- Email: [email protected]
Voice Problems Amongst Primary School Teachers in Singapore
Voice Problems Amongst Primary School Teachers in Singapore Tze Choong Charn and Paul Kan Hwei Mok, Singapore, Singapore Summary: Hypothesis/Objectives. Teachers are often cited to be at high risk of vocal disturbances. Many studies were performed in the West, but none locoregionally. The aim of this study is to determine the prevalence of voice problems amongst primary school teachers in six schools in Singapore and explore the associated risk factors. Methods. A cross-sectional study was conducted across six primary schools and 214 full-time teachers were surveyed. Teachers were asked to report if they have voice problems at the day of the interview, during the past 1 year and throughout their careers. Teachers who reported having current voice problems were invited for videolaryngostroboscopy. Multivariate analyses were performed to analyze risk factors associated with voice problems. Results. The point prevalence was 13.1% (95% confidence intervals [CI]: 9.3–18.3%). The past year prevalence and career prevalence were 25.4% (95% CI: 20–31.6%) and 32.1% (95% CI: 26.2–38.6%), respectively. Results showed that reflux symptoms (adjusted odds ratio [adj OR] ¼ 6.1, CI: 2.5–14.6) and microphone usage (adj OR ¼ 5.6, CI: 1.8–17.6) are risk factors for current voice disorders. Allergic rhinitis symptoms (adj OR ¼ 2.1, CI: 1.1–4.3), hypothyroid-like symptoms (adj OR ¼ 2.6, CI: 1.3–5.1), and microphone use (adj OR ¼ 3.4, CI: 1.1– 10.0); allergic rhinitis symptoms (adj OR ¼ 2.6, CI: 1.4–5.1), reflux symptoms (adj OR ¼ 3.1, CI: 1.5–6.5), and the teaching of arts and theater studies (adj OR ¼ 2.8, CI: 1.4–5.9) are risk factors for voice problems in past 1 year and throughout the career, respectively. Conclusion. The findings confirm that teaching is a high-risk profession for acquiring voice problems. The risk factors uncovered are statistically and clinically significant and biologically plausible. There is a need for educational authorities and healthcare providers to develop effective and comprehensive prevention programs to arrest vocal attrition and its detrimental effects on the quality of teaching.
INTRODUCTION Teachers are often cited as a vulnerable population at risk of vocal disturbances. Teachers are believed to have a high prevalence of voice problems because of the disadvantageous acoustic environment and the high vocal demands in teaching.1–4 Compared with other professions, teachers have much higher rates of voice problems, ranging from 11% to 81.0%, compared with nonteachers’ rates of 1.0–36.1%.1,5–13 Results vary because of differences in methodology, definitions, and populations surveyed. Over 20% of teachers missed days of work because of a voice problem. In comparison, between 0 and 4% of the nonteacher group missed work because of their voice.5,6 It is well recognized that teachers are at great risk of vocal disability, and voice problems are a major occupational hazard of teaching. Sick leave, medical treatment, and job attrition can be costly. In the United States, the economic burden from voice problems is estimated to be approximately 2.5 billion dollars annually.14 Many studies have been performed on teachers2,4–6,9,10,12–17 to assess the prevalence and disease burden of losing their voices. To date no published data has emerged from Singapore or South East Asia as far as we are aware of. An indepth knowledge of local high-risk populations and risk factors Accepted for publication May 10, 2011. The results of this study were presented at the World Voice Congress (oral presentation), held 6–9 September, 2010, in Seoul, Korea. From the Department of Otolaryngology, Head and Neck Surgery, Khoo Teck Puat Hospital, Yishun Central, Singapore, Singapore. Address correspondence and reprint requests to Paul Mok Kan Hwei, Department of Otolaryngology, Head and Neck Surgery, Khoo Teck Puat Hospital, 90 Yishun Central, Singapore 768828, Singapore. E-mail: [email protected] Journal of Voice, Vol. 26, No. 4, pp. e141-e147 0892-1997/$36.00 Ó 2012 The Voice Foundation doi:10.1016/j.jvoice.2011.05.004
contributing to voice loss would enable the authorities to design better primary and secondary prevention community programs to minimize vocal attrition. Risk factors previously reported include the female gender,6,10,15,16,18 being older or with more years of teaching,4,10,15 teaching performing arts subjects,19 and poor classroom environment (excessive noise, chalk dust, and poor ventilation).20,21 Health factors include respiratory allergies,10 recent upper airway problems,21 stress/mental disorders,4,21 and smoking and alcohol usage.4,22 Although the above may sound intuitive, other studies disprove the associations. Old age,5,12 upper respiratory tract illnesses,23 and tobacco and alcohol were not risk factors.21,23 There is also a lack of clarity with certain health factors—a heterogeneous group of medical conditions (allergic rhinitis, tonsillitis, colds, and sinus infections) was identified as a single risk factor of recent upper airway problems.21 Previous risk factors that were reported in a study of 100 teacher-patients include generic factors of ‘‘allergies’’ and ‘‘hormonal problems.’’4 A need exists to seek further resolution on these risk factors. The aims of this study are (1) to determine the prevalence of self-reported voice problems among primary level school teachers in Singapore during their career, in the past 1 year and at present; and (2) to explore their associated risk factors. METHODS Study design and study procedures The study design is a cross-sectional study. This study was funded by the National Medical Research Council’s Enabling Grant. Ethical approval was obtained from National Healthcare
e142 Group’s Domain Specific Research Board. Permission to conduct the study in primary schools was obtained from the Ministry of Education. Six primary schools in the proximity of our hospital (Alexandra area which is in south Singapore) were selected based on the consideration that they are representative of typical primary schools in Singapore (coeducational composition public schools). Briefings were conducted before distributing the self-administered questionnaire. Informed consent was obtained from all participants. The study was conducted over a 9-month period. As Singapore is a tropical country with no seasons, the seasonal effects on the voice were not a concern. Description of questionnaire A standardized questionnaire was used for all schools. The questionnaire administered to the teachers includes: 1. Demographical data—age, gender, education level, and race. 2. Teaching characteristics—number of years teaching, medium of instruction (English, Mandarin, Malay, Tamil, or others), subjects (Sciences, Math, Humanities & Social Studies, Arts and Theater studies, English, Chinese, other languages, and Physical education), number of hours of teaching (per week), and number of hours of continuous teaching. 3. Social habits—singing (in various capacities of religious groups, choirs, bands, or social gatherings), smoking (this includes quantifying how many cigarette sticks per day, and if the subject was an ex-smoker and for how long), and alcohol drinking (this includes quantifying drinking frequency as weekly, monthly, or occasionally) 4. Symptoms of medical conditions of allergic rhinitis, reflux disease, and hypothyroidism and their current medications. For allergic rhinitis, questions as used by our ENT clinic for diagnosing allergic rhinitis were presented: ‘‘Are you prone to having runny nose, blocked nose, sneezing almost daily, or previously diagnosed with nose allergy by a doctor?’’ For reflux disease, the question was posed: ‘‘Do you have morning hoarseness, bad breath, or heart burn? The following question was asked to assess for hypothyroid-like symptoms: ‘‘Do you have problems with weight gain, fatigue, cold sensitivity, or menstrual irregularity (if applicable)?’’ Questions on the assessment of hypothyroidism and reflux disease are adapted from a previous review paper.11 Much caution has to be taken to correlate these symptoms with a definite disease process; these factors are thence reported as medical condition’s symptoms and not the medical disease per se. In particular, symptoms of hypothyroidism are myriad, nonspecific and ill-defined and hypothyroidism is never meant to be diagnosed from symptoms alone, hence the above questions are at best to elicit symptoms associated with a hypothyroid-like state. We have thus reported this variable as hypothyroid-like symptoms. Questions to assess point, year, and career prevalence were included in the questionnaire. The questions were designed to
Journal of Voice, Vol. 26, No. 4, 2012
