- Email: [email protected]
Immediate Effect of Stimulability Assessment on Acoustic, Aerodynamic, and Patient-Perceptual Measures of Voice
Immediate Effect of Stimulability Assessment on Acoustic, Aerodynamic, and Patient-Perceptual Measures of Voice Amanda I. Gillespie and Jackie Gartner-Schmidt, Pittsburgh, Pennsylvania Summary: Objectives. The purpose of the present study was to determine if patients with voice disorders could achieve immediate improvements in acoustic and aerodynamic parameters and patient’s perception of the sound and feel of voice following instructions to use clear speech. Study Design. This is a retrospective study. Methods. A total of 114 patients underwent acoustic and aerodynamic analyses of voice before and after instructions to use ‘‘clear speech,’’ while reading a standardized passage. The patient’s and speech-language pathologist’s (SLP) judgments of voice change were also measured. Results. An increase (improvement) in average vocal intensity (P < 0.001), average airflow (P < 0.001), total breaths taken (P < 0.001), total reading time (P < 0.001), and breaths/second (P < 0.001) was observed as a function of ‘‘clear speech’’ intervention. No change in acoustic analyses was observed. Most patients reported an improvement in the sound or feel of voice immediately after the assessment. Conclusions. This study is the first to quantify acoustic and aerodynamic parameters and clinician’s and patient’s perceptions of a standardized stimulability test for voice change in the voice laboratory. These results are potentially paradigm shifting in the role of the SLP in the evaluative clinic setting. Key Words: Voice–Stimulability–Aerodynamics–Acoustics–Voice laboratory. INTRODUCTION Stimulability assessment refers to the evaluation of an individual’s ability to modify a behavior when provided with models or cues.1,2 This type of assessment provides the clinician with insight into a patient’s capability to change a specific behavior.2 Findings from stimulability assessment may help guide treatment recommendations.1,2 The procedures for assessment of how stimulable a patient is can range from objective to informal, based on clinician instinct more so than objective outcomes.1 A person is considered to have ‘‘good stimulability’’ for changing a specific behavior when the stimulability production (following a model or cue) is better than the original, or spontaneous, production of the same.1 Stimulability testing was first introduced in the speechlanguage pathology literature in 1954 as a useful tool for the evaluation of children with speech sound disorders.1 Stimulability assessment has received scant mention in the voice literature.3,4 To the best of our knowledge, only two studies exist that objectively evaluated a stimulability protocol for patients with voice disorders. Bonilha and Dawson3 assessed if acoustic analyses of voice would improve immediately after a stimulability trial of resonant voice. The authors hypothesized that immediate voice improvements experienced by patients during the first meeting with the speech-language pathologist (SLP) would improve the patient’s sense of mastery experience and therefore improve self-efficacy.3,5 Acoustic improvements Accepted for publication June 8, 2015. From the Department of Otolaryngology, University of Pittsburgh Voice Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania. Address correspondence and reprint requests to Amanda I. Gillespie, Department of Otolaryngology, University of Pittsburgh Voice Center, University of Pittsburgh Medical Center Mercy, 1400 Locust Street, Suite 11-500, Building B, Pittsburgh, PA 15219. E-mail: [email protected] Journal of Voice, Vol. -, No. -, pp. 1-6 0892-1997/$36.00 Ó 2015 Published by Elsevier Inc. on behalf of The Voice Foundation http://dx.doi.org/10.1016/j.jvoice.2015.06.004
were seen in fundamental frequency, jitter, and shimmer after the stimulability trial. It is unknown, however, if patients or SLPs were subjectively sensitive to the change. It is also unknown if phonatory aerodynamics improved with the stimulability trial. Similarly, Dejonckere and Lebacq4 investigated the immediate effects of instructing patients with voice disorders to change aspects of communication, a phenomenon the authors termed, ‘‘vocal plasticity.’’ The investigators used a variety of techniques, including resonance, articulation, breathing, and changes in intensity, to determine if the patient’s ability to achieve immediate voice change could predict therapeutic success. Therapy success was determined as an improvement in acoustic analyses, stroboscopic analyses, or both. The authors found a positive correlation between the subjective rating of good ‘‘vocal plasticity’’ and therapeutic success. It is unknown if aerodynamic function improved with stimulability testing or if the patient’s perception of his/her voice problem improved with treatment. Furthermore, the stimulability techniques used were not standardized; thus, it is unclear which techniques provided the most robust assessment results. Many stimulability test results in the field of speech sound assessment are task or sound specific. For example, a pediatric patient may be stimulable to produce the phoneme /r/, but not /s/.1 Similarly, for patients with voice problems, some may complain more of a problem with the ‘‘feel’’ of voice production than with the ‘‘sound.’’ It is unknown if stimulability assessment can address these two components of voice. It is therefore important to assess the patient’s perception of both a change in the sound and feel of voice when conducting stimulability testing for patients with voice problems. Clear speech as a stimulability technique One technique that may be useful for assessing voice patient stimulability is the use of ‘‘clear speech.’’6,7 The ‘‘clear
2 speech’’ technique was initially developed as an intelligibility strategy for speaking to listeners with hearing loss.6 When speakers use clear speech, increases are observed in phrase duration, number and duration of pauses, individual speech sound durations, intensity, and intonation contour. Furthermore, speaking rate decreases, vowels are less reduced, and a wider range of fundamental frequency values are produced.7 These changes have been observed immediately after clear speech instructions in normal speakers and did not require extensive training. Of importance to the voice clinician, these changes are similar to those expected in speech after direct voice therapy. Therefore, it was hypothesized that patients with good stimulability for clear speech would likely be appropriate for further therapeutic intervention. In other words, if a patient can improve the sound or feel of his/her voice after a very brief ‘‘clear speech’’ instruction, they were considered good candidates for voice therapy because greater gains could be expected with more intensive intervention. Furthermore, unlike other traditional voice therapy techniques which build hierarchically from single phonemes to words, phrases, and finally, speech, ‘‘clear speech’’ is an instruction only for connected speech. Therefore, if the goal of voice therapy is improved, balanced phonation in connected speech, it is critical to assess a patient’s stimulability for change ‘‘in connected speech.’’ The purpose of the present study was to determine if patients with voice disorders could achieve immediate improvements in acoustic and aerodynamic parameters and patient’s perception of the sound and feel of voice following a standardized stimulability protocol, which included instructions to use clear speech. Because many voice disorders treated with voice therapy (eg, benign lesions, muscle tension dysphonia, vocal fold atrophy), present with abnormalities in at least one of these aspects of vocal communication, it is hypothesized that clear speech may be a useful treatment strategy for some voice disorders. In addition, because the effects of clear speech are observed immediately, we hypothesized that clear speech may be a fruitful tool in assessing stimulability for behavioral change in patients with voice problems.
