Physiologic Voice Rehabilitation Based on Water Resistance

Therapy With Connected Speech in Subjects With Vocal

Fatigue
*Marco Guzman, †Ilter Denizoglu, *Daniela Fridman, *Constanza Loncon, *Constanza Rivas, ‡Raimundo Garc
ıa,
§

n, §Camilo Quezada, and ‡Leandro Rodriguez, *zxSantiago, Chile, and yIzmir, Turkey
Camilo Mora

Summary: Purpose. The present study aimed to assess the effectiveness of a physiologic voice therapy pro-
gram based on water resistance therapy (WRT) exercises including connected speech in a group of subjects with
voice complaints (vocal effort and fatigue).
Methods. Twenty-four participants with behavioral dysphonia were randomly assigned to one of two treatment
groups: (1) voice treatment with WRT plus vocal hygiene program (n = 12), and (2) vocal hygiene program only
(n = 12). Laryngoscopic assessment was performed in all subjects. Before and after voice therapy, participants
underwent aerodynamic and electroglottographic assessment. The Voice Handicap Index (VHI) and self-assess-
ment of resonant voice were also performed. The treatment included six voice therapy sessions. For the experi-
mental group, the exercises consisted of a sequence of seven phonatory tasks performed with two different voice
training devices (PocketVox and MaskVox). Comparison for all variables was performed between experimental
group and control group.
Results. Significant differences were found for experimental group for VHI physical subscale, and self-perceived
resonant voice when comparing pre-post conditions. A strong negative correlation between self-perceived reso-
nant voice and VHI physical sub-score was also reported. No significant differences were found for instrumented
variables.
Conclusion. Physiologic voice therapy based on WRT exercises including connected speech seems to be an
effective tool to improve self-perceived voice in subjects diagnosed with voice complaints. Apparently, changes
are more prone to occur in perceptual variables related with physical discomfort associate with voice production.
A reduction in phonatory effort and perceptual aspects of vocal fatigue are the main improvements.
Key Words: Water resistance therapy−Semioccluded vocal tract−Vocal fatigue−Vocal effort.

INTRODUCTION production: respiration, phonation, and resonance (vocal

Physiologic voice therapy is conducted on the belief that tract configuration and sensations related to “voice place-
voice disorders are best treated by modifying the underlying ment”) in an integrated or holistic way; (2) improving the
physiology of voice production.1,2 The physiologic “strength,” balance, tone, and stamina of laryngeal muscles;
approach of voice therapy is commonly used by speech-lan- and (3) developing a healthy mucosa covering of the true
guage pathologists when treating patients with a wide vari- vocal folds.2 Examples of physiologic voice therapy pro-
ety of voice disorders. The literature suggests that grams include: Vocal Function Exercises (VFE),1 the
physiologic methods of voice therapy are supported by a Accent Method of Voice Therapy (AM),4 and Resonant
larger body of evidence than other approaches to voice ther- Voice Therapy.5 Each program approaches the voice condi-
apy.3 This involves more studies and a higher level of empir- tion in a holistic or integrated manner aiming at altering the
ical data. Stemple et al2 suggest that the physiologic overall physiology of voice production. This implies not
approach involves three key components: (1) improving the treating breathing function in an isolated way, separate
“balance” among the main subsystems involved in voice from phonation and resonance functions.
Water resistance therapy (WRT) is considered as a physi-
ologic approach of voice therapy and training.6 It includes
Funding: This research was supported by grants from CONICYT (Grant FONDE-
CYT 11180291) phonation of a sustained vowel sound and some other pho-
Conflict of interests: The second author is the founder of the DoctorVox Method natory tasks (eg, ascending/descending glissandos, loudness
and inventor/patent owner of the devices. He received no specific funding for this
work. The rest of the authors declared no potential conflicts of interest with respect to changes, pitch and loudness accents, song melody, speech
the research, authorship, and/or publication of this article. prosody) into a tube with the distal end submerged in water.
From the *Department of Communication Sciences and Disorders, Universidad de
los Andes, Chile, Santiago, Chile; yEgeSante Vocology Center, Izmir, Turkey; One approach to WRT has been described by Simberg.6
zDepartment of Otolaryngology, Clínica Universidad de los Andes, Santiago, Chile; According to the author, the therapeutic process consists of
and the xDepartment of Communication Sciences and Disorders, Universidad de
Chile, Santiago, Chile. several steps occurring during sessions throughout a several
Address correspondence and reprint requests to Marco Guzmán, Department of weeks period. At the beginning of WRT, the patient uses a
Communication Sciences and Disorders, Universidad de los Andes, Chile, Avenida
Monse~

nor Alvaro del Portillo 12455, Santiago, Chile. limited pitch range for the first week(s) of training.6 Gradu-
E-mail: [email protected] ally, the patient starts to engage in more varied intonation
Journal of Voice, Vol. 37, No. 2, pp. 300.e1−300.e10
0892-1997 productions such as glides and simple intervals in a glis-
© 2020 The Voice Foundation. Published by Elsevier Inc. All rights reserved. sando mode.6 Traditionally, two main versions of WRT are
https://doi.org/10.1016/j.jvoice.2020.12.022
Marco Guzman, et al Physiologic Voice Rehabilitation Based on Water Resistance Therapy With Connecte 300.e2

