ARTICLE IN PRESS

VHI-10 and SVHI-10 Differences in Singers’

Self-perception of Dysphonia Severity
*Elizabeth Renk, †Lucian Sulica, ‡Chad Grossman, ‡Jenny Georges, and *Thomas Murry, *Loma Linda, California, and
†‡New York, New York

Summary: Objective. Previous investigations of the Voice Handicap Index (VHI)-10 in clinical practice noted that
specific information relevant to singers was not forthcoming. Consequently, a second index, the Singing Voice Hand-
icap Index (SVHI) as well as its shortened counterpart the SVHI-10, was developed. The purpose of this study was to
directly compare the differences in scores between the VHI-10 and the SVHI-10 in a group of 50 singers.
Methods. A retrospective chart review of 50 singers (26 women, 24 men) was performed between June 2014 and
November 2014 at Weill Cornell Medical College, New York. Subjects completed both the VHI-10 and the SVHI-10
at their initial evaluation. The results from the VHI-10 and the SVHI-10 were then compared using paired t test and
two-way analysis of variance.
Results. The SVHI-10 scores from the performers were significantly higher than those of the VHI-10 (P < 0.0001).
The mean score on the VHI-10 was 12.1 compared with 20.4 on the SVHI-10 (maximum score for each questionnaire
is 40). There were no significant gender differences when comparing the VHI-10 and the SVHI-10 on the overall scores
or for individual items. The analysis of variance also found no significant gender difference (P = 0.865) and con-
firmed a significant difference between VHI-10 and SVHI-10 (P = 0.0003).
Conclusion. Whereas singers may have general complaints about their voice, they also have specific complaints that relate
only to their singing voice. Finding a significant difference between the scores of the VHI-10 and the SVHI-10 suggests
the importance of assessing the singer’s perception of voice severity using a tool that focuses on the singing voice.
Key Words: Singers–Voice–Dysphonia–Voice handicap index.

INTRODUCTION
Objective analysis of therapeutic results in the management of
voice disorders has been a challenge.1 Consequently, outcomes
research has been the mainstay for determining the need for clin-
ical management of voice disorders and for voice research.
Outcomes research places the focus of the impact of disease and
subsequent treatment on the patients’ subjective evaluation of
their voice disorder as well as the clinician’s impression of
changes. In 1997, Jacobson et al published the Voice Handicap
Index (VHI), a 30-item series of statements about the use of the
voice, which is scored from zero (never) to 4 (always).2 The VHI
was later shortened and validated to the VHI-10, a 10-item form
now used by clinicians around the world.3
During the development of the VHI and the VHI-10, the state-
ments that compose the indexes were organized into three
categories—functional, physical, and emotional. The VHI com-
prises 10 functional items, 10 physical items, and 10 emotional
items. From those 30 items, the VHI-10 was reduced to five func-
tional items, three physical items, and two emotional items.
As the VHI-10 was used in clinical practice, differences were
noted between nonsinger and singer populations. VHI-10 lacked
statements specifically about singers and the singing voice. Several
investigators identified differences between singers and nonsingers
in their use of the VHI-10. Phyland et al compared 167 vocal
Accepted for publication August 26, 2016.
From the *Department of Otolaryngology—Head and Neck Surgery, Loma Linda Uni-
versity Medical Center, Loma Linda, California; †Department of Otolaryngology—Head
and Neck Surgery, Weill Cornell Medical College, New York, New York; and the ‡Teach-
ers College, Columbia University, New York, New York.
