Evidence of acclimatization in persons with severe-to

Washington University School of Medicine
Digital Commons@Becker
Publications
Division of Adult Audiology
2003
Evidence of acclimatization in persons with severeto-profound hearing loss
Francis K. Kuk
Widex Office of Research
Lisa G. Potts
Washington University School of Medicine in St. Louis
Lidia Lee
Northern Illinois University
Michael Valente
Washington University School of Medicine in St. Louis
Jay Picirrillo
Washington University School of Medicine in St. Louis
Follow this and additional works at: http://digitalcommons.wustl.edu/audio_hapubs
Recommended Citation
Kuk, Francis K.; Potts, Lisa G.; Lee, Lidia; Valente, Michael; and Picirrillo, Jay, "Evidence of acclimatization in persons with severe-toprofound hearing loss" (2003). Publications. Paper 2.
http://digitalcommons.wustl.edu/audio_hapubs/2
This Article is brought to you for free and open access by the Division of Adult Audiology at Digital Commons@Becker. It has been accepted for
inclusion in Publications by an authorized administrator of Digital Commons@Becker. For more information, please contact engeszer@wustl.edu.
Evidence of Acclimatization in Persons with
Severe-to-Profound Hearing Loss
Francis K. Kuk *
Lisa Potts **
Michael Valente **
Lidia Lee ***
Jay Picirrillo **
Abstract
The present study examined the phenomenon of acclimatization in persons
with a severe-to-profound hearing loss. A secondary purpose was to examine
the efficacy of a digital nonlinear power hearing aid that has a low compression threshold with expansion for this population. Twenty experienced hearing
aid users wore the study hearing aids for three months and their performance
with the study hearing aids was evaluated at the initial fitting, one month, and
three months after the initial fitting. Performance of their current hearing aids
was also evaluated at the initial fitting. Speech recognition testing was conducted at input levels of 50 dB SPL and 65 dB SPL in quiet, and 75 dB SPL
in noise at a + 10 SNR. Questionnaires were used to measure subjective performance at each evaluation interval. The results showed improvement in
speech recognition score at the one-month evaluation over the initial evaluation. No significant improvement was seen at the three-month evaluation from
the one-month visit. In addition, subjective and objective performance of the
study hearing aids was significantly better than the participants' own hearing
aids at all evaluation intervals. These results provided evidence of acclimatization in persons with a severe-to-profound hearing loss and reinforced the
precaution that any trial of amplification, especially from linear to nonlinear
mode, should consider this phenomenon.
Key Words: acclimatization, digital power hearing aid, low compression threshold.
Abbreviations: CT = compression threshold; HLC = high level compression,
APHAB = Abbreviated Profile of Hearing Aid Benefit; EC = ease of communication; RV = reverberation; BN = background noise; AV = aversiveness of
sounds; SPIN = Speech Perception in Noise Test; FDA = Food and Drug
Administration; VC = volume control; HHIE-S = Hearing Handicap Inventory
for the Elderly - Screening; ANOVA = Analysis of Variance; PTA = pure-tone
average; OSPL90 = Output sound pressure level at 90 dB SPL input; WDRC
= wide dynamic range compression
Sumario:
EI presente estudio examina el efecto de la climatizaci6n en personas con
hipoacusia severas a profundas. Un prop6sito secundario fue examinar la eficacia de un auxiliar auditivo potente, digital, no lineal, con un umbral de
compresi6n bajo, pero con expansi6n para esta poblaci6n. Veinte usuarios
con experiencia con audifonos utilizaron el auxiliar auditivo del estudio por
tres meses y su rendimiento con el audffono del estudio se ~valu6 en la
adaptaci6n inicial, al mes y a los tres meses posteriores a la adaptaci6n. EI
rendimiento de su auxiliar auditivo original fue evaluado tam bien al inicio. Se
condujeron pruebas de reconocimiento dellenguaje a niveles de 50 dB SPL
• Widex Office of Research in Clinical Amplification, Lisle. IL
•• Washington University School of Medicine, Department of Otolaryngology - Head and Neck Surgery, SI. Louis, MO
••• Northern Illinois University, Department of Communications Disorders, Dekalb, IL (now at Eastern Michigan University)
Reprint address: Francis Kuk, Ph.D., Director of Audiology, Widex Office of Research in Clinical Amplification, 2300 Cabot Drive, Suite 415, Lisle, IL 60532
84
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
in Linear Hearing
Aid Users/Kuk et at
ya 65 dB SPL, en silencio, y a 75 dB SPL en ruido, con un SNR de + 10. Se utilizaron
cuestionarios
para medir el rendimiento subjetivo en cad a intervalo de evaluacion.
Los
resultados
uacion
mostraron
del primer
significativa
Ademas,
una mejorfa en los puntajes de reconocimiento
mes, comparados
en la evaluacion
el rendimiento
con los iniciales.
a los tres meses,
subjetivo
comparada
y objetivo del audffono
dellenguaje
No se observo
en la eval-
ninguna
mejoria
con la visita del primer
mes.
del estudio fue significativamente
mejor que el del propio audifono de los participantes, en todos los intervalos de evaluacion.
Estos resultados mcfstraron evidencia de la climatizacion en personas con hipoacusias severas a profundas,
amplificacion,
Palabras
bajo.
y reafirmo
la precaucion
que debe tenerse
en cualquier
sobre todo del modo lineal al no lineal, debiendo considerarse
Clave:
Abreviaturas:
Perfil Abreviado
Climatizacion,
auxiliar
auditivo
CT = umbral de compresion;
del Beneficio
potente
digital,
HLC = compresion
del Auxiliar Auditivo;
EL = facilidad
umbral
estudio
de
este fenomeno.
de compresion
de alto nivel; APHAB
de audicion;
=
RV = rever-
beracion; BN = ruido de fonda; AV = repulsion al sonido; SPIN = Prueba de Percepcion
del Lenguaje en Ruido; HHIE-S = Inventario de Impedimentos
Auditivos del Anciano Tamizaje; ANOVA = Analisis de Variancia; PTA = Promedio Tonal Puro; OSPL90 = Nivel
de presion sonora de salida a 90 dB SPL de entrada; WCRC = Com presion de rango
dinamico amplio.
ditOry
acclimatization is the systematic
change in auditory performance with
~
ime that is not linked to a change in the
acoustic information available to the listener. It
involves improvement in performance that cannot be attributed purely to task, procedural, or
training effects (Arlinger et aI., 1996, p.87S). It
has become a popular topic since the late 1980s.
Palmer et al (1998)provided an excellent review
of the literature on acclimatization. The same
author (Palmer, 1999) also provided another
update on the most recent findings in the area.
Aside from the potential physiological
underpinnings of acclimatization (Willot, 1996),
the observation of this phenomenon has many
ramifications. From a research standpoint, it
casts doubts on the validity of studies that did
not allow time or did not allow adequate time
for acclimatization before an evaluation of hearing aid performance was made. It reinforces, as
indicated by the FDA in its guidelines for hearing aid validation studies (FDA, 1994),that sufficient time must be allotted for acclimatization
prior to actual evaluation.
The phenomenon of acclimatization has far
greater clinical implication, especially in the
counseling provided to patients relative to
expectations from their new hearing aids. For
one thing, patients would need to be counseled
that (1) their initial experience with the new
hearing aids may not represent their final performance with the hearing aids; and (2) it may
take time to fully realize the benefits.
