Otolaryngology–Head and Neck Surgery (2008) 138, 430-434

ORIGINAL RESEARCH—LARYNGOLOGY AND NEUROLARYNGOLOGY

Factors associated with perception of singing voice handicap Seth M. Cohen, MD, MPH, J. Pieter Noordzij, MD, C. Gaelyn Garrett, MD, and Robert H. Ossoff, DMD, MD, Durham, NC; Boston, MA; and Nashville, TN OBJECTIVE: This study will determine factors that influence the self-perceived handicap associated with singing voice problems. STUDY DESIGN: A prospective cohort. SUBJECT AND METHODS: Singers presenting to a voice clinic prospectively completed the Singing Voice Handicap Index (SVHI) before evaluation and treatment. Demographic data, singing style, professional status, duration of symptoms, medical problems, and diagnosis were collected. Univariate and multivariate analysis was performed. RESULTS: One hundred seventy-one singers completed the SVHI. The duration of symptoms, being an amateur singer or singing teacher, benign vocal fold lesions, and neurologic voice disorders were associated with increased SVHI scores (P ⬍ 0.05, multiple linear regression). Age greater than 50 years and gospel singing were predictive of increased SVHI scores only on univariate analysis (P ⬍ 0.05, t test). CONCLUSION: Singers experience significant handicap as a result of their singing problems with certain factors associated with greater impairment. Targeting interventions at patients more severely affected may improve outcomes. © 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.

P

rior reports have suggested that singers are at a high risk for voice problems and perceive their vocal impairment differently than nonsingers.1-3 Specifically, singers appear more affected by small changes in their voice, seeking medical care more readily than nonsingers and may have lower Voice Handicap Index (VHI) scores, less voice handicap, than nonsingers.1-4 However, the VHI is not specific for singing-related vocal problems. Additionally, because most comparisons have been between singers and nonsingers, little is known about factors that influence the perception of voice problems within the singing population. The newly validated Singing Voice Handicap Index (SVHI) provides a tool to assess the voice handicap resulting from singing voice problems.5 Understanding how voice problems impact singers is essential to maximize treatment.

To further determine the impact of voice problems among singers, this study will examine potential predictors of the patient-perceived voice handicap among singers as measured on the SVHI. The hypothesis is that certain variables will be identified that predict the SVHI score.

METHODS The Vanderbilt Institutional Review Board approved the study. New voice patients who were singers presenting to a tertiary care voice clinic prospectively completed the SVHI. The SVHI is a newly validated singing voice specific health status instrument containing 36 questions, each scaled from zero to four. Raw scores range from zero to 144 and are scaled to range from zero to 100. The SVHI is scored as a single scale, and a higher score indicates more voice handicap.5 Patients also determined whether singing was a primary, secondary, or not a source of income, and classified their singing status as professional singer, singing student, amateur singer, or singing teacher; they also categorized their singing styles as country, choral, classical, pop, rock, or gospel. Patients could sing in more than one style. Patients also noted whether they had gastroesophageal reflux (GER) based on symptoms of heartburn or regurgitation at least monthly. They also noted whether they had allergic rhinitis based on symptoms of sneezing; rhinorrhea; itchy throat; or itchy, watery eyes. The length of the patients’ voice problem, age, sex, and diagnosis were collected. Diagnosis was determined by either a 90° rigid or flexible distal chip videostroboscopy. Examinations were viewed by either a masters- or PhD-level speech pathologist and a fellowshiptrained laryngologist, and a consensus diagnosis was made. Statistical analysis was performed with SigmaStat 2.03 (SPSS Inc., Chicago, IL). Univariate analysis was performed between potential predictors (age, length of voice problem, income status, singing style, singing status, neu-

Received August 17, 2007; revised December 19, 2007; accepted December 28, 2007.