used: (1) Phonation into a traditional Finnish resonance glass water resistance ventilation mask and the traditional water
tube (24-28 cm in length with an 8-9 mm inner diameter) that resistance exercise seem to be effective vocal warm-up exer-
is submerged in a recipient filled with water7; and (2) The Lax cises for musical theater students. To the best of our knowl-
Vox technique, which involves phonation into a flexible sili- edge, to date, there are no studies exploring the possible
cone tube (35 cm in length with an inner diameter of 9-12 therapeutic effectiveness (long-term) of a mask with WRT
mm) that is submerged into a bottle filled with water.8 In in subjects diagnosed with voice complaints. The present
both versions of WRT, the patient is asked to feel vibratory study was designed to assess the effectiveness of a physio-
sensations and to execute an easy phonation (voice produc- logic voice therapy program based on WRT exercises
tion with no effort). Even though resonance glass tube and including connected speech in a group of subjects with voice
Lax Vox are the most common ways to perform WRT, pho- complaints (vocal effort and fatigue). Based on the fact that
nation with an everyday drinking straw submerged into the the present program focuses on the same principles as other
water has been also reported as an effective tool for perform- physiologic programs using SOVTEs, we expected a posi-
ing WRT in subjects with voice complaints.9 tive impact in voice of subjects on objective and subjective
WRT belongs to a broad group of voice exercises called vocal features. Since SOVTEs have been suggested to
semi occluded vocal tract exercises (SOVTEs). SOVTEs improve phonatory efficiency (better conversion from aero-
with different types of tubes (eg, LaxVox, glass resonance dynamic to acoustic energy) and vocal economy (more
tube, drinking straw) with the free end either in air or sub- acoustic output without a proportional increment of vocal
merged in water (WRT) have some limitations. They only folds impact stress), we hypothesized that the present pro-
allow performing a single-phoneme task such as [u:]. Con- gram would be able to promote a higher self-perception of
nected speech or singing is not possible. To overcome this resonant voice production and lower handicap in voice
limitation, a semi-occluded ventilation mask (a mask usu- related aspects. Also, lower airflow rate, lower PTP values,
ally used for cardiopulmonary resuscitation) was first pro- and a slightly increased electroglottographic contact quo-
posed by Borragan et al10 in 1999. The semi-occluded tient were also expected after voice therapy.
ventilation mask (SOVM), which is considered a type of
SOVTE, creates a more distal occlusion allowing connected
speech and singing during voice therapy and training. To METHODS
date, four studies have explored the effect and effectiveness Participants
of the SOVM as a training and therapeutic tool. In a Canine Inclusion criteria for all participants were (1) age within 18-
larynx model, Mills et al11 reported that SOVM leads to the 50 years range (2) laryngoscopic diagnosis of behavioral
same decrease in phonation threshold flow and phonation dysphonia with the absence of organic lesions or other tissue
threshold pressure (PTP) that have been previously observed changes (3) history of voice problems for at least one year,
during tube phonation. Fantini et al12 found significant and (4) no current or previous voice therapy. Also, all sub-
improvements in acoustic and subjective self-assessed meas- jects reported a sensation of muscle tension, vocal effort,
ures after SOVM exercises in a group of singers. Frisancho and vocal fatigue from the clinical history performed by
et al13 suggested that immediate positive effect could be pro- three clinicians (co-authors of the present study). This study
duced by connected speech phonatory tasks using the SOVM was reviewed and approved by the Institutional Review
in both dysphonic subjects and subjects with normal voice. Board of Universidad de los Andes. Informed consent was
Authors concluded that SOVM exercises with connected obtained from all participants. Forty-two participants were
speech seem to promote an easy voice production and a more initially enrolled for this study, they were recruited from the
efficient phonation (more efficient conversion from aerody- general population by social network advertising. Eighteen
namic to acoustic energy). In a recent study designed to subjects were excluded because their laryngoscopic diagno-
examine the effects of the SOVM in normophonic and dys- sis did not meet inclusion criteria. Twenty-four subjects (5
phonic participants, it was reported that beneficial changes in male, 19 female) laryngoscopically diagnosed with func-
both aerodynamic and acoustic variables may be observed in tional dysphonia (non-organic dysphonia) were randomly
both groups after using SOVM.14 assigned (block randomization) to one of two treatment
Devices for voice training and therapy based on a combi- groups before starting voice therapy procedures: (1) voice
nation of a mask with WRT exercises are currently being treatment with physiologic voice therapy plus vocal hygiene
commercialized in the market. These devices (eg, Vocal Feel program (n = 12; experimental group), and (2) vocal
and MaskVox) create a more distal occlusion (similar to the hygiene program only (n = 12; control group). Mean age in
above described SOVM) allowing connected speech and the experimental group was 28 years, range 20-42. Mean
singing while doing water bubbling. There is only one pub- age in the control group was 26 years, range 21-40. All sub-
lished study exploring the immediate effects of a SOVM jects completed the whole treatment program.
combined with WRT exercises (semi-occluded water resis-
tance ventilation mask) on objective (voice range, multi-
parametric voice quality indices) and subjective (auditory- Laryngoscopic assessment
perceptual, self-report) vocal outcomes in subject with nor- Before voice therapy, all participants underwent laryngo-
mal voices.15 Authors concluded that both the innovative scopic, aerodynamic, electroglottographic, and acoustic
300.e3 Journal of Voice, Vol. 37, No. 2, 2023