Furthermore, it suggests that any validation
studies should not be performed at the time of
the initial fit. Rather, validation should be done
after the patients have had enough time to
acclimatize to the hearing aids.
Despite its potential implications, evidence
on acclimatization since Silverman and
Silman's (1990) clinical report on two cases is,
at best, mixed (Cox and Alexander, 1992;
Gatehouse, 1992; Saunders and Cienkowski,
1997; 'furner et aI., 1996). In 1995, a group of
clinical researchers met in Eriksholm and concluded that the evidence on acclimatization was
confounded by the experimental conditions
employed in each study. Issues of audibility,
optimal test conditions (stimulus types, test levels and SNR), allowance for gain change over
time, and the "right" subject population could
have affected the outcome of the evaluation.
The last possibility on the "right" subject
population to demonstrate acclimatization is
especially of interest. Previously, many of the
studies on acclimatization employed participants with relatively mild-to-moderate degrees
of hearing loss. These participants, because of
their relatively "good"hearing, may have never
been deprived of the acoustic signal to demonstrate acclimatization. Consequently, the key to
demonstrating such a phenomenon, as summarized by Palmer (1999), is to ensure that the
participants "...have enough hearing loss to
produce deprivation and yet not so much hearing loss that amplification cannot restore audibility ..." Furthermore, the "...outcome measure would have to consist of the same frequencies that were believed to be deprived."
Obviously, there can be many different
designs to examine this hypothesis (Palmer,
1999). One possible paradigm to examine the
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
85
Journal
of the American Academy of AudiologyNolume 14, Number 2, 2003
acclimatization phenomenon is to examine the
change in speech recognition scores at low input
levels in participants with a severe-to-profound
hearing loss (PTAgreater than 70 dB HL). This
group of participants are typically experienced
hearing aid wearers using linear (either peak
clipping or compression limiting) hearing aids
with a volume control (VC).In theory, while the
amount of gain provided by linear hearing aids
may be appropriate for speech presented at a
conversational level, inputs at a lower level may
not be audible unless the wearers constantly
adjust the VC on the hearing aids or leave the
VC at a higher than desirable setting. In some
situations, one can assume that even with this
adjustment, there may not be adequate
gain/output. That is to say, audibility for soft
sounds is not as assured as conversational level
sounds for people with a severe-to-profound
hearing loss when conventional linear amplification is used. There is a higher likelihood of
auditory deprivation in this group of individuals for low-intensity sounds. If one can ensure
audibility (as much as one can) or provide more
audibility than conventional linear hearing aids
to this group of individuals without risking discomfort, one may be able to observe acclimatization more consistently. Unfortunately, commercial hearing aids that meet this requirement were unavailable.
Recently, a digital nonlinear power hearing
aid that uses a low compression threshold was
introduced. This allows extra gain for low-input
sounds without risking discomfort at higher
input levels. This hearing aid would appear
ideal for use to test the hypothesis of acclimatization. Because the amplification rationale (i.e.,
nonlinear) used by this hearing aid is significantly different from the conventional linear
approach available for persons with a severe-toprofound hearing loss, it may be valuable to
compare the efficacy of the study hearing aids
to the participants' own linear hearing aids. Thus,
the primary purpose of the present study was to
examine the presence (or lack thereof) of acclimatization in persons with a severe-to-profound
hearing loss. A secondary objective of this study
was to compare the efficacyof the study hearing
aids with the participants' own hearing aids.
METHOD
Subjects
'I\venty adult hearing aid wearers participated in the study. 'I\velve of the participants
were recruited from the Washington University
School of Medicine and eight were recruited
from the Northern Illinois University. The age of
the participants ranged from 43 years to 92
years with a mean of 55 years (S.D. = 17.3 yrs).
All but one participant had a symmetrical (within 5 dB) severe-to-profound sensorineural hearing loss. Figure 1 shows the mean headphone
thresholds of the participants. Note that in
instances where thresholds were unobtainable
within the limits of the audiometer, the maximum dial reading of the audiometer at that frequency was taken instead to compute the average hearing loss. In other words, the "true" average thresholds for the higher frequencies were
poorer than those indicated in the figure.
Frequency (Hz)
250
lD
~
0
«
20
inz
...
500
1000
2000
4000
8000
Q)
..J
J: 40
CO
~
::l
o
60
..J
OJ
r:: 80
.;:
III
Q)
J:
r:: 100
III
Q)
::E
120
Figure 1. Mean audiometric thresholds measured under
headphones (N=20). Maximum dial readings were used
to calculate average threshold when "no response" was
noted.
All the participants were experienced binaural hearing aid wearers and had been wearing
hearing aids for an average of 20 years (S.D. =
13.6 yrs). Their current hearing aids included a
variety of power hearing aids from various
manufacturers with a typical output saturated
sound pressure level (OSPL90)exceeding 135 dB
SPL. The average age of their current hearing
aids was 4.5 years (S.D. = 3.9 yrs). Participants'
hearing aids were disPlinsed at their respective
sites and were judged (via electroacoustic measurement) to be functioning according to specifications at the time ofthe evaluation. Their hearing aids were mostly linear hearing aids with
either peak clipping or compression limiting as
the method of output limiting.
86
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
in Linear Hearing Aid Users/Kuk et al
Hearing aids
~
130
The Widex Senso P38 hearing aid was used
in this study. This is a digital 3-channel hearing
aid utilizing adaptive, slow-acting compression
as the processing algorithm. It has a low compression threshold and high-level compression
(i.e.,two compressionthresholds beforesaturation).
As will be elaborated in the following paragraphs, this unique input-output characteristic
made it ideal to study the phenomenon of
acclimatization in hearing-impaired subjects.
Ringdahl et al (2000)provided a detailed description of the characteristics of this hearing aid.
The study hearing aid has a maximum insitu gain exceeding 70 dB and a peak OSPL90
of 139 dB SPL. It uses a low compression
threshold (CT) at 20 dB HL in all three channels. This translates roughly to 32 dB SPL in
the low-frequency and 26 dB SPL in the midand high-frequency channels. The advantage of
the low CT is to provide greater gain to low
input signals that are below the CT (Kuk,
1999). Expansion is used below the CT to minimize the perception of microphone noise and
low-level ambient noise. Uniform gain reduction (i.e., compression) starts above the 20 dB
HL and continues until a conversational level
(second compression threshold - about 53 dB
HL in the low-frequency channel, 44 dB HL in
the mid-frequency channel, and 39 dB HL in
the high-frequency channel) is reached beyond
which less gain reduction occurs (high level
compression). The advantage ofthe second segment of compression is to minimize temporal
and saturation distortion occurring at high
input levels. Gain for conversational input was
estimated from the NAL-RP fitting formula
(Byrne et aI., 1990) modified for the device specific features (e.g., channels, release times etc.)
of the study hearing aid (Kuk and Ludvigsen,
1999). "Noise reduction" algorithm was not
available on this study hearing aid. Figure 2
compares the input - output curve of a linear
hearing aid and the P38 hearing aid matched at
a conversational input level. Note the extra output at low input levels and the reduced output
above a conversational level provided by the
study hearing aid over a conventional linear
hearing aid. Because most hearing-impaired
participants wear conventional linear hearing
aids, one may hypothesize that these individuals are deprived oflow-input sounds. By providing them with extra gain for low-input sounds,
one may be able to observe the phenomenon of
acclimatization more easily.