0194-5998/$34.00 © 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2007.12.041

Cohen et al

Factors associated with perception of singing . . .

rologic voice disorder, benign vocal fold lesion, vocal fold bowing, GER, allergic rhinitis, and sex) and the SVHI. Analysis of variance (ANOVA) or t tests were performed between categorical data and the SVHI. The Spearman correlation analyzed the relationship between continuous variables and the SVHI. Multivariate analysis was performed with multiple linear regression. Variables with a P value ⱕ0.1 on univariate analysis were entered into the multiple linear regression. For a reliable regression model, at least 10 subjects per potential predictor should be included.6 A goal of 15 patients per potential predictor with 11 potential predictors leads to a sample size of 165.

RESULTS One hundred seventy-one singers prospectively completed the SVHI. The mean age was 38.7 years with a range of 18 to 81 years with 61.4% female and 38.6% male. The patient diagnoses are shown in Table 1. Singing was not a source of income for 50.9%, a secondary source of income for 23.4%, and a primary source of income for 22.2% of patients. The singing status among the cohort was 40.9% professional singers, 24.0% amateur singers, 12.9% singing teachers, and 16.4% singing students. Singing styles var-

Table 1 Patient diagnoses and mean SVHI scores and standard deviations Diagnosis Benign vocal fold lesions* Edema Nodule Chronic laryngitis Cyst Polyp Acute laryngitis Leukoplakia Varix Scar Hemorrhage Granuloma Neurologic voice disorder* Vocal fold paralysis Superior laryngeal nerve paresis Spasmodic dysphonia Muscle tension dysphonia* Chronic laryngitis* Acute laryngitis* Bowing* Unknown

Number Mean SVHI

SD

79

45.8

18.7

19 25 17 7 22 9 1 2 1 1 1 10

42.6 38.3 29.2 67.8 49.2 34.6 44.4 44.8 50.0 51.0 56.9 73.6

18.5 16.8 18.2 18.2 19.2 25.1

23.2

5 3

86.1 83.0

10.9 10.3

2 32

60.1 42.1

18.9

19 9 7 15

30.0 34.6 54.9 47.2

14.6 25.1 18.8 22.7

*Diagnostic categories used in analysis of diagnosis and SVHI.

431

ied with 32.7% country, 35.7% classical, 49.7% choral, 34.5% pop, 20.5% rock, and 48.5% gospel singers among the cohort. Univariate analysis was performed between potential predictors and the SVHI score. Patients greater than 50 years of age had higher SVHI scores (mean 51.3, standard deviation [SD] 23.2) than those younger than 50 years (mean 42.6, SD 20.0) (P ⫽ 0.017, t test). The longer patients were symptomatic the higher the SVHI score (Spearman correlation ⫽ 0.32, P ⬍ 0.001). No statistically significant associations were seen between the presence of GER, allergic rhinitis, sex, and the SVHI score (P ⬎ 0.1). Patients for whom singing was not a source of income had higher SVHI scores (mean 48.4, SD 20.9) than those for whom singing was a secondary source of income (mean 38.0, SD 18.9) (P ⫽ 0.017, ANOVA; P ⫽ 0.026, Bonferroni t test) but not compared with those for whom singing was a primary source of income (mean 40.7, SD 20.7) (P ⫽ 0.017, ANOVA; P ⫽ 0.17, Bonferroni t test). Amateur singers (mean 55.7, SD 18.7) had higher SVHI scores than professional singers (mean 37.8, SD 19.5) (P ⬍ 0.001, ANOVA; Bonferroni t test, P ⫽ 0.001) and singing students (mean 37.8, SD 19.3) (P ⬍ 0.001, ANOVA; Bonferroni t test, P ⫽ 0.001). Additionally, singing teachers (mean 52.7, SD 19.8) had higher SVHI scores than professional singers (mean 37.8, SD 19.5) (P ⬍ 0.001, ANOVA; Bonferroni t test, P ⫽ 0.001) and singing students (mean 37.8, SD 19.3) (P ⬍ 0.001, ANOVA; Bonferroni t test, P ⫽ 0.001). Because patients who sang country, classical, pop, choral, and rock had mean SVHI scores within four points of each other, they were combined into a single group of nongospel singers. Gospel singers (mean 49.2, SD 21.4) had worse SVHI scores than nongospel singers (mean 39.2, SD 19.2) (P ⫽ 0.002, t test). To compare associations between diagnosis and SVHI scores, diagnoses were grouped into categories as shown in Table 1. Patients with a neurologic diagnosis had higher SVHI scores than any other laryngeal pathology group (P ⱕ 0.001, ANOVA; Bonferroni t test, P ⬍ 0.001 for all comparisons of neurologic versus other diagnostic categories). Within the benign vocal fold lesions, cysts had higher SVHI scores than nodules (P ⫽ 0.003, ANOVA; Bonferroni t test, P ⫽ 0.002) but not than polyps (P ⫽ 0.003, ANOVA; Bonferroni t test, P ⫽ 0.07). Multiple linear regression was performed with the dependent variable SVHI score and independent variables associated with SVHI score (with P value ⱕ 0.1) including age ⱖ50, length of voice problem, income status, amateur singer or singing teacher, gospel singer, neurologic voice problem, and benign vocal fold lesions. Vocal fold bowing was also included because of its association with elderly patients. The length of a voice problem, neurologic voice disorder, benign vocal fold lesions, and being an amateur singer or singing teacher were predictive of the SVHI score (Table 2).