enable teachers to decide for themselves if they have a voice problem. A ‘‘yes’’ answer to the question ‘‘Do you have a problem with your voice today, which is preventing you from doing all that you want to do with it?’’ is a positive case for point prevalence of voice disorders. In response to the question ‘‘During the past 12 months, how often have you had a voice problem, which prevented you from doing all that you wanted to do with it?,’’ an answer of ‘‘every 2–3 months or more frequently’’ is taken as positive for year prevalence. In response to the question ‘‘Throughout your entire teaching career, how often have you had a problem with your voice, which prevented you from doing all you wanted to do with it?,’’ an answer of ‘‘every 6 months or more’’ is considered to be positive for a career prevalence of voice disorders. STATISTICAL METHODS Statistical analysis was performed using the Statistical Package for Social Sciences Version 17.0 (SPSS Inc, Chicago, IL, USA). The respective prevalence rates were calculated and presented with corresponding 95% confidence intervals. The prevalence rates of voice disorder were associated with study variables using contingency tables. Univariate analysis of association was evaluated using Pearson chi-square (c2) tests. Multivariate analysis was next applied using multiple logistic regression, which enable us to identify the strongest combined predictors of a dependent variable. Adjusted odds ratios (adj ORs) with 95% confidence intervals (CIs) were generated. Analysis was repeated using the full saturation model and results are consistent with logistic regressions. Significance was set at the P value of 0.05. RESULTS A total of 214 teachers from six primary schools were recruited for the study. All the teachers who attended the information sessions agreed to undergo the questionnaire. Approximately 20% of the teachers did not turn up at each session because of ongoing teaching or administrative duties or absenteeism from work. Table 1 shows the demographics and teaching details of the participants. The majority are females (82.7%) of Chinese ethnicity (74.8%) who teach in English (75.6%) and are aged 31–40 years (40.7%). As mentioned under the methodology, we have used a convenient sampling of six primary schools around our hospital in south Singapore. We have taken into consideration during the selection that the surveyed schools need to be stereotypical primary schools in Singapore. As we could not perform statistical weighting because of the confidentiality of specific data of all public school teachers in Singapore, we are unable to determine if our population is representative of all teachers in Singapore. However, our study’s population data is very similar to that of the available published data. In the Yearbook of National Statistics 2008,24 17.4% of teachers in primary schools were males and 82.6% females. In our study, 17.3% and 82.7% were males and females, respectively. In the Educational Statisitcs Digest 2007,25 29.5% of primary school
Tze Choong Charn and Paul Kan Hwei Mok
Voice Problems Amongst Primary School Teachers in Singapore
TABLE 1. Characteristics of Study Population Characteristic
n (%)
Gender Male Female
37 (17.3) 177 (82.7)
Race Chinese Malay Indian Others
160 (74.8) 32 (15.0) 17 (7.9) 5 (2.3)
Age group (years) 20–30 31–40 41–50 >50 Missing
72 (33.6) 87 (40.7) 30 (14.0) 24 (11.2) 1 (0.5)
Years of teaching <1 1–4 5–9 10–19 20–29 >30 Missing
30 (14.0) 55 (25.7) 58 (27.1) 36 (16.8) 17 (7.9) 15 (7.0) 3 (1.4)
Medium of instruction English Mandarin Malay Tamil
161 (75.2) 40 (18.7) 9 (4.2) 4 (1.9)
teachers are aged 29 years and younger, 42.3% are between 30 and 39 years, 15.4% are between 40 and 49 years, and 12.9% are aged 50 years and older. In our study population, 34% are aged between 20 and 30 years, 41% are between 31 and 40 years, 14.2% are between 41 and 50 years, and 10.8% are 50 years or older. The point prevalence of voice disorder in the six primary schools was 13.1% (CI: 9.3–18.3%). The past year prevalence and career prevalence in the surveyed schools were 25.4% (CI: 20–31.6%) and 32.1% (CI: 26.2–38.6%), respectively. Univariate analyses of associations between study variables and the three different prevalence rates were carried out. The results are shown in Table 2. To assess the associations between reporting a voice disorder and the study variables, adj ORs were calculated to quantify the odds of a teacher with a voice disorder, having the studied variable against a teacher without a voice problem. Based on this model, a teacher with current voice problems is likely to be using a microphone and experiencing reflux symptoms. Voice problems over the past year are associated with microphone use and having allergic rhinitis and hypothyroid-like symptoms. Allergic rhinitis symptoms, reflux symptoms, and the teaching of arts and theater studies are associated with a teacher experiencing recurring voice problems. These are summarized in Table 3.
e143
Questions were also asked to assess the teachers’ knowledge and attitudes toward voice problems. Almost 90% of teachers agree that voice loss is an occupational hazard. Not surprisingly, 92.8% of teachers answered ‘‘yes’’ to the question ‘‘Do you feel as if your career depended on the health of your voice.’’ This implies that teachers perceive having a healthy voice as essential for optimal career performance. DISCUSSION Assessment of voice problems Voice disorders are generally assessed via two methods—selfreported voice problems by test subjects and the instrumental assessment of voice disease (eg, videostroboscopy). Both have their own inherent deficits. Self-reporting introduces the elements of subjectivity and is prone to over-reporting by patients. The use of clinical tools to objectively assess voice disorders is not standardized and various tools such as acoustic parameter measurements, voice recordings with analyses, and laryngostroboscopy have been used. Evidence from clinical practice suggests that a significant number of people who present for management of voice problems do not have vocal fold pathology. For example, a previous study showed that no vocal pathology was evident in one-third of patients who reported dysphonia.26 True prevalence will hence be underestimated if it is measured solely on the presence of vocal pathology.15 The use of a self-reported questionnaire provides a practical and realistic estimate of the disease prevalence in the teaching community. Furthermore, it is more practical logistically to administer a self-reported questionnaire to screen for voice disorders for a significant number of teachers or across several schools. We had conducted a separate arm of this study to perform videolaryngostroboscopy on teachers who self report voice problems—to investigate the strength of correlation between symptoms and signs. It was surprising and disappointing for us to find that only 55% of teachers consented to videolaryngostroboscopy. This emphasizes the fact that many teachers in our country are apprehensive about undergoing laryngoscopy and is therefore not an appropriate tool for mass screening. It is well known that the criteria to determine if the teacher has a voice disorder varies from study to study and suffers from a lack of standardized international definition.1 In-line with a frequently cited and similar community study, we have used the same definitions for career, past year, and point prevalence for our study.15 Comparison of results In our study, the point prevalence of voice disorders is 13.1%. Past year prevalence is 25.4% and career prevalence is 32.1%. The criterion to define a voice problem is not standardized across studies, thereby making it difficult to compare outcome measures between studies. Taking into account the limited role of evaluating results across different epidemiological surveys, the point prevalence uncovered in our study is broadly comparable with other similar studies, which assessed self-reported voice
e144
Journal of Voice, Vol. 26, No. 4, 2012
TABLE 2. Odds Ratios of Variables by Univariate Analysis
Variable* Female gender No of hours teaching <20 21–30 >30 Microphone use Reflux symptoms Hypothyroid-like symptomsy
Current Voice Problems (n ¼ 28), Number (%)
No Current Voice Problems (n ¼ 185, Missing ¼ 1), Number (%)
ORs
95% CIs
P Value
27 (96.4)
149 (80.5)
6.52
0.858, 49.6
0.039 0.071
8 (28.6) 17 (60.7) 3 (10.7) 7 (25.0) 15 (53.6) 14 (50.0)
57 (30.8) 74 (40.0) 52 (28.1) 11 (5.9) 30 (16.2) 47 (25.4)
2.43 3.98 1.00 5.24 5.89 3.05
0.613, 9.66 1.11, 14.3 Referent 1.83, 15.0 2.54, 13.6 1.34, 6.95
6.38 1.09 1.00 2.57 2.47 3.13 2.95
1.12, 36.2 0.526, 2.26 Referent 0.959, 6.91 1.28, 4.76 1.56, 6.29 1.53, 5.69
Voice Problems Over 1 Year (n ¼ 54), Number (%) Alcohol consumption Weekly Occasional No Microphone use Allergic rhinitis symptoms Reflux symptoms Hypothyroid-like symptomsy
Hours of continuous teaching before a rest period <0.5 0.5–1 1.1–2 2.1–3 3.1–4 4.1–5 >5 Teaching arts and drama Allergic rhinitis symptoms Hypothyroid-like symptomsy Reflux symptoms
0.001 0.000 0.006
No Voice Problems Over 1 Year (n ¼ 159, Missing ¼ 1), Number (%) 0.062
4 (7.4) 13 (24.1) 37 (68.5) 8 (14.8) 23 (42.6) 20 (37.0) 25 (75.9)
2 (1.3) 38 (23.9) 118 (74.2) 10 (6.3) 36 (22.6) 25 (15.7) 36 (22.6)
Voice Problems Throughout Career (n ¼ 68), Number (%)
No Voice Problems Throughout Career (n ¼ 144, Missing ¼ 2), Number (%)
0.053 0.006 0.001 0.001
0.024 0 2 (2.9) 7 (10.3) 40 (58.8) 13 (19.1) 3 (4.4) 3 (4.4) 22 (32.4) 29 (42.6) 30 (44.1) 24 (78.4)
0 1 (0.7) 43 (29.9) 67 (46.5) 17 (11.8) 10 (6.9) 2 (1.4) 23 (16.0) 30 (20.8) 31 (21.5) 21 (1.4)
NA 1.33 0.11 0.398 0.510 0.200 1.00 2.52 2.75 2.85 3.17
0.067, 26.6 0.015, 0.770 0.064, 2.49 0.074, 3.51 0.022, 1.82 Referent 1.28, 4.95 1.47, 5.15 1.53, 5.33 1.61, 6.25
0.006 0.001 0.001 0.001
* Insignificant variables were omitted from the above: age, education level, race, number of years teaching, medium of instruction, other subjects taught, previous voice training, teaching environment, current smoking, ex-smoker, singing, and exposure to smoking. y Hypothyroid-like symptoms—weight gain, fatigue, cold sensitivity, or menstrual irregularity.
