Voice Handicap in Singers

*Thomas Murry, †Anne Zschommler, and ‡Jan Prokop, *yzNew York, New York

Summary: The study aimed to determine the differences in responses to the Voice Handicap Index (VHI-10) between
singers and nonsingers and to evaluate the ranked order differences of the VHI-10 statements for both groups. The VHI-
10 was modified to include statements related to the singing voice for comparison to the original VHI-10. Thirty-five
nonsingers with documented voice disorders responded to the VHI-10. A second group, consisting of 35 singers with
voice complaints, responded to the VHI-10 with three statements added specifically addressing the singing voice. Data
from both groups were analyzed in terms of overall subject self-rating of voice handicap and the rank order of statements
from least to most important. The difference between the mean VHI-10 for the singers and nonsingers was not statis-
tically significant, thus, supporting the validity of the VHI-10. However, the 10 statements were ranked differently in
terms of their importance by both groups. In addition, when three statements related specifically to the singing voice
were substituted in the original VHI-10, the singers judged their voice problem to be more severe than when using
the original VHI-10. The type of statements used to assess self-perception of voice handicap may be related to the sub-
ject population. Singers with voice problems do not rate their voices to be more handicapped than nonsingers unless
statements related specifically to singing are included.
Key Words: VHI-10–Singer’s–Voice handicap.

INTRODUCTION In another study, sick singers were evaluated regarding the

Speech-language pathologists confronted with a singer who has
a voice problem are often at a disadvantage not knowing
whether to treat the patient’s speaking voice issues or attempt
to understand and treat singing voice issues. The Voice Handi-
cap Index (VHI) developed in 1997 by Jacobson et al1,5 is a pen-
cil-and-paper tool to scale the severity of the voice problem
according to the patient’s perception of his or her voice handi-
cap. The VHI consists of 30 statements, addressing emotional,
physical, and functional aspects of voice disorders. When com-
pleting the VHI, the patient is asked to rate each statement with
a score between 0 (never) and 4 (always), resulting in a mini-
mum total score of 0 and a maximum total score of 120.
Thus, the more severe the patient perceives the voice handicap,
the higher the score of the VHI is. The VHI was validated and
its reliability and internal consistency were established for the
speaking voice.1,5 None of the 30 statements are specifically
directed to issues of singing. One question asks if the voice
problem has caused the patient to lose income, which may be
considered related to the singing voice.
In 2000, Rosen and Murry presented the VHI to 100 singers
and reported that singers’ ratings of their voice handicap were
significantly lower than those of nonsingers with similar diag-
noses.2 These results were similar for classical singers and
singers with other singing styles. Rosen and Murry did find
that certain questions were generally responded to with greater
severity than other questions. Nonetheless, all of these ques-
tions were related to the speaking voice only.
Accepted for publication October 17, 2007.
Presented at the 36th Annual Symposium: Care of the Professional Voice, June 2, 2007,
Philadelphia, Pennsylvania.
From the *Department of Otolaryngology—Head and Neck Surgery, Columbia Univer-
sity, New York, New York; yDepartment of Speech and Hearing, New York Presbyterian
Hospital, New York, New York; and the zDepartment of Music, New York University, New
York, New York.
Address correspondence and reprint requests to Thomas Murry, Otolaryngology—Head
and Neck Surgery, Columbia University, College of Physicians and Surgeons, 180 Fort
Washington Avenue, HP 8-812, New York, NY 10032. E-mail: tm2103@columbia.edu
Journal of Voice, Vol. 23, No. 3, pp. 376-379
0892-1997/$36.00
Ó 2009 The Voice Foundation
doi:10.1016/j.jvoice.2007.10.013
time between their first voice symptoms and the time they con-
sulted a doctor for an examination.3 Results indicated that in
general, classically trained singers reported voice problems
sooner than those who were not classically trained. Singers
with voice problems are frequently referred to speech-language
pathologists.
Recently, the VHI was modified to a 10-statement tool that
was found to be highly correlated with the 30 statements on
the original VHI.4 Ten statements were selected on the basis
that they had the highest weighted difference between the
responses of normal subjects and patients with voice disorders.
However, because the statements of the VHI-10 were derived
from the original VHI, there are no specific statements related
to the singing voice. Singers are typically very sensitive to
minor nuances and variations in their singing voice, identifying
voice abnormalities during singing that are not perceived alike
during speaking. Yet, no data exist on the VHI-10 as it relates to
the voice problems of singers. Because the singers score low on
the VHI, it was hypothesized that statements on the VHI-10 that
would specifically target the singing voice may better represent
the singers’ degree of handicap.
The purpose of this study is first to determine the differences
in responses to statements on the VHI-10 between singers and
nonsingers. This study aims to rank the statements of the VHI-
10 by their clinical value as perceived by singers versus non-
singers. A second purpose of this study was to determine how
singers with vocal complaints (sick singers) respond differently
to an outcome survey when statements specifically related to
the singing voice are added or substituted to the original
VHI-10.
METHOD
Modification of the VHI-10
The VHI-10 was modified by substituting three statements similar
to those of the original VHI-10 but with the words ‘‘singing
voice’’ replacing ‘‘speaking voice.’’ These three statements
came from a panel of singing teachers who were asked to review
Thomas Murry, et al Voice Handicap in Singers 377