assessment. They also provided a self-assessment of their samples were included in the EGG and aerodynamic analy-
voice. They were firstly asked to undergo rigid videostro- sis. Once the stable sections were selected, the following var-
boscopy (Digital tele-endoscope Olympus WA96100A; iables were obtained: (1) Mean EGG CQ (%) from EGG
Olympus, Center Valley, PA) to confirm laryngosopic signal. Criterion level of 25% from the peak-to-peak ampli-
diagnosis. Laryngoscopic examinations were performed tude of the EGG signal was used for CQ analysis (from the
by two experienced ENT physicians who are co-authors sustained vowel [a:] task). (2) Psub (cm H2O) estimated
in the present study. Topical anesthesia was used during from the maximum peak of the oral pressure during the
endoscopic procedure. occlusion of the consonant [p:] in the syllable [pa:]. (3) PTP
(cm H2O) from the aerodynamic signal (from the repetition
of the syllable [pa:] at the softest possible voice without
Aerodynamic and electroglottographic assessment reaching whisper). (4) Mean glottal airflow (L/seg) from the
Aerodynamic and electroglottographic (EGG) signals were repetition of the syllable [pa:] at comfortable loudness.
captured simultaneously during all phonatory tasks. Aerody-
namic data were collected with a Phonatory Aerodynamic
Self-assessment of voice quality
System (PAS; KayPentax, model 4500, KayPENTAX, Lin-
Before aerodynamic, electroglottographic, and acoustic
coln Park, NJ). EGG data was obtained with an electroglot-
recordings, all participants were required to self-assess their
tograph (KayPentax, model 6103, KayPENTAX, Lincoln
voice. The perceptual assessment was performed on a
Park, NJ). Both aerodynamic and EGG systems were con-
100 mm visual analogue scale. Only one perceptual variable
nected to an interface (Computerized Speech Lab, Model
was assessed (resonant voice quality) defined as a voice that
4500, KayPENTAX, Lincoln Park, NJ), which in turn was
feels easy and with the sensation of vibration on the front
connected to a desktop computer running a Real-Time aero
part of face and mouth (0 = not resonant at all, 100 = very
dynamic and EGG analysis software (KayPENTAX, Model
resonant). Furthermore, all participants were asked to com-
6600, version 3.4, KayPENTAX, Lincoln Park, NJ). All sam-
plete the validated Spanish adaptation of the Voice Handi-
ples were digitally recorded at a sampling rate of 22.1 KHz
cap Index (VHI-30).16,17 VHI-30 is a self-administrated
with 16 bits/sample quantization. Calibration of the airflow
questionnaire designed to assess the voice handicap result-
rate and pressure was performed before every recording ses-
ing from voice problems. The VHI has strong psychometric
sion according to the manufacturer’s instructions.
measures in terms of reliability and validity.18 It contains 30
Participants from both groups were asked to engage in
items chosen to address the functional, physical, and emo-
the same assessment phonatory task before and after treat-
tional impact of voice problems. Each item is individually
ment: repetition of the syllable [pa:] (speaking voice qual-
scored on a 5-point Likert scale anchored by "never" (score
ity). Repetition of the syllable [pa:] was used to estimate the
of 0) and "always" (score of 4).16
subglottic pressure (Psub) from the oral pressure during the
occlusion of the consonant [p:]. A silicon tube inserted into
the mouth was used to acquire oral pressure. Participants Voice therapy procedures
were asked not to touch the tube with the tongue or any Voice therapy procedures were based on DoctorVox Voice
other oral structure in order to prevent a blockage of the air- Therapy Technique (DVT) described in detail in previous
flow. To avoid air leakage through the nose, a nose clip was publications.19,20 DVT is a technique designed to directly
used for all participants during data acquisition. Three repe- modify the vocal mechanism. It is a physiologic approach
titions of the phonatory task were performed by each sub- combining the three main subsystems involved in voice pro-
ject. F0 was required to be the same during pre and post- duction (phonation, resonance, and breathing). It is a multi-
assessments. PTP was also obtained. Participants were dimensional, multi-level treatment strategy, and uses an
asked to produce the same phonatory task that they per- integrative approach.19
formed to measure Psub estimated from oral pressure. They The treatment period included six voice therapy sessions
were required to produce a sequence of six syllables [pa:] at within 3 weeks, with a frequency of two sessions per week.
the softest possible voice without reaching whisper. All pho- Each session lasted 30 minutes. Therapy sessions were
natory tasks were first performed by researchers for demon- administrated by three trained clinicians. To standardize
stration purposes, and a brief practice was conducted before therapeutic performance, all clinicians underwent a 5-hour
obtaining voice recordings that best represented target pro- training period (conducted by the first author and the sec-
ductions. For PTP, a longer practice was performed. For ond author of the present study) before performing the ther-
both mean Psub and PTP, monitorization for a return to apy. One of the authors is a phoniatrician and experienced
zero during all objective measures was performed to ensure user of DVT. This training period included aspects related
validity. to: (1) sensory-motor learning principles applied to voice
All samples were analyzed with Real-Time aerodynamic rehabilitation (eg, attentional focus, amount of practice,
and EGG analysis software. Criterion level of 25% from the practice distribution, practice variability, practice schedule,
peak to peak amplitude of the EGG signal was used for target complexity, and feedback type),21 (2) use of SOVTE,
electroglottographic contact quotient (CQEGG) analysis. and (3) specific aspects related to the application of DVT
Only the most stable sections from the middle part of the method.
Marco Guzman, et al Physiologic Voice Rehabilitation Based on Water Resistance Therapy With Connecte 300.e4

FIGURE 2. Scheme of a subject performing WRT with con-

FIGURE 1. Scheme of a subject performing WRT with vowel- nected speech phonatory tasks using the MaskVox device mounted
like phonatory tasks using the PocketVox device.
to the PocketVox.