problems. These studies reported point prevalence of 11% (n ¼ 1243 in United States),10 15.9% (n ¼ 1168 in Australia),15 14.6% (n ¼ 242 in United States),5 and 15% (for probable dysphonia; n ¼ 2103 in Brazil).21 Russell et al15 reported the past year and career prevalence to be 20% and 19%, respectively. In the study of 1878 Dutch teachers by de Jong et al,2 the past year and career prevalence are 31.6% and 47.5%, respectively. Smith et al7 reported that 32% of teachers (n ¼ 554 in United States) replied positively that they feel they ever had a voice problem before and of these,
93% report that the condition was intermittent but recurrent (approximated to a lifetime prevalence of around 30%). In two studies that were conducted recently, Angelillo et al27 reported the prevalence rate of junior high school teachers as 14.3% (n ¼ 118 in Italy) and the lifetime prevalence of teachers of all levels (n ¼ 504 in Italy) as 51.4%, Van Houtee et al13 reported a similar rate of 51.2% in 994 Belgian teachers. Unlike our study, both studies defined lifetime prevalence of voice problems as teachers who have had any episode of dysphonia, and not recurrent dysphonia. Owing to the lack of universal
Tze Choong Charn and Paul Kan Hwei Mok
Voice Problems Amongst Primary School Teachers in Singapore
TABLE 3. Odds Ratios of Significant Variables by Multivariate Analyses Variable Current voice problems Microphone use Yes No Reflux symptoms Yes No Voice problems over 1 year Microphone use Yes No Hypothyroid-like symptomsy Yes No Allergic rhinitis symptoms Yes No Voice problems throughout career Allergic rhinitis symptoms Yes No Reflux symptoms Yes No Teaching arts and theater studies Yes No
P Value
ORs (Adjusted)*
95% CIs
5.61 1.00
1.79, 17.61 Referent
0.030
6.09 1.00
2.54, 14.62 Referent
0.000
3.35 1.00
1.13, 9.98 Referent
0.030
2.55 1.00
1.29, 5.06 Referent
0.007
2.13 1.00
1.06, 4.31 Referent
0.035
2.63 1.00
1.35, 5.11 Referent
0.005
3.13 1.00
1.52, 6.47 Referent
0.002
2.84 1.00
1.37, 5.88 Referent
0.005
* ORs are adjusted for the significant associated variables from Table 2. y Hypothyroid-like symptoms—weight gain, fatigue, cold sensitivity, or menstrual irregularity.
consensus, results vary significantly because of differences in methodology and definitions.1,15,21 It is also worthy to note that a significant proportion of current reported voice problems would be related to acute problems such as acute upper respiratory tract infections.15 As our study is conducted on site during the teachers’ meetings, the true point prevalence might be underestimated because of the absence of teachers who would be on medical leave because of acute infections. Risk factors As this is a cross-sectional study where the associated factors and the verification of having a significant voice disorder are
e145
assessed simultaneously, we are unable to establish a causal relationship. The risk factors associated with the different voice problem domains have been summarized in Tables 2 and 3. Multivariate analyses identified that microphone use and reflux symptoms are associated with current voice problems. Voice problems over the past year are associated with microphone use and having allergic rhinitis and hypothyroid-like symptoms. Allergic rhinitis and reflux symptoms and the teaching of arts and theater studies are associated with having recurring voice problems. Microphone usage is likely an associated factor, reactionary rather than causal. It is probable that having voice disorders will predispose teachers to using microphones to relieve the strain on their impaired voices. Studies have shown that the usage of vocal amplification measures help in improving voice endurance and decrease the need for repetition,3 and usage will improve the teachers’ self rating of their own voice impairments.28 It is counter-intuitive and less likely that the use of microphones cause one to develop voice problems. Thibeault et al19 reported that teachers of vocal music, drama, and other performing arts are at risk of developing voice disorders. Music and drama teachers are hypothesized to have increased voice disorders as they use their voices over high levels of background noise. Furthermore, the alternating use of speaking and singing voices predisposes them to developing voice problems.1 It is has been reported that ‘‘upper airway conditions’’10,21 predispose toward voice problems. However, this term is nebulous and encompasses a wide spectrum of disorders, such as nasal allergy, asthma, sinusitis, upper respiratory tract infections, and postnasal drip. These conditions often overlap, hence, making it difficult to know precisely the contribution from each of these. Allergic rhinitis is a well-established specific disease that is common and estimated to affect at least 10–25% of the population worldwide. It is screened based on symptoms. In our study, allergic rhinitis is associated with voice problems over the past year and recurring throughout career, but not a risk factor at the point of survey. This is not surprising given the intermittent nature of allergic rhinitis symptoms. Our study confirms that allergic rhinitis as a single condition is a risk factor for voice disorders in teachers. Reflux laryngitis is frequently seen in patients with voice complaints.29 Laryngeal pharyngeal reflux causes vocal fold inflammation and is hence associated with an increased likelihood of developing voice problems. In a study on the general public population of 1326,30 reflux symptoms has been found to be a risk factor for developing voice problems. Calas et al4 reported that gastrointestinal reflux was a risk factor for hoarseness in 100 dysphonic teachers presenting to his clinic. As explained above, we are unable to establish a cause-effect relationship because of the cross-sectional nature of our study. We can only establish reflux symptoms as being associated with voice disorders. It is biologically plausible that poor vocal techniques can cause the inflammation and symptoms of reflux laryngitis, and the two are often interrelated. To our knowledge, there has been only one previous community study on teachers, which has identified reflux as a risk factor for voice problems,31 and our study is in
e146 agreement. We must highlight the diagnostic difficulties associated with laryngopharayngeal reflux (LPR). There presently exists no easily administered, cheap, and reliable test for LPR. Hence, most clinicians depend mainly on clinical symptoms and signs for instituting empirical treatment. Specific investigations such as 24-hour impedance and dual-sensor pH probe testing are reserved for treatment failures.32,33 Studies on voice disorders above have used symptoms based questions to identify patients with reflux symptoms, and we have used a similar approach. We did not use the Reflux Symptom Index (RSI) score, although in retrospect it would have been very useful to incorporate it into our questionnaire. Hypothyroidism has been recognized to cause hoarseness of voice although its exact mechanism is not well understood. Endocrinologic changes have the capacity to alter the fluid content of the laryngeal mucosa, leading to a change in bulk and shape of vocal fold, hence, affecting the quality of voice produced.11 There has been no study to our knowledge, which identified hypothyroidism specifically as a risk factor for voice disorders among teachers. Calas et al4 identified teachers with ‘‘hormonal problems’’ as being at risk. We acknowledge that a history of hypothyroidal symptoms is not specific. It is not possible to make a conclusion that our patients with hypothyroid-like symptoms indeed have hypothyroidism. Because of ethical and financial constraints, we are unable to confirm the diagnoses with biochemical tests. Future studies looking into the relationship between thyroid disorders and voice problems in teachers will be interesting. Some factors that have not been significant in our study include the female gender, age of teachers, years of teaching experience, smoking, and alcohol consumption. It is rather surprising that being female is not an independent risk factor for voice problems. As mentioned above, the female gender has been shown to be a risk factor of voice disorders in multiple studies. Roy et al10 showed that women have a higher lifetime prevalence compared with men (46.3% vs 36.9%). Russell et al15 reported that female teachers are twice as likely compared with males to report voice problems. In our study, the female gender is a risk factor for current voice problems by univariate analysis. However, this relationship is lost with multivariate analysis. The female gender was not shown to be a risk factor by univariate or multivariate analyses for annual and career long voice problems. A possible reason could be because this finding is related to the very small number of men in our population sample. Any voice symptoms reported by any one male teacher tends to over-estimate the actual prevalence of male teachers in our small sample size of 214 teachers. In fact, some studies circumvent this issue by recruiting purely female interviewees.21,34 Significant risk contribution from the age of teachers and the number of years taught is more controversial. Smith et al5 and Russell et al15 found that age was associated with more vocal symptoms and a higher likelihood of reporting a current voice problem, respectively. However, in the same study, Russell et al15 also reported there was no association between the annual prevalence of voice problems and age nor the number of years teaching. Urrutikoetxea et al22 reported a decrease in