Address correspondence and reprint requests to Elizabeth Renk, Department of
Otolaryngology, Loma Linda University, 11234 Anderson Street, Suite 2586A, Loma Linda,
CA 92354. E-mail: erenk@llu.edu, lizarenk@gmail.com
Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■
0892-1997
© 2016 The Voice Foundation. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jvoice.2016.08.017
performers with 86 nonperformers and found that the vocal per-
formers were more likely to be diagnosed with a vocal condition
in the past 12 months (44% compared with 21%) and were more
likely to suffer vocal disability (69% compared with 41%).4
Despite these significant complaints regarding their voice, Rosen
and Murry found that, when compared with nonsingers, singers
scored significantly lower on the VHI (less severe).5 To explain
this paradoxical finding, Murry et al compared the VHI-10 re-
sponses of 35 singers with that of 35 nonsingers. They found
that the two groups ranked differently the 10 statements that
compose the VHI-10. In addition, when Murry et al changed three
statements to “my singing voice,” rather than simply “my voice,”
the singers judged their voice complaints as more severe.6
The above findings led to the development of a second index,
the Singing Voice Handicap Index (SVHI)7 and its shortened coun-
terpart the SVHI-10,8 which were developed and validated on
performers. To develop the SVHI, a clinical consensus confer-
ence (comprising laryngologists and speech pathologists) generated
an 81-item list of voice complaints that they felt would be of par-
ticular concern to singers. This list was then narrowed to 36 items
and then administered by Cohen et al to a group of 112 singers
with dysphonia and 129 singers with normal voice. They found
a high test-retest reliability, as well as a high internal consistency.7,8
In an effort to simplify the more cumbersome 36-item SVHI,
Cohen et al undertook the development of the SVHI-10.8 Once
again, a clinical consensus conference selected 16 items (from
the original 36), which in their combined opinion had the highest
content validity. Ultimately, 10 of the highest items in content
validity were retained to create the SVHI-10. This shortened
survey was then administered to a group of 91 singers with dys-
phonia at their initial evaluation. A second survey was mailed
to them 1 day later. Once again, there was a high test-retest re-
liability, as well as a high internal consistency.
 ARTICLE IN PRESS
2 Journal of Voice, Vol. ■■, No. ■■, 2016

TABLE 1. TABLE 2.
Voice Handicap Index-10 Patient Demographics
F1 My voice makes it difficult for 0 1 2 3 4 Groups N (%)
people to hear me. n 50
F2 People have difficulty 0 1 2 3 4 Male, n (%) 24 (48%)
understanding me in a noisy Female, n (%) 26 (52%)
room Age
F8 My voice difficulties restrict 0 1 2 3 4 11–20, n (%) 6 (12%)
personal and social life 21–30, n (%) 27 (54%)
F9 I feel left out of conversations 0 1 2 3 4 31–40, n (%) 10 (20%)
because of my voice 41–50, n (%) 5 (10%)
F10 My voice problem causes me to 0 1 2 3 4 51–60, n (%) 2 (4%)
lose income Type of performer
P5 I feel as though I have to strain to 0 1 2 3 4 Current performers, n (%) 34 (68%)
produce voice Graduate students, n (%) 14 (28%)
P6 The clarity of my voice is 0 1 2 3 4 Previous performers, n (%) 2 (4%)
unpredictable
E4 My voice problem upsets me 0 1 2 3 4
E6 My voice makes me feel 0 1 2 3 4 DATA ANALYSIS
handicapped The data were submitted to a subjects (male vs female) by con-
P3 People ask “What is wrong with 0 1 2 3 4
ditions (VHI-10 vs SVHI-10) two-way analysis of variance
your voice?”
(ANOVA), with repeated measures on subjects for each of the
Abbreviations: E, emotional; F, functional; P, physical.
two assessment tools. To have an equal number of male and
female vocal singers (n = 24) to perform the ANOVA, two female
singers whose VHI-10 and SVHI-10 scores were closest to the
The SVHI and the SVHI-10 were developed differently from mean were eliminated from the calculations. Finally, the mean
the VHI and the VHI-10. No attempt was made to identify an values of each statement were ranked from highest to lowest for
equal number of items related to physical, functional, and emo- both men and women.
tional parameters in the SVHI or the SVHI-10. Nonetheless, by
examining the statements, the current authors categorized the RESULTS
10 statements into the broad categories of physical, functional, A two-way ANOVA of gender by assessment type with repeat-
and emotional (Table 1). ed measures on gender demonstrated that there were no significant
Presently, there are no data that directly compare responses differences in VHI-10 or SVHI-10 scores between genders
of a large group of singers on both self-assessment tools. The (P = 0.865). For the assessment types, there was a significant
purpose of this study was twofold. First, to directly compare the difference between VHI-10 and SVHI-10 (P = 0.0003). There
scores from the VHI-10 and the SVHI-10 in a group of 50 singers was no significant interaction term indicating that men and women
and to examine the relative importance of each item as it relates performed similarly on both assessment tools.
to singers’ perception of voice severity. Second, to use the same
analysis to explore differences between male and female singers.