.~e
....
~
.....................
.... ~
!\,,,?>
S
~ 110
...J
:I:
In
~ 90
-.e-
Threshold
:J
:J
70
o
50
1000 Hz
o
20
40
60
80
Input (dB HL)
100
120
Figure 2. Comparison in input-output curves between
the Senso P38 hearing aid and a linear hearing aid
matched at a conversational input (about 45 dB HL). A
hearing loss of 90 dB HL is assumed.
The manufacturer's fitting software was
used to fit the study hearing aids. After the
audiogram information was entered into the
NOAH audiogram module, the fitting software
directly took the audiometric thresholds at specific frequencies as gain entry for each frequency channel. Specifically,the threshold at 500 Hz
was used to compute gain for the low-frequency
channel; the average threshold at 1000 Hz and
2000 Hz was used for the mid-frequency channel and the threshold at 4000 Hz was used for
the high-frequency channel. The filter slope and
the cross-over frequency between filters were
automatically adjusted based on the magnitude
of the entered thresholds.
Followingthis, an in-situ feedback test was
performed on the study hearing aids with the
participants wearing the proper earmolds.
During the feedback test, gain in each channel
was gradually but automatically increased until
acoustic feedback was detected at the hearing
aid microphone. The maximum gain ofthe hearing aids was then leveled at 6 dB below the feedback point. This corresponded to the desired gain
at an input of20 dB HL. Because the main objective of this study was to examine acclimatization
at low input levels, efforts were made to ensure
that the desired gain was available for all channels. However, such was not always achievable
especially in the high frequency because of earmold leakage etc. Consequently, a compromised
criterion of "-10" (i.e., the actual gain available
cannot be 10 dB less than that required to yield
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
87
Journal of the American Academy of AudiologyNolume 14, Number 2, 2003
an aided threshold at around 20-30 dB HL) was
adopted as the criterion for all but the high-frequency channels. Participants' earmolds were
remade if the feedback values did not meet this
criterion. It was recognized that even with this
criterion one might not ensure full audibility.But
it would have provided more audibility for low
input sounds than what the participants
received from their linear hearing aids.
The input-output curve in Figure 2 suggests the possibility that some participants may
initially react negatively to the study hearing
aids. Specifically,the additional gain for the low
inputs may lead to perception of "noisiness"
while the reduced output for the high inputs
may lead to perception of "inadequate loudness", "reduced clarity" etc. With time, such perception may disappear and become more
acceptable. On the other hand, it is also possible that the same perception may reflect less
than optimal gain for the participants. To
ensure that the study clinicians did not unnecessarily adjust the study hearing aids and
reduce the magnitude of the acclimatization
effect, the following instructions were written
as a reminder. They were:
potential frustrations in their daily use, all participants were given the study hearing aids
with the VC option. This allowed gain adjustment of +/- 6 dB relative to the prescribed setting. Ifparticipants, upon their return, reported
constant use of the VC, then the settings on
their hearing aids were fine-tuned in order to
meet their preference. This was done before any
data collection. It was recognized that this
action might alter the amount of gain that participants received during speech recognition
testing. However, it would be impractical to
force the participants to wear hearing aids that
provided unacceptable sound quality even after
one month of wear. The settings on the hearing
aids at all test sessions were recorded for subsequent verification of gain. As will be shown in
the Results section, the majority of participants
required less than 3 dB gain adjustment.
•
Aided sound-field thresholds
•
•
•
•
•
If conversational speech was too loud, the
gain setting was lowered in 2.5 dB steps in
all channels
If conversational speech was too soft, the
gain setting was raised in 2.5 dB steps in
all channels
If conversational speech was clear and
comfortable, but "things are not quite as
loud", the recommended settings were
maintained
If conversational speech was clear and
comfortable, but participants reported
"hearing many soft sounds like sniffling,
birds chirping etc", the recommended
settings were also maintained
If conversational speech was loud, but
speech was not clear, the mid channel gain
was increased by 5 dB and the low channel
HTL was decreased by 2.5 dB
If conversational speech was acceptable,
but the occlusion effect was reported, the
low-frequency channel gain for high input
was lowered in 2-dB steps. To compensate
for the loss of loudness, the mid-frequency
channel gain was increased by the same
amount.
In order to evaluate how the study hearing
aids perform in real-life and to minimize any
Outcome measures
Both objectiveand subjectiveoutcomemeasures were used to evaluate participant performance with their own hearing aids and with the
study hearing aids over time. These included:
Aided thresholds were obtained with the
participants' own hearing aids and the study
hearing aids worn binaurally. Aided sound-field
thresholds represented the lowest input level
that was audible to a listener at a fixed VC set
for target gain. Warble tones were presented in
an ascending manner to elicit thresholds.
Participants sat approximately one meter from
a loudspeaker that was placed directly in front
ofthem. No VC adjustment was allowed during
any sound-field threshold testing. Participants
wore their own hearing aids at the VC setting
that they typically used in their daily lives.
Speech recognition test
Sentences from the Speech Perception In
Noise (SPIN) test were used to evaluate speech
recognition. The noise stimulus was a Widex
party noise recorded at the Old Copenhagen
Stock Exchange (Valente et al., 1999). This
noise was chosen instead of the babble noise
that accompanied the SPIN test because of its
closer approximation to real-life noise. The
speech materials were presented at 0 azimuth,
first with the participants' own hearing aids,
and then with the study hearing aids at each of
88
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
the three intervals of evaluation. The speech
materials were presented at 50 dB SPL in
quiet, 65 dB SPL in quiet and 75 dB SPL in a 65
dB SPL party noise background. Both speech
and noise were presented from the same loudspeaker placed at one meter directly in front of
the participants. Order of teilting (stimulus levels and word list) was counterbalanced across
participants and visits. A practice list was used
to familiarize participants to the task.
Abbreviated Profile of Hearing Aid
Benefit (APHAB)
The APHAB questionnaire, a benefit
measure consisting of 24 items belonging to
four categories - Ease of Communication (EC),
Reverberation (RV), Background Noise (BN),
and Aversiveness of Sounds (AV), was used
(Cox and Alexander, 1995). Participants completed the unaided and aided portion of the
APHAB questionnaire on their hearing aids at
the beginning of the study. The APHAB was
not completed on the study hearing aids at initial fit because of participants' lack of experience with the study hearing aids. Instead, participants completed the unaided and aided portions of the APHAB after they have worn the
study hearing aids for one month and three
months. Because it was realized later that it
may be difficult for the participants to estimate
their difficulty in the unaided condition on
account oftheir degree of hearing loss, only the
aided scores were compared in this study.
Washington University Questionnaire
(WUQ)
This relative preference questionnaire was
administered at one month and three months
after the initial fitting ofthe study hearing aids.
The questionnaire asked for relative preference
between the study hearing aids and the participants' own hearing aids in a variety of daily
communication situations. Participants chose
either their "own aids," the "study aids," "both"
or "neither aids" as the response.