Journal of Voice, Vol. 26, No. 4, 2012
the presence of vocal nodules with age. Whereas Sapir et al,12 Thibeault et al,19 and de Jong et al2 reported that age was not correlated to vocal symptoms at all. It is generally recognized that alcohol and smoking are factors associated with voice disorders and has been proven in previous studies.4,22 However, our study did not show a positive correlation. This finding is similar to studies by Gotaas and Starr23 and Roy et al.10 However, it must be noted that our study has a very small proportion of smokers (1.9%) and regular alcohol drinkers (2.8%), thus making it difficult to establish any relationship between the voice disorders and the two variables owing to a lack of statistical power. CONCLUSION The results show that (1) the point prevalence of voice disorders amongst primary school teachers in our study is 13.1%, and prevalence in past 1 year and throughout the career are 25.4% and 32.1%, respectively, whereby a voice disorder is defined as having a problem with one’s voice that prevented one from doing all that is wished to be done with it. (2) Risk factors that cut across the different voice disorder domains include the use of microphones, allergic rhinitis symptoms, reflux symptoms, and hypothyroid-like symptoms. The significant medical risk factors of allergic rhinitis, reflux, and hypothyroid-like symptoms reinforce the information trove of current knowledge. As our study is wholly questionnaire based, the significant risk factors above are symptoms of medical conditions per se. Our study is not meant to determine if the medical conditions are also risk factors without the benefits of confirmatory diagnostic tests and investigations. This study has furnished further insight and clarification with regards to prevalence of voice disorders and associated risk factors amongst primary school teachers in Singapore. We hope that the results will enable health and educational authorities to pursue health education policies that minimize vocal attrition at the teaching workplace and identify those teachers at risk and apply primary and secondary prevention methods to arrest the associated degradation of education. Acknowledgments The authors would like to sincerely express their gratitude to A/P Ng Tze Pin and Dr. Chen Yuming for their meticulous biostatistical support and advice. REFERENCES 1. Mattiske JA, Oates JM, Greenwood KM. Vocal problems among teachers: a review of prevalence, cause, prevention, and treatment. J Voice. 1998;13: 489–499. 2. de Jong FI, Kooijman PG, Thomas G, Huinck WJ, Graamans K, Schutte HK. Epidemiology of voice problems in Dutch teachers. Folia Phoniatr Logop. 2006;58:186–198. 3. Jonsdottir V, Rantala L, Laukkanen A-M, Vilkman E. Effects of sound amplification on teachers’ speech while teaching. Logoped Phoniatr Vocol. 2001;26:118–123. 4. Calas M, Verhust J, Lecoq M, Dalleas B, Seihean M. Vocal pathology of teachers. Rev Laryngol Otol Rhinol (Bord). 1989;110:397–406. [in French]. 5. Smith E, Gray SD, Dove H, Kirchner L, Heras H. Frequency and effects of teachers’ voice problems. J Voice. 1997;11:81–87.
Tze Choong Charn and Paul Kan Hwei Mok
Voice Problems Amongst Primary School Teachers in Singapore
6. Smith E, Kirchner JA, Taylor M, Hoffman H, Lemke J. Voice problems among teachers: differences by gender and teaching characteristics. J Voice. 1998;12:328–334. 7. Smith E, Lemke J, Taylor M, Kirchner L, Hoffman H. Frequency of voice problems among teachers and other occupations. J Voice. 1998;12: 480–488. 8. Sala E, Laine A, Simberg S, Pentti J, Suonpaa J. The prevalence of voice disorders among day care center teachers compared with nurses: a questionnaire and clinical study. J Voice. 2001;15:413–423. 9. Jonsdottir VI, Boyle BE, Martin PJ, Sigurdardottir G. A comparison of the occurrence and nature of voice symptoms in two groups of Icelandic teachers. Logoped Phoniatr Vocol. 2002;27:98–105. 10. Roy N, Merrill RM, Thibeault SL, Parsa RA, Gray SD, Smith ME. Prevalence of voice problems in teachers and the general population. J Speech Lang Hear Res. 2004;47:281–293. 11. Sataloff RT. Professional voice users: the evaluation of voice disorders. Occup Med. 2001;16:633–647. 12. Sapir S, Keidar A, Van Velzen D. Vocal attrition in teachers: survey findings. Eur J Disord Commun. 1993;28:177–185. 13. Van Houtte E, Claey S, Wuyts F, Van Lierde K. The impact of voice disorders among teachers: vocal complaints, treatment-seeking behaviour, knowledge of vocal care and voice-related absenteeism. J Voice. Jul 14, 2010; [Epub ahead of print]. 14. Verdolini K, Ramig OR. Review: occupational risks for voice problems. Logoped Phoniatr Vocol. 2001;26:37–46. 15. Russell A, Oates J, Greenwood KM. Prevalence of voice problems in teachers. J Voice. 1998;12:467–479. 16. Miller MK, Verdolini K. Frequency and risk factors for voice problems in teachers of singing and control subjects. J Voice. 1995;9:348–362. 17. Herrington-Hall BL, Lee L, Stemple JC, Niemi KR, McHone MM. Description of laryngeal pathologies by age, sex and occupation in a treatment-seeking sample. J Speech Hearing Disord. 1998;53:57–64. 18. Pekkarinen E, Himberg L, Pentti J. Prevalence of vocal symptoms among teachers compared with nurses: a questionnaire study. Scand J Logoped Phonatr. 1992;17:113–117. 19. Thibeault SL, Merrill RM, Roy N, Gray SD, Smith EM. Occupational risk factors associated with voce disorders among teachers. Ann Epidemiol. 2004;14:786–792.
e147
20. Simberg S, Sala E, Vehmas K, Laine A. Changes in the prevalence of vocal symptoms among teachers during a twelve-year period. J Voice. 2005;19: 95–102. 21. de Medeiros AM, Barreto SM, Assunc¸~ao AA. Voice disorders (dysphonia) in public school female teachers working in Belo Horizonte: prevalence and associated factors. J Voice. 2008;22:676–687. 22. Urrutikoetxea A, Ispizua A, Matellanes F, Aurrekoetxea J. Prevalence of vocal nodules in teachers. Paper presented at: the First World Congress of Voice. Oporto, Portugal: 1995. 23. Gotaas C, Starr CD. Vocal fatigue among teachers. Folia Phoniatr (Basel). 1993;45:120–129. 24. Chapter 19: Education. In: Yearbook of Statistics Singapore. 2008: 243–261. 25. Chapter 10. In: Education Statistics Digest 2007. Ministry of Education. 2008:14–17. 26. Mathieson L. Vocal tract discomfort in hyperfunctional dysphonia. J Voice. 1993;2:40–48. 27. Angelillo M, Di Maio G, Costa G, Angelillo N, Barillari U. Prevalence of occupational voice disorders in teachers. J Prev Med Hyg. 2009;50:26–32. 28. Roy N, Weinrich B, Gray SD, Tanner K, Toledo SW, Dove H, Corbin-Lewis K, Stemple JC. Voice amplification versus vocal hygiene instruction for teachers with voice disorders: a treatment outcomes study. J Speech Lang Hear Res. 2002;45:625–638. 29. Santaloff RT, Castell DO, Katz PO, Sataloff DM. Reflux Laryngitis and Related Disorders. San Diego, CA: Singular Publishing Group; 1997. 30. Roy N, Merrill RM, Gray SD, Smith EM. Voice disorders in the general population: prevalence, risk factors, and occupational impact. Laryngoscope. 2005;115:1988–1995. 31. Preciado J, Perez C, Calzada M, Preciado P. Frequency and risk factors of voice disorders among teaching staff of La Rioja, Spain. Acta Otorrinolaringol Esp. 2005;56:161–170. [in Spanish]. 32. Bove MJ, Rosen C. Diagnosis and management of laryngopharyngeal reflux disease. Curr Opin Otolaryngol Head Neck Surg. 2006;14:116–123. 33. Ali Mel-S. Laryngopharyngeal reflux: diagnosis and treatment of a controversial disease. Curr Opin Allergy Clin Immunol. 2008;8:28–33. 34. Schneider B, Bigenzahn W. Vocal risk factors for occupational voice disorders in female teaching students. Eur Arch Otorhinolaryngol. 2005; 262:272–276.