Journal of Voice, Vol. -, No. -, 2015
Participants This study used a retrospective review of clinical data collected from the University of Pittsburgh Voice Center database from November 2013–October 2014. Informed consent was obtained from all patients before data entry into the database. The data were retrieved from the database by a research assistant (RA). Patients were included in the study if they had completed voice laboratory data set, operationally defined as preclear and postclear speech stimulability instruction recordings of the Rainbow Passage.8
plinary clinic consisting of fellowship-trained laryngologist and voice-specialized SLP. Part of the SLP assessment involves determination of patient’s candidacy for voice therapy. On the basis of clinical acumen, the SLP determined which patients may be able to achieve total or partial amelioration of voicerelated complaints through behavioral intervention. For example, patients with severe vocal fold paralysis and large glottal gaps and those with neurologic disorders such as spasmodic dysphonia were not included. Patients generally amenable to behavioral intervention (eg, vocal fold lesions, muscle tension dysphonia, vocal fold atrophy) were more likely to undergo standardized stimulability testing (SST).9 In addition, because voice therapy referrals are made on the basis of presenting voice behavior and are not diagnosis specific, care was taken not to exclude patients with certain diagnoses (ie, atrophy) because often these patients are using maladaptive compensatory techniques, such as hyperfunction, which may be amendable to therapeutic intervention, if not just to decrease maladaptive compensations before augmentation laryngoscopy if needed. The SST protocol was as follows. First, patients read the first four sentences of the Rainbow Passage.10 The Rainbow Passage is a commonly used reading passage to elicit an oral reading sample. Patients were instructed to use their ‘‘normal’’ voice (ie, presenting, dysphonic voice). This reading was recorded using the Phonatory Aerodynamic System (PAS) 6600 (KayPENTAX, Montvale, NJ). The system consists of a face mask coupled to a pneumotachometer with integral calibrated microphone. Patients sat upright and held the face mask snugly over their nose and mouth and read the passage, which was printed on a single sheet of paper in large font size and held in front of them. Patients were asked to check for any air leaks around the face mask by gently exhaling first before reading the passage. After the first reading, the SLP gave the patient instructions to read the passage again using ‘‘clear speech.’’ Specifically, the SLP asked the patient to use ‘‘crisp, clear consonants’’ and ‘‘precise articulation’’ when reading. The SLP also instructed the patient to vary his/her intonation/pitch when reading and used the example, ‘‘like you are reading or telling an interesting story.’’ Patients were dissuaded from having ‘‘sing-songy’’ inflection but, rather, inflection that felt ‘‘normal’’ to them. The patient read the passage again using the PAS6600. After the second reading, the SLP asked the patient if he/she noticed a change in the sound and/or feel of voice when reading the second time compared with the first and noted this information in the patient’s medical chart. The SLP also noted if he/she heard a change in the overall severity of the sound of the patient’s voice in the chart. The purpose of the SLP’s voice assessment ‘‘on line’’ was to attempt to quantify the judgments SLPs have to make quickly on appropriateness for voice therapy during a clinic visit. Both recordings were saved in an electronic passwordprotected folder.
Procedures Standard clinical care procedures at the University of Pittsburgh Voice Center include assessment of all patients in a multidisci-
Data reduction A blinded RA completed all data reduction. All patients with complete recordings were included in analysis. First, the RA
MATERIALS AND METHODS All study procedures were approved by the University of Pittsburgh institutional review board (IRB PRO13030372).
Amanda I. Gillespie and Jackie Gartner-Schmidt
Immediate Effect of Stimulability Assessment
listened to each sample to locate errors in reading (including word repetitions, mispronunciations followed by corrections) or extraneous sounds such as coughs, throat clears, and so forth. These were removed from the recording, so that a fluent reading of the passage remained. Cursors were manually placed at the beginning and end of the dB SPL tracing and analyzed using a custom protocol in the PAS6600. The protocol displayed tracings of pitch, dB SPL, and average phonatory airflow; as well as negative airflow tracings to view inhalations. The following data were collected from each reading: duration of passage, number of breaths, average dB SPL, average fundamental frequency (F0), and average phonatory airflow during speech. Next, both the prestimulability and poststimulability readings of the sentence, ‘‘The rainbow is a division of white light,’’ were viewed in the Analysis of Dysphonia in Speech and Voice (ADSV) program of the Computerized Speech Lab (KayPENTAX, Montvale, NJ). This sentence has been used by past authors for acoustic voice analyses.11–13 The cepstral spectral index of dysphonia (CSID) was calculated using the ‘‘allvoiced sentence’’ protocol in ADSV. CSID is a time- and frequency-based algorithm that has demonstrated sensitivity to change after treatment and correlates well to the overall severity auditory-perceptual voice rating on the Consensus Auditory-Perceptual Evaluation of Voice.14–17 Finally, pitch variation, measured as the standard deviation (SD) of the fundamental frequency calculated from the cepstral peak prominence (CPP F0 SD), was also analyzed for the same sentence. Statistical methods Means and SDs were calculated for all measures. Aerodynamic outcomes were analyzed with a 2 3 5 within-subjects analysis of variance, with aerodynamic measurement (five levels: duration of passage, number of breaths, average dB SPL, average fundamental frequency (F0), and average phonatory airflow during speech) and time (two levels: pre-SST and post-SST).