Knowles MarkeTrak Questionnaire
The KnowlesMarkeTrak questionnaire has
been used to evaluate consumer satisfaction
(Kochkin, 1996) for hearing aids for several
years. Through the use of a rating scale (mostly
from 1 to 5, with "I" being very satisfied and "5"
being very dissatisfied), hearing aid wearers
in Linear Hearing Aid Users/Kuk et al
indicated their overall satisfaction with the recommended hearing aids, as well as on issues
pertaining to their effectiveness in different listening environments, their cost, cosmeticissues,
and professionalism of the dispensers. Because
of the nature of this study, not all the items on
the questionnaire were included. Instead, only
the followingitems were included:
•
Overall satisfaction with hearing aids
•
Improvement in quality oflife with hearing
aids
•
Satisfaction with "Hearing Aid Features"
•
Satisfactionin specific"ListeningSituations"
Participants completed the selected items
at the beginning ofthe study on their own hearing aids. They completed the same items at one
month and three months after fitting.
Hearing Handicap
Inventory
Elderly - Screening (HHIE-S)
for
the
This was a lO-item questionnaire (Ventry
and Weinstein, 1982) that addressed the emotional and social difficulty of a hearingimpaired person. Participants answered with a
"Yes"(1) "No"(2) or "Sometimes" (3) response to
the 10 items. An example of items may be: "Do
you have difficulty hearing when someone
speaks in a whisper?" " Do you feel handicapped
by a hearing problem?" This questionnaire was
used at the beginning of the study using participants' own hearing aids as the reference.
Participants completed the same questionnaire
at one- and three- months post-fitting.
Procedure
Participants were seen for a minimum of 4
sessions. During the first session, the functional status of the participants' own hearing aids
was evaluated via electroacoustic measures.
Unaided audiometric thresholds and aided
sound-field thresholds with the participants'
own hearing aids were determined at the wearers' typical VC setting. No attempt was made to
match the output of the participants' own hearing aids to the study hearing aids at any input
levels. Participants' performance with their
own hearing aids on the SPIN test was determined at the three presentation levels (50, 65,
and 75 (+10 SNR) dB SPL) at a fixed VC setting. The practice list was used to familiarize
participants to the task. Order of testing and
SPIN lists were counterbalanced. Participants
also completed the unaided and aided portions
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
89
Journal
of the American Academy of AudiologyNolume 14, Number 2, 2003
of the APHAB questionnaire using their own
hearing aids as the criterion. In addition, they
also completed the MarkeTrak questionnaire
and the HHIE questionnaire.
The appropriateness of the participants'
earmolds for the study hearing aids was also
evaluated during the first session. New earmolds were ordered if their current earmolds
could not maintain the desired gain for soft
sounds within a 10 dB margin in the mid- and
high-frequency channels. Participants did not
wear the study hearing aids until the proper
earmolds were available.
Participants returned for the second session
(between one and three weeks afterwards) at
which time they started wearing the study hearing aids binaurally. The settings on the hearing
aids were optimized based on the criteria
described in an earlier section. The aided soundfield thresholds with the study hearing aids were
determined, along with the SPIN test at the
three test conditions presented in a counterbalanced order.Participants were also instructed on
the proper use and maintenance of the hearing
aids without being disclosedtheir features. They
were also counseled that their perception with
the study hearing aids might be different from
what they were accustomed to, and they were
asked to return with specific comments so that
the settings could be modifiedif necessary.
Participants returned for the third session
after they had worn the study hearing aids for
one month. The settings on the hearing aids
were modified based on the subjective comments whenever necessary. In addition, their
speech recognition ability at the three test conditions was again determined. The APHAB,
MarkeTrak, HHIE-S, and the Washington
University Questionnaire were completed also.
Participants completed the same procedure
during the three-month follow-up.
RESULTS
examine the issue of acclimatization. Figure 3
shows the average aided thresholds between
the participants' own hearing aids and the
study hearing aids at the initial fitting. The
average aided difference between the hearing
aids was 4.3 dB at 2000 Hz to 17.5 dB at 250
Hz. These sound-field thresholds were analyzed
using a three-way split-plot ANOVA with
repeated measures of the two within-subject
factors (hearing aids and frequency) and one
between-subject factor (site). The results suggested a significant main effect for hearing aid
W=345.66; df=2,323; p<. 0001). Post-hoc analyses showed that the average aided thresholds
for both hearing aids were significantly lower
than the unaided thresholds. In addition, the
aided thresholds for the study hearing aids were
lower than the participants' own hearing aids at
all frequencies. There was also a significant frequency effect (F=55.88; df=5,323; p<. 0001).
However, no site effect was found (p > 0.2).
Frequency (Hz)
til
til
250
o
500
1000
2000
8000
--1
20
i
c:
.;:
ctl
CIl~
4000
o
..J
Cl
:I:~
'C...!.
(jj C/) 40
;j::Z
-6<
§ ~
Stud
60
O..J
C/):I:
~Cll
:::l 'C 80
ctl~
c:
en
c:
100
::E
120
ctl
CIl
Figure 3. Mean unaided (U) and aided binaural soundfield thresholds for the participants' own hearing aids (0)
and the study hearing aids (8). The number of participants with various pure-tone averages (PTA)is listed.
Aided sound-field thresholds
A compression hearing aid with a low CT
allows more gain for low input sounds. When
comparing the aided thresholds of such a hearing aid to a linear hearing aid when both are
matched in output to a conversational input,
the aid with a lower CT would yield a lower
aided threshold (Kuk, 1999). Indeed, in this
study, one must meet the requirement of a
lower aided threshold before one can even
Because the experimenters were allowed to
modifY settings on the study hearing aids to
ensure continued ,subject participation, it is
important to examine these settings in order to
ascertain if any change in performance could be
attributed to a change in gain/output or to
acclimatization. The recorded settings on the
study hearing aids for each participant were
programmed retroactively on the study hearing
90
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
aid and its coupler response was determined
with a speech-shaped noise at 50 dB SPL and
80 dB SPL. Of the 20 participants, 10 showed
less than 2 dB overall gain change (6 increase
and 4 decrease) for both the low-and high-input
stimuli. Five participants showed a 3 dB gain
increase for the high-level ihput only and five
showed a 3 dB gain decrease for the low-level
input. Overall, the average gain was 3 - 4 dB
lower in the low-to-midfrequency regions at the
three-month follow-upthan at the initial fitting
for the 50 dB SPL input but 1- 2 dB higher at
the three-month follow-up than at the initial
visit for the 80 dB SPL input. This is not unexpected from listeners who were accustomed to
linear hearing aids. These individuals tended to
want more loudness for sounds above a conversationallevel and less audibility for soft sounds
from their new nonlinear hearing aids. Any
acclimatization effect seen in this study is possibly not a result of higher gain from the study
hearing aids (because it is lower). On the other
hand, Horwitz and Turner (1997) showed that
inexperienced wearers preferred 2 dB less gain
and experienced wearers preferred 2 dB more
gain (for conversational input) at the conclusion
of the acclimatization period.
Speech recognition scores
SPIN scores were analyzed by three-way
split-plot ANOVA with repeated measures of
the two within-subject factors (hearing aids and
test conditions) and one between-subject factor
(site). Because there was not any site effect, the
results were combined for reporting.
in Linear Hearing
LP-Item Score
Figure 4 summarizes the mean scores for
the low probability (LP) items on the SPIN test
for all participants (N=20). There was a significant main effect for hearing aids (F=6.65;
df=3,209; p=.0003). Bonferroni corrected comparisons of least squares means revealed that
the mean performance of the study hearing aids
at one-month and three-months was significantly better than the mean performance of the
study hearing aids at initial fit (p = .0252; P =
.0174 respectively). Mean differences between
the study hearing aids at one- and threemonths were not significant. This supports the
hypothesis of acclimatization between initial fit
and one-month post-fitting. In addition, the
mean performance of the study hearing aids at
one-month and three-months was significantly
better than the mean performance of the participants' own aids (p=.0004 and p=.0002
respectively). However, the mean difference
between the study hearing aids at initial fit and
the participants' own aids was not significant.