INTRODUCTION Teachers are often cited as a vulnerable population at risk of vocal disturbances. Teachers are believed to have a high prevalence of voice problems because of the disadvantageous acoustic environment and the high vocal demands in teaching.1–4 Compared with other professions, teachers have much higher rates of voice problems, ranging from 11% to 81.0%, compared with nonteachers’ rates of 1.0–36.1%.1,5–13 Results vary because of differences in methodology, definitions, and populations surveyed. Over 20% of teachers missed days of work because of a voice problem. In comparison, between 0 and 4% of the nonteacher group missed work because of their voice.5,6 It is well recognized that teachers are at great risk of vocal disability, and voice problems are a major occupational hazard of teaching. Sick leave, medical treatment, and job attrition can be costly. In the United States, the economic burden from voice problems is estimated to be approximately 2.5 billion dollars annually.14 Many studies have been performed on teachers2,4–6,9,10,12–17 to assess the prevalence and disease burden of losing their voices. To date no published data has emerged from Singapore or South East Asia as far as we are aware of. An indepth knowledge of local high-risk populations and risk factors Accepted for publication May 10, 2011. The results of this study were presented at the World Voice Congress (oral presentation), held 6–9 September, 2010, in Seoul, Korea. From the Department of Otolaryngology, Head and Neck Surgery, Khoo Teck Puat Hospital, Yishun Central, Singapore, Singapore. Address correspondence and reprint requests to Paul Mok Kan Hwei, Department of Otolaryngology, Head and Neck Surgery, Khoo Teck Puat Hospital, 90 Yishun Central, Singapore 768828, Singapore. E-mail: [email protected] Journal of Voice, Vol. 26, No. 4, pp. e141-e147 0892-1997/$36.00 Ó 2012 The Voice Foundation doi:10.1016/j.jvoice.2011.05.004
contributing to voice loss would enable the authorities to design better primary and secondary prevention community programs to minimize vocal attrition. Risk factors previously reported include the female gender,6,10,15,16,18 being older or with more years of teaching,4,10,15 teaching performing arts subjects,19 and poor classroom environment (excessive noise, chalk dust, and poor ventilation).20,21 Health factors include respiratory allergies,10 recent upper airway problems,21 stress/mental disorders,4,21 and smoking and alcohol usage.4,22 Although the above may sound intuitive, other studies disprove the associations. Old age,5,12 upper respiratory tract illnesses,23 and tobacco and alcohol were not risk factors.21,23 There is also a lack of clarity with certain health factors—a heterogeneous group of medical conditions (allergic rhinitis, tonsillitis, colds, and sinus infections) was identified as a single risk factor of recent upper airway problems.21 Previous risk factors that were reported in a study of 100 teacher-patients include generic factors of ‘‘allergies’’ and ‘‘hormonal problems.’’4 A need exists to seek further resolution on these risk factors. The aims of this study are (1) to determine the prevalence of self-reported voice problems among primary level school teachers in Singapore during their career, in the past 1 year and at present; and (2) to explore their associated risk factors. METHODS Study design and study procedures The study design is a cross-sectional study. This study was funded by the National Medical Research Council’s Enabling Grant. Ethical approval was obtained from National Healthcare
e142 Group’s Domain Specific Research Board. Permission to conduct the study in primary schools was obtained from the Ministry of Education. Six primary schools in the proximity of our hospital (Alexandra area which is in south Singapore) were selected based on the consideration that they are representative of typical primary schools in Singapore (coeducational composition public schools). Briefings were conducted before distributing the self-administered questionnaire. Informed consent was obtained from all participants. The study was conducted over a 9-month period. As Singapore is a tropical country with no seasons, the seasonal effects on the voice were not a concern. Description of questionnaire A standardized questionnaire was used for all schools. The questionnaire administered to the teachers includes: 1. Demographical data—age, gender, education level, and race. 2. Teaching characteristics—number of years teaching, medium of instruction (English, Mandarin, Malay, Tamil, or others), subjects (Sciences, Math, Humanities & Social Studies, Arts and Theater studies, English, Chinese, other languages, and Physical education), number of hours of teaching (per week), and number of hours of continuous teaching. 3. Social habits—singing (in various capacities of religious groups, choirs, bands, or social gatherings), smoking (this includes quantifying how many cigarette sticks per day, and if the subject was an ex-smoker and for how long), and alcohol drinking (this includes quantifying drinking frequency as weekly, monthly, or occasionally) 4. Symptoms of medical conditions of allergic rhinitis, reflux disease, and hypothyroidism and their current medications. For allergic rhinitis, questions as used by our ENT clinic for diagnosing allergic rhinitis were presented: ‘‘Are you prone to having runny nose, blocked nose, sneezing almost daily, or previously diagnosed with nose allergy by a doctor?’’ For reflux disease, the question was posed: ‘‘Do you have morning hoarseness, bad breath, or heart burn? The following question was asked to assess for hypothyroid-like symptoms: ‘‘Do you have problems with weight gain, fatigue, cold sensitivity, or menstrual irregularity (if applicable)?’’ Questions on the assessment of hypothyroidism and reflux disease are adapted from a previous review paper.11 Much caution has to be taken to correlate these symptoms with a definite disease process; these factors are thence reported as medical condition’s symptoms and not the medical disease per se. In particular, symptoms of hypothyroidism are myriad, nonspecific and ill-defined and hypothyroidism is never meant to be diagnosed from symptoms alone, hence the above questions are at best to elicit symptoms associated with a hypothyroid-like state. We have thus reported this variable as hypothyroid-like symptoms. Questions to assess point, year, and career prevalence were included in the questionnaire. The questions were designed to
Journal of Voice, Vol. 26, No. 4, 2012
enable teachers to decide for themselves if they have a voice problem. A ‘‘yes’’ answer to the question ‘‘Do you have a problem with your voice today, which is preventing you from doing all that you want to do with it?’’ is a positive case for point prevalence of voice disorders. In response to the question ‘‘During the past 12 months, how often have you had a voice problem, which prevented you from doing all that you wanted to do with it?,’’ an answer of ‘‘every 2–3 months or more frequently’’ is taken as positive for year prevalence. In response to the question ‘‘Throughout your entire teaching career, how often have you had a problem with your voice, which prevented you from doing all you wanted to do with it?,’’ an answer of ‘‘every 6 months or more’’ is considered to be positive for a career prevalence of voice disorders. STATISTICAL METHODS Statistical analysis was performed using the Statistical Package for Social Sciences Version 17.0 (SPSS Inc, Chicago, IL, USA). The respective prevalence rates were calculated and presented with corresponding 95% confidence intervals. The prevalence rates of voice disorder were associated with study variables using contingency tables. Univariate analysis of association was evaluated using Pearson chi-square (c2) tests. Multivariate analysis was next applied using multiple logistic regression, which enable us to identify the strongest combined predictors of a dependent variable. Adjusted odds ratios (adj ORs) with 95% confidence intervals (CIs) were generated. Analysis was repeated using the full saturation model and results are consistent with logistic regressions. Significance was set at the P value of 0.05. RESULTS A total of 214 teachers from six primary schools were recruited for the study. All the teachers who attended the information sessions agreed to undergo the questionnaire. Approximately 20% of the teachers did not turn up at each session because of ongoing teaching or administrative duties or absenteeism from work. Table 1 shows the demographics and teaching details of the participants. The majority are females (82.7%) of Chinese ethnicity (74.8%) who teach in English (75.6%) and are aged 31–40 years (40.7%). As mentioned under the methodology, we have used a convenient sampling of six primary schools around our hospital in south Singapore. We have taken into consideration during the selection that the surveyed schools need to be stereotypical primary schools in Singapore. As we could not perform statistical weighting because of the confidentiality of specific data of all public school teachers in Singapore, we are unable to determine if our population is representative of all teachers in Singapore. However, our study’s population data is very similar to that of the available published data. In the Yearbook of National Statistics 2008,24 17.4% of teachers in primary schools were males and 82.6% females. In our study, 17.3% and 82.7% were males and females, respectively. In the Educational Statisitcs Digest 2007,25 29.5% of primary school