3
Acoustic outcome was analyzed with a paired-sample t test. SPSS 22 (IBM, Armonk, NY, USA) for Windows was used for all analyses. RESULTS Participants Records from 114 patients were included in the study. Most patients were female (66% female). The average age was 50 years, with a range of 15–89 years. Patients included had the following diagnoses: muscle tension dysphonia (n ¼ 33), lesions (n ¼ 32), hypomobility (n ¼ 4), immobility (n ¼ 22), atrophy (n ¼ 19), ‘‘other’’ (ie, reflux, baseline examination; n ¼ 4). Aerodynamic analyses There was a significant difference between time (pre-SST/postSST) and measure, indicating an increase (improvement) in average vocal intensity (P < 0.001), average airflow (P < 0.001), total breaths taken (P < 0.001), total reading time (P < 0.001), and breaths/second (P < 0.001) as a function of ‘‘clear speech’’ intervention. For the patients with diagnoses that may result in glottal incompetence (eg, atrophy, hypomobility), three had prestimulability average airflow values above normal. Two patients showed a decrease in flow with SST and one showed an increase. None of these patients were deemed stimulable for behavioral voice intervention on the basis of results of SST. Figures 1 and 2 depict a typical PAS6600 tracing of a 62-year-old female patient with vocal fold atrophy and secondary muscle tension dysphonia before (Figure 1) and after (Figure 2) ‘‘clear speech’’ intervention. Acoustic analyses No significant difference in CSID following ‘‘clear speech’’ intervention was observed (t ¼ 1.114; P ¼ 0.268). Average CSID scores at baseline were 3.89; SD, 19.3; range, 26.31 to 80.71. Following ‘‘clear speech,’’ average CSID scores
FIGURE 1. Pre- stimulability tracing of a patient speaking the Rainbow Passage. Average airflow ¼ 60 ml/sec, breath number ¼ 3, dB SPL ¼ 75, total speaking duration ¼ 23 seconds, average F0 ¼ 130 Hz.
4
Journal of Voice, Vol. -, No. -, 2015
FIGURE 2. Post stimulability trial of the same patient as in Figure 1. Average airflow ¼ 150 ml/sec, total breaths ¼ 4, dB SPL ¼ 79, total speaking duration ¼ 31 sec, average F0 ¼ 171 Hz.
were 5.77; SD, 15.19; range, 16.33 to 80.46. Similarly, no significant difference in CPP F0 SD, or a measure of pitch variability, was observed following the ‘‘clear speech’’ instructions (t ¼ 0.693; P ¼ 0.490). Patient’s perception Approximately, 57% (56.5) of patients noted an improvement in the sound of voice after the SST. Furthermore, 46.1% of patients reported an improvement in the feel of voice after stimulability. Nineteen patients who reported their voice sounded better did not think the feel of voice improved. Stated differently, 43 patients (37%) noted an improvement in both the sound and feel of voice after the intervention. Of the 53.9% of patients who did not notice an improvement in the feel of voice, four (0.07%) felt the ‘‘clear speech’’ trial made their voice feel worse. All these patients were diagnosed with muscle tension dysphonia. Clinician’s perception In 64.3% of cases, the SLP noted an improvement in the sound of voice after ‘‘clear speech’’ intervention. In 59 (51%) of cases, both the SLP and patient agreed that there was an improvement in the sound of the voice after the intervention.
Stimulability by diagnosis To assess if SST was equally effective between disorders, the results of the SLP’s and patient’s perceptions of improvement were assessed per diagnosis (Table 1). Most patients were noted by the SLP to have vocal improvement after stimulability. Similarly, most patients self-perceived a better-sounding voice for all diagnoses except immobility. Finally, only most patients with MTD noted an improvement in the feel of voice after SST. The number of patients per diagnostic group recommended for behavioral voice intervention after SST is indicated in Table 1. Of note, in each diagnostic group, there were patients for whom the SLP did not perceive a change in the sound of the voice, but because of other factors including patient’s perception, they were still recommended for voice therapy. For patients with MTD, this number was seven; for lesions, two; hypomobility, one; immobility, three; atrophy, four.
DISCUSSION This study is the first to quantify acoustic and aerodynamic parameters and clinician’s and patient’s perceptions of stimulability to voice change in the voice laboratory. These results are potentially paradigm shifting in the role of the SLP in the evaluative clinic setting for the following reasons.
