relationship between voice laboratory measures and the VHI and found a poor correlation between them. Similar results were obtained by Wheeler et al,21 who correlated acoustic measures to the VHI and found that those measures could predict neither the overall VHI score nor the individual VHI items. Behrman et al22 also found a poor correlation between the harmonic-to-noise ratio and the VHI. The VHI has been used in clinical settings for self-evaluation of voice disorders and for pre- and posttreatment to assess treatment outcomes. Rosen and Murry23 investigated the use of the VHI in professional and recreational singers and in nonsingers (primarily teachers) with various voice disorders. Results showed that singers had lower VHI scores in comparison to nonsingers and professional singers had lower VHI scores in comparison to recreational singers. The results were interpreted to suggest that singers place different emphasis on their singing voice than on their speaking voice, whereas teachers focus on the speaking voice. This might be due to the fact that questions investigate voice disorders relevant to talking and not to singing. Several of the questions were particularly sensitive for singers. A low VHI score in singers should be taken under
222 consideration because as suggested by the authors this may relate to the specific voice needs of singers. The VHI also provides additional information that cannot be obtained through visual imaging and objective measures. In the study by Rosen and Murry,24 the VHI was administered before and after treatment to three groups of voice disordered patients with unilateral vocal fold paralysis, vocal cysts/polyps, and muscle tension dysphonia. Results showed a significant reduce of voice handicap for all three groups after treatment. Previous investigations of the VHI have shown that the VHI is a valid and reliable tool for self-assessment of a voice disorder. As such, it may be considered an objective test much like the Medical Outcomes Study Short Form 36 (SF-36).25,26 The SF-36 is a self-administered health measure that assesses eight health concepts: physical functioning, role functioning physical, social functioning, bodily pain, mental health, role functioning emotional, vitality, and general health perceptions. The SF-36 is a general health-related quality of life measure that has been used to measure the impact of voice disorders on the patients’ health-related quality of life.27,28 It has been translated and validated in the Greek Language.29,30 Wilson et al27 and Krischke et al28 used the SF-36 to investigate the impact of dysphonia on the patients’ health status and quality of life. Results indicated that voice disorders significantly affect the health status and the health-related quality of life in patients with voice disorders. Benninger et al31 examined the relationship between the SF-36 and the VHI and found that they correlated in the domains of social functioning, mental health, and role functioning emotional. However, it is not a specific instrument for self-assessment of voice disorders, and consequently it does not cover a wide range of conditions regarding the voice. A search in the Greek Literature showed that no standardized measures exist in the Greek language regarding self-assessment of voice disorders. Self-assessment of voice disorders is important because it provides significant information on how a patient experiences a voice disorder, and investigates the impact of a voice disorder in daily life. In Greece, the evaluation of voice disorders is performed through subjective perceptual measures of voice quality by the clinician, through visual imaging, acoustic, and aerodynamic measures. Patient self-assessment is usually limited to simply asking the patient about the severity of their voice disorder. The translation and validation of a widely used measure for self-assessment of voice disorders in Greek is important to better assess the patient’s awareness of the problem, the desire to recover, and the importance the patient places upon the voice. Knowing these factors allows the clinicians to better plan a treatment program consistent with the overall degree of handicap. Purposes The purposes of the present study were to culturally adapt the VHI to the Greek Language, and to obtain measures of reliability of the Greek VHI from a group of individuals with voice disorders and a control group of subjects without voice disorders. Further, the purpose was to evaluate the psychometric properties of the Greek VHI overall score and each of the three domains.