Tze Choong Charn and Paul Kan Hwei Mok
Voice Problems Amongst Primary School Teachers in Singapore
TABLE 1. Characteristics of Study Population Characteristic
n (%)
Gender Male Female
37 (17.3) 177 (82.7)
Race Chinese Malay Indian Others
160 (74.8) 32 (15.0) 17 (7.9) 5 (2.3)
Age group (years) 20–30 31–40 41–50 >50 Missing
72 (33.6) 87 (40.7) 30 (14.0) 24 (11.2) 1 (0.5)
Years of teaching <1 1–4 5–9 10–19 20–29 >30 Missing
30 (14.0) 55 (25.7) 58 (27.1) 36 (16.8) 17 (7.9) 15 (7.0) 3 (1.4)
Medium of instruction English Mandarin Malay Tamil
161 (75.2) 40 (18.7) 9 (4.2) 4 (1.9)
teachers are aged 29 years and younger, 42.3% are between 30 and 39 years, 15.4% are between 40 and 49 years, and 12.9% are aged 50 years and older. In our study population, 34% are aged between 20 and 30 years, 41% are between 31 and 40 years, 14.2% are between 41 and 50 years, and 10.8% are 50 years or older. The point prevalence of voice disorder in the six primary schools was 13.1% (CI: 9.3–18.3%). The past year prevalence and career prevalence in the surveyed schools were 25.4% (CI: 20–31.6%) and 32.1% (CI: 26.2–38.6%), respectively. Univariate analyses of associations between study variables and the three different prevalence rates were carried out. The results are shown in Table 2. To assess the associations between reporting a voice disorder and the study variables, adj ORs were calculated to quantify the odds of a teacher with a voice disorder, having the studied variable against a teacher without a voice problem. Based on this model, a teacher with current voice problems is likely to be using a microphone and experiencing reflux symptoms. Voice problems over the past year are associated with microphone use and having allergic rhinitis and hypothyroid-like symptoms. Allergic rhinitis symptoms, reflux symptoms, and the teaching of arts and theater studies are associated with a teacher experiencing recurring voice problems. These are summarized in Table 3.
e143
Questions were also asked to assess the teachers’ knowledge and attitudes toward voice problems. Almost 90% of teachers agree that voice loss is an occupational hazard. Not surprisingly, 92.8% of teachers answered ‘‘yes’’ to the question ‘‘Do you feel as if your career depended on the health of your voice.’’ This implies that teachers perceive having a healthy voice as essential for optimal career performance. DISCUSSION Assessment of voice problems Voice disorders are generally assessed via two methods—selfreported voice problems by test subjects and the instrumental assessment of voice disease (eg, videostroboscopy). Both have their own inherent deficits. Self-reporting introduces the elements of subjectivity and is prone to over-reporting by patients. The use of clinical tools to objectively assess voice disorders is not standardized and various tools such as acoustic parameter measurements, voice recordings with analyses, and laryngostroboscopy have been used. Evidence from clinical practice suggests that a significant number of people who present for management of voice problems do not have vocal fold pathology. For example, a previous study showed that no vocal pathology was evident in one-third of patients who reported dysphonia.26 True prevalence will hence be underestimated if it is measured solely on the presence of vocal pathology.15 The use of a self-reported questionnaire provides a practical and realistic estimate of the disease prevalence in the teaching community. Furthermore, it is more practical logistically to administer a self-reported questionnaire to screen for voice disorders for a significant number of teachers or across several schools. We had conducted a separate arm of this study to perform videolaryngostroboscopy on teachers who self report voice problems—to investigate the strength of correlation between symptoms and signs. It was surprising and disappointing for us to find that only 55% of teachers consented to videolaryngostroboscopy. This emphasizes the fact that many teachers in our country are apprehensive about undergoing laryngoscopy and is therefore not an appropriate tool for mass screening. It is well known that the criteria to determine if the teacher has a voice disorder varies from study to study and suffers from a lack of standardized international definition.1 In-line with a frequently cited and similar community study, we have used the same definitions for career, past year, and point prevalence for our study.15 Comparison of results In our study, the point prevalence of voice disorders is 13.1%. Past year prevalence is 25.4% and career prevalence is 32.1%. The criterion to define a voice problem is not standardized across studies, thereby making it difficult to compare outcome measures between studies. Taking into account the limited role of evaluating results across different epidemiological surveys, the point prevalence uncovered in our study is broadly comparable with other similar studies, which assessed self-reported voice
e144
Journal of Voice, Vol. 26, No. 4, 2012
TABLE 2. Odds Ratios of Variables by Univariate Analysis
Variable* Female gender No of hours teaching <20 21–30 >30 Microphone use Reflux symptoms Hypothyroid-like symptomsy
Current Voice Problems (n ¼ 28), Number (%)
No Current Voice Problems (n ¼ 185, Missing ¼ 1), Number (%)
ORs
95% CIs
P Value
27 (96.4)
149 (80.5)
6.52
0.858, 49.6
0.039 0.071
8 (28.6) 17 (60.7) 3 (10.7) 7 (25.0) 15 (53.6) 14 (50.0)
57 (30.8) 74 (40.0) 52 (28.1) 11 (5.9) 30 (16.2) 47 (25.4)
2.43 3.98 1.00 5.24 5.89 3.05
0.613, 9.66 1.11, 14.3 Referent 1.83, 15.0 2.54, 13.6 1.34, 6.95
6.38 1.09 1.00 2.57 2.47 3.13 2.95
1.12, 36.2 0.526, 2.26 Referent 0.959, 6.91 1.28, 4.76 1.56, 6.29 1.53, 5.69
Voice Problems Over 1 Year (n ¼ 54), Number (%) Alcohol consumption Weekly Occasional No Microphone use Allergic rhinitis symptoms Reflux symptoms Hypothyroid-like symptomsy
Hours of continuous teaching before a rest period <0.5 0.5–1 1.1–2 2.1–3 3.1–4 4.1–5 >5 Teaching arts and drama Allergic rhinitis symptoms Hypothyroid-like symptomsy Reflux symptoms
0.001 0.000 0.006
No Voice Problems Over 1 Year (n ¼ 159, Missing ¼ 1), Number (%) 0.062
4 (7.4) 13 (24.1) 37 (68.5) 8 (14.8) 23 (42.6) 20 (37.0) 25 (75.9)
2 (1.3) 38 (23.9) 118 (74.2) 10 (6.3) 36 (22.6) 25 (15.7) 36 (22.6)
Voice Problems Throughout Career (n ¼ 68), Number (%)
No Voice Problems Throughout Career (n ¼ 144, Missing ¼ 2), Number (%)
0.053 0.006 0.001 0.001
0.024 0 2 (2.9) 7 (10.3) 40 (58.8) 13 (19.1) 3 (4.4) 3 (4.4) 22 (32.4) 29 (42.6) 30 (44.1) 24 (78.4)
0 1 (0.7) 43 (29.9) 67 (46.5) 17 (11.8) 10 (6.9) 2 (1.4) 23 (16.0) 30 (20.8) 31 (21.5) 21 (1.4)
NA 1.33 0.11 0.398 0.510 0.200 1.00 2.52 2.75 2.85 3.17
0.067, 26.6 0.015, 0.770 0.064, 2.49 0.074, 3.51 0.022, 1.82 Referent 1.28, 4.95 1.47, 5.15 1.53, 5.33 1.61, 6.25
0.006 0.001 0.001 0.001
* Insignificant variables were omitted from the above: age, education level, race, number of years teaching, medium of instruction, other subjects taught, previous voice training, teaching environment, current smoking, ex-smoker, singing, and exposure to smoking. y Hypothyroid-like symptoms—weight gain, fatigue, cold sensitivity, or menstrual irregularity.