432

Otolaryngology–Head and Neck Surgery, Vol 138, No 4, April 2008

Table 2 Multiple linear regression between SVHI and possible predictors Variable

Regression coefficient

Standard error

Constant Age ⱖ50 Length of voice problem Neurologic voice disorder Benign vocal fold lesion Vocal fold bowing Gospel singing style No income from singing Amateur singer or singing teacher

26.6 ⫺3.2 0.2 32.6 8.6 8.9 5.7 5.1 11.5

3.4 4.1 0.1 7.7 3.3 8.3 3.1 3.3 3.5

DISCUSSION Singers are a unique population at risk for voice problems. Yet, the lack of a specific singing-voice health status instrument has limited the evaluation of voice problems among singers. Determining the factors that influence singers’ perceived dysphonia will facilitate their evaluation and treatment. Using the SVHI, this study explored predictors of patient-perceived singing voice handicap. The type of laryngeal pathology was associated with patients’ SVHI scores. Although they were a relatively uncommon etiology in our cohort, neurologic voice disorders had a severely negative effect on the SVHI scores. Similar negative impact has been shown between neurologic voice problems and voice handicap as measured on the VHI, with neurologic laryngeal pathology having the largest impact compared with other etiologies.7,8 Whether patients can recover their singing voice after treatment remains unknown. The presence of benign vocal fold lesions also was a predictor of patients’ SVHI scores. Among the benign vocal fold lesions, cysts, followed by polyps, had the highest SVHI scores. Likewise, Rosen et al1 found that singers with polyps or cysts had higher VHI scores than nodules. Presumably, polyps and cysts have more adverse impact on vocal fold closure and mucosal wave than nodules. Although our study was not designed to assess how specific videostroboscopic factors, such as vocal fold closure and mucosal wave, impact the singing voice handicap, understanding how laryngeal pathophysiology impacts patients’ perception of voice problems is an important avenue of future investigation. Additionally, to maximize treatment outcome of benign vocal fold lesions, studies comparing SVHI results from different surgical procedures and voice therapy are necessary. How long singing patients have had a voice problem influenced their perceived handicap. Patients with longer durations of voice complaints had more voice handicap on the SVHI. In contrast, Behrman et al2 found a nonstatistically significant inverse association between VHI scores and the chronicity of a voice problem. Our use of a 100% singing patient population and a different health status instrument may explain the disparate findings. Potentially,