All therapy sessions for the experimental group included included: (1) sustained vowels, (2) ascending and descending
three sections: (1) introduction (3 minutes), during which the glissandos throughout a comfortable vocal range, (3) inten-
clinician asked about the home practice and any voice issues sity and pitch accents, (4) counting numbers, (5) text read-
that might have emerged during the previous week, (2) core ing, (6) talking, (7) singing the song “happy birthday”, and
(24 minutes), during which the participants engaged in exer- (8) speech prosody. Water depth ranged from 2 to 4 cm of
cises that they had practiced during the previous week and H2O. These phonatory tasks, water depths and SOVTEs
also rehearsed new phonatory tasks planned for the session, were sequentially included in the treatment period during
and (3) end of practice (3 minutes), during which the clinician the six sessions as presented in Table 1.
instructed the home practice that the patient had to perform A specific number of trials for each phonatory task or
daily until the next therapy session. The first therapy session exercise was not required in sessions. Participants were
also included instructions about vocal hygiene habits (hydra- asked to perform exercises and phonatory tasks until they
tion, avoidance of high loudness speech, and avoidance of reached an appropriate execution (ie, finding the primal
laryngeal irritants) for both groups. sound, having vibratory sensations and feeling the ease of
As for the experimental group, the therapy program con- phonation). Appropriate execution was controlled by
sisted of a sequence including two devices used as SOVTE: experimenters and self-controlled by participants. Before
(1) PocketVox only (Figure 1), and 2) MaskVox mounted to and during practice, the clinicians provided individual dem-
the PocketVox (Figure 2). The silicone tube part (pocket- onstrations and verbal descriptions of each phonatory task.
VOX) has been devised for a practical use with 500 cc water For a home exercise program, the subjects took the devi-
bottles. It has a phonation channel and an inhalation chan- ces home and were required to complete, 6-8 times daily
nel. The phonation channel used in voice therapy exercises and during 5-10 minutes each time, the same exercises they
is 28 cm length, its inner diameter is 9 mm and outer diame- practiced during each session. At the end of each session,
ter is 12 mm. The oral mask (maskVOX) has been devised the subjects were given, on a paper sheet, detailed instruc-
for free articulation. The oral mask has a sufficient inner tions for their home exercise program. The instructions
space for lip and jaw movements. The edges of the mask included all phonatory tasks learned during the session. To
that contacts face have two flaps. The outer flap is thin and monitor patient compliance, a WhatsApp (WhatsApp Inc.
wider in order to allow sealing during mouth opening. The Menlo Park, CA) message was sent daily to each partici-
inner flap is shorter and thicker for support. The outlet of pant. However, data on compliance across participants
the tube has been devised to fit the tube firmly. Both devices were not gathered in the present study.
can be used together and separately when needed. The first therapy session also included instructions about
Seven phonatory tasks were also included to be per- vocal hygiene (hydration, avoidance of high loudness
formed with the two different devices. Phonatory tasks speech, and avoidance of laryngeal irritants) for both
300.e5 Journal of Voice, Vol. 37, No. 2, 2023

TABLE 1.
Voice Devices and Phonatory Tasks Included Throughout the Six Voice Therapy Sessions
Session Number Exercises Phonatory Tasks
1 PocketVox (4 cm) Exploration producing voiceless bubbles.
PocketVox (4 cm) Primal sound (Shwa) with sustained vowel at comfortable pitch
2 PocketVox (4 cm) Sustained vowel /u/ at comfortable pitch
MaskVox (2 cm) Ascending and descending glissandos
Sustained vowel /u/ at comfortable pitch
Sustained vowel /a/ at comfortable pitch
Sustained vowel /i/ at comfortable pitch
Ascending and descending glissandos
3 PocketVox (4 cm) Ascending and descending glissandos
MaskVox (2 cm) Pitch and loudness accents
Ascending and descending glissandos
Pitch and loudness accents
Counting from 1 to 10
4 MaskVox (3 cm) Ascending and descending glissandos
Pitch and loudness accents
Counting from 1 to 10
Connected speech (talking)
Text reading
5 MaskVox (4 cm) Ascending and descending glissandos
Pitch and loudness accents
Connected speech (talking)
Sing Happy Birthday
Text reading
6 MaskVox (4 cm) Ascending and descending glissandos
Pitch and loudness accents
Connected speech (talking)
Sing Happy Birthday
Text reading

groups. Although the content of the hygiene program was Statistical analysis
targeted to each individual in both control and experimental Data were statistically analyzed and plotted with R (R Core
groups, general recommendations included (1) Hydration Team, 2019). Effects for all variables of interest were
habits: sufficient intake of water (2 L of water approxi- inspected by means of mixed-factor 2 £ 2 ANOVAs con-
mately), increasing water intake with perspiration, decreas- ducted on each one of the selected dependent variables:
ing dehydrating beverages, decreasing consumption of CQEGG, Psub, PTP, Glottal airflow, Resonant Voice Qual-
nonessential diuretics; (2) Control of exogenous inflamma- ity, and VHI (Total, Physical, Emotional, and Functional).
tion: behavioral LPR precautions, reducing or quitting Group (Experimental/Control) and Measure (Pre/Post)
smoking, and avoidance of chemical exposures; (3) Control were the between-factor and the within-factor respectively
of high vocal folds impact stress: control of background in each test. Values for subjective measures were acquired
noise, avoidance of loud speech during therapy sessions, for all participants both in Pre-measure and Post-measure
avoidance of uncontrolled yelling and screaming. Vocal in both groups. However, data for objective variables were
hygiene recommendations were provided and explained in not complete, since four participants could not attend their
detail for both groups just once (at the first session). This corresponding Post assessments. At the time of conducting
session was the only treatment session for the control group. the experiment, widespread long-lasting social riots made
this unfeasible. Because of the small sample size no reliable
missing data imputation could be conducted to estimate
Post-therapy assessment missing values. Thus, ANOVAs were implemented on sam-
Once the six-session voice therapy period was completed, all ples with different sizes. Between-group initial equivalence
participants in both groups underwent the same assessment was inspected, with all dependent variables complying
procedure they underwent for the pre-therapy assessment. except for Psub (P = 0.004, with a barely complying value
The procedure included aerodynamics, EGG, and self- for PTP: P = 0.056). Follow-up analyses reported for VHI
assessment of voice. Post testing was performed 1 week after scores were implemented by conducting paired Wilcoxon
completion of voice therapy. tests (comparing Pre and Post measures) and implementing
Marco Guzman, et al Physiologic Voice Rehabilitation Based on Water Resistance Therapy With Connecte 300.e6