Journal of Voice, Vol. 24, No. 2, 2010
MATERIALS AND METHODS Translation of the VHI The translation and cultural adaptation of the VHI were carried out according to the ‘‘Minimal Translation Criteria.’’32 The VHI was forward translated into Greek by two Greek speech and language therapists, who are proficient in the English and Greek languages. A reconciliation version of the two translations was developed with the help of a moderator. A Greek/English bilingual professional translator then back translated the reconciliation version of the VHI into English. The back translation was subsequently sent to one of the authors (TM) for review and comments. A Cognitive Debriefing Method32 was used to assess the subjects’ understanding of each item of the questionnaire, identify any problems with language they had difficulty with, and offer alternative suggestions for the rewording of specific items of the questionnaire. As part of this method, the Greek version of the VHI was pilot tested with seven subjects with voice disorders and subsequently, the VHI was amended according to their suggestions (Appendix A). For example, in item P4 ‘‘My voice sounds creaky and dry,’’ the word ‘‘shaky’’ was used instead of ‘‘creaky’’ because the word ‘‘creaky’’ is not used in Greek for voice and ‘‘shaky’’ is an equivalent word in meaning. Following the Cognitive Debriefing Method,32 the final version of the VHI, along with the subjects’ suggestions regarding the wording of the questionnaire was resolved by consensus review among the authors. The final Greek version of the VHI translated into English is shown in Appendix B.
Subjects Consecutive sampling was used in the recruitment of participants. Forty-five outpatients, who attended the ENT clinic of the University Hospital of Crete for voice disorders between February and October 2007, were invited to participate in the study. Over the same time period, 22 subjects attending a private medical ENT office in Athens were also invited to participate. The subjects with voice disorders did not have previous laryngeal surgery nor did they have other recent surgeries at the time they completed the VHI. There were no refusals. The sample consisted of 18 males (27%) and 49 females (73%), a sex ratio of almost 1:3. The mean age of the voice disorder group was 47 years (SD 15 years), the age range being from 22 to 74 years. The principal investigator and an otolaryngologist classified the subjects in the voice disorder group into the following four broad subgroups according to their diagnosis: 1. ‘‘Mass lesions’’ subgroup (34 subjects, 51%): 18 with vocal nodules, 13 with vocal polyps, two with vocal fold cysts, one with leukoplakia. 2. ‘‘Inflammatory disorders’’ subgroup (15 subjects, 22%): Seven with Reinke’s edema, two with vocal fold hemorrhage, six with chronic laryngitis. 3. ‘‘Neurogenic disorders’’ subgroup (14 subjects, 21%): 10 with vocal fold paralysis, two with vocal fold paresis, one with superior laryngeal nerve paresis, one with spasmodic dysphonia.
Meropi E. Helidoni, et al
Validation of the Voice Handicap Index Into Greek
4. ‘‘Hyperfunctional disorders’’ subgroup (four subjects, 6%): four with muscle tension dysphonia. A control group consisting of 79 subjects participated in the study. These subjects were assembled from persons accompanying the voice disorder patients, those attending the ENT department for reasons other than a voice disorder, and clinical staff members. The mean age of the 79 subjects constituting the control group was 38 (SD 12.5) years (age range 19–70 years). Thirty-two of the control group subjects (40%) were men and 47 (60%) were women. The subjects in the control group did not report any voice complaints on the day of the completion of the VHI, did not present any voice disorders, had not undergone voice therapy, and had not lost any work days due to voice disorders in the past. The Ethical Committee of the Medical School of the University of Crete approved the study. All subjects were informed of the aims of the study and of the anonymity and confidentiality of the data. All subjects provided written consent before participating. Procedure The Greek VHI was completed twice with a period of approximately 2 weeks between each administration. An interval period of 2 weeks was selected because no substantial change was expected to take place in the health condition of the subjects within that period. All subjects reported no medical changes between the first and second administration. The actual interval period ranged from 5 to 28 days. The mean time interval between administrations was 13 days (SD ¼ 3 days) in the group of subjects with voice disorders and 14 days (SD ¼ 3 days) in the control group. When the subjects completed the second VHI, they had no access to their responses from the first VHI. As