TABLE 3.
Singing Voice Handicap Index-10
METHODS
P It takes a lot of effort to sing 0 1 2 3 4
A retrospective chart review of 50 singers (26 women, 24 men)
F I am unsure of what will come out 0 1 2 3 4
presenting to a laryngology clinic was performed between June
when I sing
2014 and November 2014 at Weill Cornell Medical College, New P My voice “gives out” on me while I 0 1 2 3 4
York. This group consisted of both professional singers and stu- am singing
dents in a university vocal performance program. Table 2 presents E My singing voice upsets me 0 1 2 3 4
the demographics of the group. There were 24 men and 26 F I have no confidence in my singing 0 1 2 3 4
women. The age range was between 18 and 55 years. Of the 50 voice
subjects, 34 were performers earning their living from perform- P I have trouble making my voice do 0 1 2 3 4
ing, 14 were graduate students enrolled in a voice performance what I want it to
program, and 2 were professional singers currently not perform- P I have to “push it” to produce my 0 1 2 3 4
ing because of their voice problems. At their initial visit, all voice when singing
P My singing voice tires easily 0 1 2 3 4
subjects completed the VHI-10 and the SVHI-10 before their
E I feel something is missing in my life 0 1 2 3 4
evaluation. For comparison, two of the authors (E.R., T.M.) also
because of my inability to sing
reviewed the items of the SVHI-10 and compared them to the P I am unable to use my “high voice” 0 1 2 3 4
original statements of the VHI to classify them as functional,
Abbreviations: E, emotional; F, functional; P, physical.
physical, or emotional items (Table 3).
 ARTICLE IN PRESS
Elizabeth Renk, et al VHI-10 and SVHI-10 Differences in Singers’
TABLE 4.
Comparison of VHI-10 and SVHI-10
Notes: Highlighted texts denote the average scores of items with the highest weighting among the 10 statements for each questionnaire.
The mean score for VHI-10 was 12.1 and that of SVHI-10
was 20.39 (both out of a possible 40). Five of 50 subjects had
a score equal to or lower in the SVHI-10 than in the VHI-10.
When the scores of the 24 male vocal performers were com-
pared with those of the 26 female vocal performers, there was
no significant gender difference in the overall score between the
VHI-10 or the SVHI-10. For the VHI-10, men had a mean score
of 11.75, compared with 12.42 for women. On the SVHI-10, men
had a mean score of 20.39, compared with 20.38 for women.
We provide both t test and ANOVA statistical analysis of our
data. Historically, some authors have considered the responses
to the questions of the VHI or SVHI as ordinal data, which should
be analyzed using nonparametric methods, such as a t test. More
recently, authors have begun to view the data as interval data,
which would be better analyzed using parametric methods, such
as ANOVA or Pearson correlations.9,10 As there is no concensus
in the literature, it seems to the authors of this paper that in-
cluding both methods strengthened the statistical analysis.
When ranking the most severe items in the SVHI-10, as ranked
by the singers, we found that specific statements within the SVHI-
10 were of note. The statements “My voice problem upsets me”
(statement 8 on the VHI-10) and “My singing voice tires easily”
(statement 8 on the SVHI-10) each had the highest weighting
(most severe) among the 10 statements, with average scores of
2.8 and 2.4, respectively (Table 4). Statement 8 in the VHI-10
falls under the category of emotional statements. Statement 8
on the SVHI-10 could be considered a physical statement. When
we compared men and women, both ranked statement 8 on the
SVHI-10 as the most severe, with a score of 2.2 and 2.5,
respectively.
The second highest scores on each of the two forms were for
statements “The clarity of my voice is unpredictable” (ques-
tion number 7 on the VHI-10) and “I am unable to use my high
voice” (question number 10 on the SVHI-10) with average scores
of 1.8 and 2.3, respectively. These questions both address the
physical dimension of voice.