95% confidence interval ⫺11.2, 0, 17.5, 2.1, ⫺7.4, ⫺0.4, ⫺1.4, 4.6,

4.8 0.4 47.7 15.1 25.2 11.8 11.6 18.4

P value ⬍0.001 0.4 0.04 ⬍0.001 0.01 0.3 0.07 0.1 0.001

patients adjust to their voice problem with time and experience less handicap. How patients adapt to voice problems economically and socially is worthy of continued investigation. Furthermore, determining the factors that drive patients to seek medical treatment would help treating physicians meet patient needs. The role singing served in patients’ lives had an important predictive effect on the SVHI scores. Amateur singers had worse SVHI scores than professional singers. More voice handicap in recreational singers compared with professional singers has been previously shown.1 Because of the economic impact, professional singers may be more likely to seek medical attention earlier when problems do occur and present with less singing voice handicap. In addition, Zeine and Waltar9 found that professional actors knew more about vocal hygiene than acting students. Better vocal health knowledge among professional singers may also reduce the likelihood of developing severe voice problems. Voice health education programs may prove beneficial in preventing dysphonia in targeted and more general audiences. In fact, training and education among professional voice users has led to improved voice quality.10 Additionally, prior reports have noted that teachers are at a high risk for voice problems and resultant lost work productivity.11,12 Furthermore, compared with other teachers, singing teachers had a greater risk of voice problems and were more likely to have chronic voice disorders.13 In a videostroboscopy study among singing teachers, vocal fold lesions were prevalent in those with and without symptoms.14 Our study also found that singing teachers were a high-risk population for impaired singing voice handicap. Thus, [voice-damage-prevention] programs are essential and have shown potential among teachers.15,16 Whether vocal health education and training translates into less voice handicap, better vocal health, and reduces the incidence of voice problems are important areas of continued study. Although age was not predictive of SVHI scores on multivariate analysis, it may have an important impact on voice handicap. Voice problems are highly prevalent among the aging population.17,18 Changes in neurologic and pulmonary function and laryngeal alterations may result in

Cohen et al

Factors associated with perception of singing . . .

voice symptoms. These voice problems may negatively impact patients’ voice handicap as seen in our univariate analysis. Professional singers may lose income, and amateur singers may become reluctant to sing in their choirs, limiting important socialization. Continued investigation about how age impacts patients’ perception of voice problems could lead to improved means of vocal rehabilitation in the elderly. Certain methodologic issues need to be addressed. First, patients noted whether they had GER or allergic rhinitis based on the presence of specific symptoms. Specifics regarding type and length of medication treatment and compliance with medical management were not known and could potentially influence associations between GER or allergic rhinitis and singing voice handicap. Patients with untreated GER or allergic rhinitis would intuitively be more likely to suffer impaired voice handicap than those with treated GER or allergic rhinitis. Thus, although allergic rhinitis has been suggested as contributing to voice problems in singers, our study did not find any association between allergic rhinitis and the SVHI score.19 Also, objective testing for GER as a cause of voice symptoms was not routinely conducted and may have further affected our ability to find an association between GER and the SVHI scores. Further study to assess the relationship between these variables and the SVHI is warranted. Additionally, perceptual voice analysis and acoustic and aerodynamic measures were not routinely performed, limiting our ability to assess associations between these parameters and the SVHI. Lastly, most of our patients sang more than one style. Although gospel singing was associated with higher SVHI scores than nongospel singing on univariate analysis, studies designed to specifically address how different singing styles impact the singing voice handicap and vocal health are necessary.

CONCLUSIONS Singers experience impaired voice handicap as a result of their voice problem. Several predictors of the singing voice handicap as measured on the SVHI were identified. The chronicity of vocal problems and the specific etiology, such as neurologic voice disorders and benign vocal fold lesions, were associated with worse singing voice handicap. Additionally, amateur singers and singing teachers are more prone to worse SVHI scores. Age and singing style may also be important predictors of singing voice handicap requiring further investigation. Future studies should also compare singing voice handicap outcomes after surgical and nonsurgical treatments.