the VHI-10 and suggest statements similar to the VHI-10, but that
would specifically address the singing voice. The three statements
selected were based on the number of times they were suggested
by a panel of 20 singing teachers. The three statements were
merged into the VHI-10 thus allowing comparison of the original
VHI-10 to the VHI-10 with the substituted statements now called
the VHI-10 S (Appendix A). Thus, it was possible to compare the
responses of the singers and nonsingers to the VHI-10 and to com-
pare the responses of the VHI-10 to the VHI-10 S in the group of
35 singers. The modified VHI-10 S is identical to the original
VHI-10 except that the term ‘‘speaking voice’’ was changed to
‘‘singing voice’’ within the three statements.
Subjects
Two groups of 35 subjects responded to the VHI-10 and the
VHI-10 S. One group consisted of 35 singers at various stages
in their careers from student performer to seasoned performer.
The second group consisted of 35 nonsingers. The singers
ranged from 18 to 29 years of age with a mean of 19.7 years.
The nonsingers ranged in age between 17 and 97 years with
a mean of 41.2 years. The singers consisted of 18 females
and 17 males, whereas the group of nonsingers included 19 fe-
males and 16 males. The singers were accepted into the study if
they and their voice teacher reported problems in their singing
voice and were referred for medical examination of their voice.
Not all subjects ultimately underwent a medical examination.
All subjects agreed to participate and completed the New
York University informed consent form. All subjects were
seen at New York University between January and May 2004.
The nonsingers completed the original VHI-10 with a maximum
score of 40. The singers completed the VHI-10 S (maximum
score of 40) with the three singing voice statements merged
into the VHI-10 form. Both groups received the same instruc-
tions for rating the statements, which were
‘‘These are the statements that many people have used to
describe their voices and the effects of their voices on their
lives. Check the response that indicates how frequently
you had the same experience in the past two weeks
(0 ¼ Never, 1 ¼ Almost never, 2 ¼ Sometimes, 3 ¼ Almost
always, 4 ¼ Always).’’
No assistance was given regarding the ratings itself to either
of the two groups.
Analysis
First, the scores from the original VHI-10 were compared be-
tween the two groups. This was followed by an item analysis
of the 10 statements to determine how the individual items
were ranked in their importance for each group. Age, gender,
and mean scores of both groups were calculated. A t test for
independent measures was used to determine if there was a dif-
ference between groups.
Each singer’s mean VHI-10 and mean VHI-10 S scores were
compared to determine if there was a difference in the two sets
of scores. For this comparison, statements 5, 6, and 7
(Appendix A) were substituted for speaking voice statements
shown in Appendix A.
RESULTS
Singers and nonsingers
The mean VHI-10 score for the singers was 16.3 as compared to
a mean of 14.0 for the nonsingers. The mean score for each of
the 10 statements and the total mean for the VHI-10 scores of
both study groups are displayed in Table 1. A t test indicates
that the overall difference between the singers and nonsingers
was not significant at P < 0.05. Further examination of the
mean scores for each item of the VHI-10 revealed a different
rank of importance for singers compared to nonsingers. For
example, statement 8 ‘‘My voice problem upsets me’’ was the
most important for the group of singers, but ranked eighth by
the nonsingers. Statement 10 ‘‘People ask: What’s wrong with
your voice?’’ was rated most important by the group of nonsing-
ers, whereas ranking fourth for the group of singers.
Table 2 shows the ranked order of each of the 10 VHI-10
statements for nonsingers and singers and their differences in
ranking. For example, the statement ‘‘My voice difficulties
restrict personal and social life’’ was the sixth highest mean
score for nonsingers and the fifth highest score for singers,
a difference of one. The mean difference between the singers
and nonsingers across the 10 statements was three with a maxi-
mal difference of seven for the statement: ‘‘My voice problem
upsets me.’’ Thus, although the mean scores do not distinguish
severity between the group of singers and the nonsingers with
voice problems, specific statements based on the ranked
differences show that singers and nonsingers have marked dif-
ferences in what they consider as handicap.