Benjamini-Hochberg’s P values correction considering 8 group (V = 77, P = 0.025, 95% CI 7, 13) and a nonsignifi-
contrasts, so as to control for family-wise error. 95% confi- cant difference for Control Group. A Measure main effect
dent intervals for the pseudomedian are also provided for a was also observed for VHI Total: F(1,22) = 9.34, P= 0.005,
more robust interpretation of effects. Because of the pres- GES = 0.06), with no Pre/Post statistical differences in
ence of a within-group independent variable, Generalized either group.
Eta Squared is provided as an effect size measure. Finally, Figure 5 shows the results for self-perceived Resonant
Spearman rank-correlation tests between Resonance Voice Voice Quality. A significant Measure effect main effect was
Quality and VHI measures were conducted. No P values observed for Resonant Voice Quality (F(1,22) = 23.17, P<
are provided for these tests, discussions being based on cor- 0.001, GES = 0.26). Pre/Post differences in each group were
relation coefficients as indicators of underlying effect sizes. inspected by means of Wicolxon paired-sample tests. Results
were significant both for the experimental group (V = 4.5,
P = 0.007, 95% CI 50.50, 13) and Control group (V = 10,
RESULTS P = 0.04, 95% CI 36, 0.49). However, the effect is stronger
Instrumented variables for experimental group, as expressed by P values.
Figure 3 shows the results for aerodynamic and EGG varia- Finally, the correlation between Self-perceived Resonant
bles. Only significant Group main effects were observed: Voice Quality and VHI measures was inspected by means of
Psub (F(1,18) = 15.31, P= 0.001, GES = 0.36)) and PTP (F Spearman rank correlation tests, considering all values for all
(1,18) = 6.15, P= 0.02, GES = 0.19). Since no significant subjects (ie, Pre and Post values, n = 48). Coefficients showed
Pre/Post differences were observed in either group, these negative correlations with all VHI variables: 0.49 for Func-
main effects reveal both non-manipulated between-group tional, 0.50 for Emotional, 0.74 for Physical, and 0.67
baseline differences and no differential effects of treatment for Total score. Thus, coefficients reveal a strong negative
for the experimental condition. Possible consequences are association between Self-perceived Resonant Voice Quality
addressed in the discussion section below. and Physical subscale and Total score (around 0.7) and a
milder one for the other two VHI measures (around 0.5).

Perceptual variables
Table 2 shows descriptive statistic for perceptual variables. DISCUSSION
Figure 4 shows the results for VHI subscales and total score. This randomized controlled trial assessed the effectiveness
A significant interaction was found for VHI Physical: F of a physiologic voice therapy program based on WRT
(1,22) = 9.77, P= 0.004, GES = 0.07. This interaction was exercises including connected speech, observing a group of
driven by a significant Pre/Post difference for Experimental subjects with voice complaints (vocal effort and vocal

FIGURE 3. Mean plots for objective variables. Error bars represent 95% CI.
300.e7 Journal of Voice, Vol. 37, No. 2, 2023

FIGURE 4. Mean plots for VHI total and sub-scores. Error bars represent 95% CI.

TABLE 2.
Descriptive Statistics for Perceptual Variables
Variable Group Measure Mean SD
VHI Total Control Pre 38.08 19.21
Post 32.83 19.68
Experimental Pre 41.50 13.96
Post 28.25 15.90
VHI Physical Control Pre 18.25 8.07
Post 16.33 8.62
Experimental Pre 18.92 5.82
Post 9.08 6.58
VHI Functional Control Pre 11.83 6.45
Post 10.33 6.83
Experimental Pre 13.00 5.20
Post 12.92 5.98
VHI Emotional Control Pre 8.00 6.54
Post 6.17 5.77
Experimental Pre 9.58 6.75
Post 6.25 6.38
Resonant Voice Quality (mm) Control Pre 42.08 21.96
Post 59.17 23.64
Experimental Pre 38.92 23.65
Post 71.75 17.04

fatigue). Results seem to support the role of the present ther- voice physical complaints after a 3-week treatment period
apy protocol in voice treatment as a potentially effective (six sessions in total). In addition, a significant increment
treatment for subjects with functional voice disorders. Data was observed for self-assessed resonant voice. No significant
showed significant improvements for the self-assessed out- changes were revealed for instrumented variables. Specific
come physical subscale of VHI, revealing that patients from hypotheses about instrumented measures were not sup-
the experimental group reported a significant decrease in ported by our data in the present study.
Marco Guzman, et al Physiologic Voice Rehabilitation Based on Water Resistance Therapy With Connecte 300.e8

FIGURE 5. Mean plots for self-perceived resonant voice. Error bars represent 95% CI.

