Voice Disorder Outcome Profile (V-DOP)-Translation and Validation in

Voice Disorder Outcome Profile (V-DOP)—Translation and Validation in Tamil Language Shenbagavalli Mahalingam, Prakash Boominathan, and Balasubramaniyan Subramaniyan, Chennai, Tamil Nadu, India Summary: This study sought to translate and validate the voice disorder outcome profile (V-DOP) for Tamil-speaking populations. It was implemented in two phases: the English language V-DOP developed for an Indian population was first translated into Tamil, a south Indian Dravidian language. Five Tamil language experts verified the translated version for exactness of meaning and usage. The expert’s comments and suggestions were used to select the questions for the final V-DOP, thus establishing content validity. Then the translated V-DOP was administered to 95 subjects (75 in clinical and 20 in nonclinical group) for reliability (item-total correlation) and validity (construct) measures. The overall Cronbach coefficient a for V-DOP was 0.89 whereas the mean total V-DOP score was zero for the nonclinical group and 104.28 for the clinical group (standard deviation ¼ 64.71). The emotional and functional domains indicated a statistically significant correlation (r ¼ 0.91 and r ¼ 0.90 respectively), followed by the physical domain (r ¼ 0.82) with the total scores. A significant, but moderate correlation was obtained across V-DOP domains (r ¼ 0.50 to 0.60; P < 0.05 and 0.01) and between total score and overall severity of V-DOP (r ¼ 0.62; P < 0.01). Thus the self-perception measuring VDOP in Tamil is a valid and reliable tool for evaluating the impact of voice disorders in Tamil-speaking population. Key Words: Voice disorder outcome profile–Dysphonia–Self-perception of voice.

INTRODUCTION Voice is a multidimensional entity and it reflects the speakers’ physical and emotional health, personality and identity. A voice disorder exists when the structure and/or function of laryngeal mechanism do not meet the speaker’s occupational, social, and voicing needs.1,2 The impact of voice disorders on individuals’ day-to-day activities is comparable to the impact of lifethreatening health issues such as cancer.3 Thus, measuring the health-related quality of life in these individuals is crucial. The impact of voice problems on occupational needs, physical and emotional health, and daily social communication can be evaluated using patient related outcome measures.4 In recent years, self-perception measures have gained importance over other tools in comprehensive voice assessments. Several standardized self-perception measures are available to assess the impact of voice disorders on the daily activities.3,5–9 In India, a culturally sensitive tool, voice disorder outcome profile (V-DOP), was developed by Konnai et al10 in English and Kannada language (a south Indian Dravidian language). This has evolved as a valid and reliable tool for assessing the individuals’ self-perception of voice problems. There is an inherent need to develop this culture sensitive tool (V-DOP) in Indian languages for clinical use. India has diverse languages and dialects and Tamil is another Indo- Dravidian language that is spoken in Tamil Nadu and elsewhere.11 Approximately 69 million people across the globe use Tamil as their first language and it is the 15th commonly used language worldwide.12 Thus

Accepted for publication April 8, 2014. Presented at ‘‘The Fourth World Voice Congress’’ in Seoul, South Korea (2010). From the Department of Speech Language and Hearing Sciences, Sri Ramachandra University, Chennai, Tamil Nadu, India. Address correspondence and reprint requests to Shenbagavalli Mahalingam, Department of Speech Language and Hearing Sciences, Sri Ramachandra University, Porur, Chennai 600 116, India. E-mail: [email protected] Journal of Voice, Vol. 28, No. 6, pp. 841.e21-841.e32 0892-1997/$36.00 Ó 2014 The Voice Foundation http://dx.doi.org/10.1016/j.jvoice.2014.04.006

this study sought to translate the V-DOP in Tamil language and validate it in Tamil-speaking population. AIM OF THE STUDY This study aimed to translate and validate the voice disorder outcome profile (V-DOP) in Tamil. METHOD Subjects Seventy-five subjects with dysphonia (age range, 18 to 65 years; Mean, 33.24 years; standard deviation, 15.44) were included in the clinical group. These subjects were diagnosed to have voice problems by an Otolaryngologist and a speech pathologist. Twenty individuals (age matched to clinical group; 10 males and 10 females) with no known history and/or complaint of any voice changes were included in nonclinical group. Table 1 summarizes the details of 75 subjects in clinical group. (Table 1). Procedure Phase I: translation of voice disorder outcome profile (V-DOP) in Tamil. The V-DOP developed by Konnai et al10 in English and Kannada was adapted with minor modification in this study. Certain questions in the V-DOP regarding jobs and their performance would not be applicable for homemakers, students, and so on. As suggested by Konnai et al,10 an additional column of ‘‘not applicable’’ was added to accommodate those subjects for whom jobs and their performance would not be applicable. The investigator translated the English version of V-DOP into the Tamil language. The translation and the English version of the questionnaire were given to five experts who had masters’ and/or doctoral degrees in Tamil literature studies to individually compare the V-DOP contents in English with the Tamil version. The study’s purpose was explained to them, and they were instructed to check the appropriateness of translation. These experts were encouraged to suggest pertinent changes