problems. These studies reported point prevalence of 11% (n ¼ 1243 in United States),10 15.9% (n ¼ 1168 in Australia),15 14.6% (n ¼ 242 in United States),5 and 15% (for probable dysphonia; n ¼ 2103 in Brazil).21 Russell et al15 reported the past year and career prevalence to be 20% and 19%, respectively. In the study of 1878 Dutch teachers by de Jong et al,2 the past year and career prevalence are 31.6% and 47.5%, respectively. Smith et al7 reported that 32% of teachers (n ¼ 554 in United States) replied positively that they feel they ever had a voice problem before and of these,
93% report that the condition was intermittent but recurrent (approximated to a lifetime prevalence of around 30%). In two studies that were conducted recently, Angelillo et al27 reported the prevalence rate of junior high school teachers as 14.3% (n ¼ 118 in Italy) and the lifetime prevalence of teachers of all levels (n ¼ 504 in Italy) as 51.4%, Van Houtee et al13 reported a similar rate of 51.2% in 994 Belgian teachers. Unlike our study, both studies defined lifetime prevalence of voice problems as teachers who have had any episode of dysphonia, and not recurrent dysphonia. Owing to the lack of universal
Tze Choong Charn and Paul Kan Hwei Mok
Voice Problems Amongst Primary School Teachers in Singapore
TABLE 3. Odds Ratios of Significant Variables by Multivariate Analyses Variable Current voice problems Microphone use Yes No Reflux symptoms Yes No Voice problems over 1 year Microphone use Yes No Hypothyroid-like symptomsy Yes No Allergic rhinitis symptoms Yes No Voice problems throughout career Allergic rhinitis symptoms Yes No Reflux symptoms Yes No Teaching arts and theater studies Yes No
P Value
ORs (Adjusted)*
95% CIs
5.61 1.00
1.79, 17.61 Referent
0.030
6.09 1.00
2.54, 14.62 Referent
0.000
3.35 1.00
1.13, 9.98 Referent
0.030
2.55 1.00
1.29, 5.06 Referent
0.007
2.13 1.00
1.06, 4.31 Referent
0.035
2.63 1.00
1.35, 5.11 Referent
0.005
3.13 1.00
1.52, 6.47 Referent
0.002
2.84 1.00
1.37, 5.88 Referent
0.005
* ORs are adjusted for the significant associated variables from Table 2. y Hypothyroid-like symptoms—weight gain, fatigue, cold sensitivity, or menstrual irregularity.
consensus, results vary significantly because of differences in methodology and definitions.1,15,21 It is also worthy to note that a significant proportion of current reported voice problems would be related to acute problems such as acute upper respiratory tract infections.15 As our study is conducted on site during the teachers’ meetings, the true point prevalence might be underestimated because of the absence of teachers who would be on medical leave because of acute infections. Risk factors As this is a cross-sectional study where the associated factors and the verification of having a significant voice disorder are
e145
assessed simultaneously, we are unable to establish a causal relationship. The risk factors associated with the different voice problem domains have been summarized in Tables 2 and 3. Multivariate analyses identified that microphone use and reflux symptoms are associated with current voice problems. Voice problems over the past year are associated with microphone use and having allergic rhinitis and hypothyroid-like symptoms. Allergic rhinitis and reflux symptoms and the teaching of arts and theater studies are associated with having recurring voice problems. Microphone usage is likely an associated factor, reactionary rather than causal. It is probable that having voice disorders will predispose teachers to using microphones to relieve the strain on their impaired voices. Studies have shown that the usage of vocal amplification measures help in improving voice endurance and decrease the need for repetition,3 and usage will improve the teachers’ self rating of their own voice impairments.28 It is counter-intuitive and less likely that the use of microphones cause one to develop voice problems. Thibeault et al19 reported that teachers of vocal music, drama, and other performing arts are at risk of developing voice disorders. Music and drama teachers are hypothesized to have increased voice disorders as they use their voices over high levels of background noise. Furthermore, the alternating use of speaking and singing voices predisposes them to developing voice problems.1 It is has been reported that ‘‘upper airway conditions’’10,21 predispose toward voice problems. However, this term is nebulous and encompasses a wide spectrum of disorders, such as nasal allergy, asthma, sinusitis, upper respiratory tract infections, and postnasal drip. These conditions often overlap, hence, making it difficult to know precisely the contribution from each of these. Allergic rhinitis is a well-established specific disease that is common and estimated to affect at least 10–25% of the population worldwide. It is screened based on symptoms. In our study, allergic rhinitis is associated with voice problems over the past year and recurring throughout career, but not a risk factor at the point of survey. This is not surprising given the intermittent nature of allergic rhinitis symptoms. Our study confirms that allergic rhinitis as a single condition is a risk factor for voice disorders in teachers. Reflux laryngitis is frequently seen in patients with voice complaints.29 Laryngeal pharyngeal reflux causes vocal fold inflammation and is hence associated with an increased likelihood of developing voice problems. In a study on the general public population of 1326,30 reflux symptoms has been found to be a risk factor for developing voice problems. Calas et al4 reported that gastrointestinal reflux was a risk factor for hoarseness in 100 dysphonic teachers presenting to his clinic. As explained above, we are unable to establish a cause-effect relationship because of the cross-sectional nature of our study. We can only establish reflux symptoms as being associated with voice disorders. It is biologically plausible that poor vocal techniques can cause the inflammation and symptoms of reflux laryngitis, and the two are often interrelated. To our knowledge, there has been only one previous community study on teachers, which has identified reflux as a risk factor for voice problems,31 and our study is in
e146 agreement. We must highlight the diagnostic difficulties associated with laryngopharayngeal reflux (LPR). There presently exists no easily administered, cheap, and reliable test for LPR. Hence, most clinicians depend mainly on clinical symptoms and signs for instituting empirical treatment. Specific investigations such as 24-hour impedance and dual-sensor pH probe testing are reserved for treatment failures.32,33 Studies on voice disorders above have used symptoms based questions to identify patients with reflux symptoms, and we have used a similar approach. We did not use the Reflux Symptom Index (RSI) score, although in retrospect it would have been very useful to incorporate it into our questionnaire. Hypothyroidism has been recognized to cause hoarseness of voice although its exact mechanism is not well understood. Endocrinologic changes have the capacity to alter the fluid content of the laryngeal mucosa, leading to a change in bulk and shape of vocal fold, hence, affecting the quality of voice produced.11 There has been no study to our knowledge, which identified hypothyroidism specifically as a risk factor for voice disorders among teachers. Calas et al4 identified teachers with ‘‘hormonal problems’’ as being at risk. We acknowledge that a history of hypothyroidal symptoms is not specific. It is not possible to make a conclusion that our patients with hypothyroid-like symptoms indeed have hypothyroidism. Because of ethical and financial constraints, we are unable to confirm the diagnoses with biochemical tests. Future studies looking into the relationship between thyroid disorders and voice problems in teachers will be interesting. Some factors that have not been significant in our study include the female gender, age of teachers, years of teaching experience, smoking, and alcohol consumption. It is rather surprising that being female is not an independent risk factor for voice problems. As mentioned above, the female gender has been shown to be a risk factor of voice disorders in multiple studies. Roy et al10 showed that women have a higher lifetime prevalence compared with men (46.3% vs 36.9%). Russell et al15 reported that female teachers are twice as likely compared with males to report voice problems. In our study, the female gender is a risk factor for current voice problems by univariate analysis. However, this relationship is lost with multivariate analysis. The female gender was not shown to be a risk factor by univariate or multivariate analyses for annual and career long voice problems. A possible reason could be because this finding is related to the very small number of men in our population sample. Any voice symptoms reported by any one male teacher tends to over-estimate the actual prevalence of male teachers in our small sample size of 214 teachers. In fact, some studies circumvent this issue by recruiting purely female interviewees.21,34 Significant risk contribution from the age of teachers and the number of years taught is more controversial. Smith et al5 and Russell et al15 found that age was associated with more vocal symptoms and a higher likelihood of reporting a current voice problem, respectively. However, in the same study, Russell et al15 also reported there was no association between the annual prevalence of voice problems and age nor the number of years teaching. Urrutikoetxea et al22 reported a decrease in