in the original study by Jacobson et al,1 all the subjects completed a four-point self-rating scale regarding the perception of their voice, in which 0 corresponded to a normal voice, 1 to a mild voice disorder, 2 to a moderate voice disorder, and 3 to a severe voice disorder. Statistical analysis The internal consistency of the Greek version of the VHI was assessed using Cronbach’s alpha coefficient. A value greater than 0.8 is considered ‘‘good’’ and greater than 0.9 ‘‘excellent.’’33,34 A value greater than 0.7 is often considered satisfactory. The test-retest reliability, that is, the repeatability of the questionnaire, was assessed by estimating the intraclass correlation coefficient (ICC). The ICC is a measure of the reproducibility of a variable and is defined as the proportion of variability that is associated with differences between subjects, as a proportion of the total variability (between- and withinsubject variation).35 The ICC was also calculated for each domain separately. Pearson’s rank correlation coefficient was calculated to assess the degree of association between the VHI score and the self-rating dysphonia severity scale. The distributions of both overall and individual domain VHI scores in the group of subjects with voice disorders were positively skewed. Possible correlations between VHI scores and the self-rating scale and age were assessed using the
223
nonparametric Spearman’s rho test. Comparisons of scores in the four voice disorder diagnosis subgroups were made using the nonparametric Kruskal-Wallis test and the distributions of VHI scores in male and female subjects with voice disorders were compared using the nonparametric Mann-Whitney test. Possible differences in scores between the three 10-item domains were assessed using the paired-sample t test. Average individual domain and overall VHI scores in voice disordered and control groups were compared using the nonparametric MannWhitney test. The average age of subjects in voice disordered and control groups was compared using the independent-sample t test and possible sex differences were assessed using the chisquared test of independence. The significance level was set to 0.05 throughout. The Statistical Package for the Social Sciences, Version 15 (SPSS Inc, Chicago, IL) was used for all statistical analyses.
RESULTS The overall median VHI score in the group of 67 Greek subjects with voice disorders was 32 (mean 37, SD 21.3), the range being from 4 to 97 out of a maximum possible score of 120. Median scores for the functional, physical, and emotional domains were 9, 18, and 6, respectively (Table 1), out of a maximum possible score of 40. Eight patients (12%) had a score of zero for the functional and/or the emotional domains, whereas the minimum score for the physical domain was 2. Of the 67 subjects with voice disorders, 13 (19%) perceived their voice as being severely impaired, 29 (43%) moderately impaired, 22 (33%) mildly impaired, and 3 (5%) normal, according to the self-rating scale. The overall estimated internal consistency of the VHI was excellent (a ¼ 0.95) and the three separate domains had internal consistency ranging from 0.88 to 0.91 (see Table 2). The estimated ICC for the overall VHI score was 0.96 with 95% confidence interval (CI) from 0.93 to 0.97, whereas the estimates for the three domains were as follows: 0.93 (95% CI: 0.89–0.96), 0.88 (95% CI: 0.82–0.93), and 0.93 (95% CI: 0.88–0.96) for the functional, physical, and emotional domains, respectively. Also, the overall VHI score positively correlated to the self-rating scale with Spearman’s rho equal to 0.55 (P < 0.0001, n ¼ 67). Each of the three separate domains (with rho ¼ 0.49, 0.52, and 0.51 for the functional, physical, and emotional domains, respectively, and P < 0.0001 in each case) was also positively correlated to the self-rating scale. The median VHI score was 66.5 in the neurogenic subgroup (min 7, max 97), 39 in the hyperfunctional subgroup (min 7, max 70), 29.5 in the mass lesions subgroup (min 10, max 72), and 29 in the inflammatory disorders subgroup (min 4, max 57). VHI score distribution differed to a statistically significant extent between the four diagnosis subgroups (P ¼ 0.042), these distributions being illustrated in Figure 1. Subjects with neurogenic disorders appeared to have higher functional domain scores with median score 19 (min 0, max 30) compared to 5.5, 10, and 10.5 in the mass lesion (min 0, max 20), inflammatory (min 0, max 19), and hyperfunctional (min 0, max 20) subgroups, respectively. The distribution of functional domain scores differed significantly between the four diagnosis
224
Journal of Voice, Vol. 24, No. 2, 2010
TABLE 1. A Summary of Scores for the Functional, Physical, and Emotional Domains and Overall, in the 67 Subjects With Voice Disorders Who Completed the Greek Version of the VHI VHI Domain (Maximum Possible Score)