Instrumental variables “I feel as though I have to strain to produce voice,” “my

Since instrumented variables were, on average, within nor- voice sounds creaky and dry,” my voice “gives out” on me
mal range before voice therapy for both groups, it seems in the middle of speaking”). A significant reduction in the
natural that values remained mostly unaltered or just physical subscale of VHI for the experimental group was
slightly changed after voice therapy, as was the case in our reported when comparing pre-post conditions in the present
study (no significant pre-post differences). Recall that the study. No significant differences were found for the other
main complaint of the recruited participants was the sensa- two subscales (functional and emotional). This could be
tion of vocal fatigue, vocal effort, and vocal tract discomfort interpreted that the main improvement caused by WRT was
in general. As recently defined by Hunter et al,22 vocal effort the reduction of physical discomfort related to voice pro-
is the perceived exertion of a vocalist’s response (vocal duction. A previous study measuring the effectiveness of
demand response) to a perceived communication scenario WRT also reported a significant reduction not only in total
(vocal demand). Previous definitions of vocal effort have VHI but also in the physical subscale.9 Interestingly, corre-
emphasized that it is a perceptual phenomenon (not a physi- lation analysis from our data showed a strong negative cor-
ological phenomenon) experienced by the speaker and not relation between self-perceived voice and VHI physical
the listener.23,24 score. This association has also been previously reported.9
Reduction of VHI physical score and increase on the self-
perceived resonant voice in our findings could have an asso-
Perceptual variables ciation not only with the perceived vocal effort but also
The vocal effort is usually described by subjects as a physi- with the perceptual part of vocal fatigue. A universally
cal perception of exertion and work associated with voice accepted definition of vocal fatigue is currently lacking.
production, it is by definition measured via self-report.20 However, the most current definition of this term was pro-
Results from VHI total, VHI physical subscale, and self-per- vided by Hunter et al.22 According to these authors, vocal
ceived resonant voice therapy could reflect an important fatigue is a quantifiable decline in function (performance or
reduction of vocal effort among participants from experi- perceptual) that influences vocal task performance and is
mental group in the present study. According to Verdolini individual-specific. Previous studies underscore the percep-
et al,25 a self-perceived resonant voice includes not only tual subjective aspects of vocal fatigue such as a feeling of
vibratory sensations on the front part of face and mouth, localized tiredness and a weak voice after a period of vocal
but also the sensation of an easy (effortless) phonation. As use that involves a set of symptoms during or after phona-
matter of fact, since a pressed voice production is also capa- tion.26 These symptoms are detected by self-perception of
ble to produce vibratory sensations (with high perceived effort from prolonged vocal use and improve with vocal
phonatory effort), the ease of phonation possibly constitutes rest.26−28 Because vocal fatigue can occur despite a normal-
the most relevant aspect of resonant voice in voice rehabili- appearing larynx and a normal-sounding voice, the voice
tation. VHI includes three subscales, being one of them the user’s report of increased effort with continued voice use
physical subscale which is related to the degree of physical and alleviation of symptoms after resting is commonly the
vocal discomfort (eg, “I use a great deal of effort to speak,” only element to diagnose vocal fatigue.28 In the present
300.e9 Journal of Voice, Vol. 37, No. 2, 2023

study, laryngoscopic assessment previous to voice therapy variables (decrease in sensations associated with phonatory
showed an absence of organic lesions or other tissue changes effort), it seems likely that relaxing sensations attributed to
for all participants. the water bubbling positively contribute to the increased
The positive treatment effects on perceptual variables self-perceived resonant voice (including easy of phonation)
observed in this research are consistent with results previ- and the decrease in VHI physical score. Water bubbling
ously reported by Guzman et al9 who used a comparable constitutes a good source of sensory stimulation for subjects
research design to assess the efficacy of WRT and tube pho- with voice complaints, especially for those reporting sensa-
nation with the free end in the air. The authors reported sig- tions of vocal effort and vocal fatigue. Paes et al43 studied
nificant improvements for both groups when pre- and post- the immediate effects of WRT on teachers with voice com-
voice therapy conditions were compared, for the total score plaints. Significantly greater phonatory comfort after the
of VHI (decrease) and self-perception of resonant voice exercises were reported.
(increase). Similarly, in a recent study designed to assess the WRT could be a useful tool for voice rehabilitation not
effectiveness of a physiologic voice therapy program based only because of the presence of water bubbling but also
on different SOVTE in subjects with behavioral dysphonia, because of the increased airflow resistance linked to the
results revealed significant lowering after therapy for VHI, degree of tube submersion into water.44 The resistance
Voice Symptom Severity Scale (VoiSS), and Vocal Tract opposed by semi-occlusions creates an increased Poral,
Discomfort Scale (VTDS). Data also showed a significant which in turn, causes a compensatory increment of subglot-
improvement (increase) in the visual analogue scale assess- tic pressure (Psub) in order to maintain phonation and over-
ing the self-perceived resonant voice after eight weeks of come resistance by the water.44,45 Maxfield et al46 ranked 13
voice therapy.29 semi-occluded postures according to intraoral pressure lev-
An important number of previous voice therapy studies els. Semi-occlusion showing the highest level of Poral was
have also reported positive effects on VHI scores after phys- phonation into a tube submerged in water (compared to
iologic programs for voice rehabilitation.30−36 Two studies phonation into a tube in the air and other SOVTE).46 When
conducted with elderly subjects obtained positive outcomes the depth of immersion gradually increases throughout the
in VHI after voice therapy with VFE.32, 33 VFE also has therapy sessions (as occurred in the present study using
been shown to be an effective therapeutical tool for teachers MaskVox) while keeping an easy voice production (effort-
with voice complaints31 and subjects diagnosed with behav- less) and feeling the massage-like sensation, WRT could
ioral dysphonia.34 In both studies, the total VHI score constitute an appropriate tool for voice training. Both the
decreased. Also, studies have reported improvement in VHI proper sensory stimulation and the increasing resistance
scores after treatment with resonant voice therapy.30,36 Sim- possibly promote a change in the voice production pattern
ilar results from Resonant Voice Therapy were shown by contributing to an increased vocal economy47 and a reduc-
Chen et al30 in a group of female teachers with voice disor- tion in vocal fatigue.
ders. Also, in a study designed to compare the effectiveness
of VFE and a rehabilitation program based on phonation
into a thin straw (stirring straw), findings showed a signifi- CONCLUSION
cant reduction of the VHI total score after treatment for Physiologic voice therapy based on WRT exercises includ-
both groups.35 ing connected speech seems to be an effective tool to
improve voice in subjects diagnosed with voice complaints.
Apparently, changes are more prone to occur in perceptual
Water bubbling and massage-like sensation variables related to the physical discomfort associated with
The present voice treatment protocol was based on WRT. voice production. A reduction in phonatory effort and per-
Specifically, the therapy program consisted of a sequence of ceptual aspects of vocal fatigue are the main improvements.
several phonatory tasks including two devices used as
SOVTE: (1) PocketVox and (2) MaskVox. Water bubbling
produced during WRT voice exercises has been linked to a SUPPLEMENTARY DATA
massage-like effect due to the oscillation in oral pressure Supplementary data related to this article can be found
(Poral) caused by bubbles during phonation.36−40 Patients online at doi:10.1016/j.jvoice.2020.12.022.
usually report that water bubbling positively impacts their
voice production because of the relaxing effect in both
REFERENCES
laryngeal and pharyngeal areas.39,41 A massage-like effect 1. Stemple JC. Voice Therapy: Clinical Studies. San Diego, CA: Singular
with the reduction of muscle hypertension could be desir- Publishing Group; 2000.
able in patients with voice complaints, especially in subjects 2. Stemple J, Graze L, Klaben B. Clinical Voice Pathology: Theory and
reporting vocal fatigue and vocal effort. Previous data have Management. San Diego, CA: Singular Publishing Group; 2000. (Ter-
cera ed.).
shown improvement in muscle relaxation sensation in sub-
3. Thomas L, Stemple J. Voice therapy: does science support the art?
jects diagnosed with functional dysphonia immediately after Commun Disord Rev. 2007;1:49–77.
voice exercises with WRT and after 1 week of home prac- 4. Kotby N. The Accent Method of Voice Therapy. San Diego, CA: Sin-
tice.42 Since our main findings were observed on subjective gular Publishing; 1995.
Marco Guzman, et al Physiologic Voice Rehabilitation Based on Water Resistance Therapy With Connecte 300.e10