Journal of Voice, Vol. 26, No. 4, 2012
the presence of vocal nodules with age. Whereas Sapir et al,12 Thibeault et al,19 and de Jong et al2 reported that age was not correlated to vocal symptoms at all. It is generally recognized that alcohol and smoking are factors associated with voice disorders and has been proven in previous studies.4,22 However, our study did not show a positive correlation. This finding is similar to studies by Gotaas and Starr23 and Roy et al.10 However, it must be noted that our study has a very small proportion of smokers (1.9%) and regular alcohol drinkers (2.8%), thus making it difficult to establish any relationship between the voice disorders and the two variables owing to a lack of statistical power. CONCLUSION The results show that (1) the point prevalence of voice disorders amongst primary school teachers in our study is 13.1%, and prevalence in past 1 year and throughout the career are 25.4% and 32.1%, respectively, whereby a voice disorder is defined as having a problem with one’s voice that prevented one from doing all that is wished to be done with it. (2) Risk factors that cut across the different voice disorder domains include the use of microphones, allergic rhinitis symptoms, reflux symptoms, and hypothyroid-like symptoms. The significant medical risk factors of allergic rhinitis, reflux, and hypothyroid-like symptoms reinforce the information trove of current knowledge. As our study is wholly questionnaire based, the significant risk factors above are symptoms of medical conditions per se. Our study is not meant to determine if the medical conditions are also risk factors without the benefits of confirmatory diagnostic tests and investigations. This study has furnished further insight and clarification with regards to prevalence of voice disorders and associated risk factors amongst primary school teachers in Singapore. We hope that the results will enable health and educational authorities to pursue health education policies that minimize vocal attrition at the teaching workplace and identify those teachers at risk and apply primary and secondary prevention methods to arrest the associated degradation of education. Acknowledgments The authors would like to sincerely express their gratitude to A/P Ng Tze Pin and Dr. Chen Yuming for their meticulous biostatistical support and advice. REFERENCES 1. Mattiske JA, Oates JM, Greenwood KM. Vocal problems among teachers: a review of prevalence, cause, prevention, and treatment. J Voice. 1998;13: 489–499. 2. de Jong FI, Kooijman PG, Thomas G, Huinck WJ, Graamans K, Schutte HK. Epidemiology of voice problems in Dutch teachers. Folia Phoniatr Logop. 2006;58:186–198. 3. Jonsdottir V, Rantala L, Laukkanen A-M, Vilkman E. Effects of sound amplification on teachers’ speech while teaching. Logoped Phoniatr Vocol. 2001;26:118–123. 4. Calas M, Verhust J, Lecoq M, Dalleas B, Seihean M. Vocal pathology of teachers. Rev Laryngol Otol Rhinol (Bord). 1989;110:397–406. [in French]. 5. Smith E, Gray SD, Dove H, Kirchner L, Heras H. Frequency and effects of teachers’ voice problems. J Voice. 1997;11:81–87.
Tze Choong Charn and Paul Kan Hwei Mok
Voice Problems Amongst Primary School Teachers in Singapore
6. Smith E, Kirchner JA, Taylor M, Hoffman H, Lemke J. Voice problems among teachers: differences by gender and teaching characteristics. J Voice. 1998;12:328–334. 7. Smith E, Lemke J, Taylor M, Kirchner L, Hoffman H. Frequency of voice problems among teachers and other occupations. J Voice. 1998;12: 480–488. 8. Sala E, Laine A, Simberg S, Pentti J, Suonpaa J. The prevalence of voice disorders among day care center teachers compared with nurses: a questionnaire and clinical study. J Voice. 2001;15:413–423. 9. Jonsdottir VI, Boyle BE, Martin PJ, Sigurdardottir G. A comparison of the occurrence and nature of voice symptoms in two groups of Icelandic teachers. Logoped Phoniatr Vocol. 2002;27:98–105. 10. Roy N, Merrill RM, Thibeault SL, Parsa RA, Gray SD, Smith ME. Prevalence of voice problems in teachers and the general population. J Speech Lang Hear Res. 2004;47:281–293. 11. Sataloff RT. Professional voice users: the evaluation of voice disorders. Occup Med. 2001;16:633–647. 12. Sapir S, Keidar A, Van Velzen D. Vocal attrition in teachers: survey findings. Eur J Disord Commun. 1993;28:177–185. 13. Van Houtte E, Claey S, Wuyts F, Van Lierde K. The impact of voice disorders among teachers: vocal complaints, treatment-seeking behaviour, knowledge of vocal care and voice-related absenteeism. J Voice. Jul 14, 2010; [Epub ahead of print]. 14. Verdolini K, Ramig OR. Review: occupational risks for voice problems. Logoped Phoniatr Vocol. 2001;26:37–46. 15. Russell A, Oates J, Greenwood KM. Prevalence of voice problems in teachers. J Voice. 1998;12:467–479. 16. Miller MK, Verdolini K. Frequency and risk factors for voice problems in teachers of singing and control subjects. J Voice. 1995;9:348–362. 17. Herrington-Hall BL, Lee L, Stemple JC, Niemi KR, McHone MM. Description of laryngeal pathologies by age, sex and occupation in a treatment-seeking sample. J Speech Hearing Disord. 1998;53:57–64. 18. Pekkarinen E, Himberg L, Pentti J. Prevalence of vocal symptoms among teachers compared with nurses: a questionnaire study. Scand J Logoped Phonatr. 1992;17:113–117. 19. Thibeault SL, Merrill RM, Roy N, Gray SD, Smith EM. Occupational risk factors associated with voce disorders among teachers. Ann Epidemiol. 2004;14:786–792.
e147
20. Simberg S, Sala E, Vehmas K, Laine A. Changes in the prevalence of vocal symptoms among teachers during a twelve-year period. J Voice. 2005;19: 95–102. 21. de Medeiros AM, Barreto SM, Assunc¸~ao AA. Voice disorders (dysphonia) in public school female teachers working in Belo Horizonte: prevalence and associated factors. J Voice. 2008;22:676–687. 22. Urrutikoetxea A, Ispizua A, Matellanes F, Aurrekoetxea J. Prevalence of vocal nodules in teachers. Paper presented at: the First World Congress of Voice. Oporto, Portugal: 1995. 23. Gotaas C, Starr CD. Vocal fatigue among teachers. Folia Phoniatr (Basel). 1993;45:120–129. 24. Chapter 19: Education. In: Yearbook of Statistics Singapore. 2008: 243–261. 25. Chapter 10. In: Education Statistics Digest 2007. Ministry of Education. 2008:14–17. 26. Mathieson L. Vocal tract discomfort in hyperfunctional dysphonia. J Voice. 1993;2:40–48. 27. Angelillo M, Di Maio G, Costa G, Angelillo N, Barillari U. Prevalence of occupational voice disorders in teachers. J Prev Med Hyg. 2009;50:26–32. 28. Roy N, Weinrich B, Gray SD, Tanner K, Toledo SW, Dove H, Corbin-Lewis K, Stemple JC. Voice amplification versus vocal hygiene instruction for teachers with voice disorders: a treatment outcomes study. J Speech Lang Hear Res. 2002;45:625–638. 29. Santaloff RT, Castell DO, Katz PO, Sataloff DM. Reflux Laryngitis and Related Disorders. San Diego, CA: Singular Publishing Group; 1997. 30. Roy N, Merrill RM, Gray SD, Smith EM. Voice disorders in the general population: prevalence, risk factors, and occupational impact. Laryngoscope. 2005;115:1988–1995. 31. Preciado J, Perez C, Calzada M, Preciado P. Frequency and risk factors of voice disorders among teaching staff of La Rioja, Spain. Acta Otorrinolaringol Esp. 2005;56:161–170. [in Spanish]. 32. Bove MJ, Rosen C. Diagnosis and management of laryngopharyngeal reflux disease. Curr Opin Otolaryngol Head Neck Surg. 2006;14:116–123. 33. Ali Mel-S. Laryngopharyngeal reflux: diagnosis and treatment of a controversial disease. Curr Opin Allergy Clin Immunol. 2008;8:28–33. 34. Schneider B, Bigenzahn W. Vocal risk factors for occupational voice disorders in female teaching students. Eur Arch Otorhinolaryngol. 2005; 262:272–276.