Mean (SD)
Median (Q1, Q3)
Minimum, Maximum Score
Functional (40) Physical (40) Emotional (40) VHI overall (120)
10 (7.7) 18 (7.4) 9 (8.9) 37 (21.3)
9 (3, 14) 18 (13, 23) 6 (2, 16) 32 (21, 51)
0, 30 2, 33 0, 36 4, 97
Q1 ¼ lower quartile; Q3 ¼ upper quartile
subgroups (P ¼ 0.003), and are illustrated in Figure 2. The physical and emotional domain scores did not differ to a statistically significant extent among the four diagnosis subgroups (P ¼ 0.14 in each case). There was a difference in age distribution among the four subgroups (P < 0.0001), with median ages being 36, 57, 59, and 55 years in the mass lesion, inflammatory, neurogenic, and hyperfunctional subgroups, respectively. Age, however, was not correlated to either the individual domain scores or the overall VHI score (rho ¼ 0.10, P ¼ 0.45, rho ¼ 0.11, P ¼ 0.40, rho ¼ 0.12, P ¼ 0.35, and rho ¼ 0.33, P ¼ 0.79 for domains F, P, and E and the total score, respectively). The median VHI score in male subjects was 29 (min 4, max 27) and in female subjects 33 (min 10, max 97). The distribution of F, P, E, and total scores did not differ according to gender (with Mann-Whitney P values of 0.44, 0.15, 0.38, and 0.19, respectively). The average physical domain score was significantly higher than average scores in the functional and emotional domains, with mean difference between physical and functional domain scores being 8 (95% CI: 6–10) units and between physical and emotional domain scores again 8 (95% CI: 7–10) units. In contrast, the mean difference between functional and emotional domain scores was 0 (95% CI: 1 to 2) units. Forty (51%) of the control group subjects had an overall VHI score of zero, the range of scores being from 0 to 14. When the three separate domains were considered, 61 (77%), 49 (62%), and 68 (86%) of subjects had a score of zero in the F, P, and E domains, respectively. The maximum scores were 7, 7, and 3 for the F, P, and E domains, respectively. On the self-rating
FIGURE 1. Dot plots of individual VHI scores according to diagnosis group: mass lesions (n ¼ 34), inflammatory disorders (n ¼ 15), neurogenic disorders (n ¼ 14), and hyperfunctional disorders (n ¼ 4). The horizontal lines represent the medians. scale, 71 (90%) subjects perceived themselves as having a normal voice, whereas the remaining eight (10%) subjects stated having mild voice disorder. In the control group, both, overall scores and scores in each of the separate domains were negatively correlated to age, with rho ¼ 0.40 (P < 0.001) for the correlation of age with the overall VHI score, and rho ¼ 0.34 (P ¼ 0.02), 0.30 (P ¼ 0.008), and 0.29 (P ¼ 0.01) for the F, P, and E domains, respectively. Score distributions, however, did not differ between male and female control group subjects (P ¼ 0.81, 0.95, 0.17, and 0.62, respectively). The controls were 8 years younger, on average, than those in the voice disordered group, the 95% CI for the difference being from 4 to 13 years. VHI score distributions differed to a statistically significant extent between the voice disorder and the control group, both overall and for each of the F, P, and E domains separately (Mann-Whitney P < 0.0001 for each domain and overall). DISCUSSION The results of the present study indicate that the Greek version of the VHI has strong internal consistency for subjects with voice disorders. These findings are in agreement with the original study by Jacobson et al1 and with the Chinese version of the
TABLE 2. Cronbach’s Alpha Coefficient and the ICC Coefficients (With 95% CL) and Spearman’s Rank Correlation Coefficient (to Assess Correlation With the Self-Rating Scale) for Each of the Three Separate Domains and for the Overall VHI Score (n ¼ 67)
Domain
No. of Items
Cronbach’s Alpha Coefficient
Functional Physical Emotional Overall VHI
10 10 10 30
0.882 0.845 0.911 0.945
ICC (95% CI)
Correlation With Self-Rating Scale, r (P Value)
0.93 (0.89–0.96) 0.88 (0.82–0.93) 0.93 (0.88–0.96) 0.96 (0.93–0.97)
0.49 (<0.0001) 0.52 (<0.0001) 0.51 (<0.0001) 0.55 (<0.0001)
Meropi E. Helidoni, et al
Validation of the Voice Handicap Index Into Greek