There was a significant test condition effect
(F=35.61; df = 2,209; p<.OOOl)suggesting that
performance at the 65 dB SPL test condition
was better than the 50 dB SPL (p < 0.0001) and
the 75 dB SPL (+10 SNR) (p = .0027)conditions.
HP-Item Score
Figure 5 summarizes the mean
the high probability (HP) items on
test. There was a significant main
hearing aids (F=10.57; df=3,209;
~50
Legends:
~
l!!
40J
0
u
I
fD
Z
it
•
Own aid
~
study aid - initial
~
study aid - 1 month
aid - 3 months
IIIlII study
I
30-1
en
...
20~
:lIE
10-
C
IlII
III
•
Own aid
~ 40
~
study aid - initial
1;l
~
study aid - 1 month
aid - 3 months
IIIlII study
Z
it
.
en
30
!i: 20
c
IlII
III
I
1:1.
scores for
the SPIN
effect for
p<.OOOl).
Legends:
~
50-
Aid Users/Kuk et al
:lIE
10
0
50 (quiet)
0
50 (quiet)
65 (quiet)
75 (+10SNR)
Input Level (dB SPL)
65 (quiet)
75 (+10SNR)
Input Level (dB SPL)
Figure 4. Mean low-probability (LPJ SPIN scores for
own aids and study hearing aids at different evaluation
intervals (initial, one month, and three months) and test
conditions (50 dB, 65 dB, and 75 dB SPL (+10 SNR)).
Figure 5. Mean high-probability (HPJ SPIN scores for
own aids and study hearing aids at different evaluation
intervals (initial, one month, and three months) and test
conditions (50 dB, 65 dB, and 75 dB SPL (+10 SNRJJ.
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
91
Journal
of the American Academy of AudiologyNolume 14, Number 2, 2003
Bonferroni corrected comparisons of least
squares means revealed that the mean performance of the study hearing aids at one
month (p = .0248) and three months (p = .0029)
was significantly better than their mean performance at initial fit. However, there was no
significant difference between performance at
one- and three-months post-fitting. This supports the hypothesis of acclimatization between
initial fit and one-month post-fitting. In addition,
the mean performance of the study hearing aids
at initial fit (p=.0406), one-month post-fitting
(p=.0004) and three-months
post-fitting
(p<.OOOl)was significantly better than the mean
performance of the participants' own aids. There
was a significant test condition effect (F=58.50;
df = 2,209; p<.OOOl)suggesting that the performance at the 65 dB SPL condition was better
than other test conditions (p < .0001). The
ANaVA revealed a significant two-factor type by
condition (F=3.75, d=6,209;p=.0014)interaction.
These results showed significantly better
SPIN scores at the one-month and three-month
visits than the initial visit, but not between the
one-month and three-month visits. This sug-
gests that any acclimatization would have stabilized after using the study hearing aids for one
month. In addition, significant differences were
also noted between the study hearing aids and
the participants' aids at all test intervals for the
50 dB SPL and 65 dB SPL test conditions.
APHAB
Figure 6 reports the mean APHAB aided scores
for the participants' own hearing aids and the
study hearing aids at one month and three
months. Separate two-way (aid by site) ANaVAs
were performed for each of the four subscales.
For the background noise subscale (BN), there
was a significant hearing aid main effect
(F=6.54; df=2,36; p =.0038). Bonferroni corrected comparisons revealed that the mean difficulty score for the study hearing aids at three
months was significantly lower than the participants' own aids (p=. 0028). No differences were
observed between the participants' own aids and
the study hearing aids at one month, or between
the study aids at one- and three-months post-fitting. Mean differences reported for the other
three subscales were not significant.
Table 1. Frequency of relative preference on the Washington University Questionnaire.
i Washington
University
Hearing Aid Comparison
Current HA
1 month
3 months
EXDeriment HA
1 month
3 months
Both
1 month
12*
7
5
3
11**
11*
5
5
3
7
11*
7
4
3
10*
4
7
12
2
5
4
3
Pleasant
4
Natural
4
~~.
Comfortably
2
loud
t
"
8
4
Distinct
Neither
1 month
3 months
"
,
'
.
a close friend j·on·j
4
4
7
9
9
with
a stranger
3
3
8
11*
10
5
7
2
10*
8**
7
8
3
to TV with no one
else talking
listening
14
,
,
-,--
'
1
1
with
listening
'
1
15
was better.,.
j-on-j
.~-
3 months
to TV with one
_.------- ---.----.
or
:~
3
7*
.. F·
5
4
7
8
1
4
7*
6
10
6
1
1
3
5
5
7
4
5
1
3
6
6
9
9
3
1
1
2
13*
9*
2
5
2
2
in car
2
3
listening in an elegant restauranl
1
3
2
--.-
3
more people
listening
at
talking
one speaker
listening
at
a meeting with
several speakers
f---.. ~--~.
listening at a family gathering
listening
to the radio in the car
fj~~~~i'!fl __
~~ a passenger
listening
in
a family restau.~ant
L~~~~f1~~JLi'!_~_
~.~use
, listening
in
of worship
a movie theater
13*
11*
4
4
7
9
8
2
1
9'
6
5
8
3
2
11*
8*
3
6'
4
5
2
17**
4
3
3
3
5
6
5
2
4
5
4
5
3
1
-
-.--
1
?
1
9'
10*
4
5
8
5
quiet sounds were more audible
2
1
15*
13*
Overall
4
5
II*"
12*"
4
2
1
TOTAL
31
47
151"
135"
99
95
42
to recorded
music
6
5
1
2
-,-._--
was less frustrating
listening
performance
..
a meeting with
5 - -- ~---_.2
...
..
92
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
3
--
1
..- --.-.-_._---~
45
Evidence of Acclimatization
in Linear Hearing
Aid Users/Kuk et al
50
ell
ou 80
en
D
Own aid
D
study aid - 1 month
•
study aid - 3 months
~60
::::l
~
u
IE
i:5
40
eo
Legends:
•
30
c
aid - 1 month
IIIllll Study
aid - 3 months
I'll
40-
::I
'C
ell
'C
:( 20r:
20
EC
RV
BN
AV
..i:
Ii
10
C1l
ell
:E 0
Own aid
mJ Study
U
o
Social
Emotional
Total
Listening environments
Figure 6. Mean APHAB aided difficulty scores between
the participants' own aids and the study hearing aids at
one- and three-month follow-up.
Figure 7. Mean scores on the HHIE-S questionnaire for
the participants' own aids and the study hearing aids at
one- and three-month intervals.
WU questionnaire
HHIE - S
Table 1 summarizes the number of participants who preferred their own "current hearing
aids," the "experimental hearing aids," "both
hearing aids," and "neither hearing aids" at the
one-month and three-month follow-up evaluations. A McNemar's Chi-square test of significance was performed on each item to determine
if significant differences were present between
related measures. Items showing the study hearing aids to be statistically better than the participants' own hearing aids at the 0.05 level were
identified by one asterisk (*) in the cell under the
"experimental HA" column. Items that were significant at the p < 0.1 level were indicated by two
asterisks (**). Table 1 shows that there is an
overall preference for the study hearing aids.