TABLE 1. Clinician’s and Patient’s Perceptions of Voice Improvement by Diagnosis Diagnosis MTD (n ¼ 33) Lesion (n ¼ 32) Hypomobility (n ¼ 4) Immobility (n ¼ 22) Atrophy (n ¼ 19)
Sound Better Per SLP
Sound Better Per Patient
Feel Better Per Patient
Recommended for Voice Therapy
24 (73%) 21 (66%) 3 (75%) 11 (50%) 10 (53%)
22 (67%) 19 (59%) 2 (50%) 7 (32%) 10 (53%)
21 (64%) 14 (44%) 1 (25%) 6 (27%) 5 (26%)
21 (87%) 24 (75%) 3 (75%) 9 (41%) 11 (57%)
Amanda I. Gillespie and Jackie Gartner-Schmidt
Immediate Effect of Stimulability Assessment
First, results indicate that immediate voice improvement is possible after ‘‘clear speech’’ instructions in not only the sound but also the feel of voice. These results support past literature on stimulability testing, which demonstrated immediate voice change after resonant voice instruction.3 Results also indicate that the use of ‘‘clear speech’’ techniques improves aerodynamic aspects of voice production but did not significantly influence acoustic outcomes. The lack of significant change in acoustic measures is not surprising and is in agreement with the literature showing that many cepstral outcomes do not change, even after intensive voice therapy, or even surgery.14 In most cases, both the SLP and patient heard an improvement in the sound of the voice as a result of the stimulability task, which is an important indicator of success in voice. Despite not focusing specifically on laryngeal musculature (ie, via massage or reposturing), 46% of patients noted an improvement in the feel of voice after the stimulability trial. Overall, patients with MTD and lesions, after atrophy, demonstrated the greatest voice improvement after stimulability assessment. Patients with immobility and hypomobility improved the least. These findings are not surprising, given that many patients with immobility have glottal incompetence and may require surgical augmentation to improve phonation, whereas patients with MTD and lesions are most often treated with voice therapy.9 Second, the study adds to the literature on ‘‘clear speech’’ by demonstrating that some individuals with voice disorders, such as vocally healthy people,6,7 are able to achieve immediate voice improvement with use of the technique. Results are also similar to those observed in the literature following trials of resonant voice or flow phonation.3,4 ‘‘Clear speech’’ may be similar to these techniques in its use of forward focus (oral consonants production), which, such as resonant voice and flow phonation therapy, provides the speaker with kinesthetic feedback external to the gesture, which enhances learning.18– 20 However, unlike these techniques, the ‘‘clear speech’’ instruction is one that can only be given in connected speech; therefore, the clinician is able to assess the patient’s ability to manipulate his/her voice in conversation, which may relate to improved therapy recommendations and outcomes. This correlation will be the focus of future studies. Third, results of stimulability testing—both subjective and objective—provide the SLP with data to present to physician colleagues to support SLP’s recommendations for or against behavioral therapy. These data may assist in patients being appropriately referred for voice therapy, and thus improve voice therapy outcomes.21 Furthermore, the individual data can be presented to patients, which can assist with improved patient understanding of the therapeutic process, and as an extension, improve therapeutic buy-in, potentially removing some of the unknowns. Our past work demonstrated that the presence of an SLP in clinic improves not only therapy attendance but also voice outcomes after treatment.22 This study is the first to standardize stimulability testing and report its objective findings. Finally, this study introduces voice clinicians to a new way to use traditional voice laboratory techniques and equipment—not just as outcomes following intervention, but as stimulability for
5
change testing. This new method of SST in the diagnostic clinic provides further support for voice clinics to use a fully equipped voice laboratory staffed by a voice-specialized SLP. Limitations The major limitation in this study is the potential selection bias of patients chosen for stimulability testing. The examining SLP used her discretion based on experience and clinical acumen to determine which patients would most likely be stimulable for behavioral intervention based on clinical instinct and observation of presenting phonatory habits and techniques. However, despite this limitation, a range of voice disorders was represented in the sample (eg, MTD, lesions, atrophy, hypomobility, immobility), indicating that patients with many voice diagnoses may present with aberrant phonatory techniques that may improve with behavioral intervention, regardless of medical diagnosis, and voice assessment by an SLP should not be limited only to patients with certain voice disorders. A second limitation is that the present study did not follow the participants to track therapeutic success. The goal of the present study was to determine if patients with voice problems are able to make immediate changes in connected speech using the ‘‘clear speech’’ technique and investigate its use as part of a standardized stimulability test. A prospective investigation correlating SST with therapy results will be addressed in future study protocols. CONCLUSIONS In conclusion, this study was the first to operationalize SST performed by an SLP as part of the voice diagnostic clinic for patients with voice problems. It was also the first to provide objective and subjective outcomes of stimulability testing. Future research should explore the predictive value of stimulability testing on voice therapy outcomes. Acknowledgments The authors would like to thank Maurice Goodwin for his assistance. REFERENCES 1. Powell TW, Miccio AW. Stimulability: a useful clinical tool. J Commun Dis. 1996;29:237–253. 2. Tyler AA. Assessing stimulability in toddlers. J Commun Dis. 1996;29: 279–297. 3. Bonilha HS, Dawson AE. Creating a mastery experience during the voice evaluation. J Voice. 2012;26:665.e1–665.e7. 4. Dejonckere PH, Lebacq J. Plasticity of voice quality: a prognostic factor for outcome of voice therapy? J Voice. 2001;15:251–256. 5. Bandura A. Toward a unifying theory of behavioral change. Psychol Rev. 1977;84:191–215. 6. Picheny M, Durlach N, Braida L. Speaking clearly for the hard of hearing I: intelligibility differences between clear and conversational speech. J Speech Hear Res. 1985;28:96–103. 7. Picheny M, Durlach N, Braida L. Speaking clearly for the hard of hearing. II: acoustic characteristics of clear and conversational speech. J Speech Hear Res. 1986;29:434–446. 8. Fairbanks G. Voice and Articulation Drill Book. 2nd ed. New York: Harper and Row; 1960.
6 9. Rosen CA, Gartner-Schmidt J, Hathaway B, et al. A nomenclature paradigm for benign midmembranous vocal fold lesions. Laryngoscope. 2012;122:1335–1341. 10. Baken RJ, Orlikoff RF. Clinical Measurement of Speech & Voice. 2nd ed. San Diego: Singular; 2000. 11. Lowell SY, Colton RH, Kelley RT, Mizia SA. Predictive value and discriminant capacity of cepstral- and spectral-based measures during continuous speech. J Voice. 2013;27:393–400. 12. Awan SN, Roy N, Cohen SM. Exploring the relationship between spectral and cepstral measures of voice and the Voice Handicap Index (VHI). J Voice. 2014;28:430–439. 13. Eadie TL, Doyle PC. Classification of dysphonic voice: acoustic and auditory-perceptual measures. J Voice. 2005;19:1–14. 14. Gillespie AI, Dastolfo C, Magid N, Gartner-Schmidt J. Acoustic analysis of four common voice diagnoses: moving toward disorder-specific assessment. J Voice. 2014;28:582–588. 15. Peterson EA, Roy N, Awan SN, Merrill RM, Banks R, Tanner K. Toward validation of the cepstral spectral index of dysphonia (CSID) as an objective treatment outcomes measure. J Voice. 2013;27:401–410.