TABLE 1.
Results of the Mean Scores for Each Item of the VHI-10 Obtained from the 35 Singers and 35 Nonsingers
VHI-10 Q # 1 2 3 4 5 6 7 8 9 10 Total
Singers Mean 1.23 1.66 1.66 1.31 0.74 1.74 1.69 2.65 2.12 1.54 16.33
SD 0.97 1.08 1.11 1.13 1.10 1.15 1.08 1.30 1.37 1.04 11.32
Sum 43 58 58 46 17 61 59 90 72 54 558
Nonsingers Mean 1.57 1.8 1.4 0.89 0.91 1.74 1.34 1.06 1.43 1.89 14.03
SD 0.74 0.83 1.03 0.93 0.92 0.92 0.76 0.73 0.74 0.87 8.47
Sum 55 63 79 31 32 61 47 37 50 66 491
t with P < 0.05 (df ¼ 18) ¼ 1.17 NS.
Abbreviation: SD, standard deviation; NS, nonsingers.
378 Journal of Voice, Vol. 23, No. 3, 2009

responded differently to five of the VHI-10 statements as indi-

TABLE 2.
Differences in the Severity of Statements as Scored by
cated by the rank of each of the 10 items. Overall, the singers
the NS and S were more concerned about emotional factors in their singing
voice, whereas nonsingers were more concerned about loud-
NS S Difference ness. Emotional statements evoked the greatest response by
My voice difficulties restrict 6 5 1 singers when responding to the VHI-10 (‘‘My voice problem
personal and social life upsets me’’ and ‘‘My voice makes me feel handicapped’’).
I feel left out of conversations 10 8 2 For the nonsingers, statements relating to loudness and use of
because of my voice the voice in social situations were more important.
My voice problem causes me to 9 10 1
Based on this study, there was an 11% higher perceived hand-
lose income
icap when statements specifically referred to the singing voice
I feel as though I have to strain to 3 3 0
produce voice (VHI-10 S). Thus, the VHI-10 may not reflect the degree of
The clarity of my voice is 7 7 0 handicap that sick singers perceive in their voices.
unpredictable The fact that there was no significant difference between the
My voice problem upsets me 8 1 7 VHI-10 and the VHI-10 S scores within the singer subjects may
My voice makes me feel 5 2 3 lead one to think that statements regarding singing may be
handicapped substituted for those related to the speaking voice. We do not
People ask: ‘‘what’s wrong with 1 4 3 interpret these data in that light. Rather, we interpret the data
your voice?’’ to suggest that the VHI-10 is a stable tool to assess self-
My voice makes it difficult for 4 9 5 perceived voice handicap. Although the ranked orders of the
people to hear me
statements in terms of handicap were different for the singers
People have difficulty 2 6 4
compared to the nonsingers, the lack of a significant difference
understanding me in a noisy room
implies the strength of the VHI-10.
For example, the statement ‘‘My voice difficulties restrict personal and so-
cial life’’ was sixth highest in severity for nonsingers and fifth for singers,
Although the statements related to singing may give the cli-
a difference of one. nician greater insight into the specific voice problem, the results
Abbreviations: NS, nonsingers; S, singers. of this study must be interpreted in the same light of other ques-
tionnaires that have yet to be validated.6 The authors suggest
that although the clinical value of the VHI-10 S may be helpful
Singers VHI-10 and VHI-10 S in treatment planning, the results of this study should not be
Examination of the VHI-10 S (Table 3) shows that there was an considered a valid indication of self-perceived handicap in
increase in the mean scores of the singers from 16.33 to 18.05 a population of singers with voice problems. The three items
when the three singer statements were substituted for the three substituted may have construct validity; however, no attempt
original VHI-10 statements. Thus, there was an increase of 11% was made to obtain construct validity of the VHI-10 S within
in perceived handicap when singers were presented with the 10 the normal procedures of test development.7 As indicated by
statements of the VHI-10 S compared to the original VHI-10. Pusic et al, a number of factors are involved in test construction
Although this change does not reflect a statistically significant and validation.8 In this study, the authors were merely interested
difference, the clinical impact of these findings may have in determining the differences between singers and nonsingers
treatment implications. In addition, the trend is in the direction responses to the VHI-10. Substituting three statements in the
that statements related to the singing voice reflect a greater self- VHI-10 was primarily done to determine the strength of the
perception of handicap. VHI-10, not to suggest that it should be substituted as a new
assessment tool. Although the scores on the VHI-10 S were
DISCUSSION 11% higher, the difference was not statistically significant. In
The average scores on the VHI-10 for singers and nonsingers fact, to some degree, the results of this study point out the
did not differ significantly. However, singers and nonsingers strength of the VHI-10.