5. Verdolini K. Resonant voice therapy. In: Verdolini K, ed. National 28. Solomon NP. Vocal fatigue and its relation to vocal hyperfunction. Int
Center for Voice and Speech’s Guide to Vocology. Iowa City, IA: J Speech Lang Pathol. 2008;10:254–266.
National Center for Voice and Speech; 1998. 29. Guzman M, Bertucci T, Pacheco C, Leiva F, Quintana F, Ansaldi R,
6. Simberg S, Laine A. The resonance tube method in voice therapy: Quezada C, Mu~ noz D. Effectiveness of a physiologic voice therapy
description and practical Implementations. Logoped Phoniatr Vocol. program based on different semioccluded vocal tract exercises in sub-
2007;32:165–170. jects with behavioral dysphonia: a randomized controlled trial. Journal
7. Laukkanen A-M. About the so called ‘resonance tubes’ used in Finnish of Communication Disorders. 2020;87:106023. https://doi.org/10.1016/j.
voice training practice. Scand J Logoped Phoniatr. 1992;17:151–161. jcomdis.2020.106023.
8. Sihvo M. Terve € a€
aani, €a€anen hoidon A B C [Healthy voice. The A B C 30. Chen SH, Hsiao TY, Hsiao LC, et al. Outcome of resonant voice ther-
for voice care. Helsinki: Kirjapaja; 2006. [In Finnish]. apy for female teachers with voice disorders: perceptual, physiological,
9. Guzman M, Jara R, Olavarria C, et al. Efficacy of water resistance acoustic, aerodynamic, and functional measurements. J Voice. 2007;
therapy in subjects diagnosed with behavioral dysphonia: a random- 21:415–425.
ized-controlled trial. J Voice.. 2017;31. 385.e1-385.e10. 31. Roy N, Gray S, Simon M, et al. An evaluation of the effects of two
10. Borrogan A, del Barrio JA, Gutierrez JN. El juego Vocal Para Prevenir treatment approaches for teachers with voice disorders: a prospective
Problemas De La Voz. Archidona, Malaga: Aljibe; 1999. randomized clinical trial. J Speech Lang Hear Res. 2001;44:286–296.
11. Mills R, Hays C, Al-Ramahi J, et al. Validation and evaluation of the 32. Sauder C, Roy N, Tanner K, et al. Vocal function exercises for presby-
effects of semi-occluded face mask straw phonation therapy methods laryngis: a multidimensional assessment of treatment outcomes. Ann
on aerodynamic parameters in comparison to traditional methods. J Otol Rhinol Laryngol. 2010;119:460–467.
Voice. 2017;31:323–328. 33. Kaneko M, Hirano S, Tateya I, et al. Multidimensional analysis on the
12. Fantini M, Succo G, Crosetti E, et al. Voice quality after a semioc- effect of vocal function exercises on aged vocal fold atrophy. J Voice.
cluded vocal tract exercise with a ventilation mask in contemporary 2015;29:638–644.
commercial singers: acoustic analysis and self-assessments. J Voice.. 34. Pedrosa V, Pontes A, Pontes P, et al. The effectiveness of the compre-
2017;31:336–341. hensive voice rehabilitation program compared with the vocal function
13. Frisancho K, Salfate L, Lizana K, et al. Immediate effects of the semi- exercises method in behavioral dysphonia: a randomized clinical trial.
occluded ventilation mask on subjects diagnosed with functional dys- J Voice. 2016;30. 377, e11-e19.
phonia and subjects with normal voices. J Voice. 2020;34:398–409. 35. Kapsner-Smith MR, Hunter EJ, Kirkham K, et al. A randomized con-
14. Awan SN, Gartner-Schmidt JL, Timmons LK, et al. Effects of a vari- trolled trial of two semi-occluded vocal tract voice therapy protocols. J
ably occluded face mask on the aerodynamic and acoustic characteris- Speech Lang Hear Res. 2015;58:535–549.
tics of connected speech in patients with and without voice disorders. J 36. Enflo L, Sundberg J, Romedahl C, et al. Effects on vocal fold collision
Voice. 2019;33. 809.e1-809.e10. and phonation threshold pressure of resonance tube phonation with
15. Meerschman I, Van Lierde K, Redman Y, et al. Immediate effects of a tube end in water. J Speech Lang Hear Res. 2013;56:1530–1538.
semi-occluded water resistance ventilation mask on objective and sub- 37. Andrade P, Wood G, Ratcliffe P, et al. Electroglottographic study of
jective vocal outcomes in musical theater students. J Speech Lang seven semi-occluded exercises: LaxVox, straw, lip-trill, tongue-trill,
Hear Res. 2020;63:661–673. humming, hand-over-mouth, and tongue-trill combined with hand-