However,statistical significancewas reached on
only half of the items. Expectedly, items that
were found to be statistically significant related
to enhanced audibility (quiet sounds more audible), sound quality (distinct, pleasant, natural,
music), and moderate noise and reverberation
(car, house of worship, family restaurant). For
most items (exception: comfortably loud, one
speaker, family restaurant), preference for the
study hearing aids at one month post-fitting was
also seen at the three month post-fitting interval.
This suggests no difference in preference for the
study hearing aids between one-month and
three-month post-fitting.
Figure 7 summarizes the mean scores for
the social, emotional, and total subscales for the
participants' own aids and the study hearing
aids at one month and three months on the
HHIE-S questionnaire. In general, fewer difficulties were reported with the study hearing
aids than the participant's own hearing aids,
especially at the three-month evaluation interval (p =0.011).A two-way (aid by site) ANOVA
was performed on the Total HHIE-S score.
There was a significant hearing aid effect
(F=6.83; df=2,36; p=.0031). Bonferroni corrected comparisons revealed that the mean HHIES Total score for the study hearing aids at 3
months was significantly better than participants' own aids (p=. 0023). There was no significant hearing aid by site interaction.
MarkeTrak
Table 2 summarizes the mean satisfaction
ratings on selected items of the MarkeTrak questionnaire for the participants' own hearing aids
and for the study hearing aids at one and three
months post-fitting. Two, 2-way ANOVAwere
performed - one to compare the ratings for the
study hearing aids between one-month and
three-month post-fitting;and another to evaluate
if the differences in ratings between the participants' own hearing aids and the study hearing
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
93
Journal
of the American Academy of AudiologyNolume 14, Number 2, 2003
Table 2. Mean satisfaction ratings on selected items on the MarkeTrak.
- -P38 - one month
P38 - three months
2.45
2.05"
2.15"
1.4
1.4
Own Aid
Items
Overall
--... -
Overall satisfaction
01 life
1.6
III il))>i
Fit/comfort
2.05
2.4"
2.2
Visibility to others
2.25
2.2
2.15
Ease of battery change
1.6
2.1"
2.25"
Battery life
2.25
1.95"
1.85"
2.65
2.4
2.1"
Feedback
3.2
2.65"
2.75"
Volume control ease
2.35
2.55
2.6
Reliability
2.2
1.95
2.05
2.05
~ofsounds
19
1.95
Use in noise
3.1
2.7"
Naturalness
2.4
2.3
Improves hearing
"-_._-
2.25
3.15
2.7"
2.4"
2.3
2.1
Hear soft sounds
3.15
2.35"
2.45"
Comfort loud sounds
2.5
2.3
Own voice
---.
i,i·,,,,,,,,,,,,,,,,,
Small group
I
Outdoors
Large group
Concert
House of worship
-- .
2.5
iii;;;;)
i)))
One person
--
3.15
,
2.2
Localization
i»
I))..
1.6
1.8
1.6
2.7
2.4
2.25"
2.4
2.35
2.55
3.8
3.25"
3.25"
3.68
3.33
3.33
3.4
3.03
2.98"
TV
3.1
2.79"
2.7"
Restaurant
3.15
3
3.15
Car
3
2.55"
2.7"
Telephone
3.12
2.95
3
I
Music
3.25
2.41"
2.3"
!
Leisure
2.5
2.3
2.35
=
--.-"-
--
-~
i»~
A))
--
=
(1 very satisfied and 5 very dissatisfied)
Asterisk (") within cell indicates that the study hearing aids were rated significantly higher than subjects' own hearing aids (P< 0.05)
aids were significant. None of the items showed
any significant difference when the comparison
was made between the study hearing aids at one
month and three months. This suggests no subjective acclimatization occurred between one
month and three months post-fitting.
On the other hand, many items on the
MarkeTrak questionnaire showed significant
difference between the study hearing aids and
the participants' own hearing aids (significant
differences at the p < 0.05 level were indicated
with an asterisk). The significant difference
was seen at both the one-month and threemonth post-fitting. These items included: overall satisfaction, battery life, feedback, localization, hearing soft sounds, large groups, TV, car,
and music. Only four items showed a significant
change at one time and not the other time of
evaluation (fitJcomfort, clarity of sounds, small
group, house of worship). There was no significant difference in rating for the other items.
DISCUSSION
The present study showed an improvement
in SPIN scores when performance was compared between the initial visit and the onemonth follow-up with the study hearing aids.
No improvement in SPIN scores or in subjective
ratings was observed between the one-month
and three-month visits. This supports the
hypothesis of acclimatization on objective tasks,
and further suggests that this phenomenon, if
present in subjective rating tasks, takes about a
month to plateau. In addition, this study also
showed significantly higher SPIN scores for the
study hearing aids than the subjects' own hearing aids at the 50 dB ~PL and 65 dB SPL test
conditions. Results on the subjective questionnaires also supported the advantage of the
study hearing aids in situations where the audibility of low level input sounds and perception
in noise may be important. This supports the
94
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
efficacyof the study hearing aids and suggests
that multichannel nonlinear processing as used
in the study hearing aids may be preferable to
single channel linear processing commonly
used by individuals with a severe-to-profound
degree of hearing loss.
Advantage of a lower CT
A key feature of the study hearing aid that
prompted this study is its low compression
threshold at 20 dB HL in all three channels.
Even though not all the participants in this
study were able to achieve this low CT because
of earmold related feedback issues, the average
participant benefited from this low CT
nonetheless. This is evidenced in the better
aided sound-field thresholds with the study
hearing aids, higher SPIN scores at the 50 dB
SPL input level at all evaluation intervals, and
higher ratings on items in the Washington
University Questionnaire and MarkeTrak questionnaire that related to listening at low input
levels (soft sounds, localization, hearing at a
distance).
The advantages of the low CT may appear
at odds with the report of Dillon et al (1998)
who showed that adult hearing-impaired people preferred a higher compression threshold
(> 65 dB SPL) than a lower one. There are both
hearing aid specific issues and criteria issues
that made comparisons difficult. In addition to
the difference in subject population (moderate
versus severe-to-profound hearing loss), Dillon
et al (1998) used a single channel fast-acting
WDRC hearing aid that employed linear processing below the CT as the main processing
strategy. A problem with linear processing
below the CT is that same gain is applied to
sounds at and below the CT. This gives rise to
the possibility that typical ambient noise
would be amplified to an audible level (Kuk,
1999). The effect would be exacerbated with
the use of short release times because of the
more frequent gain fluctuation. On the other
hand, the study hearing aids employed a long
adaptive release time (from 10 ms to 20 s) with
expansion processing below the compression
threshold. In contrast to compression, an
expansion circuit increases its gain as input
level increases. This type of processing allows
the desirable gain to be achieved at the target
input level (CT) while providing less gain for
input levels below the CT.This has the effect of
minimizing the perception of circuit noise
while maximizing the advantages of a low CT.
in Linear Hearing Aid Users/Kuk et al
These two studies were also different in
their objectives. Dillon et al (1998) examined
wearer preference for compressionthreshold. In
this study, the experimenters recognized potential objections to a low CT by the participants,
but insisted that they try the recommended
amplification for at least one month. Such
preparation may have neutralized any negative
comments on the use of a low CT.Although one
may argue that such practice may have positively biased the subjective findings, that could
not explain the positive findings on the SPIN
test.