TABLE 3.
Mean, SD, and t Test Results for the Mean VHI-10 Compared to the Mean VHI-10 S Scores for 35 Singers
VHI-10 Q # 1 2 3 4 5 6 7 8 9 10 Total
VHI-10 Mean 1.23 1.66 1.66 1.31 0.74 1.74 1.69 2.65 2.12 1.54 16.33
VHI-10 S Mean 1.23 1.66 1.66 1.31 1.38 2.29 2.23 2.65 2.12 1.54 18.05
t Test 0.777
SD 0.498
P value 0.45
Statements 5, 6, and 7 are the statements related to the singing voice substituted from the original VHI-10 statements as shown in Appendix A.
Abbreviation: SD, standard deviation.
Thomas Murry, et al Voice Handicap in Singers
In summary, this study addressed the need for a tool to assess
singer’s self-perception of their singing handicap. Singers esti-
mated their handicap different from nonsingers based on the
ranked importance of specific statements, but not in terms of
their overall handicap. They do not, however, estimate their
handicap as significantly less when the statements are not spe-
cific to the singing voice. Thus, the VHI-10 S may provide cli-
nicians with additional information that may not be gathered by
the VHI-10. Although the VHI-10 S may reflect a more severe
self-perceived voice handicap than the original VHI-10, test
validation and reliability studies related to the singing voice re-
main to be done.
REFERENCES
1. Benninger MS, Gardner GM, Jacobson BH, Grywalski C. New dimensions
in measuring voice treatment outcomes. In: Sataloff RT, ed. Professional
Voice: The Science and Art of Clinical Care. 2nd ed).. San Diego, CA: Sin-
gular Publishing Group; 1997:789-794.
2. Rosen CA, Murry T. Voice Handicap Index in singers. J Voice. 2000;14:
370-377.
3. Mishra S, Rosen CA, Murry T. 24 Hours prior to curtain. J Voice. 2000;14:
92-98.
4. Rosen CA, Lee AS, Osborne J, Zullo T, Murry T. Development and valida-
tion of the Voice Handicap Index-10. Laryngoscope. 2004;114:1549-1556.
5. Jacobsen BH, Johnson A, Grywalski C, Silbergleit A, Jacobson G,
Benninger MS. The Voice Handicap Index (VHI): development and valida-
tion. Am J Speech Lang Pathol. 1997;6:66-70.
6. Scientific Advisory Committee of the Medical Outcomes Trust. Assessing
health status and quality of life instruments: attributes and review criteria.
Qual Life Res. 2002;11:193-205.
7. Franic DM, Bramlett RE, Boothe AC. Psychometric evaluation of disease
specific quality of life instruments in voice disorders. J Voice. 2005;19:
300-315.
8. Pusic A, Liu JC, Chen CM, et al. A systematic review of patient-reported out-
come measures in head and neck cancer surgery. Otolaryngol Head Neck
Surg. 2007;136:525-535.
379
Appendix A
The VHI-10 S with the three additional statements shown in
bold, followed by the three original statements used to calculate
the VHI-10.
1. My voice makes it difficult for people to hear me in nor-
mal conversation.
2. People have difficulty understanding me in a noisy
room.
3. My voice difficulties restrict personal and social life.
4. I feel left out of conversations because of my voice.
5. My singing voice problem causes me to lose income.
6. I feel I have to strain to produce my singing voice.
7. The clarity of my singing voice is unpredictable.
8. My voice problem upsets me.
9. My voice makes me feel handicapped.
10. People ask: ‘‘What’s wrong with your voice?’’
Substituted statements:
My voice problem causes me to lose income.
I feel I have to strain to produce my speaking voice.
The clarity of my voice is unpredictable.
Scoring: 0 1 2 3 4.
0 ¼ Never 1 ¼ Almost never 2 ¼ Sometimes 3 ¼ Al-
most always 4 ¼ Always.
Note: Statements 5, 6, and 7 were not in bold when presented
to the subjects.