16. Jacobson B, Johnson A, Grywalski C, et al. The Voice Handicap Index over- mouth. J Voice. 2014;28:589–595.
(VHI): development and validation. Am J Speech Lang Pathol. 1997; 38. Wistbacka G, Sundberg J, Simberg S. Vertical laryngeal position and
6:66–70. oral pressure variations during resonance tube phonation in water and
17. N
~ez-Batalla F, Corte-Santos P, Se~
un naris-González B, et al. Adapta- in air. A pilot study. Logoped Phoniatr Vocol. 2016;41:117–123.
tion and validation to the Spanish of the Voice Handicap Index (VHI- 39. Guzman M, Castro C, Madrid S, et al. Air pressure and contact quo-
30) and its shortened version (VHI-10). Acta Otorrinolaringol Esp. tient measures during different semioccluded postures in subjects with
2007;58:386–392. different voice conditions. J Voice. 2016;30. 759, e1-e10.
18. Mathieson L, Hirani SP, Epstein R, et al. Laryngeal manual therapy: a 40. Radolf V, Laukkanen A-M, Horácek J, et al. Air-pressure, vocal fold
preliminary study to examine its treatment effects in the management vibration and acoustic characteristics of phonation during vocal
of muscle tension dysphonia. J Voice. 2009;23:353–366. exercising. Part 1: measurement in vivo. Eng Mech. 2014;21:53–59.
19. Denizoglu II, Sahin M, Bayrak S, et al. Efficacy of doctorvox voice ther- 41. Guzman M, Castro C, Acevedo K, Moran C, Espinoza V, Quezada C.
apy technique for mutational falsetto. J Voice. 2019;33. 950.e1-950.e8. How do tube diameter and vocal tract configuration affect oral pressure
20. Denizoglu I. LaxVox therapy used in a case of muscle tension dyspho- characteristics caused by bubbling during water resistance therapy?
nia hiding and underlying glottal incompetence. In: Stemple JC, Journal of Voice. In press. https://doi.org/10.1016/j.jvoice.2020.03.004.
Hapner ER, eds. Voice Therapy Clinical Studies. 5th ed San Diego, 42. Guzman M, Acu~ na G, Pacheco F, et al. The impact of double source of
CA: Plural Publishing; 2019:201–212. vibration semioccluded voice exercises on objective and subjective out-
21. Titze I, Verdolini K. Vocology: The Science and Practice of Voice comes in subjects with voice complaints. J Voice. 2018;32. 770.e1-770.e9.
Habilitation. Salt Lake City, UT: National Center for Voice and 43. Paes SM, Zambon F, Yamasaki R, et al. Immediate effects of the
Speech; 2012. Finnish resonance tube method on behavioral dysphonia. J Voice.
22. Hunter EJ, Cantor-Cutiva LC, van Leer E, et al. Toward a consensus 2013;27:717–722.
description of vocal effort, vocal load, vocal loading, and vocal fatigue. 44. Amarante Andrade P, Wistbacka G, Larsson H, et al. The flow and
J Speech Lang Hear Res. 2020;63:509–532. pressure relationships in different tubes commonly used for semi-
23. Abbiss CR, Peiffer JJ, Meeusen R, et al. Role of ratings of perceived occluded vocal tract exercises. J Voice. 2016;30:36–41.
exertion during self-paced exercise: what are we actually measuring? 45. Radolf V, Laukkanen AM, Hor.cek J, et al. Air-pressure, vocal fold
Sports Med. 2015;45:1235–1243. vibration and acoustic characteristics of phonation during vocal
24. Pageaux B. Perception of effort in exercise science: definition, measure- exercising. 1. Measurement in vivo. Eng Mech. 2014;21:53–59.
ment and perspectives. Eur J Sport Sci. 2016;16:885–894. 46. Maxfield L, Titze I, Hunter E, et al. Intraoral pressures produced by
25. Verdolini K, D ruker D, P almer P, et al. Laryngeal adduction in reso- thirteen semi-occluded vocal tract gestures. Logoped Phoniatr Vocol.
nant voice. J Voice. 1998;12:315–327. 2015;40:86–92.
26. Nanjundeswaran C, Jacobson BH, Gartner-Schmidt J, et al. Vocal Fatigue 47. Guzman M, Calvache C, Ugalde N, Pacheco F, Ortiz V, Del lago J,
Index (VFI): development and validation. J Voice. 2015;29:433–440. Bobadilla M. A Voice Rehabilitation Protocol with the Semioccluded
27. Paolillo NP, Pantaleo G. Development and validation of the Voice Ventilation Mask in Subjects with Symptoms of Vocal Fatigue and
Fatigue Handicap Questionnaire (VFHQ): clinical, psychometric, and Phonatory Effort. Journal of Voice. In press. https://doi.org/10.1016/j.
psychosocial facets. J Voice. 2015;29:91–100. jvoice.2020.10.011.