Evidence of acclimatization?
The first documented suggestion of
acclimatization with hearing aid use was made
by Watson and Knudsen (1940) who reported
that one of their seventeen subjects showed
improved speech scores over time with hearing
aid use. Throughout the years, many authors
(e.g., Gatehouse, 1992, 1994; Horwitz and
Turner, 1997; Silverman and Silman, 1990)
reported the same observation. On the other
hand, probably an equal number, if not many
more questioned the validity of such observations (e.g., Arlinger et aI., 1996; Holte et aI.,
1997; Neuman et aI., 1997; Saunder and
Cienkowski, 1997). As indicated earlier, such
differences in observation could be related to
the stimulus test conditions, study protocol,
and the choice of the "right" study participants
(Palmer et aI., 1998).
The results of this study supported the
hypothesis of acclimatization in the severe-toprofound hearing loss population. The significant improvement in SPIN scores at the one
month follow-up over the initial evaluation
attests to this conclusion. Indeed, if one examines Figures 4 and 5 (LP and HP item scores
respectively), one can offer the followingexplanation for the observed findings. With the low
CT, more speech information that was previously inaudible (with the linear hearing aids)
became audible. If audibility were the only
determinant of speech recognition, one would
expect an increase in SPIN score (over the participants' own hearing aids) at the initial visit,
but no further increase at subsequent visits.
This was not the case. Figure 4 shows minimal
or no change in LP-item scores between the
participants' aids and the test aids at the initial visit at the 50 dB and 75 dB SPL test conditions. However, a small but significant
increase was noted for the 65 dB SPL condi-
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
95
Journal
of the American Academy of AudiologyNolume 14, Number 2, 2003
tion. This suggests that the additional audibility did not fully improve the intelligibility of
LP items. The same is not true for the HP
items (Figure 5).A difference of9% in HP-item
scores was noted between the study hearing
aids and the participants' own aids at the initial visit for the 50 dB SPL condition and 2%
for the 65 dB SPL condition. A decrease of 2%
was noted at the 75 dB SPL (+10 SNR) condition. The observation that LP item scores were
not improved but HP item scores did initially
may suggest that overall speech was more
audible, but not necessarily those ofthe target
words. Because the identification of the LP
items required identification of the target
words whereas identification of the HP items
may not have such a stringent requirement,
the extra audibility yielded higher scores for
the HP but not the LP items at the 50 dB SPL
test condition. With time and use of the hearing aids, the audible cues that were available
at the initial visit became "usable" by the
wearers. This may explain the increase in item
scores (LP, HP) at the one-month post-fitting
interval over the initial fitting. The non-significant difference in item scores between the
one-month and three-month post-fitting intervals suggests that most of the acclimatization
effect plateaued by about one month.
The phenomenon of acclimatization was
observable in all test conditions, although the
pattern at the 75 dB SPL (+10 SNR) condition
may not be as clear as the 50 dB SPL condition. However, if one examines the word scores
at the different test intervals, one would find
improvement ranging from 2% to 5% between
the initial and three-month evaluation intervals. This improvement, however, was non-significant.
The above-mentioned observations may be
explained by the design of the study hearing
aids. The low CT accounted for the additional
audibility and improved SPIN scores at the initial and one month visits for the low- and medium- presentation levels. At a higher input level
(and in noise), the nonlinear characteristics of
the study hearing aids (high-level compression)
would result in less gain and possibly less distortion than what the participants had previously experienced with their linear hearing
aids. One may hypothesize that the loss ofloudness and possible loss of distortion (temporal as
well as saturation) cues at the initial visit may
have accounted for the poorer initial performance with the study hearing aids. Fortunately,
this loss of information was not detrimental,
but rather beneficial as evidenced by the gradual improvement in SPIN scores over time as
participants acclimatized to the new temporal
cues. It would be of interest to examine if such
an improvement extends beyond the threemonth evaluation. Gatehouse (1992) suggested
that 18 weeks (41/2 months) may be necessary
for acclimatization.
One may question, because of the availability of the VC (+/- 6 dB adjustment) and the
allowancefor clinician gain adjustment at followup visits, that the noted changes may be a function of additional gain desired by the participants and not a better use of existing gain.
Clearly, an alternative study design is to prohibit any gain adjustment in the study hearing aids
during the study period. This was impossible
given our desire to evaluate the real-life efficacy
of the study hearing aids in this study as well.
However, our choice may not alter the present
findings for three reasons. First, Horwitz et al
(1997)found that acclimatization was evidenced
in subjects regardless of gain adjustment.
Secondly,the majority of participants required
less than 2 dB gain change with the average participant desiring 2 dB less gain for low- and
medium-level sounds and 4 dB more gain for
high- level sounds in the mid frequencies. This
amount of gain change is possibly too small to
result in any measurable difference in speech
scores.Furthermore, the direction ofgain change
is opposite to the direction of the observed
improvement in SPIN scores for the 50 dB SPL
and 65 dB SPL test conditions. One may argue
that acclimatization may have been diminished
because of such an adjustment. Lastly,there was
not a significant difference in subjective ratings
between the one-month and three-month followup when gain changes were made.
A limitation of the current study design
that may have prohibited the full observation
of acclimatization is that no subjective data on
the study hearing aids were collected during
the initial visit. Rather, all the questionnaires
were collected one month after wearing the
study hearing aids. Such a decision was made
because subjective data obtained without any
participant experience may be meaningless.
On the other hand, the observation of no subjective acclimatization between one-month
and three-month visits leaves the question on
subjective acclimatization unresolved. Is there
no subjective acclimatization at all, or does
acclimatization
on subjective tasks also
plateau before one month? Future studies on
subjective acclimatization will need to exam-
96
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
100
100
A-50
dB SPL
1
'6
'6
B - 65 dB SPL
1
80
80
iii 60
}40
~20
0
iii 60
IL
Initial
100
i
80
in Linear Hearing Aid Users/Kuk et al
}40
~ 20
0
'1-month
Initial
3-month
'I-month
C - 75 dB SPL
(+10 SNR)
.a
i 60
Legends:
1: 40
IIstudy aid better
D
~
~20
0
3-month
-'
Initial
D•
1-month
~
study aid same
~
study aid worse
3-month
Figure 8. Mean percentage of subjects showing the "Same," "Better," and "Worse"total SPIN scores with the study hearing aids re: their own hearing aids at the three evaluation intervals when the stimuli were presented at (Al 50 dB SPL in
quiet, (B) 65 dB SPL in quiet, and (el 75 dB SPL in noise (+10 SNRl,
ine subjective performance closer to the initial
fit in order to detail the time course of subjective acclimatization.
Implications
There are several clinical implications
from the results of the current study. First, the
phenomenon of acclimatization is real and its
expression is often determined by the test conditions, choice of subjects, evaluation materials,
and the type of hearing aid processing used in
the evaluation, In general, acclimatization is
evidenced when the wearer has been sufficiently deprived of the acoustic stimuli and is given
a chance to utilize them. The choice of participants with a severe-to-profound degree of hearing loss and the use of sentence materials with
target words of different contextual cues may
have optimized this evaluation. The use ofmultichannel nonlinear processing with a low CT
(and expansion) may be more effective than the
use of single channel linear processing to
demonstrate this phenomenon in this population also. Participants with a different degree of
hearing loss or hearing aid experience may
require different evaluation materials or hearing aid processing to demonstrate this phenomenon. Furthermore, because the effect of
acclimatization may be subtle, a relatively large
sample size may be necessary to allow sufficient
statistical power. These factors are important to
consider when studying acclimatization or comparing the results of this study to other studies.