Journal of Voice, Vol. -, No. -, 2015 16. Awan SN, Roy N, Jiang JJ. Nonlinear dynamic analysis of disordered voice: the relationship between the correlation dimension (D(2)) and pre-/posttreatment change in perceived dysphonia severity. J Voice. 2009;24:285–293. 17. Awan SN, Roy N. Outcomes measurement in voice disorders: application of an acoustic index of dysphonia severity. J Speech Lang Hear Res. 2009;52:482–499. 18. Verdolini Abbott K. Lessac-Madsen Resonant Voice Therapy. San Diego, CA: Plural Publishing; 2008. 19. Wulf G, Prinz W. Directing attention to movement effects enhances learning: a review. Psychon Bull Rev. 2001;8:648–660. 20. Schmidt RA, Lee TD. Motor Control and Learning. A Behavioral Emphasis. 4th ed. Champaign, IL: Human Kinetics Publisher; 2005. 21. Gartner-Schmidt J. Voice therapy for voice disorders. In: Johnson JT, Rosen CA, eds. Bailey’s Head and Neck Surgery-Otolaryngology. 5th ed. Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins; 2013. 22. Litts J, Gartner-Schmidt J, Clary M, Gillespie AI. Impact of combined laryngologist and speech-language pathologist co-assessment on treatment outcomes and billing revenue. Laryngoscope. 2015; http://dx.doi.org/10. 1002/lary.25349 [Epub ahead of print].
were seen in fundamental frequency, jitter, and shimmer after the stimulability trial. It is unknown, however, if patients or SLPs were subjectively sensitive to the change. It is also unknown if phonatory aerodynamics improved with the stimulability trial. Similarly, Dejonckere and Lebacq4 investigated the immediate effects of instructing patients with voice disorders to change aspects of communication, a phenomenon the authors termed, ‘‘vocal plasticity.’’ The investigators used a variety of techniques, including resonance, articulation, breathing, and changes in intensity, to determine if the patient’s ability to achieve immediate voice change could predict therapeutic success. Therapy success was determined as an improvement in acoustic analyses, stroboscopic analyses, or both. The authors found a positive correlation between the subjective rating of good ‘‘vocal plasticity’’ and therapeutic success. It is unknown if aerodynamic function improved with stimulability testing or if the patient’s perception of his/her voice problem improved with treatment. Furthermore, the stimulability techniques used were not standardized; thus, it is unclear which techniques provided the most robust assessment results. Many stimulability test results in the field of speech sound assessment are task or sound specific. For example, a pediatric patient may be stimulable to produce the phoneme /r/, but not /s/.1 Similarly, for patients with voice problems, some may complain more of a problem with the ‘‘feel’’ of voice production than with the ‘‘sound.’’ It is unknown if stimulability assessment can address these two components of voice. It is therefore important to assess the patient’s perception of both a change in the sound and feel of voice when conducting stimulability testing for patients with voice problems. Clear speech as a stimulability technique One technique that may be useful for assessing voice patient stimulability is the use of ‘‘clear speech.’’6,7 The ‘‘clear
2 speech’’ technique was initially developed as an intelligibility strategy for speaking to listeners with hearing loss.6 When speakers use clear speech, increases are observed in phrase duration, number and duration of pauses, individual speech sound durations, intensity, and intonation contour. Furthermore, speaking rate decreases, vowels are less reduced, and a wider range of fundamental frequency values are produced.7 These changes have been observed immediately after clear speech instructions in normal speakers and did not require extensive training. Of importance to the voice clinician, these changes are similar to those expected in speech after direct voice therapy. Therefore, it was hypothesized that patients with good stimulability for clear speech would likely be appropriate for further therapeutic intervention. In other words, if a patient can improve the sound or feel of his/her voice after a very brief ‘‘clear speech’’ instruction, they were considered good candidates for voice therapy because greater gains could be expected with more intensive intervention. Furthermore, unlike other traditional voice therapy techniques which build hierarchically from single phonemes to words, phrases, and finally, speech, ‘‘clear speech’’ is an instruction only for connected speech. Therefore, if the goal of voice therapy is improved, balanced phonation in connected speech, it is critical to assess a patient’s stimulability for change ‘‘in connected speech.’’ The purpose of the present study was to determine if patients with voice disorders could achieve immediate improvements in acoustic and aerodynamic parameters and patient’s perception of the sound and feel of voice following a standardized stimulability protocol, which included instructions to use clear speech. Because many voice disorders treated with voice therapy (eg, benign lesions, muscle tension dysphonia, vocal fold atrophy), present with abnormalities in at least one of these aspects of vocal communication, it is hypothesized that clear speech may be a useful treatment strategy for some voice disorders. In addition, because the effects of clear speech are observed immediately, we hypothesized that clear speech may be a fruitful tool in assessing stimulability for behavioral change in patients with voice problems.