FIGURE 2. Dot plots of individual functional domain scores according to diagnosis group: mass lesions (n ¼ 34), inflammatory disorders (n ¼ 15), neurogenic disorders (n ¼ 14), and hyperfunctional disorders (n ¼ 4). The horizontal lines represent median scores. VHI.2 The three domains had strong internal consistency as found in the Hebrew-translated version of the VHI.8 The Greek VHI also had excellent overall test-retest reliability. These findings are in agreement with the original study by Jacobson et al1 and with the Chinese version of the VHI.2 The test-retest reliability was found to be slightly lower for the physical domain, compared to the other two domains. This was also found in the Hakkesteegt et al’s study.10 In the voice disorder group, higher physical domain scores were found on average, in comparison to the functional and emotional domains, also in agreement with previous reports.3,5,10 It may be that subjects with voice disorders are more familiar with their physical symptoms and when confronted with statements about functional or emotional issues, they do not associate them specifically with a voice disorder. In other words, the physical symptoms are the prominent perceptual parameters patients directly associate with dysphonia. As in the original study by Jacobson et al,1 there was a moderate correlation between the overall VHI scores and the subjects’ self-rating of the severity of their voice disorder, supporting the construct validity of the Greek VHI. This finding supported the prediction of Jacobson et al1 and Lam et al2 that higher self-perception of voice handicap would result in higher scores in the functional, physical, and emotional domains of the VHI. This finding is also in agreement with the Chinese version of the VHI that also found a significant correlation with the subjects’ self-rating of dysphonia severity.2 When the subjects were grouped according to the type of voice disorder, the neurogenic subgroup had the highest overall scores, followed by subjects in the hyperfunctional subgroup, the mass lesion subgroup, and the inflammatory subgroup, in that order. Even though these results can only be preliminary due to the small numbers of subjects in each group, they are in agreement with the findings of previous studies. Hsiung et al3 reported that the glottic insufficiency group received the highest VHI scores in that study, followed by the vocal mass
225
group and the functional voice disorder group. Rosen et al24 evaluated three dysphonic groups of patients before and after treatment, and found that the highest pretreatment handicap scores were obtained by patients with unilateral vocal fold paralysis followed by patients with benign vocal fold lesions. The lowest scores were returned by patients with muscular tension dysphonia. Results obtained by Lam et al2 and Amir et al8 also indicated that the neurogenic group had the highest VHI scores. The Lam et al’s2 and the Amir et al’s8 study results were similar to the current study that found the lowest scores occurred in the inflammatory group. Thus, as can be seen from the present study and previous studies, there is overwhelming evidence that indicates the VHI can be affected by the type of the voice disorder.8 Patients sense the greatest handicap when there is a neurogenic-based voice disorder. The data from the present study indicate that the VHI may be a sensitive tool to identify voice disorders. With regard to the control group, VHI scores were significantly lower than the voice disorder group. These findings agree with those of Guimara˜es and Abberton,5 Amir et al,8 and Lam et al2 according to which subjects with voice disorders had significantly higher scores in relation to a control comparison group of subjects without voice disorders. The current subjects in the control group were not age matched to the voice disorder subjects; they were 8 years younger as compared to the subjects in the voice disorder group. Older control group subjects tended to have lower scores. However, in the voice disorder group age and gender were not correlated to the overall VHI score and to the three domains (F, P, and E). Future studies of the Greek population may lead to increased understanding of how this population responds to voice changes and to treatments. The proportion of male subjects in the voice disorder group of the present study was low. This reflects a general clinical impression in Greece. Although one can speculate on the reasons for the higher prevalence of voice disorders in females, additional research into lifestyle, employment, and environment is needed to reach definitive answers. It has been postulated that the physical aspects of higher fundamental frequency36 and reduced amounts of hyaluronic acid37 may ultimately play a role in the increased incidence of voice disorders in females compared to males. In recent years, the importance for the use of patient-centered measures along with the objective and perceptual measures for the evaluation and treatment effectiveness of voice disorders has been cited.38,39 By using the VHI in Greece, clinicians will have a better understanding of the reasons why patients seek help for their voice. With a valid and reliable patient questionnaire, the clinician may better address the problems confronting the patient. The opportunity to provide a multidimensional voice assessment for the voice-disordered patients invites a more effective treatment program in accordance with the patients’ needs.