AUTHOR INFORMATION From the Duke Voice Care Center, Duke University Medical Division of Otolaryngology–Head and Neck Surgery Center, Durham, NC (Dr Cohen); Boston University, Department of Otolaryngology–Head and Neck Sur-

433

gery, Boston, MA (Dr Noordzij); and Vanderbilt University Medical Center, Vanderbilt Voice Center, Nashville, TN (Drs Garrett and Ossoff). Presented at the American Academy of Otolaryngology–Head and Neck Surgery Foundation Annual Meeting, Washington, DC, September 16-19, 2007. Corresponding author: Dr Seth Cohen, DUMC Box 2805, Durham, NC 27710. E-mail address: [email protected].

AUTHOR CONTRIBUTIONS Seth M. Cohen, design, data collection, data analysis, writer; J. Pieter Noordzij, design, data collection; C. Gaelyn Garrett, design, data collection; Robert H. Ossoff, design, data collection.

FINANCIAL DISCLOSURE None.

REFERENCES 1. Rosen CA, Murry T. Voice Handicap Index in singers. J Voice 2000; 14:370 –7. 2. Behrman A, Sulica L, He T. Factors predicting patient perception of dysphonia caused by benign vocal fold lesions. Laryngoscope 2004; 114:1693–700. 3. Phyland DJ, Oates J, Greenwood KM. Self-reported voice problems among three groups of professional singers. J Voice 1999;13:602–11. 4. 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. 5. Cohen SM, Jacobson BH, Garrett CG, et al. Creation and validation of the Singing Voice Handicap Index. Ann Otol Rhinol Laryngol 2007; 116:402– 6. 6. Harrell FE Jr. Regression Modeling Strategies: with Applications to Linear Models, Logistic Regression, and Survival Analysis. SpringerVerlag. New York, NY; 2001. p. 60 – 61. 7. Cohen SM, Dupont WD, Courey MS. Quality of life impact of nonneoplastic voice disorders: a meta-analysis. Ann Otol Rhinol Laryngol 2006;115:128 –34. 8. Benninger MS, Ahuja AS, Gardner G, et al. Assessing outcomes for dysphonic patients. J Voice 1998;12:540 –50. 9. Zeine L, Waltar KL. The voice and its care: survey findings from actors’ perspectives. J Voice 2002;16:229 – 43. 10. Timmermans B, de Bodt MS, Wuyts FL, et al. Analysis and evaluation of a voice-training program in future professional voice users. J Voice 2005;19:202–10. 11. Roy N, Merrill RM, Thibeault S, et al. Voice disorders in teachers and the general population: effects on work performance, attendance, and future career choices. J Speech Lang Hear Res 2004;47:542–51. 12. Roy N, Merrill RM, Thibeault S, et al. Prevalence of voice disorders in teachers and the general population. J Speech Lang Hear Res 2004;47:281–93. 13. Thibeault SL, Merrill RM, Roy N, et al. Occupational risk factors associated with voice disorders among teachers. Ann Epidemiol 2004; 14:786 –92. 14. Heman-Ackah YD, Dean CM, Sataloff RT. Strobovideolaryngoscopic findings in singing teachers. J Voice 2002;16:81– 6. 15. Bovo R, Galceran M, Petruccelli J, et al. Vocal problems among teachers: evaluation of a preventive voice program. J Voice 2007;21:705–22.

434

Otolaryngology–Head and Neck Surgery, Vol 138, No 4, April 2008

16. Chan RW. Does the voice improve with vocal hygiene education? A study of some instrumental voice measures in a group of kindergarten teachers. J Voice 1994;8:279 –91. 17. Golub JS, Chen PH, Otto KJ, et al. Prevalence of perceived dysphonia in a geriatric population. J Am Geriatr Soc 2006;54:1736 –9.

18. Roy N, Stemple J, Merrill RM, et al. Epidemiology of voice disorders in the elderly: preliminary findings. Laryngoscope 2007;117:628 –33. 19. Hamdan AL, Sibai A, Youssef M, et al. The use of a screening questionnaire to determine the incidence of allergic rhinitis in singers with dysphonia. Arch Otolaryngol Head Neck Surg 2006;132:547–9.