Interestingly, performers gave the statement “People ask,
‘What’s wrong with your voice?’” (question 10 on the VHI-
10) the lowest (least severe) average score, 0.7, despite their
overall concern with their singing voice. The lowest average score
on the SVHI-10 was an average score of 1, for the statement “I
have no confidence in my singing voice” (question 5).
3
DISCUSSION
This study demonstrates the significant difference in how singers
perceive their singing voice in comparison with their speaking
voice. The VHI-10 was designed to analyze spoken voice and
uses words like “my voice” or “when I talk,” whereas the SVHI-
10 uses “singing” or “my singing voice.” Our data indicate that
when singers respond to the SVHI-10, they identify their voice
problem as more severe than when responding to the VHI-10,
with scoring of 20.4 and 12.1, respectively.
Whereas singers may have general complaints about their voice,
they also have specific complaints that relate only to their singing
voice. This is demonstrated by the fact that performers rated the
statement “People ask, ‘What’s wrong with your voice?’” on the
VHI-10 very low in severity (average score of only 0.7) even
though their singing voice was still very bothersome to them.
The statements in the two assessment tools demonstrate a dif-
ferent focus. The VHI-10 lists five statements related to voice
function, whereas the SVHI-10 appears to have only two state-
ments related to function, and the remaining eight are related
to emotional and physical needs. Although our data show that
certain questions were more concerning to singers, no further
statistical analysis was performed. Further research is war-
ranted to explore differences in ranking of the individual items
that compose each survey.
In the current study, all of the 36 professional singers come
to the voice center with a primary complaint related to singing.
They were either not able to perform well in their current roles
or not performing at all because of their singing voice. All of
the 14 graduate students were in a vocal performance program
and could not maintain their class load because of their singing
voice. Clearly, these singers perceive an inability to perform a
physically demanding task, much like an athlete having little
trouble walking but having a significant problem running or
jumping. Moreover, it appears that the inability to perform the
physical task of singing contributes to emotional concern. These
data suggest the need to assess the physical needs of singers.
CONCLUSION
It is important for clinicians to understand that their patients who
are vocal performers perceive a marked distinction between talking
and singing. This study highlights the importance of using the
 ARTICLE IN PRESS
4 Journal of Voice, Vol. ■■, No. ■■, 2016
VHI-10 and the SVHI-10 as complementary questionnaires for
performers. When used in this manner, the clinician can iden-
tify specific complaints relating to the voice concerns of the patient
when singing and speaking. Furthermore, clinicians can use this
knowledge to guide clinical decision-making.
REFERENCES
1. Roy N, Barkmeier-Kraemer J, Eadie T, et al. Evidence-based clinical voice
assessment: a systematic review. Am J Speech Lang Pathol. 2012;22:212–
226.
2. Jacobson BJ, Johnson A, Grywalski C, et al. The voice handicap index (VHI).
Am J Speech Lang Pathol. 1997;6:66–70.
3. Rosen CA, Lee AS, Osborne J, et al. Development and validation of the
voice handicap index-10. Laryngoscope. 2004;114:1549–1556.
4. Phyland DJ, Oates J, Greenwood KM. Self-reported voice problems among
three groups of professional singers. J Voice. 1999;13:602–611.
5. Rosen CA, Murry T. Voice handicap index in singers. J Voice. 2000;14:370–
377.
6. Murry T, Zschommler A, Prokop J. Voice handicap in singers. J Voice.
2009;23:376–379.
7. Cohen SM, Jacobson BH, Garrett CG, et al. Creation and validation of the
Singing Voice Handicap Index. Ann Otol Rhinol Laryngol. 2007;116:402–
406.
8. Cohen SM, Statham M, Rosen CA, et al. Development and validation of
the Singing Voice Handicap-10. Laryngoscope. 2009;119:1864–1869.
9. Niebudek-Bogusz E, Woznicka E, Zamyslowska-Szmytke E, et al.
Correlation between acoustic parameters and Voice Handicap Index in
dysphonic teachers. Folia Phoniatr Logop. 2010;62:55–60.
10. Timmermans B, De Bodt M, Wuyts F, et al. Analysis and evaluation of a
voice-training program in future professional voice users. J Voice.
2004;19:202–210.