841.e22

Journal of Voice, Vol. 28, No. 6, 2014

TABLE 1. Subjects in Phase II Diagnosis Vocal fold nodule Hyperfunctional voice Gastro-esophageal reflux disease Chronic laryngitis Muscle tension dysphonia Vocal fold edema Vocal fold palsy Plica ventricularis Keratosis Sulcus vocalis Contact ulcer Hematoma Puberphonia Intracordal cyst Ectasia Spasmodic dysphonia (adductor type) Functional aphonia Total

No. of Subjects

Age Range (y)

Mean (y)

SD

18 (7 M and 11 F) 15 (8 M and 7 F) 13 (10 M and 3 F) 8 (8 F) 4 (2 M and 2 F) 3 (3 F) 2 (2 M) 2 (2 M) 2 (1 M and 1 F) 1 (1 M) 1 (1 M) 1 (1 M) 1 (1 M) 1 (1 F) 1 (1 F) 1 (1 F) 1 (1 F) 75 (36 M and 39 F)

19 to 36 20 to 45 18 to 36 19 to 33 22 to 58 25 to 42 43 to 60 39 to 65 33 to 64 28 36 49 23 36 49 60 20 18 to 65

25.82 30.43 27.20 24.38 47.50 31.67 51.5 52 59.50 — — — — — — — — 33.24

5.98 11.56 7.29 4.47 36.06 9.07 12.02 18.38 37.48 — — — — — — — — 15.44

by changing words and rephrasing sentences that they felt were inappropriate. Items with 80% agreement among experts were included in the translated version, so four out of five experts needed to agree that the translated version was correct (Appendix A and Appendix B). Phase II: reliability and validity of the V-DOP in Tamil. Reliability of V-DOP in Tamil was measured using internal consistency and item-total correlation. Construct validity of the V-DOP in Tamil was established by administering it to clinical and nonclinical groups. Administration of V-DOP V-DOP began with one question regarding the overall severity of the voice problem, followed by 32 questions under three

domains: physical, emotional, and functional. Subjects were instructed to rate the severity of their voice problem by putting an ‘X’ on a 10 cm line (visual analog scale) to reflect their perception. If the subjects did not feel a question was applicable to them, they could choose a ‘‘not applicable’’ option. The subjects were asked to complete the questionnaire; further clarification and assistance were provided only on request. The average time taken to complete the questionnaire was 10 minutes. Scoring and statistical analysis The distance (in centimeters) measured from the left end of the line to the individual’s mark was used as the score for each question. The scores for all items in each domain were acquired

TABLE 2. Item-Total Correlation of Each Item of V-DOP Using Cronbach a Coefficient Physical Items

Emotional

Functional

Item-Total Correlation

Items


Items


0.49 0.72 0.77 0.59 0.84 0.53 0.70 0.25* 0.64 0.58 —

11 12 13 14 15 16 17 18 19 20

0.77 0.70 0.55 0.66 0.56 0.42* 0.52 0.78 0.50 0.85 —

21 22 23 24 25 26 27 28 29 30 31 32

0.77 0.49 0.82 0.77 0.78 0.81 0.63 0.67 0.79 0.81 0.43* 0.42*

1 2 3 4 5 6 7 8 9 10

* Items with a < .5.

Shenbagavalli Mahalingam, et al

841.e23

V-DOP in Tamil Language

TABLE 3. Mean and Standard Deviation of Each Item in V-DOP for Clinical Group Physical Items 1 2 3 4 5 6 7 8 9 10