Journal of Voice, Vol. -, No. -, 2015
Participants This study used a retrospective review of clinical data collected from the University of Pittsburgh Voice Center database from November 2013–October 2014. Informed consent was obtained from all patients before data entry into the database. The data were retrieved from the database by a research assistant (RA). Patients were included in the study if they had completed voice laboratory data set, operationally defined as preclear and postclear speech stimulability instruction recordings of the Rainbow Passage.8
plinary clinic consisting of fellowship-trained laryngologist and voice-specialized SLP. Part of the SLP assessment involves determination of patient’s candidacy for voice therapy. On the basis of clinical acumen, the SLP determined which patients may be able to achieve total or partial amelioration of voicerelated complaints through behavioral intervention. For example, patients with severe vocal fold paralysis and large glottal gaps and those with neurologic disorders such as spasmodic dysphonia were not included. Patients generally amenable to behavioral intervention (eg, vocal fold lesions, muscle tension dysphonia, vocal fold atrophy) were more likely to undergo standardized stimulability testing (SST).9 In addition, because voice therapy referrals are made on the basis of presenting voice behavior and are not diagnosis specific, care was taken not to exclude patients with certain diagnoses (ie, atrophy) because often these patients are using maladaptive compensatory techniques, such as hyperfunction, which may be amendable to therapeutic intervention, if not just to decrease maladaptive compensations before augmentation laryngoscopy if needed. The SST protocol was as follows. First, patients read the first four sentences of the Rainbow Passage.10 The Rainbow Passage is a commonly used reading passage to elicit an oral reading sample. Patients were instructed to use their ‘‘normal’’ voice (ie, presenting, dysphonic voice). This reading was recorded using the Phonatory Aerodynamic System (PAS) 6600 (KayPENTAX, Montvale, NJ). The system consists of a face mask coupled to a pneumotachometer with integral calibrated microphone. Patients sat upright and held the face mask snugly over their nose and mouth and read the passage, which was printed on a single sheet of paper in large font size and held in front of them. Patients were asked to check for any air leaks around the face mask by gently exhaling first before reading the passage. After the first reading, the SLP gave the patient instructions to read the passage again using ‘‘clear speech.’’ Specifically, the SLP asked the patient to use ‘‘crisp, clear consonants’’ and ‘‘precise articulation’’ when reading. The SLP also instructed the patient to vary his/her intonation/pitch when reading and used the example, ‘‘like you are reading or telling an interesting story.’’ Patients were dissuaded from having ‘‘sing-songy’’ inflection but, rather, inflection that felt ‘‘normal’’ to them. The patient read the passage again using the PAS6600. After the second reading, the SLP asked the patient if he/she noticed a change in the sound and/or feel of voice when reading the second time compared with the first and noted this information in the patient’s medical chart. The SLP also noted if he/she heard a change in the overall severity of the sound of the patient’s voice in the chart. The purpose of the SLP’s voice assessment ‘‘on line’’ was to attempt to quantify the judgments SLPs have to make quickly on appropriateness for voice therapy during a clinic visit. Both recordings were saved in an electronic passwordprotected folder.
Procedures Standard clinical care procedures at the University of Pittsburgh Voice Center include assessment of all patients in a multidisci-
Data reduction A blinded RA completed all data reduction. All patients with complete recordings were included in analysis. First, the RA
MATERIALS AND METHODS All study procedures were approved by the University of Pittsburgh institutional review board (IRB PRO13030372).
Amanda I. Gillespie and Jackie Gartner-Schmidt
Immediate Effect of Stimulability Assessment
listened to each sample to locate errors in reading (including word repetitions, mispronunciations followed by corrections) or extraneous sounds such as coughs, throat clears, and so forth. These were removed from the recording, so that a fluent reading of the passage remained. Cursors were manually placed at the beginning and end of the dB SPL tracing and analyzed using a custom protocol in the PAS6600. The protocol displayed tracings of pitch, dB SPL, and average phonatory airflow; as well as negative airflow tracings to view inhalations. The following data were collected from each reading: duration of passage, number of breaths, average dB SPL, average fundamental frequency (F0), and average phonatory airflow during speech. Next, both the prestimulability and poststimulability readings of the sentence, ‘‘The rainbow is a division of white light,’’ were viewed in the Analysis of Dysphonia in Speech and Voice (ADSV) program of the Computerized Speech Lab (KayPENTAX, Montvale, NJ). This sentence has been used by past authors for acoustic voice analyses.11–13 The cepstral spectral index of dysphonia (CSID) was calculated using the ‘‘allvoiced sentence’’ protocol in ADSV. CSID is a time- and frequency-based algorithm that has demonstrated sensitivity to change after treatment and correlates well to the overall severity auditory-perceptual voice rating on the Consensus Auditory-Perceptual Evaluation of Voice.14–17 Finally, pitch variation, measured as the standard deviation (SD) of the fundamental frequency calculated from the cepstral peak prominence (CPP F0 SD), was also analyzed for the same sentence. Statistical methods Means and SDs were calculated for all measures. Aerodynamic outcomes were analyzed with a 2 3 5 within-subjects analysis of variance, with aerodynamic measurement (five levels: duration of passage, number of breaths, average dB SPL, average fundamental frequency (F0), and average phonatory airflow during speech) and time (two levels: pre-SST and post-SST).
3
Acoustic outcome was analyzed with a paired-sample t test. SPSS 22 (IBM, Armonk, NY, USA) for Windows was used for all analyses. RESULTS Participants Records from 114 patients were included in the study. Most patients were female (66% female). The average age was 50 years, with a range of 15–89 years. Patients included had the following diagnoses: muscle tension dysphonia (n ¼ 33), lesions (n ¼ 32), hypomobility (n ¼ 4), immobility (n ¼ 22), atrophy (n ¼ 19), ‘‘other’’ (ie, reflux, baseline examination; n ¼ 4). Aerodynamic analyses There was a significant difference between time (pre-SST/postSST) and measure, indicating an increase (improvement) in average vocal intensity (P < 0.001), average airflow (P < 0.001), total breaths taken (P < 0.001), total reading time (P < 0.001), and breaths/second (P < 0.001) as a function of ‘‘clear speech’’ intervention. For the patients with diagnoses that may result in glottal incompetence (eg, atrophy, hypomobility), three had prestimulability average airflow values above normal. Two patients showed a decrease in flow with SST and one showed an increase. None of these patients were deemed stimulable for behavioral voice intervention on the basis of results of SST. Figures 1 and 2 depict a typical PAS6600 tracing of a 62-year-old female patient with vocal fold atrophy and secondary muscle tension dysphonia before (Figure 1) and after (Figure 2) ‘‘clear speech’’ intervention. Acoustic analyses No significant difference in CSID following ‘‘clear speech’’ intervention was observed (t ¼ 1.114; P ¼ 0.268). Average CSID scores at baseline were 3.89; SD, 19.3; range, 26.31 to 80.71. Following ‘‘clear speech,’’ average CSID scores
FIGURE 1. Pre- stimulability tracing of a patient speaking the Rainbow Passage. Average airflow ¼ 60 ml/sec, breath number ¼ 3, dB SPL ¼ 75, total speaking duration ¼ 23 seconds, average F0 ¼ 130 Hz.
