gms | German Medical Science

100 Jahre Phoniatrie in Deutschland
22. Wissenschaftliche Jahrestagung der Deutschen Gesellschaft für Phoniatrie und Pädaudiologie
24. Kongress der Union Europäischer Phoniater

Deutsche Gesellschaft für Phoniatrie und Pädaudiologie e. V.

16. bis 18.09.2005, Berlin

Voice pathology morbidity screening in teachers

Stimmpathologie-Morbiditätsscreening bei Lehrern

Vortrag

Suche in Medline nach

  • corresponding author presenting/speaker Dosio Doskov - Queen Joanna Hospital, Medical University, Audiology and Phoniatrics, Sofia, Bulgaria
  • author Peter Dyakov - Queen Joanna Hospital, Medical University, Audiology and Phoniatrics, Sofia, Bulgaria

100 Jahre Phoniatrie in Deutschland. 22. Jahrestagung der Deutschen Gesellschaft für Phoniatrie und Pädaudiologie, 24. Kongress der Union der Europäischen Phoniater. Berlin, 16.-18.09.2005. Düsseldorf, Köln: German Medical Science; 2005. Doc05dgppV28

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter: http://www.egms.de/de/meetings/dgpp2005/05dgpp083.shtml

Veröffentlicht: 15. September 2005

© 2005 Doskov et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.de). Er darf vervielf&aauml;ltigt, verbreitet und &oauml;ffentlich zug&aauml;nglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.


Gliederung

Abstract

The authors share their experience concerning a screening study of the voice signal in 250 teachers, examined twice with an interval of two years. We compare the results of the ENT exam and the acoustic analysis performed by a system running Praat. A voice sample was recorded and Formants, Frequency range, Max fonation time, Voice breaks, Jitter, Shimmer & Additive noise were compared. The comparative analysis reveals that subjective methods are at least twice less accurate for early diagnosis of laryngeal pathology compared to acoustic screening. The proposed modification allows the fast processing of a great number of subjects for timely diagnosis of professional voice pathology.


Text

Introduction

Teachers represent a group of professionals who are very likely to develop voice associated pathology and indeed occupational voice disorders are extremely widely spread among teachers worldwide. The unbeneficial work environment is probably the main reason for that, second as a direct result comes voice strain and misuse - raised speech pressure level, combined with increased mechanical demand to the vocal cords. This gradually leads to functional changes, ordinary vocal fatigue at the beginning, that later with the continuation of the voice strain develop to objective pathological changes of the vocal cords.

The aims of the current study are as follow:

1. to investigate the incidence of vocal symptoms and voice disorders among the observed group of teachers,

2. to develop a protocol for voice screening,

3. to estimate the number of teachers reporting voice symptoms and to track the change in their number and pattern over the 2 year period,

4. to assess the prediction value of the voice report for developing of voice disorder.

Material and methods

The study encompasses 250 teachers who were followed for 2 years. All participants were subject to the following examination and analysis:

• questionnaire - a modified Sataloff's questionnaire for professional voice users, consisting of 50 closed yes/no questions, covering present voice status, habits, drug usage, work environment etc.

• standard ENT examination, including laryngoscopy

• recording of a voice sample consisting of a sustained vowel (a:) for 10 sec, again sustained (a:) to measure max phonation time, and 3 test phrases designed to contain all vowels and consonants of Bulgarian language.

Initially the group (group 1) consisted of 231 women and 19 men. The first year 250 teachers were evaluated, the second 242 and therefore 8 subjects (7 women and 1 men) were excluded from the survey due to unavailability of second examination (group 2). Mean age of participants was 39 years and the mean weekly work time was 16 hours. The mean years of practice were 17 (1-41). No prophylactic or therapeutic measures were prescribed for the period between the two examinations.

The symptoms monitored in the questionnaire were voice fatigue, throat clearing or coughing, tension on the articulary organs, voice hoarseness, voice breaks, present of past colds, pain in the throat and loss of voice.

Laryngoscopy was performed of all subjects. Voice recordings were made using a condenser microphone and a portable computer using an external analog/digital convertor.

Voice samples were later processed and for the purpose of this study only recording of the sustained vowel (a:) was used for further analysis. The sample was analysed using the freely available software for voice analysis Praat and the Voice report function was used. We created a script that automatically batch processed all sound files and saved voice report variables to a standard text file. Data were than imported in statistical software (SPSS 11 for Windows) and statistically analyzed.

Results and discussion

Frequency of vocal symptoms among both groups was high, in the second being more frequent. Of all the subjects in group 1, 69% reported at least two symptoms occurring frequently, and 21% reported three or more symptoms during the past year. In group 2 72% reported at least two symptoms, and 13% reported three or more symptoms.

The most common symptoms occurring in the first group were throat clearing or coughing (39%), voice fatigue (25%), tension in the articulary organs (19%) and voice hoarseness (15%). In the second group throat clearing remains in the leading position with 41%, followed by tension in 22%, voice fatigue in 17% and hoarsness in 17%.

Of all 250 subjects in the first group , 63 (25%) had marked disturbance of voice quality noted during the examination by the ENT specialist. In the second this percent was 71 (29%), giving a significance level lower than 0.01.

Voice analysis showed that these percents should be considerably higher since in group 1, 145 of 250 subjects (58%) demonstrated abnormal reading of the observed voice parameters, while in group 2 the percent was even higher - 151 (62%). Of all 145 subjects in the first group, 142 (97%) remained with abnormal values of voice parameters and in 34% the deterioration was higher than 10% compared to the initial value. Shimmer (apq3), Shimmer (apq11) and Shimmer (ddp) (PRAAT's original Get shimmer) were the most sensitive parameters. All newly discovered pathological voices in the second group had borderline or slightly elevated values of these three parameters in the first group. The differences in jitter and harmonicity were not significant.

Traditionally for screening purposes only perceptual analysis was used.There are studies however showing that traditional screening might not be enough to detect the earliest changes leading to voice pathology. Further more from the perspective of the current study, which shows over a double sensitivity of objective voice analysis over perceptual one and a correlation between parameter changes and future noticeable the practical consequences should be to more widely introduce in practice the machine screening for voice pathology.

Conclusion

The present study indicates that the number of teachers who might need voice therapy and medical care might actually be even higher than is revealed by statistics from treatment-seeking populations.

The comparative analysis reveals that subjective methods are at least twice less accurate for early diagnosis of laryngeal pathology compared to acoustic screening. The proposed modification allows the fast processing of a great number of subjects for timely diagnosis of professional voice pathology.