CONCLUSIONS The Greek translation of the VHI appears to be a valid and reliable assessment of patients’ self-perception of voice handicap. In this study, a control group and a voice-disordered group demonstrated that the Greek VHI significantly discriminated
226 between the two groups of subjects. Application of the Greek VHI provides an additional valid tool for the clinician to understand the manner in which patients perceive their voice problem. Further work with the Greek VHI may also show that this assessment tool can also provide valid and reliable outcome data when used before and after treatments. REFERENCES 1. Jacobson BH, Johnson A, Grywalski C, et al. The Voice Handicap Index (VHI): development and validation. Am J Speech Lang Pathol. 1997;6: 66–70. 2. Lam PKY, Chan KM, Ho WK, Kwong E, Yiu EM, Wei WI. Cross-cultural adaptation and validation of the Chinese Voice Handicap Index-10. Laryngoscope. 2006;116:1192–1198. 3. Hsiung MW, Lu P, Kang BH, Wang HW. Measurement and validation of the Voice Handicap Index in voice-disordered patients in Taiwan. J Laryngol Otol. 2003;117:478–481. 4. Nawka T, Wiesmann U, Gonnermann U. Validierung des Voice Handicap Index (VHI) in der deutschen Fassung. HNO. 2003;51:921–929. 5. Guimara˜es I, Abberton E. An investigation of the Voice Handicap Index with speakers of Portuguese: preliminary data. J Voice. 2004;18:71–82. 6. Woisard V, Bodin S, Puech M. The Voice Handicap Index: impact of the translation in French on the validation. Rev Laryngol Otol Rhinol. 2004;125:307–312. 7. Pruszewicz A, Obrebowski A, Wiskirska-Woznica B, Wojnowski W. Complex voice assessment—Polish version of the Voice Handicap Index (VHI). Otolaryngol Pol. 2004;58:547–549. 8. Amir O, Ashkenazi O, Leibovitzh T, Michael O, Tavor Y, Wolf M. Applying the Voice Handicap Index (VHI) to dysphonic and nondysphonic Hebrew speakers. J Voice. 2006;20:318–324. 9. Amir O, Tavor Y, Leibovitzh T, et al. Evaluating the validity of the Voice Handicap Index-10 (VHI-10) among Hebrew speakers. Otolaryngol Head Neck Surg. 2006;135:603–607. 10. Hakkesteegt MM, Wieringa MH, Gerritsma EJ, Feenstra L. Reproducibility of the Dutch version of the Voice Handicap Index. Folia Phoniatr Logop. 2006;58:132–138. 11. Nunez-Batalla F, Corte-Santos P, Senaris-Gonzalez B, Liorente-Pendas JL, Gorriz-Gil C, Suarez-Nieto C. Adaptation and validation to the Spanish of the Voice Handicap Index (VHI-30) and its shortened version (VHI-10). Acta Otorrinolaringol Esp. 2007;58:386–392. 12. Agency for Healthcare Research and Quality. Criteria for determining disability in speech-language disorders. Evidence Report/Technology Assessment. January 2002, Number 52. 13. Rosen CA, Lee AS, Osborne J, Zullo T, Murry T. Development and validation of the Voice Handicap Index-10. Laryngoscope. 2004;114:1549–1556. 14. Hogikyan ND, Sethuraman G. Validation of an instrument to measure Voice-Related Quality of Life (V-RQOL). J Voice. 1999;13:557–569. 15. Gliklich RE, Glovsky RM, Montgomery WW. Validation of a voice outcome survey for unilateral vocal cord paralysis. Otolaryngol Head Neck Surg. 1999;120:153–158. 16. Ma EP, Yiu EM. Voice activity and participation profile: assessing the impact of voice disorders on daily activities. J Speech Lang Hear Res. 2001;44:511–524. 17. Deary IJ, Wilson JA, Carding PN, Mackenzie K. VoiSS A patient-derived Voice Symptom Scale. J Psychosom Res. 2003;54:483–489. 18. Deary IJ, Webb A, Mackenzie K, Wilson JA, Carding PN. Short, self-report voice symptom scales: psychometric characteristics of the Voice Handicap Index-10 and the Vocal Performance Questionnaire. Otolaryngol Head Neck Surg. 2004;131:232–235. 19. Murry T, Rosen CA. Outcome measurements and quality of life in voice disorders. In: Rosen CA, Murry T, eds. Philadelphia, PA: WB Saunders; 2000:905–916. 20. Hsiung MW, Pai L, Wang HW. Correlation between voice handicap index and voice laboratory measurements in dysphonic patients. Eur Arch Otorhinolaryngol. 2002;259:97–99.