4
Journal of Voice, Vol. -, No. -, 2015
FIGURE 2. Post stimulability trial of the same patient as in Figure 1. Average airflow ¼ 150 ml/sec, total breaths ¼ 4, dB SPL ¼ 79, total speaking duration ¼ 31 sec, average F0 ¼ 171 Hz.
were 5.77; SD, 15.19; range, 16.33 to 80.46. Similarly, no significant difference in CPP F0 SD, or a measure of pitch variability, was observed following the ‘‘clear speech’’ instructions (t ¼ 0.693; P ¼ 0.490). Patient’s perception Approximately, 57% (56.5) of patients noted an improvement in the sound of voice after the SST. Furthermore, 46.1% of patients reported an improvement in the feel of voice after stimulability. Nineteen patients who reported their voice sounded better did not think the feel of voice improved. Stated differently, 43 patients (37%) noted an improvement in both the sound and feel of voice after the intervention. Of the 53.9% of patients who did not notice an improvement in the feel of voice, four (0.07%) felt the ‘‘clear speech’’ trial made their voice feel worse. All these patients were diagnosed with muscle tension dysphonia. Clinician’s perception In 64.3% of cases, the SLP noted an improvement in the sound of voice after ‘‘clear speech’’ intervention. In 59 (51%) of cases, both the SLP and patient agreed that there was an improvement in the sound of the voice after the intervention.
Stimulability by diagnosis To assess if SST was equally effective between disorders, the results of the SLP’s and patient’s perceptions of improvement were assessed per diagnosis (Table 1). Most patients were noted by the SLP to have vocal improvement after stimulability. Similarly, most patients self-perceived a better-sounding voice for all diagnoses except immobility. Finally, only most patients with MTD noted an improvement in the feel of voice after SST. The number of patients per diagnostic group recommended for behavioral voice intervention after SST is indicated in Table 1. Of note, in each diagnostic group, there were patients for whom the SLP did not perceive a change in the sound of the voice, but because of other factors including patient’s perception, they were still recommended for voice therapy. For patients with MTD, this number was seven; for lesions, two; hypomobility, one; immobility, three; atrophy, four.
DISCUSSION This study is the first to quantify acoustic and aerodynamic parameters and clinician’s and patient’s perceptions of stimulability to voice change in the voice laboratory. These results are potentially paradigm shifting in the role of the SLP in the evaluative clinic setting for the following reasons.
TABLE 1. Clinician’s and Patient’s Perceptions of Voice Improvement by Diagnosis Diagnosis MTD (n ¼ 33) Lesion (n ¼ 32) Hypomobility (n ¼ 4) Immobility (n ¼ 22) Atrophy (n ¼ 19)
Sound Better Per SLP
Sound Better Per Patient
Feel Better Per Patient
Recommended for Voice Therapy
24 (73%) 21 (66%) 3 (75%) 11 (50%) 10 (53%)
22 (67%) 19 (59%) 2 (50%) 7 (32%) 10 (53%)
21 (64%) 14 (44%) 1 (25%) 6 (27%) 5 (26%)
21 (87%) 24 (75%) 3 (75%) 9 (41%) 11 (57%)
Amanda I. Gillespie and Jackie Gartner-Schmidt
Immediate Effect of Stimulability Assessment
First, results indicate that immediate voice improvement is possible after ‘‘clear speech’’ instructions in not only the sound but also the feel of voice. These results support past literature on stimulability testing, which demonstrated immediate voice change after resonant voice instruction.3 Results also indicate that the use of ‘‘clear speech’’ techniques improves aerodynamic aspects of voice production but did not significantly influence acoustic outcomes. The lack of significant change in acoustic measures is not surprising and is in agreement with the literature showing that many cepstral outcomes do not change, even after intensive voice therapy, or even surgery.14 In most cases, both the SLP and patient heard an improvement in the sound of the voice as a result of the stimulability task, which is an important indicator of success in voice. Despite not focusing specifically on laryngeal musculature (ie, via massage or reposturing), 46% of patients noted an improvement in the feel of voice after the stimulability trial. Overall, patients with MTD and lesions, after atrophy, demonstrated the greatest voice improvement after stimulability assessment. Patients with immobility and hypomobility improved the least. These findings are not surprising, given that many patients with immobility have glottal incompetence and may require surgical augmentation to improve phonation, whereas patients with MTD and lesions are most often treated with voice therapy.9 Second, the study adds to the literature on ‘‘clear speech’’ by demonstrating that some individuals with voice disorders, such as vocally healthy people,6,7 are able to achieve immediate voice improvement with use of the technique. Results are also similar to those observed in the literature following trials of resonant voice or flow phonation.3,4 ‘‘Clear speech’’ may be similar to these techniques in its use of forward focus (oral consonants production), which, such as resonant voice and flow phonation therapy, provides the speaker with kinesthetic feedback external to the gesture, which enhances learning.18– 20 However, unlike these techniques, the ‘‘clear speech’’ instruction is one that can only be given in connected speech; therefore, the clinician is able to assess the patient’s ability to manipulate his/her voice in conversation, which may relate to improved therapy recommendations and outcomes. This correlation will be the focus of future studies. Third, results of stimulability testing—both subjective and objective—provide the SLP with data to present to physician colleagues to support SLP’s recommendations for or against behavioral therapy. These data may assist in patients being appropriately referred for voice therapy, and thus improve voice therapy outcomes.21 Furthermore, the individual data can be presented to patients, which can assist with improved patient understanding of the therapeutic process, and as an extension, improve therapeutic buy-in, potentially removing some of the unknowns. Our past work demonstrated that the presence of an SLP in clinic improves not only therapy attendance but also voice outcomes after treatment.22 This study is the first to standardize stimulability testing and report its objective findings. Finally, this study introduces voice clinicians to a new way to use traditional voice laboratory techniques and equipment—not just as outcomes following intervention, but as stimulability for