Secondly,the time course of acclimatization
may be differentfor differentevaluation materials.
In this case, it took approximately one month for
the participants to acclimatizeto the extra audibility cue that affected their SPIN scores in quiet.
However,it may take more than three months for
the average wearer to improve SPIN scores in
noisy background.This suggests that acclimatiza-
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
97
Journal
of the American Academy of AudiologyNolume 14, Number 2, 2003
tion may not be a uniform and immediate phenomenon.Rather, it may be a processthat has different rates ofexpressionfor differenttasks.
From the clinical perspective, the results of
this study highlight the importance of postponing validation measures on hearing aids beyond
the initial visit. The authors re-analyzed the
SPIN total scores by counting the percentage of
participants who achieved a higher, the same,
and a lower SPIN score with the study hearing
aids than their own hearing aids at the initial,
one-month, and three-month evaluation intervals. Based on the within-subject test-retest
variability that was observed in a few subjects
and from the use of the same test on other occasions, an arbitrary criterion difference of 5% was
set for significant differences. The results are
summarized in Figure 8 (a, b, and c) for the
three test conditions. For the 50 and 65 dB SPL
conditions (Figures 8a and 8b respectively), only
about 40% of participants showed better performance with the study hearing aids than their
own hearing aids at the initial visit. However, at
the one-month follow-up,as many as 75% ofthe
participants showed more than 5% improvement with the study hearing aids. If one had
made a decision on the appropriateness of the
study hearing aids based simply on the initial
test results, one would have concluded that twothirds of the participants would not do better
with the study hearing aids. In real-life, the
study hearing aids yielded better performance
in three-quarters of the participants when they
were evaluated at a later time. Similarly, only
15% of the participants performed better with
the study hearing aids than their own hearing
aids during the initial visit in the 75 dB SPL
(+ 10 SNR) condition. On the other hand, 35% 40% of them performed better with the study
hearing aids at the one-month and three-month
visits (Fig 8c). It seems necessary to delay efficacy measures to at least one month post-fitting
before one can conclude if the new hearing aids
are better than the participants' previous ones.
Along the same line, the results of the current study cast doubt on the appropriateness to
fine-tune the recommended settings on hearing
aids during initial fittings with the first sign of
patient complaint. While such an action may be
necessary because the initial estimate may not be
optimal, it is also possible that the settings are
optimal, but the p~tients are unaccustomed to
the processed sounds. In the later case, adjustment may compromisethe benefits offeredby the
design ofthe hearing aids. One must understand
whether their patients' complaints are expected
reactions or true complaints that originate from a
sub-optimal fit. Counseling and proper instructions would be necessary if they are expected
reactions. Especially now that digital signal processing is used extensively to achieve different
sound processing, it is important that clinicians
are fully familiar with their hearing aids to avoid
compromises on their fittings.
REFERENCES
American National Standards Institute (1992) Testing
Hearing Aids with a Broad-Band Noise Signal. ANSI
S3.42-1992.
Arlinger S, Gatehouse S, Bentler R, Byrne D, Cox R,
Dirks D, Humes L, Neuman A, Ponton C, Robinson K,
Silman S, Summerfield A, Turner C, Tyler R, Willot J.
(1996). Report of the Eriksholm Workshop on Auditory
Deprivation and Acclimatization. Ear Hear 17:87S-90S.
Byrne D, Parkinson A, Newall P. (1990). Hearing aid gain
and
frequency
response
requirements
for the
severely/profoundly hearing impaired. Ear Hear 11:40-49.
Cox R, Alexander G. (1992) Maturation of hearing aid
benefit: Objective and subjective measurements. Ear
Hear 13:131-141.
Cox R, Alexander G. (1995). The abbreviated profile of
hearing aid benefit. Ear Hear 16(2):176-183.
Dillon H, Storey L, Grant F, Phillips A-M, Skelt L, Mavrias
G, WoytowychW, Walsh M. (1998). Preferred compression
threshold with 2:1 wide dynamic range compression in
everyday environments. Austr J Audiol 20, 33-44.
Food and Drug Administration. (1994). Draft: Hearing
Aids: Proposed Clinical Protocol.
Gatehouse S. (1992). The time course and magnitude of
perceptual acclimatization
to frequency responses:
Evidence from monaural fitting of hearing aids. J Acoust
Soc Amer 92:1258-1268.
Gatehouse S. (1994). Components and determinants
hearing aid benefit. Ear Hear 15:30-49.
of
Horwitz A, Turner C. (1997). The time course of hearing
aid benefit. Ear Hear 18(1):1-11.
Kochkin S. (1996). Customer satisfaction and subjective
benefit with high performance hearing instruments.
Hear Rev. 3(12): 16-26.
Kuk F. (1999). Optimizing compression: the advantages
of a low compression threshold. Hear Rev Suppl (High
Performance Hearing Solutions III: marketing and technology), edited by K. Strom & s. Kochkin. P.44-47.
Kuk F, Ludvigsen C. (1999). Variables affecting the use of
general-purpose prescriptive formulae to fit modern nonlinear hearing aids. J Amer Acad Audiol10(8):458-465.
Palmer C. (1999). Depriv~tion and acclimatization in the
human auditory system: do they happen? Do they matter? Hear J 52(11): 23-24.
Palmer C, Nelson C, Lindley G. (1998). The functionally
and physiologically plastic adult auditory system. J
Acoust Soc Amer 103(4):1705-1721.
98
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
Evidence of Acclimatization
in Linear Hearing Aid Users/Kuk et al
Ringdahl A, Magnusson L, Thelin L, Edberg P, Israelsson
B. (1999). Clinical evaluation of a digital power hearing
aid. Hear Rev 7(3):59-64.
Saunders G, Cienkowski K. (1997). Acclimatization to
hearing aids. Ear Hear 18:129-138.
Silverman C, Silman S. (1990). Apparent auditory deprivation from monaural amplificaiion and recovery with
binaural amplification: Two case studies. J Amer Acad
Audiol 1: 175-180.
Turner C, Humes L, Bentler R, Cox R. (1996). A review of
past research on changes in hearing aid benefit over
time. Ear Hear 17:14S-25S.
Valente M, Sweetow R, Potts L, Bingea B. (1999). Digital
versus analog signal processing: effect of directional
microphone. J Amer Acad Audioll0:133-150.
Ventry I, Weinstein B. (1982). The hearing handicap inventory for the elderly: a new tool. Ear Hear 3:128-134.
Watson N, Knudsen V. (1940). Selective amplification in
hearing aids. J Acoust Soc Amer 11:406-419.
Willot J. (1996). Physiological plasticity in the auditory
system and its possible relevance to hearing aid use, deprivation effects, and acclimatization. Ear Hear 17(3):
66S-77S.
Delivered by Ingenta to: Washington University School of Medicine Library
IP : 128.252.10.42 On: Tue, 20 Sep 2011 18:54:58
99
Download PDF
Similar pages