Journal of Voice, Vol. 24, No. 2, 2010 21. Wheeler KM, Collins SP, Sapienza CM. The relationship between VHI scores and specific acoustic measures of mildly disordered voice production. J Voice. 2006;20:308–317. 22. Behrman A, Sulica L, He T. Factors predicting patient perception of dysphonia caused by benign vocal fold lesions. Laryngoscope. 2004;114: 1693–1700. 23. Rosen CA, Murry T. Voice Handicap Index in singers. J Voice. 2000;14: 370–377. 24. Rosen CA, Murry T, Zinn A, Zullo T, Sonbolian M. Voice Handicap Index change following treatment of voice disorders. J Voice. 2000;14:619–623. 25. Ware JE, Sherbourne CD. The MOS 36-Item Short-Form Health Survey (SF-36): I. Conceptual Framework and Item Selection. Med Care. 1992;30:473–483. 26. McHorney CA, Ware JE, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31:247–263. 27. Wilson JA, Deary IJ, Millar A, Mackenzie K. The quality of life impact of dysphonia. Clin Otolaryngol. 2002;27:179–182. 28. Krischke S, Weigelt S, Hoppe U, et al. Quality of life in dysphonic patients. J Voice. 2005;19:132–137. 29. Pappa E, Kontodimopoulos N, Niakas D. Validating and norming of the Greek SF-36 Health Survey. Qual Life Res. 2005;14:1433–1438. 30. Anagnostopoulos F, Niakas D, Pappa E. Construct validation of the Greek SF-36 Health Survey. Qual Life Res. 2005;14:1959–1965. 31. Benninger MS, Ahuja AS, Gardner G, Grywalski C. Assessing outcomes for dysphonic patients. J Voice. 1998;12:540–550. 32. Medical Outcomes Trust. Trust introduces new translation criteria. Trust Bull. 1997;5:1–4. 33. Feinstein AR. Clinimetrics. New Haven: Yale University Press; 1987. 180. 34. Kianifard F. Evaluation of Clinimetric Scales: basic principles and methods. Statistician. 1994;4:475–482. 35. Streiner DR, Norman GR. Health Measurement Scales: A practical Guide to Their Development and Use. 3rd ed.). New York: OUP; 2003. 133–137. 36. Woo P. Quantification of videostrobolaryngoscopic findings—measurements of the normal glottal cycle. Laryngoscope. 1996;106(Suppl 79):1–27. 37. Butler JE, Hammond TH, Gray SD. Gender-related differences of hyaluronic acid distribution in the human vocal fold. Laryngoscope. 2001;111: 907–911. 38. Dejonckere PH, Bradley P, Clemente P, et al. A basic protocol for functional assessment of voice pathology, especially for investigating the efficacy of (phonosurgical) treatments and evaluating new assessment techniques. Eur Arch Otorhinolaryngol. 2001;258:77–82. 39. Hogikyan ND, Rosen CA. A review of outcome measurements for voice disorders. Otolaryngol Head Neck Surg. 2002;126:562–572.
Appendix A Voice Handicap Index Physical part
Q4. My voice sounds creaky and dry. Emotional Q3. I find part other people don’t understand my voice problem. Emotional Q5. I am less part outgoing because of my voice problem. Emotional Q9. My voice part makes me feel incompetent.