5
change testing. This new method of SST in the diagnostic clinic provides further support for voice clinics to use a fully equipped voice laboratory staffed by a voice-specialized SLP. Limitations The major limitation in this study is the potential selection bias of patients chosen for stimulability testing. The examining SLP used her discretion based on experience and clinical acumen to determine which patients would most likely be stimulable for behavioral intervention based on clinical instinct and observation of presenting phonatory habits and techniques. However, despite this limitation, a range of voice disorders was represented in the sample (eg, MTD, lesions, atrophy, hypomobility, immobility), indicating that patients with many voice diagnoses may present with aberrant phonatory techniques that may improve with behavioral intervention, regardless of medical diagnosis, and voice assessment by an SLP should not be limited only to patients with certain voice disorders. A second limitation is that the present study did not follow the participants to track therapeutic success. The goal of the present study was to determine if patients with voice problems are able to make immediate changes in connected speech using the ‘‘clear speech’’ technique and investigate its use as part of a standardized stimulability test. A prospective investigation correlating SST with therapy results will be addressed in future study protocols. CONCLUSIONS In conclusion, this study was the first to operationalize SST performed by an SLP as part of the voice diagnostic clinic for patients with voice problems. It was also the first to provide objective and subjective outcomes of stimulability testing. Future research should explore the predictive value of stimulability testing on voice therapy outcomes. Acknowledgments The authors would like to thank Maurice Goodwin for his assistance. REFERENCES 1. Powell TW, Miccio AW. Stimulability: a useful clinical tool. J Commun Dis. 1996;29:237–253. 2. Tyler AA. Assessing stimulability in toddlers. J Commun Dis. 1996;29: 279–297. 3. Bonilha HS, Dawson AE. Creating a mastery experience during the voice evaluation. J Voice. 2012;26:665.e1–665.e7. 4. Dejonckere PH, Lebacq J. Plasticity of voice quality: a prognostic factor for outcome of voice therapy? J Voice. 2001;15:251–256. 5. Bandura A. Toward a unifying theory of behavioral change. Psychol Rev. 1977;84:191–215. 6. Picheny M, Durlach N, Braida L. Speaking clearly for the hard of hearing I: intelligibility differences between clear and conversational speech. J Speech Hear Res. 1985;28:96–103. 7. Picheny M, Durlach N, Braida L. Speaking clearly for the hard of hearing. II: acoustic characteristics of clear and conversational speech. J Speech Hear Res. 1986;29:434–446. 8. Fairbanks G. Voice and Articulation Drill Book. 2nd ed. New York: Harper and Row; 1960.
6 9. Rosen CA, Gartner-Schmidt J, Hathaway B, et al. A nomenclature paradigm for benign midmembranous vocal fold lesions. Laryngoscope. 2012;122:1335–1341. 10. Baken RJ, Orlikoff RF. Clinical Measurement of Speech & Voice. 2nd ed. San Diego: Singular; 2000. 11. Lowell SY, Colton RH, Kelley RT, Mizia SA. Predictive value and discriminant capacity of cepstral- and spectral-based measures during continuous speech. J Voice. 2013;27:393–400. 12. Awan SN, Roy N, Cohen SM. Exploring the relationship between spectral and cepstral measures of voice and the Voice Handicap Index (VHI). J Voice. 2014;28:430–439. 13. Eadie TL, Doyle PC. Classification of dysphonic voice: acoustic and auditory-perceptual measures. J Voice. 2005;19:1–14. 14. Gillespie AI, Dastolfo C, Magid N, Gartner-Schmidt J. Acoustic analysis of four common voice diagnoses: moving toward disorder-specific assessment. J Voice. 2014;28:582–588. 15. Peterson EA, Roy N, Awan SN, Merrill RM, Banks R, Tanner K. Toward validation of the cepstral spectral index of dysphonia (CSID) as an objective treatment outcomes measure. J Voice. 2013;27:401–410.
Journal of Voice, Vol. -, No. -, 2015 16. Awan SN, Roy N, Jiang JJ. Nonlinear dynamic analysis of disordered voice: the relationship between the correlation dimension (D(2)) and pre-/posttreatment change in perceived dysphonia severity. J Voice. 2009;24:285–293. 17. Awan SN, Roy N. Outcomes measurement in voice disorders: application of an acoustic index of dysphonia severity. J Speech Lang Hear Res. 2009;52:482–499. 18. Verdolini Abbott K. Lessac-Madsen Resonant Voice Therapy. San Diego, CA: Plural Publishing; 2008. 19. Wulf G, Prinz W. Directing attention to movement effects enhances learning: a review. Psychon Bull Rev. 2001;8:648–660. 20. Schmidt RA, Lee TD. Motor Control and Learning. A Behavioral Emphasis. 4th ed. Champaign, IL: Human Kinetics Publisher; 2005. 21. Gartner-Schmidt J. Voice therapy for voice disorders. In: Johnson JT, Rosen CA, eds. Bailey’s Head and Neck Surgery-Otolaryngology. 5th ed. Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins; 2013. 22. Litts J, Gartner-Schmidt J, Clary M, Gillespie AI. Impact of combined laryngologist and speech-language pathologist co-assessment on treatment outcomes and billing revenue. Laryngoscope. 2015; http://dx.doi.org/10. 1002/lary.25349 [Epub ahead of print].