Emotional

Functional

Mean

SD

Items

Mean

SD

Items

Mean

SD

4.85 2.49 3.58 3.44 4.03 4.64 3.53 2.96 4.44 2.38 — —

3.24 2.91 3.12 3.16 3.46 3.28 3.16 2.84 2.95 2.75

11 12 13 14 15 16 17 18 19 20

4.43 4.68 3.82 2.99 2.97 3.30 2.55 3.31 2.56 2.59 — —

3.53 3.79 3.28 3.39 3.33 3.47 3.27 3.46 3.38 3.40

21 22 23 24 25 26 27 28 29 30 31 32

2.56 0.99 2.16 2.94 2.06 1.94 3.81 1.53 3.18 2.31 0.86 1.70

3.38 2.24 2.44 3.32 2.63 2.73 3.92 2.54 3.10 3.07 1.24 2.31

in this manner. The total V-DOP score was obtained by adding together the three domain scores. The total V-DOP score could be a maximum of 320 (as there were 32 items with a maximum score of 10 for each item) and a minimum of zero. Questions rated as ‘‘not applicable’’ by the subjects were not given any score and were excluded for the composite scoring and analysis. The data were then subjected to statistical analysis: 1. Reliability measure (internal consistency) of V-DOP was evaluated using item to total correlation and Cronbach alpha coefficient correlation test. 2. Validation of V-DOP was analyzed using Kendaul tau-b correlation test.

RESULTS Phase I: translation of voice disorder outcome profile (V-DOP) in Tamil Content validity of developed V-DOP in Tamil was established in Phase I. Appropriateness of the translation was analyzed with five experts in the Tamil language. Three translated questions in the emotional domain (items 12, 18, and 20) did not comply with 80% agreement. Experts were allowed to make changes such as replace words and rephrase sentences for these questions. The rephrased translated questions were analyzed for 80% agreement and questions for final Tamil version of VDOP evolved. Phase II: reliability and validity of the V-DOP in Tamil Reliability measures of V-DOP. The internal consistency of the V-DOP was estimated using item-total correlation and Cronbach coefficient a using SPSS version (14.0; Chicago, IL). Item-total correlation of V-DOP ranged from 0.49 to 0.85 for individual items; Table 2 presents the results. The V-DOP items had a high item-total correlation (a > .5) except for items 08, 16, 31, and 32. The overall Cronbach coefficient a for VDOP was .89. However, deleting these items from the question-

naire led to an insignificant decrease in overall a from .89 to .88. Thus, these items were considered to be relatively homogeneous and were included in the Tamil V-DOP. (Table 2) Validity measures of V-DOP. The mean score of each item of V-DOP for the nonclinical group was zero. The mean (SD) of each item, domains and total V-DOP scores for clinical group is summarized in Table 3 and Table 4. To establish the construct validity, the scores of each domain (physical, emotional, and functional) were correlated with the total V-DOP score and between the domains. A correlation matrix was obtained with the scores from the domains and with the total V-DOP scores using Spearman correlation coefficient. Results are in Table 5. Significant correlation was obtained across domains and the total V-DOP score with correlations ranging from 0.52 to 0.91. The emotional and functional domains indicated a statistically significant correlation (r ¼ 0.91 and r ¼ 0.90 respectively), followed by the physical domain (r ¼ 0.82) with the total scores (Tables 3 and 4). Overall severity score of V-DOP was correlated with domain score and total score. The correlation between self-perceived severity of dysphonia and the total V-DOP score was 0.62 (P < 0.01). The self-perceived severity of dysphonia correlated significantly with each domain at 0.53 for physical (P < 0.01), 0.60 for functional, and 0.50 for emotional domain (P < 0.05) (Table 5).

TABLE 4. Mean and Standard Deviation of Each Domain and Total V-DOP Score of Clinical Group Domains

Mean

SD

Severity Physical Emotional Functional Total

4.46 39.06 37.74 27.83 104.28

2.49 21.34 27.60 24.61 64.71

841.e24


TABLE 5. Spearman Correlation Matrix Across Domains and Total V-DOP Score Domains Severity Physical Emotional Functional Total

Severity

Physical

Emotional

Functional

Total

1 0.53* 0.60y 0.50y 0.62y

0.53* 1 0.60y 0.58y 0.82y

0.60y 0.60y 1 0.52y 0.91y

0.50y 0.58y 0.52y 1 0.90y

0.62y 0.82y 0.91y 0.90y 1

* Correlation is significant at P < 0.05. y Correlation is significant at P < 0.01.