Q4. My voice sounds shaky and dry. Q3. I find that other people do not comprehend the problem of my voice. Q5. I am less social because of the problem of my voice. Q9. My voice makes me feel insufficient.
Meropi E. Helidoni, et al
227
Validation of the Voice Handicap Index Into Greek
Appendix B Voice Handicap Index Original version Instructions These are statements that many people have used to describe their voices and the effects of their voices on their lives. Circle the response that indicates how frequently you have the same experience. 0: never, 1: almost never, 2: sometimes, 3: almost always, 4: always Greek Translation (final version)
. . 0:
, 1:
, 2:
, 3:
, 4:
Back – Translation into English (final version) Instructions Many people have used the following sentences in order to describe their voice and its effects on their lives. Circle the answer that indicates how often you have the same experience. 0: never, 1: almost never, 2: sometimes, 3: almost always, 4: always
Original Version F1. My voice makes it difficult for people to hear me. F2. People have difficulty understanding me in a noisy room F3. My family has difficulty hearing me when I call them throughout the house F4. I use the phone less often than I would like to
Greek Translation (final version)
Back-Translation into English (final version) F1. It’s hard for other people to hear me because of my voice. F2. It’s hard for other people to understand what I am saying in a noisy room. F3. It’s hard for my family to hear me when I call them inside the house. F4. I use the phone less often than I would like to. F5. I have the tendency to avoid being with groups of people because of my voice. F6. I talk to friends, neighbors or relatives less often because of my voice. F7. People ask me to repeat what I am saying when we speak face-to-face. F8. The difficulties of my voice limit my personal and social life. F9. I feel excluded from discussions due to my voice. F10. The problem of my voice causes me loss of income. P1. When I speak I become short of breath. P2. The sound of my voice varies during the day. P3. People ask me: “What’s wrong with your voice?” P4. My voice sounds shaky and dry. P5. I feel that I must make an effort to speak.
F1. . F2. . F3. . F4. .
F5. I tend to avoid groups of people because of my voice F6. I speak with friends, neighbors, or relatives less often because of my voice F7. People ask me to repeat myself when speaking face-to-face F8. My voice difficulties restrict my personal and social life F9. I feel left out of conversations because of my voice F10.My voice problem causes me to lose income
F5. . F6.
, .
F7. . F8. . F9. . F10.
, .
P1. I run out of air when I talk P2. The sound of my voice varies throughout the day P3. People ask, “What’s wrong with your voice?” P4. My voice sounds creaky and dry P5. I feel as though I have to strain to produce voice P6. The clarity of my voice is unpredictable
P1. P2.
. .
P3.
“
;”
P4. P5.
. .
P6. I cannot foresee the clarity of my voice.
P6. .
P7. I try to change my voice to sound different
P7. I try to change my voice in order to be heard differently. P8. I make a great effort to manage to speak.
P7. .
P8. I use a great deal of effort to speak
P8.
P9. My voice is worse in the evening P10. My voice ‘gives out’ on me in the middle of speaking E1. I am tense when talking to others because of my voice E2. People seem irritated with my voice
P9. P10. . E1.
E3. I find other people don’t understand my voice problem E4. My voice problem upsets me E5. I am less outgoing because of my voice problem E6. My voice makes me feel handicapped
E3. E4. E5.
E7. I feel annoyed when people ask me to repeat
E7.
. .
P9. My voice sounds worse at night. P10. My voice weakens while I speak. E1. I feel stress when I talk to other people, because of my voice. E2. People seem to be annoyed by my voice.
. E2. .
E3. I find that people do not comprehend the problem of my voice. E4. The problem of my voice upsets me. E5. I am less social because of the problem of my voice. E6. My voice makes me feel inferior.
. . / . E6. . .
E8. I feel embarrassed when people ask me to repeat E9. My voice makes me feel incompetent E10. I am ashamed of my voice problem
E8. . E9. E10.
. .
E7. I am annoyed when people ask me to repeat what I said. E8. I feel embarrassed when people ask me to repeat what I said. E9. My voice makes me feel insufficient. E10. I am ashamed of the problem of my voice.