DISCUSSION Phase I: translation of voice disorder outcome profile (V-DOP) in Tamil Experts’ verification of the translated version of Tamil V-DOP was based on the exactness of meaning and its acceptability in usage. The translated version of V-DOP agreed on by four out of five experts evolved as a final Tamil version of V-DOP and was used in the study. Phase II: reliability and validity of the V-DOP in Tamil Overall, Cronbach coefficient a of .89 and high item-total correlation indicated that V-DOP is a reliable measure to assess the impact of potential voice problems on quality of life in Tamilspeaking populations. The mean V-DOP scores for the nonclinical group (normal) and clinical group (individuals with dysphonia) were significantly different. The mean total V-DOP score for the dysphonia group was 104.28 (SD ¼ 64.71), whereas the mean was zero for the nonclinical group (there were no indications of any vocal difficulties for the nonclinical group). These results indicated that the V-DOP in Tamil can differentiate normal voice from dysphonia. Subjects in the nonclinical group (normal) rated the items to completely normal voice (zero score for all items). It can be noted that the scores of physical domain were greater followed by emotional and functional domain. Lesser scores in functional domain could be attributed to job related questions. Out of 75 subjects, 21 subjects were students or homemakers. They had rated job related questions as not applicable and were not included for analysis. A statistically significant correlation (r ¼ 0.82 to 0.91; P < 0.01) was obtained between domains of V-DOP with the total scores and among all the three domains (r ¼ 0.52 to 0.60). These findings support previous studies by Jacobson et al3 and Konnai et al.10 A high correlation within domains would indicate that a person’s voice problem might affect all parameters equally. However, the present study revealed that the impact of voice problems in these domains was not equally distributed and depended on environmental and personal factors such as personality of the individual, occupation, need for voice use, and so on. A statistically significant correlation was found between overall severity of the voice problem reported in V-DOP with the domains (r ¼ 0.50 to 0.60) and the total V-DOP score (r ¼ 0.62, P < 0.01). Similar findings were reported in Konnai et al,10 Jacobson et al,3 and Ma and Yiu6 suggesting that voice assessment tools share this trait.

CONCLUSION Development of language- and culture-specific self-perception measures is essential in assessing the impact of voice problems. V-DOP in Tamil has been shown as a valid and reliable tool for measuring quality of life in Tamil-speaking population. Future work should focus on evaluating the therapeutic outcomes in voice disorders using V-DOP and on correlating subjective and objective measures of voice assessment. Acknowledgment Our sincere thanks to Dr. Vallaurie Crawford (International student advisor/visiting professor, at the College of Humanities and Social Sciences, Taipei Medical University, Taipei, Taiwan) and Prof. Roopa Nagarajan (Course chairperson, Dept of Speech, Language and Hearing Sciences, Sri Ramachandra University, Porur, Chennai) for helping in technical editing of the manuscript. REFERENCES 1. Aronson AE, Bless DM, eds. Clinical Voice Disorders. New York, NY: Thieme; 2009. 2. Stemple JC, Glaze LE, Gerdeman BK, eds. Clinical Voice Pathology: Theory and Management. Connecticut: Cengage Learning; 2000. 3. Jacobson BH, Johnson A, Grywalski C, Silbergleit A, Jacobson G, Benninger MS, Newman CW. The Voice Handicap Index: development and validation. Amer J Sp Lang Path. 1997;6:66–70. 4. Ma EPM, Yiu EML, Abbott KV. Application of the ICF in voice disorders. Sem Sp Lang. 2007;28:343–350. 5. Hogikyan ND, Sethuraman G. Validation of an instrument to measure voice-related quality of life (V-RQOL). J Voice. 1999;13:557–569. 6. Ma EPM, Yiu EML. Voice activity and participation profile: Assessing the impact of voice disorders on daily activities. J Speech Lang Hear Res. 2001; 44:511–524. 7. Hartnick CJ. Validation of a pediatric voice quality-of-life instrument: the pediatric voice outcome survey. Arch Otolaryngol Head Neck Surg. 2002; 128:919–922. 8. Deary IJ, Wilson JA, Carding PN, MacKenzie K. VoiSS- A patient derived Voice Symptom Scale. J Psychosom Res. 2003;54:483–489. 9. Rosen CA, Lee AS, Osborne J, Zullo T, Murry T. Development and validation of the Voice Handicap Index-10. Laryngoscope. 2004;114:1549–1556. 10. Konnai RM, Jayaram M, Scherer RC. Development and validation of a voice disorder outcome profile (Voice-DOP) for an Indian population. J Voice. 2010;24:206–220. 11. Shanmugam SV. A Brief history of Tamil Langauge. Int J Dravidian Linguist. 2009;38:35–42. 12. Lewis MP, Gary FS, Charles DF, eds. Ethnologue: Languages of the World. 17th ed. Dallas, TX: SIL International; 2013. Available at: http://www. ethnologue.com. Accessed on 2nd April, 2014.


APPENDIX A


841.e25

841.e26




841.e27

841.e28



APPENDIX B


841.e29

841.e30




841.e31

841.e32
